ASUS-AICS / LibMultiLabel-Old-Archive

This library has been moved to https://github.com/ntumlgroup/LibMultiLabel
MIT License
153 stars 29 forks source link

Don't Calculate Loss in Prediction #363

Closed donglihe-hub closed 8 months ago

donglihe-hub commented 8 months ago

What does this PR do?

The original prediction function uses shared_step which calculates losses, though it later discards the loss. This procedure introduces complexity to algorithms using reduced label space. With the current prediction function, test labels has to be mapped to the reduced space before doing prediction, which is unnecessary and could be avoided.

Test CLI & API (bash tests/autotest.sh)

Test APIs used by main.py.

Check API Document

If any new APIs are added, please check if the description of the APIs is added to API document.

Test quickstart & API (bash tests/docs/test_changed_document.sh)

If any APIs in quickstarts or tutorials are modified, please run this test to check if the current examples can run correctly after the modified APIs are released.