{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8a5758a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# ============================================================\n",
    "# CELL 1 — Install + Clone\n",
    "# ============================================================\n",
    "import os, sys, subprocess, time\n",
    "from pathlib import Path\n",
    "\n",
    "# ---------- CONFIG ----------\n",
    "REPO_URL          = \"https://github.com/Imsachin010/salespath_env.git\"\n",
    "MODEL_NAME        = \"unsloth/Qwen2.5-7B-Instruct\"   # 7B Model for final submission\n",
    "ENV_URL           = \"http://127.0.0.1:8000\"\n",
    "OUTPUT_DIR        = \"/content/salespath_out\"\n",
    "# -----------------------------------------\n",
    "\n",
    "def run(cmd, check=True, cwd=None):\n",
    "    print(f\"\\n$ {cmd}\")\n",
    "    r = subprocess.run(cmd, shell=True, text=True, capture_output=True, cwd=cwd)\n",
    "    if r.stdout: print(r.stdout.strip())\n",
    "    if r.stderr: print(r.stderr.strip())\n",
    "    if check and r.returncode != 0:\n",
    "        raise RuntimeError(f\"Command failed ({r.returncode}): {cmd}\")\n",
    "    return r\n",
    "\n",
    "!nvidia-smi\n",
    "print(\"Python:\", sys.version)\n",
    "\n",
    "# Install dependencies\n",
    "!pip install -q -U pip\n",
    "!pip uninstall -y openenv 2>/dev/null || true\n",
    "!pip install -q fastapi uvicorn pydantic httpx openenv-core torch transformers trl unsloth datasets pyarrow huggingface_hub matplotlib\n",
    "\n",
    "# Clone repo\n",
    "if not Path(\"/content/salespath_env\").exists():\n",
    "    run(f\"git clone {REPO_URL} /content/salespath_env\")\n",
    "else:\n",
    "    print(\"Repo already cloned.\")\n",
    "\n",
    "REPO_ROOT = \"/content/salespath_env\"\n",
    "os.chdir(REPO_ROOT)\n",
    "print(\"Working dir:\", os.getcwd())\n",
    "\n",
    "# Install package in editable mode\n",
    "run(\"pip install -q -e .\")\n",
    "run(\"python -c \\\"import salespath_env; print('salespath_env import OK')\\\"\")\n",
    "run(\"python -c \\\"import openenv.core; print('openenv.core import OK')\\\"\")\n",
    "\n",
    "# HF Login\n",
    "hf_token = os.environ.get(\"HF_TOKEN\")\n",
    "if hf_token:\n",
    "    from huggingface_hub import login\n",
    "    login(token=hf_token)\n",
    "    print(\"HF login OK\")\n",
    "else:\n",
    "    print(\"HF_TOKEN not set.\")\n",
    "\n",
    "print(\"\\n✅ Setup complete.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "48a62b2d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pull the fix we just pushed\n",
    "!git pull origin main"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "82e04114",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "97e47c56",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "!PYTORCH_ALLOC_CONF=expandable_segments:True \\\n",
    "python -m training.grpo_train \\\n",
    "    --mode grpo \\\n",
    "    --model-name unsloth/Qwen2.5-7B-Instruct \\\n",
    "    --grpo-steps 150 \\\n",
    "    --grpo-dataset-size 128 \\\n",
    "    --num-generations 2 \\\n",
    "    --max-completion-length 128 \\\n",
    "    --per-device-train-batch-size 2 \\\n",
    "    --gradient-accumulation-steps 8 \\\n",
    "    --output-dir /content/salespath_out \\\n",
    "    --logging-steps 10\n"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}