{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "17e1f6f6",
   "metadata": {},
   "source": [
    "# 00 - Environment Setup\n",
    "\n",
    "## CyberForge AI - ML Pipeline Environment Configuration\n",
    "\n",
    "This notebook sets up the complete environment for the CyberForge AI machine learning pipeline.\n",
    "\n",
    "### What this notebook does:\n",
    "1. Validates Python version and system requirements\n",
    "2. Installs and pins all dependencies\n",
    "3. Configures GPU/CPU detection\n",
    "4. Sets up Gemini API connectivity\n",
    "5. Validates Web Scraper API connection\n",
    "6. Creates necessary directories\n",
    "\n",
    "### Prerequisites:\n",
    "- Python 3.10+ (3.11 recommended)\n",
    "- Access to Gemini API (API key required)\n",
    "- Access to WebScrapper.live API"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33029fa4",
   "metadata": {},
   "source": [
    "## 1. System Validation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "076fa991",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "import platform\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "print(\"=\" * 60)\n",
    "print(\"CYBERFORGE AI - ENVIRONMENT VALIDATION\")\n",
    "print(\"=\" * 60)\n",
    "\n",
    "# Python version check\n",
    "python_version = sys.version_info\n",
    "print(f\"\\n✓ Python Version: {python_version.major}.{python_version.minor}.{python_version.micro}\")\n",
    "\n",
    "if python_version.major < 3 or (python_version.major == 3 and python_version.minor < 10):\n",
    "    raise EnvironmentError(\"Python 3.10+ is required. Please upgrade your Python installation.\")\n",
    "\n",
    "# System info\n",
    "print(f\"✓ Platform: {platform.system()} {platform.release()}\")\n",
    "print(f\"✓ Architecture: {platform.machine()}\")\n",
    "print(f\"✓ Processor: {platform.processor() or 'Unknown'}\")\n",
    "\n",
    "# Memory info\n",
    "try:\n",
    "    import psutil\n",
    "    memory = psutil.virtual_memory()\n",
    "    print(f\"✓ Available Memory: {memory.available / (1024**3):.2f} GB / {memory.total / (1024**3):.2f} GB\")\n",
    "except ImportError:\n",
    "    print(\"⚠ psutil not installed - memory check skipped\")\n",
    "\n",
    "print(\"\\n\" + \"=\" * 60)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "45e95831",
   "metadata": {},
   "source": [
    "## 2. Install Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "faa9b079",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "\n",
    "# Core dependencies with pinned versions for reproducibility\n",
    "# NOTE: torch/transformers intentionally excluded - not needed for sklearn models\n",
    "# and too heavy for HF Space Docker containers\n",
    "DEPENDENCIES = \"\"\"\n",
    "# Core ML/AI\n",
    "numpy>=1.24.0,<2.0.0\n",
    "pandas>=2.0.0\n",
    "scikit-learn>=1.3.0\n",
    "scipy>=1.11.0\n",
    "\n",
    "# Gemini API (new SDK)\n",
    "google-genai>=1.0.0\n",
    "\n",
    "# Data Processing\n",
    "joblib>=1.3.0\n",
    "tqdm>=4.65.0\n",
    "pyarrow>=14.0.0\n",
    "\n",
    "# Feature Engineering\n",
    "tldextract>=5.0.0\n",
    "validators>=0.22.0\n",
    "\n",
    "# Web/API\n",
    "httpx>=0.25.0\n",
    "requests>=2.31.0\n",
    "\n",
    "# Hugging Face\n",
    "huggingface_hub>=0.19.0\n",
    "\n",
    "# Utilities\n",
    "python-dotenv>=1.0.0\n",
    "pyyaml>=6.0.0\n",
    "psutil>=5.9.0\n",
    "\"\"\"\n",
    "\n",
    "# Write requirements file\n",
    "requirements_path = Path(\"../requirements_notebooks.txt\")\n",
    "requirements_path.write_text(DEPENDENCIES.strip())\n",
    "print(f\"✓ Requirements written to: {requirements_path.absolute()}\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7dc8c6ca",
   "metadata": {},
   "outputs": [],
   "source": [
    "import subprocess\n",
    "import sys\n",
    "from pathlib import Path\n",
    "\n",
    "# Install dependencies\n",
    "requirements_path = Path(\"../requirements_notebooks.txt\")\n",
    "\n",
    "if requirements_path.exists():\n",
    "    print(\"Installing dependencies... This may take a few minutes.\")\n",
    "    result = subprocess.run(\n",
    "        [sys.executable, \"-m\", \"pip\", \"install\", \"-q\", \"-r\", str(requirements_path)],\n",
    "        capture_output=True,\n",
    "        text=True\n",
    "    )\n",
    "\n",
    "    if result.returncode == 0:\n",
    "        print(\"✓ All dependencies installed successfully!\")\n",
    "    else:\n",
    "        print(f\"⚠ Installation warnings: {result.stderr[:500] if result.stderr else 'None'}\")\n",
    "else:\n",
    "    print(\"⚠ Requirements file not found. Run previous cell first or skip if deps installed.\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c11760cc",
   "metadata": {},
   "source": [
    "## 3. GPU/CPU Detection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d1b948c4",
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"=\" * 60)\n",
    "print(\"COMPUTE DEVICE DETECTION\")\n",
    "print(\"=\" * 60)\n",
    "\n",
    "# CyberForge uses sklearn (CPU-only) — torch is optional\n",
    "try:\n",
    "    import torch\n",
    "    cuda_available = torch.cuda.is_available()\n",
    "    print(f\"\\n✓ PyTorch Version: {torch.__version__}\")\n",
    "    print(f\"✓ CUDA Available: {cuda_available}\")\n",
    "\n",
    "    if cuda_available:\n",
    "        print(f\"✓ CUDA Version: {torch.version.cuda}\")\n",
    "        DEVICE = \"cuda\"\n",
    "    elif hasattr(torch.backends, 'mps') and torch.backends.mps.is_available():\n",
    "        print(\"✓ Apple MPS (Metal) available\")\n",
    "        DEVICE = \"mps\"\n",
    "    else:\n",
    "        DEVICE = \"cpu\"\n",
    "except ImportError:\n",
    "    print(\"\\n⚠ PyTorch not installed (not required — sklearn models use CPU)\")\n",
    "    DEVICE = \"cpu\"\n",
    "\n",
    "print(f\"\\n✓ Selected Device: {DEVICE}\")\n",
    "print(\"  (Note: CyberForge models use scikit-learn which runs on CPU)\")\n",
    "print(\"=\" * 60)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d39ddbf5",
   "metadata": {},
   "source": [
    "## 4. Environment Variables & API Configuration"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0f63a5ce",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "# Load configuration from notebook_config.json first (for HF Spaces)\n",
    "config_json_path = Path(\"notebook_config.json\")\n",
    "if config_json_path.exists():\n",
    "    with open(config_json_path, \"r\") as f:\n",
    "        loaded_config = json.load(f)\n",
    "    print(f\"✓ Loaded configuration from: {config_json_path.absolute()}\")\n",
    "else:\n",
    "    loaded_config = {}\n",
    "    print(f\"⚠ No notebook_config.json found, using defaults\")\n",
    "\n",
    "# Try loading .env file as fallback (for local dev)\n",
    "try:\n",
    "    from dotenv import load_dotenv\n",
    "    env_path = Path(\"../.env\")\n",
    "    if env_path.exists():\n",
    "        load_dotenv(env_path)\n",
    "        print(f\"✓ Loaded environment from: {env_path.absolute()}\")\n",
    "except ImportError:\n",
    "    pass\n",
    "\n",
    "# Detect device (torch is optional)\n",
    "DEVICE = \"cpu\"\n",
    "try:\n",
    "    import torch\n",
    "    if torch.cuda.is_available():\n",
    "        DEVICE = \"cuda\"\n",
    "    elif hasattr(torch.backends, 'mps') and torch.backends.mps.is_available():\n",
    "        DEVICE = \"mps\"\n",
    "except ImportError:\n",
    "    pass\n",
    "\n",
    "# Configuration class\n",
    "class Config:\n",
    "    # API Keys - priority: config.json > env vars > defaults\n",
    "    GEMINI_API_KEY = loaded_config.get(\"gemini_api_key\") or os.getenv(\"GEMINI_API_KEY\", \"\")\n",
