Moved code to lightning
Updating codebase to make use of pytorch-lightning
- Main train/valid loops are handled by lightning
- Removed device-handling code
- Each model implements its own train/valid/predict steps
- Updated config files for use with lightning
- Using lightning checkpointing and logging callbacks
- Updated DeviceMonitor to be compatible with lightning
Merge request reports
Activity
added 1 commit
- 51efda26 - Removed checkpointer tests as we are using the lightning callback
added 1 commit
- 2a7513e8 - Assign CUDA_VISIBLE_DEVICES if device specified
added 2 commits
added 1 commit
- 0a858361 - Updated accelerator selection during prediction
added 1 commit
- b3b1462b - Changed pytorch-lightning requirement to lightning
requested review from @ojimenez
mentioned in merge request !5 (merged)
added 37 commits
- 507a6207...a6ff407d - 27 earlier commits
- 37b723ca - completed documentation for tbx11k
- 78010d30 - added complete tbx11k dataset to project .toml
- 727dfa94 - added tests for all tbx11k configs
- 47647a7c - fixed formatting in tests
- 5f644c0a - fixed wrong paranthesis
- 9b297231 - removed unusable tbx11k custom split 3
- 2271683d - fixed small bugs in documentation
- f9a3e2d0 - more doc bug fixes
- 2577fb3c - more doc bug fixes
- f87e433e - Merge branch 'tbx11k' into 'move-to-lightning'
Toggle commit list
Please register or sign in to reply