alu/cse151b-final-project
1
0
Fork 0
mirror of https://github.com/ltcptgeneral/cse151b-final-project.git synced 2024-09-20 00:47:55 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #1 from ltcptgeneral/gym-2

Browse Source 
Gym 2
This commit is contained in:
Arthur Lu 2024-03-20 21:07:19 -07:00 committed by GitHub
commit 2244de94ac
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
21 changed files with 389502 additions and 16844 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
.gitignore vendored
View File

@ -1,3 +1,6 @@
**/data/* **/data/*
**/*.zip **/*.zip
**/__pycache__ **/__pycache__
/env
**/runs/*
**/wandb/*

545
dqn_letter_gssr.ipynb Normal file
View File

@ -0,0 +1,545 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"def load_valid_words(file_path='wordle_words.txt'):\n",
" \"\"\"\n",
" Load valid five-letter words from a specified text file.\n",
"\n",
" Parameters:\n",
" - file_path (str): The path to the text file containing valid words.\n",
"\n",
" Returns:\n",
" - list[str]: A list of valid words loaded from the file.\n",
" \"\"\"\n",
" with open(file_path, 'r') as file:\n",
" valid_words = [line.strip() for line in file if len(line.strip()) == 5]\n",
" return valid_words"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"from stable_baselines3 import PPO, DQN # Or any other suitable RL algorithm\n",
"from stable_baselines3.common.env_checker import check_env\n",
"from letter_guess import LetterGuessingEnv\n",
"from tqdm import tqdm"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"env = LetterGuessingEnv(valid_words=load_valid_words()) # Make sure to load your valid words\n",
"check_env(env) # Optional: Verify the environment is compatible with SB3"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"initial_state = env.clone_state()"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"obs, _ = env.reset()"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"model_save_path = \"wordle_ppo_model\"\n",
"model = PPO.load(model_save_path)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"action, _ = model.predict(obs)\n",
"obs, reward, done, _, info = env.step(action)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"5"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"action % 26"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"5"
]
},
"execution_count": 28,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ord('f') - ord('a')"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"'f'"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"chr(ord('a') + action % 26)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
" 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
" 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1,\n",
" 1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1,\n",
" 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1,\n",
" 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1,\n",
" 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1, 1, 1, 0, 1,\n",
" 1, 1])"
]
},
"execution_count": 16,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"obs"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"env.set_state(initial_state)"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"False"
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all(env.get_obs() == obs)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"# Perform your action to see the outcome\n",
"action = # Define your action\n",
"observation, reward, done, info = env.step(action)\n",
"\n",
"# Revert to the initial state\n",
"env.env.set_state(initial_state)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"import wandb\n",
"from wandb.integration.sb3 import WandbCallback"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Failed to detect the name of this notebook, you can set it manually with the WANDB_NOTEBOOK_NAME environment variable to enable code saving.\n",
"\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mltcptgeneral\u001b[0m (\u001b[33mfulltime\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n"
]
},
{
"data": {
"text/html": [
"Tracking run with wandb version 0.16.4"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"Run data is saved locally in <code>/home/art/cse151b-final-project/wandb/run-20240319_211220-cyh5nscz</code>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
"Syncing run <strong><a href='https://wandb.ai/fulltime/wordle/runs/cyh5nscz' target=\"_blank\">distinctive-flower-20</a></strong> to <a href='https://wandb.ai/fulltime/wordle' target=\"_blank\">Weights & Biases</a> (<a href='https://wandb.me/run' target=\"_blank\">docs</a>)<br/>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" View project at <a href='https://wandb.ai/fulltime/wordle' target=\"_blank\">https://wandb.ai/fulltime/wordle</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"text/html": [
