|
Download lrm/lrm_sana/docs/migration_notes.md from aryadomain/all_code_base: direct link, hf CLI and curl.
- Browser
- Download file 1.87 kB
-
https://huggingface.co/aryadomain/all_code_base/resolve/main/lrm/lrm_sana/docs/migration_notes.md
- Command line
-
hf download hf://aryadomain/all_code_base/lrm/lrm_sana/docs/migration_notes.md
-
curl -L -o migration_notes.md https://huggingface.co/aryadomain/all_code_base/resolve/main/lrm/lrm_sana/docs/migration_notes.md
1.87 kB
| # Migration Notes: Flux -> SANA LRM | |
| ## Reused Components | |
| - Training loop orchestration from `flux/trainer/scripts/train.py`. | |
| - Accelerator stack (`base`, `debug`, `deepspeed`). | |
| - Optimizer and LR scheduler modules. | |
| - Pairwise loss structure and evaluation flow. | |
| - Dataset filtering logic including pseudo-preference support. | |
| ## SANA-Specific Adaptations | |
| - Replace Flux model wrapper with SANA latent-transformer wrapper. | |
| - Replace Flux tokenizer/text path with variant-aware SANA path. | |
| - Keep pairwise reward objective unchanged; adapt only feature extraction. | |
| - Add model profile support for four requested checkpoints. | |
| ## Naming Changes | |
| - `step_flux` -> `step_sana` | |
| - `step_flux_base` -> `step_sana_base` | |
| - `step_clip_flux` -> `step_clip_sana` | |
| - `FluxPreferenceModel` -> `SanaPreferenceModel` | |
| ## Config Baseline Choice | |
| - Dataset source follows flux working setting: | |
| - `pickapic-anonymous/pickapic_v1` | |
| - Pseudo preference CSV path follows flux launcher fallback behavior. | |
| - Logging/output path remains flux-style to simplify operational parity. | |
| ## Runtime Notes | |
| - Quick profile should force minimal run: | |
| - test split for train/valid/test | |
| - image size downscaled | |
| - `max_steps=1` | |
| - Main profile should use production defaults with DeepSpeed sharded mode. | |
| ## Known Risks During Migration | |
| - Mismatch between tokenizer outputs and model forward signature. | |
| - Caption embedding/channel shape mismatch in SANA transformer conditioning. | |
| - Missing subfolders for some checkpoints (tokenizer_2 or text_encoder_2). | |
| - Diffusers/transformers version compatibility for SANA classes. | |
| ## Mitigation Strategy | |
| - Use profile metadata for each checkpoint (explicit, not implicit only). | |
| - Fail fast with clear validation errors at model init time. | |
| - Keep quick-run smoke checks as gate before longer runs. | |
| - Preserve fallback behavior from flux launcher for data and cache paths. | |