nickjeliopoulos
Developing methods to make transformer neural networks more efficient for deployment on edge devices.
Purdue UniversityWest Lafayette, Indiana
Pinned Repositories
embeddedvision
An introductory graduate course on "Computer Vision for Embedded Systems"
CrossSectionalImageToSTL
Program that converts 2D bitmaps representing cross sections of a 3D model into vertices. A modeling program such as Blender can then use boolean operations to synthesize the model.
cutlass
CUDA Templates for Linear Algebra Subroutines
HardwareAwareToMe
A method to increase the speed and lower the memory footprint of existing vision transformers.
nickjeliopoulos
nickjeliopoulos.github.io
PyTorchCustomCUDAKernelPlayground
Simple implementation of C++/CUDA code that is called from Python (using PyTorch)
SViTE-Configurable-Patch-Removal
[NeurIPS'21] "Chasing Sparsity in Vision Transformers: An End-to-End Exploration" by Tianlong Chen, Yu Cheng, Zhe Gan, Lu Yuan, Lei Zhang, Zhangyang Wang
cutlass
CUDA Templates for Linear Algebra Subroutines
nutils
nickjeliopoulos's Repositories
nickjeliopoulos/CrossSectionalImageToSTL
Program that converts 2D bitmaps representing cross sections of a 3D model into vertices. A modeling program such as Blender can then use boolean operations to synthesize the model.
nickjeliopoulos/cutlass
CUDA Templates for Linear Algebra Subroutines
nickjeliopoulos/HardwareAwareToMe
A method to increase the speed and lower the memory footprint of existing vision transformers.
nickjeliopoulos/nickjeliopoulos
nickjeliopoulos/nickjeliopoulos.github.io
nickjeliopoulos/PyTorchCustomCUDAKernelPlayground
Simple implementation of C++/CUDA code that is called from Python (using PyTorch)
nickjeliopoulos/SViTE-Configurable-Patch-Removal
[NeurIPS'21] "Chasing Sparsity in Vision Transformers: An End-to-End Exploration" by Tianlong Chen, Yu Cheng, Zhe Gan, Lu Yuan, Lei Zhang, Zhangyang Wang