2. QSAR_ANN.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 85,
   "id": "f4d8480c-bfda-45b5-8f29-37fe5c4e314a",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from torch.utils.data import Dataset\n",
    "from torch.utils.data import DataLoader\n",
    "from torch.utils.data import TensorDataset\n",
    "import os \n",
    "import sys\n",
    "sys.path.append('./Codebase/ann')\n",
    "from Featurizer import featurizer\n",
    "from TargetNormalize import target_bin"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc29e737-1f54-43de-8f22-7ed9330ad22c",
   "metadata": {},
   "source": [
    "# 1. Fingerprints Calculation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "id": "2b612f3d-c5dd-4509-963a-5eac5944b9ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_0 = pd.read_csv('./Data/QSAR/Data_qsar.csv', index_col=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "515dbde8-4673-414c-83c9-6f438a765cf1",
   "metadata": {},
   "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>ID</th>\n",
       "      <th>pIC50</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>...</th>\n",
       "      <th>2038</th>\n",
       "      <th>2039</th>\n",
       "      <th>2040</th>\n",
       "      <th>2041</th>\n",
       "      <th>2042</th>\n",
       "      <th>2043</th>\n",
       "      <th>2044</th>\n",
       "      <th>2045</th>\n",
       "      <th>2046</th>\n",
       "      <th>2047</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>10.193820</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>8.197705</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>7.980178</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>9.225483</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>9.349692</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2050 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID         pIC50  0      1      2      3      4      5      6      7      \\\n",
       "0      0  10.193820      0      0      0      0      1      0      0      0   \n",
       "1      1   8.197705      0      0      0      0      0      0      0      0   \n",
       "2      2   7.980178      0      0      0      0      0      0      0      0   \n",
       "3      3   9.225483      0      0      0      0      0      0      0      0   \n",
       "4      4   9.349692      0      0      0      0      0      0      0      0   \n",
       "\n",
       "   ...  2038   2039   2040   2041   2042   2043   2044   2045   2046   2047   \n",
       "0  ...      0      0      0      0      0      0      0      1      0      0  \n",
       "1  ...      0      0      0      0      0      0      0      0      0      0  \n",
       "2  ...      0      0      0      0      0      0      0      0      0      0  \n",
       "3  ...      0      0      0      0      0      0      0      0      0      0  \n",
       "4  ...      0      0      0      0      0      0      0      0      0      0  \n",
       "\n",
       "[5 rows x 2050 columns]"
      ]
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature= featurizer(data=df_0, ID_col='ID', smiles_col='Smiles', type_fp = 'secfp', active_col = 'pIC50')\n",
    "df=feature.fit()\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c5e1b799-288e-4a22-aa84-4cc76d49f4d2",
   "metadata": {},
   "source": [
    "# 2. Target normalization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "id": "569bfd08-bc5f-4c77-b164-5da4378a06d0",
   "metadata": {},
   "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>ID</th>\n",
       "      <th>pIC50</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>5</th>\n",
       "      <th>6</th>\n",
       "      <th>7</th>\n",
       "      <th>...</th>\n",
       "      <th>2038</th>\n",
       "      <th>2039</th>\n",
       "      <th>2040</th>\n",
       "      <th>2041</th>\n",
       "      <th>2042</th>\n",
       "      <th>2043</th>\n",
       "      <th>2044</th>\n",
       "      <th>2045</th>\n",
       "      <th>2046</th>\n",
       "      <th>2047</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 2050 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   ID     pIC50  0      1      2      3      4      5      6      7      ...  \\\n",
       "0      0      1      0      0      0      0      1      0      0      0  ...   \n",
       "1      1      0      0      0      0      0      0      0      0      0  ...   \n",
       "2      2      0      0      0      0      0      0      0      0      0  ...   \n",
       "3      3      0      0      0      0      0      0      0      0      0  ...   \n",
       "4      4      0      0      0      0      0      0      0      0      0  ...   \n",
       "\n",
       "   2038   2039   2040   2041   2042   2043   2044   2045   2046   2047   \n",
       "0      0      0      0      0      0      0      0      1      0      0  \n",
       "1      0      0      0      0      0      0      0      0      0      0  \n",
       "2      0      0      0      0      0      0      0      0      0      0  \n",
       "3      0      0      0      0      0      0      0      0      0      0  \n",
       "4      0      0      0      0      0      0      0      0      0      0  \n",
       "\n",
       "[5 rows x 2050 columns]"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = target_bin(data = df, thresh = 10, Y_name = \"pIC50\")\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dc560a4a-72bb-4470-abc0-df9ace4a7bc2",
   "metadata": {},
   "source": [
    "# 3. Data preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "3bcb87cf-b341-4cb7-8ed2-8448cc26227e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X = df.drop(['ID','pIC50'],axis = 1)\n",
    "y = df[\"pIC50\"]\n",
    "#Split Data train, Data_test, Data_validation\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2,\n",
    "                                                                random_state =42, stratify=y)\n",
    "X_train, X_val, y_train, y_val = train_test_split(X_train, y_train, test_size = 0.1,\n",
    "                                                                random_state =42, stratify=y_train)\n",
    "\n",
    "#setting device\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(device)\n",
    "# change numpy to tensor gpu\n",
    "X_train = torch.tensor(X_train.values , device=device).float()\n",
    "X_val = torch.tensor(X_val.values , device=device).float()\n",
    "X_test = torch.tensor(X_test.values, device=device).float()\n",
    "\n",
    "y_train = torch.tensor(y_train.values , device=device).float()\n",
    "y_val = torch.tensor(y_val.values, device=device).float()\n",
    "y_test = torch.tensor(y_test.values, device=device).float()\n",
    "\n",
    "# convert into dataloader\n",
    "train_dataset = TensorDataset(X_train, y_train)\n",
    "val_dataset = TensorDataset(X_val, y_val)\n",
    "test_dataset = TensorDataset(X_test, y_test)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(dataset=train_dataset,\n",
    "                                          batch_size=64,\n",
    "                                          shuffle=True)\n",
    "valid_loader = torch.utils.data.DataLoader(dataset=val_dataset,\n",
    "                                          batch_size=64,\n",
    "                                          shuffle=True)\n",
    "test_loader = torch.utils.data.DataLoader(dataset=test_dataset,\n",
    "                                          batch_size=64,\n",
    "                                          shuffle=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "af9fa453-5dec-402d-8570-31c0ea083281",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "id": "626c371d-53d7-42a5-b642-485cb9928cbd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cuda:0\n"
     ]
    }
   ],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X = df.drop(['ID','pIC50'],axis = 1)\n",
    "y = df[\"pIC50\"]\n",
    "#Split Data train, Data_test, Data_validation\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2,\n",
    "                                                                random_state =42, stratify=y)\n",
    "\n",
    "\n",
    "#setting device\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")\n",
    "print(device)\n",
    "# change numpy to tensor gpu\n",
    "X_train = torch.tensor(X_train.values , device=device).float()\n",
    "\n",
    "X_test = torch.tensor(X_test.values, device=device).float()\n",
    "\n",
    "y_train = torch.tensor(y_train.values , device=device).float()\n",
    "\n",
    "y_test = torch.tensor(y_test.values, device=device).float()\n",
    "\n",
    "# convert into dataloader\n",
    "train_dataset = TensorDataset(X_train, y_train)\n",
    "\n",
    "test_dataset = TensorDataset(X_test, y_test)\n",
    "\n",
    "train_loader = torch.utils.data.DataLoader(dataset=train_dataset,\n",
    "                                          batch_size=32,\n",
    "                                          shuffle=True)\n",
    "\n",
    "test_loader = torch.utils.data.DataLoader(dataset=test_dataset,\n",
    "                                          batch_size=32,\n",
    "                                          shuffle=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a6418abe-a5b4-4e39-848a-a7d87494492a",
   "metadata": {},
   "source": [
    "# 4. Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "faaa543e-6781-4867-826e-67e535168201",
   "metadata": {},
   "outputs": [],
   "source": [
    "from ModelArchitecture import Net\n",
    "from annpipeline import model_pipeline\n",
    "\n",
    "from sklearn.metrics import f1_score, average_precision_score, classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "40562402-fa7e-49bc-89d4-9bbf541aebb2",
   "metadata": {
    "jupyter": {
     "source_hidden": true
    },
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch: 5/170..  Training Loss: 0.542..  validation Loss: 0.504..  validation f1_score: 0.732..  validation average precision: 0.824.. \n",
      "Epoch: 10/170..  Training Loss: 0.411..  validation Loss: 0.397..  validation f1_score: 0.756..  validation average precision: 0.843.. \n",
      "Epoch: 15/170..  Training Loss: 0.384..  validation Loss: 0.403..  validation f1_score: 0.731..  validation average precision: 0.848.. \n",
      "Epoch: 20/170..  Training Loss: 0.379..  validation Loss: 0.383..  validation f1_score: 0.771..  validation average precision: 0.852.. \n",
      "Epoch: 25/170..  Training Loss: 0.366..  validation Loss: 0.375..  validation f1_score: 0.769..  validation average precision: 0.842.. \n",
      "Epoch: 30/170..  Training Loss: 0.367..  validation Loss: 0.373..  validation f1_score: 0.772..  validation average precision: 0.853.. \n",
      "Epoch: 35/170..  Training Loss: 0.365..  validation Loss: 0.363..  validation f1_score: 0.780..  validation average precision: 0.860.. \n",
      "Epoch: 40/170..  Training Loss: 0.360..  validation Loss: 0.372..  validation f1_score: 0.765..  validation average precision: 0.852.. \n",
      "Epoch: 45/170..  Training Loss: 0.354..  validation Loss: 0.363..  validation f1_score: 0.788..  validation average precision: 0.857.. \n",
      "Epoch: 50/170..  Training Loss: 0.358..  validation Loss: 0.367..  validation f1_score: 0.770..  validation average precision: 0.853.. \n",
      "Epoch: 55/170..  Training Loss: 0.354..  validation Loss: 0.367..  validation f1_score: 0.777..  validation average precision: 0.856.. \n",
      "Epoch: 60/170..  Training Loss: 0.351..  validation Loss: 0.364..  validation f1_score: 0.775..  validation average precision: 0.852.. \n",
      "Epoch: 65/170..  Training Loss: 0.346..  validation Loss: 0.361..  validation f1_score: 0.779..  validation average precision: 0.856.. \n",
      "Epoch: 70/170..  Training Loss: 0.354..  validation Loss: 0.363..  validation f1_score: 0.774..  validation average precision: 0.857.. \n",
      "Epoch: 75/170..  Training Loss: 0.349..  validation Loss: 0.363..  validation f1_score: 0.779..  validation average precision: 0.856.. \n",
      "Epoch: 80/170..  Training Loss: 0.350..  validation Loss: 0.362..  validation f1_score: 0.773..  validation average precision: 0.859.. \n",
      "Epoch: 85/170..  Training Loss: 0.352..  validation Loss: 0.361..  validation f1_score: 0.779..  validation average precision: 0.859.. \n",
      "Epoch: 90/170..  Training Loss: 0.347..  validation Loss: 0.390..  validation f1_score: 0.758..  validation average precision: 0.856.. \n",
      "Epoch: 95/170..  Training Loss: 0.344..  validation Loss: 0.360..  validation f1_score: 0.771..  validation average precision: 0.860.. \n",
      "Epoch: 100/170..  Training Loss: 0.349..  validation Loss: 0.365..  validation f1_score: 0.778..  validation average precision: 0.852.. \n",
      "Epoch: 105/170..  Training Loss: 0.340..  validation Loss: 0.369..  validation f1_score: 0.784..  validation average precision: 0.851.. \n",
      "Epoch: 110/170..  Training Loss: 0.349..  validation Loss: 0.363..  validation f1_score: 0.781..  validation average precision: 0.855.. \n",
      "Epoch: 115/170..  Training Loss: 0.334..  validation Loss: 0.363..  validation f1_score: 0.790..  validation average precision: 0.855.. \n",
      "Epoch: 120/170..  Training Loss: 0.345..  validation Loss: 0.359..  validation f1_score: 0.782..  validation average precision: 0.856.. \n",
      "Epoch: 125/170..  Training Loss: 0.339..  validation Loss: 0.365..  validation f1_score: 0.781..  validation average precision: 0.860.. \n",
      "Epoch: 130/170..  Training Loss: 0.338..  validation Loss: 0.356..  validation f1_score: 0.784..  validation average precision: 0.858.. \n",
      "Epoch: 135/170..  Training Loss: 0.345..  validation Loss: 0.357..  validation f1_score: 0.789..  validation average precision: 0.857.. \n",
      "Epoch: 140/170..  Training Loss: 0.334..  validation Loss: 0.357..  validation f1_score: 0.777..  validation average precision: 0.857.. \n",
      "Epoch: 145/170..  Training Loss: 0.337..  validation Loss: 0.363..  validation f1_score: 0.772..  validation average precision: 0.855.. \n",
      "Epoch: 150/170..  Training Loss: 0.336..  validation Loss: 0.355..  validation f1_score: 0.789..  validation average precision: 0.856.. \n",
      "Epoch: 155/170..  Training Loss: 0.332..  validation Loss: 0.372..  validation f1_score: 0.770..  validation average precision: 0.845.. \n",
      "Epoch: 160/170..  Training Loss: 0.346..  validation Loss: 0.368..  validation f1_score: 0.768..  validation average precision: 0.857.. \n",
      "Epoch: 165/170..  Training Loss: 0.333..  validation Loss: 0.360..  validation f1_score: 0.784..  validation average precision: 0.853.. \n",
      "Epoch: 170/170..  Training Loss: 0.341..  validation Loss: 0.359..  validation f1_score: 0.790..  validation average precision: 0.854.. \n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saving...\n",
      "Complete the process...\n"
     ]
    }
   ],
   "source": [
    "net = Net(in_features = 2048, hidden_nodes_1 =2048,hidden_nodes_2 =1024, hidden_nodes_3=256, drop_out=0.5)\n",
    "pip = model_pipeline(device = device, seed = 42,\n",
    "                     save_dir = '.')\n",
    "pip.fit(\n",
    "    data_train = train_loader,data_valid = test_loader,epochs =170, show_progress = True,\n",
    "    model = net,lr=0.0001, weight_decay=0.05, factor=0.9, patience=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "946e0899-b593-4405-a465-82fec4b18b93",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.86      0.93      0.90       263\n",
      "         1.0       0.86      0.73      0.79       146\n",
      "\n",
      "    accuracy                           0.86       409\n",
      "   macro avg       0.86      0.83      0.84       409\n",
      "weighted avg       0.86      0.86      0.86       409\n",
      "\n",
      "Average precision score 0.853623637192412\n",
      "F1 score 0.7896678966789669\n",
      "Recall score 0.7328767123287672\n"
     ]
    }
   ],
   "source": [
    "criterion = torch.nn.BCELoss()\n",
    "checkpoint = torch.load('./ANN_model.pth')\n",
    "model  =  Net(in_features = 2048, hidden_nodes_1 =2048,hidden_nodes_2 =1024, hidden_nodes_3=256, drop_out=0.5)\n",
    "\n",
    "y_pre , y_true = pip.predict(test_loader, model =model, checkpoint=checkpoint,)\n",
    "print(classification_report(y_true, y_pre))\n",
    "\n",
    "y_proba, y_true = pip.predict_proba(test_loader, model =model, checkpoint=checkpoint,\n",
    "                                   criterion=criterion)\n",
    "\n",
    "print(\"Average precision score\", average_precision_score(y_true ,y_proba))\n",
    "print(\"F1 score\", f1_score(y_true ,y_pre))\n",
    "print(\"Recall score\", recall_score(y_true ,y_pre))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "b611dcd4-ed11-4b1d-bad8-a2ded97d2830",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best Threshold=0.491328, F-Score=0.799\n"
     ]
    }
   ],
   "source": [
    "from sklearn.metrics import precision_recall_curve\n",
    "precision, recall, thresholds = precision_recall_curve(y_true, y_proba)\n",
    "# convert to f score\n",
    "fscore = (2 * precision * recall) / (precision + recall)\n",
    "# locate the index of the largest f score\n",
    "ix = np.argmax(fscore)\n",
    "print('Best Threshold=%f, F-Score=%.3f' % (thresholds[ix], fscore[ix]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "id": "05647bf2-5720-46c6-9676-6100ef3137af",
   "metadata": {},
   "outputs": [],
   "source": [
    "y_prob_flat = []\n",
    "y_pred_new = []\n",
    "for key, values in enumerate(y_proba):\n",
    "    y_prob_flat.append(values[0])\n",
    "    if values[0] < 0.491328:\n",
    "        y_pred_new.append(0)\n",
    "    else:\n",
    "        y_pred_new.append(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "3725613e-749b-47b5-8857-012059ec5350",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "         0.0       0.87      0.93      0.90       263\n",
      "         1.0       0.86      0.75      0.80       146\n",
      "\n",
      "    accuracy                           0.87       409\n",
      "   macro avg       0.86      0.84      0.85       409\n",
      "weighted avg       0.87      0.87      0.86       409\n",
      "\n",
      "F1 score 0.7985347985347986\n",
      "Recall score 0.7465753424657534\n"
     ]
    }
   ],
   "source": [
    "print(classification_report(y_true, y_pred_new))\n",
    "print(\"F1 score\", f1_score(y_true ,y_pred_new))\n",
    "print(\"Recall score\", recall_score(y_true ,y_pred_new))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "851fc8ea-c99d-4c16-9e8c-56926fcbba09",
   "metadata": {},
   "source": [
    "# 5. Comparison"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6dfaaadb-ebfc-448c-93bf-0e3460832b34",
   "metadata": {},
   "source": [
    "## 5.1. Evaluation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "id": "95db1e41-3d62-4b24-bace-44f3dbd24e9a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.model_selection import RepeatedStratifiedKFold\n",
    "from sklearn.model_selection import cross_val_score\n",
    "from sklearn.metrics import recall_score\n",
    "cv= RepeatedStratifiedKFold(n_splits=10, n_repeats=3,\n",
    "                            random_state=42)\n",
    "X = df.drop(['ID','pIC50'],axis = 1)\n",
    "y = df[\"pIC50\"]\n",
    "#Split Data train, Data_test, Data_validation\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.2,\n",
    "                                                                random_state =42, stratify=y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "id": "17a2a6d4-8df1-4abb-a4fe-218c97c05fe3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Internal Validation - SVC\n",
      ">F1-cv: 0.734 ± 0.047 (0.734)\n",
      ">AP-cv: 0.858 ± 0.034 (0.849)\n",
      ">Recall-cv: 0.645 ± 0.062 (0.650)\n",
      "External Validation - SVC\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.86      0.89      0.88       263\n",
      "           1       0.80      0.75      0.77       146\n",
      "\n",
      "    accuracy                           0.84       409\n",
      "   macro avg       0.83      0.82      0.82       409\n",
      "weighted avg       0.84      0.84      0.84       409\n",
      "\n",
      "Average precision score 0.8403695047099706\n",
      "F1 score 0.7715355805243447\n",
      "Recall score 0.7054794520547946\n"
     ]
    }
   ],
   "source": [
    "from sklearn.svm import SVC\n",
    "model = SVC(probability=True)\n",
    "\n",
    "print('Internal Validation - SVC')\n",
    "scores_f1 = cross_val_score(model, X_train, y_train, scoring = \"f1\", cv=cv, n_jobs = -1)\n",
    "scores_ap=cross_val_score(model, X_train, y_train, scoring = \"average_precision\",\n",
    "                        cv= cv, n_jobs = -1)\n",
    "scores_recall=cross_val_score(model, X_train, y_train, scoring = \"recall\",\n",
    "                        cv= cv, n_jobs = -1)\n",
    "\n",
    "\n",
    "print('>%s %.3f ± %.3f (%.3f)' % ('F1-cv:', np.mean(scores_f1), np.std(scores_f1), np.median(scores_f1)))\n",
    "print('>%s %.3f ± %.3f (%.3f)' % ('AP-cv:', np.mean(scores_ap), np.std(scores_ap), np.median(scores_ap)))\n",
    "print('>%s %.3f ± %.3f (%.3f)' % ('Recall-cv:', np.mean(scores_recall), np.std(scores_recall), np.median(scores_recall)))\n",
    "\n",
    "print('External Validation - SVC')\n",
    "model.fit(X_train, y_train)\n",
    "print(classification_report(y_test,model.predict(X_test)))\n",
    "print(\"Average precision score\", average_precision_score(y_test, model.predict_proba(X_test)[:,1]))\n",
    "print(\"F1 score\", f1_score(y_test, model.predict(X_test)))\n",
    "print(\"Recall score\", recall_score(y_test, model.predict(X_test)))\n",
    "\n",
    "record = {\"F1_score\": scores_f1,\n",
    "         \"AP_score\": scores_ap,\n",
    "         \"Recall_score\": scores_recall}\n",
    "df_record = pd.DataFrame(record)\n",
    "df_record.to_csv(\"./Data/QSAR/Compare/SVC.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "id": "c1e76527-cdb5-42d8-9caf-0e9a38b45559",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Internal Validation - Random Forest\n",
      ">F1-cv: 0.740 ± 0.049 (0.742)\n",
      ">AP-cv: 0.848 ± 0.037 (0.832)\n",
      ">Recall-cv: 0.678 ± 0.063 (0.675)\n",
      "External Validation - Random Forest\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.86      0.89      0.88       263\n",
      "           1       0.80      0.75      0.77       146\n",
      "\n",
      "    accuracy                           0.84       409\n",
      "   macro avg       0.83      0.82      0.82       409\n",
      "weighted avg       0.84      0.84      0.84       409\n",
      "\n",
      "Average precision score 0.8644579562657897\n",
      "F1 score 0.7703180212014135\n",
      "Recall score 0.7465753424657534\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "model = RandomForestClassifier(random_state=42)\n",
    "\n",
    "print('Internal Validation - Random Forest')\n",
    "scores_f1 = cross_val_score(model, X_train, y_train, scoring = \"f1\", cv=cv, n_jobs = -1)\n",
    "scores_ap=cross_val_score(model, X_train, y_train, scoring = \"average_precision\",\n",
    "                        cv= cv, n_jobs = -1)\n",
    "scores_recall=cross_val_score(model, X_train, y_train, scoring = \"recall\",\n",
    "                        cv= cv, n_jobs = -1)\n",
    "\n",
    "\n",
    "print('>%s %.3f ± %.3f (%.3f)' % ('F1-cv:', np.mean(scores_f1), np.std(scores_f1), np.median(scores_f1)))\n",
    "print('>%s %.3f ± %.3f (%.3f)' % ('AP-cv:', np.mean(scores_ap), np.std(scores_ap), np.median(scores_ap)))\n",
    "print('>%s %.3f ± %.3f (%.3f)' % ('Recall-cv:', np.mean(scores_recall), np.std(scores_recall), np.median(scores_recall)))\n",
    "\n",
    "print('External Validation - Random Forest')\n",
    "model.fit(X_train, y_train)\n",
    "print(classification_report(y_test, model.predict(X_test)))\n",
    "print(\"Average precision score\", average_precision_score(y_test, model.predict_proba(X_test)[:,1]))\n",
    "print(\"F1 score\", f1_score(y_test, model.predict(X_test)))\n",
    "print(\"Recall score\", recall_score(y_test, model.predict(X_test)))\n",
    "\n",
    "record = {\"F1_score\": scores_f1,\n",
    "         \"AP_score\": scores_ap,\n",
    "         \"Recall_score\": scores_recall}\n",
    "df_record = pd.DataFrame(record)\n",
    "df_record.to_csv(\"./Data/QSAR/Compare/RF.csv\", index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "id": "13e9e611-c952-4318-9448-2740fd330880",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Fold: 1..  validation f1_score: 0.796..  validation average precision: 0.899.. \n",
      "Fold: 2..  validation f1_score: 0.706..  validation average precision: 0.821.. \n",
      "Fold: 3..  validation f1_score: 0.733..  validation average precision: 0.835.. \n",
      "Fold: 4..  validation f1_score: 0.738..  validation average precision: 0.852.. \n",
      "Fold: 5..  validation f1_score: 0.706..  validation average precision: 0.841.. \n",
      "Fold: 6..  validation f1_score: 0.800..  validation average precision: 0.894.. \n",
      "Fold: 7..  validation f1_score: 0.733..  validation average precision: 0.881.. \n",
      "Fold: 8..  validation f1_score: 0.733..  validation average precision: 0.877.. \n",
      "Fold: 9..  validation f1_score: 0.760..  validation average precision: 0.849.. \n",
      "Fold: 10..  validation f1_score: 0.714..  validation average precision: 0.853.. \n",
      "Fold: 11..  validation f1_score: 0.800..  validation average precision: 0.910.. \n",
      "Fold: 12..  validation f1_score: 0.729..  validation average precision: 0.909.. \n",
      "Fold: 13..  validation f1_score: 0.730..  validation average precision: 0.839.. \n",
      "Fold: 14..  validation f1_score: 0.763..  validation average precision: 0.830.. \n",
      "Fold: 15..  validation f1_score: 0.722..  validation average precision: 0.865.. \n",
      "Fold: 16..  validation f1_score: 0.720..  validation average precision: 0.862.. \n",
      "Fold: 17..  validation f1_score: 0.741..  validation average precision: 0.847.. \n",
      "Fold: 18..  validation f1_score: 0.745..  validation average precision: 0.880.. \n",
      "Fold: 19..  validation f1_score: 0.725..  validation average precision: 0.805.. \n",
      "Fold: 20..  validation f1_score: 0.784..  validation average precision: 0.905.. \n",
      "Fold: 21..  validation f1_score: 0.762..  validation average precision: 0.869.. \n",
      "Fold: 22..  validation f1_score: 0.788..  validation average precision: 0.848.. \n",
      "Fold: 23..  validation f1_score: 0.709..  validation average precision: 0.796.. \n",
      "Fold: 24..  validation f1_score: 0.771..  validation average precision: 0.870.. \n",
      "Fold: 25..  validation f1_score: 0.724..  validation average precision: 0.852.. \n",
      "Fold: 26..  validation f1_score: 0.777..  validation average precision: 0.896.. \n",
      "Fold: 27..  validation f1_score: 0.659..  validation average precision: 0.830.. \n",
      "Fold: 28..  validation f1_score: 0.796..  validation average precision: 0.909.. \n",
      "Fold: 29..  validation f1_score: 0.824..  validation average precision: 0.916.. \n",
      "Fold: 30..  validation f1_score: 0.674..  validation average precision: 0.838.. \n",
      "Overall AP score = 0.8626\n",
      "Overall F1 score = 0.7454\n"
     ]
    }
   ],
   "source": [
    "net = Net(in_features = 2048, hidden_nodes_1 =2048,hidden_nodes_2 =1024, hidden_nodes_3=256, drop_out=0.5)\n",
    "pip = model_pipeline(device = device, seed = 42,\n",
    "                     save_dir = '.')\n",
    "\n",
    "pip.cross_val_score(model = net,epochs=170, X_train=X_train, y_train=y_train, cv=cv)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 134,
   "id": "862a2474-49eb-4db9-945a-e3959152de7a",
   "metadata": {},
   "outputs": [],
   "source": [
    "f1_ann =pd.DataFrame(pip.History['F1_record'], columns =['F1_score'])\n",
    "ap_ann =pd.DataFrame(pip.History['AP_record'], columns =['AP_score'])\n",
    "df_record_ann = pd.concat([f1_ann, ap_ann], axis=1)\n",
    "df_record_ann.to_csv(\"./Data/QSAR/Compare/ANN.csv\", index=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc320311-758f-4299-9292-03f0c37537ac",
   "metadata": {},
   "source": [
    "## 5.2. Statistical test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 135,
   "id": "e9fcb882-441e-489e-b2f1-b97baa888f87",
   "metadata": {},
   "outputs": [],
   "source": [
    "svc = pd.read_csv(\"./Data/QSAR/Compare/SVC.csv\",)\n",
    "rf = pd.read_csv(\"./Data/QSAR/Compare/RF.csv\",)\n",
    "ann = pd.read_csv(\"./Data/QSAR/Compare/ANN.csv\",)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "id": "23982a57-2d7f-4967-9ad2-37e0674a8796",
   "metadata": {},
   "outputs": [],
   "source": [
    "F1_internal =pd.DataFrame({\"SVC\":svc[\"F1_score\"].values,\n",
    "                           \"ANN\":ann[\"F1_score\"].values,\n",
    "                          \"RF\":rf[\"F1_score\"].values})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "id": "939f8783-607e-49f8-84b6-8edaabcae6c7",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/labhhc/anaconda3/envs/pytorch_gpu/lib/python3.10/site-packages/scipy/stats/_morestats.py:3414: UserWarning: Exact p-value calculation does not work if there are zeros. Switching to normal approximation.\n",
      "  warnings.warn(\"Exact p-value calculation does not work if there are \"\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVC vs. ANN: Custom statistical test, P_val:1.161e-01\n",
      "ANN vs. RF: Custom statistical test, P_val:4.174e-01\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from statannotations.Annotator import Annotator\n",
    "from scipy.stats import mannwhitneyu, normaltest, wilcoxon\n",
    "from statannot import add_stat_annotation\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "df_melt = pd.melt(F1_internal.reset_index(), id_vars=['index'], value_vars=F1_internal.columns)\n",
    "df_melt.columns = ['index', 'Model', 'F1']\n",
    "subcat_palette = sns.dark_palette(\"#8BF\", reverse=True, n_colors=5)\n",
    "\n",
    "stat_results = [wilcoxon(F1_internal['SVC'].astype('float'), F1_internal['ANN'].astype('float'), alternative=\"two-sided\"),\n",
    "                wilcoxon(F1_internal['RF'].astype('float'), F1_internal['ANN'].astype('float'), alternative=\"two-sided\"),\n",
    "               ]\n",
    "\n",
    "pvalues = [result.pvalue for result in stat_results]\n",
    "plotting_parameters = {\n",
    "'data':    df_melt,\n",
    "'x':       'Model',\n",
    "'y':       'F1',\n",
    "'palette': subcat_palette[1:]}\n",
    "\n",
    "pairs = [('SVC', 'ANN'),\n",
    "          ('RF', 'ANN'),]\n",
    "\n",
    "\n",
    "sns.set_style(\"whitegrid\")\n",
    "plt.figure(figsize = (10,7))\n",
    "\n",
    "my_colors = {'SVC': 'salmon', \n",
    "             'RF': 'orange', \n",
    "             # 'SS': 'lightgreen',\n",
    "             # 'GS': 'lightblue',\n",
    "             'ANN': 'lightgreen'}\n",
    "\n",
    "\n",
    "ax = sns.boxplot(x='Model', y='F1', data=df_melt, palette=my_colors, showmeans=True ,meanprops={\"marker\":\"d\",\n",
    "                           \"markerfacecolor\":\"white\", \n",
    "                           \"markeredgecolor\":\"black\",\n",
    "                          \"markersize\":\"5\"})\n",
    "\n",
    "mean = round(F1_internal.mean(),3)\n",
    "data = np.array(mean)   \n",
    "ser = pd.Series(data, index =F1_internal.columns)\n",
    "\n",
    "dict_columns = {'Mean':mean,}\n",
    "df = pd.DataFrame(dict_columns)\n",
    "\n",
    "vertical_offset = df[\"Mean\"].median()*0.008\n",
    "\n",
    "for xtick in ax.get_xticks():\n",
    "    ax.text(xtick,ser[xtick]+ vertical_offset,ser[xtick], horizontalalignment='center',color='k',weight='semibold', fontsize = 15)\n",
    "\n",
    "\n",
    "annotator = Annotator(ax, pairs, **plotting_parameters)\n",
    "annotator.configure(text_format=\"simple\")\n",
    "annotator.set_pvalues_and_annotate(pvalues)\n",
    "\n",
    "ax.set_ylabel('F1 SCORE', fontsize = 12)\n",
    "ax.set_xlabel(None)\n",
    "ax.set_xticklabels(labels = F1_internal.columns, fontsize = 12)\n",
    "ax.set_title(\"F1 score - Cross validation - Wilcoxon\",fontsize = 16, weight ='semibold')\n",
    "plt.savefig(\"./Data/QSAR/Compare/f1_compare.png\", dpi = 600)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "id": "1931860d-90b7-4f52-aa8f-fe07d658908d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SVC vs. ANN: Custom statistical test, P_val:8.407e-02\n",
      "ANN vs. RF: Custom statistical test, P_val:1.341e-03\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x700 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "from statannotations.Annotator import Annotator\n",
    "from scipy.stats import mannwhitneyu, normaltest, wilcoxon\n",
    "from statannot import add_stat_annotation\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "F1_internal =pd.DataFrame({\"SVC\":svc[\"AP_score\"].values,\n",
    "                           \"ANN\":ann[\"AP_score\"].values,\n",
    "                          \"RF\":rf[\"AP_score\"].values})\n",
    "\n",
    "df_melt = pd.melt(F1_internal.reset_index(), id_vars=['index'], value_vars=F1_internal.columns)\n",
    "df_melt.columns = ['index', 'Model', 'F1']\n",
    "subcat_palette = sns.dark_palette(\"#8BF\", reverse=True, n_colors=5)\n",
    "\n",
    "stat_results = [wilcoxon(F1_internal['SVC'].astype('float'), F1_internal['ANN'].astype('float'), alternative=\"two-sided\"),\n",
    "                wilcoxon(F1_internal['RF'].astype('float'), F1_internal['ANN'].astype('float'), alternative=\"two-sided\"),\n",
    "               ]\n",
    "\n",
    "pvalues = [result.pvalue for result in stat_results]\n",
    "plotting_parameters = {\n",
    "'data':    df_melt,\n",
    "'x':       'Model',\n",
    "'y':       'F1',\n",
    "'palette': subcat_palette[1:]}\n",
    "\n",
    "pairs = [('SVC', 'ANN'),\n",
    "          ('RF', 'ANN'),]\n",
    "\n",
    "\n",
    "sns.set_style(\"whitegrid\")\n",
    "plt.figure(figsize = (10,7))\n",
    "\n",
    "my_colors = {'SVC': 'salmon', \n",
    "             'RF': 'orange', \n",
    "             # 'SS': 'lightgreen',\n",
    "             # 'GS': 'lightblue',\n",
    "             'ANN': 'lightgreen'}\n",
    "\n",
    "\n",
    "ax = sns.boxplot(x='Model', y='F1', data=df_melt, palette=my_colors, showmeans=True ,meanprops={\"marker\":\"d\",\n",
    "                           \"markerfacecolor\":\"white\", \n",
    "                           \"markeredgecolor\":\"black\",\n",
    "                          \"markersize\":\"5\"})\n",
    "\n",
    "mean = round(F1_internal.mean(),3)\n",
    "data = np.array(mean)   \n",
    "ser = pd.Series(data, index =F1_internal.columns)\n",
    "\n",
    "dict_columns = {'Mean':mean,}\n",
    "df = pd.DataFrame(dict_columns)\n",
    "\n",
    "vertical_offset = df[\"Mean\"].median()*0.008\n",
    "\n",
    "for xtick in ax.get_xticks():\n",
    "    ax.text(xtick,ser[xtick]+ vertical_offset,ser[xtick], horizontalalignment='center',color='k',weight='semibold', fontsize = 15)\n",
    "\n",
    "\n",
    "annotator = Annotator(ax, pairs, **plotting_parameters)\n",
    "annotator.configure(text_format=\"simple\")\n",
    "annotator.set_pvalues_and_annotate(pvalues)\n",
    "\n",
    "ax.set_ylabel('AP SCORE', fontsize = 12)\n",
    "ax.set_xlabel(None)\n",
    "ax.set_xticklabels(labels = F1_internal.columns, fontsize = 12)\n",
    "ax.set_title(\"AP score - Cross validation - Wilcoxon\",fontsize = 16, weight ='semibold')\n",
    "plt.savefig(\"./Data/QSAR/Compare/ap_compare.png\", dpi = 600)\n",
    "plt.show()\n",
    "#plt.savefig(\"ML_f1_compare_tune.png\", dpi = 600)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f77fd7b0-23f3-4dce-843e-3cfcb83fb951",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "54167894-46ab-42c0-be44-8d7bd485a22f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ca3452ee-f17c-4343-a99b-5f79a91f4c23",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4c012ba1-20f6-41c4-a169-681f6dd632f7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a38d2bce-076a-4ea8-98df-1a9d0ec2c63e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fd99db41-8dfd-4c54-9e8c-dcabf7fa2cd7",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "fbe54351-143b-4542-8f15-f0f27cd71fc2",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "229c8f64-efea-46ac-8513-0e02cd7c7c06",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "be697cd5-3391-410d-a5cb-9fe636908ac0",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b961af3-0ddc-4278-8413-8bbf152e2ac6",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2550a63d-65f2-42a4-9e89-520da65eaebd",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "398c9b7c-0027-421d-81b6-6cf29b26be02",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "221fbd8a-dd16-4b15-ac54-c2c3b314eb07",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "d7d24261-967e-4026-bbbb-094bbc886ee8",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "3cb80cf1-5ff7-4c37-8344-82ca3afe7e38",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "76b11b20-a37d-409a-bdda-5fbb8f9b69c8",
   "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": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}