Name: Xinfeng
Type: User
Company: UC, Santa Barbara
Bio: Ph.D. student in SEAL Lab, Dept. of Electrical and Computer Engineering at UC, Santa Barbara. Research interests include the computer system and architecture.
Location: https://seal.ece.ucsb.edu/location
Xinfeng's Projects
Caffe: a fast open framework for deep learning.
Caffe models in TensorFlow
Testing the performance of the cuBLAS
Automatically exported from code.google.com/p/cuda-convnet2
Codes for auto-tuning cudnn conv forward implementations
Reference workloads for modern deep learning methods.
FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/
Fast and memory-efficient exact attention
Homepage
GPGPU-Sim provides a detailed simulation model of contemporary NVIDIA GPUs running CUDA and/or OpenCL workloads. It includes support for features such as TensorCores and CUDA Dynamic Parallelism as well as a performance visualization tool, AerialVisoin, and an integrated energy model, GPUWattch.
a language for fast, portable data-parallel computation
LevelDB is a fast key-value storage library written at Google that provides an ordered mapping from string keys to string values.
Testing the performance of the MKL-DNN
Models and examples built with TensorFlow
Source code of MPU simulator and compiler for ASPLOS 2021 submission.
Homepage of the MPU project based on the Cayman theme.
Lightweight, Portable, Flexible Distributed/Mobile Deep Learning with Dynamic, Mutation-aware Dataflow Dep Scheduler; for Python, R, Julia, Scala, Go, Javascript and more
NiftyRec is a software toolbox for Tomographic image reconstruction. NiftyRec is written in C and computationally intensive functions have a GPU accelerated version based on NVidia CUDA. NiftyRec includes a Matlab Toolbox and a Python Package that access the low level routines, hiding the complexity of the GPU accelerated algorithms.
Tensors and Dynamic neural networks in Python with strong GPU acceleration
95.16% on CIFAR10 with PyTorch
Pytorch domain library for recommendation systems
Hackable and optimized Transformers building blocks, supporting a composable construction.