sklearn/doc/modules/learning_curve.rst

190 lines
8.2 KiB
ReStructuredText

.. _learning_curves:
=====================================================
Validation curves: plotting scores to evaluate models
=====================================================
.. currentmodule:: sklearn.model_selection
Every estimator has its advantages and drawbacks. Its generalization error
can be decomposed in terms of bias, variance and noise. The **bias** of an
estimator is its average error for different training sets. The **variance**
of an estimator indicates how sensitive it is to varying training sets. Noise
is a property of the data.
In the following plot, we see a function :math:`f(x) = \cos (\frac{3}{2} \pi x)`
and some noisy samples from that function. We use three different estimators
to fit the function: linear regression with polynomial features of degree 1,
4 and 15. We see that the first estimator can at best provide only a poor fit
to the samples and the true function because it is too simple (high bias),
the second estimator approximates it almost perfectly and the last estimator
approximates the training data perfectly but does not fit the true function
very well, i.e. it is very sensitive to varying training data (high variance).
.. figure:: ../auto_examples/model_selection/images/sphx_glr_plot_underfitting_overfitting_001.png
:target: ../auto_examples/model_selection/plot_underfitting_overfitting.html
:align: center
:scale: 50%
Bias and variance are inherent properties of estimators and we usually have to
select learning algorithms and hyperparameters so that both bias and variance
are as low as possible (see `Bias-variance dilemma
<https://en.wikipedia.org/wiki/Bias-variance_dilemma>`_). Another way to reduce
the variance of a model is to use more training data. However, you should only
collect more training data if the true function is too complex to be
approximated by an estimator with a lower variance.
In the simple one-dimensional problem that we have seen in the example it is
easy to see whether the estimator suffers from bias or variance. However, in
high-dimensional spaces, models can become very difficult to visualize. For
this reason, it is often helpful to use the tools described below.
.. rubric:: Examples
* :ref:`sphx_glr_auto_examples_model_selection_plot_underfitting_overfitting.py`
* :ref:`sphx_glr_auto_examples_model_selection_plot_validation_curve.py`
* :ref:`sphx_glr_auto_examples_model_selection_plot_learning_curve.py`
.. _validation_curve:
Validation curve
================
To validate a model we need a scoring function (see :ref:`model_evaluation`),
for example accuracy for classifiers. The proper way of choosing multiple
hyperparameters of an estimator is of course grid search or similar methods
(see :ref:`grid_search`) that select the hyperparameter with the maximum score
on a validation set or multiple validation sets. Note that if we optimize
the hyperparameters based on a validation score the validation score is biased
and not a good estimate of the generalization any longer. To get a proper
estimate of the generalization we have to compute the score on another test
set.
However, it is sometimes helpful to plot the influence of a single
hyperparameter on the training score and the validation score to find out
whether the estimator is overfitting or underfitting for some hyperparameter
values.
The function :func:`validation_curve` can help in this case::
>>> import numpy as np
>>> from sklearn.model_selection import validation_curve
>>> from sklearn.datasets import load_iris
>>> from sklearn.svm import SVC
>>> np.random.seed(0)
>>> X, y = load_iris(return_X_y=True)
>>> indices = np.arange(y.shape[0])
>>> np.random.shuffle(indices)
>>> X, y = X[indices], y[indices]
>>> train_scores, valid_scores = validation_curve(
... SVC(kernel="linear"), X, y, param_name="C", param_range=np.logspace(-7, 3, 3),
... )
>>> train_scores
array([[0.90..., 0.94..., 0.91..., 0.89..., 0.92...],
[0.9... , 0.92..., 0.93..., 0.92..., 0.93...],
[0.97..., 1... , 0.98..., 0.97..., 0.99...]])