    "    HUGGINGFACE_TOKEN = loaded_config.get(\"hf_token\") or os.getenv(\"HF_TOKEN\", \"\")\n",
    "    WEBSCRAPER_API_KEY = loaded_config.get(\"webscraper_api_key\", \"sk-fd14eaa7bceb478db7afc7256e514d2b\")\n",
    "    WEBSCRAPER_API_URL = loaded_config.get(\"webscraper_api_url\", \"http://webscrapper.live/api/scrape\")\n",
    "    \n",
    "    # Gemini model\n",
    "    GEMINI_MODEL = loaded_config.get(\"gemini_model\", os.getenv(\"GEMINI_MODEL\", \"gemini-2.5-flash\"))\n",
    "    \n",
    "    # HF repos\n",
    "    HF_REPO = loaded_config.get(\"hf_repo\", \"Che237/cyberforge-models\")\n",
    "    HF_DATASETS_REPO = loaded_config.get(\"hf_datasets_repo\", \"Che237/cyberforge-datasets\")\n",
    "    \n",
    "    # Paths\n",
    "    BASE_DIR = Path(\"..\").resolve()\n",
    "    DATASETS_DIR = BASE_DIR / \"datasets\"\n",
    "    MODELS_DIR = BASE_DIR / \"models\"\n",
    "    ARTIFACTS_DIR = BASE_DIR / \"artifacts\"\n",
    "    \n",
    "    # ML Settings\n",
    "    RANDOM_STATE = loaded_config.get(\"random_state\", 42)\n",
    "    TEST_SIZE = loaded_config.get(\"test_size\", 0.2)\n",
    "    CV_FOLDS = loaded_config.get(\"cv_folds\", 5)\n",
    "    \n",
    "    # Device\n",
    "    DEVICE = DEVICE\n",
    "\n",
    "config = Config()\n",
    "\n",
    "# Validate required API keys\n",
    "print(\"\\n\" + \"=\" * 60)\n",
    "print(\"API CONFIGURATION STATUS\")\n",
    "print(\"=\" * 60)\n",
    "print(f\"  Gemini API Key: {'✓ Set' if config.GEMINI_API_KEY else '✗ Missing'}\")\n",
    "hf_status = '✓ Set' if config.HUGGINGFACE_TOKEN else '⚠ Not set (models will not upload)'\n",
    "print(f\"  HuggingFace Token: {hf_status}\")\n",
    "print(f\"  Gemini Model: {config.GEMINI_MODEL}\")\n",
    "print(f\"  HF Model Repo: {config.HF_REPO}\")\n",
    "print(f\"  Device: {config.DEVICE}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "126b5f7f",
   "metadata": {},
   "source": [
    "## 5. Gemini API Connectivity Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "14cef3bc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Gemini Integration — using google-genai (new SDK)\n",
    "import json\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "try:\n",
    "    from google import genai\n",
    "except ImportError:\n",
    "    import subprocess, sys\n",
    "    subprocess.run([sys.executable, '-m', 'pip', 'install', 'google-genai', '-q'])\n",
    "    from google import genai\n",
    "\n",
    "# Load config (self-contained)\n",
    "config_json_path = Path('notebook_config.json')\n",
    "if config_json_path.exists():\n",
    "    with open(config_json_path, 'r') as f:\n",
    "        loaded_config = json.load(f)\n",
    "else:\n",
    "    loaded_config = {}\n",
    "\n",
    "GEMINI_API_KEY = loaded_config.get('gemini_api_key') or os.getenv('GEMINI_API_KEY', '')\n",
    "GEMINI_MODEL = loaded_config.get('gemini_model', os.getenv('GEMINI_MODEL', 'gemini-2.5-flash'))\n",
    "\n",
    "def test_gemini_connection():\n",
    "    if not GEMINI_API_KEY:\n",
    "        return False, 'API key not configured'\n",
    "    try:\n",
    "        client = genai.Client(api_key=GEMINI_API_KEY)\n",
    "        response = client.models.generate_content(\n",
    "            model=GEMINI_MODEL,\n",
    "            contents='Respond with only: OK'\n",
    "        )\n",
    "        return True, f'Model: {GEMINI_MODEL}, Response: {response.text.strip()}'\n",
    "    except Exception as e:\n",
    "        # Try fallback model\n",
    "        try:\n",
    "            client = genai.Client(api_key=GEMINI_API_KEY)\n",
    "            response = client.models.generate_content(\n",
    "                model='gemini-2.5-flash',\n",
    "                contents='Respond with only: OK'\n",
    "            )\n",
