预创建的 Estimators

在 tensorFlow.google.cn 上查看

在 Google Colab 中运行

在 GitHub 上查看源代码

下载笔记本

警告：不建议将 Estimator 用于新代码。Estimator 运行 v1.Session 风格的代码，此类代码更加难以正确编写，并且可能会出现意外行为，尤其是与 TF 2 代码结合使用时。Estimator 确实在我们的兼容性保证范围内，但除了安全漏洞之外不会得到任何修复。请参阅迁移指南以了解详情。

本教程向您展示了如何使用 Estimator 在 TensorFlow 中解决鸢尾花分类问题。Estimator 是完整模型在旧版 TensorFlow 中的高级表示。有关更多详细信息，请参阅 Estimator。

注：在 TensorFlow 2.0 中，Keras API 可以完成这些相同的任务，并且被认为是一个更容易学习的 API。如果您刚入门，建议您从 Keras 开始。

首先要做的事

为了开始，您将首先导入 Tensorflow 和一系列您需要的库。

import tensorflow as tf

import pandas as pd

2023-11-07 19:05:32.884593: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered
2023-11-07 19:05:32.884639: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered
2023-11-07 19:05:32.886222: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered

数据集

本文档中的示例程序构建并测试了一个模型，该模型根据花萼和花瓣的大小将鸢尾花分成三种物种。

您将使用鸢尾花数据集训练模型。该数据集包括四个特征和一个标签。这四个特征确定了单个鸢尾花的以下植物学特征：

• 花萼长度
• 花萼宽度
• 花瓣长度
• 花瓣宽度

根据这些信息，您可以定义一些有用的常量来解析数据：

CSV_COLUMN_NAMES = ['SepalLength', 'SepalWidth', 'PetalLength', 'PetalWidth', 'Species']
SPECIES = ['Setosa', 'Versicolor', 'Virginica']

接下来，使用 Keras 与 Pandas 下载并解析鸢尾花数据集。注意为训练和测试保留不同的数据集。

train_path = tf.keras.utils.get_file(
    "iris_training.csv", "https://storage.googleapis.com/download.tensorflow.org/data/iris_training.csv")
test_path = tf.keras.utils.get_file(
    "iris_test.csv", "https://storage.googleapis.com/download.tensorflow.org/data/iris_test.csv")

train = pd.read_csv(train_path, names=CSV_COLUMN_NAMES, header=0)
test = pd.read_csv(test_path, names=CSV_COLUMN_NAMES, header=0)

Downloading data from https://storage.googleapis.com/download.tensorflow.org/data/iris_training.csv
2194/2194 [==============================] - 0s 0us/step
Downloading data from https://storage.googleapis.com/download.tensorflow.org/data/iris_test.csv
573/573 [==============================] - 0s 0us/step

通过检查数据您可以发现有四列浮点型特征和一列 int32 型标签。

train.head()

对于每个数据集都分割出标签，模型将被训练来预测这些标签。

train_y = train.pop('Species')
test_y = test.pop('Species')

# The label column has now been removed from the features.
train.head()

Estimator 编程概述

现在您已经设置了数据，可以使用 TensorFlow Estimator 定义模型。 Estimator 是从 tf.estimator.Estimator 派生的任何类。TensorFlow 提供了一组 tf.estimator（例如 LinearRegressor）来实现常见的 ML 算法。除此之外，您可以编写自己的自定义 Estimator。建议在刚开始时使用预制的 Estimator。

为了编写基于预创建的 Estimator 的 Tensorflow 项目，您必须完成以下工作：

• 创建一个或多个输入函数
• 定义模型的特征列
• 实例化一个 Estimator，指定特征列和各种超参数。
• 在 Estimator 对象上调用一个或多个方法，传递合适的输入函数以作为数据源。

