通过相对路径保存 h5 模型 - Keras tensorflow

Question

Hey to my beloved community.

How do we save a model to a relative file path using tensorflow embedded keras?

    model.save('/models/model.h5')

I tried:

'./models/'
'/models/'
'models/'

Neither seems to work and I always end up with:

InvalidArgumentError (see above for traceback): Failed to create a NewWriteableFile:

I don't want to provide an absolute path as it might change dynamically.

Answer 1

The error might be related to one of the reasons below:

• Do the specified folder exists? If not, you should make it using:

   import os os.makedirs('models/') # Creating a directory model.save('models/model.h5') # Saving model

• Do you have permission to write to that folder? If you are using Unix-based system (ie Mac OS or Linux), then you can check it by:

   ls -l models/

  A friendly tutorial on Unix file permission can be found here .

• Does your path contains special symbols? (ie ~ for user's folder) you will need to use os.path.expanduser to resolve the path.

Answer 2

Adding to what @FalconUA mentioned, Here I am providing an example of saving a tensorflow.keras model to be saved in model_path folder under current directory. This works well with most recent tensorflow ( TF2.0.0rc2 ).

import tensorflow as tf
from tensorflow import keras
mnist = tf.keras.datasets.mnist

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

# create a model
def create_model():
  model = tf.keras.models.Sequential([
    tf.keras.layers.Flatten(input_shape=(28, 28)),
    tf.keras.layers.Dense(512, activation=tf.nn.relu),
    tf.keras.layers.Dropout(0.2),
    tf.keras.layers.Dense(10, activation=tf.nn.softmax)
    ])
# compile the model
  model.compile(optimizer='adam',
              loss='sparse_categorical_crossentropy',
              metrics=['accuracy'])
  return model

# Create a basic model instance
model=create_model()

model.fit(x_train, y_train, epochs=1)
loss, acc = model.evaluate(x_test, y_test,verbose=1)
print("Original model, accuracy: {:5.2f}%".format(100*acc))

# Save entire model to a HDF5 file
model.save('./model_path/my_model.h5')

# Recreate the exact same model, including weights and optimizer.
new_model = keras.models.load_model('./model_path/my_model.h5')
loss, acc = new_model.evaluate(x_test, y_test)
print("Restored model, accuracy: {:5.2f}%".format(100*acc))

Answer 3

Here is a simple example of a function that responds to the question. If the folder does not exist create a new one else use the existing folder.

def save_model(path, model):
    if not os.path.exists(path):
        print('save directories...', flush=True)
        os.makedirs(path)
    model.save(path + '/handrecognition_model.h5')