    "            return True, f'Model: gemini-2.5-flash (fallback), Response: {response.text.strip()}'\n",
    "        except Exception as e2:\n",
    "            return False, str(e2)\n",
    "\n",
    "print('Testing Gemini API connection...')\n",
    "success, message = test_gemini_connection()\n",
    "if success:\n",
    "    print(f'✓ Gemini API: {message}')\n",
    "else:\n",
    "    print(f'⚠ Gemini API: Connection failed - {message}')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "628ac121",
   "metadata": {},
   "source": [
    "## 6. Web Scraper API Connectivity Test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "beb1b036",
   "metadata": {},
   "outputs": [],
   "source": [
    "import httpx\n",
    "import json\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "# Load config (self-contained)\n",
    "config_json_path = Path('notebook_config.json')\n",
    "if config_json_path.exists():\n",
    "    with open(config_json_path, 'r') as f:\n",
    "        loaded_config = json.load(f)\n",
    "else:\n",
    "    loaded_config = {}\n",
    "\n",
    "WEBSCRAPER_API_KEY = loaded_config.get('webscraper_api_key', 'sk-fd14eaa7bceb478db7afc7256e514d2b')\n",
    "WEBSCRAPER_API_URL = loaded_config.get('webscraper_api_url', 'http://webscrapper.live/api/scrape')\n",
    "\n",
    "def test_webscraper_connection_sync():\n",
    "    try:\n",
    "        with httpx.Client(timeout=30.0) as client:\n",
    "            response = client.post(\n",
    "                WEBSCRAPER_API_URL,\n",
    "                json={'url': 'https://example.com'},\n",
    "                headers={'Content-Type': 'application/json', 'X-API-Key': WEBSCRAPER_API_KEY}\n",
    "            )\n",
    "            if response.status_code == 200:\n",
    "                return True, 'Connected'\n",
    "            else:\n",
    "                return False, f'Status {response.status_code}'\n",
    "    except Exception as e:\n",
    "        return False, str(e)\n",
    "\n",
    "print('Testing Web Scraper API connection...')\n",
    "success, message = test_webscraper_connection_sync()\n",
    "if success:\n",
    "    print(f'✓ WebScraper API: Connected successfully')\n",
    "else:\n",
    "    print(f'⚠ WebScraper API: {message}')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "75ee0f51",
   "metadata": {},
   "source": [
    "## 7. Create Directory Structure"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "776236f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "from pathlib import Path\n",
    "\n",
    "# Define directories (self-contained)\n",
    "BASE_DIR = Path('..').resolve()\n",
    "DATASETS_DIR = BASE_DIR / 'datasets'\n",
    "MODELS_DIR = BASE_DIR / 'models'\n",
    "ARTIFACTS_DIR = BASE_DIR / 'artifacts'\n",
    "\n",
    "# Create necessary directories\n",
    "directories = [\n",
    "    DATASETS_DIR,\n",
    "    MODELS_DIR,\n",
    "    ARTIFACTS_DIR,\n",
    "    BASE_DIR / 'logs',\n",
    "    BASE_DIR / 'cache',\n",
    "]\n",
    "\n",
    "print('Creating directory structure...')\n",
    "for directory in directories:\n",
    "    directory.mkdir(parents=True, exist_ok=True)\n",
    "    print(f'  ✓ {directory}')\n",
    "\n",
    "print('\\n✓ Directory structure ready!')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6fe27eb",
   "metadata": {},
   "source": [
    "## 8. Save Configuration for Other Notebooks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6b854bac",
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import sys\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "# Get values (self-contained)\n",
    "python_version = sys.version_info\n",
    "\n",
    "DEVICE = 'cpu'\n",
    "torch_version = 'not installed (not required)'\n",
    "cuda_available = False\n",
    "try:\n",
    "    import torch\n",
