Polynomial fitting in Python

Question

i am trying to do a polynomial fit to my data and the output is a linear fit. i am trying to understand where my mistake is.

regr2 = PolynomialFeatures(degree=2)
regr2.fit_transform(diabetes_X_train, diabetes_y_train)
regr2 = PolynomialFeatures(interaction_only=True)
regr2.fit_transform(diabetes_X_train, diabetes_y_train)

regr = LinearRegression()
regr.fit(diabetes_X_train, diabetes_y_train)

diabetes_y_pred = regr.predict(diabetes_X_test)

plt.scatter(diabetes_X_test, diabetes_y_test,  color='blue')
plt.plot(diabetes_X_test, diabetes_y_pred, color='red', linewidth=1)

angrymantis · Accepted Answer · 2020-11-21 02:03:10Z

1

It's doing it because you told it to.

regr = LinearRegression()

You are plotting "diabetes_y_pred," which comes from "regr.fit" and regr is defined above with "LinearRegression()."

https://scikit-learn.org/stable/modules/generated/sklearn.linear_model.LinearRegression.html

answered Nov 21, 2020 at 2:03

angrymantis

3821 silver badge9 bronze badges

Sign up to request clarification or add additional context in comments.

4 Comments

satya Over a year ago

Hi,Thanks for the help. How do i fix it?

angrymantis Over a year ago

Don't use scikit-learn much. My guess is change diabetes_y_pred = regr.predict(diabetes_X_test) to diabetes_y_pred = regr2.predict(diabetes_X_test).

satya Over a year ago

AttributeError: 'PolynomialFeatures' object has no attribute 'predict' getting this error when i run that.

tandem Over a year ago

stackoverflow.com/questions/56201362/…

Collectives™ on Stack Overflow

Polynomial fitting in Python

1 Answer 1

4 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

4 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related