" View run at <a href='https://wandb.ai/fulltime/wordle/runs/cyh5nscz' target=\"_blank\">https://wandb.ai/fulltime/wordle/runs/cyh5nscz</a>"
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"model_save_path = \"wordle_ppo_model_test\"\n",
"config = {\n",
" \"policy_type\": \"MlpPolicy\",\n",
" \"total_timesteps\": 200_000\n",
"}\n",
"run = wandb.init(\n",
" project=\"wordle\",\n",
" config=config,\n",
" sync_tensorboard=True\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Using cuda device\n",
"Wrapping the env with a `Monitor` wrapper\n",
"Wrapping the env in a DummyVecEnv.\n",
"Logging to runs/cyh5nscz/PPO_1\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "ca60c274a90b4dddaf275fe164012f16",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
"Output()"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"---------------------------------\n",
"| rollout/ | |\n",
"| ep_len_mean | 2.54 |\n",
"| ep_rew_mean | -3.66 |\n",
"| time/ | |\n",
"| fps | 721 |\n",
"| iterations | 1 |\n",
"| time_elapsed | 2 |\n",
"| total_timesteps | 2048 |\n",
"---------------------------------\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"-----------------------------------------\n",
"| rollout/ | |\n",
"| ep_len_mean | 2.53 |\n",
"| ep_rew_mean | -3.61 |\n",
"| time/ | |\n",
"| fps | 718 |\n",
"| iterations | 2 |\n",
"| time_elapsed | 5 |\n",
"| total_timesteps | 4096 |\n",
"| train/ | |\n",
"| approx_kl | 0.011673957 |\n",
"| clip_fraction | 0.0292 |\n",
"| clip_range | 0.2 |\n",
"| entropy_loss | -3.25 |\n",
"| explained_variance | -0.126 |\n",
"| learning_rate | 0.0003 |\n",
"| loss | 0.576 |\n",
"| n_updates | 10 |\n",
"| policy_gradient_loss | -0.0197 |\n",
"| value_loss | 3.58 |\n",
"-----------------------------------------\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"-----------------------------------------\n",
"| rollout/ | |\n",
"| ep_len_mean | 2.7 |\n",
"| ep_rew_mean | -3.56 |\n",
"| time/ | |\n",
"| fps | 698 |\n",
"| iterations | 3 |\n",
"| time_elapsed | 8 |\n",
"| total_timesteps | 6144 |\n",
"| train/ | |\n",
"| approx_kl | 0.019258872 |\n",
"| clip_fraction | 0.198 |\n",
"| clip_range | 0.2 |\n",
"| entropy_loss | -3.22 |\n",
"| explained_variance | -0.211 |\n",
"| learning_rate | 0.0003 |\n",
"| loss | 0.187 |\n",
"| n_updates | 20 |\n",
"| policy_gradient_loss | -0.0215 |\n",
"| value_loss | 0.637 |\n",
"-----------------------------------------\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"-----------------------------------------\n",
"| rollout/ | |\n",
"| ep_len_mean | 2.73 |\n",
"| ep_rew_mean | -3.43 |\n",
"| time/ | |\n",
"| fps | 681 |\n",
"| iterations | 4 |\n",
"| time_elapsed | 12 |\n",
"| total_timesteps | 8192 |\n",
"| train/ | |\n",
"| approx_kl | 0.021500897 |\n",
"| clip_fraction | 0.171 |\n",
"| clip_range | 0.2 |\n",
"| entropy_loss | -3.17 |\n",
"| explained_variance | 0.378 |\n",
"| learning_rate | 0.0003 |\n",
"| loss | 0.185 |\n",
"| n_updates | 30 |\n",
"| policy_gradient_loss | -0.0214 |\n",
"| value_loss | 0.479 |\n",
"-----------------------------------------\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"-----------------------------------------\n",
"| rollout/ | |\n",
"| ep_len_mean | 2.92 |\n",
"| ep_rew_mean | -3.36 |\n",
"| time/ | |\n",
"| fps | 682 |\n",
"| iterations | 5 |\n",
"| time_elapsed | 14 |\n",
"| total_timesteps | 10240 |\n",
"| train/ | |\n",
"| approx_kl | 0.018113121 |\n",
"| clip_fraction | 0.101 |\n",
"| clip_range | 0.2 |\n",
"| entropy_loss | -3.11 |\n",
"| explained_variance | 0.448 |\n",
"| learning_rate | 0.0003 |\n",
"| loss | 0.203 |\n",
"| n_updates | 40 |\n",
"| policy_gradient_loss | -0.0183 |\n",
"| value_loss | 0.455 |\n",
"-----------------------------------------\n"
]
}
],
"source": [
"model = PPO(config[\"policy_type\"], env=env, verbose=2, tensorboard_log=f\"runs/{run.id}\", batch_size=64)\n",
"\n",
"# Train for a certain number of timesteps\n",
"model.learn(\n",
" total_timesteps=config[\"total_timesteps\"],\n",
" callback=WandbCallback(\n",
" model_save_path=f\"models/{run.id}\",\n",
" verbose=2,\n",
" ),\n",
"\tprogress_bar=True\n",
")\n",
"\n",
"run.finish()"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"model.save(model_save_path)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"model = PPO.load(model_save_path)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": [
"rewards = 0\n",
"for i in tqdm(range(1000)):\n",
" obs, _ = env.reset()\n",
" done = False\n",
" while not done:\n",
" action, _ = model.predict(obs)\n",
" obs, reward, done, _, info = env.step(action)\n",
" rewards += reward\n",
"print(rewards / 1000)"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "env",
"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": "3.11.5"
}
},
"nbformat": 4,
"nbformat_minor": 2
}

