Sign up / in
Logo support
  • Product Features Releases What's New
  • Getting Started
  • Pricing Subscriptions Private Deployments Training Certifications BigML for Education
  • Support
  • Dashboard
  1. Support
  2. Predicting with your Models
  3. Evaluations

Which evaluation metric should I trust the most?

There is not just one metric more important than the rest. In any evaluation you should consider them all to make sure you evaluate your model properly. Nevertheless, accuracy (the number of correct predictions over the number of total instances that have been evaluated) is the most common metric used to measure the performance of classification models, however sometimes it is not the best metric. For instance, when we have a two class model but these classes are unbalanced ("true" class has 850 instances and "false" class has 150) you can get a high accuracy if you classify all your instances as the bigger class, so you would be predicting "true" most of the time. Since this is not correct, when evaluating your model you need to take into account other metrics such as Precision, Recall, F-measure or F-score, and Phi coefficient.
 
The evaluation metrics are not the same for classification and regression models. Please read this question to learn the evaluation metrics for regression models, and this blog post for further details.

Related articles

  • How can I interpret the evaluation results for my classification model?
  • What is the difference between lift and leverage?
  • How can I interpret the evaluation results for my regression model?
  • How can I evaluate my model?
  • Why can't I find the confusion matrix in the evaluation panel?

Any other questions?

We are happy to help! Please contact us at support@bigml.com and we'll get back to you shortly.

Live support

Feel free to ask us your doubts about BigML. Our team will instantly answer your requests in our live chat.

COMPANY

  • About
  • Brand Style Guide
  • Blog
  • Contact
  • Internships
  • Milestones
  • Openings
  • Team

PRODUCT

  • API
  • Documentation
  • Features
  • Labs
  • Organizations
  • Private Deployments
  • Releases
  • Tools
  • WhizzML

BUSINESS

  • Customers
  • Events
  • Partners
  • Pricing

TRAINING

  • Certifications
  • Education
  • ML Schools

GALLERY

  • Datasets
  • Models
  • WhizzML
Terms of Service Privacy Policy FAQ Copyright © 2019 BigML, Inc.