{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "M8NNCivsv-38"
      },
      "outputs": [],
      "source": [
        "!pip install -q huggingface-hub joblib pandas scikit-learn matplotlib seaborn"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from huggingface_hub import hf_hub_download\n",
        "import joblib, os, pandas as pd\n",
        "\n",
        "REPO_ID = \"dr-rakshith-truth-zeeker/truth-zeeker-ai-demo\"\n",
        "FNAME = \"model.joblib\"\n",
        "\n",
        "path = hf_hub_download(repo_id=REPO_ID, filename=FNAME)\n",
        "print(\"✅ Downloaded model from Hugging Face:\", path)\n",
        "\n",
        "model = joblib.load(path)\n",
        "print(\"Model loaded successfully:\", type(model))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4cRLizbOwmYH",
        "outputId": "e8c03cc1-4573-4d17-95aa-73955dbafc76"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "✅ Downloaded model from Hugging Face: /root/.cache/huggingface/hub/models--dr-rakshith-truth-zeeker--truth-zeeker-ai-demo/snapshots/c53c401c56ba50cac5bb805104dfe5ecf380d022/model.joblib\n",
            "Model loaded successfully: <class 'tuple'>\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "csv_url = \"https://raw.githubusercontent.com/dr-rakshith-truth-zeeker/Truth-Zeeker-AI/refs/heads/release/demo_data/sample_features.csv?token=GHSAT0AAAAAADNHI2OE4ZDLAJ26R6E2CJZY2HP242Q\"\n",
        "\n",
        "df = pd.read_csv(csv_url)\n",
        "print(\"Loaded demo data shape:\", df.shape)\n",
        "print(\"\\nFirst 10 rows:\\n\")\n",
        "print(df.head(10).to_string(index=False))\n",
        "\n",
        "# Run inference (handle tuple (scaler, model) or plain model)\n",
        "if isinstance(model, tuple) and len(model) == 2:\n",
        "    scaler, clf = model\n",
        "    Xs = scaler.transform(df[['duration','orig_bytes','resp_bytes']])\n",
        "    preds = clf.predict(Xs)\n",
        "else:\n",
        "    clf = model\n",
        "    preds = clf.predict(df[['duration','orig_bytes','resp_bytes']])\n",
        "\n",
        "df['anomaly_flag'] = preds\n",
        "print(\"\\nResults with anomaly_flag:\\n\")\n",
        "print(df.to_string(index=False))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4NqEdftzx-dN",
        "outputId": "dcff4174-28dc-4ea5-b900-18575399a712"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Loaded demo data shape: (10, 3)\n",
            "\n",
            "First 10 rows:\n",
            "\n",
            " duration  orig_bytes  resp_bytes\n",
            "     0.12         456         789\n",
            "     0.08         123          80\n",
            "     0.30        2000         150\n",
            "     0.05          60           0\n",
            "     0.90        4000        1200\n",
            "     0.02          50          10\n",
            "     0.20         300         400\n",
            "     0.25        1500        1200\n",
            "     0.15         800         200\n",
            "     0.07          90          70\n",
            "\n",
            "Results with anomaly_flag:\n",
            "\n",
            " duration  orig_bytes  resp_bytes  anomaly_flag\n",
            "     0.12         456         789             1\n",
            "     0.08         123          80             1\n",
            "     0.30        2000         150             1\n",
            "     0.05          60           0             1\n",
            "     0.90        4000        1200            -1\n",
            "     0.02          50          10             1\n",
            "     0.20         300         400             1\n",
            "     0.25        1500        1200            -1\n",
            "     0.15         800         200             1\n",
            "     0.07          90          70             1\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.12/dist-packages/sklearn/utils/validation.py:2732: UserWarning: X has feature names, but StandardScaler was fitted without feature names\n",
            "  warnings.warn(\n"
          ]
        }
      ]
    }
  ]
}