Skip to content

Instantly share code, notes, and snippets.

View ipsec's full-sized avatar

Fernando Ribeiro ipsec

View GitHub Profile
@ipsec
ipsec / lunarlander_v2_image.py
Created September 14, 2022 17:28
LunarLander-v2 using image
import gym
import numpy as np
from gym.spaces import Box
class ValidateLunarLanderImageEnv(gym.Env):
def __init__(self):
self.env = gym.make('LunarLander-v2')
obs = self.reset()
self.observation_space = Box(
@ipsec
ipsec / muzero.py
Created July 17, 2022 02:43 — forked from Mononofu/muzero.py
MuZero pseudocode
"""Pseudocode description of the MuZero algorithm."""
# pylint: disable=unused-argument
# pylint: disable=missing-docstring
# pylint: disable=g-explicit-length-test
import collections
import math
import typing
from typing import Any, Dict, List, Optional
@ipsec
ipsec / muzero_unplugged.py
Created July 17, 2022 02:42 — forked from Mononofu/muzero_unplugged.py
Pseudocode for MuZero Unplugged
"""Pseudocode description of the MuZero Unplugged algorithm."""
# pylint: disable=unused-argument
# pylint: disable=missing-docstring
# pylint: disable=g-explicit-length-test
import collections
import math
import random
import typing
from typing import Dict, List, Optional
@ipsec
ipsec / stochastic_muzero.py
Created July 17, 2022 02:42 — forked from Mononofu/stochastic_muzero.py
Pseudocode for Stochastic MuZero
"""Pseudocode description of the Stochastic MuZero algorithm.
This pseudocode was adapted from the original MuZero pseudocode.
"""
# pylint: disable=unused-argument
# pylint: disable=missing-docstring
# pylint: disable=g-explicit-length-test
import abc
import math
from typing import Any, Dict, Callable, List, NamedTuple, Tuple, Union, Optional, Sequence