Trying to create u-net like structure for signal processing using Conv1D

Question

I have tried to create a u-net-like neural network (autoencoder) structure for signal processing applications using Tensorflow2. However, I have run into some problems when training the model.

Here is how I define my neural network structure.

 def build_unet(input_shape, n_filters_list = [16, 32]):
    inputs = Input(shape=input_shape)
    print("in", inputs)
    contraction = {}
    for f in n_filters_list:
        x = Conv1D(f, 5, activation='relu', kernel_initializer='he_normal', padding='same')(inputs)
        x = Dropout(0.1)(x)
        x = Conv1D(f, 5, activation='relu', kernel_initializer='he_normal', padding='same')(x)
        contraction[f'conv{f}'] = x
        x = MaxPooling1D(pool_size=4,strides=2)(x)
        print("enc", x)
        inputs = x

    c5 = Conv1D(160, 5, activation='relu', kernel_initializer='he_normal', padding='same')(inputs)
    c5 = Dropout(0.2)(c5)
    c5 = Conv1D(160, 5, activation='relu', kernel_initializer='he_normal', padding='same')(c5)
    print("c5",c5)
    inputs = c5
    print(inputs)
    for i,f in zip([0,0],reversed(n_filters_list)):
        x = Conv1DTranspose(f, 4 + i, 2)(inputs)
        print("dec",x)
        x = concatenate([x, contraction[f'conv{f}']])
        x = Conv1D(f, 5, activation='relu', kernel_initializer='he_normal', padding='same')(x)
        x = Dropout(0.2)(x)
        x = Conv1D(f, 5, activation='relu', kernel_initializer='he_normal', padding='same')(x)
        inputs = x

    outputs = Conv1D(filters=1, kernel_size=3, activation="tanh", padding="same")(inputs)
    print("out",outputs)

    return Model(inputs=inputs, outputs=outputs)

It can compile as usual.

model = build_unet(input_shape=(3490,1))
model.compile(optimizer="Adam", loss='mean_squared_error')

I also try to see the model summary. However, it displays the unexpected result as it shows that I have only 2 layers.

And when I try to train my model as follows:

history = model.fit(training_generator,
                validation_data=validation_generator,
                epochs=100)

It has this error:

ValueError: in user code:

File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1021, in train_function  *
    return step_function(self, iterator)
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1010, in step_function  **
    outputs = model.distribute_strategy.run(run_step, args=(data,))
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 1000, in run_step  **
    outputs = model.train_step(data)
File "/usr/local/lib/python3.7/dist-packages/keras/engine/training.py", line 859, in train_step
    y_pred = self(x, training=True)
File "/usr/local/lib/python3.7/dist-packages/keras/utils/traceback_utils.py", line 67, in error_handler
    raise e.with_traceback(filtered_tb) from None
File "/usr/local/lib/python3.7/dist-packages/keras/engine/input_spec.py", line 228, in assert_input_compatibility
    raise ValueError(f'Input {input_index} of layer "{layer_name}" '

ValueError: Exception encountered when calling layer "model_1" (type Functional).

Input 0 of layer "conv1d_77" is incompatible with the layer: expected min_ndim=3, found ndim=2. Full shape received: (None, None)

Call arguments received:
  • inputs=tf.Tensor(shape=(None, None), dtype=float32)
  • training=True
  • mask=None

Could anyone provide the reason I have this unexpected model summary and error and how can I fix this?

Much appreciated.

Answer 1

You override inputs variable several times so when you create a model here Model(inputs = inputs, outputs = outputs) your inputs variable actually doesn't contain Input(shape=input_shape).

You can basically remain code as it is but add additional variable to memorize initial input layer like this initial_input = inputs = Input(shape=input_shape) and then use initial_input variable when creating model like this Model(inputs = initial_input, outputs = outputs)