warfarin_linUCB_Gage.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "source": [
    "import math\n",
    "import pandas as pd\n",
    "import random\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "import time\n",
    "warnings.simplefilter(action='ignore', category=FutureWarning)\n",
    "warnings.simplefilter(action='ignore', category=DeprecationWarning)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:40.130115Z",
     "start_time": "2024-05-27T02:54:40.126989Z"
    }
   },
   "id": "b8220aa2932ad65b",
   "outputs": [],
   "execution_count": 1
  },
  {
   "cell_type": "markdown",
   "source": [
    "# Prepare data\n",
    "\n",
    "## Data"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "3ea94f32c86c7864"
  },
  {
   "cell_type": "code",
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:40.232530Z",
     "start_time": "2024-05-27T02:54:40.180029Z"
    }
   },
   "source": [
    "data = pd.DataFrame(columns=[\n",
    "    'Age',\n",
    "    'Height (cm)',\n",
    "    'Weight (kg)',\n",
    "    'Amiodarone',\n",
    "    'Smokes',\n",
    "    'Race Black or African American',\n",
    "    'CYP2C9 *2',\n",
    "    'CYP2C9 *3',\n",
    "    'VKORC1',\n",
    "    'Target INR',\n",
    "    'dose',\n",
    "    'type'])"
   ],
   "outputs": [],
   "execution_count": 2
  },
  {
   "cell_type": "markdown",
   "source": [
    "**Explain**\n",
    "\n",
    "sample = {  \n",
    "'Age': int 20-90 (for adult only),  \n",
    "'Height (cm)': double 140-200,  \n",
    "'Weight (kg)': double with bmi in range (16 < $\\frac{weight}{height^2} < 30$),  \n",
    "\n",
    "'Amiodarone': 0-1,  \n",
    "'Smoke': 0-1,  \n",
    " \n",
    "'Race Black or African American': 0-1,   \n",
    "If true is 1 else 0  \n",
    "  \n",
    "'CYP2C9 *2': 0-2,  \n",
    "'CYP2C9 *3': 0-2,  \n",
    "sum cant bigger than 2\n",
    "\n",
    "'VKORC1': 0-2,  \n",
    "\n",
    "'Target INR': 1-3.5\n",
    "\n",
    "'dose': double calc from below function,  \n",
    "'type': 0 - low, 1 - medium, 2 - high calc from below function  \n",
    "}"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "4a9757c10c0facdb"
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Calc dose function\n"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "89978b4d04ce10d7"
  },
  {
   "cell_type": "code",
   "source": [
    "def calc_dose(sampl):\n",
    "    dose = math.exp(0.9751\n",
    "                    - 0.3238 * sampl['VKORC1']\n",
    "                    + 0.4317 * (0.0235 * sampl['Weight (kg)'] ** 0.51456 * sampl['Height (cm)'] ** 0.42246)  # BSA\n",
    "                    - 0.00745 * sampl['Age']\n",
    "                    - 0.0901 * sampl['Race Black or African American']\n",
    "                    + 0.0922 * sampl['Smokes']\n",
    "                    - 0.2538 * sampl['Amiodarone']\n",
    "                    - 0.2066 * sampl['CYP2C9 *2']\n",
    "                    - 0.4008 * sampl['CYP2C9 *3']\n",
    "                    + 0.2029 * sampl['Target INR']\n",
    "                    ) * 7\n",
    "    if dose < 21:\n",
    "        level = 0\n",
    "    elif 21 <= dose < 49:\n",
    "        level = 1\n",
    "    else:\n",
    "        level = 2\n",
    "    return dose, level"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:40.323026Z",
     "start_time": "2024-05-27T02:54:40.233860Z"
    }
   },
   "id": "848642d0f825b4ae",
   "outputs": [],
   "execution_count": 3
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Random data function"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "e5a13cc7ee35ab1d"
  },
  {
   "cell_type": "code",
   "source": [
    "def create_data():\n",
    "    sample = {\n",
    "        'Age': 0,\n",
    "        'Height (cm)': 0,\n",
    "        'Weight (kg)': 0,\n",
    "\n",
    "        'Amiodarone': 0,\n",
    "        'Smokes': 0,\n",
    "\n",
    "        'Race Black or African American': 0,\n",
    "\n",
    "        'CYP2C9 *2': 0,\n",
    "        'CYP2C9 *3': 0,\n",
    "\n",
    "        'VKORC1': 0,\n",
    "        'Target INR': 0,\n",
    "\n",
    "        'dose': 0,\n",
    "        'type': 0\n",
    "    }\n",
    "\n",
    "    sample['Age'] = random.randint(20, 90)\n",
    "    sample['Height (cm)'] = random.uniform(1.4, 2) * 100\n",
    "    while not (16 <= sample['Weight (kg)'] / ((sample['Height (cm)'] / 100) ** 2) <= 30):\n",
    "        sample['Weight (kg)'] = random.uniform(31.36, 200)\n",
    "\n",
    "    sample['Amiodarone'] = random.randint(0, 1)\n",
    "    sample['Smokes'] = random.randint(0, 1)\n",
    "    sample['Target INR'] = random.uniform(1, 3.5)\n",
    "\n",
    "    race = random.randint(0, 12)\n",
    "    match race:\n",
    "        case 2 | 5 | 8 | 11:\n",
    "            sample['Race Black or African American'] = 1\n",
    "\n",
    "    cyp2d9 = random.randint(0, 24)\n",
    "    match cyp2d9:\n",
    "        case 2 | 8 | 14 | 20:\n",
    "            sample['CYP2C9 *2'] = 1\n",
    "        case 3 | 9 | 15 | 21:\n",
    "            sample['CYP2C9 *3'] = 1\n",
    "        case 4 | 10 | 16 | 22:\n",
    "            sample['CYP2C9 *2'] = 2\n",
    "        case 5 | 11 | 17 | 23:\n",
    "            sample['CYP2C9 *2'] = 1\n",
    "            sample['CYP2C9 *3'] = 1\n",
    "        case 6 | 12 | 18 | 24:\n",
    "            sample['CYP2C9 *3'] = 2\n",
    "\n",
    "    vkorc1 = random.randint(0, 12)\n",
    "    match vkorc1:\n",
    "        case 1 | 4 | 7 | 10:\n",
    "            sample['VKORC1'] = 2\n",
    "        case 2 | 5 | 8 | 11:\n",
    "            sample['VKORC1'] = 1\n",
    "        case 3 | 6 | 9 | 12:\n",
    "            sample['VKORC1'] = 0\n",
    "\n",
    "    sample['dose'], sample['type'] = calc_dose(sample)\n",
    "\n",
    "    return sample\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:40.416002Z",
     "start_time": "2024-05-27T02:54:40.324015Z"
    }
   },
   "id": "2f3adfb53dca3b5f",
   "outputs": [],
   "execution_count": 4
  },
  {
   "cell_type": "markdown",
   "source": [
    "## Generate data"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "dc84cc50ddb94ae0"
  },
  {
   "cell_type": "code",
   "source": [
    "dict_data = []\n",
    "data_size = 100000\n",
    "for i in range(data_size):\n",
    "    dict_data.append(create_data())\n",
    "data = data._append(dict_data, ignore_index=True)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:41.746691Z",
     "start_time": "2024-05-27T02:54:40.417614Z"
    }
   },
   "id": "886cc20f73174858",
   "outputs": [],
   "execution_count": 5
  },
  {
   "cell_type": "code",
   "source": [
    "print(\"Age: %.2f ± %.2f\" % (data['Age'].mean(), data['Age'].std()))\n",
    "print(\"Height (cm): %.2f ± %.2f\" % (data['Height (cm)'].mean(), data['Height (cm)'].std()))\n",
    "print(\"Weight (kg): %.2f ± %.2f\" % (data['Weight (kg)'].mean(), data['Weight (kg)'].std()))\n",
    "aminodarone = data['Amiodarone'].sum()\n",
    "print(\"Amiodarone: %2.2f%% (%d)\" % (aminodarone / data_size * 100, aminodarone))\n",
    "smoke = data['Smokes'].sum()\n",
    "print(\"Smokes: %2.2f%% (%d)\" % (smoke / data_size * 100, smoke))"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:41.788523Z",
     "start_time": "2024-05-27T02:54:41.747525Z"
    }
   },
   "id": "9d5b6b1e5544348d",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Age: 55.03 ± 20.54\n",
      "Height (cm): 170.05 ± 17.35\n",
      "Weight (kg): 67.21 ± 18.14\n",
      "Amiodarone: 50.01% (50014)\n",
      "Smokes: 50.03% (50026)\n"
     ]
    }
   ],
   "execution_count": 6
  },
  {
   "cell_type": "code",
   "source": [
    "def pct(x):\n",
    "    print('{:.4f}% ({:.0f})'.format(x, data_size * x / 100))\n",
    "    return '{:.4f}%\\n({:.0f})'.format(x, data_size * x / 100)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:41.869430Z",
     "start_time": "2024-05-27T02:54:41.789620Z"
    }
   },
   "id": "1518927f5043b9bf",
   "outputs": [],
   "execution_count": 7
  },
  {
   "cell_type": "code",
   "source": [
    "values = [data['Race Black or African American'].sum(),\n",
    "          data_size - data['Race Black or African American'].sum()\n",
    "          ]\n",
    "labels = [\n",
    "    'Race Black or African American',\n",
    "    'Race Other']\n",
    "_ = plt.pie(values, labels=labels, autopct=pct)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:42.048630Z",
     "start_time": "2024-05-27T02:54:41.870682Z"
    }
   },
   "id": "3f0f1e89b982b495",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "30.6340% (30634)\n",
      "69.3660% (69366)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 8
  },
  {
   "cell_type": "code",
   "source": [
    "values = [\n",
    "    len(data[(data['CYP2C9 *2'] == 0) & (data['CYP2C9 *3'] == 0)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 1) & (data['CYP2C9 *3'] == 0)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 0) & (data['CYP2C9 *3'] == 1)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 1) & (data['CYP2C9 *3'] == 1)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 2) & (data['CYP2C9 *3'] == 0)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 0) & (data['CYP2C9 *3'] == 2)]),\n",
    "]\n",
    "labels = ['Other', '*2/*?', '*?/*3', '*2/*3', '*2/*2', '*3/*3']\n",
    "_ = plt.pie(values, labels=labels, autopct=pct)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:42.211472Z",
     "start_time": "2024-05-27T02:54:42.050095Z"
    }
   },
   "id": "6808762c6246ed69",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19.8110% (19811)\n",
      "15.9010% (15901)\n",
      "16.2350% (16235)\n",
      "16.0160% (16016)\n",
      "16.0680% (16068)\n",
      "15.9690% (15969)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 9
  },
  {
   "cell_type": "code",
   "source": [
    "values = [data['VKORC1'].value_counts()[2],\n",
    "          data['VKORC1'].value_counts()[1],\n",
    "          data['VKORC1'].value_counts()[0],\n",
    "          ]\n",
    "labels = ['A/A', 'A/?', 'unknown']\n",
    "_ = plt.pie(values, labels=labels, autopct=pct)\n",
    "plt.show()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:42.287971Z",
     "start_time": "2024-05-27T02:54:42.212471Z"
    }
   },
   "id": "3862467b583d77c",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "31.0280% (31028)\n",
      "30.4840% (30484)\n",
      "38.4880% (38488)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 10
  },
  {
   "cell_type": "code",
   "source": [
    "ratio = 0.7\n",
    "pointer = int(ratio * data_size)\n",
    "train_set = data.iloc[:pointer, :-2]\n",
    "test_set = data.iloc[pointer:, :-2]\n",
    "train_dose = data.iloc[:pointer, -1:].values\n",
    "test_dose = data.iloc[pointer:, -1:].values"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T02:54:42.358169Z",
     "start_time": "2024-05-27T02:54:42.290879Z"
    }
   },
   "id": "36c424f1385a05ec",
   "outputs": [],
   "execution_count": 11
  },
  {
   "cell_type": "code",
   "source": [
    "arms = [0, 1, 2]\n",
    "accuracy = []\n",
    "A_map = {}\n",
    "b_map = {}\n",
    "for arm in arms:\n",
    "    A_map[arm] = np.identity(len(list(train_set)))\n",
    "    b_map[arm] = np.zeros(len(list(train_set)))\n",
    "\n",
    "for epoch in range(200):\n",
    "    start_time = time.time()\n",
    "    # train\n",
    "    indices = np.random.permutation(list(range(len(train_set))))\n",
    "    for i in indices:\n",
    "        feature = np.array(train_set.iloc[i], dtype=float)\n",
    "        # iterate through actions\n",
    "        p = {}\n",
    "        max_value = -np.inf\n",
    "        max_arm = 0\n",
    "        for arm in arms:\n",
    "            theta = np.matmul(np.linalg.inv(A_map[arm]), b_map[arm])\n",
    "            p = np.matmul(theta.T, feature) + \\\n",
    "                np.sqrt(np.matmul(np.matmul(feature.T, np.linalg.inv(A_map[arm])), feature))\n",
    "\n",
    "            if p == max_value:\n",
    "                if np.random.choice(a=[False, True]): max_arm = arm\n",
    "            elif p > max_value:\n",
    "                max_value = p\n",
    "                max_arm = arm\n",
    "\n",
    "        # observe reward and update A and b\n",
    "        r = 0 if max_arm == train_dose[i] else -1\n",
    "        A_map[max_arm] += np.outer(feature, feature)\n",
    "        b_map[max_arm] += r * feature\n",
    "\n",
    "    # test\n",
    "    correct_count = 0\n",
    "    count = 0\n",
    "    higher = 0\n",
    "    lower = 0\n",
    "    for i in range(len(test_set)):\n",
    "        count += 1\n",
    "        feature = np.array(test_set.iloc[i], dtype=float)\n",
    "        # iterate through actions\n",
    "        p = {}\n",
    "        max_value = -np.inf\n",
    "        max_arm = 0\n",
    "        for arm in arms:\n",
    "            theta = np.matmul(np.linalg.inv(A_map[arm]), b_map[arm])\n",
    "            p = np.matmul(theta.T, feature) + \\\n",
    "                np.sqrt(np.matmul(np.matmul(feature.T, np.linalg.inv(A_map[arm])), feature))\n",
    "            if p == max_value:\n",
    "                if np.random.choice(a=[False, True]): max_arm = arm\n",
    "            elif p > max_value:\n",
    "                max_value = p\n",
    "                max_arm = arm\n",
    "\n",
    "        if max_arm == test_dose[i]:\n",
    "            correct_count += 1\n",
    "        elif max_arm < test_dose[i]:\n",
    "            lower += test_dose[i] - max_arm\n",
    "        else:\n",
    "            higher += max_arm - test_dose[i]\n",
    "\n",
    "    end_time = time.time()\n",
    "    accu = correct_count / count\n",
    "    print(\"Epoch %d: Accuracy: %.5f%% (%d/%d), Higher: %d, Lower: %d, Time: %.5fs\" % (\n",
    "        epoch, accu * 100, correct_count, count, higher, lower, end_time - start_time))\n",
    "    accuracy.append(accu)\n"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:39.793953Z",
     "start_time": "2024-05-27T02:54:42.359762Z"
    }
   },
   "id": "ef0157a1766ae6da",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 0: Accuracy: 93.84333% (28153/30000), Higher: 599, Lower: 1248, Time: 15.56986s\n",
      "Epoch 1: Accuracy: 97.53000% (29259/30000), Higher: 331, Lower: 410, Time: 15.44681s\n",
      "Epoch 2: Accuracy: 97.24333% (29173/30000), Higher: 323, Lower: 504, Time: 15.53597s\n",
      "Epoch 3: Accuracy: 97.21667% (29165/30000), Higher: 311, Lower: 524, Time: 15.00262s\n",
      "Epoch 4: Accuracy: 97.21333% (29164/30000), Higher: 408, Lower: 428, Time: 15.21497s\n",
      "Epoch 5: Accuracy: 97.53000% (29259/30000), Higher: 245, Lower: 496, Time: 15.01596s\n",
      "Epoch 6: Accuracy: 97.49000% (29247/30000), Higher: 299, Lower: 454, Time: 14.49568s\n",
      "Epoch 7: Accuracy: 97.25667% (29177/30000), Higher: 361, Lower: 462, Time: 14.44342s\n",
      "Epoch 8: Accuracy: 97.54667% (29264/30000), Higher: 192, Lower: 544, Time: 14.40658s\n",
      "Epoch 9: Accuracy: 97.40333% (29221/30000), Higher: 331, Lower: 448, Time: 14.80215s\n",
      "Epoch 10: Accuracy: 97.37333% (29212/30000), Higher: 347, Lower: 441, Time: 14.44979s\n",
      "Epoch 11: Accuracy: 97.40000% (29220/30000), Higher: 375, Lower: 405, Time: 14.44049s\n",
      "Epoch 12: Accuracy: 97.34333% (29203/30000), Higher: 362, Lower: 435, Time: 14.58267s\n",
      "Epoch 13: Accuracy: 97.34333% (29203/30000), Higher: 385, Lower: 412, Time: 14.62040s\n",
      "Epoch 14: Accuracy: 97.48000% (29244/30000), Higher: 293, Lower: 463, Time: 14.90237s\n",
      "Epoch 15: Accuracy: 97.34000% (29202/30000), Higher: 382, Lower: 416, Time: 14.86539s\n",
      "Epoch 16: Accuracy: 97.44667% (29234/30000), Higher: 303, Lower: 463, Time: 15.05223s\n",
      "Epoch 17: Accuracy: 97.48333% (29245/30000), Higher: 278, Lower: 477, Time: 14.93093s\n",
      "Epoch 18: Accuracy: 97.46667% (29240/30000), Higher: 338, Lower: 422, Time: 15.15998s\n",
      "Epoch 19: Accuracy: 97.44667% (29234/30000), Higher: 345, Lower: 421, Time: 15.04848s\n",
      "Epoch 20: Accuracy: 97.54333% (29263/30000), Higher: 239, Lower: 498, Time: 14.22242s\n",
      "Epoch 21: Accuracy: 97.52333% (29257/30000), Higher: 297, Lower: 446, Time: 14.54722s\n",
      "Epoch 22: Accuracy: 97.57333% (29272/30000), Higher: 320, Lower: 408, Time: 14.43366s\n",
      "Epoch 23: Accuracy: 97.53000% (29259/30000), Higher: 320, Lower: 421, Time: 14.76007s\n",
      "Epoch 24: Accuracy: 97.54667% (29264/30000), Higher: 314, Lower: 422, Time: 14.55979s\n",
      "Epoch 25: Accuracy: 97.45000% (29235/30000), Higher: 358, Lower: 407, Time: 14.81233s\n",
      "Epoch 26: Accuracy: 97.53000% (29259/30000), Higher: 298, Lower: 443, Time: 14.57552s\n",
      "Epoch 27: Accuracy: 97.44333% (29233/30000), Higher: 349, Lower: 418, Time: 14.34470s\n",
      "Epoch 28: Accuracy: 97.40333% (29221/30000), Higher: 400, Lower: 379, Time: 14.58954s\n",
      "Epoch 29: Accuracy: 97.50000% (29250/30000), Higher: 348, Lower: 402, Time: 14.96406s\n",
      "Epoch 30: Accuracy: 97.63667% (29291/30000), Higher: 250, Lower: 459, Time: 14.81633s\n",
      "Epoch 31: Accuracy: 97.54333% (29263/30000), Higher: 301, Lower: 436, Time: 15.26718s\n",
      "Epoch 32: Accuracy: 97.45333% (29236/30000), Higher: 383, Lower: 381, Time: 15.06951s\n",
      "Epoch 33: Accuracy: 97.60333% (29281/30000), Higher: 296, Lower: 423, Time: 15.18043s\n",
      "Epoch 34: Accuracy: 97.66333% (29299/30000), Higher: 245, Lower: 456, Time: 14.92329s\n",
      "Epoch 35: Accuracy: 97.56333% (29269/30000), Higher: 277, Lower: 454, Time: 15.05722s\n",
