0

I have thousands image size of (750,750,3). I want to feed these images to 1D CNN. How can I convert this input shape to be utilized in 1D CNN? What should be the input shape? How can I do that before training process?

I have tried to reshape the size of the images but not sure how to do so as I am new to CNN.

Didar
  • 1

1 Answers1

0

You are likely looking for a flatten operation. Flattening reduces a M x N x O shaped array to a M * N * O 1-D array.

Consider this example

So your (750, 750, 3) RGB image turns into a (1687500) array. Keep in mind that there are different options (channel first, etc.).

code-lukas
  • 1,586
  • 9
  • 19