Getting an infinite p value while selecting best feature using SelectKbest model

machine_learning
python

#1

Hi,
i am using SelectKbest model to select the best feature amongst other varaibles.
But for some of the features i am getting pvalues equals to ‘INF’ which means inifinity.
What does it states?


#2

Hi @Kush mostly in statistics when your p-value is extremely small then system just show it as ‘INF’, or say machine is not able to render that small value, it can be up-to 10**-29 also, so just take p value as the lowest and state your hypothesis. in this case my guess is,
Low P values: your data are unlikely with a true null.
A low P value suggests that your sample provides enough evidence that you can reject the null hypothesis for the entire population. and Null hypothesis would be your variables are not important.
Now you can reject this hypothesis, and can state this variables are explaining some variance and are important for your model.