Posted on 17th September 2024| views
The input shape:
Within Keras, the input layer is not a layer; it is just a tensor. It is the origin tensor you transmit to the initial hidden layer. That tensor should possess the identical shape as your training data. For Example: if you have 15 images of 60x60 pixels in RGB 3 channels, then the shape of your input data will be something like this (15,60,60,3).