5ea8eb958c
torchao 0.17 removed MXLinearConfig from prototype.mx_formats in favour of MXFP8TrainingOpConfig.from_recipe shared with MoE training. _mxfp8_training_config tries the 0.16 API first and falls back to the 0.17 one; both feed quantize_. mxfp8 still degrades to bf16 with a warning when neither import resolves |
||
|---|---|---|
| .. | ||
| __init__.py | ||
| diffusion_dit_trainer.py | ||
| diffusion_lora_trainer.py | ||
| diffusion_train_common.py | ||
| diffusion_training_service.py | ||
| resume.py | ||
| s3_dataset.py | ||
| trainer.py | ||
| training.py | ||
| worker.py | ||