per

Massively Parallel Deep Reinforcement Learning. 🔥

ASN.1 parsing, encoding and decoding.

The Ultimate Click and Aim Trainer 2026: Precision CPS Drills

ASN.1 data types and UPER encoder/decoder

road-map & paper review for Reinforcement Learning

A minimal, retro-inspired Hugo theme with an Apple System 7 aesthetic and zero JS

My reproduction of various reinforcement learning algorithms (DQN variants, A3C, DPPO, RND with PPO) in Tensorflow.

An ASN1 compiler producing Rust bindings for the rasn framework

ORM for accessing indexedDB as a promise base api implementation.

A small project to gather counter statistics from the file system or output from commands. Useful for retrieving packets per second and bytes per second on a network interface.

Collection of reinforcement learning algorithms implementations with TensorFlow2

Keras Implementation of DDPG(Deep Deterministic Policy Gradient) with PER(Prioritized Experience Replay) option on OpenAI gym framework

Library to decode/encode ASN.1 protocol (PER only)

An advanced ROS 2 package for autonomous drone navigation using SAC with Prioritized Experience Replay. Built on PX4, Gazebo Harmonic, and ROS 2 Humble, it enables intelligent obstacle avoidance with training scripts, test environments, and PX4 bridge configs.

Keras Implementation of TD3(Twin Delayed DDPG) with PER(Prioritized Experience Replay) option on OpenAI gym framework

A Beat saber Quest mod which let's you play in 3rd person

Deep Q-Network Prioritized Experience Replay (PER) binary sum-tree engine enabling O(log N) proportional transition sampling and TD-error ranking.

Moteur de calcul pour simulateurs fiscaux et sociaux français : simulateur de plafond Madelin, PER, indemnités journalières, TMI, capitalisation. Fonctions pures, typées, testées. Barèmes 2026.

Deep Q-Network Prioritized Experience Replay (PER) binary sum-tree engine enabling O(log N) proportional transition sampling and TD-error ranking.

Implementation code when learning deep reinforcement learning