tutorials/explainers/RISE/rise_tabular_penguin.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Colab setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img width=\"150\" alt=\"Logo_ER10\" src=\"https://user-images.githubusercontent.com/3244249/151994514-b584b984-a148-4ade-80ee-0f88b0aefa45.png\">\n",
    "\n",
    "### Model Interpretation using LIME for penguin dataset classifier\n",
    "This notebook demonstrates the use of DIANNA with the RISE tabular method on the penguins dataset.\n",
    "\n",
    "RISE is short for Randomized Input Sampling for Explanation of Black-box Models. It estimates each feature's relevance to the model's decision empirically by probing the model with randomly masked versions of the input text and obtaining the corresponding outputs. More details about this method can be found in the [paper introducing RISE](https://arxiv.org/abs/1806.07421)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.815285Z",
     "start_time": "2024-05-30T11:11:52.728074Z"
    }
   },
   "outputs": [],
   "source": [
    "running_in_colab = 'google.colab' in str(get_ipython())\n",
    "if running_in_colab:\n",
    "  # install dianna\n",
    "  !python3 -m pip install dianna[notebooks]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 0 - Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.815285Z",
     "start_time": "2024-05-30T11:11:52.734426Z"
    }
   },
   "outputs": [],
   "source": [
    "import dianna\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "from sklearn.model_selection import train_test_split\n",
    "from dianna.utils.onnx_runner import SimpleModelRunner\n",
    "from dianna.utils.downloader import download"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 1 - Loading the data\n",
    "Load penguins dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.831285Z",
     "start_time": "2024-05-30T11:11:52.748720Z"
    }
   },
   "outputs": [],
   "source": [
    "penguins = sns.load_dataset('penguins')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Prepare the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.891178Z",
     "start_time": "2024-05-30T11:11:52.756731Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>bill_length_mm</th>\n",
       "      <th>bill_depth_mm</th>\n",
       "      <th>flipper_length_mm</th>\n",
       "      <th>body_mass_g</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>39.1</td>\n",
       "      <td>18.7</td>\n",
       "      <td>181.0</td>\n",
       "      <td>3750.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>39.5</td>\n",
       "      <td>17.4</td>\n",
       "      <td>186.0</td>\n",
       "      <td>3800.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>40.3</td>\n",
       "      <td>18.0</td>\n",
       "      <td>195.0</td>\n",
       "      <td>3250.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>36.7</td>\n",
       "      <td>19.3</td>\n",
       "      <td>193.0</td>\n",
       "      <td>3450.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>39.3</td>\n",
       "      <td>20.6</td>\n",
       "      <td>190.0</td>\n",
       "      <td>3650.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>338</th>\n",
       "      <td>47.2</td>\n",
       "      <td>13.7</td>\n",
       "      <td>214.0</td>\n",
       "      <td>4925.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>340</th>\n",
       "      <td>46.8</td>\n",
       "      <td>14.3</td>\n",
       "      <td>215.0</td>\n",
       "      <td>4850.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>341</th>\n",
       "      <td>50.4</td>\n",
       "      <td>15.7</td>\n",
       "      <td>222.0</td>\n",
       "      <td>5750.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>342</th>\n",
       "      <td>45.2</td>\n",
       "      <td>14.8</td>\n",
       "      <td>212.0</td>\n",
       "      <td>5200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>343</th>\n",
       "      <td>49.9</td>\n",
       "      <td>16.1</td>\n",
       "      <td>213.0</td>\n",
       "      <td>5400.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>342 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     bill_length_mm  bill_depth_mm  flipper_length_mm  body_mass_g\n",
       "0              39.1           18.7              181.0       3750.0\n",
       "1              39.5           17.4              186.0       3800.0\n",
       "2              40.3           18.0              195.0       3250.0\n",
       "4              36.7           19.3              193.0       3450.0\n",
       "5              39.3           20.6              190.0       3650.0\n",
       "..              ...            ...                ...          ...\n",
       "338            47.2           13.7              214.0       4925.0\n",
       "340            46.8           14.3              215.0       4850.0\n",
       "341            50.4           15.7              222.0       5750.0\n",
       "342            45.2           14.8              212.0       5200.0\n",
       "343            49.9           16.1              213.0       5400.0\n",
       "\n",
       "[342 rows x 4 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Remove categorial columns and NaN values\n",
    "penguins_filtered = penguins.drop(columns=['island', 'sex']).dropna()\n",
    "\n",
    "# Get the species\n",
    "species = penguins['species'].unique()\n",
    "\n",
    "# Extract inputs and target\n",
    "input_features = penguins_filtered.drop(columns=['species'])\n",
    "target = pd.get_dummies(penguins_filtered['species'])\n",
    "\n",
    "# Let's explore the features of the dataset\n",
