Visitar URL original
Improvements to the training UX by n-poulsen · Pull Request #2775 · DeepLabCut/DeepLabCut · GitHub
Skip to content
Merged
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
typo fixes
  • Loading branch information
Niels Poulsen
Niels Poulsen committed Nov 8, 2024
commit 7ec0d170090ad65da5830b94f35e7f27446e0434
4 changes: 2 additions & 2 deletions docs/pytorch/pytorch_config.md
Original file line number Diff line number Diff line change
Expand Up @@ -399,7 +399,7 @@ targets and heatmaps for that image.
### Restarting Training at a Specific Checkpoint

If you wish to restart the training at a specific checkpoint, you can specify the
full path of the checkpoint to the `init_weights` variable, as shown below. In this
full path of the checkpoint to the `resume_training_from` variable, as shown below. In this
example, `snapshot-010.pt` will be loaded before training starts, and the model will
continue to train from the 10th epoch on.

Expand Down Expand Up @@ -468,7 +468,7 @@ detector that brings enough performance. The recommended variants are the follow
### Restarting Training of an Object Detector at a Specific Checkpoint

If you wish to restart the training of a detector at a specific checkpoint, you can
specify the full path of the checkpoint to the detector's `init_weights` variable, as
specify the full path of the checkpoint to the detector's `resume_training_from` variable, as
shown below. In this example, `snapshot-detector-020.pt` will be loaded before training
starts, and the model will continue to train from the 20th epoch on.

Expand Down