Incompatible shapes: [100,1] vs. [100,1,4,1]

Viewed 35

I get the error

Incompatible shapes: [100,1] vs. [100,1,4,1]

when I try to train on my data.

My model is like this:

train_set = windowed_dataset(x_train, window_size=60, batch_size=100, shuffle_buffer=1000)

model = tf.keras.models.Sequential([
    tf.keras.layers.Input(shape=(100, 4,)),
    tf.keras.layers.Normalization(axis=None),
    # tf.keras.layers.LSTM(64, return_sequences=True),
    # tf.keras.layers.LSTM(64),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(60, return_sequences=True)),
    tf.keras.layers.Bidirectional(tf.keras.layers.LSTM(60)),
    tf.keras.layers.Flatten(),
    tf.keras.layers.Dense(120, activation="relu"),
    tf.keras.layers.Dense(100, activation="relu"),
    tf.keras.layers.Dense(60, activation="relu"),
    tf.keras.layers.Dropout(.5),
    tf.keras.layers.Dense(30, activation="relu"),
    tf.keras.layers.Dense(10, activation="relu"),
    tf.keras.layers.Dense(1, input_shape=[None, 1, 1]),
])

And here is the compiling code

optimizer = tf.keras.optimizers.SGD(learning_rate=.01, momentum=.9, decay=.01)
model.compile(loss=tf.keras.losses.Huber(), optimizer=optimizer, metrics=["mae"])

print('Samples : %d' % len(x_train))
history = model.fit(train_set, epochs=50, steps_per_epoch=5, callbacks=[under_mae()])
0 Answers
Related