rmarkello

# -*- coding: utf-8 -*-

import numpy as np
import matplotlib.patches as mpatch
import matplotlib.pyplot as plt
from scipy import stats
import seaborn as sns


def distbox_plot(data, groups=None, bw='scott', cut=3, gridsize=100,

rmarkello

# -*- coding: utf-8 -*-
"""
Implementation of the pre-processing code used by Smith et al., 2015, Nature
Neuroscience (available at https://www.fmrib.ox.ac.uk/datasets/HCP-CCA/).

Assumes you have access to both the unrestricted AND restricted behavioral data
from HCP. And can load it into a dataframe + re-order it according to
instructions at the above link (see `smith_cols.csv` for up-to-date column
names that work with the S1200 release of HCP.

rmarkello

# -*- coding: utf-8 -*-
"""
Python implementation of PermCCA (https://github.com/andersonwinkler/PermCCA),
licensed under GPLv3 (https://www.gnu.org/licenses/gpl-3.0.en.html) as a 
derivative of the original MATLAB code.

Since the Schur decomposition is non-unique there are numerical differences
betwen the MATLAB code and this implementation.

This is a rough translation so there are likely a few bugs, but it generates