- 👋 Hi, I’m Harshal Raut
- 👀 I’m interested in High Performance Computing, Machine Learning and CFD
- 📫 How to reach me hrshl212@gmail.com
-
Johns Hopkins Univeristy
- Baltimore, Maryland
Popular repositories Loading
-
TD3-libtorch
TD3-libtorch PublicTD3 reinforcement learning algorithm using libtorch in simple environment
Makefile 3
-
Preconditioned-Conjugate-Gradient-Method-in-CUDA
Preconditioned-Conjugate-Gradient-Method-in-CUDA PublicPreconditioned conjugate gradient method with ILU preconditioner implemented in CUDA
Cuda 2
-
Transformer-Attention-Mechanism-in-CUDA
Transformer-Attention-Mechanism-in-CUDA PublicCustom CUDA kernel for transformer's attention mechanism and integrating it with pytorch
Cuda 2
-
Matrix_transpose_using_GPU
Matrix_transpose_using_GPU PublicPerforming matrix transpose, first with each thread performing transpose of one element of matrix and later using the unrolling option as well.
Cuda 1
-
Optimized-matrix-multiplication-in-CUDA
Optimized-matrix-multiplication-in-CUDA PublicThe repository contains different ways to implement matrix-matrix multiplication in CUDA starting from basic implementation to using tensor cores in NVIDIA A100 GPUs
Cuda 1
-
Custom-CUDA-kernels-with-Neural-Network-Implementation
Custom-CUDA-kernels-with-Neural-Network-Implementation PublicThe repository contains custom CUDA kernels for linear layer, softmax and relu which are integrated with python to develop a Neural Network
Python 1
If the problem persists, check the GitHub status page or contact support.