你好，有些问题想咨询一下您~

[MYmengyun]

i download this code and change de dataset ,but my dataset is gray picture,so i know that maybe my picture is not RGB, but when i change the input channel to 1, it is not still run through this code so can you give me some advice to help me?
this is my question:
ValueError: Cannot feed value of shape (150, 32, 32) for Tensor 'Placeholder:0', which has shape '(?, 32, 32, 3)'

[NabsTao]

i download this code and change de dataset ,but my dataset is gray picture,so i know that maybe my picture is not RGB, but when i change the input channel to 1, it is not still run through this code so can you give me some advice to help me?
this is my question:
ValueError: Cannot feed value of shape (150, 32, 32) for Tensor 'Placeholder:0', which has shape '(?, 32, 32, 3)'

You can try to add something to your data set’ pictures, such as a red point.
And then your pictures will have 3 channels.

[hujunxianligong]

You should expand a dimension by np.expand or tf.expand after the last dimension as the channel.
The input should have 4 dimensions.