1160
dqn_wordle.ipynb
View File

File diff suppressed because it is too large Load Diff

129
eric_wordle/.gitignore vendored Normal file
View File

@ -0,0 +1,129 @@
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class
# C extensions
*.so
# Distribution / packaging
.Python
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
pip-wheel-metadata/
share/python-wheels/
*.egg-info/
.installed.cfg
*.egg
MANIFEST
# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec
# Installer logs
pip-log.txt
pip-delete-this-directory.txt
# Unit test / coverage reports
htmlcov/
.tox/
.nox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
*.py,cover
.hypothesis/
.pytest_cache/
# Translations
*.mo
*.pot
# Django stuff:
*.log
local_settings.py
db.sqlite3
db.sqlite3-journal
# Flask stuff:
instance/
.webassets-cache
# Scrapy stuff:
.scrapy
# Sphinx documentation
docs/_build/
# PyBuilder
target/
# Jupyter Notebook
.ipynb_checkpoints
# IPython
profile_default/
ipython_config.py
# pyenv
.python-version
# pipenv
# According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control.
# However, in case of collaboration, if having platform-specific dependencies or dependencies
# having no cross-platform support, pipenv may install dependencies that don't work, or not
# install all needed dependencies.
#Pipfile.lock
# PEP 582; used by e.g. github.com/David-OConnor/pyflow
__pypackages__/
# Celery stuff
celerybeat-schedule
celerybeat.pid
# SageMath parsed files
*.sage.py
# Environments
.env
.venv
env/
venv/
ENV/
env.bak/
venv.bak/
# Spyder project settings
.spyderproject
.spyproject
# Rope project settings
.ropeproject
# mkdocs documentation
/site
# mypy
.mypy_cache/
.dmypy.json
dmypy.json
# Pyre type checker
.pyre/

11
eric_wordle/README.md Normal file
View File

@ -0,0 +1,11 @@
# N-dle Solver
A solver designed to beat New York Time's Wordle (link [here](https://www.nytimes.com/games/wordle/index.html)). If you are bored enough, can extend to solve the more general N-dle problem (for quordle, octordle, etc.)
I originally made this out of frustration for the game (and my own lack of lingual talent). One day, my friend thought she could beat my bot. To her dismay, she learned that she is no better than a machine. Let's see if you can do any better (the average number of attempts is 3.6).
## Usage:
1. Run `python main.py --n 1`
2. Follow the prompts
Currently only supports solving for 1 word at a time (i.e. wordle).

