Hi,
Please see question in my screenshot below: -
Best wishes,
Brett
Hi @Brett_Hunter,
input_shape defines the image shape - not the matrix. Reason for that is our model should be able to deal with any batch size, so you define only image dimensions, i.e. (28, 28, 1)
If you check the summary, it will show (None, 26, 26, 64), i.e. will be capable to deal with datasets of arbitrary number of images.
Note: In my case it shows (None, 26, 26, 32) because Iām using 32 units instead of 64.
Hope it helps,