将 Keras 模型导出为 TF Estimator:找不到训练好的模型 [英] Exporting a Keras model as a TF Estimator: trained model cannot be found

问题描述

在尝试将 Keras 模型导出为 TensorFlow Estimator 以便为模型提供服务时，我遇到了以下问题.由于同样的问题也出现了 在对这个问题的回答 中，我将说明在一个玩具示例中会发生什么，并提供我的解决方案以用于文档目的.Tensorflow 1.12.0 和 Keras 2.2.4 会出现此行为.实际的 Keras 和 tf.keras 都会发生这种情况.

当尝试使用 tf.keras.estimator.model_to_estimator 导出从 Keras 模型创建的 Estimator 时会出现问题.调用 estimator.export_savedmodel 时，会抛出 NotFoundError 或 ValueError.

以下代码为一个玩具示例重现了这一点.

创建一个 Keras 模型并保存:

导入keras模型 = keras.Sequential()模型.添加(keras.layers.Dense(单位=1，激活='sigmoid',input_shape=(10, )))model.compile(loss='binary_crossentropy', 优化器='sgd')model.save('./model.h5')

接下来，使用tf.keras.estimator.model_to_estimator将模型转换为estimator，添加输入接收器函数，并使用以Savedmodel格式导出estimator.export_savedmodel:

# 将 keras 模型转换为 TF estimatortf_files_path = './tf'估计量 =	f.keras.estimator.model_to_estimator(keras_model=model,model_dir=tf_files_path)defserving_input_receiver_fn():返回 tf.estimator.export.build_raw_serving_input_receiver_fn({model.input_names[0]: tf.placeholder(tf.float32, shape=[None, 10])})# 导出估计器export_path = './export'estimator.export_savedmodel(出口路径，serving_input_receiver_fn=serving_input_receiver_fn())

这会抛出:

ValueError: 无法在 ./tf.

解决方案

我的解决方案如下.检查 ./tf 文件夹清楚地表明对 model_to_estimator 的调用将必要的文件存储在 keras 子文件夹中，而 export_model 期望这些文件直接位于 ./tf 文件夹中，因为这是我们为 model_dir 参数指定的路径:

$ 树 ./tf./tf└── 凯拉斯├──检查站├── keras_model.ckpt.data-00000-of-00001├── keras_model.ckpt.index└── keras_model.ckpt.meta1个目录，4个文件

简单的解决方法是将这些文件向上移动一个文件夹.这可以用 Python 来完成:

导入操作系统进口木材从 pathlib 导入路径def up_one_dir(路径):"""将path中的所有文件向上移动一个文件夹，并删除空文件夹"""parent_dir = str(Path(path).parents[0])对于 os.listdir(path) 中的 f:Shutil.move(os.path.join(path, f), parent_dir)关闭.rmtree(路径)up_one_dir('./tf/keras')

这将使 model_dir 目录看起来像这样:

$ 树 ./tf./tf├──检查站├── keras_model.ckpt.data-00000-of-00001├── keras_model.ckpt.index└── keras_model.ckpt.meta0 个目录，4 个文件

在 model_to_estimator 和 export_savedmodel 调用之间执行此操作允许根据需要导出模型:

export_path = './export'estimator.export_savedmodel(出口路径，serving_input_receiver_fn=serving_input_receiver_fn())

<块引用>

INFO:tensorflow:SavedModel 写入:./export/temp-b'1549796240'/saved_model.pb

I encountered the following issue when trying to export a Keras model as a TensorFlow Estimator with the purpose of serving the model. Since the same problem also popped up in an answer to this question, I will illustrate what happens on a toy example and provide my workaround solution for documentation purposes. This behaviour occurs with Tensorflow 1.12.0 and Keras 2.2.4. This happens with actual Keras as well as with tf.keras.

The problem occurs when trying to export an Estimator that was created from a Keras model with tf.keras.estimator.model_to_estimator. Upon calling estimator.export_savedmodel, either a NotFoundError or a ValueError is thrown.

The below code reproduces this for a toy example.

Create a Keras model and save it:

import keras
model = keras.Sequential()
model.add(keras.layers.Dense(units=1,
                                activation='sigmoid',
                                input_shape=(10, )))
model.compile(loss='binary_crossentropy', optimizer='sgd')
model.save('./model.h5')

Next, convert the model to an estimator with tf.keras.estimator.model_to_estimator, add an input receiver function and export it in the Savedmodel format with estimator.export_savedmodel:

# Convert keras model to TF estimator
tf_files_path = './tf'
estimator =
    tf.keras.estimator.model_to_estimator(keras_model=model,
                                          model_dir=tf_files_path)
def serving_input_receiver_fn():
    return tf.estimator.export.build_raw_serving_input_receiver_fn(
        {model.input_names[0]: tf.placeholder(tf.float32, shape=[None, 10])})

# Export the estimator
export_path = './export'
estimator.export_savedmodel(
    export_path,
    serving_input_receiver_fn=serving_input_receiver_fn())

This will throw:

ValueError: Couldn't find trained model at ./tf.

解决方案

My workaround solution is as follows. Inspecting the ./tf folder makes clear that the call to model_to_estimator stored the necessary files in a keras subfolder, while export_model expects those files to be in the ./tf folder directly, as this is the path we specified for the model_dir argument:

$ tree ./tf
./tf
└── keras
    ├── checkpoint
    ├── keras_model.ckpt.data-00000-of-00001
    ├── keras_model.ckpt.index
    └── keras_model.ckpt.meta

1 directory, 4 files

The simple workaround is to move these files up one folder. This can be done with Python:

import os
import shutil
from pathlib import Path

def up_one_dir(path):
    """Move all files in path up one folder, and delete the empty folder
    """
    parent_dir = str(Path(path).parents[0])
    for f in os.listdir(path):
        shutil.move(os.path.join(path, f), parent_dir)
    shutil.rmtree(path)

up_one_dir('./tf/keras')

Which will make the model_dir directory look like this:

$ tree ./tf
./tf
├── checkpoint
├── keras_model.ckpt.data-00000-of-00001
├── keras_model.ckpt.index
└── keras_model.ckpt.meta

0 directories, 4 files

Doing this manipulation in between the model_to_estimator and the export_savedmodel calls allows to export the model as desired:

export_path = './export'
estimator.export_savedmodel(
    export_path,
    serving_input_receiver_fn=serving_input_receiver_fn())

INFO:tensorflow:SavedModel written to: ./export/temp-b'1549796240'/saved_model.pb

这篇关于将 Keras 模型导出为 TF Estimator:找不到训练好的模型的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！