我们来看看这些任务是如何在鸢尾花分类中实现的。

创建输入函数

您必须创建输入函数来提供用于训练、评估和预测的数据。

输入函数是一个返回 tf.data.Dataset 对象的函数，此对象会输出下列含两个元素的元组：

• features——Python字典，其中：
  • 每个键都是特征名称
  • 每个值都是包含此特征所有值的数组
• label 包含每个样本的标签的值的数组。

为了向您展示输入函数的格式，请查看下面这个简单的实现：

def input_evaluation_set():
    features = {'SepalLength': np.array([6.4, 5.0]),
                'SepalWidth':  np.array([2.8, 2.3]),
                'PetalLength': np.array([5.6, 3.3]),
                'PetalWidth':  np.array([2.2, 1.0])}
    labels = np.array([2, 1])
    return features, labels

您的输入函数可以用您喜欢的任何方式生成 features字典和label 列表。但是，推荐使用 TensorFlow 的 Dataset API，它可以解析各种数据。

Dataset API 可以为您处理很多常见情况。例如，使用 Dataset API，您可以轻松地从大量文件中并行读取记录，并将它们合并为单个数据流。

为了简化此示例，我们将使用 pandas 加载数据，并利用此内存数据构建输入管道。

def input_fn(features, labels, training=True, batch_size=256):
    """An input function for training or evaluating"""
    # Convert the inputs to a Dataset.
    dataset = tf.data.Dataset.from_tensor_slices((dict(features), labels))

    # Shuffle and repeat if you are in training mode.
    if training:
        dataset = dataset.shuffle(1000).repeat()

    return dataset.batch(batch_size)

定义特征列（feature columns）

特征列（feature columns）是一个对象，用于描述模型应该如何使用特征字典中的原始输入数据。当您构建一个 Estimator 模型的时候，您会向其传递一个特征列的列表，其中包含您希望模型使用的每个特征。tf.feature_column 模块提供了许多为模型表示数据的选项。

对于鸢尾花，4 个原始特征是数值，因此您将构建一个特征列列表来告诉 Estimator 模型将四个特征中的每一个表示为 32 位浮点值。因此，创建特征列的代码为：

# Feature columns describe how to use the input.
my_feature_columns = []
for key in train.keys():
    my_feature_columns.append(tf.feature_column.numeric_column(key=key))

WARNING:tensorflow:From /tmpfs/tmp/ipykernel_420919/1593920324.py:4: numeric_column (from tensorflow.python.feature_column.feature_column_v2) is deprecated and will be removed in a future version.
Instructions for updating:
Use Keras preprocessing layers instead, either directly or via the `tf.keras.utils.FeatureSpace` utility. Each of `tf.feature_column.*` has a functional equivalent in `tf.keras.layers` for feature preprocessing when training a Keras model.

特征列可能比这里显示的要复杂得多。您可以在此指南中阅读有关特征列的更多信息。

我们已经介绍了如何使模型表示原始特征，现在您可以构建 Estimator 了。

实例化 Estimator

鸢尾花为题是一个经典的分类问题。幸运的是，Tensorflow 提供了几个预创建的 Estimator 分类器，其中包括：

• tf.estimator.DNNClassifier 用于多类别分类的深度模型
• tf.estimator.DNNLinearCombinedClassifier 用于广度与深度模型
• tf.estimator.LinearClassifier 用于基于线性模型的分类器

对于鸢尾花问题，tf.estimator.DNNClassifier 似乎是最好的选择。您可以这样实例化该 Estimator：

# Build a DNN with 2 hidden layers with 30 and 10 hidden nodes each.
classifier = tf.estimator.DNNClassifier(
    feature_columns=my_feature_columns,
    # Two hidden layers of 30 and 10 nodes respectively.
    hidden_units=[30, 10],
    # The model must choose between 3 classes.
    n_classes=3)

