'Is it relevant to standardize data before PCA on MNIST digits dataset?

Since the features for the MNIST digits dataset (the pixels) are all expressed in the same units and are comprised between [0:255], is it relevant to standardize them (apply StandardScaler from scikitlearn) or a simple normalization would be sufficient (like minmaxscaler)?

python data-science

Sources

This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.

Source: Stack Overflow

Solution	Source

'Is it relevant to standardize data before PCA on MNIST digits dataset?

Sources

Related Questions