xw285cornell
I'm a research scientist at Facebook, working on hardware efficiency for machine learning.
Facebook Inc. Menlo Park, CA
Pinned Repositories
amdsmi
AMD SMI
audio
Data manipulation and transformation for audio signal processing, powered by PyTorch
buckit
Makes building C++ projects easier with Buck.
composable_kernel
Composable Kernel: Performance Portable Programming Model for Machine Learning Tensor Operators
convolution-visualizer
Convolution visualizations
d2go
D2Go is a toolkit for efficient deep learning
DeepLearningExamples
Deep Learning Examples
FBGEMM
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
gloo
Collective communications library with various primitives for multi-machine training.
glow
Compiler for Neural Network hardware accelerators
xw285cornell's Repositories
xw285cornell/amdsmi
AMD SMI
xw285cornell/audio
Data manipulation and transformation for audio signal processing, powered by PyTorch
xw285cornell/buckit
Makes building C++ projects easier with Buck.
xw285cornell/composable_kernel
Composable Kernel: Performance Portable Programming Model for Machine Learning Tensor Operators
xw285cornell/convolution-visualizer
Convolution visualizations
xw285cornell/d2go
D2Go is a toolkit for efficient deep learning
xw285cornell/DeepLearningExamples
Deep Learning Examples
xw285cornell/FBGEMM
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
xw285cornell/gloo
Collective communications library with various primitives for multi-machine training.
xw285cornell/glow
Compiler for Neural Network hardware accelerators
xw285cornell/kineto
A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.
xw285cornell/onnx
Open Neural Network Exchange
xw285cornell/ossci-job-dsl
Jenkins job definitions for OSSCI
xw285cornell/pybind11
Seamless operability between C++11 and Python
xw285cornell/pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
xw285cornell/rocm_smi_lib
ROCm SMI LIB
xw285cornell/torchrec
Pytorch domain library for recommendation systems
xw285cornell/triton
Development repository for the Triton language and compiler
xw285cornell/xformers
Hackable and optimized Transformers building blocks, supporting a composable construction.