mode.save() 只创建一个文件

Question

我阅读了 model.save() 的文档。它说它创建了文件夹，但是当我保存它时它只创建一个文件。当我使用此文件将代码从 keras 更改为 tensorflow-lite 时，它​​会出现此错误：

OSError: SavedModel file does not exist at: C:\Users\Munib\New folder\3_layer_model_try3/{saved_model.pbtxt|saved_model.pb}

我也在上传文件的图片及其位置。

Answer 1

我认为您保存的模型是 keras model 而不是 'SavedModel. So, you need to usefrom_keras_model`，如下所示。

我尝试使用简单模型来模拟问题，并成功重现了您的问题。检查下面的代码。

!pip install tf-nightly
import tensorflow as tf
mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model = tf.keras.models.Sequential([
  tf.keras.layers.Flatten(input_shape=(28, 28)),
  tf.keras.layers.Dense(128, activation='relu'),
  tf.keras.layers.Dropout(0.2),
  tf.keras.layers.Dense(10, activation='softmax')
])

model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)
model.evaluate(x_test, y_test)

训练模型后，我猜你保存如下

model.save('my_model.h5',save_format='h5')

然后尝试将模型转换为 tflite 模型（以下是您的代码）

# Converting a SavedModel to a TensorFlow Lite model. 
saved_model_dir = '/content/my_model.h5' #r"C:\Users\Munib\New folder" 
converter = tf.lite.TFLiteConverter.from_saved_model(saved_model_dir) 
tflite_model = converter.convert()

使用上面的代码，我得到了以下错误（看起来和你面临的错误一样）

---------------------------------------------------------------------------
OSError                                   Traceback (most recent call last)
<ipython-input-4-a9a46a590f90> in <module>()
      1 # Converting a SavedModel to a TensorFlow Lite model.
      2 saved_model_dir = '/content/my_model.h5' #r"C:\Users\Munib\New folder"
----> 3 converter = tf.lite.TFLiteConverter.from_saved_model(saved_model_dir)
      4 tflite_model = converter.convert()

4 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/saved_model/loader_impl.py in parse_saved_model(export_dir)
    111                   (export_dir,
    112                    constants.SAVED_MODEL_FILENAME_PBTXT,
--> 113                    constants.SAVED_MODEL_FILENAME_PB))
    114 
    115 

OSError: SavedModel file does not exist at: /content/my_model.h5/{saved_model.pbtxt|saved_model.pb}

这是更新后的代码

# Converting a SavedModel to a TensorFlow Lite model. 
saved_model_dir = '/content/my_model.h5' #r"C:\Users\Munib\New folder" 
loaded_model = tf.keras.models.load_model(saved_model_dir)
converter = tf.lite.TFLiteConverter.from_keras_model(loaded_model)# .from_saved_model(saved_model_dir) 
tflite_model = converter.convert()
open("converted_model.tflite", "wb").write(tflite_model)

完整代码是here。