inf-sele-1-eval-clustering-on-nbi-inf.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "1197f5c1",
   "metadata": {
    "_cell_guid": "b1076dfc-b9ad-4769-8c92-a6c4dae69d19",
    "_uuid": "8f2839f25d086af736a60e9eeb907d3b93b6e0e5",
    "papermill": {
     "duration": 0.046872,
     "end_time": "2022-04-26T05:19:05.612531",
     "exception": false,
     "start_time": "2022-04-26T05:19:05.565659",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Introduction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a60d6ed4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:05.705496Z",
     "iopub.status.busy": "2022-04-26T05:19:05.704174Z",
     "iopub.status.idle": "2022-04-26T05:19:29.340180Z",
     "shell.execute_reply": "2022-04-26T05:19:29.339137Z",
     "shell.execute_reply.started": "2022-04-26T05:01:27.697703Z"
    },
    "papermill": {
     "duration": 23.681461,
     "end_time": "2022-04-26T05:19:29.340414",
     "exception": false,
     "start_time": "2022-04-26T05:19:05.658953",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# for everything else\n",
    "import os\n",
    "import random\n",
    "import scipy\n",
    "\n",
    "from functools import partial\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from pandas_profiling import ProfileReport\n",
    "\n",
    "from imageio import imread\n",
    "from keras.preprocessing.image import load_img \n",
    "from keras.preprocessing.image import img_to_array \n",
    "from keras.applications.vgg16 import preprocess_input \n",
    "\n",
    "# models \n",
    "from keras.applications.vgg16 import VGG16 \n",
    "from keras.models import Model\n",
    "import pickle\n",
    "# from keras.layers import Activation, Dense, Dropout\n",
    "\n",
    "from sklearn.decomposition import PCA\n",
    "from sklearn.manifold import TSNE\n",
    "import umap\n",
    "\n",
    "# clustering and dimension reduction\n",
    "from sklearn.cluster import AgglomerativeClustering\n",
    "from sklearn.cluster import KMeans\n",
    "from sklearn.cluster import MiniBatchKMeans\n",
    "from sklearn.cluster import SpectralClustering\n",
    "from sklearn.cluster import DBSCAN\n",
    "# import hdbscan\n",
    "\n",
    "from sklearn import metrics\n",
    "from sklearn.metrics import roc_curve, auc\n",
    "from sklearn.metrics import silhouette_score\n",
    "from sklearn.metrics import calinski_harabasz_score\n",
    "from sklearn.metrics import precision_score, recall_score, f1_score, accuracy_score\n",
    "from sklearn.metrics import classification_report\n",
    "from sklearn.metrics import precision_recall_fscore_support\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "from keras.utils.np_utils import to_categorical\n",
    "from tqdm import trange\n",
    "from hyperopt import fmin, tpe, hp, STATUS_OK, space_eval, Trials\n",
    "\n",
    "from sklearn.metrics import plot_confusion_matrix\n",
    "import plotly.graph_objs as go\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from mpl_toolkits.mplot3d import Axes3D\n",
    "\n",
    "%matplotlib inline\n",
    "sns.set(style='white', context='notebook', rc={'figure.figsize':(14,10)})\n",
    "\n",
    "from warnings import filterwarnings\n",
    "filterwarnings('ignore')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "f47525a5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:29.436652Z",
     "iopub.status.busy": "2022-04-26T05:19:29.435838Z",
     "iopub.status.idle": "2022-04-26T05:19:29.438325Z",
     "shell.execute_reply": "2022-04-26T05:19:29.438774Z",
     "shell.execute_reply.started": "2022-04-26T05:01:51.448003Z"
    },
    "papermill": {
     "duration": 0.053822,
     "end_time": "2022-04-26T05:19:29.438959",
     "exception": false,
     "start_time": "2022-04-26T05:19:29.385137",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "random.seed(10)\n",
    "np.random.seed(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "883f0e1f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:29.533030Z",
     "iopub.status.busy": "2022-04-26T05:19:29.531758Z",
     "iopub.status.idle": "2022-04-26T05:19:29.535108Z",
     "shell.execute_reply": "2022-04-26T05:19:29.534545Z",
     "shell.execute_reply.started": "2022-04-26T05:01:51.455583Z"
    },
    "papermill": {
     "duration": 0.052244,
     "end_time": "2022-04-26T05:19:29.535281",
     "exception": false,
     "start_time": "2022-04-26T05:19:29.483037",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "config = {\n",
    "    'method': 'AgglomerativeClustering',\n",
    "    'n_neighbors': 60,\n",
    "    'min_dist': 0.15,\n",
    "    'n_components': 2,\n",
    "    'n_clusters': 4,\n",
    "    'n_init': 10,\n",
    "    'random_state': 42\n",
    "}"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9dbca03b",
   "metadata": {
    "papermill": {
     "duration": 0.043634,
     "end_time": "2022-04-26T05:19:29.625418",
     "exception": false,
     "start_time": "2022-04-26T05:19:29.581784",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Load Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "0d449e60",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:29.726283Z",
     "iopub.status.busy": "2022-04-26T05:19:29.725539Z",
     "iopub.status.idle": "2022-04-26T05:19:29.906663Z",
     "shell.execute_reply": "2022-04-26T05:19:29.907559Z",
     "shell.execute_reply.started": "2022-04-26T05:01:51.482086Z"
    },
    "papermill": {
     "duration": 0.236694,
     "end_time": "2022-04-26T05:19:29.907775",
     "exception": false,
     "start_time": "2022-04-26T05:19:29.671081",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "base_dir = '../input/nbiinfframes/FRAMES'\n",
    "Fold1 = '../input/nbiinfframes/FRAMES/Fold1'\n",
    "Fold2 = '../input/nbiinfframes/FRAMES/Fold2'\n",
    "Fold3 = '../input/nbiinfframes/FRAMES/Fold3'\n",
    "\n",
    "pkl_file = r\"../input/feature-embedding/files_feature_embeddings_vgg16.pkl\"\n",
    "\n",
    "class_ = ['B', 'I', 'S', 'U']\n",
    "\n",
    "Folds = [Fold1, Fold2, Fold3]\n",
    "all_data = []\n",
    "\n",
    "# Define path to the data directory\n",
    "for fold in Folds:\n",
    "    for case in class_:\n",
    "        path = os.path.join(fold, case)\n",
    "        if os.path.isdir(path):\n",
    "            for img in os.listdir(path):\n",
    "                all_data.append([os.path.join(path, img), case, os.path.basename(fold)])\n",
    "        \n",
    "all_df = pd.DataFrame(all_data, columns=['image_path', 'class', 'folder'],index=None)\n",
    "all_df.dropna(inplace = True)\n",
    "\n",
    "all_df['name'] = all_df['image_path'].map(lambda x: os.path.basename(x))\n",
    "img_name = all_df['name'] .to_list()\n",
    "files = all_df['image_path'].to_list()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "be60e536",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:30.000585Z",
     "iopub.status.busy": "2022-04-26T05:19:29.999502Z",
     "iopub.status.idle": "2022-04-26T05:19:30.013562Z",
     "shell.execute_reply": "2022-04-26T05:19:30.014066Z",
     "shell.execute_reply.started": "2022-04-26T05:01:51.717553Z"
    },
    "papermill": {
     "duration": 0.063469,
     "end_time": "2022-04-26T05:19:30.014236",
     "exception": false,
     "start_time": "2022-04-26T05:19:29.950767",
     "status": "completed"
    },
    "tags": []
   },
   "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>image_path</th>\n",
       "      <th>class</th>\n",
       "      <th>folder</th>\n",
       "      <th>name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...</td>\n",
       "      <td>B</td>\n",
       "      <td>Fold1</td>\n",
       "      <td>mvideo28_03796.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...</td>\n",
       "      <td>B</td>\n",
       "      <td>Fold1</td>\n",
       "      <td>mvideo28_04694.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...</td>\n",
       "      <td>B</td>\n",
       "      <td>Fold1</td>\n",
       "      <td>mvideo28_04039.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>../input/nbiinfframes/FRAMES/Fold1/B/video8_02...</td>\n",
       "      <td>B</td>\n",
       "      <td>Fold1</td>\n",
       "      <td>video8_02675.jpg</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...</td>\n",
       "      <td>B</td>\n",
       "      <td>Fold1</td>\n",
       "      <td>mvideo28_04040.jpg</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          image_path class folder  \\\n",
       "0  ../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...     B  Fold1   \n",
