New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

add score function for multinomial #97

Closed
jasmainak opened this Issue May 19, 2016 · 3 comments

Comments

Projects
None yet
4 participants
@jasmainak
Member

jasmainak commented May 19, 2016

'multinomial' is typically evaluated using a score like accuracy or F1: deviance and pseudo_R2 are not common.

cc @pavanramkumar

@jasmainak jasmainak added this to the version 0.2 milestone May 19, 2016

@daniel-acuna

This comment has been minimized.

Show comment
Hide comment
@daniel-acuna

daniel-acuna May 19, 2016

Collaborator

There is no clear way of measuring the performance of a multiclass classifier. I like log likelihood ratio between a null model and full model but I agree it is not standard.

Collaborator

daniel-acuna commented May 19, 2016

There is no clear way of measuring the performance of a multiclass classifier. I like log likelihood ratio between a null model and full model but I agree it is not standard.

@jasenjackson

This comment has been minimized.

Show comment
Hide comment
@jasenjackson

jasenjackson May 25, 2016

Could we measure performance with the sci-kit log_loss function?

http://scikit-learn.org/stable/modules/model_evaluation.html#log-loss

I was looking through their documentation, and it says that logistic regression loss is a common way of evaluating performance for multinomial classifiers and neural networks.

Could we measure performance with the sci-kit log_loss function?

http://scikit-learn.org/stable/modules/model_evaluation.html#log-loss

I was looking through their documentation, and it says that logistic regression loss is a common way of evaluating performance for multinomial classifiers and neural networks.

@pavanramkumar pavanramkumar changed the title from implement accuracy or F1 to add score function for multinomial Jun 8, 2016

@pavanramkumar

This comment has been minimized.

Show comment
Hide comment
@pavanramkumar

pavanramkumar Aug 27, 2017

Collaborator

closing this since we will track these issues while implementing a separate MultinomialGLM() class.

Collaborator

pavanramkumar commented Aug 27, 2017

closing this since we will track these issues while implementing a separate MultinomialGLM() class.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment