I comes from CNN model. I am going to study LSTM. In my case, I am going to make a classification of PNG images. What I know about CNN is:
- The input of CNN is the shape of image. e.g. (width, height, channels). In my case is (128, 128, 3)
But how can I build a LSTM model for such data. What I know about LSTM is:
- The "unit" parameter stands for how many notes in the hidden layers.
How can I set the input_shape of a RNN model?