205
eric_wordle/ai.py Normal file
View File

@ -0,0 +1,205 @@
import re
import string
import numpy as np
from stable_baselines3 import PPO, DQN
from letter_guess import LetterGuessingEnv
import torch
def load_valid_words(file_path='wordle_words.txt'):
"""
Load valid five-letter words from a specified text file.
Parameters:
- file_path (str): The path to the text file containing valid words.
Returns:
- list[str]: A list of valid words loaded from the file.
"""
with open(file_path, 'r') as file:
valid_words = [line.strip() for line in file if len(line.strip()) == 5]
return valid_words
class AI:
def __init__(self, vocab_file, model_file, num_letters=5, num_guesses=6, use_q_model=False):
self.vocab_file = vocab_file
self.num_letters = num_letters
self.num_guesses = 6
self.vocab, self.vocab_scores, self.letter_scores = self.get_vocab(self.vocab_file)
self.best_words = sorted(list(self.vocab_scores.items()), key=lambda tup: tup[1])[::-1]
self.domains = None
self.possible_letters = None
self.use_q_model = use_q_model
if use_q_model:
# we initialize the same q env as the model train ONLY to simplify storing/calculating the gym state, not used to control the game at all
self.q_env = LetterGuessingEnv(load_valid_words(vocab_file))
self.q_env_state, _ = self.q_env.reset()
# load model
self.q_model = PPO.load(model_file)
self.reset("")
def solve_eval(self, results_callback):
num_guesses = 0
while [len(e) for e in self.domains] != [1 for _ in range(self.num_letters)]:
num_guesses += 1
if self.use_q_model:
self.freeze_state = self.q_env.clone_state()
# sample a word, this would use the q_env_state if the q_model is used
word = self.sample(num_guesses)
# get emulated results
results = results_callback(word)
if self.use_q_model:
self.q_env.set_state(self.freeze_state)
# step the q_env to match the guess we just made
for i in range(len(word)):
char = word[i]
action = ord(char) - ord('a')
self.q_env_state, _, _, _, _ = self.q_env.step(action)
self.arc_consistency(word, results)
return num_guesses, word
def solve(self):
num_guesses = 0
while [len(e) for e in self.domains] != [1 for _ in range(self.num_letters)]:
num_guesses += 1
if self.use_q_model:
self.freeze_state = self.q_env.clone_state()
# sample a word, this would use the q_env_state if the q_model is used
word = self.sample(num_guesses)
print('-----------------------------------------------')
print(f'Guess #{num_guesses}/{self.num_guesses}: {word}')
print('-----------------------------------------------')
print(f'Performing arc consistency check on {word}...')
print(f'Specify 0 for completely nonexistent letter at the specified index, 1 for existent letter but incorrect index, and 2 for correct letter at correct index.')
results = []
# Collect results
for l in word:
while True:
result = input(f'{l}: ')
if result not in ['0', '1', '2']:
print('Incorrect option. Try again.')
continue
results.append(result)
break
if self.use_q_model:
self.q_env.set_state(self.freeze_state)
# step the q_env to match the guess we just made
for i in range(len(word)):
char = word[i]
action = ord(char) - ord('a')
self.q_env_state, _, _, _, _ = self.q_env.step(action)
self.arc_consistency(word, results)
return num_guesses, word
def arc_consistency(self, word, results):
self.possible_letters += [word[i] for i in range(len(word)) if results[i] == '1']
for i in range(len(word)):
if results[i] == '0':
if word[i] in self.possible_letters:
if word[i] in self.domains[i]:
self.domains[i].remove(word[i])
else:
for j in range(len(self.domains)):
if word[i] in self.domains[j] and len(self.domains[j]) > 1:
self.domains[j].remove(word[i])
if results[i] == '1':
if word[i] in self.domains[i]:
self.domains[i].remove(word[i])
if results[i] == '2':
self.domains[i] = [word[i]]
def reset(self, target_word):
self.domains = [list(string.ascii_lowercase) for _ in range(self.num_letters)]
self.possible_letters = []
if self.use_q_model:
self.q_env_state, _ = self.q_env.reset()
self.q_env.target_word = target_word
def sample(self, num_guesses):
"""
Samples a best word given the current domains
:return:
"""
# Compile a regex of possible words with the current domain
regex_string = ''
for domain in self.domains:
regex_string += ''.join(['[', ''.join(domain), ']', '{1}'])
pattern = re.compile(regex_string)
# From the words with the highest scores, only return the best word that match the regex pattern
max_qval = float('-inf')
best_word = None
for word, _ in self.best_words:
# reset the state back to before we guessed a word
if pattern.match(word) and False not in [e in word for e in self.possible_letters]:
if self.use_q_model and num_guesses == 3:
self.q_env.set_state(self.freeze_state)
# Use policy to grade word
# get the state and action pairs
curr_qval = 0
for l in word:
action = ord(l) - ord('a')
q_val, _, _ = self.q_model.policy.evaluate_actions(self.q_model.policy.obs_to_tensor(self.q_env.get_obs())[0], torch.Tensor(np.array([action])).to("cuda"))
_, _, _, _, _ = self.q_env.step(action)
curr_qval += q_val
if curr_qval > max_qval:
max_qval = curr_qval
best_word = word
else:
# otherwise return the word from eric heuristic
return word
return best_word
def get_vocab(self, vocab_file):
vocab = []
with open(vocab_file, 'r') as f:
for l in f:
vocab.append(l.strip())
# Count letter frequencies at each index
letter_freqs = [{letter: 0 for letter in string.ascii_lowercase} for _ in range(self.num_letters)]
for word in vocab:
for i, l in enumerate(word):
letter_freqs[i][l] += 1
# Assign a score to each letter at each index by the probability of it appearing
letter_scores = [{letter: 0 for letter in string.ascii_lowercase} for _ in range(self.num_letters)]
for i in range(len(letter_scores)):
max_freq = np.max(list(letter_freqs[i].values()))
for l in letter_scores[i].keys():
letter_scores[i][l] = letter_freqs[i][l] / max_freq
# Find a sorted list of words ranked by sum of letter scores
vocab_scores = {} # (score, word)
for word in vocab:
score = 0
for i, l in enumerate(word):
score += letter_scores[i][l]
# # Optimization: If repeating letters, deduct a couple points
# if len(set(word)) < len(word):
# score -= 0.25 * (len(word) - len(set(word)))
vocab_scores[word] = score
return vocab, vocab_scores, letter_scores

37
eric_wordle/dist.py Normal file
View File

@ -0,0 +1,37 @@
import string
import numpy as np
words = []
with open('words.txt', 'r') as f:
for l in f:
words.append(l.strip())
# Count letter frequencies at each index
letter_freqs = [{letter: 0 for letter in string.ascii_lowercase} for _ in range(5)]
for word in words:
for i, l in enumerate(word):
letter_freqs[i][l] += 1
# Assign a score to each letter at each index by the probability of it appearing
letter_scores = [{letter: 0 for letter in string.ascii_lowercase} for _ in range(5)]
for i in range(len(letter_scores)):
max_freq = np.max(list(letter_freqs[i].values()))
for l in letter_scores[i].keys():
letter_scores[i][l] = letter_freqs[i][l] / max_freq
# Find a sorted list of words ranked by sum of letter scores
word_scores = [] # (score, word)
for word in words:
score = 0
for i, l in enumerate(word):
score += letter_scores[i][l]
word_scores.append((score, word))
sorted_by_second = sorted(word_scores, key=lambda tup: tup[0])[::-1]
print(sorted_by_second[:10])
for i, (score, word) in enumerate(sorted_by_second):
if word == 'soare':
print(f'{word} with a score of {score} is found at index {i}')

