CNN model testing data shape error

1

Hi, I am working with CNN in pytorch.
I trained my CNN model but while I am trying to predict my testing data it shows an error message which says:

data = data.view(-1,1,32,200)
RuntimeError: shape ‘[-1,1,32,200]’ is invalid for input of size 3200

I have total 10000 data for testing. During the model training, everything was okay and my model has been trained as well. But now It shoes some error in the testing data.

My training data shape was [-1,1,32,200] as well.

2

It seems that 3200 input does not fit the shape [-1, 1, 32, 200].

3

My test data size is 10000. I am trying to figure it out where this input size 3200 coming from, my batch loader size is 32, and I passed [-1, 1, 32, 200] data each iteration for testing part with trained CNN model.
Also I am confused during the training time it was okay but the problem started from testing part.