تاریخچه Commit ها

نویسنده SHA1 پیام تاریخ
  heiway d76b9174d1 fix a dtype mismatch when use mps (#790) 1 سال پیش
  Whale and Dolphin 40665e1a39 [fix]fix problems to let version 1.5 support sft (#774) 1 سال پیش
  spicysama b951de3b72 v1.5 (#696) 1 سال پیش
  PoTaTo 834b07257c Agent inference (#650) 1 سال پیش
  spicysama e37a445f51 Fix backend (#627) 1 سال پیش
  bfs18 dfc95d0f5c Add preprocessing and validation steps when loading the checkpoint. (#433) 1 سال پیش
  spicysama ea53678446 Quantization Support (#316) 1 سال پیش
  Lengyue c9209e94d9 support finetuning 1 سال پیش
  spicysama 584be1d047 Adaptation to version 1.2 (#301) 1 سال پیش
  Leng Yue 5e7914472f This PR brings V1.2 inference into main (#300) 1 سال پیش
  Lengyue 26af12b8c6 Make AR and naive decoder configurable 2 سال پیش
  Lengyue 3d002fd18b Optimize amp training 2 سال پیش
  Lengyue 87ca8cf62e Rollback decoding 2 سال پیش
  Lengyue 6e67ebda73 Optimize training speed by skip unused decode 2 سال پیش
  Lengyue 0a1986bb14 Fix data leaking 2 سال پیش
  Lengyue 362a1a7116 Update dual ar generate & config 2 سال پیش
  Lengyue 9e1a9debfd fix bug & implement generate 2 سال پیش
  Lengyue 7a31b4043a Fix target rotate 2 سال پیش
  Lengyue 0855b454e4 Add VITS vq 2 سال پیش
  Lengyue cecee062c4 Remove flash attn deps as PyTorch ships in 2.2.0 2 سال پیش
  Lengyue 415c78a7d4 Disable neft on eval 2 سال پیش
  Lengyue f5a2df2d23 Add 4 in 8 out support 2 سال پیش
  Lengyue 8093258065 Optimize lora & add auto dpo training 2 سال پیش
  Lengyue 7ac4d4b918 Add neft and save lora only 2 سال پیش
  Lengyue 39f6902119 Add dropout options to optimize overfitting 2 سال پیش
  Lengyue 2919eafd19 Disable flash attention 2 by default 2 سال پیش
  Lengyue dcc5e80ce2 Add flash attention & gradient checkpointing 2 سال پیش
  spicysama 06a35aef53 add fastapi for inference (#12) 2 سال پیش
  Lengyue c583555995 Add half precision inference and document 2 سال پیش
  Lengyue 9649873532 Update generate examples 2 سال پیش