jku-vds-lab / paradime

A framework for parametric dimensionality reduction
MIT License
9 stars 3 forks source link

Add method to run training phase #12

Closed einbandi closed 2 years ago

einbandi commented 2 years ago

Should be relatively straightforward, if all else is properly set up. Just create the loader for each phase, iterate through its batches, compute batchwise losses and call the optimizer. Don't forget to write placeholders or TODOs, wherever a loss report is necessary (as this is not yet implemented; see #3).