Tennis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# RL Tennis playing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\ProgramData\\Anaconda3\\lib\\site-packages\\h5py\\__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.\n",
      "  from ._conv import register_converters as _register_converters\n"
     ]
    }
   ],
   "source": [
    "from unityagents import UnityEnvironment\n",
    "import numpy as np\n",
    "from brain.agent import Agent, Config\n",
    "from brain.teacher import Teacher"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Next, we will start the environment!  **_Before running the code cell below_**, change the `file_name` parameter to match the location of the Unity environment that you downloaded.\n",
    "\n",
    "- **Mac**: `\"path/to/Tennis.app\"`\n",
    "- **Windows** (x86): `\"path/to/Tennis_Windows_x86/Tennis.exe\"`\n",
    "- **Windows** (x86_64): `\"path/to/Tennis_Windows_x86_64/Tennis.exe\"`\n",
    "- **Linux** (x86): `\"path/to/Tennis_Linux/Tennis.x86\"`\n",
    "- **Linux** (x86_64): `\"path/to/Tennis_Linux/Tennis.x86_64\"`\n",
    "- **Linux** (x86, headless): `\"path/to/Tennis_Linux_NoVis/Tennis.x86\"`\n",
    "- **Linux** (x86_64, headless): `\"path/to/Tennis_Linux_NoVis/Tennis.x86_64\"`\n",
    "\n",
    "For instance, if you are using a Mac, then you downloaded `Tennis.app`.  If this file is in the same folder as the notebook, then the line below should appear as follows:\n",
    "```\n",
    "env = UnityEnvironment(file_name=\"Tennis.app\")\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "INFO:unityagents:\n",
      "'Academy' started successfully!\n",
      "Unity Academy name: Academy\n",
      "        Number of Brains: 1\n",
      "        Number of External Brains : 1\n",
      "        Lesson number : 0\n",
      "        Reset Parameters :\n",
      "\t\t\n",
      "Unity brain name: TennisBrain\n",
      "        Number of Visual Observations (per agent): 0\n",
      "        Vector Observation space type: continuous\n",
      "        Vector Observation space size (per agent): 8\n",
      "        Number of stacked Vector Observation: 3\n",
      "        Vector Action space type: continuous\n",
      "        Vector Action space size (per agent): 2\n",
      "        Vector Action descriptions: , \n"
     ]
    }
   ],
   "source": [
    "env = UnityEnvironment(file_name=\"Tennis_Windows_x86_64/Tennis.exe\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Environments contain **_brains_** which are responsible for deciding the actions of their associated agents. Here we check for the first brain available, and set it as the default brain we will be controlling from Python."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# get the default brain\n",
    "brain_name = env.brain_names[0]\n",
    "brain = env.brains[brain_name]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of agents: 2\n",
      "Size of each action: 2\n",
      "There are 2 agents. Each observes a state with length: 24\n",
      "The state for the first agent looks like: [ 0.          0.          0.          0.          0.          0.\n",
      "  0.          0.          0.          0.          0.          0.\n",
      "  0.          0.          0.          0.         -6.65278625 -1.5\n",
      " -0.          0.          6.83172083  6.         -0.          0.        ]\n"
     ]
    }
   ],
   "source": [
    "# reset the environment\n",
    "env_info = env.reset(train_mode=True)[brain_name]\n",
    "\n",
    "# number of agents\n",
    "num_agents = len(env_info.agents)\n",
    "print('Number of agents:', num_agents)\n",
    "\n",
    "# size of each action\n",
    "action_size = brain.vector_action_space_size\n",
    "print('Size of each action:', action_size)\n",
    "\n",
    "# examine the state space \n",
    "states = env_info.vector_observations\n",
    "state_size = states.shape[1]\n",
    "print('There are {} agents. Each observes a state with length: {}'.format(states.shape[0], state_size))\n",
    "print('The state for the first agent looks like:', states[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train agent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = Config()\n",
    "config.BUFFER_SIZE=int(3e4)\n",
    "\n",
    "agent = Agent(state_size, action_size, config)\n",
    "teacher = Teacher(agent, env, brain_name, num_agents)\n",
    "# teacher.set_epsilon(1.0, 0.01, 80)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  2%|▊                                   | 24/1000 [00:02<01:48,  8.99it/s, actor_loss=0, critic_loss=0, scores=0.0321]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Learning started\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 76%|██████████▌   | 756/1000 [02:56<00:57,  4.28it/s, actor_loss=-0.063044585, critic_loss=0.044002566, scores=0.0484]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Memory filled\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 89%|██████████████▎ | 891/1000 [06:55<00:50,  2.15it/s, actor_loss=-0.07792923, critic_loss=0.052385934, scores=0.511]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Environment solved in 791 episodes!\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "res = teacher.train(1000, 100, 0.5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x396 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "Teacher.visualise_scores(res, 100, 0.5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When finished, you can close the environment."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.600000038743019"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "teacher.display()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "torch.save(agent.actor_local.state_dict(), 'pytorch_models/checkpoint_actor.pth')\n",
    "torch.save(agent.critic_local.state_dict(), 'pytorch_models/checkpoint_critic.pth')"
   ]
  }
 ],
 "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": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}