চরম

তথ্যসূত্র:

• কোড
• আলিঙ্গন মুখ

এক্সএনএলআই

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XNLI')

• বর্ণনা :

The Cross-lingual Natural Language Inference (XNLI) corpus is a crowd-sourced collection of 5,000 test and
2,500 dev pairs for the MultiNLI corpus. The pairs are annotated with textual entailment and translated into
14 languages: French, Spanish, German, Greek, Bulgarian, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese,
Hindi, Swahili and Urdu. This results in 112.5k annotated pairs. Each premise can be associated with the
corresponding hypothesis in the 15 languages, summing up to more than 1.5M combinations. The corpus is made to
evaluate how to perform inference in any language (including low-resources ones like Swahili or Urdu) when only
English NLI data is available at training time. One solution is cross-lingual sentence encoding, for which XNLI
is an evaluation benchmark.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "language": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "gold_label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

টাইডিকা

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tydiqa')

• বর্ণনা :

Gold passage task (GoldP): Given a passage that is guaranteed to contain the
             answer, predict the single contiguous span of characters that answers the question. This is more similar to
             existing reading comprehension datasets (as opposed to the information-seeking task outlined above).
             This task is constructed with two goals in mind: (1) more directly comparing with prior work and (2) providing
             a simplified way for researchers to use TyDi QA by providing compatibility with existing code for SQuAD 1.1,
             XQuAD, and MLQA. Toward these goals, the gold passage task differs from the primary task in several ways:
             only the gold answer passage is provided rather than the entire Wikipedia article;
             unanswerable questions have been discarded, similar to MLQA and XQuAD;
             we evaluate with the SQuAD 1.1 metrics like XQuAD; and
            Thai and Japanese are removed since the lack of whitespace breaks some tools.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

স্কোয়াড

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/SQuAD')

• বর্ণনা :

Stanford Question Answering Dataset (SQuAD) is a reading comprehension     dataset, consisting of questions posed by crowdworkers on a set of Wikipedia     articles, where the answer to every question is a segment of text, or span,     from the corresponding reading passage, or the question might be unanswerable.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.af

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.af')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ar')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.bg

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.bg')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.bn

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.bn')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.de')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.el

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.el')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.en')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.es')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.et

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.et')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.eu

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.eu')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.fa

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.fa')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.fi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.fi')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.fr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.fr')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.he

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.he')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.hi')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.hu

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.hu')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.id

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.id')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.it

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.it')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ja

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ja')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.jv

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.jv')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ka

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ka')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.kk

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.kk')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ko

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ko')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ml

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ml')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.mr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.mr')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ms

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ms')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.my

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.my')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.nl

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.nl')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.pt

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.pt')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ru

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ru')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.sw

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.sw')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ta

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ta')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.te

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.te')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.th

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.th')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.tl

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.tl')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.tr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.tr')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.ur

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.ur')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.vi')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.yo

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.yo')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

PAN-X.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAN-X.zh')

• বর্ণনা :

The WikiANN dataset (Pan et al. 2017) is a dataset with NER annotations for PER, ORG and LOC. It has been
constructed using the linked entities in Wikipedia pages for 282 different languages including Danish. The dataset
can be loaded with the DaNLP package:
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "ner_tags": {
        "feature": {
            "num_classes": 7,
            "names": [
                "O",
                "B-PER",
                "I-PER",
                "B-ORG",
                "I-ORG",
                "B-LOC",
                "I-LOC"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "langs": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.ar.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.ar.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.de.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.de.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.vi.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.vi.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.zh.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.zh.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.en.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.en.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.es.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.es.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.ar')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.de')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.vi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.zh')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.en')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.es')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

MLQA.hi.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/MLQA.hi.hi')

• বর্ণনা :

MLQA (MultiLingual Question Answering) is a benchmark dataset for evaluating cross-lingual question answering performance.
MLQA consists of over 5K extractive QA instances (12K in English) in SQuAD format in seven languages - English, Arabic,
German, Spanish, Hindi, Vietnamese and Simplified Chinese. MLQA is highly parallel, with QA instances parallel between
4 different languages on average.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "title": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.ar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.ar')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.de')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.vi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.vi')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.zh')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.en')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.es')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.hi

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.hi')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.el

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.el')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.ru

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.ru')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.th

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.th')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

XQuAD.tr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/XQuAD.tr')

• বর্ণনা :

XQuAD (Cross-lingual Question Answering Dataset) is a benchmark dataset for evaluating cross-lingual question
answering performance. The dataset consists of a subset of 240 paragraphs and 1190 question-answer pairs from
the development set of SQuAD v1.1 (Rajpurkar et al., 2016) together with their professional translations into
ten languages: Spanish, German, Greek, Russian, Turkish, Arabic, Vietnamese, Thai, Chinese, and Hindi. Consequently,
the dataset is entirely parallel across 11 languages.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "context": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "question": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "answers": {
        "feature": {
            "answer_start": {
                "dtype": "int32",
                "id": null,
                "_type": "Value"
            },
            "text": {
                "dtype": "string",
                "id": null,
                "_type": "Value"
            }
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

bucc18.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/bucc18.de')

• বর্ণনা :

Building and Using Comparable Corpora

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

bucc18.fr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/bucc18.fr')

• বর্ণনা :

Building and Using Comparable Corpora

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

bucc18.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/bucc18.zh')

