search

cjiang2 / VDCNN

Implementation of Very Deep Convolutional Neural Network for Text Classification
171 stars 41 forks source link

downsampling return should use same padding not valid #12

Closed Lapis-Hong closed 6 years ago

Lapis-Hong commented 6 years ago

return tf.layers.conv1d( inputs=pool, filters=pool.get_shape()[2] * 2, kernel_size=1, strides=1, padding='same', use_bias=False)