add Live Training Cartpole DQN NB

notebooks/live_training/cartpole_dqn_LT.ipynb

@@ -0,0 +1,273 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Cartpole DQN"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Deep Q-Learning Network with Keras and OpenAI Gym, based on [Keon Kim's code](https://github.com/keon/deep-q-learning/blob/master/dqn.py)."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Import dependencies"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import random\n",
+    "import gym\n",
+    "import numpy as np\n",
+    "from collections import deque\n",
+    "from keras.models import Sequential\n",
+    "from keras.layers import Dense\n",
+    "from keras.optimizers import Adam\n",
+    "import os "
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Set parameters"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "env = # FILL IN"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "state_size = env.observation_space.shape[0]\n",
+    "state_size"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "action_size = env.action_space.n\n",
+    "action_size"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "batch_size = # FILL IN"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "n_episodes = # FILL IN"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "output_dir = 'model_output/cartpole/'"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "if not os.path.exists(output_dir):\n",
+    "    os.makedirs(output_dir)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Define agent"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "class DQNAgent:\n",
+    "    \n",
+    "    def __init__(self, state_size, action_size):\n",
+    "        self.state_size = state_size\n",
+    "        self.action_size = action_size\n",
+    "        \n",
+    "        self.memory = deque(maxlen=2000) \n",
+    "        \n",
+    "        self.gamma = # FILL IN\n",
+    "        \n",
+    "        self.epsilon = # FILL IN\n",
+    "        self.epsilon_decay = # FILL IN\n",
+    "        self.epsilon_min = # FILL IN\n",
+    "        \n",
+    "        self.learning_rate = # FILL IN\n",
+    "        \n",
+    "        self.model = self._build_model() \n",
+    "    \n",
+    "    def _build_model(self):\n",
+    "\n",
+    "        model = Sequential()\n",
+    "\n",
+    "        # FILL IN NEURAL NETWORK ARCHITECTURE\n",
+    "\n",
+    "        # COMPILE MODEL\n",
+    "        \n",
+    "        return model\n",
+    "    \n",
+    "    def remember(self, state, action, reward, next_state, done):\n",
+    "        self.memory.append((state, action, reward, next_state, done)) \n",
+    "\n",
+    "    def act(self, state):\n",
+    "        if np.random.rand() <= self.epsilon: \n",
+    "            return random.randrange(self.action_size)\n",
+    "        act_values = self.model.predict(state)\n",
+    "        return np.argmax(act_values[0]) \n",
+    "\n",
+    "    def replay(self, batch_size): \n",
+    "        \n",
+    "        minibatch = random.sample(self.memory, batch_size) \n",
+    "        \n",
+    "        for state, action, reward, next_state, done in minibatch: \n",
+    "            target = reward # N.B.: if done\n",
+    "            if not done: \n",
+    "                target = (reward + self.gamma * np.amax(self.model.predict(next_state)[0])) # (maximum target Q based on future action a')\n",
+    "            target_f = self.model.predict(state) \n",
+    "            target_f[0][action] = target\n",
+    "            \n",
+    "            self.model.fit(state, target_f, epochs=1, verbose=0)\n",
+    "\n",
+    "        if self.epsilon > self.epsilon_min:\n",
+    "            self.epsilon *= self.epsilon_decay\n",
+    "\n",
+    "    def load(self, name):\n",
+    "        self.model.load_weights(name)\n",
+    "\n",
+    "    def save(self, name):\n",
+    "        self.model.save_weights(name)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Interact with environment"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "agent = DQNAgent(state_size, action_size) "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "done = False\n",
+    "\n",
+    "for e in range(n_episodes): \n",
+    "\n",
+    "    state = env.reset() \n",
+    "    state = np.reshape(state, [1, state_size])\n",
+    "    \n",
+    "    for time in range(5000):  \n",
+    "        \n",
+    "        #  env.render()\n",
+    "\n",
+    "        action = agent.act(state) \n",
+    "    \n",
+    "        next_state, reward, done, _ = env.step(action) \n",
+    "        \n",
+    "        reward = reward if not done else -10 \n",
+    "        \n",
+    "        next_state = np.reshape(next_state, [1, state_size])\n",
+    "        \n",
+    "        agent.remember(state, action, reward, next_state, done) \n",
+    "\n",
+    "        state = next_state \n",
+    "        \n",
+    "        if done: \n",
+    "            print(\"episode: {}/{}, score: {}, e: {:.2}\".format(e, n_episodes, time, agent.epsilon))\n",
+    "            break \n",
+    "\n",
+    "    if len(agent.memory) > batch_size:\n",
+    "        agent.replay(batch_size)\n",
+    "\n",
+    "    if e % 50 == 0:\n",
+    "        agent.save(output_dir + \"weights_\" + '{:04d}'.format(e) + \".hdf5\")         "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# saved agents can be loaded with agent.load(\"./path/filename.hdf5\") "
+   ]
+  }
+ ],
+ "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": "3.6.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}