62
eric_wordle/eval.py Normal file
View File

@ -0,0 +1,62 @@
import argparse
from ai import AI
import numpy as np
from tqdm import tqdm
global solution
def result_callback(word):
global solution
result = ['0', '0', '0', '0', '0']
for i, letter in enumerate(word):
if solution[i] == word[i]:
result[i] = '2'
elif letter in solution:
result[i] = '1'
else:
pass
return result
def main(args):
global solution
if args.n is None:
raise Exception('Need to specify n (i.e. n = 1 for wordle, n = 4 for quordle, n = 16 for sedecordle).')
ai = AI(args.vocab_file, args.model_file, use_q_model=args.q_model)
total_guesses = 0
wins = 0
num_eval = args.num_eval
np.random.seed(0)
for i in tqdm(range(num_eval)):
idx = np.random.choice(range(len(ai.vocab)))
solution = ai.vocab[idx]
ai.reset(solution)
guesses, word = ai.solve_eval(results_callback=result_callback)
if word != solution:
total_guesses += 5
else:
total_guesses += guesses
wins += 1
print(f"q_model?: {args.q_model} \t average guesses per game: {total_guesses / num_eval} \t win rate: {wins / num_eval}")
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--n', dest='n', type=int, default=None)
parser.add_argument('--vocab_file', dest='vocab_file', type=str, default='wordle_words.txt')
parser.add_argument('--num_eval', dest="num_eval", type=int, default=1000)
parser.add_argument('--model_file', dest="model_file", type=str, default='wordle_ppo_model')
parser.add_argument('--q_model', dest="q_model", type=bool, default=False)
args = parser.parse_args()
main(args)

1
eric_wordle/letter_guess.py Symbolic link
View File

@ -0,0 +1 @@
../letter_guess.py

21
eric_wordle/main.py Normal file
View File

@ -0,0 +1,21 @@
import argparse
from ai import AI
def main(args):
if args.n is None:
raise Exception('Need to specify n (i.e. n = 1 for wordle, n = 4 for quordle, n = 16 for sedecordle).')
print(f"using q model? {args.q_model}")
ai = AI(args.vocab_file, args.model_file, use_q_model=args.q_model)
ai.reset("lingo")
ai.solve()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument('--n', dest='n', type=int, default=None)
parser.add_argument('--vocab_file', dest='vocab_file', type=str, default='wordle_words.txt')
parser.add_argument('--model_file', dest="model_file", type=str, default='wordle_ppo_model')
parser.add_argument('--q_model', dest="q_model", type=bool, default=False)
args = parser.parse_args()
main(args)

15
eric_wordle/process.py Normal file
View File

@ -0,0 +1,15 @@
import pandas
print('Loading in words dictionary; this may take a while...')
df = pandas.read_json('words_dictionary.json')
print('Done loading words dictionary.')
words = []
for word in df.axes[0].tolist():
if len(word) != 5:
continue
words.append(word)
words.sort()
with open('words.txt', 'w') as f:
for word in words:
f.write(word + '\n')

8677
gym_wordle/dictionary/guess_list.csv → eric_wordle/words.txt
View File

File diff suppressed because it is too large Load Diff

370104
eric_wordle/words_dictionary.json Normal file
View File

File diff suppressed because it is too large Load Diff

2
eval.sh Executable file
View File

@ -0,0 +1,2 @@
python eric_wordle/eval.py --n 1 --vocab_file wordle_words.txt --num_eval 5000
python eric_wordle/eval.py --n 1 --vocab_file wordle_words.txt --num_eval 5000 --q_model True --model_file wordle_ppo_model

7
gym_wordle/__init__.py
View File

@ -1,7 +0,0 @@
from gym.envs.registration import register
from .wordle import WordleEnv
register(
id='Wordle-v0',
entry_point='gym_wordle.wordle:WordleEnv'
)

BIN
gym_wordle/dictionary/guess_list.npy
View File

Binary file not shown.

BIN
gym_wordle/dictionary/solution_list.npy
View File

Binary file not shown.