       "1  ../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...     B  Fold1   \n",
       "2  ../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...     B  Fold1   \n",
       "3  ../input/nbiinfframes/FRAMES/Fold1/B/video8_02...     B  Fold1   \n",
       "4  ../input/nbiinfframes/FRAMES/Fold1/B/mvideo28_...     B  Fold1   \n",
       "\n",
       "                 name  \n",
       "0  mvideo28_03796.jpg  \n",
       "1  mvideo28_04694.jpg  \n",
       "2  mvideo28_04039.jpg  \n",
       "3    video8_02675.jpg  \n",
       "4  mvideo28_04040.jpg  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "all_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7994ad21",
   "metadata": {
    "papermill": {
     "duration": 0.043659,
     "end_time": "2022-04-26T05:19:30.100846",
     "exception": false,
     "start_time": "2022-04-26T05:19:30.057187",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Display"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e54345db",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:30.194081Z",
     "iopub.status.busy": "2022-04-26T05:19:30.193259Z",
     "iopub.status.idle": "2022-04-26T05:19:33.221970Z",
     "shell.execute_reply": "2022-04-26T05:19:33.222491Z",
     "shell.execute_reply.started": "2022-04-26T05:01:51.739734Z"
    },
    "papermill": {
     "duration": 3.077719,
     "end_time": "2022-04-26T05:19:33.222693",
     "exception": false,
     "start_time": "2022-04-26T05:19:30.144974",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x720 with 5 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "image_example = [all_df.loc[all_df['class']== 'I']['image_path'].iloc[15],\n",
    "                all_df.loc[all_df['class']== 'B']['image_path'].iloc[15],\n",
    "                all_df.loc[all_df['class']== 'S']['image_path'].iloc[30],\n",
    "                all_df.loc[all_df['class']== 'U']['image_path'].iloc[15]]\n",
    "\n",
    "fig, axs = plt.subplots(1, 4, constrained_layout=True)\n",
    "pcm = [i for i in range(4)]\n",
    "for i in range(4):\n",
    "    axs[i].set_xticks([])\n",
    "    axs[i].set_yticks([])\n",
    "    axs[i].set_xlabel(['(a){}', '(b){}', '(c){}', '(d){}'][i].format([' I', ' B', ' S', ' U'][i]))\n",
    "    pcm = axs[i].imshow(imread(image_example[i]))\n",
    "    \n",
    "plt.tight_layout()\n",
    "cbar = fig.colorbar(pcm, \n",
    "                    ax=axs[-1:], \n",
    "#                     shrink=1.5, \n",
    "                    cax = fig.add_axes([0.01, 0.3, 0.98, 0.02]), ticks=[], orientation='horizontal') #shrink=1.0, #location='bottom',  orientation='horizontal' #fraction=0.19, aspect=15\n",
    "\n",
    "## remove white margin\n",
    "plt.gca().set_axis_off()\n",
    "plt.subplots_adjust(top = 1, bottom = 0, right = 1, left = 0, \n",
    "            hspace = 0, wspace = 0)\n",
    "plt.margins(0,0)\n",
    "plt.gca().xaxis.set_major_locator(plt.NullLocator())\n",
    "plt.gca().yaxis.set_major_locator(plt.NullLocator())\n",
    "plt.savefig(\"dataset_overview.pdf\", bbox_inches = 'tight',pad_inches = 0)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cb59c866",
   "metadata": {
    "papermill": {
     "duration": 0.05035,
     "end_time": "2022-04-26T05:19:33.324849",
     "exception": false,
     "start_time": "2022-04-26T05:19:33.274499",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Methods: Building Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "b94e5da0",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:33.435647Z",
     "iopub.status.busy": "2022-04-26T05:19:33.429797Z",
     "iopub.status.idle": "2022-04-26T05:19:38.498670Z",
     "shell.execute_reply": "2022-04-26T05:19:38.497764Z",
     "shell.execute_reply.started": "2022-04-26T05:01:54.816976Z"
    },
    "papermill": {
     "duration": 5.123517,
     "end_time": "2022-04-26T05:19:38.498899",
     "exception": false,
     "start_time": "2022-04-26T05:19:33.375382",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-04-26 05:19:33.498180: I tensorflow/core/common_runtime/process_util.cc:146] Creating new thread pool with default inter op setting: 2. Tune using inter_op_parallelism_threads for best performance.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/vgg16/vgg16_weights_tf_dim_ordering_tf_kernels.h5\n",
      "553467904/553467096 [==============================] - 3s 0us/step\n",
      "553476096/553467096 [==============================] - 3s 0us/step\n"
     ]
    }
   ],
   "source": [
    "model = VGG16()\n",
    "model = Model(inputs = model.inputs, outputs = model.layers[-2].output)\n",
    "out_features = model.output.shape[1] # to obtain the number of the output feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "cbdc187f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:38.668237Z",
     "iopub.status.busy": "2022-04-26T05:19:38.667562Z",
     "iopub.status.idle": "2022-04-26T05:19:38.682814Z",
     "shell.execute_reply": "2022-04-26T05:19:38.683378Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.766751Z"
    },
    "papermill": {
     "duration": 0.097657,
     "end_time": "2022-04-26T05:19:38.683585",
     "exception": false,
     "start_time": "2022-04-26T05:19:38.585928",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"model\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "input_1 (InputLayer)         [(None, 224, 224, 3)]     0         \n",
      "_________________________________________________________________\n",
      "block1_conv1 (Conv2D)        (None, 224, 224, 64)      1792      \n",
      "_________________________________________________________________\n",
      "block1_conv2 (Conv2D)        (None, 224, 224, 64)      36928     \n",
      "_________________________________________________________________\n",
      "block1_pool (MaxPooling2D)   (None, 112, 112, 64)      0         \n",
      "_________________________________________________________________\n",
      "block2_conv1 (Conv2D)        (None, 112, 112, 128)     73856     \n",
      "_________________________________________________________________\n",
      "block2_conv2 (Conv2D)        (None, 112, 112, 128)     147584    \n",
      "_________________________________________________________________\n",
      "block2_pool (MaxPooling2D)   (None, 56, 56, 128)       0         \n",
      "_________________________________________________________________\n",
      "block3_conv1 (Conv2D)        (None, 56, 56, 256)       295168    \n",
      "_________________________________________________________________\n",
      "block3_conv2 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_conv3 (Conv2D)        (None, 56, 56, 256)       590080    \n",
      "_________________________________________________________________\n",
      "block3_pool (MaxPooling2D)   (None, 28, 28, 256)       0         \n",
      "_________________________________________________________________\n",
      "block4_conv1 (Conv2D)        (None, 28, 28, 512)       1180160   \n",
      "_________________________________________________________________\n",
      "block4_conv2 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_conv3 (Conv2D)        (None, 28, 28, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block4_pool (MaxPooling2D)   (None, 14, 14, 512)       0         \n",
      "_________________________________________________________________\n",
      "block5_conv1 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv2 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_conv3 (Conv2D)        (None, 14, 14, 512)       2359808   \n",
      "_________________________________________________________________\n",
      "block5_pool (MaxPooling2D)   (None, 7, 7, 512)         0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 25088)             0         \n",
      "_________________________________________________________________\n",
      "fc1 (Dense)                  (None, 4096)              102764544 \n",
      "_________________________________________________________________\n",
      "fc2 (Dense)                  (None, 4096)              16781312  \n",
      "=================================================================\n",
      "Total params: 134,260,544\n",
      "Trainable params: 134,260,544\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "files[0],  os.path.basename(files[0])\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88e381df",
   "metadata": {
    "papermill": {
     "duration": 0.077445,
     "end_time": "2022-04-26T05:19:38.839895",
     "exception": false,
     "start_time": "2022-04-26T05:19:38.762450",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Feature Extraction by CNN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "956f5648",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:39.007947Z",
     "iopub.status.busy": "2022-04-26T05:19:39.006964Z",
     "iopub.status.idle": "2022-04-26T05:19:39.012259Z",
     "shell.execute_reply": "2022-04-26T05:19:39.011409Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.785325Z"
    },
    "papermill": {
     "duration": 0.095986,
     "end_time": "2022-04-26T05:19:39.012506",
     "exception": false,
     "start_time": "2022-04-26T05:19:38.916520",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def extract_features(files, model, pkl=\"\", out_feature=out_features):\n",
    "    \n",
    "    data = {}\n",
    "    if os.path.exists(pkl):\n",
    "        with open(pkl, 'rb') as f:\n",
    "            data = pickle.load(f)\n",
    "    else:\n",
    "        print(\"File data pkl file doesn't exist..., extracting by model\")\n",
    "        for file in files:\n",
    "            resized_img = load_img(file, target_size=(224,224)) ##reshaped\n",
    "            img_arr = np.array(resized_img) \n",
    "            reshaped_img = img_arr.reshape(1,224,224,3)  ##reshape\n",
    "            imgx = preprocess_input(reshaped_img)\n",
    "            embeddings = model.predict(imgx, use_multiprocessing=True)\n",
    "            data[os.path.basename(file)] = embeddings\n",
    "            \n",
    "        pickle.dump(data, open(pkl, \"wb\"))\n",
    " \n",
    "    filenames = np.array(list(data.keys()))\n",
    "    feat = np.array(list(data.values()))\n",
    "    feat = feat.reshape(-1, out_features)\n",
    "\n",
    "    return filenames, feat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "0bee1ea1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:39.214552Z",
     "iopub.status.busy": "2022-04-26T05:19:39.213824Z",
     "iopub.status.idle": "2022-04-26T05:19:39.472383Z",
     "shell.execute_reply": "2022-04-26T05:19:39.473351Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.799663Z"
    },
    "papermill": {
     "duration": 0.358619,
     "end_time": "2022-04-26T05:19:39.473729",
     "exception": false,
     "start_time": "2022-04-26T05:19:39.115110",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "extracted feature shape: (720, 4096)\n",
      "CPU times: user 18 ms, sys: 13.8 ms, total: 31.7 ms\n",
      "Wall time: 254 ms\n"
     ]
    }
   ],
   "source": [
    "%%time\n",
    "\n",
    "filenames, feat = extract_features(files, model, pkl_file, out_features)    \n",
    "print('extracted feature shape: {}'.format(feat.shape))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e283a90",
   "metadata": {
    "papermill": {
     "duration": 0.076398,
     "end_time": "2022-04-26T05:19:39.629812",
     "exception": false,
     "start_time": "2022-04-26T05:19:39.553414",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "8ff1c6bb",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:39.794181Z",
     "iopub.status.busy": "2022-04-26T05:19:39.793087Z",
     "iopub.status.idle": "2022-04-26T05:19:39.795333Z",
     "shell.execute_reply": "2022-04-26T05:19:39.795817Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.948045Z"
    },
    "papermill": {
     "duration": 0.087857,
     "end_time": "2022-04-26T05:19:39.795985",
     "exception": false,
     "start_time": "2022-04-26T05:19:39.708128",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def generate_mkmclusters(feature_embeddings,\n",
    "                       n_neighbors = 190,\n",
    "                       min_dist = 0.25,\n",
    "                       metric='manhattan',\n",
    "                       n_components = 2,\n",
    "                       n_clusters = 4,\n",
    "                       n_init = 10,\n",
    "                       random_state = 42,\n",
    "                       batch_size = 512):\n",
    "    \n",
    "    umap_embeddings = umap.UMAP(n_neighbors = n_neighbors,\n",
    "                                min_dist = min_dist,\n",
    "                                n_components = n_components,\n",
    "                                metric = metric,\n",
    "                                random_state=random_state).fit_transform(feature_embeddings)\n",
    "    \n",
    "    clusters = MiniBatchKMeans(init=\"k-means++\",\n",
    "                          n_clusters = n_clusters,\n",
    "                          batch_size = batch_size,\n",
    "                          max_iter=400,\n",
    "                          n_init = n_init,\n",
    "                          max_no_improvement=10,\n",
    "                          verbose=0,\n",
    "                          random_state=random_state).fit(umap_embeddings)\n",
    "    \n",
    "    return clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "4223d3f7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:39.957648Z",
     "iopub.status.busy": "2022-04-26T05:19:39.956560Z",
     "iopub.status.idle": "2022-04-26T05:19:39.964655Z",
     "shell.execute_reply": "2022-04-26T05:19:39.965135Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.958538Z"
    },
    "papermill": {
     "duration": 0.090621,
     "end_time": "2022-04-26T05:19:39.965335",
     "exception": false,
     "start_time": "2022-04-26T05:19:39.874714",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def generate_speclusters(feature_embeddings,\n",
    "                       n_neighbors = 260,\n",
    "                       min_dist = 0.0,\n",
    "                       metric = 'manhattan',\n",
    "                       eigen_solver = 'arpack',\n",
    "                       affinity = 'rbf',\n",
    "                       assign_labels = 'kmeans',\n",
    "                       random_state = 42,\n",
    "                       n_components = 2,\n",
    "                       n_clusters = 4,\n",
    "                       n_init = 10):\n",
    "        \n",
    "    umap_embeddings = umap.UMAP(n_neighbors = n_neighbors,\n",
    "                                min_dist = min_dist,\n",
    "                                n_components = n_components,\n",
    "                                metric = metric,\n",
    "                                random_state=random_state).fit_transform(feature_embeddings)\n",
    "    \n",
    "    clusters = SpectralClustering(n_clusters = n_clusters,\n",
    "                                  n_components = n_components,\n",
    "                                  n_init = n_init,\n",
    "                                  eigen_solver = eigen_solver,\n",
    "                                  affinity = affinity,\n",
    "                                  assign_labels = assign_labels,\n",
    "                                  gamma=1.0,\n",
    "                                  n_jobs=None, \n",
    "                                  n_neighbors=10,\n",
    "                                  random_state=random_state).fit(umap_embeddings) #tsne_sc_s\n",
    "\n",
    "    return clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "9a5d4a6c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:40.131308Z",
     "iopub.status.busy": "2022-04-26T05:19:40.130362Z",
     "iopub.status.idle": "2022-04-26T05:19:40.132880Z",
     "shell.execute_reply": "2022-04-26T05:19:40.133474Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.974863Z"
    },
    "papermill": {
     "duration": 0.089511,
     "end_time": "2022-04-26T05:19:40.133646",
     "exception": false,
     "start_time": "2022-04-26T05:19:40.044135",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def generate_aggclusters(feature_embeddings,\n",
    "                       n_neighbors = 60,\n",
    "                       min_dist = 0.15000000000000002,\n",
    "                       metric = 'hellinger',\n",
    "                       affinity = 'euclidean',\n",
    "                       linkage = 'ward',\n",
    "                       random_state = 42,\n",
    "                       n_components = 2,\n",
    "                       n_clusters = 4,\n",
    "                       n_init = 10):\n",
    "        \n",
    "    umap_embeddings = umap.UMAP(n_neighbors = n_neighbors,\n",
    "                                min_dist = min_dist,\n",
    "                                n_components = n_components,\n",
    "                                metric = metric,\n",
    "                                random_state=random_state).fit_transform(feature_embeddings)\n",
    "    \n",
    "    clusters = AgglomerativeClustering(n_clusters=n_clusters,\n",
    "                                       affinity=affinity,\n",
    "                                       linkage=linkage,\n",
    "                                       memory=None,\n",
    "                                       connectivity=None,\n",
    "                                       compute_full_tree='auto',\n",
    "                                       distance_threshold=None, \n",
    "                                       compute_distances=False).fit(umap_embeddings) \n",
    "    \n",
    "    return clusters"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba0f9e58",
   "metadata": {
    "papermill": {
     "duration": 0.075818,
     "end_time": "2022-04-26T05:19:40.287158",
     "exception": false,
     "start_time": "2022-04-26T05:19:40.211340",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "{'dbscan_metric': 'l2', 'eps': 0.6001000000000001, 'min_dist': 0.0, 'min_samples': 15, 'n_neighbors': 120, 'umap_metric': 'correlation'}\n",
    "\n",
    "1️⃣ UMAP, n_neighbors: 80, min_dist: 0.0, umap_metric: hellinger\n",
    "\n",
    "2️⃣ DBSCAN, eps: 0.7000000000000002, min_samples: 15, dbscan_metric:l2, algorithm: auto\n",
    "\n",
    "3️⃣ LABLE&COUNTS: {0: 203, 1: 143, 2: 217, 3: 157}\n",
    "\n",
    "4️⃣ LOSS: out_counts: 0, label_dev: 0, cluster_var: 949.0, total: 9.49"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "6a9ab4ff",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:40.444017Z",
     "iopub.status.busy": "2022-04-26T05:19:40.443012Z",
     "iopub.status.idle": "2022-04-26T05:19:40.448992Z",
     "shell.execute_reply": "2022-04-26T05:19:40.449641Z",
     "shell.execute_reply.started": "2022-04-26T05:01:59.991821Z"
    },
    "papermill": {
     "duration": 0.08603,
     "end_time": "2022-04-26T05:19:40.449814",
     "exception": false,
     "start_time": "2022-04-26T05:19:40.363784",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def generate_hdbclusters(feature_embeddings,\n",
    "                       random_state = config['random_state'],\n",
    "                       n_components = config['n_components']):\n",
    "    \n",
    "\n",
    "    umap_embeddings = umap.UMAP(n_neighbors = 80,\n",
    "                                min_dist = 0.0,\n",
    "                                n_components = n_components,\n",
    "                                metric = 'hellinger',\n",
    "                                random_state=random_state).fit_transform(feature_embeddings)\n",
    "    \n",
    "    clusters = DBSCAN(eps=0.7000000000000002,\n",
    "                      min_samples = 15,\n",
    "                      metric = 'l2').fit(umap_embeddings) #algorithm = algorithm\n",
    "    \n",
    "    return clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "07a1d4dd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:40.607049Z",
     "iopub.status.busy": "2022-04-26T05:19:40.606021Z",
     "iopub.status.idle": "2022-04-26T05:19:40.613835Z",
     "shell.execute_reply": "2022-04-26T05:19:40.614416Z",
     "shell.execute_reply.started": "2022-04-26T05:02:00.008444Z"
    },
    "papermill": {
     "duration": 0.088162,
     "end_time": "2022-04-26T05:19:40.614602",
     "exception": false,
     "start_time": "2022-04-26T05:19:40.526440",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def auto_labeling(gt_df, clustered_df, class_ = ['B', 'I', 'S', 'U']):\n",
    "    target2label = {}\n",
    "    \n",
    "    g1 = clustered_df.groupby('clustered_target')['name']\n",
    "    g2 = gt_df.groupby('class')['name']\n",
    "    \n",
    "    for n in list(g1.groups.keys()):\n",
    "        group_ins_num = [len(set(g1.get_group(n)) & set(g2.get_group(label))) for label in list(g2.groups.keys())]\n",
    "        target2label[n] = class_[group_ins_num.index(max(group_ins_num))]\n",
    "        print('the group {} intersect with {} is {}, the likely label is {}'.format(n, g2.groups.keys(),group_ins_num, target2label[n]))\n",
    "\n",
    "    return target2label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "52142560",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:40.775885Z",
     "iopub.status.busy": "2022-04-26T05:19:40.774897Z",
     "iopub.status.idle": "2022-04-26T05:19:52.372557Z",
     "shell.execute_reply": "2022-04-26T05:19:52.371569Z",
     "shell.execute_reply.started": "2022-04-26T05:02:00.025089Z"
    },
    "papermill": {
     "duration": 11.679027,
     "end_time": "2022-04-26T05:19:52.372772",
     "exception": false,
     "start_time": "2022-04-26T05:19:40.693745",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the group 0 intersect with dict_keys(['B', 'I', 'S', 'U']) is [178, 0, 23, 2], the likely label is B\n",
      "the group 1 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 142, 1, 0], the likely label is I\n",
      "the group 2 intersect with dict_keys(['B', 'I', 'S', 'U']) is [2, 38, 0, 177], the likely label is U\n",
      "the group 3 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 0, 156, 1], the likely label is S\n"
     ]
    },
    {
     "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>y_Actual</th>\n",
       "      <th>y_Predicted</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>715</th>\n",
       "      <td>U</td>\n",
       "      <td>U</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>716</th>\n",
       "      <td>U</td>\n",
       "      <td>U</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>717</th>\n",
       "      <td>U</td>\n",
       "      <td>U</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>718</th>\n",
       "      <td>U</td>\n",
       "      <td>U</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>719</th>\n",
       "      <td>U</td>\n",
       "      <td>U</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>720 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    y_Actual y_Predicted\n",
       "0          B           B\n",
       "1          B           B\n",
       "2          B           B\n",
       "3          B           B\n",
       "4          B           B\n",
       "..       ...         ...\n",
       "715        U           U\n",
       "716        U           U\n",
       "717        U           U\n",
       "718        U           U\n",
       "719        U           U\n",
       "\n",
       "[720 rows x 2 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clusters = generate_hdbclusters(feat)\n",
    "clustered_target = clusters.labels_ #group id generated from cluster\n",
    "clustered_df = pd.DataFrame({'name':filenames, 'clustered_target':clustered_target})\n",
    "\n",
    "clustered_df = clustered_df[clustered_df['clustered_target'] != -1] # for the dbscan\n",
    "\n",
    "target2label = auto_labeling(all_df, clustered_df, class_)\n",
    "\n",
    "clustered_df['predict_class'] = clustered_df['clustered_target'].map(target2label)\n",
    "eval_df = pd.merge(clustered_df, all_df, on='name', how='left')\n",
    "\n",
    "\n",
    "clustered_df.shape[0]\n",
    "\n",
    "eval_df\n",
    "\n",
    "df = pd.DataFrame({'y_Actual': eval_df['class'], \n",
    "                   'y_Predicted': eval_df['predict_class']},\n",
    "                  columns=['y_Actual','y_Predicted'])\n",
    "\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "761a24cc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:52.537699Z",
     "iopub.status.busy": "2022-04-26T05:19:52.536986Z",
     "iopub.status.idle": "2022-04-26T05:19:52.538979Z",
     "shell.execute_reply": "2022-04-26T05:19:52.539504Z",
     "shell.execute_reply.started": "2022-04-26T05:02:11.811879Z"
    },
    "papermill": {
     "duration": 0.087943,
     "end_time": "2022-04-26T05:19:52.539681",
     "exception": false,
     "start_time": "2022-04-26T05:19:52.451738",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def evalue_score_(clusters, filenames, df, class_):\n",
    "    clustered_target = clusters.labels_ #group id generated from cluster\n",
    "    clustered_df = pd.DataFrame({'name':filenames, 'clustered_target':clustered_target})\n",
    "    clustered_df = clustered_df[clustered_df['clustered_target'] != -1] # for the dbscan\n",
    "    \n",
    "    target2label = auto_labeling(df, clustered_df, class_)\n",
    "\n",
    "    clustered_df['predict_class'] = clustered_df['clustered_target'].map(target2label)\n",
    "    eval_df = pd.merge(clustered_df, df, on='name', how='left')\n",
    "    \n",
    "    df = pd.DataFrame({'y_Actual': eval_df['class'], \n",
    "                   'y_Predicted': eval_df['predict_class']},\n",
    "                  columns=['y_Actual','y_Predicted'])\n",
    "    \n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "3f72a4a8",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:52.710603Z",
     "iopub.status.busy": "2022-04-26T05:19:52.709486Z",
     "iopub.status.idle": "2022-04-26T05:19:52.711972Z",
     "shell.execute_reply": "2022-04-26T05:19:52.712452Z",
     "shell.execute_reply.started": "2022-04-26T05:02:11.823689Z"
    },
    "papermill": {
     "duration": 0.092323,
     "end_time": "2022-04-26T05:19:52.712633",
     "exception": false,
     "start_time": "2022-04-26T05:19:52.620310",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def prf_report(df, method = config['method'], results = []):\n",
    "\n",
    "    y_test = df['y_Actual']\n",
    "    y_pred = df['y_Predicted']\n",
    "    \n",
    "    report = classification_report(y_test, y_pred)\n",
    "    print('------------- classification report of {} ------------- '.format(method))\n",
    "    print(report)\n",
    "    print('------------------------ End ----------------------------')\n",
    "    \n",
    "    precision_score_ = precision_recall_fscore_support(y_test, y_pred)[0]\n",
    "    recall_score_ = precision_recall_fscore_support(y_test, y_pred)[1]\n",
    "    f1_score_ = precision_recall_fscore_support(y_test, y_pred)[2]\n",
    "\n",
    "    avg_precision = np.average(precision_score_)\n",
    "    avg_recall = np.average(recall_score_)\n",
    "    avg_f1 = np.average(f1_score_)\n",
    "    \n",
    "    results.append([method, precision_score_, recall_score_, f1_score_])\n",
    "\n",
    "    return results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "d9391545",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:52.873054Z",
     "iopub.status.busy": "2022-04-26T05:19:52.872035Z",
     "iopub.status.idle": "2022-04-26T05:19:52.880042Z",
     "shell.execute_reply": "2022-04-26T05:19:52.880525Z",
     "shell.execute_reply.started": "2022-04-26T05:02:13.751213Z"
    },
    "papermill": {
     "duration": 0.090063,
     "end_time": "2022-04-26T05:19:52.880713",
     "exception": false,
     "start_time": "2022-04-26T05:19:52.790650",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def cm_plot(confusion_matrix, method):\n",
    "    group_counts = [\"{0:0.0f}\".format(value) for value in confusion_matrix.to_numpy().flatten()]\n",
    "    group_percentages = [\"{0:.2f}\".format(value) for value in (confusion_matrix/np.sum(confusion_matrix)).to_numpy().flatten()]\n",
    "    labels = [f\"{v1}\\n{v2}\\n\" for v1, v2 in zip(group_counts, group_percentages)]\n",
    "    labels = np.asarray(labels).reshape(4,4)\n",
    "    ax = sns.heatmap(confusion_matrix, annot=labels, fmt='', cmap='Blues')\n",
    "    ax.set_xlabel('Predicted label')\n",
    "    ax.set_ylabel('True label')\n",
    "    plt.title('the Confusion Maxtrix of the {}'.format(config['method']))\n",
    "    plt.savefig('{}_confusion_matrix.png'.format(method))\n",
    "    return ax"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "3b7cfb95",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:19:53.044491Z",
     "iopub.status.busy": "2022-04-26T05:19:53.043479Z",
     "iopub.status.idle": "2022-04-26T05:20:22.552500Z",
     "shell.execute_reply": "2022-04-26T05:20:22.553181Z",
     "shell.execute_reply.started": "2022-04-26T05:02:13.769513Z"
    },
    "papermill": {
     "duration": 29.593843,
     "end_time": "2022-04-26T05:20:22.553453",
     "exception": false,
     "start_time": "2022-04-26T05:19:52.959610",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "the group 0 intersect with dict_keys(['B', 'I', 'S', 'U']) is [2, 12, 7, 167], the likely label is U\n",
      "the group 1 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 168, 0, 0], the likely label is I\n",
      "the group 2 intersect with dict_keys(['B', 'I', 'S', 'U']) is [177, 0, 19, 2], the likely label is B\n",
      "the group 3 intersect with dict_keys(['B', 'I', 'S', 'U']) is [1, 0, 154, 11], the likely label is S\n",
      "------------- classification report of MinibatchKMeans ------------- \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           B       0.89      0.98      0.94       180\n",
      "           I       1.00      0.93      0.97       180\n",
      "           S       0.93      0.86      0.89       180\n",
      "           U       0.89      0.93      0.91       180\n",
      "\n",
      "    accuracy                           0.93       720\n",
      "   macro avg       0.93      0.93      0.92       720\n",
      "weighted avg       0.93      0.93      0.92       720\n",
      "\n",
      "------------------------ End ----------------------------\n",
      "Predicted    B    I    S    U\n",
      "Actual                       \n",
      "B          177    0    1    2\n",
      "I            0  168    0   12\n",
      "S           19    0  154    7\n",
      "U            2    0   11  167\n",
      "the group 0 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 170, 0, 2], the likely label is I\n",
      "the group 1 intersect with dict_keys(['B', 'I', 'S', 'U']) is [177, 0, 20, 2], the likely label is B\n",
      "the group 2 intersect with dict_keys(['B', 'I', 'S', 'U']) is [2, 10, 0, 175], the likely label is U\n",
      "the group 3 intersect with dict_keys(['B', 'I', 'S', 'U']) is [1, 0, 160, 1], the likely label is S\n",
      "------------- classification report of Agglomerative ------------- \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           B       0.89      0.98      0.93       180\n",
      "           I       0.99      0.94      0.97       180\n",
      "           S       0.99      0.89      0.94       180\n",
      "           U       0.94      0.97      0.95       180\n",
      "\n",
      "    accuracy                           0.95       720\n",
      "   macro avg       0.95      0.95      0.95       720\n",
      "weighted avg       0.95      0.95      0.95       720\n",
      "\n",
      "------------------------ End ----------------------------\n",
      "Predicted    B    I    S    U\n",
      "Actual                       \n",
      "B          177    0    1    2\n",
      "I            0  170    0   10\n",
      "S           20    0  160    0\n",
      "U            2    2    1  175\n",
      "the group 0 intersect with dict_keys(['B', 'I', 'S', 'U']) is [3, 0, 153, 10], the likely label is S\n",
      "the group 1 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 169, 0, 1], the likely label is I\n",
      "the group 2 intersect with dict_keys(['B', 'I', 'S', 'U']) is [177, 0, 21, 2], the likely label is B\n",
      "the group 3 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 11, 6, 167], the likely label is U\n",
      "------------- classification report of Spectral ------------- \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           B       0.89      0.98      0.93       180\n",
      "           I       0.99      0.94      0.97       180\n",
      "           S       0.92      0.85      0.88       180\n",
      "           U       0.91      0.93      0.92       180\n",
      "\n",
      "    accuracy                           0.93       720\n",
      "   macro avg       0.93      0.93      0.92       720\n",
      "weighted avg       0.93      0.93      0.92       720\n",
      "\n",
      "------------------------ End ----------------------------\n",
      "Predicted    B    I    S    U\n",
      "Actual                       \n",
      "B          177    0    3    0\n",
      "I            0  169    0   11\n",
      "S           21    0  153    6\n",
      "U            2    1   10  167\n",
      "the group 0 intersect with dict_keys(['B', 'I', 'S', 'U']) is [178, 0, 23, 2], the likely label is B\n",
      "the group 1 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 142, 1, 0], the likely label is I\n",
      "the group 2 intersect with dict_keys(['B', 'I', 'S', 'U']) is [2, 38, 0, 177], the likely label is U\n",
      "the group 3 intersect with dict_keys(['B', 'I', 'S', 'U']) is [0, 0, 156, 1], the likely label is S\n",
      "------------- classification report of DBSCAN ------------- \n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           B       0.88      0.99      0.93       180\n",
      "           I       0.99      0.79      0.88       180\n",
