Hands-On Machine Learning with TensorFlow.js

The Keras HDF5 model

To convert a Keras HDF5 model into the web format, you need to specify the keras input format option. Let's assume we want to convert the following model trained through the Keras API into the web format:

import tensorflow as tffrom tensorflow.keras import layersimport numpy as npmodel = tf.keras.Sequential()model.add(layers.Dense(32, activation='relu'))# Add another:model.add(layers.Dense(10, activation='softmax'))model.compile(optimizer=tf.train.AdamOptimizer(0.001),              loss='categorical_crossentropy',              metrics=['accuracy'])data = np.random.random((1000, 32))labels = np.random.random((1000, 10))model.fit(data, labels, epochs=10, batch_size=32)model.save('my_keras_model.h5')

You do not need to specify additional options ...