storytold/FineTrainers-Conditioning

A change to existing behaviour in (repo-wide), finetrainers, training — 1 commit, 35 files: “LTX Video (\#123)” (refs \#123).

training/cogvideox · low confidence · unverified

A change to existing behaviour in finetrainers — 3 commits, 15 files: “reposition private methods (\#213)”; “Rename lora files (\#214)”; “Full Finetuning for LTX possibily extended to other models. (\#192)” (refs \#213, \#214, \#192). Also in this area: 2 commits (1 fix) modifying finetrainers; 1 commit (0 fixes) modifying finetrainers; 1 commit (1 fix) fixing finetrainers; 7 commits (6 fixes) fixing finetrainers.

finetrainers, training/mochi-1 · low confidence · unverified

A change to existing behaviour in training — 3 commits (1 fix), 7 files: “Hunyuan Video LoRA (\#126)”; “Update README.md to fix graph paths (\#171)”; “LTX Video (\#123)” (refs \#126, \#171, \#123).

training · medium confidence · unverified

A change to existing behaviour in (repo-wide) — 42 commits (2 fixs), 4 files: “Update README.md”; “docs: update README.md (\#21)”; “Update README.md (\#59)”; and 37 more (refs \#21, \#59, \#20, \#163, \#126, …).

(repo-wide) · low confidence · unverified

A change to existing behaviour in assets — 6 commits, 16 files: “.github page”; “Update README and Contribution guide (\#20)”; “CogVideoX LoRA and full finetuning (\#1)”; and 3 more (refs \#20, \#1, \#3, \#14, \#43).

assets · medium confidence · unverified

Test coverage in tests — 1 commit, 5 files: “Shell script to minimally test supported models on a real dataset (\#204)” (refs \#204).

tests · high confidence · unverified

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

• First survey — no prior run to compare against. CAI 45.

• Code Health 66
• Architecture 98
• Maturity 29
• Readiness 41
• Security 91

• 151 commits — 122 feature/other, 29 fixes

• (root) — 46 commits
• training/mochi-1 — 25 commits
• finetrainers/trainer.py — 18 commits
• (repo) — 16 commits
• training/dataset.py — 7 commits
• finetrainers/utils — 6 commits
• training/cogvideox_image_to_video_lora.py — 6 commits
• finetrainers/args.py — 4 commits
• docs/training — 3 commits
• .github/workflows — 2 commits
• accelerate_configs/deepspeed.yaml — 2 commits
• finetrainers/dataset.py — 2 commits
• finetrainers/hunyuan_video — 2 commits
• training/args.py — 2 commits
• training/cogvideox_text_to_video_sft.py — 2 commits
• docs/dataset — 1 commit
• finetrainers/cogvideox — 1 commit
• finetrainers/ltx_video — 1 commit
• tests/scripts — 1 commit
• training/README.md — 1 commit

• fix: Allow images; Remove LLM generated prefixes; Allow JSON/JSONL; Fix bugs (#158)
• fix: Fix minor bug with function call that doesn't exist. (#195)
• fix: Fix scheduler bugs (#177)
• fix: Fix the checkpoint dir bug in get_intermediate_ckpt_path so that checkpointing_limit can actually work. (#207)
• fix: Fixed optimizers parsing error in bash scripts (#61)
• fix: Merge pull request #100 from a-r-r-o-w/wandb-tracker-fix
• fix: Merge pull request #104 from a-r-r-o-w/fix-format-specifier
• fix: RoPE fixes for 1.5, bfloat16 support in prepare_dataset, gradient_accumulation grad norm undefined fix (#107)
• fix: Unbound fix (#105)
• fix: Update README.md to fix graph paths (#171)
• fix: [core] Fix loading of precomputed conditions and latents (#199)
• fix: add I2V sft and fix an error (#97)
• fix: fix
• fix: fix deadlock; make style
• fix: fix format specifier.
• fix: fix grad norm related logging
• fix: fix help message
• fix: fix validation bug (#149)
• fix: fix: correct type in .py files (#52)
• fix: fix: resuming from a checkpoint when using deepspeed. (#38)
• …and 131 more

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

storytold/FineTrainers-Conditioning was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.