Pin handoff to ef477d68d (review fixes)
Browse files- HANDOFF.md +3 -3
HANDOFF.md
CHANGED
|
@@ -30,7 +30,7 @@ FastH3 Omni is a reference-conditioned video+audio model that runs through FastV
|
|
| 30 |
```bash
|
| 31 |
mkdir -p ~/fasth3-mlx && cd ~/fasth3-mlx
|
| 32 |
git clone https://github.com/aryan5v/FastVideo.git && cd FastVideo
|
| 33 |
-
git checkout
|
| 34 |
uv venv --python 3.12 --seed && source .venv/bin/activate
|
| 35 |
uv pip install -e ".[mlx]"
|
| 36 |
python -c "import mlx.core as mx; print(mx.__version__, mx.device_info())" # record this output
|
|
@@ -109,7 +109,7 @@ Make the dataset public, or add the requester as a collaborator if the operator
|
|
| 109 |
gh pr create --repo aryan5v/FastVideo --base aryan/fasth3-omni-mlx --head <you>:fasth3-omni-mlx-m5-fixes
|
| 110 |
```
|
| 111 |
Give it the title `[bugfix]: fixes found running FastH3 Omni MLX on M5` and list each fix in the body.
|
| 112 |
-
- If GitHub is not available, run `git format-patch
|
| 113 |
|
| 114 |
**c. Results comment on the upstream draft PR** [hao-ai-lab/FastVideo#2000](https://github.com/hao-ai-lab/FastVideo/pull/2000):
|
| 115 |
```bash
|
|
@@ -117,7 +117,7 @@ gh pr comment 2000 --repo hao-ai-lab/FastVideo --body-file ~/fasth3-mlx/m5_repor
|
|
| 117 |
```
|
| 118 |
Write `m5_report.md` with exactly these parts:
|
| 119 |
1. **Hardware:** chip, GPU cores, unified memory (`sysctl -n machdep.cpu.brand_string hw.memsize`), macOS (`sw_vers`), mlx version and `mx.device_info()`.
|
| 120 |
-
2. **Code:** commit `
|
| 121 |
3. **Results table:** rows are the compositions; columns are mean ± SD end-to-end seconds over seeds 0/1/2, then mean denoise s, decode s, conditioning s, and peak memory GiB. Use the generated `results-480p-36l/README.md` and `per-request-records.jsonl`.
|
| 122 |
4. **Link** to the HF dataset with the videos.
|
| 123 |
5. **Issues:** anything that failed, was skipped (for example `--seeds 0`) or was changed. Also note the `prepare_model_dir.sh` verification summary, including any hash warnings.
|
|
|
|
| 30 |
```bash
|
| 31 |
mkdir -p ~/fasth3-mlx && cd ~/fasth3-mlx
|
| 32 |
git clone https://github.com/aryan5v/FastVideo.git && cd FastVideo
|
| 33 |
+
git checkout ef477d68d # branch aryan/fasth3-omni-mlx (hao-ai-lab/FastVideo#2000)
|
| 34 |
uv venv --python 3.12 --seed && source .venv/bin/activate
|
| 35 |
uv pip install -e ".[mlx]"
|
| 36 |
python -c "import mlx.core as mx; print(mx.__version__, mx.device_info())" # record this output
|
|
|
|
| 109 |
gh pr create --repo aryan5v/FastVideo --base aryan/fasth3-omni-mlx --head <you>:fasth3-omni-mlx-m5-fixes
|
| 110 |
```
|
| 111 |
Give it the title `[bugfix]: fixes found running FastH3 Omni MLX on M5` and list each fix in the body.
|
| 112 |
+
- If GitHub is not available, run `git format-patch ef477d68d -o ~/fasth3-mlx/results-480p-36l/fixes/` before uploading in step 7a.
|
| 113 |
|
| 114 |
**c. Results comment on the upstream draft PR** [hao-ai-lab/FastVideo#2000](https://github.com/hao-ai-lab/FastVideo/pull/2000):
|
| 115 |
```bash
|
|
|
|
| 117 |
```
|
| 118 |
Write `m5_report.md` with exactly these parts:
|
| 119 |
1. **Hardware:** chip, GPU cores, unified memory (`sysctl -n machdep.cpu.brand_string hw.memsize`), macOS (`sw_vers`), mlx version and `mx.device_info()`.
|
| 120 |
+
2. **Code:** commit `ef477d68d`, plus any fixes (link the fix PR or patch).
|
| 121 |
3. **Results table:** rows are the compositions; columns are mean ± SD end-to-end seconds over seeds 0/1/2, then mean denoise s, decode s, conditioning s, and peak memory GiB. Use the generated `results-480p-36l/README.md` and `per-request-records.jsonl`.
|
| 122 |
4. **Link** to the HF dataset with the videos.
|
| 123 |
5. **Issues:** anything that failed, was skipped (for example `--seeds 0`) or was changed. Also note the `prepare_model_dir.sh` verification summary, including any hash warnings.
|