Show Sidebar Hide Sidebar

# Underfitting vs. Overfitting in Scikit-learn

This example demonstrates the problems of underfitting and overfitting and how we can use linear regression with polynomial features to approximate nonlinear functions. The plot shows the function that we want to approximate, which is a part of the cosine function. In addition, the samples from the real function and the approximations of different models are displayed. The models have polynomial features of different degrees. We can see that a linear function (polynomial with degree 1) is not sufficient to fit the training samples. This is called underfitting. A polynomial of degree 4 approximates the true function almost perfectly. However, for higher degrees the model will overfit the training data, i.e. it learns the noise of the training data. We evaluate quantitatively overfitting / underfitting by using cross-validation. We calculate the mean squared error (MSE) on the validation set, the higher, the less likely the model generalizes correctly from the training data.

#### New to Plotly?¶

You can set up Plotly to work in online or offline mode, or in jupyter notebooks.
We also have a quick-reference cheatsheet (new!) to help you get started!

### Version¶

In [1]:
import sklearn
sklearn.__version__

Out[1]:
'0.18.1'

### Imports¶

In [2]:
print(__doc__)

import plotly.plotly as py
import plotly.graph_objs as go
from plotly import tools

import numpy as np
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import PolynomialFeatures
from sklearn.linear_model import LinearRegression
from sklearn.model_selection import cross_val_score

Automatically created module for IPython interactive environment


### Calculations¶

In [3]:
np.random.seed(0)

n_samples = 30
degrees = [1, 4, 15]

true_fun = lambda X: np.cos(1.5 * np.pi * X)
X = np.sort(np.random.rand(n_samples))
y = true_fun(X) + np.random.randn(n_samples) * 0.1
titles = []
data = []


### Plot Results¶

In [4]:
for i in range(len(degrees)):
polynomial_features = PolynomialFeatures(degree=degrees[i],
include_bias=False)
linear_regression = LinearRegression()
pipeline = Pipeline([("polynomial_features", polynomial_features),
("linear_regression", linear_regression)])
pipeline.fit(X[:, np.newaxis], y)

# Evaluate the models using crossvalidation
scores = cross_val_score(pipeline, X[:, np.newaxis], y,
scoring="neg_mean_squared_error", cv=10)

X_test = np.linspace(0, 1, 100)

if(i==1):
leg=True
else:
leg=False
data.append([])
trace1 = go.Scatter(x=X_test, y=pipeline.predict(X_test[:, np.newaxis]),
name="Model", mode='lines',
line=dict(color='blue', width=1),
showlegend=leg)
data[i].append(trace1)
trace2 = go.Scatter(x=X_test, y=true_fun(X_test),
name="True function", mode='lines',
line=dict(color='green', width=1),
showlegend=leg)
data[i].append(trace2)

trace3 = go.Scatter(x=X, y=y,
name="Samples", mode='markers',
marker=dict(color='blue',
line=dict(color='black', width=1)),
showlegend=leg)
data[i].append(trace3)

titles.append("Degree {}<br>MSE = {:.2e}(+/- {:.2e})".format(
degrees[i], -scores.mean(), scores.std()))

In [5]:
fig = tools.make_subplots(rows=1, cols=3,
subplot_titles=tuple(titles[:3]),
print_grid=False)

for i in range(0, len(data)):
for j in range(0, len(data[i])):
fig.append_trace(data[i][j], 1, i+1)

for i in map(str,range(1, 4)):
y = 'yaxis'+i
x = 'xaxis'+i
fig['layout'][y].update(title='y', showgrid=False,
showticklabels=False, ticks='')
fig['layout'][x].update(title='x', showgrid=False,
showticklabels=False, ticks='')

In [6]:
py.iplot(fig)

Out[6]:
Still need help?