sklearn metrics for multiclass classification

Question

I have performed GaussianNB classification using a sklearn. I tried to calculate the metrics using the following code:

print accuracy_score(y_test, y_pred)

print precision_score(y_test, y_pred)

Accuracy score is working correctly but the precision score calculation is showing error as:

ValueError: Target is multiclass but average='binary'. Please choose another average setting.

As the target is multiclass, can I have the metric scores of precision, recall, etc.?

Answer 1

In your code, notice that the 

precision_score(y_test, y_pred) 

is equivalent to 

precision_score(y_test, y_pred, pos_label=1, average='binary'). 

'binary':

There report results for the class are specified by pos_label. This is applicable only if targets (y_{true,pred}) are binary.

Here labels are not binary, but probably one-hot encoded. There are other options which should work with your data:

For example:

precision_score(y_test, y_pred, average=None) 

will return the precision scores for each class, while

precision_score(y_test, y_pred, average='micro') 

will return the total ratio of tp/(tp + fp)

The pos_label argument will be ignored if you choose another average option than binary.