[FIX] Normalize data to zero mean and unit variance before dimension …

…estimation (#636)

* Normalize data to zero mean and unit variance in ma_pca

* Use normalized data as input to ma_pca

* Revert first change: do not re-scale input in ma_pca

* Use consistent normalization strategy in ma_pca

* Remove outdated comments

* Update regression test data; normalization produces fewer figures

* Normalize along spatial dimension in mapca; Matches GIFT

* Update test output; updated mapca finds 3 more components

tedana/decomposition/ma_pca.py

@@ -478,9 +478,7 @@ def ma_pca(data_nib, mask_nib, criteria='mdl'):
     maskvec = np.reshape(mask_nib, Nx * Ny * Nz, order='F')
     data_non_normalized = data_nib_V[maskvec == 1, :]
     scaler = StandardScaler(with_mean=True, with_std=True)
-    # TODO: determine if tedana is already normalizing before this
     data = scaler.fit_transform(data_non_normalized)  # This was X_sc
-    data = data_non_normalized
 
     LGR.info('Performing SVD on original OC data...')
     V, EigenValues = _icatb_svd(data, Nt)