      "           S       0.99      0.87      0.93       180\n",
      "           U       0.82      0.98      0.89       180\n",
      "\n",
      "    accuracy                           0.91       720\n",
      "   macro avg       0.92      0.91      0.91       720\n",
      "weighted avg       0.92      0.91      0.91       720\n",
      "\n",
      "------------------------ End ----------------------------\n",
      "Predicted    B    I    S    U\n",
      "Actual                       \n",
      "B          178    0    0    2\n",
      "I            0  142    0   38\n",
      "S           23    1  156    0\n",
      "U            2    0    1  177\n"
     ]
    }
   ],
   "source": [
    "results = []\n",
    "methods = ['MinibatchKMeans', 'Agglomerative', 'Spectral', 'DBSCAN']\n",
    "\n",
    "for method in methods:\n",
    "    if method == 'MinibatchKMeans':\n",
    "        clusters = generate_mkmclusters(feat)\n",
    "    elif method == 'Agglomerative':\n",
    "        clusters = generate_aggclusters(feat)\n",
    "    elif method == 'Spectral':\n",
    "        clusters = generate_speclusters(feat)\n",
    "    elif method == 'DBSCAN':\n",
    "        clusters = generate_hdbclusters(feat)\n",
    "        \n",
    "    df = evalue_score_(clusters, filenames, all_df, class_)\n",
    "    prf_report(df, method, results)\n",
    "    confusion_matrix = pd.crosstab(df['y_Actual'], df['y_Predicted'], rownames=['Actual'], colnames=['Predicted'])\n",
    "    print(confusion_matrix)\n",
    "\n",
    "#     for cm, ax in zip(confusion_matrix, axs):\n",
    "#         cm_img  = cm_plot(confusion_matrix, method)\n",
    "#         ax.set_axis_off()\n",
    "#         ax.imshow(cm_img)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "f3b19392",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:22.720660Z",
     "iopub.status.busy": "2022-04-26T05:20:22.719986Z",
     "iopub.status.idle": "2022-04-26T05:20:22.723502Z",
     "shell.execute_reply": "2022-04-26T05:20:22.724111Z",
     "shell.execute_reply.started": "2022-04-26T05:02:44.053085Z"
    },
    "papermill": {
     "duration": 0.087756,
     "end_time": "2022-04-26T05:20:22.724307",
     "exception": false,
     "start_time": "2022-04-26T05:20:22.636551",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# fig, axes = plt.subplots(nrows=2, ncols=2, figsize=(15,10))\n",
    "\n",
    "# for cls, ax in zip(classifiers, axes.flatten()):\n",
    "#     plot_confusion_matrix(cls, \n",
    "#                           X_test, \n",
    "#                           y_test, \n",
    "#                           ax=ax, \n",
    "#                           cmap='Blues',\n",
    "#                          display_labels=data.target_names)\n",
    "#     ax.title.set_text(type(cls).__name__)\n",
    "# plt.tight_layout()  \n",
    "# plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "e58e8bdc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:22.890372Z",
     "iopub.status.busy": "2022-04-26T05:20:22.889663Z",
     "iopub.status.idle": "2022-04-26T05:20:22.897676Z",
     "shell.execute_reply": "2022-04-26T05:20:22.898183Z",
     "shell.execute_reply.started": "2022-04-26T05:02:44.061047Z"
    },
    "papermill": {
     "duration": 0.094026,
     "end_time": "2022-04-26T05:20:22.898369",
     "exception": false,
     "start_time": "2022-04-26T05:20:22.804343",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def draw_confusionx(df, method):\n",
    "    confusion_matrix = pd.crosstab(df['y_Actual'], df['y_Predicted'], rownames=['Actual'], colnames=['Predicted'])\n",
    "    group_counts = [\"{0:0.0f}\".format(value) for value in confusion_matrix.to_numpy().flatten()]\n",
    "    group_percentages = [\"{0:.2f}\".format(value) for value in (confusion_matrix/np.sum(confusion_matrix)).to_numpy().flatten()]\n",
    "    labels = [f\"{v1}\\n{v2}\\n\" for v1, v2 in zip(group_counts, group_percentages)]\n",
    "    labels = np.asarray(labels).reshape(4,4)\n",
    "    ax = sns.heatmap(confusion_matrix, annot=labels, fmt='', cmap='Blues')\n",
    "    ax.set_xlabel('Predicted label')\n",
    "    ax.set_ylabel('True label')\n",
    "#     plt.title('the Confusion Maxtrix of the {}'.format(config['method']))\n",
    "    plt.savefig('{}_confusion_matrix.png'.format(method))\n",
    "    return plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "aee26732",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:23.063339Z",
     "iopub.status.busy": "2022-04-26T05:20:23.062562Z",
     "iopub.status.idle": "2022-04-26T05:20:23.555388Z",
     "shell.execute_reply": "2022-04-26T05:20:23.555927Z",
     "shell.execute_reply.started": "2022-04-26T05:02:44.076218Z"
    },
    "papermill": {
     "duration": 0.577213,
     "end_time": "2022-04-26T05:20:23.556120",
     "exception": false,
     "start_time": "2022-04-26T05:20:22.978907",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x720 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt = draw_confusionx(df, 'MinibatchKMeans')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "277774f6",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:23.722704Z",
     "iopub.status.busy": "2022-04-26T05:20:23.721985Z",
     "iopub.status.idle": "2022-04-26T05:20:23.739468Z",
     "shell.execute_reply": "2022-04-26T05:20:23.738770Z",
     "shell.execute_reply.started": "2022-04-26T05:02:44.609107Z"
    },
    "papermill": {
     "duration": 0.101561,
     "end_time": "2022-04-26T05:20:23.739628",
     "exception": false,
     "start_time": "2022-04-26T05:20:23.638067",
     "status": "completed"
    },
    "tags": []
   },
   "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>method</th>\n",
       "      <th>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>f1</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MinibatchKMeans</td>\n",
       "      <td>[0.8939393939393939, 1.0, 0.927710843373494, 0...</td>\n",
       "      <td>[0.9833333333333333, 0.9333333333333333, 0.855...</td>\n",
       "      <td>[0.9365079365079364, 0.9655172413793104, 0.890...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Agglomerative</td>\n",
       "      <td>[0.8894472361809045, 0.9883720930232558, 0.987...</td>\n",
       "      <td>[0.9833333333333333, 0.9444444444444444, 0.888...</td>\n",
       "      <td>[0.9340369393139842, 0.9659090909090908, 0.935...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Spectral</td>\n",
       "      <td>[0.885, 0.9941176470588236, 0.9216867469879518...</td>\n",
       "      <td>[0.9833333333333333, 0.9388888888888889, 0.85,...</td>\n",
       "      <td>[0.9315789473684211, 0.9657142857142857, 0.884...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>DBSCAN</td>\n",
       "      <td>[0.8768472906403941, 0.993006993006993, 0.9936...</td>\n",
       "      <td>[0.9888888888888889, 0.7888888888888889, 0.866...</td>\n",
       "      <td>[0.9295039164490861, 0.8792569659442725, 0.925...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            method                                          precision  \\\n",
       "0  MinibatchKMeans  [0.8939393939393939, 1.0, 0.927710843373494, 0...   \n",
       "1    Agglomerative  [0.8894472361809045, 0.9883720930232558, 0.987...   \n",
       "2         Spectral  [0.885, 0.9941176470588236, 0.9216867469879518...   \n",
       "3           DBSCAN  [0.8768472906403941, 0.993006993006993, 0.9936...   \n",
       "\n",
       "                                              recall  \\\n",
       "0  [0.9833333333333333, 0.9333333333333333, 0.855...   \n",
       "1  [0.9833333333333333, 0.9444444444444444, 0.888...   \n",
       "2  [0.9833333333333333, 0.9388888888888889, 0.85,...   \n",
       "3  [0.9888888888888889, 0.7888888888888889, 0.866...   \n",
       "\n",
       "                                                  f1  \n",
       "0  [0.9365079365079364, 0.9655172413793104, 0.890...  \n",
       "1  [0.9340369393139842, 0.9659090909090908, 0.935...  \n",
       "2  [0.9315789473684211, 0.9657142857142857, 0.884...  \n",
       "3  [0.9295039164490861, 0.8792569659442725, 0.925...  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_df = pd.DataFrame(results, columns=['method', 'precision', 'recall', 'f1'])\n",
    "result_df"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf2253f3",
   "metadata": {
    "papermill": {
     "duration": 0.085201,
     "end_time": "2022-04-26T05:20:23.907062",
     "exception": false,
