Skip to content

Alpaca-lora for huggingface implementation using Deepspeed and FullyShardedDataParallel

License

Notifications You must be signed in to change notification settings

naem1023/alpaca-lora-for-huggingface

Repository files navigation

😊 Alpaca-lora for huggingface implementation using Deepspeed and FullyShardedDataParallel 😊

Update Logs

!!! There's some error when you update torch to 2.0. I'll fix it as soon as possible !!!

Features

Dependency

pip install -r requirements.txt

Train

# Use PEFT, LORA
accelerate launch --config_file peft_config.yaml finetune.py

or you can use Huggingface Arguments for controll all situations during training. All the HFArguments can be used.

# You can use train.sh.
# Stil updating...
python train.py \
    --model_name_or_path decapoda-research/llama-7b-hf \
    --dataset_name alpaca_data.hf \
    --is_dataset_from_disk True \
    --per_device_train_batch_size 8 \
    --per_device_eval_batch_size 8 \
    --do_train \
    --do_eval \
    --output_dir test-clm

The codes are still updated, so maybe there're can be some unexpected error. I used base code from https://github.com/tloen/alpaca-lora. Thanks a lot!!

About

Alpaca-lora for huggingface implementation using Deepspeed and FullyShardedDataParallel

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages