Color images have 3 channel, read out the shape of (x, y, 3), where x and y is the height to width of the picture, 3 is RGB three channels, but LSTM can receive (1, 1, 1) of the 3 d data, the general how to deal with? Need to manually write a loop three channels into strip? Or there is a better way??
CodePudding user response:
https://blog.csdn.net/mangobar/article/details/105200963
CodePudding user response:
https://blog.csdn.net/mangobar/article/details/105248260
CodePudding user response:
https://blog.csdn.net/mangobar/article/details/105406137