Andrei Pokrovsky's starred repositories
chatgpt-demo
Minimal web UI for ChatGPT.
AITemplate
AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.
open_spiel
OpenSpiel is a collection of environments and algorithms for research in general reinforcement learning and search/planning in games.
tiny-cuda-nn
Lightning fast C++/CUDA neural network framework
machine-learning-articles
🧠💬 Articles I wrote about machine learning, archived from MachineCurve.com.
muzero-general
MuZero
Arcade-Learning-Environment
The Arcade Learning Environment (ALE) -- a platform for AI research.
torch-mlir
The Torch-MLIR project aims to provide first class support from the PyTorch ecosystem to the MLIR ecosystem.
ConcurrentDeque
Fast, generalized, implementation of the Chase-Lev lock-free work-stealing deque for C++17
cudnn-python-wrappers
Python wrappers for the NVIDIA cuDNN libraries
malloc-survey
:chart_with_upwards_trend: Allocation benchmarks
model-based-rl
Implementation of MuZero with PyTorch, based on the pseudocode from DeepMind (https://arxiv.org/src/1911.08265v2/anc/pseudocode.py).
CudaSharedPtr
Shared Pointer for Cuda Device Pointers and Cuda Streams, Smart Wrapper to Allocate and Deallocate Cuda Device Buffer.
MutexShootout
A benchmark to measure lock overhead and compare mutex performance under varying levels of contention.