llm-finetuning docker image

make docker-build --> build the image

make docker-push --> push the image

llm-finetuning script

ovhai job run --gpu 1 --volume llm-datasets@1azgra:/workspace/datasets:ro --volume llm-jobs@1azgra:/workspace/jobs:rwd --env HF_TOKEN=<huggingface_token> ghcr.io/dataesr/llm-finetuning:latest -- ./venv/bin/python3 train.py --mode ["train", "push", "delete"] --model_name <huggingface_model_name> --dataset_name <dataset_name_from_object_storage> --output_model_name <name_of_finetuned_model> --hf_hub <huggingface_hub_id> --hf_hub_private

examples:

• Simple Fine-Tuning ovhai job run --gpu 1 --volume llm-datasets@1azgra:/workspace/datasets:ro --volume llm-jobs@1azgra:/workspace/jobs:rwd --env HF_TOKEN=hf_abcdefg ghcr.io/dataesr/llm-finetuning:latest -- ./venv/bin/python3 train.py --model_name meta-llama/Llama-3.2-1B --dataset_name test.json --output_model_name Llama-3.2-1B-finetuned

• Fine-tuning and push to hub ovhai job run --gpu 1 --volume llm-datasets@1azgra:/workspace/datasets:ro --volume llm-jobs@1azgra:/workspace/jobs:rwd --env HF_TOKEN=hf_abcdefg ghcr.io/dataesr/llm-finetuning:latest -- ./venv/bin/python3 train.py --model_name meta-llama/Llama-3.2-1B --dataset_name test.json --output_model_name Llama-3.2-1B-finetuned --hf_hub dataesr/hub_name --hf_hub_private

• Push trained model to hub ovhai job run --cpu 1 --volume llm-jobs@1azgra:/workspace/jobs:ro --env HF_TOKEN=hf_abcdefg ghcr.io/dataesr/llm-finetuning:latest -- ./venv/bin/python3 train.py --mode push --output_model_name Llama-3.2-1B-finetuned --hf_hub dataesr/hub_name --hf_hub_private

• Delete all jobs files ovhai bucket object delete llm-jobs@1azgra --all --yes

• Inference with fine-tuned model ovhai app run --gpu 1 --env HF_TOKEN=hf_abcdef --env MODEL_NAME=dataesr/openchat-3.6-8b-acknowledgments --default-http-port 8000 --unsecure-http ghcr.io/dataesr/llm-finetuning:latest