Using the same dataset, we can now test the performance of the ICA. However, in this case, as explained, we need to zero-center and whiten the dataset, but fortunately these preprocessing steps are done by the Scikit-Learn implementation (if the parameter whiten=True is omitted).
To perform the ICA on the MNIST dataset, we're going to instantiate the FastICA class, passing the arguments n_components=64 and the maximum number of iterations max_iter=5000. It's also possible to specify which function will be used to approximate the negentropy; however, the default is log cosh(x), which is normally a good choice:
from sklearn.decomposition import FastICAfastica = FastICA(n_components=64, max_iter=5000, ...