     "start_time": "2022-04-26T05:20:23.821861",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "#### [x]TODO clouds of the image name "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1539b7af",
   "metadata": {
    "papermill": {
     "duration": 0.083156,
     "end_time": "2022-04-26T05:20:24.074115",
     "exception": false,
     "start_time": "2022-04-26T05:20:23.990959",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "### Draw Confusion mx"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "7c1f34a2",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:24.254602Z",
     "iopub.status.busy": "2022-04-26T05:20:24.253495Z",
     "iopub.status.idle": "2022-04-26T05:20:24.256446Z",
     "shell.execute_reply": "2022-04-26T05:20:24.255907Z",
     "shell.execute_reply.started": "2022-04-26T05:02:44.62959Z"
    },
    "papermill": {
     "duration": 0.100157,
     "end_time": "2022-04-26T05:20:24.256597",
     "exception": false,
     "start_time": "2022-04-26T05:20:24.156440",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def draw_roc_auc(df, method = config['method']):\n",
    "    # # Compute ROC curve and ROC area for each class\n",
    "\n",
    "    le = LabelEncoder()\n",
    "    df['y_Actual_enc']= le.fit_transform(df['y_Actual'])\n",
    "    df['y_Predicted_enc']= le.fit_transform(df['y_Predicted'])\n",
    "    y_test = to_categorical(df['y_Actual_enc'], num_classes = len(le.classes_),dtype='int64')\n",
    "    y_pred = to_categorical(df['y_Predicted_enc'], num_classes = len(le.classes_),dtype='int64')\n",
    "    le.classes_, le.classes_[0]\n",
    "\n",
    "    fpr = dict()\n",
    "    tpr = dict()\n",
    "\n",
    "    roc_auc = dict()\n",
    "    for i in range(len(le.classes_)):\n",
    "        fpr[i], tpr[i], _ = roc_curve(y_test[:, i], y_pred[:, i])\n",
    "        roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "        \n",
    "    # Compute micro-average ROC curve and ROC area\n",
    "    fpr[\"micro\"], tpr[\"micro\"], _ = roc_curve(y_test.ravel(), y_pred.ravel())\n",
    "    roc_auc[\"micro\"] = auc(fpr[\"micro\"], tpr[\"micro\"])\n",
    "\n",
    "    plt.figure()\n",
    "    plt.plot(fpr[\"micro\"], tpr[\"micro\"],\n",
    "            label='micro-average ROC curve (AUC = {0:0.2f})'\n",
    "                ''.format(roc_auc[\"micro\"]))\n",
    "    for i in range(len(le.classes_)):\n",
    "        plt.plot(fpr[i], tpr[i], label='{0} (AUC = {1:0.2f})'\n",
    "                                    ''.format(le.classes_[i], roc_auc[i]))\n",
    "    plt.plot([0, 1], [0, 1], 'k--')\n",
    "    plt.xlim([0.0, 1.0])\n",
    "    plt.ylim([0.0, 1.05])\n",
    "    plt.xlabel('False Positive Rate')\n",
    "    plt.ylabel('True Positive Rate')\n",
    "    plt.title('ROC/AUC of the {}'.format(method))\n",
    "    plt.legend(loc=\"lower right\")\n",
    "    plt.savefig('{}_roc_curve.png'.format(method))\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "1f60649b",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:24.426689Z",
     "iopub.status.busy": "2022-04-26T05:20:24.425988Z",
     "iopub.status.idle": "2022-04-26T05:20:24.911237Z",
     "shell.execute_reply": "2022-04-26T05:20:24.910535Z",
     "shell.execute_reply.started": "2022-04-26T05:02:44.648455Z"
    },
    "papermill": {
     "duration": 0.570608,
     "end_time": "2022-04-26T05:20:24.911515",
     "exception": false,
     "start_time": "2022-04-26T05:20:24.340907",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "draw_roc_auc(df)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09c1838f",
   "metadata": {
    "papermill": {
     "duration": 0.086453,
     "end_time": "2022-04-26T05:20:25.084094",
     "exception": false,
     "start_time": "2022-04-26T05:20:24.997641",
     "status": "completed"
    },
    "tags": []
   },
   "source": [
    "## Boxplot Recall"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "b585cd7f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:25.259147Z",
     "iopub.status.busy": "2022-04-26T05:20:25.258432Z",
     "iopub.status.idle": "2022-04-26T05:20:25.273766Z",
     "shell.execute_reply": "2022-04-26T05:20:25.274344Z",
     "shell.execute_reply.started": "2022-04-26T05:02:45.169576Z"
    },
    "papermill": {
     "duration": 0.103089,
     "end_time": "2022-04-26T05:20:25.274526",
     "exception": false,
     "start_time": "2022-04-26T05:20:25.171437",
     "status": "completed"
    },
    "tags": []
   },
   "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>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>method</th>\n",
       "      <td>MinibatchKMeans</td>\n",
       "      <td>Agglomerative</td>\n",
       "      <td>Spectral</td>\n",
       "      <td>DBSCAN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>precision</th>\n",
       "      <td>[0.8939393939393939, 1.0, 0.927710843373494, 0...</td>\n",
       "      <td>[0.8894472361809045, 0.9883720930232558, 0.987...</td>\n",
       "      <td>[0.885, 0.9941176470588236, 0.9216867469879518...</td>\n",
       "      <td>[0.8768472906403941, 0.993006993006993, 0.9936...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>recall</th>\n",
       "      <td>[0.9833333333333333, 0.9333333333333333, 0.855...</td>\n",
       "      <td>[0.9833333333333333, 0.9444444444444444, 0.888...</td>\n",
       "      <td>[0.9833333333333333, 0.9388888888888889, 0.85,...</td>\n",
       "      <td>[0.9888888888888889, 0.7888888888888889, 0.866...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>f1</th>\n",
       "      <td>[0.9365079365079364, 0.9655172413793104, 0.890...</td>\n",
       "      <td>[0.9340369393139842, 0.9659090909090908, 0.935...</td>\n",
       "      <td>[0.9315789473684211, 0.9657142857142857, 0.884...</td>\n",
       "      <td>[0.9295039164490861, 0.8792569659442725, 0.925...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                           0  \\\n",
       "method                                       MinibatchKMeans   \n",
       "precision  [0.8939393939393939, 1.0, 0.927710843373494, 0...   \n",
       "recall     [0.9833333333333333, 0.9333333333333333, 0.855...   \n",
       "f1         [0.9365079365079364, 0.9655172413793104, 0.890...   \n",
       "\n",
       "                                                           1  \\\n",
       "method                                         Agglomerative   \n",
       "precision  [0.8894472361809045, 0.9883720930232558, 0.987...   \n",
       "recall     [0.9833333333333333, 0.9444444444444444, 0.888...   \n",
       "f1         [0.9340369393139842, 0.9659090909090908, 0.935...   \n",
       "\n",
       "                                                           2  \\\n",
       "method                                              Spectral   \n",
       "precision  [0.885, 0.9941176470588236, 0.9216867469879518...   \n",
       "recall     [0.9833333333333333, 0.9388888888888889, 0.85,...   \n",
       "f1         [0.9315789473684211, 0.9657142857142857, 0.884...   \n",
       "\n",
       "                                                           3  \n",
       "method                                                DBSCAN  \n",
       "precision  [0.8768472906403941, 0.993006993006993, 0.9936...  \n",
       "recall     [0.9888888888888889, 0.7888888888888889, 0.866...  \n",
       "f1         [0.9295039164490861, 0.8792569659442725, 0.925...  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "result_df.T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "1ffb6102",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:25.448941Z",
     "iopub.status.busy": "2022-04-26T05:20:25.448254Z",
     "iopub.status.idle": "2022-04-26T05:20:25.456704Z",
     "shell.execute_reply": "2022-04-26T05:20:25.457233Z",
     "shell.execute_reply.started": "2022-04-26T05:02:45.196566Z"
    },
    "papermill": {
     "duration": 0.09697,
     "end_time": "2022-04-26T05:20:25.457446",
     "exception": false,
     "start_time": "2022-04-26T05:20:25.360476",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def extract_recall(result_df):\n",
    "    ## SOTA\n",
    "    recall_results = []\n",
    "\n",
    "    recall_results.append(['Moccia, 2018', 0.83, 0.91, 0.62, 0.85])\n",
    "    recall_results.append(['Galdran, 2019', 0.94, 1, 0.91, 0.94])\n",
    "    recall_results.append(['Patrini, 2020', 0.96,0.97,0.88,0.93])\n",
    "    \n",
    "    for i in range(result_df.shape[0]):\n",
    "        eval_method = []\n",
    "        eval_method.extend([result_df['method'].astype('str').iloc[i]])\n",
    "        eval_method.extend(result_df['recall'].iloc[i].tolist())\n",
