Merge pull request #31 from BradenEverson/rl/research

New Reinforcement Learning Branch

earthmover-research/.gitignore

@@ -0,0 +1 @@
+.ipynb_checkpoints/

earthmover-research/lunar-lander/Lunar Landing Reinforcement Learning.ipynb

@@ -0,0 +1,120 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "01c32df2",
+   "metadata": {},
+   "source": [
+    "# Reinforcement Learning with Lunar Landing\n",
+    "This is a small sample code for the `LunarLander-v3` environment in OpenAI's Gym which is designed to simulate a reinforcement learning scenario where a agent must control a lunar lander to safely land on a designated landing pad."
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "86ca8de7",
+   "metadata": {},
+   "source": [
+    "### Library Imports\n",
+    "Here, we make the following library imports, where we manage to import the `gymansium` library, which gives us access to the environment. We import the `imageio` library which allows us to produce a `.gif` file."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "05c326ff",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import gymnasium as gym\n",
+    "import imageio"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "d5772728",
+   "metadata": {},
+   "source": [
+    "### Creating a Environment\n",
+    "Here, we create a Gym environment using `gym.make()`. Although the `.make()` can accept multiple parameters, we can insert the lunar lander environment, as well as set the render mode, which allows us to create image arrays."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "72e3d14d",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "env = gym.make(\"LunarLander-v3\", render_mode=\"rgb_array\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "2fac2cd5",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def run_multiple_episodes(num_episodes=10, max_steps=200):\n",
+    "    frames = [] \n",
+    "\n",
+    "    for _ in range(num_episodes):\n",
+    "        observation, info = env.reset()\n",
+    "        episode_over = False\n",
+    "        steps = 0\n",
+    "\n",
+    "        while not episode_over and steps < max_steps:\n",
+    "            frame = env.render()  \n",
+    "            frames.append(frame) \n",
+    "            action = env.action_space.sample() \n",
+    "            observation, reward, terminated, truncated, info = env.step(action)\n",
+    "            episode_over = terminated or truncated\n",
+    "            steps += 1  \n",
+    "\n",
+    "    env.close()\n",
+    "    return frames"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "id": "ca785b50",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "frames = run_multiple_episodes(num_episodes=10)  "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "581bd473",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "imageio.mimsave('lunar_lander_multiple_episodes.gif', frames, fps=50)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "RL-lunar-lander",
+   "language": "python",
+   "name": "rl-lunar-lander"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.13.0"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}