WARNING:tensorflow:From /tmpfs/tmp/ipykernel_420919/2221267581.py:2: DNNClassifierV2.__init__ (from tensorflow_estimator.python.estimator.canned.dnn) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/head/head_utils.py:59: MultiClassHead.__init__ (from tensorflow_estimator.python.estimator.head.multi_class_head) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/canned/dnn.py:759: Estimator.__init__ (from tensorflow_estimator.python.estimator.estimator) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/estimator.py:1844: RunConfig.__init__ (from tensorflow_estimator.python.estimator.run_config) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Using default config.
WARNING:tensorflow:Using temporary folder as model directory: /tmpfs/tmp/tmpoulu7cx6
INFO:tensorflow:Using config: {'_model_dir': '/tmpfs/tmp/tmpoulu7cx6', '_tf_random_seed': None, '_save_summary_steps': 100, '_save_checkpoints_steps': None, '_save_checkpoints_secs': 600, '_session_config': allow_soft_placement: true
graph_options {
  rewrite_options {
    meta_optimizer_iterations: ONE
  }
}
, '_keep_checkpoint_max': 5, '_keep_checkpoint_every_n_hours': 10000, '_log_step_count_steps': 100, '_train_distribute': None, '_device_fn': None, '_protocol': None, '_eval_distribute': None, '_experimental_distribute': None, '_experimental_max_worker_delay_secs': None, '_session_creation_timeout_secs': 7200, '_checkpoint_save_graph_def': True, '_service': None, '_cluster_spec': ClusterSpec({}), '_task_type': 'worker', '_task_id': 0, '_global_id_in_cluster': 0, '_master': '', '_evaluation_master': '', '_is_chief': True, '_num_ps_replicas': 0, '_num_worker_replicas': 1}

训练、评估和预测

我们已经有一个 Estimator 对象，现在可以调用方法来执行下列操作：

• 训练模型。
• 评估经过训练的模型。
• 使用经过训练的模型进行预测。

训练模型

通过调用 Estimator 的 Train 方法来训练模型，如下所示：

# Train the Model.
classifier.train(
    input_fn=lambda: input_fn(train, train_y, training=True),
    steps=5000)

WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/estimator.py:385: StopAtStepHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Calling model_fn.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/keras/src/optimizers/legacy/adagrad.py:93: calling Constant.__init__ (from tensorflow.python.ops.init_ops) with dtype is deprecated and will be removed in a future version.
Instructions for updating:
Call initializer instance with the dtype argument instead of passing it to the constructor
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/model_fn.py:250: EstimatorSpec.__new__ (from tensorflow_estimator.python.estimator.model_fn) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Done calling model_fn.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/estimator.py:1416: NanTensorHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/estimator.py:1419: LoggingTensorHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/basic_session_run_hooks.py:232: SecondOrStepTimer.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/estimator.py:1456: CheckpointSaverHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Create CheckpointSaverHook.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/monitored_session.py:579: StepCounterHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/monitored_session.py:586: SummarySaverHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
2023-11-07 19:05:37.897085: W tensorflow/core/common_runtime/type_inference.cc:339] Type inference failed. This indicates an invalid graph that escaped type checking. Error message: INVALID_ARGUMENT: expected compatible input types, but input 1:
type_id: TFT_OPTIONAL
args {
  type_id: TFT_PRODUCT
  args {
    type_id: TFT_TENSOR
    args {
      type_id: TFT_INT64
    }
  }
}
 is neither a subtype nor a supertype of the combined inputs preceding it:
type_id: TFT_OPTIONAL
args {
  type_id: TFT_PRODUCT
  args {
    type_id: TFT_TENSOR
    args {
      type_id: TFT_INT32
    }
  }
}