    "input_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data-set currently has four features that were used to train the model: bill length, bill depth, flipper length, and body mass. These features were used to classify the different species."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Training, validation, and test data split."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.893178Z",
     "start_time": "2024-05-30T11:11:52.775984Z"
    }
   },
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(input_features, target, test_size=0.2,\n",
    "                                                    random_state=0, shuffle=True, stratify=target)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get an instance to explain."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.893178Z",
     "start_time": "2024-05-30T11:11:52.781926Z"
    }
   },
   "outputs": [],
   "source": [
    "# get an instance from test data\n",
    "data_instance = X_test.iloc[10].to_numpy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 2. Loading ONNX model\n",
    "DIANNA supports ONNX models. Here we demonstrate the use of LIME explainer for tabular data with a pre-trained ONNX model, which is a MLP classifier for the penguins dataset. <br>\n",
    "\n",
    "The model is trained following this notebook: <br>\n",
    "https://github.com/dianna-ai/dianna-exploration/blob/main/example_data/model_generation/penguin_species/generate_model.ipynb"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.921188Z",
     "start_time": "2024-05-30T11:11:52.786936Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Downloading data from 'doi:10.5281/zenodo.10580742/penguin_model.onnx' to file '/home/elenar/.cache/dianna/penguin_model.onnx'.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'Gentoo'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load onnx model and check the prediction with it\n",
    "model_path = download('penguin_model.onnx', 'model')\n",
    "loaded_model = SimpleModelRunner(model_path)\n",
    "predictions = loaded_model(data_instance.reshape(1,-1).astype(np.float32))\n",
    "species[np.argmax(predictions)]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A runner function is created to prepare data for the ONNX inference session."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:52.922190Z",
     "start_time": "2024-05-30T11:11:52.881179Z"
    }
   },
   "outputs": [],
   "source": [
    "import onnxruntime as ort\n",
    "\n",
    "def run_model(data):\n",
    "    # get ONNX predictions\n",
    "    sess = ort.InferenceSession(model_path)\n",
    "    input_name = sess.get_inputs()[0].name\n",
    "    output_name = sess.get_outputs()[0].name\n",
    "\n",
    "    onnx_input = {input_name: data.astype(np.float32)}\n",
    "    pred_onnx = sess.run([output_name], onnx_input)[0]\n",
    "    \n",
    "    return pred_onnx"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 3. Applying RISE with DIANNA\n",
    "The simplest way to run DIANNA on image data is with `dianna.explain_tabular`.\n",
    "\n",
    "DIANNA requires input in numpy format, so the input data is converted into a numpy array.\n",
    "\n",
    "Note that the training data is also required for proper perturbation of the input data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:53.364609Z",
     "start_time": "2024-05-30T11:11:52.885179Z"
    }
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Explaining: 100%|██████████████████████████████████████████████████████████████████████| 10/10 [00:00<00:00, 738.01it/s]\n"
     ]
    }
   ],
   "source": [
    "explanation = dianna.explain_tabular(run_model, input_tabular=data_instance, method='rise',\n",
    "                                     mode ='classification', training_data = X_train.to_numpy(),\n",
    "                                     feature_names=input_features.columns, class_names=species)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:53.369138Z",
     "start_time": "2024-05-30T11:11:53.332880Z"
    },
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1.3488360e-05, 3.1694275e-05, 9.9995482e-01]], dtype=float32)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "run_model(data_instance[None,...])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### 4. Visualization\n",
    "The output can be visualized with the DIANNA built-in visualization function. It shows the importance of each feature contributing to the prediction.\n",
    "\n",
    "The prediction is \"Gentoo\", so let's visualize the feature importance scores for \"Gentoo\".\n",
    "\n",
    "It can be noticed that the body mass feature has the biggest weight in the prediction."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:53.560552Z",
     "start_time": "2024-05-30T11:11:53.370141Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from dianna.visualization import plot_tabular\n",
    "\n",
    "_ = plot_tabular(explanation[np.argmax(predictions)], X_test.columns, num_features=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-30T11:11:53.565301Z",
     "start_time": "2024-05-30T11:11:53.561551Z"
    },
    "collapsed": false,
    "jupyter": {
     "outputs_hidden": false
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[6.3201445e-01, 6.6387820e-01, 6.6702765e-01, 8.4758140e-06],\n",
       "       [1.9935470e-02, 2.0091312e-02, 9.9996408e-04, 7.2206632e-05],\n",
       "       [3.3405048e-01, 3.2203084e-01, 3.4197244e-01, 9.9791914e-01]],\n",
       "      dtype=float32)"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "explanation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}