• বর্ণনা :

Building and Using Comparable Corpora

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

bucc18.ru

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/bucc18.ru')

• বর্ণনা :

Building and Using Comparable Corpora

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.de

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.de')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.en

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.en')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.es

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.es')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.fr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.fr')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.ja

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.ja')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.ko

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.ko')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

PAWS-X.zh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/PAWS-X.zh')

• বর্ণনা :

This dataset contains 23,659 human translated PAWS evaluation pairs and 296,406 machine translated training
pairs in six typologically distinct languages: French, Spanish, German, Chinese, Japanese, and Korean. All
translated pairs are sourced from examples in PAWS-Wiki.
The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "sentence1": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "sentence2": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "label": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.afr

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.afr')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.ara

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.ara')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.ben

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.ben')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.bul

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.bul')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.deu

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.deu')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.cmn

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.cmn')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.ell

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.ell')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.est

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.est')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.eus

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.eus')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.fin

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.fin')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.fra

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.fra')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.heb

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.heb')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.hin

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.hin')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.hun

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.hun')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.ind

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.ind')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.ita

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.ita')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.jav

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.jav')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.jpn

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.jpn')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.kat

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.kat')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.kaz

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.kaz')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.kor

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.kor')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.mal

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.mal')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.mar

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.mar')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.nld

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.nld')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.pes

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.pes')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.por

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.por')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.rus

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.rus')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.spa

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.spa')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.swh

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.swh')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.tam

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.tam')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.tel

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.tel')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.tgl

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.tgl')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.tha

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.tha')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.tur

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.tur')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.urd

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.urd')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tatoeba.vie

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/tatoeba.vie')

• বর্ণনা :

his data is extracted from the Tatoeba corpus, dated Saturday 2018/11/17.
For each languages, we have selected 1000 English sentences and their translations, if available. Please check
this paper for a description of the languages, their families and scripts as well as baseline results.
Please note that the English sentences are not identical for all language pairs. This means that the results are
not directly comparable across languages. In particular, the sentences tend to have less variety for several
low-resource languages, e.g. "Tom needed water", "Tom needs water", "Tom is getting water", ...

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "source_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_sentence": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "source_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    },
    "target_lang": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

udpos.আফ্রিকান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Afrikaans')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.আরবি

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Arabic')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.Basque

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Basque')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.বুলগেরিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Bulgarian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ডাচ

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Dutch')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ইংরেজি

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.English')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.এস্তোনিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Estonian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ফিনিশ

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Finnish')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ফরাসি

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.French')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.জার্মান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.German')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.গ্রীক

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Greek')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.হিব্রু

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Hebrew')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.হিন্দি

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Hindi')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.হাঙ্গেরিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Hungarian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ইন্দোনেশিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Indonesian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ইটালিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Italian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.জাপানিজ

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Japanese')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.কাজাখ

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Kazakh')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.কোরিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Korean')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.Chinese

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Chinese')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.মারাঠি

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Marathi')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.Persian

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Persian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.পর্তুগিজ

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Portuguese')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.রাশিয়ান

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Russian')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.স্প্যানিশ

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Spanish')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.Tagalog

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Tagalog')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.তামিল

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Tamil')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.তেলেগু

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Telugu')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.থাই

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Thai')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.তুর্কি

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Turkish')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.উর্দু

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Urdu')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.ভিয়েতনামী

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Vietnamese')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}

udpos.Yoruba

TFDS এ এই ডেটাসেট লোড করতে নিম্নলিখিত কমান্ডটি ব্যবহার করুন:

ds = tfds.load('huggingface:xtreme/udpos.Yoruba')

• বর্ণনা :

Universal Dependencies (UD) is a framework for consistent annotation of grammar (parts of speech, morphological
features, and syntactic dependencies) across different human languages. UD is an open community effort with over 200
contributors producing more than 100 treebanks in over 70 languages. If you’re new to UD, you should start by reading
the first part of the Short Introduction and then browsing the annotation guidelines.

The Cross-lingual TRansfer Evaluation of Multilingual Encoders (XTREME) benchmark is a benchmark for the evaluation of
the cross-lingual generalization ability of pre-trained multilingual models. It covers 40 typologically diverse languages
(spanning 12 language families) and includes nine tasks that collectively require reasoning about different levels of
syntax and semantics. The languages in XTREME are selected to maximize language diversity, coverage in existing tasks,
and availability of training data. Among these are many under-studied languages, such as the Dravidian languages Tamil
(spoken in southern India, Sri Lanka, and Singapore), Telugu and Malayalam (spoken mainly in southern India), and the
Niger-Congo languages Swahili and Yoruba, spoken in Africa.

• লাইসেন্স : কোনো পরিচিত লাইসেন্স নেই
• সংস্করণ : 1.0.0
• বিভাজন :

• বৈশিষ্ট্য :

{
    "tokens": {
        "feature": {
            "dtype": "string",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "pos_tags": {
        "feature": {
            "num_classes": 17,
            "names": [
                "ADJ",
                "ADP",
                "ADV",
                "AUX",
                "CCONJ",
                "DET",
                "INTJ",
                "NOUN",
                "NUM",
                "PART",
                "PRON",
                "PROPN",
                "PUNCT",
                "SCONJ",
                "SYM",
                "VERB",
                "X"
            ],
            "names_file": null,
            "id": null,
            "_type": "ClassLabel"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    }
}