Instructions to use PriyaMishra/temp with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use PriyaMishra/temp with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Unsloth Desktop
Download project/run_training_gpu2.sh from PriyaMishra/temp: direct link, hf CLI and curl.
- Browser
- Download file 881 Bytes
-
https://huggingface.co/PriyaMishra/temp/resolve/main/project/run_training_gpu2.sh
- Command line
-
hf download hf://PriyaMishra/temp/project/run_training_gpu2.sh
-
curl -L -o run_training_gpu2.sh https://huggingface.co/PriyaMishra/temp/resolve/main/project/run_training_gpu2.sh
881 Bytes
| set -euo pipefail | |
| cd /home/priya/qwen_unsloth | |
| export CUDA_VISIBLE_DEVICES="${TRAIN_GPU_UUID:-2}" | |
| export TRAIN_BATCH_SIZE="${TRAIN_BATCH_SIZE:-8}" | |
| export TRAIN_GRAD_ACC="${TRAIN_GRAD_ACC:-4}" | |
| export HF_HOME=/22tdata/priya/qwen_unsloth/hf_cache | |
| export HF_DATASETS_CACHE=/22tdata/priya/qwen_unsloth/datasets_cache | |
| export XDG_CACHE_HOME=/22tdata/priya/qwen_unsloth/cache | |
| export TORCHINDUCTOR_CACHE_DIR=/22tdata/priya/qwen_unsloth/torchinductor_cache | |
| export TRITON_CACHE_DIR=/22tdata/priya/qwen_unsloth/triton_cache | |
| export TORCHINDUCTOR_COMPILE_THREADS=4 | |
| export OMP_NUM_THREADS=8 | |
| export TOKENIZERS_PARALLELISM=false | |
| export PYTHONUNBUFFERED=1 | |
| export TRAIN_OUTPUT_DIR="${TRAIN_OUTPUT_DIR:-/22tdata/priya/qwen_unsloth/outputs/Qwen38-27B-3shotTrained-GPU2}" | |
| export TRAIN_RECOVERY_SAVE_STEPS="${TRAIN_RECOVERY_SAVE_STEPS:-10}" | |
| exec .venv/bin/python -u train_qwen38-3S.py | |