סופר_אפוטרופוס

ds = tfds.load('huggingface:guardian_authorship/cross_topic_1')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_genre_1')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_2')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_3')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_4')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_5')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_6')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_7')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_8')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_9')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_10')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_11')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_topic_12')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

ds = tfds.load('huggingface:guardian_authorship/cross_genre_2')

A dataset cross-topic authorship attribution. The dataset is provided by Stamatatos 2013. 
1- The cross-topic scenarios are based on Table-4 in Stamatatos 2017 (Ex. cross_topic_1 => row 1:P S U&W ).
2- The cross-genre scenarios are based on Table-5 in the same paper. (Ex. cross_genre_1 => row 1:B P S&U&W).

3- The same-topic/genre scenario is created by grouping all the datasts as follows. 
For ex., to use same_topic and split the data 60-40 use:
train_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[:60%]+validation[:60%]+test[:60%]')
tests_ds = load_dataset('guardian_authorship', name="cross_topic_<<#>>", 
                        split='train[-40%:]+validation[-40%:]+test[-40%:]')            

Important: train+validation+test[:60%] will generate the wrong splits becasue the data is imbalanced

* See https://huggingface.co/docs/datasets/splits.html for detailed/more examples

{
    "author": {
        "num_classes": 13,
        "names": [
            "catherinebennett",
            "georgemonbiot",
            "hugoyoung",
            "jonathanfreedland",
            "martinkettle",
            "maryriddell",
            "nickcohen",
            "peterpreston",
            "pollytoynbee",
            "royhattersley",
            "simonhoggart",
            "willhutton",
            "zoewilliams"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "topic": {
        "num_classes": 5,
        "names": [
            "Politics",
            "Society",
            "UK",
            "World",
            "Books"
        ],
        "names_file": null,
        "id": null,
        "_type": "ClassLabel"
    },
    "article": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}