93
gym_wordle/utils.py
View File

@ -1,93 +0,0 @@
import numpy as np
import numpy.typing as npt
from pathlib import Path
_chars = ' abcdefghijklmnopqrstuvwxyz'
_char_d = {c: i for i, c in enumerate(_chars)}
def to_english(array: npt.NDArray[np.int64]) -> str:
"""Converts a numpy integer array into a corresponding English string.
Args:
array: Word in array (int) form. It is assumed that each integer in the
array is between 0,...,26 (inclusive).
Returns:
A (lowercase) string representation of the word.
"""
return ''.join(_chars[i] for i in array)
def to_array(word: str) -> npt.NDArray[np.int64]:
"""Converts a string of characters into a corresponding numpy array.
Args:
word: Word in string form. It is assumed that each character in the
string is either an empty space ' ' or lowercase alphabetical
character.
Returns:
An array representation of the word.
"""
return np.array([_char_d[c] for c in word])
def get_words(category: str, build: bool=False) -> npt.NDArray[np.int64]:
"""Loads a list of words in array form.
If specified, this will recompute the list from the human-readable list of
words, and save the results in array form.
Args:
category: Either 'guess' or 'solution', which corresponds to the list
of acceptable guess words and the list of acceptable solution words.
build: If True, recomputes and saves the array-version of the computed
list for future access.
Returns:
An array representation of the list of words specified by the category.
This array has two dimensions, and the number of columns is fixed at
five.
"""
assert category in {'guess', 'solution'}
arr_path = Path(__file__).parent / f'dictionary/{category}_list.npy'
if build:
list_path = Path(__file__).parent / f'dictionary/{category}_list.csv'
with open(list_path, 'r') as f:
words = np.array([to_array(line.strip()) for line in f])
np.save(arr_path, words)
return np.load(arr_path)
def play():
"""Play Wordle yourself!"""
import gym
import gym_wordle
env = gym.make('Wordle-v0') # load the environment
env.reset()
solution = to_english(env.unwrapped.solution_space[env.solution]).upper() # no peeking!
done = False
while not done:
action = -1
# in general, the environment won't be forgiving if you input an
# invalid word, but for this function I want to let you screw up user
# input without consequence, so just loops until valid input is taken
while not env.action_space.contains(action):
guess = input('Guess: ')
action = env.unwrapped.action_space.index_of(to_array(guess))
state, reward, done, info = env.step(action)
env.render()
print(f"The word was {solution}")

