{"cells": [{"cell_type": "markdown", "metadata": {}, "source": ["# Alfa-Code Tiny-50M \u00e2\u20ac\u201d Colab Free T4\n", "Runtime \u00e2\u2020\u2019 Change runtime type \u00e2\u2020\u2019 T4 GPU. Then Run all."]}, {"cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": ["from huggingface_hub import login\n", "from google.colab import userdata\n", "tok = userdata.get('HF_TOKEN')\n", "login(token=tok)\n", "print('logged in')\n"]}, {"cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": ["!nvidia-smi --query-gpu=name,memory.total --format=csv\n", "!pip install -q torch transformers datasets tokenizers accelerate huggingface_hub"]}, {"cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": ["from huggingface_hub import snapshot_download\n", "snapshot_download('mrsandip/Alfa-Code', local_dir='/content/alfa', repo_type='model', token=tok)\n", "%cd /content/alfa\n", "!ls scripts/ data/processed/"]}, {"cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": ["# bigger data (500 rows) for real train\n", "!python scripts/download_data.py --dataset code_small_test --max-rows 2000\n", "!python scripts/train_tokenizer.py\n", "!ls -lh tokenizers/ data/processed/"]}, {"cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": ["!python scripts/train_hf.py --data data/processed/train.jsonl --out /content/outputs/tiny-50M --epochs 1\n", "!ls -lh /content/outputs/tiny-50M"]}, {"cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], "source": ["# push back to Hub (set HF_TOKEN in Colab Secrets)\n", "import os\n", "from huggingface_hub import HfApi\n", "tok = os.getenv('HF_TOKEN')\n", "api = HfApi(token=tok)\n", "api.upload_folder(folder_path='/content/outputs/tiny-50M', repo_id='mrsandip/Alfa-Code', path_in_repo='outputs/tiny-50M')\n", "print('uploaded')"]}], "metadata": {"kernelspec": {"display_name": "Python 3", "language": "python", "name": "python3"}, "accelerator": "GPU", "colab": {"provenance": [], "gpuType": "T4"}}, "nbformat": 4, "nbformat_minor": 5}