`Chapter 11 part 02 sequence models: tf.one_hot() inside the keras model error

[gayanlanke]

import tensorflow as tf
inputs = keras.Input(shape=(None,), dtype="int64")
embedded = tf.one_hot(inputs, depth=max_tokens)
x = layers.Bidirectional(layers.LSTM(32))(embedded)
x = layers.Dropout(0.5)(x)
outputs = layers.Dense(1, activation="sigmoid")(x)
model = keras.Model(inputs, outputs)
model.compile(optimizer="rmsprop",
              loss="binary_crossentropy",
              metrics=["accuracy"])
model.summary()

Error message:

ValueError: A KerasTensor cannot be used as input to a TensorFlow function. A KerasTensor is a symbolic placeholder for a shape and dtype, used when constructing Keras Functional models or Keras Functions. You can only use it as input to a Keras layer or a Keras operation (from the namespaces keras.layers and keras.operations). You are likely doing something like:

x = Input(...)
...
tf_fn(x)  # Invalid.

What you should do instead is wrap tf_fn in a layer:

class MyLayer(Layer):
    def call(self, x):
        return tf_fn(x)

x = MyLayer()(x)

The solution can be found here

class EmbeddedLayer(keras.Layer):
    def call(self, x):
        return tf.one_hot(x, depth=max_tokens)

inputs = keras.Input(shape=(None,), dtype="int64")
embedded = EmbeddedLayer()(inputs)
x = layers.Bidirectional(layers.LSTM(32))(embedded)
x = layers.Dropout(0.5)(x)
outputs = layers.Dense(1, activation="sigmoid")(x)
model = keras.Model(inputs, outputs)
model.compile(
    optimizer="rmsprop",
    loss="binary_crossentropy",
    metrics=["accuracy"],
)
model.summary()

[ifond]

​ I have received your E-mail——Steven Lee

[gayanlanke]

Found the solution in #239. Thanks.