296
gym_wordle/wordle.py
View File

@ -1,296 +0,0 @@
import gym
import numpy as np
import numpy.typing as npt
from sty import fg, bg, ef, rs
from collections import Counter
from gym_wordle.utils import to_english, to_array, get_words
from typing import Optional
class WordList(gym.spaces.Discrete):
"""Super class for defining a space of valid words according to a specified
list.
TODO: Fix these paragraphs
The space is a subclass of gym.spaces.Discrete, where each element
corresponds to an index of a valid word in the word list. The obfuscation
is necessary for more direct implementation of RL algorithms, which expect
spaces of less sophisticated form.
In addition to the default methods of the Discrete space, it implements
a __getitem__ method for easy index lookup, and an index_of method to
convert potential words into their corresponding index (if they exist).
"""
def __init__(self, words: npt.NDArray[np.int64], **kwargs):
"""
Args:
words: Collection of words in array form with shape (_, 5), where
each word is a row of the array. Each array element is an integer
between 0,...,26 (inclusive).
kwargs: See documentation for gym.spaces.MultiDiscrete
"""
super().__init__(words.shape[0], **kwargs)
self.words = words
def __getitem__(self, index: int) -> npt.NDArray[np.int64]:
"""Obtains the (int-encoded) word associated with the given index.
Args:
index: Index for the list of words.
Returns:
Associated word at the position specified by index.
"""
return self.words[index]
def index_of(self, word: npt.NDArray[np.int64]) -> int:
"""Given a word, determine its index in the list (if it exists),
otherwise returning -1 if no index exists.
Args:
word: Word to find in the word list.
Returns:
The index of the given word if it exists, otherwise -1.
"""
try:
index, = np.nonzero((word == self.words).all(axis=1))
return index[0]
except:
return -1
class SolutionList(WordList):
"""Space for *solution* words to the Wordle environment.
In the game Wordle, there are two different collections of words:
* "guesses", which the game accepts as valid words to use to guess the
answer.
* "solutions", which the game uses to choose solutions from.
Of course, the set of solutions is a strict subset of the set of guesses.
Reference: https://fivethirtyeight.com/features/when-the-riddler-met-wordle/
This class represents the set of solution words.
"""
def __init__(self, **kwargs):
"""
Args:
kwargs: See documentation for gym.spaces.MultiDiscrete
"""
words = get_words('solution')
super().__init__(words, **kwargs)
class WordleObsSpace(gym.spaces.Box):
"""Implementation of the state (observation) space in terms of gym
primatives, in this case, gym.spaces.Box.
The Wordle observation space can be thought of as a 6x5 array with two
channels:
- the character channel, indicating which characters are placed on the
board (unfilled rows are marked with the empty character, 0)
- the flag channel, indicating the in-game information associated with
each character's placement (green highlight, yellow highlight, etc.)
where there are 6 rows, one for each turn in the game, and 5 columns, since
the solution will always be a word of length 5.
For simplicity, and compatibility with the stable_baselines algorithms,
this multichannel is modeled as a 6x10 array, where the two channels are
horizontally appended (along columns). Thus each row in the observation
should be interpreted as
c0 c1 c2 c3 c4 f0 f1 f2 f3 f4
when the word is c0...c4 and its associated flags are f0...f4.
While the superclass method `sample` is available to the WordleObsSpace, it
should be emphasized that the output of `sample` will (almost surely) not
correspond to a real game configuration, because the sampling is not out of
possible game configurations. Instead, the Box superclass just samples the
integer array space uniformly.
"""
def __init__(self, **kwargs):
self.n_rows = 6
self.n_cols = 5
self.max_char = 26
self.max_flag = 4
low = np.zeros((self.n_rows, 2*self.n_cols))
high = np.c_[np.full((self.n_rows, self.n_cols), self.max_char),
np.full((self.n_rows, self.n_cols), self.max_flag)]
super().__init__(low, high, dtype=np.int64, **kwargs)
class GuessList(WordList):
"""Space for *solution* words to the Wordle environment.
In the game Wordle, there are two different collections of words:
* "guesses", which the game accepts as valid words to use to guess the
answer.
* "solutions", which the game uses to choose solutions from.
Of course, the set of solutions is a strict subset of the set of guesses.
Reference: https://fivethirtyeight.com/features/when-the-riddler-met-wordle/
This class represents the set of guess words.
"""
def __init__(self, **kwargs):
"""
Args:
kwargs: See documentation for gym.spaces.MultiDiscrete
"""
words = get_words('guess')
super().__init__(words, **kwargs)
class WordleEnv(gym.Env):
metadata = {'render.modes': ['human']}
# character flag codes
no_char = 0
right_pos = 1
wrong_pos = 2
wrong_char = 3
def __init__(self):
super().__init__()
self.seed()
self.action_space = GuessList()
self.solution_space = SolutionList()
self.observation_space = WordleObsSpace()
self._highlights = {
self.right_pos: (bg.green, bg.rs),
self.wrong_pos: (bg.yellow, bg.rs),
self.wrong_char: ('', ''),
self.no_char: ('', ''),
}
self.n_rounds = 6
self.n_letters = 5
def _highlighter(self, char: str, flag: int) -> str:
"""Terminal renderer functionality. Properly highlights a character
based on the flag associated with it.
Args:
char: Character in question.
flag: Associated flag, one of:
- 0: no character (render no background)
- 1: right position (render green background)
- 2: wrong position (render yellow background)
- 3: wrong character (render no background)
Returns:
Correct ASCII sequence producing the desired character in the
correct background.
"""
front, back = self._highlights[flag]
return front + char + back
def reset(self):
self.round = 0
self.solution = self.solution_space.sample()
self.state = np.zeros((self.n_rounds, 2 * self.n_letters), dtype=np.int64)
return self.state
def render(self, mode: str ='human'):
"""Renders the Wordle environment.
Currently supported render modes:
- human: renders the Wordle game to the terminal.
Args:
mode: the mode to render with
"""
if mode == 'human':
for row in self.states:
text = ''.join(map(
self._highlighter,
to_english(row[:self.n_letters]).upper(),
row[self.n_letters:]
))
print(text)
else:
super(WordleEnv, self).render(mode=mode)
def step(self, action):
"""Run one step of the Wordle game. Every game must be previously
initialized by a call to the `reset` method.
Args:
action: Word guessed by the agent.
Returns:
state (object): Wordle game state after the guess.
reward (float): Reward associated with the guess (-1 for incorrect,
0 for correct)
done (bool): Whether the game has ended (by a correct guess or
after six guesses).
info (dict): Auxiliary diagnostic information (empty).
"""
assert self.action_space.contains(action), 'Invalid word!'
# transform the action, solution indices to their words
action = self.action_space[action]
solution = self.solution_space[self.solution]
# populate the word chars into the row (character channel)
self.state[self.round][:self.n_letters] = action
# populate the flag characters into the row (flag channel)
counter = Counter()
for i, char in enumerate(action):
flag_i = i + self.n_letters # starts at 5
counter[char] += 1
if char == solution[i]: # character is in correct position
self.state[self.round, flag_i] = self.right_pos
elif counter[char] <= (char == solution).sum():
# current character has been seen within correct number of
# occurrences
self.state[self.round, flag_i] = self.wrong_pos
else:
# wrong character, or "correct" character too many times
self.state[self.round, flag_i] = self.wrong_char
self.round += 1
correct = (action == solution).all()
game_over = (self.round == self.n_rounds)
done = correct or game_over
# Total reward equals -(number of incorrect guesses)
# reward = 0. if correct else -1.
# correct +10
# guesses new letter +1
# guesses correct letter +1
# spent another guess -1
reward = 0
reward += np.sum(self.state[:, 5:] == 1) * 1
reward += np.sum(self.state[:, 5:] == 2) * 0.5
reward += np.sum(self.state[:, 5:] == 3) * -1
reward += 10 if correct else -10 if done else 0
info = {'correct': correct}
return self.state, reward, done, info

