I am creating a small CNN for 2-class text classification. I was able to create and run (successfully) the CNN with a single convolution layer, but when I try adding a second, I get an error that I cannot resolve. The error is on the output from the second conv.
The NN compiles and starts to fit but then fails with an error.
I have tried removing the first conv and maxpool layers, and everything worked.
Suggestions on what to do would be appreciated.
kerCNN2 = keras.Sequential()
kerCNN2.add(keras.layers.Embedding(len(dictChck), 32))
kerCNN2.add(keras.layers.Conv1D(24,5,activation=tf.nn.relu))
kerCNN2.add(keras.layers.MaxPooling1D(5))
kerCNN2.add(keras.layers.Conv1D(16,5,activation=tf.nn.relu))
kerCNN2.add(keras.layers.GlobalAveragePooling1D())
kerCNN2.add(keras.layers.Dense(16, activation=tf.nn.relu))
kerCNN2.add(keras.layers.Dense(1, activation=tf.nn.sigmoid))
kerCNN2.summary()
kerCNN2.compile(optimizer="adam", loss="binary_crossentropy", metrics=["acc"])
trainHistCNN2 = kerCNN2.fit(encTrain, trainYPartial, epochs = 1, batch_size = 128, validation_data=(encTrainEval, trainYEval), verbose=1)
The results of compilation:
_________________________________________________________________
Layer (type) Output Shape Param #
=================================================================
embedding_23 (Embedding) (None, None, 32) 76915776
_________________________________________________________________
conv1d_32 (Conv1D) (None, None, 24) 3864
_________________________________________________________________
max_pooling1d_13 (MaxPooling (None, None, 24) 0
_________________________________________________________________
conv1d_33 (Conv1D) (None, None, 16) 1936
_________________________________________________________________
global_average_pooling1d_3 ( (None, 16) 0
_________________________________________________________________
dense_31 (Dense) (None, 16) 272
_________________________________________________________________
dense_32 (Dense) (None, 1) 17
=================================================================
Total params: 76,921,865
Trainable params: 76,921,865
Non-trainable params: 0
The (relevant portion of) error:
InvalidArgumentError (see above for traceback): computed output size would be negative
[[Node: conv1d_33/convolution/Conv2D = Conv2D[T=DT_FLOAT, data_format="NHWC", padding="VALID", strides=[1, 1, 1, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/cpu:0"](conv1d_33/convolution/ExpandDims, conv1d_33/convolution/ExpandDims_1)]]
That's because your Tensor shape is less then the size of conv kernel.
e.g. Tensor shape is (None, None, 10, None), but the filter of conv is (X, 16, X, X).
10 is less then 16.
If you love us? You can donate to us via Paypal or buy me a coffee so we can maintain and grow! Thank you!
Donate Us With