Pinned Repositories
A2C_TF2.0_Keras_LunarLander
This is a TF2.0 Keras implementation of a A2C agent (tested for openai lunar lander v2)
LSTM-Autoencoder-for-rare-events
An LSTM (binary classifier) Autoencoder to identify rare events written using TF2.0/Keras
LSTM-Cell-using-Numpy
An implementation of an LSTM cell puerly written using Numpy (and scipy).
LunarLanderDuelingDQN
A minimalistic tensorflow.keras (tf2.0) implentaion of a dueling DQN to solve the LunarLander-v2 gym env.
MonteCarloGridworld
This is my implementaion of a Monte Carlo Tree Search ML solution to gridworld. Part of Move37 - Capter 3-4. Credits go to LazyProgrammer
ProximalPolicyOptimizationContinuousKeras
This is an Tensorflow 2.0 (Keras) implementation of a Open Ai's proximal policy optimization PPO algorithem for continuous action spaces.
ProximalPolicyOptimizationKeras
This is a deterministic Tensorflow 2.0 (keras) implementation of a Open Ai's proximal policy optimization actor critic algorithm PPO.
TF2.0_Keras_DDPG
A commented Tensorflow 2.0 Keras implementation of DDPG for open AI gym continuous environments.
VAE_CNN_Keras_TF2.0
Two different implementations of a Variational Autoencoder VAE with convolutional Neural networks via Tesorflow 2.0/Keras.
VanillaPolicyGradientAlgorithm
This is an Tensorflow.Keras (TF 2.0) based implementation of a vanilla Policy Gradient learner to solve OpenAi Gym's Cartpole.
nric's Repositories
nric/ProximalPolicyOptimizationKeras
This is a deterministic Tensorflow 2.0 (keras) implementation of a Open Ai's proximal policy optimization actor critic algorithm PPO.
nric/ProximalPolicyOptimizationContinuousKeras
This is an Tensorflow 2.0 (Keras) implementation of a Open Ai's proximal policy optimization PPO algorithem for continuous action spaces.
nric/A2C_TF2.0_Keras_LunarLander
This is a TF2.0 Keras implementation of a A2C agent (tested for openai lunar lander v2)
nric/LSTM-Autoencoder-for-rare-events
An LSTM (binary classifier) Autoencoder to identify rare events written using TF2.0/Keras
nric/TF2.0_Keras_DDPG
A commented Tensorflow 2.0 Keras implementation of DDPG for open AI gym continuous environments.
nric/LSTM-Cell-using-Numpy
An implementation of an LSTM cell puerly written using Numpy (and scipy).
nric/VanillaPolicyGradientAlgorithm
This is an Tensorflow.Keras (TF 2.0) based implementation of a vanilla Policy Gradient learner to solve OpenAi Gym's Cartpole.
nric/LunarLanderDuelingDQN
A minimalistic tensorflow.keras (tf2.0) implentaion of a dueling DQN to solve the LunarLander-v2 gym env.
nric/MonteCarloGridworld
This is my implementaion of a Monte Carlo Tree Search ML solution to gridworld. Part of Move37 - Capter 3-4. Credits go to LazyProgrammer
nric/VAE_CNN_Keras_TF2.0
Two different implementations of a Variational Autoencoder VAE with convolutional Neural networks via Tesorflow 2.0/Keras.
nric/A3C_Atari
This is an implementaion of an asyncronous advantage actor critic A3C algorithm to play open ai gym atari games.
nric/AugmentedRandomSearchGym
An implementation for Augmented Radom Search algorithm solivng 2d open ai gym enviroments - tested for Box2d envs.
nric/AugmentedRandomSearchGymBipedalWalker
Augmented Random Search was chosen to make the Open AI Bipedal Walker walk. However, the Agent is general. It could be used for other gym environments as well without change. But probably hyper paramter would require adaptation.
nric/CartpoleSimpleDQN
This is a simple implementation of a Deep Q Network learning agent tested on Open AI Gym's cart pole.
nric/Coursera_Capstone
Capstone Project for IBM Data Science Certificate
nric/deep-reinforcement-learning
Repo for the Deep Reinforcement Learning Nanodegree program
nric/DeepQLeaningAtari
This is an TensorFlow Keras (TF2) implementaion of a vanilla Deep Q Leaning algorithm to play the OpenAI Gym Atari Games.
nric/DuelingDQNDoom
This is an implementation of a Dueling Deep Q Learning agent that learns to play Doom.
nric/GridworldQLearning
This is a solution to Gridworld using off policy Q Learning. As enviroment it emplys LazyProgrammers Grid_world.py https://github.com/lazyprogrammer/machine_learning_examples/tree/master/rl Solution is part of The School of AI's Move 37 Course https://www.theschool.ai/courses/move-37-course/ Written as a jupyter cell in visual studio code. Just run the cell. If you want to run using python interpreter directly, replace def main(): to if name == 'main': and remove the last line (call of main()).
nric/MonteCarloFrozenLake
A solution of Open Ai Gym FrozenLake-v0 using Monte Carlo first visit method in Python 3.6.7
nric/MonteCarloPolicyGradientAgent
This is a Monte Carlo Policy Gradient algorithm (somewhat) written using TF2.0 keras optimized to solve Open Ai Gym Lunar Lander.
nric/MPG_DataAnalysis_With_TF2.0
A test project for some data anysis and a simple Neural Network (2 fully connected layers) written with Tensor Flow 2.0 to predict consumption of a vehicle.
nric/NeuroEvolutionLunarLander
This is a my minimalisic solution to the Lular Lander gym environment using an evolutionaly Neural Net aproach.
nric/performer-pytorch
An implementation of Performer, a linear attention-based transformer, in Pytorch
nric/QLearningTaxiV2
A Q Learning Agent solving the OpenAI Gym TaxiV2 environment. Despite the Hyper parameters, the agent should be able to solve all gym toy text enviroments as it is written very general. Run code using a Jupyter Notebook or VS Code connected to a jupyter server.