evanthebouncy’s gists

evanthebouncy / random_agent.py

Created April 24, 2026 15:09

the random agent that plays the game

	import ale_py
	import gymnasium as gym
	import numpy as np
	from pong_features import PongFeaturesWrapper

	gym.register_envs(ale_py)


	def make_env(render: bool):
	mode = "human" if render else None

evanthebouncy / pong_features.py

Created April 24, 2026 15:08

state extraction for pong

	import numpy as np
	import gymnasium as gym


	# ALE Pong pixel colors
	_PLAYER_COLOR = np.array([92, 186, 92], dtype=np.uint8) # green - right paddle
	_OPPONENT_COLOR = np.array([213, 130, 74], dtype=np.uint8) # orange - left paddle
	_BALL_COLOR = np.array([236, 236, 236], dtype=np.uint8) # white - ball

	# Frame regions

evanthebouncy / qlearning.html

Last active April 22, 2026 07:57

q-learning html

	<!DOCTYPE html>
	<html lang="en">
	<head>
	<meta charset="UTF-8">
	<title>Q-Learning GridWorld</title>
	<style>
	* { box-sizing: border-box; margin: 0; padding: 0; }
	body {
	font-family: 'Segoe UI', Arial, sans-serif;
	background: #12192b;

evanthebouncy / random_policy.py

Last active April 18, 2026 01:56

random policy

	"""
	random_policy.py – Random policy and rollout utility for the Crawler
	=======================================================================

	Classes
	-------
	RandomPolicy – uniform random policy; .act() and .learn() interface

	Functions
	---------

evanthebouncy / crawler_env.py

Created April 17, 2026 15:19

crawler environment

	#!/usr/bin/env python3
	"""
	crawler_env.py – 2D Crawler RL Environment (Gymnasium-compatible)
	=====================================================================

	Robot description
	-----------------
	Body : rectangle, slides left/right on the floor (no tipping).
	Arm : L-shaped, two rigid segments attached at the front-top of the body.
	segment-1 longer, ~horizontal angle θ1 from horizontal [−30°, +30°]

evanthebouncy / speedup.py

Created March 23, 2026 15:03

speedup

	from mlagents_envs.environment import UnityEnvironment
	from mlagents_envs.envs.unity_gym_env import UnityToGymWrapper
	from mlagents_envs.side_channel.engine_configuration_channel import EngineConfigurationChannel

	engine_channel = EngineConfigurationChannel()

	unity_env = UnityEnvironment(
	file_name=r"Simple/Assignment2", # use the path to your built game here
	side_channels=[engine_channel],
	no_graphics=False, # set to True to disable rendering, which can speed up training

evanthebouncy / reinforce.py

Last active March 15, 2026 05:16

reinforce_game+AI

	import torch
	import torch.nn as nn
	import torch.optim as optim
	from torch.distributions import Categorical
	import numpy as np
	from mlagents_envs.environment import UnityEnvironment
	from mlagents_envs.envs.unity_gym_env import UnityToGymWrapper

	# ── Policy network ────────────────────────────────────────────────────────────
	class Policy(nn.Module):

evanthebouncy / tutorial_game_ai.py

Created March 8, 2026 00:54

	import random
	from time import time
	from sklearn.neighbors import KNeighborsClassifier
	import numpy as np

	from mlagents_envs.environment import UnityEnvironment
	from mlagents_envs.envs.unity_gym_env import UnityToGymWrapper

	unity_env = UnityEnvironment("gg_det")
	env = UnityToGymWrapper(unity_env)

evanthebouncy / snake html

Created March 4, 2026 03:41


	<!DOCTYPE html>
	<html>
	<body>

	<div>

	<span style="float: left">
	<canvas id="myCanvas" width="800" height="800"
	style="border:1px solid #d3d3d3;">

evanthebouncy / grid_manual.py

Created January 23, 2026 15:09

manually play the grid game

	# play_gridworld_terminal.py
	# Terminal-playable GridWorld + trajectory tracking (only dependency: numpy)

	from __future__ import annotations
	from dataclasses import dataclass
	from typing import Dict, List, Optional, Tuple
	import numpy as np


	@dataclass(frozen=True)