ndlessrain

250x250
  • 분류 전체보기 (910)
    • Ndlessrain (43)
    • Programmer's (29)
    • WebLogic (7)
    • Algorithm (25)
    • OS (43)
    • PowerBuilder (21)
    • WEB (103)
    • UNIX (81)
    • JAVA (83)
    • DB (122)
    • UTIL (79)
    • Flex (35)
    • Android (42)
    • Network (12)
    • Python (137)
    • Perl (1)
    • C# (20)
    • SE (10)
    • BMW (2)
    • React (3)
  • 방명록
  • 관리
  • 글쓰기
/ /

meta llama 파인 튜닝(fine tuning)

2023. 12. 6. 15:31
728x90

https://github.com/OpenPipe/OpenPipe/tree/main/examples/classify-recipes

따라해보려나...

메타 : https://ai.meta.com/llama/get-started/

 

Getting started with Llama 2 - AI at Meta

import torch model_id = "meta-llama/Llama-2-7b-hf" new_model = "output/llama-2-guanaco-7b/checkpoint-1875/adapter_model" # change if needed quantization_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_use_dou

ai.meta.com

728x90
저작자표시 (새창열림)

'Python' 카테고리의 다른 글

DNN 이진분류 시작  (1) 2024.01.02
meta llama 로컬 세팅  (0) 2024.01.02
OPENAI GPT3 ADA를 활용 이진분류 예제  (0) 2023.12.06
openai fine_tunes stream interrupted(client disconnected)  (0) 2023.12.06
The positive class specified `baseball` is not found among the classes in the validation file [' hockey', ' baseball']. Fine-tune failed. For help, please contact OpenAI and include your fine-tune ID  (0) 2023.12.06

+ Recent posts

Powered by Tistory, Designed by wallel
Rss Feed and Twitter, Facebook, Youtube, Google+

티스토리툴바