130
letter_guess.py Normal file
View File

@ -0,0 +1,130 @@
import gymnasium as gym
from gymnasium import spaces
import numpy as np
import random
import re
import copy
class LetterGuessingEnv(gym.Env):
"""
Custom Gymnasium environment for a letter guessing game with a focus on forming
valid prefixes and words from a list of valid Wordle words. The environment tracks
the current guess prefix and validates it against known valid words, ending the game
early with a negative reward for invalid prefixes.
"""
metadata = {'render_modes': ['human']}
def __init__(self, valid_words, seed=None):
self.action_space = spaces.Discrete(26)
self.observation_space = spaces.Box(low=0, high=1, shape=(26*2 + 26*4,), dtype=np.int32)
self.valid_words = valid_words # List of valid Wordle words
self.target_word = '' # Target word for the current episode
self.valid_words_str = ' '.join(self.valid_words) + ' '
self.letter_flags = None
self.letter_positions = None
self.guessed_letters = set()
self.guess_prefix = "" # Tracks the current guess prefix
self.reset()
def clone_state(self):
# Clone the current state
return {
'target_word': self.target_word,
'letter_flags': copy.deepcopy(self.letter_flags),
'letter_positions': copy.deepcopy(self.letter_positions),
'guessed_letters': copy.deepcopy(self.guessed_letters),
'guess_prefix': self.guess_prefix,
'round': self.round
}
def set_state(self, state):
# Restore the state
self.target_word = state['target_word']
self.letter_flags = copy.deepcopy(state['letter_flags'])
self.letter_positions = copy.deepcopy(state['letter_positions'])
self.guessed_letters = copy.deepcopy(state['guessed_letters'])
self.guess_prefix = state['guess_prefix']
self.round = state['round']
def step(self, action):
letter_index = action # Assuming action is the letter index directly
position = len(self.guess_prefix) # The next position in the prefix is determined by its current length
letter = chr(ord('a') + letter_index)
reward = 0
done = False
# Check if the letter has already been used in the guess prefix
if letter in self.guessed_letters:
reward = -1 # Penalize for repeating letters in the prefix
else:
# Add the new letter to the prefix and update guessed letters set
self.guess_prefix += letter
self.guessed_letters.add(letter)
# Update letter flags based on whether the letter is in the target word
if self.target_word[position] == letter:
self.letter_flags[letter_index, :] = [1, 0] # Update flag for correct guess
elif letter in self.target_word:
self.letter_flags[letter_index, :] = [0, 1] # Update flag for correct guess wrong position
else:
self.letter_flags[letter_index, :] = [0, 0] # Update flag for incorrect guess
reward = 1 # Reward for adding new information by trying a new letter
# Update the letter_positions matrix to reflect the new guess
if position == 4:
self.letter_positions[:, :] = 1
else:
self.letter_positions[:, position] = 0
self.letter_positions[letter_index, position] = 1
# Use regex to check if the current prefix can lead to a valid word
if not re.search(r'\b' + self.guess_prefix, self.valid_words_str):
reward = -5 # Penalize for forming an invalid prefix
done = True # End the episode if the prefix is invalid
# guessed a full word so we reset our guess prefix to guess next round
if len(self.guess_prefix) == len(self.target_word):
self.guess_prefix = ''
self.round += 1
# end after 3 rounds of total guesses
if self.round == 3:
# reward = 5
done = True
obs = self.get_obs()
if reward < -5:
print(obs, reward, done)
exit(0)
return obs, reward, done, False, {}
def reset(self, seed=None):
self.target_word = random.choice(self.valid_words)
# self.target_word_encoded = self.encode_word(self.target_word)
self.letter_flags = np.ones((26, 2), dtype=np.int32)
self.letter_positions = np.ones((26, 4), dtype=np.int32)
self.guessed_letters = set()
self.guess_prefix = "" # Reset the guess prefix for the new episode
self.round = 0
return self.get_obs(), {}
def encode_word(self, word):
encoded = np.zeros((26,))
for char in word:
index = ord(char) - ord('a')
encoded[index] = 1
return encoded
def get_obs(self):
return np.concatenate([self.letter_flags.flatten(), self.letter_positions.flatten()])
def render(self, mode='human'):
pass # Optional: Implement rendering logic if needed

4606
gym_wordle/dictionary/solution_list.csv → wordle_words.txt
View File

File diff suppressed because it is too large Load Diff