>>> valid_scores
array([[0.9..., 0.9... , 0.9... , 0.96..., 0.9... ],
[0.9..., 0.83..., 0.96..., 0.96..., 0.93...],
[1.... , 0.93..., 1.... , 1.... , 0.9... ]])
If you intend to plot the validation curves only, the class
:class:`~sklearn.model_selection.ValidationCurveDisplay` is more direct than
using matplotlib manually on the results of a call to :func:`validation_curve`.
You can use the method
:meth:`~sklearn.model_selection.ValidationCurveDisplay.from_estimator` similarly
to :func:`validation_curve` to generate and plot the validation curve:
.. plot::
:context: close-figs
:align: center
from sklearn.datasets import load_iris
from sklearn.model_selection import ValidationCurveDisplay
from sklearn.svm import SVC
from sklearn.utils import shuffle
X, y = load_iris(return_X_y=True)
X, y = shuffle(X, y, random_state=0)
ValidationCurveDisplay.from_estimator(
SVC(kernel="linear"), X, y, param_name="C", param_range=np.logspace(-7, 3, 10)
)
If the training score and the validation score are both low, the estimator will
be underfitting. If the training score is high and the validation score is low,
the estimator is overfitting and otherwise it is working very well. A low
training score and a high validation score is usually not possible. Underfitting,
overfitting, and a working model are shown in the in the plot below where we vary
the parameter `gamma` of an SVM with an RBF kernel on the digits dataset.
.. figure:: ../auto_examples/model_selection/images/sphx_glr_plot_validation_curve_001.png
:target: ../auto_examples/model_selection/plot_validation_curve.html
:align: center
:scale: 50%
.. _learning_curve:
Learning curve
==============
A learning curve shows the validation and training score of an estimator
for varying numbers of training samples. It is a tool to find out how much
we benefit from adding more training data and whether the estimator suffers
more from a variance error or a bias error. Consider the following example
where we plot the learning curve of a naive Bayes classifier and an SVM.
For the naive Bayes, both the validation score and the training score
converge to a value that is quite low with increasing size of the training
set. Thus, we will probably not benefit much from more training data.
In contrast, for small amounts of data, the training score of the SVM is
much greater than the validation score. Adding more training samples will
most likely increase generalization.
.. figure:: ../auto_examples/model_selection/images/sphx_glr_plot_learning_curve_001.png
:target: ../auto_examples/model_selection/plot_learning_curve.html
:align: center
:scale: 50%
We can use the function :func:`learning_curve` to generate the values
that are required to plot such a learning curve (number of samples
that have been used, the average scores on the training sets and the
average scores on the validation sets)::
>>> from sklearn.model_selection import learning_curve
>>> from sklearn.svm import SVC
>>> train_sizes, train_scores, valid_scores = learning_curve(
... SVC(kernel='linear'), X, y, train_sizes=[50, 80, 110], cv=5)
>>> train_sizes
array([ 50, 80, 110])
>>> train_scores
array([[0.98..., 0.98 , 0.98..., 0.98..., 0.98...],
[0.98..., 1. , 0.98..., 0.98..., 0.98...],
[0.98..., 1. , 0.98..., 0.98..., 0.99...]])
>>> valid_scores
array([[1. , 0.93..., 1. , 1. , 0.96...],
[1. , 0.96..., 1. , 1. , 0.96...],
[1. , 0.96..., 1. , 1. , 0.96...]])
If you intend to plot the learning curves only, the class
:class:`~sklearn.model_selection.LearningCurveDisplay` will be easier to use.
You can use the method
:meth:`~sklearn.model_selection.LearningCurveDisplay.from_estimator` similarly
to :func:`learning_curve` to generate and plot the learning curve:
.. plot::
:context: close-figs
:align: center
from sklearn.datasets import load_iris
from sklearn.model_selection import LearningCurveDisplay
from sklearn.svm import SVC
from sklearn.utils import shuffle
X, y = load_iris(return_X_y=True)
X, y = shuffle(X, y, random_state=0)
LearningCurveDisplay.from_estimator(
SVC(kernel="linear"), X, y, train_sizes=[50, 80, 110], cv=5)