File: plot_cv_indices.py

package info (click to toggle)
scikit-learn 0.23.2-5
  • links: PTS, VCS
  • area: main
  • in suites: bookworm, bullseye, sid
  • size: 21,892 kB
  • sloc: python: 132,020; cpp: 5,765; javascript: 2,201; ansic: 831; makefile: 213; sh: 44
file content (150 lines) | stat: -rw-r--r-- 5,678 bytes parent folder | download
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
"""
Visualizing cross-validation behavior in scikit-learn
=====================================================

Choosing the right cross-validation object is a crucial part of fitting a
model properly. There are many ways to split data into training and test
sets in order to avoid model overfitting, to standardize the number of
groups in test sets, etc.

This example visualizes the behavior of several common scikit-learn objects
for comparison.
"""

from sklearn.model_selection import (TimeSeriesSplit, KFold, ShuffleSplit,
                                     StratifiedKFold, GroupShuffleSplit,
                                     GroupKFold, StratifiedShuffleSplit)
import numpy as np
import matplotlib.pyplot as plt
from matplotlib.patches import Patch
np.random.seed(1338)
cmap_data = plt.cm.Paired
cmap_cv = plt.cm.coolwarm
n_splits = 4

###############################################################################
# Visualize our data
# ------------------
#
# First, we must understand the structure of our data. It has 100 randomly
# generated input datapoints, 3 classes split unevenly across datapoints,
# and 10 "groups" split evenly across datapoints.
#
# As we'll see, some cross-validation objects do specific things with
# labeled data, others behave differently with grouped data, and others
# do not use this information.
#
# To begin, we'll visualize our data.

# Generate the class/group data
n_points = 100
X = np.random.randn(100, 10)

percentiles_classes = [.1, .3, .6]
y = np.hstack([[ii] * int(100 * perc)
               for ii, perc in enumerate(percentiles_classes)])

# Evenly spaced groups repeated once
groups = np.hstack([[ii] * 10 for ii in range(10)])


def visualize_groups(classes, groups, name):
    # Visualize dataset groups
    fig, ax = plt.subplots()
    ax.scatter(range(len(groups)),  [.5] * len(groups), c=groups, marker='_',
               lw=50, cmap=cmap_data)
    ax.scatter(range(len(groups)),  [3.5] * len(groups), c=classes, marker='_',
               lw=50, cmap=cmap_data)
    ax.set(ylim=[-1, 5], yticks=[.5, 3.5],
           yticklabels=['Data\ngroup', 'Data\nclass'], xlabel="Sample index")


visualize_groups(y, groups, 'no groups')

###############################################################################
# Define a function to visualize cross-validation behavior
# --------------------------------------------------------
#
# We'll define a function that lets us visualize the behavior of each
# cross-validation object. We'll perform 4 splits of the data. On each
# split, we'll visualize the indices chosen for the training set
# (in blue) and the test set (in red).


def plot_cv_indices(cv, X, y, group, ax, n_splits, lw=10):
    """Create a sample plot for indices of a cross-validation object."""

    # Generate the training/testing visualizations for each CV split
    for ii, (tr, tt) in enumerate(cv.split(X=X, y=y, groups=group)):
        # Fill in indices with the training/test groups
        indices = np.array([np.nan] * len(X))
        indices[tt] = 1
        indices[tr] = 0

        # Visualize the results
        ax.scatter(range(len(indices)), [ii + .5] * len(indices),
                   c=indices, marker='_', lw=lw, cmap=cmap_cv,
                   vmin=-.2, vmax=1.2)

    # Plot the data classes and groups at the end
    ax.scatter(range(len(X)), [ii + 1.5] * len(X),
               c=y, marker='_', lw=lw, cmap=cmap_data)

    ax.scatter(range(len(X)), [ii + 2.5] * len(X),
               c=group, marker='_', lw=lw, cmap=cmap_data)

    # Formatting
    yticklabels = list(range(n_splits)) + ['class', 'group']
    ax.set(yticks=np.arange(n_splits+2) + .5, yticklabels=yticklabels,
           xlabel='Sample index', ylabel="CV iteration",
           ylim=[n_splits+2.2, -.2], xlim=[0, 100])
    ax.set_title('{}'.format(type(cv).__name__), fontsize=15)
    return ax


###############################################################################
# Let's see how it looks for the :class:`~sklearn.model_selection.KFold`
# cross-validation object:

fig, ax = plt.subplots()
cv = KFold(n_splits)
plot_cv_indices(cv, X, y, groups, ax, n_splits)

###############################################################################
# As you can see, by default the KFold cross-validation iterator does not
# take either datapoint class or group into consideration. We can change this
# by using the ``StratifiedKFold`` like so.

fig, ax = plt.subplots()
cv = StratifiedKFold(n_splits)
plot_cv_indices(cv, X, y, groups, ax, n_splits)

###############################################################################
# In this case, the cross-validation retained the same ratio of classes across
# each CV split. Next we'll visualize this behavior for a number of CV
# iterators.
#
# Visualize cross-validation indices for many CV objects
# ------------------------------------------------------
#
# Let's visually compare the cross validation behavior for many
# scikit-learn cross-validation objects. Below we will loop through several
# common cross-validation objects, visualizing the behavior of each.
#
# Note how some use the group/class information while others do not.

cvs = [KFold, GroupKFold, ShuffleSplit, StratifiedKFold,
       GroupShuffleSplit, StratifiedShuffleSplit, TimeSeriesSplit]


for cv in cvs:
    this_cv = cv(n_splits=n_splits)
    fig, ax = plt.subplots(figsize=(6, 3))
    plot_cv_indices(this_cv, X, y, groups, ax, n_splits)

    ax.legend([Patch(color=cmap_cv(.8)), Patch(color=cmap_cv(.02))],
              ['Testing set', 'Training set'], loc=(1.02, .8))
    # Make the legend fit
    plt.tight_layout()
    fig.subplots_adjust(right=.7)
plt.show()