ddqn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 140,
   "id": "2f30013f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Here we import all libraries\n",
    "import numpy as np\n",
    "import gym\n",
    "import matplotlib.pyplot as plt\n",
    "import os\n",
    "import torch\n",
    "import random\n",
    "from torch import nn\n",
    "from torch.utils.data import DataLoader\n",
    "from torchvision import datasets, transforms\n",
    "from collections import deque \n",
    "import sys\n",
    "env = gym.make(\"CartPole-v0\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "id": "63c20bea",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Hyperparameters\n",
    "episodes = 20000*5\n",
    "eps = 1.0\n",
    "learning_rate = 0.001\n",
    "tot_rewards = []\n",
    "tot_loss = []\n",
    "decay_val = 0.0001\n",
    "mem_size = 1000000\n",
    "batch_size = 300\n",
    "gamma = 0.99\n",
    "update_target = 100\n",
    "max_steps = 200"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "id": "356612d0",
   "metadata": {},
   "outputs": [],
   "source": [
    "class NeuralNetwork(nn.Module):\n",
    "    def __init__(self, state_size, action_size):\n",
    "        super(NeuralNetwork, self).__init__()\n",
    "        self.state_size = state_size\n",
    "        self.action_size = action_size\n",
    "        self.linear_relu_stack = nn.Sequential(\n",
    "            nn.Linear(state_size, 300),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(300, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, 128),\n",
    "            nn.ReLU(),\n",
    "            nn.Linear(128, action_size)\n",
    "        )\n",
    "    def forward(self, x):\n",
    "        x = self.linear_relu_stack(x)\n",
    "        return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "id": "2a48df98",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = NeuralNetwork(env.observation_space.shape[0], env.action_space.n)\n",
    "target = NeuralNetwork(env.observation_space.shape[0], env.action_space.n)\n",
    "\n",
    "opt = torch.optim.Adam(params=model.parameters(), lr=learning_rate)\n",
    "replay_buffer = deque(maxlen=mem_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "id": "5f5953d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Testing code\n",
    "\n",
    "# state = torch.tensor(env.reset(), dtype=torch.float32)\n",
    "# print(\"state = \", state)\n",
    "# out = model(state)\n",
    "# print(\"out = \", out)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "id": "9582dac6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_td_loss(batch_size):\n",
    "    state, next_state, reward, done, action = zip(*random.sample(replay_buffer, batch_size))\n",
    "    state = torch.stack(list(state), dim=0).reshape(batch_size, -1)\n",
    "\n",
    "    next_state = torch.from_numpy(np.array(next_state).reshape(batch_size, -1)).type(torch.float32)\n",
    "    \n",
    "    reward = torch.from_numpy(np.array(reward))\n",
    "    done = torch.from_numpy(np.array(done)).long()\n",
    "    action = torch.from_numpy(np.array(action)).type(torch.int64)\n",
    "    \n",
    "    q_values = model(state)\n",
    "    \n",
    "    next_q_values_m = model(next_state)\n",
    "    action_indices = torch.argmax(next_q_values_m, dim = -1).reshape(-1,1)\n",
    "\n",
    "\n",
    "    next_q_values_t = target(next_state)\n",
    "\n",
    "    max_next_q_values = next_q_values_t.gather(dim = -1, index = action_indices).reshape(-1)\n",
    "    q_vals = q_values.gather(dim=-1, index=action.reshape(-1,1))\n",
    "\n",
    "    loss = ((reward + gamma*max_next_q_values*(1-done).detach() - q_vals.squeeze())**2).mean()\n",
    "\n",
    "    opt.zero_grad()\n",
    "    loss.backward()\n",
    "    opt.step()\n",
    "    return loss\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "id": "6d21954b",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode =  0  Epsilon =  1.0\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode =  1  Epsilon =  0.9982017088605984\n",
      "Episode =  2  Epsilon =  0.9939188703524563\n",
      "Episode =  3  Epsilon =  0.9903473735423333\n",
      "Episode =  4  Epsilon =  0.9879735083147863\n",
      "Episode =  5  Epsilon =  0.9864927329490045\n",
      "Episode =  6  Epsilon =  0.9850141769703474\n",
      "Episode =  7  Epsilon =  0.9835378370523961\n",
      "Episode =  8  Epsilon =  0.9818673265897263\n",
      "Episode =  9  Epsilon =  0.9779478973357154\n",
      "Episode =  10  Epsilon =  0.9752136097242452\n",
      "Episode =  11  Epsilon =  0.9726814741462475\n",
      "Episode =  12  Epsilon =  0.9711265058410872\n",
      "Episode =  13  Epsilon =  0.9676369099381467\n",
      "Episode =  14  Epsilon =  0.9659934067385525\n",
      "Episode =  15  Epsilon =  0.9649314512706114\n",
      "Episode =  16  Epsilon =  0.962907322506727\n",
      "Episode =  17  Epsilon =  0.9614641163572715\n",
      "Episode =  18  Epsilon =  0.9582008522146408\n",
      "Episode =  19  Epsilon =  0.9518026842039732\n",
      "Episode =  20  Epsilon =  0.9472452058877892\n",
      "Episode =  21  Epsilon =  0.9458254741293678\n",
      "Episode =  22  Epsilon =  0.9446912209601311\n",
      "Episode =  23  Epsilon =  0.9431809990161535\n",
      "Episode =  24  Epsilon =  0.9419557215831661\n",
      "Episode =  25  Epsilon =  0.9396978512710986\n",
      "Episode =  26  Epsilon =  0.9381018017515383\n",
      "Episode =  27  Epsilon =  0.9356660267333141\n",
      "Episode =  28  Epsilon =  0.9337966581384062\n",
      "Episode =  29  Epsilon =  0.9328633748630615\n",
      "Episode =  30  Epsilon =  0.9306272989297062\n",
      "Episode =  31  Epsilon =  0.9296971832711197\n",
      "Episode =  32  Epsilon =  0.9270978019568776\n",
      "Episode =  33  Epsilon =  0.9228431579796557\n",
      "Episode =  34  Epsilon =  0.9215521460272273\n",
      "Episode =  35  Epsilon =  0.9204470019273496\n",
      "Episode =  36  Epsilon =  0.9175981766208114\n",
      "Episode =  37  Epsilon =  0.9163145021379915\n",
      "Episode =  38  Epsilon =  0.9143923567481823\n",
      "Episode =  39  Epsilon =  0.9100140087906435\n",
      "Episode =  40  Epsilon =  0.9085592232534118\n",
      "Episode =  41  Epsilon =  0.9067440113828967\n",
      "Episode =  42  Epsilon =  0.9052039330036876\n",
      "Episode =  43  Epsilon =  0.9039375974549135\n",
      "Episode =  44  Epsilon =  0.9028535770803838\n",
      "Episode =  45  Epsilon =  0.9018610337708306\n",
      "Episode =  46  Epsilon =  0.8996992705296256\n",
      "Episode =  47  Epsilon =  0.89673530436655\n",
      "Episode =  48  Epsilon =  0.8945858275120446\n",
      "Episode =  49  Epsilon =  0.8907475585560833\n",
      "Episode =  50  Epsilon =  0.8891457351140882\n",
      "Episode =  51  Epsilon =  0.8862165364815612\n",
      "Episode =  52  Epsilon =  0.8816204006042064\n",
      "Episode =  53  Epsilon =  0.8765220566826577\n",
      "Episode =  54  Epsilon =  0.8750333094161187\n",
      "Episode =  55  Epsilon =  0.87223781777024\n",
      "Episode =  56  Epsilon =  0.8710176001868247\n",
      "Episode =  57  Epsilon =  0.8665002897133373\n",
      "Episode =  58  Epsilon =  0.8644232862675134\n",
      "Episode =  59  Epsilon =  0.8609726714804701\n",
      "Episode =  60  Epsilon =  0.8595962854267146\n",
      "Episode =  61  Epsilon =  0.8567644343989702\n",
      "Episode =  62  Epsilon =  0.852832564823808\n",
      "Episode =  63  Epsilon =  0.8498530170718644\n",
      "Episode =  64  Epsilon =  0.8489187394131038\n",
      "Episode =  65  Epsilon =  0.8464605640524834\n",
      "Episode =  66  Epsilon =  0.8454454713068644\n",
      "Episode =  67  Epsilon =  0.8414813986596024\n",
      "Episode =  68  Epsilon =  0.8383738254161801\n",
      "Episode =  69  Epsilon =  0.8328590526701477\n",
      "Episode =  70  Epsilon =  0.8316107629559164\n",
      "Episode =  71  Epsilon =  0.8268016209765505\n",
      "Episode =  72  Epsilon =  0.8241602163587296\n",
      "Episode =  73  Epsilon =  0.8205420580968275\n",
      "Episode =  74  Epsilon =  0.818738939880682\n",
      "Episode =  75  Epsilon =  0.8162865261395134\n",
      "Episode =  76  Epsilon =  0.8149815771815924\n",
      "Episode =  77  Epsilon =  0.8117283243758593\n",
      "Episode =  78  Epsilon =  0.8095397232692787\n",
      "Episode =  79  Epsilon =  0.8059051601397533\n",
      "Episode =  80  Epsilon =  0.8046972689379669\n",
      "Episode =  81  Epsilon =  0.8030092620996244\n",
      "Episode =  82  Epsilon =  0.8018057112717893\n",
      "Episode =  83  Epsilon =  0.799244161736065\n",
      "Episode =  84  Epsilon =  0.7982058712744848\n",
      "Episode =  85  Epsilon =  0.795894542858262\n",
      "Episode =  86  Epsilon =  0.7915293563943553\n",
      "Episode =  87  Epsilon =  0.7882120701363552\n",
      "Episode =  88  Epsilon =  0.7867946353546594\n",
      "Episode =  89  Epsilon =  0.7825576064451758\n",
      "Episode =  90  Epsilon =  0.7781877493574582\n",
      "Episode =  91  Epsilon =  0.7731461838980623\n",
      "Episode =  92  Epsilon =  0.770059930094993\n",
      "Episode =  93  Epsilon =  0.7689057637483616\n",
      "Episode =  94  Epsilon =  0.7680604746662125\n",
      "Episode =  95  Epsilon =  0.7655301788966425\n",
      "Episode =  96  Epsilon =  0.7604945672889574\n",
      "Episode =  97  Epsilon =  0.7594306729883525\n",
      "Episode =  98  Epsilon =  0.7567017739735613\n",
      "Episode =  99  Epsilon =  0.755567628840404\n",
      "Episode =  100  Epsilon =  0.7490979302885523\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode =  101  Epsilon =  0.7458838850372449\n",
      "Episode =  102  Epsilon =  0.7420155486394936\n",
      "Episode =  103  Epsilon =  0.7399409146761808\n",
      "Episode =  104  Epsilon =  0.7370609087685835\n",
      "Episode =  105  Epsilon =  0.7325787458242249\n",
      "Episode =  106  Epsilon =  0.7255074368399572\n",
      "Episode =  107  Epsilon =  0.7197268196683503\n",
      "Episode =  108  Epsilon =  0.7124945249640197\n",
      "Episode =  109  Epsilon =  0.7113555021355117\n",
      "Episode =  110  Epsilon =  0.7097923183213503\n",
      "Episode =  111  Epsilon =  0.7085867566750061\n",
      "Episode =  112  Epsilon =  0.7052643787385493\n",
      "Episode =  113  Epsilon =  0.7040665076661337\n",
      "Episode =  114  Epsilon =  0.7013261325075206\n",
      "Episode =  115  Epsilon =  0.6990855876924104\n",
      "Episode =  116  Epsilon =  0.6973401437079447\n",
      "Episode =  117  Epsilon =  0.6895738945784886\n",
      "Episode =  118  Epsilon =  0.6879897762599176\n",
      "Episode =  119  Epsilon =  0.6868899277228646\n",
      "Episode =  120  Epsilon =  0.67537959966366\n",
      "Episode =  121  Epsilon =  0.6744347769946587\n",
      "Episode =  122  Epsilon =  0.670400485022109\n",
      "Episode =  123  Epsilon =  0.66906109086126\n",
      "Episode =  124  Epsilon =  0.6652584805052616\n",
      "Episode =  125  Epsilon =  0.662271691994027\n",
      "Episode =  126  Epsilon =  0.6609485383612801\n",
      "Episode =  127  Epsilon =  0.6539798316765466\n",
      "Episode =  128  Epsilon =  0.6522165557869328\n",
      "Episode =  129  Epsilon =  0.6504580340834758\n",
      "Episode =  130  Epsilon =  0.6445015597812893\n",
      "Episode =  131  Epsilon =  0.6421214289718492\n",
      "Episode =  132  Epsilon =  0.6401980473672821\n",
      "Episode =  133  Epsilon =  0.6393663721947541\n",
      "Episode =  134  Epsilon =  0.6373875042127369\n",
      "Episode =  135  Epsilon =  0.6332581269151074\n",
      "Episode =  136  Epsilon =  0.6303519741122208\n",
      "Episode =  137  Epsilon =  0.6262055566289257\n",
      "Episode =  138  Epsilon =  0.6178707865703827\n",
      "Episode =  139  Epsilon =  0.6159584484028765\n",
      "Episode =  140  Epsilon =  0.6150352494617471\n",
      "Episode =  141  Epsilon =  0.6123351738878097\n",
      "Episode =  142  Epsilon =  0.6053945367265627\n",
      "Episode =  143  Epsilon =  0.6005709417130629\n",
      "Episode =  144  Epsilon =  0.5964414717395492\n",
      "Episode =  145  Epsilon =  0.5937042781984301\n",
      "Episode =  146  Epsilon =  0.5893862329063079\n",
      "Episode =  147  Epsilon =  0.580611850503372\n",
      "Episode =  148  Epsilon =  0.5786412207104343\n",
      "Episode =  149  Epsilon =  0.5776584153958713\n",
      "Episode =  150  Epsilon =  0.5745476221903025\n",
      "Episode =  151  Epsilon =  0.5705970861430193\n",
      "Episode =  152  Epsilon =  0.5661639621818487\n",
      "Episode =  153  Epsilon =  0.5650893256126474\n",
      "Episode =  154  Epsilon =  0.5594109314978001\n",
      "Episode =  155  Epsilon =  0.5529595774456594\n",
      "Episode =  156  Epsilon =  0.551578974004397\n",
      "Episode =  157  Epsilon =  0.550036790050143\n",
      "Episode =  158  Epsilon =  0.5487732222703059\n",
      "Episode =  159  Epsilon =  0.5401711706036744\n",
      "Episode =  160  Epsilon =  0.53887637890424\n",
      "Episode =  161  Epsilon =  0.5379073222868237\n",
      "Episode =  162  Epsilon =  0.5369937023161675\n",
      "Episode =  163  Epsilon =  0.532768375346328\n",
      "Episode =  164  Epsilon =  0.5283649176279261\n",
      "Episode =  165  Epsilon =  0.5263084100690735\n",
      "Episode =  166  Epsilon =  0.5215931207674138\n",
      "Episode =  167  Epsilon =  0.5140335582581925\n",
      "Episode =  168  Epsilon =  0.5097339748748769\n",
      "Episode =  169  Epsilon =  0.5062797143513968\n",
      "Episode =  170  Epsilon =  0.5010920549788501\n",
      "Episode =  171  Epsilon =  0.49809465371209205\n",
      "Episode =  172  Epsilon =  0.49541232550012343\n",
      "Episode =  173  Epsilon =  0.48808576657616437\n",
      "Episode =  174  Epsilon =  0.48005081915438147\n",
      "Episode =  175  Epsilon =  0.4786128965541388\n",
      "Episode =  176  Epsilon =  0.4776566751114731\n",
      "Episode =  177  Epsilon =  0.47475198294310506\n",
      "Episode =  178  Epsilon =  0.4731406475888012\n",
      "Episode =  179  Epsilon =  0.46993436521097703\n",
      "Episode =  180  Epsilon =  0.46605024557590025\n",
      "Episode =  181  Epsilon =  0.46095203931928114\n",
      "Episode =  182  Epsilon =  0.45920383267493736\n",
      "Episode =  183  Epsilon =  0.4555450396348683\n",
      "Episode =  184  Epsilon =  0.4470165132381376\n",
      "Episode =  185  Epsilon =  0.4381654313582386\n",
      "Episode =  186  Epsilon =  0.4364163575646064\n",
      "Episode =  187  Epsilon =  0.43220379348501947\n",
      "Episode =  188  Epsilon =  0.43065073466321924\n",
      "Episode =  189  Epsilon =  0.42713401273640494\n",
      "Episode =  190  Epsilon =  0.4186766112569079\n",
      "Episode =  191  Epsilon =  0.41817452573881536\n",
      "Episode =  192  Epsilon =  0.4176312792039801\n",
      "Episode =  193  Epsilon =  0.41617219756191065\n",
      "Episode =  194  Epsilon =  0.41248488225751173\n",
      "Episode =  195  Epsilon =  0.40821748179243034\n",
      "Episode =  196  Epsilon =  0.40715754784677977\n",
      "Episode =  197  Epsilon =  0.40658795456730823\n",
      "Episode =  198  Epsilon =  0.40322740812799507\n",
      "Episode =  199  Epsilon =  0.402784123993861\n",
      "Episode =  200  Epsilon =  0.39869681102268983\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode =  201  Epsilon =  0.39480885875014965\n",
      "Episode =  202  Epsilon =  0.39029478822602376\n",
      "Episode =  203  Epsilon =  0.3867207207927243\n",
      "Episode =  204  Epsilon =  0.38602528434709976\n",
      "Episode =  205  Epsilon =  0.3854081685716841\n",
      "Episode =  206  Epsilon =  0.38429215967733493\n",
      "Episode =  207  Epsilon =  0.38340934747314015\n",
      "Episode =  208  Epsilon =  0.3821462450457608\n",
      "Episode =  209  Epsilon =  0.37910142285096915\n",
      "Episode =  210  Epsilon =  0.3721156547007904\n",
      "Episode =  211  Epsilon =  0.37170657297056386\n",
      "Episode =  212  Epsilon =  0.36992674541343407\n",
      "Episode =  213  Epsilon =  0.36536820306325085\n",
      "Episode =  214  Epsilon =  0.3647111646610632\n",
      "Episode =  215  Epsilon =  0.36155209734685817\n",
      "Episode =  216  Epsilon =  0.3575255061824052\n",
      "Episode =  217  Epsilon =  0.35688257123252726\n",
      "Episode =  218  Epsilon =  0.35449956912441277\n",
      "Episode =  219  Epsilon =  0.3511830510520948\n",
      "Episode =  220  Epsilon =  0.3470636492830648\n",
      "Episode =  221  Epsilon =  0.3436104800668229\n",
      "Episode =  222  Epsilon =  0.3425812447124519\n",
      "Episode =  223  Epsilon =  0.33822436612616325\n",
      "Episode =  224  Epsilon =  0.33439069348333605\n",
      "Episode =  225  Epsilon =  0.3309974132185606\n",
      "Episode =  226  Epsilon =  0.3271147903941614\n",
      "Episode =  227  Epsilon =  0.3232777110072471\n",
      "Episode =  228  Epsilon =  0.31919824746920045\n",
      "Episode =  229  Epsilon =  0.31561178824829506\n",
      "Episode =  230  Epsilon =  0.31187845203647235\n",
      "Episode =  231  Epsilon =  0.30905337297865115\n",
      "Episode =  232  Epsilon =  0.3052449625118284\n",
      "Episode =  233  Epsilon =  0.301996414561423\n",
      "Episode =  234  Epsilon =  0.2987226915275634\n",
      "Episode =  235  Epsilon =  0.29539582892836075\n",
      "Episode =  236  Epsilon =  0.29178489355134746\n",
      "Episode =  237  Epsilon =  0.28850644628449923\n",
      "Episode =  238  Epsilon =  0.2850937361026767\n",
      "Episode =  239  Epsilon =  0.2816087340930258\n",
      "Episode =  240  Epsilon =  0.27827761616999075\n",
      "Episode =  241  Epsilon =  0.27490342239070203\n",
      "Episode =  242  Epsilon =  0.27157014165291804\n",
      "Episode =  243  Epsilon =  0.26795555429980045\n",
      "Episode =  244  Epsilon =  0.26470651840216697\n",
      "Episode =  245  Epsilon =  0.26317574067950406\n",
      "Episode =  246  Epsilon =  0.2597507921113287\n",
      "Episode =  247  Epsilon =  0.256421692203576\n",
      "Episode =  248  Epsilon =  0.2530846402956218\n",
      "Episode =  249  Epsilon =  0.2520491692092778\n",
      "Episode =  250  Epsilon =  0.2486695382381634\n",
      "Episode =  251  Epsilon =  0.24486955233322924\n",
      "Episode =  252  Epsilon =  0.24098300907431736\n",
      "Episode =  253  Epsilon =  0.23621146197971377\n",
      "Episode =  254  Epsilon =  0.23197470466649225\n",
      "Episode =  255  Epsilon =  0.22838415787536667\n",
      "Episode =  256  Epsilon =  0.2249391393887312\n",
      "Episode =  257  Epsilon =  0.22141320550503218\n",
      "Episode =  258  Epsilon =  0.21702914730320025\n",
      "Episode =  259  Epsilon =  0.21418332857010391\n",
      "Episode =  260  Epsilon =  0.2099424244371654\n",
      "Episode =  261  Epsilon =  0.20578549167578447\n",
      "Episode =  262  Epsilon =  0.20243829896701906\n",
      "Episode =  263  Epsilon =  0.19842995049057244\n",
      "Episode =  264  Epsilon =  0.1945009687030905\n",
      "Episode =  265  Epsilon =  0.19145215488693512\n",
      "Episode =  266  Epsilon =  0.18873400605840024\n",
      "Episode =  267  Epsilon =  0.1861288811885559\n",
      "Episode =  268  Epsilon =  0.1856270360969126\n",
      "Episode =  269  Epsilon =  0.18332125458889806\n",
      "Episode =  270  Epsilon =  0.18075469002475322\n",
      "Episode =  271  Epsilon =  0.1786165632029653\n",
      "Episode =  272  Epsilon =  0.17630969038311772\n",
      "Episode =  273  Epsilon =  0.17415447072201337\n",
      "Episode =  274  Epsilon =  0.1739108372277118\n",
      "Episode =  275  Epsilon =  0.17176776450726722\n",
      "Episode =  276  Epsilon =  0.16943070837047797\n",
      "Episode =  277  Epsilon =  0.16734284349154047\n",
      "Episode =  278  Epsilon =  0.1654460621081816\n",
      "Episode =  279  Epsilon =  0.1635544247354552\n",
      "Episode =  280  Epsilon =  0.16173292568627343\n",
      "Episode =  281  Epsilon =  0.15970797603986736\n",
      "Episode =  282  Epsilon =  0.15778724939396155\n",
      "Episode =  283  Epsilon =  0.15598317951106433\n",
      "Episode =  284  Epsilon =  0.1540764322724223\n",
      "Episode =  285  Epsilon =  0.1519648862585535\n",
      "Episode =  286  Epsilon =  0.1500322277460788\n",
      "Episode =  287  Epsilon =  0.1479465148963155\n",
      "Episode =  288  Epsilon =  0.14605035617783674\n",
      "Episode =  289  Epsilon =  0.1440920227755511\n",
      "Episode =  290  Epsilon =  0.14164911769199584\n",
      "Episode =  291  Epsilon =  0.13952638908195705\n",
      "Episode =  292  Epsilon =  0.13747670607180187\n",
      "Episode =  293  Epsilon =  0.13528116242872196\n",
      "Episode =  294  Epsilon =  0.13342719686718263\n",
      "Episode =  295  Epsilon =  0.1315591673463262\n",
      "Episode =  296  Epsilon =  0.12917364793070776\n",
      "Episode =  297  Epsilon =  0.1266159649353353\n",
      "Episode =  298  Epsilon =  0.12410892494966032\n",
      "Episode =  299  Epsilon =  0.12237135464127341\n",
      "Episode =  300  Epsilon =  0.1206822437810194\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode =  301  Epsilon =  0.11905215648595577\n",
      "Episode =  302  Epsilon =  0.1171860071194116\n",
      "Episode =  303  Epsilon =  0.11531451193786006\n",
      "Episode =  304  Epsilon =  0.11337083051720523\n",
      "Episode =  305  Epsilon =  0.11112604878323525\n",
      "Episode =  306  Epsilon =  0.10892571450554812\n",
      "Episode =  307  Epsilon =  0.10676894760910573\n",
      "Episode =  308  Epsilon =  0.10465488544466081\n",
      "Episode =  309  Epsilon =  0.10258268244372001\n",
      "Episode =  310  Epsilon =  0.10055150978033942\n",
      "Episode =  311  Epsilon =  0.0985605550396157\n",
      "Episode =  312  Epsilon =  0.09660902189274238\n",
      "Episode =  313  Epsilon =  0.09492366204305164\n",
      "Episode =  314  Epsilon =  0.09304414063790985\n",
      "Episode =  315  Epsilon =  0.0912018344079557\n",
      "Episode =  316  Epsilon =  0.08939600647982325\n",
      "Episode =  317  Epsilon =  0.08762593457049453\n",
      "Episode =  318  Epsilon =  0.08589091069840553\n",
      "Episode =  319  Epsilon =  0.0841902409002728\n",
      "Episode =  320  Epsilon =  0.08252324495352638\n",
      "Episode =  321  Epsilon =  0.08088925610423851\n",
      "Episode =  322  Epsilon =  0.07928762080044091\n",
      "Episode =  323  Epsilon =  0.07771769843072028\n",
      "Episode =  324  Epsilon =  0.0761788610679915\n",
      "Episode =  325  Epsilon =  0.07467049321834339\n",
      "Episode =  326  Epsilon =  0.07319199157485733\n",
      "Episode =  327  Epsilon =  0.07174276477630102\n",
      "Episode =  328  Epsilon =  0.07032223317059927\n",
      "Episode =  329  Epsilon =  0.06892982858298899\n",
      "Episode =  330  Epsilon =  0.06756499408876422\n",
      "Episode =  331  Epsilon =  0.06622718379052142\n",
      "Episode =  332  Epsilon =  0.06491586259981467\n",
      "Episode =  333  Epsilon =  0.06363050602313453\n",
      "Episode =  334  Epsilon =  0.06237059995212512\n",
      "Episode =  335  Epsilon =  0.06113564045795407\n",
      "Episode =  336  Epsilon =  0.05992513358975449\n",
      "Episode =  337  Epsilon =  0.058738595177057776\n",
      "Episode =  338  Epsilon =  0.0577138909827791\n",
      "Episode =  339  Epsilon =  0.056746769771689014\n",
      "Episode =  340  Epsilon =  0.05599707176036509\n",
      "Episode =  341  Epsilon =  0.055351290787079566\n",
      "Episode =  342  Epsilon =  0.054740319172699024\n",
      "Episode =  343  Epsilon =  0.05400632715588297\n",
      "Episode =  344  Epsilon =  0.05297934757364454\n",
      "Episode =  345  Epsilon =  0.05193033813119438\n",
      "Episode =  346  Epsilon =  0.05090209943924885\n",
      "Episode =  347  Epsilon =  0.049939142994067445\n",
      "Episode =  348  Epsilon =  0.04895033065590443\n",
      "Episode =  349  Epsilon =  0.047981097144719265\n",
      "Episode =  350  Epsilon =  0.04703105479295262\n",
      "Episode =  351  Epsilon =  0.04609982360899704\n",
      "Episode =  352  Epsilon =  0.04518703112521071\n",
      "Episode =  353  Epsilon =  0.044292312248940216\n",
      "Episode =  354  Epsilon =  0.0434153091164931\n",
      "Episode =  355  Epsilon =  0.04255567095000215\n",
      "Episode =  356  Epsilon =  0.04171305391712349\n",
      "Episode =  357  Epsilon =  0.04088712099351265\n",
      "Episode =  358  Epsilon =  0.04007754182802415\n",
      "Episode =  359  Epsilon =  0.0392839926105796\n",
      "Episode =  360  Epsilon =  0.0385061559426524\n",
      "Episode =  361  Epsilon =  0.037743720710316814\n",
      "Episode =  362  Epsilon =  0.036996381959810566\n",
      "Episode =  363  Epsilon =  0.03626384077556161\n",
      "Episode =  364  Epsilon =  0.035545804160629815\n",
      "Episode =  365  Epsilon =  0.03484198491951601\n",
      "Episode =  366  Epsilon =  0.03415210154329148\n",
      "Episode =  367  Epsilon =  0.03347587809700178\n",
      "Episode =  368  Epsilon =  0.03281304410930027\n",
      "Episode =  369  Epsilon =  0.03216333446426658\n",
      "Episode =  370  Epsilon =  0.03152648929536751\n",
      "Episode =  371  Epsilon =  0.030902253881517357\n",
      "Episode =  372  Epsilon =  0.03029037854519609\n",
      "Episode =  373  Epsilon =  0.029690618552585164\n",
      "Episode =  374  Epsilon =  0.029128936955723744\n",
      "Episode =  375  Epsilon =  0.028552173909092912\n",
      "Episode =  376  Epsilon =  0.027986830970668005\n",
      "Episode =  377  Epsilon =  0.02743268201834886\n",
      "Episode =  378  Epsilon =  0.026889505407331163\n",
      "Episode =  379  Epsilon =  0.026357083881454597\n",
      "Episode =  380  Epsilon =  0.02583520448630608\n",
      "Episode =  381  Epsilon =  0.025323658484043682\n",
      "Episode =  382  Epsilon =  0.024822241269907185\n",
      "Episode =  383  Epsilon =  0.024330752290381473\n",
      "Episode =  384  Epsilon =  0.023848994962980492\n",
      "Episode =  385  Epsilon =  0.02347516060200335\n",
      "Episode =  386  Epsilon =  0.023125678336229007\n",
      "Episode =  387  Epsilon =  0.022667781890749986\n",
      "Episode =  388  Epsilon =  0.022218951953579744\n",
      "Episode =  389  Epsilon =  0.02177900900471174\n",
      "Episode =  390  Epsilon =  0.021347777078697693\n",
      "Episode =  391  Epsilon =  0.02092508369426622\n",
      "Episode =  392  Epsilon =  0.02051075978533488\n",
      "Episode =  393  Epsilon =  0.020104639633388235\n",
      "Episode =  394  Epsilon =  0.019706560801194865\n",
      "Episode =  395  Epsilon =  0.01931636406783689\n",
      "Episode =  396  Epsilon =  0.018933893365025774\n",
      "Episode =  397  Epsilon =  0.018558995714679137\n",
      "Episode =  398  Epsilon =  0.018226116183088243\n",
      "Episode =  399  Epsilon =  0.017865232766230848\n",
      "Episode =  400  Epsilon =  0.01754128833508224\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Episode =  401  Epsilon =  0.017287058629241323\n",
      "Episode =  402  Epsilon =  0.0170501475118061\n",
      "Episode =  403  Epsilon =  0.016899081990776943\n",
      "Episode =  404  Epsilon =  0.01673261477456141\n",
      "Episode =  405  Epsilon =  0.016557850177215356\n",
      "Episode =  406  Epsilon =  0.01622999900542755\n",
      "Episode =  407  Epsilon =  0.015932519070873702\n",
      "Episode =  408  Epsilon =  0.015665535215216866\n",
      "Episode =  409  Epsilon =  0.015470944480096052\n",
      "Episode =  410  Epsilon =  0.015252820321051507\n",
      "Episode =  411  Epsilon =  0.014998726012326108\n",
      "Episode =  412  Epsilon =  0.0148227899064028\n",
      "Episode =  413  Epsilon =  0.014615265430308355\n",
      "Episode =  414  Epsilon =  0.014439495062133559\n",
      "Episode =  415  Epsilon =  0.014153588055760273\n",
      "Episode =  416  Epsilon =  0.013873342107196932\n",
      "Episode =  417  Epsilon =  0.013669538699122748\n",
      "Episode =  418  Epsilon =  0.013474117521924692\n",
      "Episode =  419  Epsilon =  0.013256280492607093\n",
      "Episode =  420  Epsilon =  0.01299380154476458\n",
      "Episode =  421  Epsilon =  0.01273651977105389\n",
      "Episode =  422  Epsilon =  0.012484332265626114\n",
      "Episode =  423  Epsilon =  0.01223713816020375\n",
      "Episode =  424  Epsilon =  0.011994838583736208\n",
      "Episode =  425  Epsilon =  0.01175733662285392\n",
      "Episode =  426  Epsilon =  0.01152453728310565\n",
      "Episode =  427  Epsilon =  0.011296347450963177\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "Input \u001b[1;32mIn [146]\u001b[0m, in \u001b[0;36m<cell line: 1>\u001b[1;34m()\u001b[0m\n\u001b[0;32m     18\u001b[0m replay_buffer\u001b[38;5;241m.\u001b[39mappend((state, next_state, reward, done, action))\n\u001b[0;32m     19\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(replay_buffer)\u001b[38;5;241m>\u001b[39mbatch_size \u001b[38;5;129;01mand\u001b[39;00m steps\u001b[38;5;241m%\u001b[39m\u001b[38;5;241m4\u001b[39m\u001b[38;5;241m==\u001b[39m\u001b[38;5;241m0\u001b[39m:\n\u001b[1;32m---> 20\u001b[0m     loss \u001b[38;5;241m=\u001b[39m \u001b[43mcompute_td_loss\u001b[49m\u001b[43m(\u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     21\u001b[0m     eps_loss \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m loss\u001b[38;5;241m.\u001b[39mdetach()\u001b[38;5;241m.\u001b[39mnumpy()\n\u001b[0;32m     22\u001b[0m eps \u001b[38;5;241m=\u001b[39m eps\u001b[38;5;241m/\u001b[39m(\u001b[38;5;241m1\u001b[39m \u001b[38;5;241m+\u001b[39m decay_val)\n",
      "Input \u001b[1;32mIn [145]\u001b[0m, in \u001b[0;36mcompute_td_loss\u001b[1;34m(batch_size)\u001b[0m\n\u001b[0;32m     22\u001b[0m loss \u001b[38;5;241m=\u001b[39m ((reward \u001b[38;5;241m+\u001b[39m gamma\u001b[38;5;241m*\u001b[39mmax_next_q_values\u001b[38;5;241m*\u001b[39m(\u001b[38;5;241m1\u001b[39m\u001b[38;5;241m-\u001b[39mdone)\u001b[38;5;241m.\u001b[39mdetach() \u001b[38;5;241m-\u001b[39m q_vals\u001b[38;5;241m.\u001b[39msqueeze())\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m2\u001b[39m)\u001b[38;5;241m.\u001b[39mmean()\n\u001b[0;32m     24\u001b[0m opt\u001b[38;5;241m.\u001b[39mzero_grad()\n\u001b[1;32m---> 25\u001b[0m \u001b[43mloss\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m     26\u001b[0m opt\u001b[38;5;241m.\u001b[39mstep()\n\u001b[0;32m     27\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m loss\n",
      "File \u001b[1;32m~\\anaconda3\\envs\\torch_1\\lib\\site-packages\\torch\\_tensor.py:396\u001b[0m, in \u001b[0;36mTensor.backward\u001b[1;34m(self, gradient, retain_graph, create_graph, inputs)\u001b[0m\n\u001b[0;32m    387\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m has_torch_function_unary(\u001b[38;5;28mself\u001b[39m):\n\u001b[0;32m    388\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m handle_torch_function(\n\u001b[0;32m    389\u001b[0m         Tensor\u001b[38;5;241m.\u001b[39mbackward,\n\u001b[0;32m    390\u001b[0m         (\u001b[38;5;28mself\u001b[39m,),\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    394\u001b[0m         create_graph\u001b[38;5;241m=\u001b[39mcreate_graph,\n\u001b[0;32m    395\u001b[0m         inputs\u001b[38;5;241m=\u001b[39minputs)\n\u001b[1;32m--> 396\u001b[0m \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mautograd\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mbackward\u001b[49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgradient\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43minputs\u001b[49m\u001b[43m)\u001b[49m\n",
      "File \u001b[1;32m~\\anaconda3\\envs\\torch_1\\lib\\site-packages\\torch\\autograd\\__init__.py:173\u001b[0m, in \u001b[0;36mbackward\u001b[1;34m(tensors, grad_tensors, retain_graph, create_graph, grad_variables, inputs)\u001b[0m\n\u001b[0;32m    168\u001b[0m     retain_graph \u001b[38;5;241m=\u001b[39m create_graph\n\u001b[0;32m    170\u001b[0m \u001b[38;5;66;03m# The reason we repeat same the comment below is that\u001b[39;00m\n\u001b[0;32m    171\u001b[0m \u001b[38;5;66;03m# some Python versions print out the first line of a multi-line function\u001b[39;00m\n\u001b[0;32m    172\u001b[0m \u001b[38;5;66;03m# calls in the traceback and some print out the last line\u001b[39;00m\n\u001b[1;32m--> 173\u001b[0m \u001b[43mVariable\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_execution_engine\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mrun_backward\u001b[49m\u001b[43m(\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# Calls into the C++ engine to run the backward pass\u001b[39;49;00m\n\u001b[0;32m    174\u001b[0m \u001b[43m    \u001b[49m\u001b[43mtensors\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mgrad_tensors_\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mretain_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcreate_graph\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[0;32m    175\u001b[0m \u001b[43m    \u001b[49m\u001b[43mallow_unreachable\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43maccumulate_grad\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mTrue\u001b[39;49;00m\u001b[43m)\u001b[49m\n",
      "\u001b[1;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "for i in range(episodes):\n",
    "    print(\"Episode = \", i, \" Epsilon = \", eps)\n",
    "    state = torch.tensor(env.reset(), dtype=torch.float32).unsqueeze(0)\n",
    "\n",
    "    done = False\n",
    "    steps = 0\n",
    "    eps_rew = 0 \n",
    "    eps_loss = 0\n",
    "    while not done and steps<max_steps:\n",
    "        if np.random.uniform(0,1)<eps:\n",
    "            action = env.action_space.sample()\n",
    "        else:\n",
    "#             action = env.action_space.sample()\n",
    "            action = np.argmax(model(state).detach().numpy())\n",
    "                \n",
    "\n",
    "        next_state, reward, done, info = env.step(action)\n",
    "        replay_buffer.append((state, next_state, reward, done, action))\n",
    "        if len(replay_buffer)>batch_size and steps%4==0:\n",
    "            loss = compute_td_loss(batch_size)\n",
    "            eps_loss += loss.detach().numpy()\n",
    "        eps = eps/(1 + decay_val)\n",
    "        eps_rew += reward \n",
    "        \n",
    "        if steps%50==0:\n",
    "            target.load_state_dict(model.state_dict())\n",
    "        \n",
    "        if done:\n",
    "            tot_rewards.append(eps_rew)\n",
    "            break\n",
    "        \n",
    "        state = torch.tensor(next_state, dtype=torch.float32).unsqueeze(0)\n",
    "        steps += 1\n",
    "    tot_rewards.append(eps_rew)\n",
    "    tot_loss.append(eps_loss)\n",
    "    \n",
    "    if(i%100)==0:\n",
    "        plt.scatter(np.arange(len(tot_rewards)), tot_rewards)\n",
    "        plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5e4b9d1e",
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.plot(tot_loss)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "3190cdaf",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x219c28c4730>]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Plot\n",
    "plt.plot(tot_rewards)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "id": "2878f86c",
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "#Render the final environment\n",
    "state, info = env.reset(seed=42, return_info=True)\n",
    "import time\n",
    "for i in range(1000):\n",
    "    \n",
    "    env.render()\n",
    "    \n",
    "    state = torch.tensor(state, dtype=torch.float32)\n",
    "    action = np.argmax(model(state).detach().numpy())\n",
    "    state, reward, done, info = env.step(action)\n",
    "\n",
    "    if done:\n",
    "        state, info = env.reset(return_info=True)\n",
    "env.close()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}