# -*- coding: utf-8 -*-
"""
=============================
Normalizing your features
=============================

Often times its useful to normalize (z-score) you features before plotting, so
that they are on the same scale.  Otherwise, some features will be weighted more
heavily than others when doing PCA, and that may or may not be what you want.
The `normalize` kwarg can be passed to the plot function.  If `normalize` is
set to 'across', the zscore will be computed for the column across all of the
lists passed.  Conversely, if `normalize` is set to 'within', the z-score will
be computed separately for each column in each list.  Finally, if `normalize` is
set to 'row', each row of the matrix will be zscored.  Alternatively, you can use
the normalize function found in tools (see the third example).
"""

# Code source: Andrew Heusser
# License: MIT

# import
import hypertools as hyp
import numpy as np

# simulate data (seeded so the figures are reproducible)
np.random.seed(123)
cluster1 = np.random.multivariate_normal(np.zeros(3), np.eye(3), size=100)
cluster2 = np.random.multivariate_normal(np.zeros(3)+10, np.eye(3), size=100)
data = [cluster1, cluster2]

# plot normalized across lists
hyp.plot(data, '.', normalize='across', title='Normalized across datasets')

# plot normalized within list
hyp.plot(data, '.', normalize='within', title='Normalized within dataset')

# normalize by row
normalized_row = hyp.normalize(data, normalize='row')

# plot normalized by row
hyp.plot(normalized_row, '.', title='Normalized across row')