My model is the following:
Model: "sequential_24"
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
embedding_17 (Embedding) (None, 100, 100) 500100
_________________________________________________________________
conv1d_14 (Conv1D) (None, 96, 128) 64128
_________________________________________________________________
global_max_pooling1d_13 (Glo (None, 128) 0
_________________________________________________________________
dense_37 (Dense) (None, 10) 1290
_________________________________________________________________
dense_38 (Dense) (None, 1) 11
=================================================================
When fitting the model
history = model.fit(train_df.sentence, train_df.label, epochs=10, validation_data=(val_df.sentence, val_df.label))
I get the following error
Epoch 1/10
WARNING:tensorflow:Model was constructed with shape (None, 100) for input KerasTensor(type_spec=TensorSpec(shape=(None, 100), dtype=tf.float32, name='embedding_14_input'), name='embedding_14_input', description="created by layer 'embedding_14_input'"), but it was called on an input with incompatible shape (None, 1)
ValueError: Negative dimension size caused by subtracting 5 from 1 for '{{node sequential_21/conv1d_11/conv1d}} = Conv2D[T=DT_FLOAT, data_format="NHWC", dilations=[1, 1, 1, 1], explicit_paddings=[], padding="VALID", strides=[1, 1, 1, 1], use_cudnn_on_gpu=true](sequential_21/conv1d_11/conv1d/ExpandDims, sequential_21/conv1d_11/conv1d/ExpandDims_1)' with input shapes: [?,1,1,100], [1,5,100,128].