Gouryella ccdf6c576f Improve training performance with torch.compile and torch.amp ... - Update AMP usage from `torch.cuda.amp` to `torch.amp` for broader device compatibility. - Add `torch.compile` to models in various training scripts (SFT, LoRA, DPO, Distillation, Pretrain, Distill Reason) to leverage potential performance gains. - Implement a custom `load_state_dict` in `MiniMindForCausalLM` to handle state dict keys potentially modified by `torch.compile`.		2025-05-21 12:16:02 +08:00
..
train_distill_reason.py	Improve training performance with torch.compile and torch.amp	2025-05-21 12:16:02 +08:00
train_distillation.py	Improve training performance with torch.compile and torch.amp	2025-05-21 12:16:02 +08:00
train_dpo.py	Improve training performance with torch.compile and torch.amp	2025-05-21 12:16:02 +08:00
train_full_sft.py	Improve training performance with torch.compile and torch.amp	2025-05-21 12:16:02 +08:00
train_lora.py	Improve training performance with torch.compile and torch.amp	2025-05-21 12:16:02 +08:00
train_pretrain.py	Improve training performance with torch.compile and torch.amp	2025-05-21 12:16:02 +08:00