    "        recall_results.append(eval_method)\n",
    "\n",
    "    recall_df = pd.DataFrame(np.array(recall_results), columns=['method', 'B', 'I', 'S', 'U'])\n",
    "    data = recall_df.iloc[:,1:].T.to_numpy().astype('float')\n",
    "    labels = recall_df.T.iloc[0].astype('str').tolist()\n",
    "    return data, labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "63e3668f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:25.633340Z",
     "iopub.status.busy": "2022-04-26T05:20:25.632663Z",
     "iopub.status.idle": "2022-04-26T05:20:25.639161Z",
     "shell.execute_reply": "2022-04-26T05:20:25.639742Z",
     "shell.execute_reply.started": "2022-04-26T05:02:45.207829Z"
    },
    "papermill": {
     "duration": 0.096049,
     "end_time": "2022-04-26T05:20:25.639929",
     "exception": false,
     "start_time": "2022-04-26T05:20:25.543880",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "data, labels = extract_recall(result_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "fd024dd7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:25.816632Z",
     "iopub.status.busy": "2022-04-26T05:20:25.815949Z",
     "iopub.status.idle": "2022-04-26T05:20:27.892957Z",
     "shell.execute_reply": "2022-04-26T05:20:27.893846Z",
     "shell.execute_reply.started": "2022-04-26T05:02:45.229169Z"
    },
    "papermill": {
     "duration": 2.167143,
     "end_time": "2022-04-26T05:20:27.894133",
     "exception": false,
     "start_time": "2022-04-26T05:20:25.726990",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "import matplotlib as mpl\n",
    "mpl.rcParams['axes.linewidth'] = 1.5 #set the value globally\n",
    "tips = sns.load_dataset(\"tips\")\n",
    "\n",
    "def boxplot_classic(data, labels):\n",
    "    # data = np.random.lognormal(size=(37, 4))\n",
    "    fig, ax1 = plt.subplots()\n",
    "\n",
    "    boxprops = dict(linewidth=3,  color='blue')\n",
    "    medianprops = dict(linewidth=3, color='red')\n",
    "    whiskerprops = dict(linewidth=3, color='black')\n",
    "    capprops = dict(linewidth=3, color='black')\n",
    "    \n",
    "    with plt.style.context('classic'):\n",
    "        ax1.boxplot(data, showfliers=False,\n",
    "                    boxprops = boxprops,\n",
    "                    labels = labels, \n",
    "                    medianprops = medianprops,\n",
    "                    whiskerprops = whiskerprops,\n",
    "                    capprops=capprops)\n",
    "        \n",
    "        ax1.set_title('Recall Comparison of Different methods')\n",
    "        ax1.set_xlabel('methods', fontweight ='bold')\n",
    "        ax1.set_ylabel('recall', fontweight ='bold')\n",
    "        ax1.grid(True, linestyle='-.')\n",
    "\n",
    "        ax1.tick_params(direction='in',\n",
    "                        length=6,\n",
    "                        width=2,\n",
    "                        colors='black',\n",
    "                        labelsize='small', #medium, large\n",
    "                        grid_color='grey',\n",
    "                        grid_alpha=0.5)\n",
    "        \n",
    "    fig.savefig(\"boxplot.pdf\", bbox_inches = 'tight',pad_inches = 0)\n",
    "    fig.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "3d030c71",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:28.079881Z",
     "iopub.status.busy": "2022-04-26T05:20:28.079096Z",
     "iopub.status.idle": "2022-04-26T05:20:28.581840Z",
     "shell.execute_reply": "2022-04-26T05:20:28.582402Z",
     "shell.execute_reply.started": "2022-04-26T05:02:46.131523Z"
    },
    "papermill": {
     "duration": 0.601469,
     "end_time": "2022-04-26T05:20:28.582620",
     "exception": false,
     "start_time": "2022-04-26T05:20:27.981151",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "boxplot_classic(data, labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "dd07dae5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:28.769524Z",
     "iopub.status.busy": "2022-04-26T05:20:28.768626Z",
     "iopub.status.idle": "2022-04-26T05:20:28.772079Z",
     "shell.execute_reply": "2022-04-26T05:20:28.772621Z",
     "shell.execute_reply.started": "2022-04-26T05:02:46.669304Z"
    },
    "papermill": {
     "duration": 0.100464,
     "end_time": "2022-04-26T05:20:28.772810",
     "exception": false,
     "start_time": "2022-04-26T05:20:28.672346",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def convert_pvalue_to_asterisks(pvalue):\n",
    "    if pvalue <= 0.0001:\n",
    "        return \"****\"\n",
    "    elif pvalue <= 0.001:\n",
    "        return \"***\"\n",
    "    elif pvalue <= 0.01:\n",
    "        return \"**\"\n",
    "    elif pvalue <= 0.05:\n",
    "        return \"*\"\n",
    "    return \"ns\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "eb8133f5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:28.961311Z",
     "iopub.status.busy": "2022-04-26T05:20:28.960644Z",
     "iopub.status.idle": "2022-04-26T05:20:28.965805Z",
     "shell.execute_reply": "2022-04-26T05:20:28.966435Z",
     "shell.execute_reply.started": "2022-04-26T05:02:46.680969Z"
    },
    "papermill": {
     "duration": 0.103479,
     "end_time": "2022-04-26T05:20:28.966610",
     "exception": false,
     "start_time": "2022-04-26T05:20:28.863131",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.98333333, 0.94444444, 0.88888889, 0.97222222])"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[:,4]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "7d898e68",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:29.155738Z",
     "iopub.status.busy": "2022-04-26T05:20:29.151521Z",
     "iopub.status.idle": "2022-04-26T05:20:29.160691Z",
     "shell.execute_reply": "2022-04-26T05:20:29.161173Z",
     "shell.execute_reply.started": "2022-04-26T05:02:46.694495Z"
    },
    "papermill": {
     "duration": 0.10681,
     "end_time": "2022-04-26T05:20:29.161409",
     "exception": false,
     "start_time": "2022-04-26T05:20:29.054599",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "ns\n",
      "ns\n",
      "ns\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "[0.05714285714285714, 0.8845494388529623, 0.4857142857142857]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy.stats import mannwhitneyu\n",
    "\n",
    "#only compare with Moccia, 2018\n",
    "\n",
    "pvalues = []\n",
    "for x in range(3):\n",
    "#     stat, pvalue = scipy.stats.ttest_ind(data[:,x].tolist(), data[:,3].tolist())\n",
    "    s, p = mannwhitneyu(data[:,x].tolist(), data[:,4].tolist())\n",
    "    pvalues.append(p)\n",
    "    print(convert_pvalue_to_asterisks(p))\n",
    "pvalues"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "68cc4148",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2022-04-26T05:20:29.348905Z",
     "iopub.status.busy": "2022-04-26T05:20:29.347975Z",
     "iopub.status.idle": "2022-04-26T05:20:29.353405Z",
     "shell.execute_reply": "2022-04-26T05:20:29.352858Z",
     "shell.execute_reply.started": "2022-04-26T05:02:46.714844Z"
    },
    "papermill": {
     "duration": 0.102066,
     "end_time": "2022-04-26T05:20:29.353548",
     "exception": false,
     "start_time": "2022-04-26T05:20:29.251482",
     "status": "completed"
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Statistics=0.000, p=0.125\n",
      "Same distribution (fail to reject H0)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'ns'"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from scipy.stats import wilcoxon\n",
    "\n",
    "data1 = data[:,0].tolist()\n",
    "data2 = data[:,4].tolist()\n",
    "# stat, pvalue = scipy.stats.ttest_ind(data1, data2)\n",
    "# stat, pvalue = mannwhitneyu(data1, data2)\n",
    "stat, pvalue = wilcoxon(data1, data2)\n",
    "\n",
    "print('Statistics=%.3f, p=%.3f' % (stat, pvalue))\n",
    "\n",
    "alpha = 0.05\n",
    "if pvalue > alpha:\n",
    "    print('Same distribution (fail to reject H0)')\n",
    "else:\n",
    "    print('Different distribution (reject H0)')\n",
    "\n",
    "\n",
    "convert_pvalue_to_asterisks(pvalue)\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.7.12"
  },
  "papermill": {
   "default_parameters": {},
   "duration": 97.786088,
   "end_time": "2022-04-26T05:20:32.897985",
   "environment_variables": {},
   "exception": null,
   "input_path": "__notebook__.ipynb",
   "output_path": "__notebook__.ipynb",
   "parameters": {},
   "start_time": "2022-04-26T05:18:55.111897",
   "version": "2.3.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}