      "Epoch 36: Accuracy: 97.55667% (29267/30000), Higher: 330, Lower: 403, Time: 15.01278s\n",
      "Epoch 37: Accuracy: 97.38000% (29214/30000), Higher: 412, Lower: 374, Time: 15.35729s\n",
      "Epoch 38: Accuracy: 97.56667% (29270/30000), Higher: 306, Lower: 424, Time: 15.01968s\n",
      "Epoch 39: Accuracy: 97.64000% (29292/30000), Higher: 276, Lower: 432, Time: 15.33887s\n",
      "Epoch 40: Accuracy: 97.60667% (29282/30000), Higher: 290, Lower: 428, Time: 14.56339s\n",
      "Epoch 41: Accuracy: 97.48667% (29246/30000), Higher: 358, Lower: 396, Time: 15.03366s\n",
      "Epoch 42: Accuracy: 97.60000% (29280/30000), Higher: 321, Lower: 399, Time: 14.66865s\n",
      "Epoch 43: Accuracy: 97.64667% (29294/30000), Higher: 277, Lower: 429, Time: 14.59375s\n",
      "Epoch 44: Accuracy: 97.58000% (29274/30000), Higher: 298, Lower: 428, Time: 15.20090s\n",
      "Epoch 45: Accuracy: 97.46333% (29239/30000), Higher: 387, Lower: 374, Time: 14.46807s\n",
      "Epoch 46: Accuracy: 97.43000% (29229/30000), Higher: 378, Lower: 393, Time: 14.97720s\n",
      "Epoch 47: Accuracy: 97.50333% (29251/30000), Higher: 334, Lower: 415, Time: 15.07027s\n",
      "Epoch 48: Accuracy: 97.54667% (29264/30000), Higher: 315, Lower: 421, Time: 14.84799s\n",
      "Epoch 49: Accuracy: 97.52000% (29256/30000), Higher: 316, Lower: 428, Time: 14.64690s\n",
      "Epoch 50: Accuracy: 97.53000% (29259/30000), Higher: 343, Lower: 398, Time: 14.83434s\n",
      "Epoch 51: Accuracy: 97.52333% (29257/30000), Higher: 320, Lower: 423, Time: 15.04540s\n",
      "Epoch 52: Accuracy: 97.43000% (29229/30000), Higher: 373, Lower: 398, Time: 14.44238s\n",
      "Epoch 53: Accuracy: 97.54667% (29264/30000), Higher: 313, Lower: 423, Time: 14.90407s\n",
      "Epoch 54: Accuracy: 97.56667% (29270/30000), Higher: 313, Lower: 417, Time: 14.57496s\n",
      "Epoch 55: Accuracy: 97.58000% (29274/30000), Higher: 300, Lower: 426, Time: 14.76584s\n",
      "Epoch 56: Accuracy: 97.60667% (29282/30000), Higher: 305, Lower: 413, Time: 14.80161s\n",
      "Epoch 57: Accuracy: 97.54333% (29263/30000), Higher: 320, Lower: 417, Time: 14.75170s\n",
      "Epoch 58: Accuracy: 97.53667% (29261/30000), Higher: 320, Lower: 419, Time: 14.71220s\n",
      "Epoch 59: Accuracy: 97.47333% (29242/30000), Higher: 346, Lower: 412, Time: 14.62200s\n",
      "Epoch 60: Accuracy: 97.54667% (29264/30000), Higher: 287, Lower: 449, Time: 14.82935s\n",
      "Epoch 61: Accuracy: 97.47000% (29241/30000), Higher: 373, Lower: 386, Time: 14.85456s\n",
      "Epoch 62: Accuracy: 97.48000% (29244/30000), Higher: 362, Lower: 394, Time: 15.01894s\n",
      "Epoch 63: Accuracy: 97.61667% (29285/30000), Higher: 290, Lower: 425, Time: 14.66514s\n",
      "Epoch 64: Accuracy: 97.55667% (29267/30000), Higher: 304, Lower: 429, Time: 15.11585s\n",
      "Epoch 65: Accuracy: 97.47667% (29243/30000), Higher: 369, Lower: 388, Time: 15.21975s\n",
      "Epoch 66: Accuracy: 97.49333% (29248/30000), Higher: 359, Lower: 393, Time: 15.09603s\n",
      "Epoch 67: Accuracy: 97.49000% (29247/30000), Higher: 351, Lower: 402, Time: 14.90727s\n",
      "Epoch 68: Accuracy: 97.46667% (29240/30000), Higher: 344, Lower: 416, Time: 15.15722s\n",
      "Epoch 69: Accuracy: 97.52333% (29257/30000), Higher: 337, Lower: 406, Time: 15.14604s\n",
      "Epoch 70: Accuracy: 97.48000% (29244/30000), Higher: 333, Lower: 423, Time: 14.89002s\n",
      "Epoch 71: Accuracy: 97.47000% (29241/30000), Higher: 363, Lower: 396, Time: 14.97560s\n",
      "Epoch 72: Accuracy: 97.59667% (29279/30000), Higher: 303, Lower: 418, Time: 14.75999s\n",
      "Epoch 73: Accuracy: 97.57333% (29272/30000), Higher: 301, Lower: 427, Time: 15.10892s\n",
      "Epoch 74: Accuracy: 97.49000% (29247/30000), Higher: 355, Lower: 398, Time: 15.07115s\n",
      "Epoch 75: Accuracy: 97.55667% (29267/30000), Higher: 308, Lower: 425, Time: 14.87535s\n",
      "Epoch 76: Accuracy: 97.57667% (29273/30000), Higher: 288, Lower: 439, Time: 15.46172s\n",
      "Epoch 77: Accuracy: 97.57000% (29271/30000), Higher: 329, Lower: 400, Time: 15.05567s\n",
      "Epoch 78: Accuracy: 97.63333% (29290/30000), Higher: 273, Lower: 437, Time: 15.58545s\n",
      "Epoch 79: Accuracy: 97.45667% (29237/30000), Higher: 368, Lower: 395, Time: 14.91566s\n",
      "Epoch 80: Accuracy: 97.58333% (29275/30000), Higher: 291, Lower: 434, Time: 15.47745s\n",
      "Epoch 81: Accuracy: 97.63000% (29289/30000), Higher: 280, Lower: 431, Time: 15.21982s\n",
      "Epoch 82: Accuracy: 97.59667% (29279/30000), Higher: 294, Lower: 427, Time: 15.28479s\n",
      "Epoch 83: Accuracy: 97.55333% (29266/30000), Higher: 328, Lower: 406, Time: 15.19210s\n",
      "Epoch 84: Accuracy: 97.54667% (29264/30000), Higher: 332, Lower: 404, Time: 14.66328s\n",
      "Epoch 85: Accuracy: 97.63333% (29290/30000), Higher: 281, Lower: 429, Time: 14.60298s\n",
      "Epoch 86: Accuracy: 97.47333% (29242/30000), Higher: 368, Lower: 390, Time: 14.55249s\n",
      "Epoch 87: Accuracy: 97.60333% (29281/30000), Higher: 310, Lower: 409, Time: 14.69119s\n",
      "Epoch 88: Accuracy: 97.50333% (29251/30000), Higher: 355, Lower: 394, Time: 14.64640s\n",
      "Epoch 89: Accuracy: 97.53333% (29260/30000), Higher: 331, Lower: 409, Time: 14.86114s\n",
      "Epoch 90: Accuracy: 97.56333% (29269/30000), Higher: 352, Lower: 379, Time: 14.79045s\n",
      "Epoch 91: Accuracy: 97.52667% (29258/30000), Higher: 328, Lower: 414, Time: 14.57164s\n",
      "Epoch 92: Accuracy: 97.55333% (29266/30000), Higher: 326, Lower: 408, Time: 14.52602s\n",
      "Epoch 93: Accuracy: 97.57667% (29273/30000), Higher: 327, Lower: 400, Time: 14.91416s\n",
      "Epoch 94: Accuracy: 97.58667% (29276/30000), Higher: 310, Lower: 414, Time: 14.66555s\n",
      "Epoch 95: Accuracy: 97.65333% (29296/30000), Higher: 286, Lower: 418, Time: 14.87132s\n",
      "Epoch 96: Accuracy: 97.58000% (29274/30000), Higher: 299, Lower: 427, Time: 14.96584s\n",
      "Epoch 97: Accuracy: 97.63333% (29290/30000), Higher: 298, Lower: 412, Time: 14.76198s\n",
      "Epoch 98: Accuracy: 97.54667% (29264/30000), Higher: 329, Lower: 407, Time: 15.02201s\n",
      "Epoch 99: Accuracy: 97.64000% (29292/30000), Higher: 290, Lower: 418, Time: 15.27873s\n",
      "Epoch 100: Accuracy: 97.47333% (29242/30000), Higher: 378, Lower: 380, Time: 15.19281s\n",
      "Epoch 101: Accuracy: 97.54667% (29264/30000), Higher: 332, Lower: 404, Time: 14.60968s\n",
      "Epoch 102: Accuracy: 97.63000% (29289/30000), Higher: 285, Lower: 426, Time: 14.47196s\n",
      "Epoch 103: Accuracy: 97.66667% (29300/30000), Higher: 273, Lower: 427, Time: 14.66670s\n",
      "Epoch 104: Accuracy: 97.68333% (29305/30000), Higher: 261, Lower: 434, Time: 14.67640s\n",
      "Epoch 105: Accuracy: 97.58333% (29275/30000), Higher: 317, Lower: 408, Time: 15.13743s\n",
      "Epoch 106: Accuracy: 97.52000% (29256/30000), Higher: 364, Lower: 380, Time: 15.53562s\n",
      "Epoch 107: Accuracy: 97.54333% (29263/30000), Higher: 327, Lower: 410, Time: 18.73018s\n",
      "Epoch 108: Accuracy: 97.65667% (29297/30000), Higher: 277, Lower: 426, Time: 15.72484s\n",
      "Epoch 109: Accuracy: 97.61667% (29285/30000), Higher: 285, Lower: 430, Time: 14.95572s\n",
      "Epoch 110: Accuracy: 97.54000% (29262/30000), Higher: 357, Lower: 381, Time: 14.71096s\n",
      "Epoch 111: Accuracy: 97.60667% (29282/30000), Higher: 296, Lower: 422, Time: 14.94641s\n",
      "Epoch 112: Accuracy: 97.53667% (29261/30000), Higher: 337, Lower: 402, Time: 14.67097s\n",
      "Epoch 113: Accuracy: 97.57667% (29273/30000), Higher: 329, Lower: 398, Time: 14.31530s\n",
      "Epoch 114: Accuracy: 97.56000% (29268/30000), Higher: 322, Lower: 410, Time: 14.73175s\n",
      "Epoch 115: Accuracy: 97.61333% (29284/30000), Higher: 290, Lower: 426, Time: 15.09669s\n",
      "Epoch 116: Accuracy: 97.57000% (29271/30000), Higher: 290, Lower: 439, Time: 15.08641s\n",
      "Epoch 117: Accuracy: 97.52667% (29258/30000), Higher: 350, Lower: 392, Time: 15.29273s\n",
      "Epoch 118: Accuracy: 97.55667% (29267/30000), Higher: 334, Lower: 399, Time: 15.05770s\n",
      "Epoch 119: Accuracy: 97.53333% (29260/30000), Higher: 332, Lower: 408, Time: 15.30790s\n",
      "Epoch 120: Accuracy: 97.61000% (29283/30000), Higher: 283, Lower: 434, Time: 15.37713s\n",
      "Epoch 121: Accuracy: 97.51667% (29255/30000), Higher: 349, Lower: 396, Time: 14.99678s\n",
      "Epoch 122: Accuracy: 97.43000% (29229/30000), Higher: 389, Lower: 382, Time: 15.18773s\n",
      "Epoch 123: Accuracy: 97.65333% (29296/30000), Higher: 274, Lower: 430, Time: 15.17799s\n",
      "Epoch 124: Accuracy: 97.55667% (29267/30000), Higher: 328, Lower: 405, Time: 15.13734s\n",
      "Epoch 125: Accuracy: 97.60667% (29282/30000), Higher: 299, Lower: 419, Time: 14.87904s\n",
      "Epoch 126: Accuracy: 97.57333% (29272/30000), Higher: 322, Lower: 406, Time: 15.18810s\n",
      "Epoch 127: Accuracy: 97.60000% (29280/30000), Higher: 305, Lower: 415, Time: 14.77991s\n",
      "Epoch 128: Accuracy: 97.50000% (29250/30000), Higher: 351, Lower: 399, Time: 14.76241s\n",
      "Epoch 129: Accuracy: 97.57333% (29272/30000), Higher: 316, Lower: 412, Time: 15.20135s\n",
      "Epoch 130: Accuracy: 97.60333% (29281/30000), Higher: 303, Lower: 416, Time: 15.00950s\n",
      "Epoch 131: Accuracy: 97.47333% (29242/30000), Higher: 380, Lower: 378, Time: 15.53671s\n",
      "Epoch 132: Accuracy: 97.62000% (29286/30000), Higher: 291, Lower: 423, Time: 14.88802s\n",
      "Epoch 133: Accuracy: 97.59667% (29279/30000), Higher: 315, Lower: 406, Time: 14.88267s\n",
      "Epoch 134: Accuracy: 97.42667% (29228/30000), Higher: 394, Lower: 378, Time: 14.85320s\n",
      "Epoch 135: Accuracy: 97.55333% (29266/30000), Higher: 336, Lower: 398, Time: 14.72088s\n",
      "Epoch 136: Accuracy: 97.62333% (29287/30000), Higher: 299, Lower: 414, Time: 14.76333s\n",
      "Epoch 137: Accuracy: 97.62000% (29286/30000), Higher: 278, Lower: 436, Time: 15.04984s\n",
      "Epoch 138: Accuracy: 97.61333% (29284/30000), Higher: 298, Lower: 418, Time: 14.50321s\n",
      "Epoch 139: Accuracy: 97.61333% (29284/30000), Higher: 323, Lower: 393, Time: 14.99361s\n",
      "Epoch 140: Accuracy: 97.49333% (29248/30000), Higher: 353, Lower: 399, Time: 15.28965s\n",
      "Epoch 141: Accuracy: 97.55667% (29267/30000), Higher: 326, Lower: 407, Time: 14.86143s\n",
      "Epoch 142: Accuracy: 97.53333% (29260/30000), Higher: 331, Lower: 409, Time: 15.28519s\n",
      "Epoch 143: Accuracy: 97.59333% (29278/30000), Higher: 330, Lower: 392, Time: 15.24245s\n",
      "Epoch 144: Accuracy: 97.60000% (29280/30000), Higher: 302, Lower: 418, Time: 15.14496s\n",
      "Epoch 145: Accuracy: 97.60667% (29282/30000), Higher: 289, Lower: 429, Time: 14.83430s\n",
      "Epoch 146: Accuracy: 97.57333% (29272/30000), Higher: 330, Lower: 398, Time: 14.98981s\n",
      "Epoch 147: Accuracy: 97.49333% (29248/30000), Higher: 350, Lower: 402, Time: 14.89897s\n",
      "Epoch 148: Accuracy: 97.63667% (29291/30000), Higher: 291, Lower: 418, Time: 14.64344s\n",
      "Epoch 149: Accuracy: 97.61000% (29283/30000), Higher: 318, Lower: 399, Time: 14.91196s\n",
      "Epoch 150: Accuracy: 97.61333% (29284/30000), Higher: 291, Lower: 425, Time: 14.86096s\n",
      "Epoch 151: Accuracy: 97.56667% (29270/30000), Higher: 333, Lower: 397, Time: 15.16549s\n",
      "Epoch 152: Accuracy: 97.62667% (29288/30000), Higher: 278, Lower: 434, Time: 15.20249s\n",
      "Epoch 153: Accuracy: 97.60333% (29281/30000), Higher: 303, Lower: 416, Time: 15.06742s\n",
      "Epoch 154: Accuracy: 97.60000% (29280/30000), Higher: 286, Lower: 434, Time: 14.96122s\n",
      "Epoch 155: Accuracy: 97.65000% (29295/30000), Higher: 289, Lower: 416, Time: 15.09208s\n",
      "Epoch 156: Accuracy: 97.56667% (29270/30000), Higher: 321, Lower: 409, Time: 15.05609s\n",
      "Epoch 157: Accuracy: 97.63000% (29289/30000), Higher: 295, Lower: 416, Time: 15.41929s\n",
      "Epoch 158: Accuracy: 97.59333% (29278/30000), Higher: 303, Lower: 419, Time: 15.60691s\n",
      "Epoch 159: Accuracy: 97.57667% (29273/30000), Higher: 324, Lower: 403, Time: 15.33196s\n",
      "Epoch 160: Accuracy: 97.51333% (29254/30000), Higher: 367, Lower: 379, Time: 15.06325s\n",
      "Epoch 161: Accuracy: 97.64667% (29294/30000), Higher: 265, Lower: 441, Time: 14.47985s\n",
      "Epoch 162: Accuracy: 97.55667% (29267/30000), Higher: 325, Lower: 408, Time: 14.98964s\n",
      "Epoch 163: Accuracy: 97.50667% (29252/30000), Higher: 343, Lower: 405, Time: 15.11756s\n",
      "Epoch 164: Accuracy: 97.50000% (29250/30000), Higher: 356, Lower: 394, Time: 14.65327s\n",
      "Epoch 165: Accuracy: 97.57333% (29272/30000), Higher: 335, Lower: 393, Time: 15.22639s\n",
      "Epoch 166: Accuracy: 97.60000% (29280/30000), Higher: 296, Lower: 424, Time: 15.00091s\n",
      "Epoch 167: Accuracy: 97.53667% (29261/30000), Higher: 330, Lower: 409, Time: 15.20324s\n",
      "Epoch 168: Accuracy: 97.66333% (29299/30000), Higher: 292, Lower: 409, Time: 14.98915s\n",
      "Epoch 169: Accuracy: 97.54667% (29264/30000), Higher: 330, Lower: 406, Time: 15.16564s\n",
      "Epoch 170: Accuracy: 97.55333% (29266/30000), Higher: 331, Lower: 403, Time: 15.28238s\n",
      "Epoch 171: Accuracy: 97.65333% (29296/30000), Higher: 275, Lower: 429, Time: 15.14553s\n",
      "Epoch 172: Accuracy: 97.58333% (29275/30000), Higher: 310, Lower: 415, Time: 15.47635s\n",
      "Epoch 173: Accuracy: 97.57000% (29271/30000), Higher: 328, Lower: 401, Time: 15.32724s\n",
      "Epoch 174: Accuracy: 97.57667% (29273/30000), Higher: 319, Lower: 408, Time: 15.46750s\n",
      "Epoch 175: Accuracy: 97.60667% (29282/30000), Higher: 291, Lower: 427, Time: 15.27580s\n",
      "Epoch 176: Accuracy: 97.60000% (29280/30000), Higher: 301, Lower: 419, Time: 15.33070s\n",
      "Epoch 177: Accuracy: 97.61000% (29283/30000), Higher: 285, Lower: 432, Time: 15.01757s\n",
      "Epoch 178: Accuracy: 97.66000% (29298/30000), Higher: 272, Lower: 430, Time: 15.38488s\n",
      "Epoch 179: Accuracy: 97.55333% (29266/30000), Higher: 315, Lower: 419, Time: 15.59652s\n",
      "Epoch 180: Accuracy: 97.45000% (29235/30000), Higher: 371, Lower: 394, Time: 15.46246s\n",
      "Epoch 181: Accuracy: 97.64333% (29293/30000), Higher: 296, Lower: 411, Time: 15.05955s\n",
      "Epoch 182: Accuracy: 97.64333% (29293/30000), Higher: 287, Lower: 420, Time: 14.75176s\n",
      "Epoch 183: Accuracy: 97.63000% (29289/30000), Higher: 292, Lower: 419, Time: 15.19836s\n",
      "Epoch 184: Accuracy: 97.61000% (29283/30000), Higher: 316, Lower: 401, Time: 15.57669s\n",
      "Epoch 185: Accuracy: 97.55333% (29266/30000), Higher: 338, Lower: 396, Time: 14.36560s\n",
      "Epoch 186: Accuracy: 97.54333% (29263/30000), Higher: 329, Lower: 408, Time: 14.98146s\n",
      "Epoch 187: Accuracy: 97.58667% (29276/30000), Higher: 306, Lower: 418, Time: 14.97741s\n",
      "Epoch 188: Accuracy: 97.64667% (29294/30000), Higher: 283, Lower: 423, Time: 14.93967s\n",
      "Epoch 189: Accuracy: 97.54667% (29264/30000), Higher: 345, Lower: 391, Time: 14.91928s\n",
      "Epoch 190: Accuracy: 97.56000% (29268/30000), Higher: 333, Lower: 399, Time: 14.86379s\n",
      "Epoch 191: Accuracy: 97.53667% (29261/30000), Higher: 356, Lower: 383, Time: 15.08708s\n",
      "Epoch 192: Accuracy: 97.56667% (29270/30000), Higher: 328, Lower: 402, Time: 15.24622s\n",
      "Epoch 193: Accuracy: 97.56333% (29269/30000), Higher: 314, Lower: 417, Time: 14.78886s\n",
      "Epoch 194: Accuracy: 97.61333% (29284/30000), Higher: 307, Lower: 409, Time: 15.26291s\n",
      "Epoch 195: Accuracy: 97.59000% (29277/30000), Higher: 314, Lower: 409, Time: 15.57620s\n",
      "Epoch 196: Accuracy: 97.56333% (29269/30000), Higher: 324, Lower: 407, Time: 14.83665s\n",
      "Epoch 197: Accuracy: 97.58333% (29275/30000), Higher: 302, Lower: 423, Time: 14.88044s\n",
      "Epoch 198: Accuracy: 97.60000% (29280/30000), Higher: 307, Lower: 413, Time: 14.94570s\n",
      "Epoch 199: Accuracy: 97.62667% (29288/30000), Higher: 279, Lower: 433, Time: 15.11564s\n"
     ]
    }
   ],
   "execution_count": 12
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:39.974615Z",
     "start_time": "2024-05-27T03:44:39.795245Z"
    }
   },
   "cell_type": "code",
   "source": [
    "plt.plot(accuracy)\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.show()"
   ],
   "id": "ca130e6f21c3b2fd",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 13
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:54.085807Z",
     "start_time": "2024-05-27T03:44:39.975637Z"
    }
   },
   "cell_type": "code",
   "source": [
    "val_set = data.iloc[:,:-2]\n",
    "val_dose = data.iloc[:,-1:].values\n",
    "arms = [0, 1, 2]\n",
    "correct_count = 0\n",
    "count = 0\n",
    "heatmap = np.zeros((3,3))\n",
    "\n",
    "for i in range(len(data)):\n",
    "    count += 1\n",
    "    feature = np.array(val_set.iloc[i], dtype=float)\n",
    "    # iterate through actions\n",
    "    p = {}\n",
    "    max_value = -np.inf\n",
    "    max_arm = 0\n",
    "    for arm in arms:\n",
    "        theta = np.matmul(np.linalg.inv(A_map[arm]), b_map[arm])\n",
    "        p = np.matmul(theta.T, feature) + \\\n",
    "            np.sqrt(np.matmul(np.matmul(feature.T,np.linalg.inv(A_map[arm])), feature))\n",
    "        if p == max_value:\n",
    "            if np.random.choice(a=[False, True]): max_arm = arm\n",
    "        elif p > max_value:\n",
    "            max_value = p\n",
    "            max_arm = arm\n",
    "\n",
    "    if max_arm == val_dose[i]:\n",
    "        correct_count += 1\n",
    "    heatmap[max_arm][val_dose[i][0]] += 1\n",
    "\n",
    "accu = correct_count/count\n",
    "print(\"Validate: Accuracy: %.5f%% (%d/%d)\"%(accu*100,correct_count,count))\n",
    "ax = sns.heatmap(heatmap, linewidth=0.5,annot=True,cmap='Blues', fmt='.20g')\n",
    "plt.show()"
   ],
   "id": "ae3f2b95561b7b31",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validate: Accuracy: 97.63100% (97631/100000)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 14
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": [
    "# Test real data\n",
    "## Filter"
   ],
   "id": "76ff9ae1799651d5"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:54.611484Z",
     "start_time": "2024-05-27T03:44:54.086707Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# https://api.pharmgkb.org/v1/download/submission/553247439\n",
    "warfarin = pd.read_excel('PS206767-553247439.xls', sheet_name=1)"
   ],
   "id": "6e046944ff587c75",
   "outputs": [],
   "execution_count": 15
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:54.617768Z",
     "start_time": "2024-05-27T03:44:54.612533Z"
    }
   },
   "cell_type": "code",
   "source": [
    "warfarin_clean = warfarin.loc[:,\n",
    "                 [\n",
    "                     'Race (OMB)',\n",
    "                     'Age',\n",
    "                     'Height (cm)',\n",
    "                     'Weight (kg)',\n",
    "\n",
    "                     # drug\n",
    "                     'Amiodarone (Cordarone)',\n",
    "                     'Current Smoker',\n",
    "\n",
    "                     'VKORC1 -1639 consensus',\n",
    "                     'CYP2C9 consensus',\n",
    "                     'Target INR',\n",
    "                     'Estimated Target INR Range Based on Indication',\n",
    "                     'Therapeutic Dose of Warfarin', ]]\n",
    "warfarin_clean = warfarin_clean.dropna(subset=['Height (cm)',\n",
    "                                               'Weight (kg)',\n",
    "                                               'Therapeutic Dose of Warfarin',\n",
    "                                              ])"
   ],
   "id": "45f23b94f87a69c2",
   "outputs": [],
   "execution_count": 16
  },
  {
   "metadata": {},
   "cell_type": "markdown",
   "source": "## Convert to Gage Form",
   "id": "d2165e55c59e233f"
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:54.728536Z",
     "start_time": "2024-05-27T03:44:54.618750Z"
    }
   },
   "cell_type": "code",
   "source": [
    "data = pd.DataFrame(columns=[\n",
    "    'Age',\n",
    "    'Height (cm)',\n",
    "    'Weight (kg)',\n",
    "    'Amiodarone',\n",
    "    'Smokes',\n",
    "    'Race Black or African American',\n",
    "    'CYP2C9 *2',\n",
    "    'CYP2C9 *3',\n",
    "    'VKORC1',\n",
    "    'Target INR',\n",
    "    'dose',\n",
    "    'type'])"
   ],
   "id": "39cce0e625d0a81e",
   "outputs": [],
   "execution_count": 17
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:55.112284Z",
     "start_time": "2024-05-27T03:44:54.729559Z"
    }
   },
   "cell_type": "code",
   "source": [
    "warfarin_data = []\n",
    "\n",
    "for index, row in warfarin_clean.iterrows():\n",
    "    sample = {\n",
    "        'Age': 0,\n",
    "        'Height (cm)': row['Height (cm)'],\n",
    "        'Weight (kg)': row['Weight (kg)'],\n",
    "\n",
    "        'Amiodarone': row['Amiodarone (Cordarone)'],\n",
    "        'Smokes': row['Current Smoker'],\n",
    "\n",
    "        'Race Black or African American': 0,\n",
    "\n",
    "        'CYP2C9 *2': 0,\n",
    "        'CYP2C9 *3': 0,\n",
    "        'VKORC1': 0,\n",
    "        'Target INR': 0,\n",
    "\n",
    "        'dose': row['Therapeutic Dose of Warfarin'],\n",
    "        'type': 0\n",
    "    }\n",
    "    if not pd.isnull(row['Target INR']):\n",
    "        sample['Target INR'] = row['Target INR']\n",
    "    else:\n",
    "        if not pd.isnull(row['Estimated Target INR Range Based on Indication']):\n",
    "            sample['Target INR'] = float(str(row['Estimated Target INR Range Based on Indication']).split('-')[1])\n",
    "        else:\n",
    "            continue\n",
    "\n",
    "    age = str(row['Age']).split('-')\n",
    "    if len(age) == 2:\n",
    "        sample['Age'] = int(age[1])\n",
    "    else:\n",
    "        sample['Age'] = 100\n",
    "    match row['Race (OMB)']:\n",
    "        case 'Black or African American':\n",
    "            sample['Race Black or African American'] = 1\n",
    "\n",
    "    if pd.isnull(row['VKORC1 -1639 consensus']):\n",
    "        pass\n",
    "    else:\n",
    "        sample['VKORC1'] = str(row['VKORC1 -1639 consensus']).count('A')\n",
    "\n",
    "    sample['CYP2C9 *2'] = str(row['CYP2C9 consensus']).count('*2')\n",
    "    sample['CYP2C9 *3'] = str(row['CYP2C9 consensus']).count('*3')\n",
    "\n",
    "    if sample['dose'] < 21:\n",
    "        sample['type'] = 0\n",
    "    elif 21 <= sample['dose'] < 49:\n",
    "        sample['type'] = 1\n",
    "    else:\n",
    "        sample['type'] = 2\n",
    "    warfarin_data.append(sample)\n",
    "\n",
    "data = data._append(warfarin_data, ignore_index=True)"
   ],
   "id": "8bb79d9c1ce668ad",
   "outputs": [],
   "execution_count": 18
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:55.119599Z",
     "start_time": "2024-05-27T03:44:55.113098Z"
    }
   },
   "cell_type": "code",
   "source": [
    "data_size = len(data)\n",
    "print(\"Sample:\", data_size)\n",
    "print(\"Age: %.2f ± %.2f\" % (data['Age'].mean(), data['Age'].std()))\n",
    "print(\"Height (cm): %.2f ± %.2f\" % (data['Height (cm)'].mean(), data['Height (cm)'].std()))\n",
    "print(\"Weight (kg): %.2f ± %.2f\" % (data['Weight (kg)'].mean(), data['Weight (kg)'].std()))\n",
    "print(\"Target INR: %.2f ± %.2f\" % (data['Target INR'].mean(), data['Target INR'].std()))\n",
    "aminodarone = data['Amiodarone'].sum()\n",
    "print(\"Amiodarone: %2.2f%% (%d)\" % (aminodarone / data_size * 100, aminodarone))\n",
    "smoke = data['Smokes'].sum()\n",
    "print(\"Smokes: %2.2f%% (%d)\" % (smoke / data_size * 100, smoke))"
   ],
   "id": "348d6e89ab35ffae",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Sample: 3663\n",
      "Age: 67.92 ± 15.13\n",
      "Height (cm): 169.61 ± 10.55\n",
      "Weight (kg): 81.90 ± 22.09\n",
      "Target INR: 2.95 ± 0.35\n",
      "Amiodarone: 5.43% (199)\n",
      "Smokes: 10.37% (380)\n"
     ]
    }
   ],
   "execution_count": 19
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:55.254552Z",
     "start_time": "2024-05-27T03:44:55.120495Z"
    }
   },
   "cell_type": "code",
   "source": [
    "values = [data['Race Black or African American'].sum(),\n",
    "          data_size - data['Race Black or African American'].sum()]\n",
    "labels = ['Race Black or African American',\n",
    "          'Race Other']\n",
    "_ = plt.pie(values, labels=labels, autopct=pct)\n",
    "plt.show()"
   ],
   "id": "76feafc133f1a560",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "12.1485% (445)\n",
      "87.8515% (3218)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 20
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:55.340230Z",
     "start_time": "2024-05-27T03:44:55.255567Z"
    }
   },
   "cell_type": "code",
   "source": [
    "values = [\n",
    "    len(data[(data['CYP2C9 *2'] == 0) & (data['CYP2C9 *3'] == 0)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 1) & (data['CYP2C9 *3'] == 0)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 0) & (data['CYP2C9 *3'] == 1)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 1) & (data['CYP2C9 *3'] == 1)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 2) & (data['CYP2C9 *3'] == 0)]),\n",
    "    len(data[(data['CYP2C9 *2'] == 0) & (data['CYP2C9 *3'] == 2)]),\n",
    "]\n",
    "labels = ['Other', '*2/*?', '*?/*3', '*2/*3', '*2/*2', '*3/*3']\n",
    "_ = plt.pie(values, labels=labels, autopct=pct)\n",
    "plt.show()"
   ],
   "id": "637bf8f6bffe0a1d",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "71.9083% (2634)\n",
      "15.8613% (581)\n",
      "9.2547% (339)\n",
      "1.4742% (54)\n",
      "1.2012% (44)\n",
      "0.3003% (11)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 21
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:55.407103Z",
     "start_time": "2024-05-27T03:44:55.341230Z"
    }
   },
   "cell_type": "code",
   "source": [
    "values = [data['VKORC1'].value_counts()[2],\n",
    "          data['VKORC1'].value_counts()[1],\n",
    "          data['VKORC1'].value_counts()[0],\n",
    "          ]\n",
    "labels = ['A/A', 'A/?', 'unknown']\n",
    "_ = plt.pie(values, labels=labels, autopct=pct)\n",
    "plt.show()"
   ],
   "id": "b4c1926d19ab93b7",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "18.1545% (665)\n",
      "30.4668% (1116)\n",
      "51.3786% (1882)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 22
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:55.443575Z",
     "start_time": "2024-05-27T03:44:55.408228Z"
    }
   },
   "cell_type": "code",
   "source": [
    "val_set = data.iloc[:, :-2]\n",
    "val_dose = data.iloc[:, -1:].values"
   ],
   "id": "f271a7c94ef177d0",
   "outputs": [],
   "execution_count": 23
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:56.197595Z",
     "start_time": "2024-05-27T03:44:55.444921Z"
    }
   },
   "cell_type": "code",
   "source": [
    "arms = [0, 1, 2]\n",
    "correct_count = 0\n",
    "count = 0\n",
    "heatmap = np.zeros((3,3))\n",
    "\n",
    "for i in range(len(data)):\n",
    "    count += 1\n",
    "    feature = np.array(val_set.iloc[i], dtype=float)\n",
    "    # iterate through actions\n",
    "    p = {}\n",
    "    max_value = -np.inf\n",
    "    max_arm = 0\n",
    "    for arm in arms:\n",
    "        theta = np.matmul(np.linalg.inv(A_map[arm]), b_map[arm])\n",
    "        p = np.matmul(theta.T, feature) + \\\n",
    "            np.sqrt(np.matmul(np.matmul(feature.T,np.linalg.inv(A_map[arm])), feature))\n",
    "        if p == max_value:\n",
    "            if np.random.choice(a=[False, True]): max_arm = arm\n",
    "        elif p > max_value:\n",
    "            max_value = p\n",
    "            max_arm = arm\n",
    "\n",
    "    if max_arm == val_dose[i]:\n",
    "        correct_count += 1\n",
    "    heatmap[max_arm][val_dose[i][0]] += 1\n",
    "\n",
    "accu = correct_count/count\n",
    "print(\"Validate: Accuracy: %.5f%% (%d/%d)\"%(accu*100,correct_count,count))\n",
    "ax = sns.heatmap(heatmap, linewidth=0.5,annot=True,cmap='Blues', fmt='.20g')\n",
    "plt.show()"
   ],
   "id": "a89759ad1d9d5dba",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Validate: Accuracy: 58.28556% (2135/3663)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 24
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:56.417705Z",
     "start_time": "2024-05-27T03:44:56.198871Z"
    }
   },
   "cell_type": "code",
   "source": [
    "correct_count = 0\n",
    "for index, row in data.iterrows():\n",
    "    if calc_dose(row)[1] == row['type']:\n",
    "        correct_count += 1\n",
    "print(\"Correct: %f (%d/%d)\"%(correct_count/count*100,correct_count,count))"
   ],
   "id": "21a767ded90acbdd",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Correct: 51.324051 (1880/3663)\n"
     ]
    }
   ],
   "execution_count": 25
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-05-27T03:44:56.420647Z",
     "start_time": "2024-05-27T03:44:56.418665Z"
    }
   },
   "cell_type": "code",
   "source": "",
   "id": "9ea3a9160ee901dc",
   "outputs": [],
   "execution_count": 25
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}