Pinned Repositories
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.
ao
Custom data types and layouts for training and inference
csharp
practice
dotfiles
repo for unix/linux learning
echo_hack
VZ hackathon using Amazon Echo
einops
Deep learning operations reinvented (for pytorch, tensorflow, jax and others)
FBGEMM
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
flash-attention
Fast and memory-efficient exact attention
functorch
functorch is a prototype of JAX-like composable function transforms for PyTorch.
pytorch_open_registration_example
Example of using pytorch's open device registration API
bdhirsh's Repositories
bdhirsh/pytorch_open_registration_example
Example of using pytorch's open device registration API
bdhirsh/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.
bdhirsh/ao
Custom data types and layouts for training and inference
bdhirsh/csharp
practice
bdhirsh/dotfiles
repo for unix/linux learning
bdhirsh/echo_hack
VZ hackathon using Amazon Echo
bdhirsh/einops
Deep learning operations reinvented (for pytorch, tensorflow, jax and others)
bdhirsh/FBGEMM
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
bdhirsh/flash-attention
Fast and memory-efficient exact attention
bdhirsh/functorch
functorch is a prototype of JAX-like composable function transforms for PyTorch.
bdhirsh/icpc-trd
bdhirsh/practice
bdhirsh/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration