rl_cart_pole_actor_critic.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Reinforcement Learning: CartPole Actor Critic Solution (Temporal Difference)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"./screenshots/actor_critic.png\" alt=\"drawing\" width=\"1500\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gym\n",
    "import numpy as np\n",
    "import random\n",
    "import math\n",
    "\n",
    "from IPython.display import clear_output\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "import torch.nn.functional as F\n",
    "import torch.optim as optim"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Memory Buffer to store Experience Replay"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "class RingBuffer():\n",
    "    def __init__(self, size):\n",
    "        self.size = size\n",
    "        self.buffer_ar = []\n",
    "        for i in range(0, size):\n",
    "            self.buffer_ar.append(0)\n",
    "        self.counter = 0\n",
    "        self.full = False\n",
    "    \n",
    "    def add(self, value):\n",
    "        self.buffer_ar[self.counter] = value\n",
    "        self.counter += 1\n",
    "        if self.counter > self.size-1:\n",
    "            self.counter = 0\n",
    "            self.full = True\n",
    "    \n",
    "    def get_random(self, last_n, sample_size):\n",
    "        if not self.full:\n",
    "            last_n = self.counter\n",
    "        if self.counter - last_n < 0:\n",
    "            end = np.linspace(self.counter-last_n, -1, -1*(self.counter-last_n))\n",
    "            start = np.linspace(0, self.counter-1, self.counter)\n",
    "            window = np.concatenate((end, start))\n",
    "        else:\n",
    "            window = np.linspace(self.counter-last_n, self.counter-1, last_n)\n",
    "            #print(window)\n",
    "        random_index = np.random.choice(a=window, size=sample_size, replace=False)\n",
    "        random_choices = []\n",
    "        for i in random_index:\n",
    "            random_choices.append(self.buffer_ar[int(i)])\n",
    "        return random_choices\n",
    "    \n",
    "    def __repr__(self):\n",
    "        return str(self.buffer_ar)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Plot rewards and losses iteratively"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot(frame, rewards, v_losses, theta_losses):\n",
    "    clear_output(True)\n",
    "    plt.figure(figsize=(20, 5))\n",
    "    plt.subplot(131)\n",
    "    plt.title(\"frame %s. reward: %s\" % (frame, np.mean(rewards[-10:])))\n",
    "    plt.plot(rewards)\n",
    "    plt.subplot(132)\n",
    "    plt.title('delta_v_loss')\n",
    "    plt.plot(v_losses)\n",
    "    plt.subplot(133)\n",
    "    plt.title('theta_loss')\n",
    "    plt.plot(theta_losses)\n",
    "    plt.ylim(0, 200)\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Build Actor & Critic Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "class DVN(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(DVN, self).__init__()\n",
    "        self.fc1 = nn.Linear(4, 512)\n",
    "        self.fc2 = nn.Linear(512, 512)\n",
    "        self.fc3 = nn.Linear(512, 512)\n",
    "        self.fc4 = nn.Linear(512, 1)\n",
    "        \n",
    "    def forward(self, states):\n",
    "        out = F.relu(self.fc1(states))\n",
    "        out = F.relu(self.fc2(out))\n",
    "        out = F.relu(self.fc3(out))\n",
    "        out = self.fc4(out)\n",
    "        return out\n",
    "\n",
    "class PolicyGradient(nn.Module):\n",
    "    def __init__(self):\n",
    "        super(PolicyGradient, self).__init__()\n",
    "        self.fc1 = nn.Linear(4, 512)\n",
    "        self.fc2 = nn.Linear(512, 512)\n",
    "        self.fc3 = nn.Linear(512, 512)\n",
    "        self.fc4 = nn.Linear(512, 2)\n",
    "        \n",
    "    def forward(self, states):\n",
    "        out = F.relu(self.fc1(states))\n",
    "        out = F.relu(self.fc2(out))\n",
    "        out = F.relu(self.fc3(out))\n",
    "        out = self.fc4(out)\n",
    "        return out"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Initialize Actor & Critic models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "dvn = DVN()\n",
    "policy_gradient = PolicyGradient()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Define Adam optimizer for both networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "dvn_optimizer = optim.Adam(dvn.parameters(), lr=0.01)\n",