    "    torch_version = torch.__version__\n",
    "    cuda_available = torch.cuda.is_available()\n",
    "    if cuda_available:\n",
    "        DEVICE = 'cuda'\n",
    "    elif hasattr(torch.backends, 'mps') and torch.backends.mps.is_available():\n",
    "        DEVICE = 'mps'\n",
    "except ImportError:\n",
    "    pass\n",
    "\n",
    "# Load config\n",
    "config_json_path = Path('notebook_config.json')\n",
    "if config_json_path.exists():\n",
    "    with open(config_json_path, 'r') as f:\n",
    "        loaded_config = json.load(f)\n",
    "else:\n",
    "    loaded_config = {}\n",
    "\n",
    "BASE_DIR = Path('..').resolve()\n",
    "DATASETS_DIR = BASE_DIR / 'datasets'\n",
    "MODELS_DIR = BASE_DIR / 'models'\n",
    "ARTIFACTS_DIR = BASE_DIR / 'artifacts'\n",
    "RANDOM_STATE = loaded_config.get('random_state', 42)\n",
    "TEST_SIZE = loaded_config.get('test_size', 0.2)\n",
    "CV_FOLDS = loaded_config.get('cv_folds', 5)\n",
    "\n",
    "# Export configuration for other notebooks\n",
    "notebook_config = {\n",
    "    'device': str(DEVICE),\n",
    "    'python_version': f'{python_version.major}.{python_version.minor}.{python_version.micro}',\n",
    "    'torch_version': torch_version,\n",
    "    'cuda_available': cuda_available,\n",
    "    'base_dir': str(BASE_DIR),\n",
    "    'datasets_dir': str(DATASETS_DIR),\n",
    "    'models_dir': str(MODELS_DIR),\n",
    "    'artifacts_dir': str(ARTIFACTS_DIR),\n",
    "    'random_state': RANDOM_STATE,\n",
    "    'test_size': TEST_SIZE,\n",
    "    'cv_folds': CV_FOLDS,\n",
    "}\n",
    "\n",
    "config_path = Path('notebook_runtime_config.json')\n",
    "with open(config_path, 'w') as f:\n",
    "    json.dump(notebook_config, f, indent=2)\n",
    "\n",
    "print(f'✓ Configuration exported to: {config_path.absolute()}')\n",
    "print(json.dumps(notebook_config, indent=2))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac7ada25",
   "metadata": {},
   "source": [
    "## 9. Environment Summary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f409be56",
   "metadata": {},
   "outputs": [],
   "source": [
    "import sys\n",
    "import json\n",
    "import os\n",
    "from pathlib import Path\n",
    "\n",
    "python_version = sys.version_info\n",
    "\n",
    "try:\n",
    "    import torch\n",
    "    torch_version = torch.__version__\n",
    "    if torch.cuda.is_available():\n",
    "        DEVICE = 'cuda'\n",
    "    elif hasattr(torch.backends, 'mps') and torch.backends.mps.is_available():\n",
    "        DEVICE = 'mps'\n",
    "    else:\n",
    "        DEVICE = 'cpu'\n",
    "except ImportError:\n",
    "    torch_version = 'not installed'\n",
    "    DEVICE = 'cpu'\n",
    "\n",
    "# Load config\n",
    "config_json_path = Path('notebook_config.json')\n",
    "if config_json_path.exists():\n",
    "    with open(config_json_path, 'r') as f:\n",
    "        loaded_config = json.load(f)\n",
    "else:\n",
    "    loaded_config = {}\n",
    "\n",
    "GEMINI_API_KEY = loaded_config.get('gemini_api_key') or os.getenv('GEMINI_API_KEY', '')\n",
    "HUGGINGFACE_TOKEN = os.getenv('HF_TOKEN', '')\n",
    "\n",
    "print('\\n' + '=' * 60)\n",
    "print('ENVIRONMENT SETUP COMPLETE')\n",
    "print('=' * 60)\n",
    "print(f'''\n",
    "✅ Python: {python_version.major}.{python_version.minor}.{python_version.micro}\n",
    "✅ Device: {DEVICE}\n",
    "✅ PyTorch: {torch_version}\n",
    "✅ Gemini API: {'Ready' if GEMINI_API_KEY else 'Not configured'}\n",
    "✅ HuggingFace: {'Ready' if HUGGINGFACE_TOKEN else 'Using public access'}\n",
    "✅ WebScraper API: Ready\n",
    "✅ Directories: Created\n",
    "\n",
    "You can now proceed to the next notebook:\n",
    "  → 01_data_acquisition.ipynb\n",
    "''')\n",
    "print('=' * 60)\n"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}