    for Tuple type infernce function 0
    while inferring type of node 'dnn/zero_fraction/cond/output/_18'
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 0...
INFO:tensorflow:Saving checkpoints for 0 into /tmpfs/tmp/tmpoulu7cx6/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 0...
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/monitored_session.py:1455: SessionRunArgs.__new__ (from tensorflow.python.training.session_run_hook) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/monitored_session.py:1454: SessionRunContext.__init__ (from tensorflow.python.training.session_run_hook) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/monitored_session.py:1474: SessionRunValues.__new__ (from tensorflow.python.training.session_run_hook) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:loss = 1.5073682, step = 0
INFO:tensorflow:global_step/sec: 439.465
INFO:tensorflow:loss = 1.0172048, step = 100 (0.229 sec)
INFO:tensorflow:global_step/sec: 606.207
INFO:tensorflow:loss = 0.88746536, step = 200 (0.165 sec)
INFO:tensorflow:global_step/sec: 616.132
INFO:tensorflow:loss = 0.8561147, step = 300 (0.162 sec)
INFO:tensorflow:global_step/sec: 609.728
INFO:tensorflow:loss = 0.8380361, step = 400 (0.164 sec)
INFO:tensorflow:global_step/sec: 612.992
INFO:tensorflow:loss = 0.83253044, step = 500 (0.163 sec)
INFO:tensorflow:global_step/sec: 613.225
INFO:tensorflow:loss = 0.7997909, step = 600 (0.163 sec)
INFO:tensorflow:global_step/sec: 620.106
INFO:tensorflow:loss = 0.8083163, step = 700 (0.161 sec)
INFO:tensorflow:global_step/sec: 600.24
INFO:tensorflow:loss = 0.7587497, step = 800 (0.166 sec)
INFO:tensorflow:global_step/sec: 622.992
INFO:tensorflow:loss = 0.76957697, step = 900 (0.160 sec)
INFO:tensorflow:global_step/sec: 608.274
INFO:tensorflow:loss = 0.7207211, step = 1000 (0.164 sec)
INFO:tensorflow:global_step/sec: 613.67
INFO:tensorflow:loss = 0.7105023, step = 1100 (0.163 sec)
INFO:tensorflow:global_step/sec: 623.223
INFO:tensorflow:loss = 0.7275356, step = 1200 (0.161 sec)
INFO:tensorflow:global_step/sec: 584.544
INFO:tensorflow:loss = 0.7416762, step = 1300 (0.171 sec)
INFO:tensorflow:global_step/sec: 595.923
INFO:tensorflow:loss = 0.716245, step = 1400 (0.168 sec)
INFO:tensorflow:global_step/sec: 594.104
INFO:tensorflow:loss = 0.69990337, step = 1500 (0.168 sec)
INFO:tensorflow:global_step/sec: 596.8
INFO:tensorflow:loss = 0.69416165, step = 1600 (0.168 sec)
INFO:tensorflow:global_step/sec: 597.293
INFO:tensorflow:loss = 0.67331016, step = 1700 (0.167 sec)
INFO:tensorflow:global_step/sec: 601.12
INFO:tensorflow:loss = 0.6699522, step = 1800 (0.166 sec)
INFO:tensorflow:global_step/sec: 589.327
INFO:tensorflow:loss = 0.66161495, step = 1900 (0.170 sec)
INFO:tensorflow:global_step/sec: 586.803
INFO:tensorflow:loss = 0.6554887, step = 2000 (0.170 sec)
INFO:tensorflow:global_step/sec: 587.731
INFO:tensorflow:loss = 0.6613943, step = 2100 (0.170 sec)
INFO:tensorflow:global_step/sec: 601.215
INFO:tensorflow:loss = 0.6285989, step = 2200 (0.166 sec)
