On http://cs231n.github.io/neural-networks-2/ it is mentioned that for convolutional neural networks it is preferred to preprocess data using mean subtraction and normalization techniques.
I was just wondering how would it be best approached using Tensorflow.
X -= np.mean(X)
X /= np.std(X, axis = 0)
You're looking for
Linearly scales image to have zero mean and unit norm.
This op computes (x - mean) / adjusted_stddev, where mean is the average of all values in image, and adjusted_stddev = max(stddev, 1.0/sqrt(image.NumElements())).