    "policy_gradient_optimizer = optim.Adam(policy_gradient.parameters(), lr=0.01)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Training Loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x360 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "memory_buffer = RingBuffer(2000)\n",
    "GAMMA = 1.0\n",
    "frame = 0\n",
    "all_rewards = []\n",
    "all_v_losses = []\n",
    "all_theta_losses = []\n",
    "env = gym.make('CartPole-v0')\n",
    "\n",
    "for episode in range(1000):\n",
    "    observation = env.reset()\n",
    "    env.render()\n",
    "    life_counter = 0\n",
    "    \n",
    "    while True:\n",
    "        frame += 1\n",
    "        life_counter += 1\n",
    "        env.render()\n",
    "        \n",
    "        dvn_optimizer.zero_grad()\n",
    "        policy_gradient_optimizer.zero_grad()\n",
    "        \n",
    "        #sample action from current policy\n",
    "        observation_tensor = torch.tensor([observation], dtype=torch.float)\n",
    "        output = policy_gradient(observation_tensor)\n",
    "        softmax = nn.Softmax()\n",
    "        output = softmax(output)\n",
    "        output = output.view(output.size(1))\n",
    "        output = output.detach().numpy()\n",
    "        action = np.random.choice(a=[0, 1], size=1, p=output).item()\n",
    "        \n",
    "        #take action\n",
    "        next_observation, reward, done, info = env.step(action)\n",
    "            \n",
    "        memory_buffer.add([observation, action, reward, next_observation, done])\n",
    "        \n",
    "        #start learning\n",
    "        if frame >= 512:\n",
    "            samples = memory_buffer.get_random(2000, 512)\n",
    "            current_states = []\n",
    "            actions = []\n",
    "            rewards = []\n",
    "            next_states = []\n",
    "            is_done = []\n",
    "            for sample in samples:\n",
    "                current_states.append(sample[0])\n",
    "                actions.append(sample[1])\n",
    "                rewards.append(sample[2])\n",
    "                next_states.append(sample[3])\n",
    "                is_done.append(sample[4])\n",
    "            \n",
    "            current_states_tensor = torch.tensor(current_states, dtype=torch.float)\n",
    "            v_current = dvn(current_states_tensor)\n",
    "            v_current = v_current.view(v_current.size(0))\n",
    "            \n",
    "            next_states_tensor = torch.tensor(next_states, dtype=torch.float)\n",
    "            v_next = dvn(next_states_tensor)\n",
    "            v_next = v_next.view(v_next.size(0))\n",
    "            \n",
    "            rewards = torch.tensor(rewards, dtype=torch.float)\n",
    "            is_done = torch.tensor(is_done, dtype=torch.float)\n",
    "            \n",
    "            delta_v = rewards + (GAMMA * (v_next * (1-is_done))) - v_current\n",
    "            delta_v_copy = delta_v.clone()\n",
    "            delta_v_copy = delta_v_copy.detach()\n",
    "            \n",
    "            action_probs = policy_gradient(current_states_tensor)\n",
    "            action_probs = action_probs[range(0, action_probs.size(0)), actions]\n",
    "            log_softmax = nn.LogSoftmax()\n",
    "            log_action_probs = log_softmax(action_probs)\n",
    "            \n",
    "            theta = -1 * log_action_probs * delta_v_copy\n",
    "            \n",
    "            #backprop delta_v\n",
    "            delta_v_loss = torch.mean(torch.pow(delta_v, 2))\n",
    "            all_v_losses.append(delta_v_loss)\n",
    "            delta_v_loss.backward()\n",
    "            dvn_optimizer.step()\n",
    "            \n",
    "            #backprop theta\n",
    "            theta_loss = torch.mean(theta)\n",
    "            all_theta_losses.append(theta_loss)\n",
    "            theta_loss.backward()\n",
    "            policy_gradient_optimizer.step()\n",
    "        \n",
    "        #point to next state\n",
    "        observation = next_observation.copy()\n",
    "        \n",
    "        if done:\n",
    "            break\n",
    "            \n",
    "        if frame % 200 == 0:\n",
    "            plot(frame, all_rewards, all_v_losses, all_theta_losses)\n",
    "    \n",
    "    all_rewards.append(life_counter)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "i"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}