INFO:tensorflow:global_step/sec: 603.741
INFO:tensorflow:loss = 0.64100504, step = 2300 (0.166 sec)
INFO:tensorflow:global_step/sec: 604.138
INFO:tensorflow:loss = 0.62196255, step = 2400 (0.165 sec)
INFO:tensorflow:global_step/sec: 599.965
INFO:tensorflow:loss = 0.59547615, step = 2500 (0.167 sec)
INFO:tensorflow:global_step/sec: 593.705
INFO:tensorflow:loss = 0.5903188, step = 2600 (0.168 sec)
INFO:tensorflow:global_step/sec: 604.55
INFO:tensorflow:loss = 0.616672, step = 2700 (0.165 sec)
INFO:tensorflow:global_step/sec: 627.945
INFO:tensorflow:loss = 0.60870504, step = 2800 (0.159 sec)
INFO:tensorflow:global_step/sec: 623.905
INFO:tensorflow:loss = 0.59756136, step = 2900 (0.160 sec)
INFO:tensorflow:global_step/sec: 613.108
INFO:tensorflow:loss = 0.5934744, step = 3000 (0.163 sec)
INFO:tensorflow:global_step/sec: 616.35
INFO:tensorflow:loss = 0.59139955, step = 3100 (0.162 sec)
INFO:tensorflow:global_step/sec: 599.788
INFO:tensorflow:loss = 0.584731, step = 3200 (0.167 sec)
INFO:tensorflow:global_step/sec: 594.577
INFO:tensorflow:loss = 0.5786096, step = 3300 (0.168 sec)
INFO:tensorflow:global_step/sec: 605.614
INFO:tensorflow:loss = 0.58198833, step = 3400 (0.165 sec)
INFO:tensorflow:global_step/sec: 594.772
INFO:tensorflow:loss = 0.57257384, step = 3500 (0.168 sec)
INFO:tensorflow:global_step/sec: 585.277
INFO:tensorflow:loss = 0.5604176, step = 3600 (0.171 sec)
INFO:tensorflow:global_step/sec: 604.04
INFO:tensorflow:loss = 0.550858, step = 3700 (0.166 sec)
INFO:tensorflow:global_step/sec: 585.579
INFO:tensorflow:loss = 0.57899547, step = 3800 (0.171 sec)
INFO:tensorflow:global_step/sec: 593.479
INFO:tensorflow:loss = 0.54325897, step = 3900 (0.168 sec)
INFO:tensorflow:global_step/sec: 605.619
INFO:tensorflow:loss = 0.57464546, step = 4000 (0.165 sec)
INFO:tensorflow:global_step/sec: 615.744
INFO:tensorflow:loss = 0.54382163, step = 4100 (0.162 sec)
INFO:tensorflow:global_step/sec: 612.668
INFO:tensorflow:loss = 0.5404015, step = 4200 (0.163 sec)
INFO:tensorflow:global_step/sec: 589.679
INFO:tensorflow:loss = 0.5463786, step = 4300 (0.169 sec)
INFO:tensorflow:global_step/sec: 587.872
INFO:tensorflow:loss = 0.54900044, step = 4400 (0.170 sec)
INFO:tensorflow:global_step/sec: 602.697
INFO:tensorflow:loss = 0.52490914, step = 4500 (0.166 sec)
INFO:tensorflow:global_step/sec: 601.161
INFO:tensorflow:loss = 0.51717925, step = 4600 (0.166 sec)
INFO:tensorflow:global_step/sec: 615.13
INFO:tensorflow:loss = 0.51362664, step = 4700 (0.163 sec)
INFO:tensorflow:global_step/sec: 620.598
INFO:tensorflow:loss = 0.5242693, step = 4800 (0.161 sec)
INFO:tensorflow:global_step/sec: 602.749
INFO:tensorflow:loss = 0.5284901, step = 4900 (0.166 sec)
INFO:tensorflow:Calling checkpoint listeners before saving checkpoint 5000...
INFO:tensorflow:Saving checkpoints for 5000 into /tmpfs/tmp/tmpoulu7cx6/model.ckpt.
INFO:tensorflow:Calling checkpoint listeners after saving checkpoint 5000...
INFO:tensorflow:Loss for final step: 0.512492.
<tensorflow_estimator.python.estimator.canned.dnn.DNNClassifierV2 at 0x7fa8337c7ac0>

