Startsida Utforska Hjälp
Logga in
dongzenan
/
alpaca-lora
1
0
Fork 0
Filer Ärenden 0 Pull-förfrågningar 0 Wiki

Ingen beskrivning

Träd: 357ec81a17
Grenar Taggar
alpaca-lora
Eric Wang 357ec81a17 decapoda 3 år sedan
.gitignore 26f64780ad initial commit 3 år sedan
DATA_LICENSE 63121244c8 Licenses and whatnot 3 år sedan
LICENSE 63121244c8 Licenses and whatnot 3 år sedan
README.md 357ec81a17 decapoda 3 år sedan
alpaca_data.json 26f64780ad initial commit 3 år sedan
conversion.py 26f64780ad initial commit 3 år sedan
finetune.py 357ec81a17 decapoda 3 år sedan
generate.py 357ec81a17 decapoda 3 år sedan
iteration.ipynb 26f64780ad initial commit 3 år sedan
lengths.ipynb 26f64780ad initial commit 3 år sedan
loss.ipynb 357ec81a17 decapoda 3 år sedan

README.md

alpaca-lora (WIP)

This repository contains code for reproducing the Stanford Alpaca results. Users will need to be ready to fork transformers.

Setup

  1. Install dependencies (install zphang's transformers fork)

    pip install -q datasets accelerate loralib sentencepiece

pip install -q git+https://github.com/zphang/transformers@llama_push
pip install -q git+https://github.com/huggingface/peft.git

  2. Install bitsandbytes from source

Inference

See generate.py.

Training

Under construction.