How to Fix ValueError: Classification Metrics Can't Handle a Mix of Multiclass and Continuous-Multioutput Targets

1. Use 1d-array to Fix ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets in Python
2. Fix the ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets Error in Python

The ValueError is raised in Python when you provide a valid argument to a function, but it is an invalid value. For instance, you will get the ValueError when you input a negative number to the sqrt() function of a math module.

The error ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets occurs when you provide an invalid array in the sklearn.metrics.accuracy_score() function. Since the accuracy score is a classification metric, the ValueError can also be thrown when you use it with regression problems.

This tutorial will teach you to solve this error in Python.

Use 1d-array to Fix ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets in Python

Firstly, we will recreate this error in Python.

from sklearn.metrics import accuracy_score

y_pred = [[0.5, 1], [-1, 1], [7, -6]]
y_true = [[0, 2], [-1, 2], [8, -5]]
accuracy_score(y_true, y_pred)

Output:

ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets

The function accuracy_score() does not support the multiclass-multioutput format. When the input given in the function is not 1d-array, it displays the above error in the evaluation of the classification model.

You can solve it using the 1d-array in the accuracy_score() function.

Fix the ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets Error in Python

Another possible cause of the error might be that you are using the accuracy_score() function for the regression problems. Accuracy score is not a measure of regression models; it is only for classification models.

The regression metrics are R2 Score, MSE (Mean Squared Error), and RMSE (Root Mean Squared Error), which can be used to evaluate the performance of a regression model.

from sklearn.metrics import r2_score

y_pred = [[0.5, 1], [-1, 1], [7, -6]]
y_true = [[0, 2], [-1, 2], [8, -5]]
print(r2_score(y_true, y_pred))

Output:

0.9412391668996365

Now you know how to handle ValueError: Classification metrics can't handle a mix of multiclass and continuous-multioutput targets in Python. We hope you find these answers helpful.