How to evaluate NLP model performance - Githubissues

TUM-IDP-WS-20 / doc

0 stars 0 forks source link

How to evaluate NLP model performance #25

Open farukcankaya opened 3 years ago

farukcankaya commented 3 years ago

We measure top-1, top-10, top-20, and top-100 accuracy based on our initial dataset. Check for measurement results: https://docs.google.com/spreadsheets/d/1scM6RqGp_qVxDsGLxMZr-M-3TdYnD19_2Ij4uN0f7XQ/edit#gid=1784565335

farukcankaya commented 3 years ago

Make complete research based on the 'performance' section of the https://en.wikipedia.org/wiki/Recommender_system