How I downsample

downsample.py

import numpy as np

def downsample(data, labels):
    """
    >>> data = np.arange(100)
    >>> label = np.array([1] * 95 + [0] * 5)
    >>> print downsample(data, label)
    """
    zero_index = np.array([i for i, val in enumerate(labels) if val == 0])
    one_index = np.array([i for i, val in enumerate(labels) if val == 1])
    smaller, larger = sorted([zero_index, one_index], key=len) 
    selected = np.sort(np.concatenate((smaller, np.random.choice(larger, size=len(smaller), replace=False))))
    return data[selected], labels[selected]