注意将 input_fn 调用封装在 lambda 中以获取参数，同时提供不带参数的输入函数，如 Estimator 所预期的那样。step 参数告知该方法在训练多少步后停止训练。

评估经过训练的模型

现在模型已经经过训练，您可以获取一些关于模型性能的统计信息。代码块将在测试数据上对经过训练的模型的准确率（accuracy）进行评估：

eval_result = classifier.evaluate(
    input_fn=lambda: input_fn(test, test_y, training=False))

print('\nTest set accuracy: {accuracy:0.3f}\n'.format(**eval_result))

INFO:tensorflow:Calling model_fn.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Starting evaluation at 2023-11-07T19:05:47
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow/python/training/evaluation.py:260: FinalOpsHook.__init__ (from tensorflow.python.training.basic_session_run_hooks) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmpfs/tmp/tmpoulu7cx6/model.ckpt-5000
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
INFO:tensorflow:Inference Time : 0.70216s
INFO:tensorflow:Finished evaluation at 2023-11-07-19:05:47
INFO:tensorflow:Saving dict for global step 5000: accuracy = 0.53333336, average_loss = 0.6654332, global_step = 5000, loss = 0.6654332
INFO:tensorflow:Saving 'checkpoint_path' summary for global step 5000: /tmpfs/tmp/tmpoulu7cx6/model.ckpt-5000

Test set accuracy: 0.533

与对 train 方法的调用不同，我们没有传递 steps 参数来进行评估。用于评估的 input_fn 只生成一个 epoch 的数据。

eval_result 字典亦包含 average_loss（每个样本的平均误差），loss（每个 mini-batch 的平均误差）与 Estimator 的 global_step（经历的训练迭代次数）值。

利用经过训练的模型进行预测（推理）

我们已经有一个经过训练的模型，可以生成准确的评估结果。我们现在可以使用经过训练的模型，根据一些无标签测量结果预测鸢尾花的品种。与训练和评估一样，我们使用单个函数调用进行预测：

# Generate predictions from the model
expected = ['Setosa', 'Versicolor', 'Virginica']
predict_x = {
    'SepalLength': [5.1, 5.9, 6.9],
    'SepalWidth': [3.3, 3.0, 3.1],
    'PetalLength': [1.7, 4.2, 5.4],
    'PetalWidth': [0.5, 1.5, 2.1],
}

def input_fn(features, batch_size=256):
    """An input function for prediction."""
    # Convert the inputs to a Dataset without labels.
    return tf.data.Dataset.from_tensor_slices(dict(features)).batch(batch_size)

predictions = classifier.predict(
    input_fn=lambda: input_fn(predict_x))

predict 方法返回一个 Python 可迭代对象，为每个样本生成一个预测结果字典。以下代码输出了一些预测及其概率：

for pred_dict, expec in zip(predictions, expected):
    class_id = pred_dict['class_ids'][0]
    probability = pred_dict['probabilities'][class_id]

    print('Prediction is "{}" ({:.1f}%), expected "{}"'.format(
        SPECIES[class_id], 100 * probability, expec))

INFO:tensorflow:Calling model_fn.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/head/base_head.py:786: ClassificationOutput.__init__ (from tensorflow.python.saved_model.model_utils.export_output) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
WARNING:tensorflow:From /tmpfs/src/tf_docs_env/lib/python3.9/site-packages/tensorflow_estimator/python/estimator/head/multi_class_head.py:455: PredictOutput.__init__ (from tensorflow.python.saved_model.model_utils.export_output) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.keras instead.
INFO:tensorflow:Done calling model_fn.
INFO:tensorflow:Graph was finalized.
INFO:tensorflow:Restoring parameters from /tmpfs/tmp/tmpoulu7cx6/model.ckpt-5000
INFO:tensorflow:Running local_init_op.
INFO:tensorflow:Done running local_init_op.
Prediction is "Setosa" (78.5%), expected "Setosa"
Prediction is "Virginica" (40.7%), expected "Versicolor"
Prediction is "Virginica" (75.4%), expected "Virginica"