YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
cjtest-trained-outputs
Training and experiment artifacts from /data/alice/cjtest, organized by project.
Downloaded pretrained models, raw datasets, environment dependencies, cache files, and GitHub-original files were not uploaded. See manifests/delete_candidates_not_our_training.tsv and manifests/delete_non_training_downloaded.log for the filtered/deleted large files.
Layout
| Project | Files | Size | Documentation |
|---|---|---|---|
FinalTraj/Trajectory_Generation_multi_agent |
31 | 10.60 GiB | FinalTraj/Trajectory_Generation_multi_agent/README.md |
FinalTraj/finetune |
14 | 3.13 GiB | FinalTraj/finetune/README.md |
VideoX-Fun/Wan2.1-RoboTwin-LoRA |
44 | 10.42 GiB | VideoX-Fun/Wan2.1-RoboTwin-LoRA/README.md |
dreamzero/checkpoints |
2 | 5.81 GiB | dreamzero/checkpoints/README.md |
lara-wm/experiments |
49 | 21.09 GiB | lara-wm/experiments/README.md |
Restore
Use huggingface_hub.snapshot_download or hf download to fetch the needed project folder, then place it back under the same relative project path in a restored cjtest tree. Optimizer states are only needed for resuming training; adapter/model files are enough for inference or evaluation when paired with the corresponding base model.