Pinned Loading

1. GPT-2 GPT-2 Public

   PyTorch implementation of a GPT-2-inspired transformer language model, trained on English (Shakespeare) and Hebrew corpora, with attention visualizations and evaluation.

   Python 1

2. Denoising-Diffusion-Probabilistic-Model Denoising-Diffusion-Probabilistic-Model Public

   PyTorch implementation of the Denoising Diffusion Probabilistic Model (DDPM) with DDIM sampling and classifier-free guidance, based on Ho et al.'s paper, evaluated on the MNIST dataset.

   Python 2

3. Multi-Layer-Perceptron Multi-Layer-Perceptron Public

   Implementation of MLP from scratch using Python and NumPy with examples.

   Python

4. Deep-Reinforcement-Learning-Minigrid Deep-Reinforcement-Learning-Minigrid Public

   Deep RL agents for the MiniGrid 6 Rooms. Implements DQN, Double DQN, Huber Loss, Soft Updates, PER, and Actor-Critic. Best model solves 90% of episodes with improved training stability.

   Python

5. Auto-Encoding-Variational-Bayes Auto-Encoding-Variational-Bayes Public

   PyTorch implementation of the Variational Autoencoder (VAE) model from the Kingma & Welling paper 'Auto-Encoding Variational Bayes', evaluated on MNIST and Fashion MNIST datasets.

   Python

6. RL-FlappyBird RL-FlappyBird Public

   Reinforcement Learning agent that plays Flappy Bird using Q-Learning and SARSA, implemented from scratch. Includes a custom environment, tabular Q-table updates, and performance visualization.

   Jupyter Notebook