Skip to content
View kalfazed's full-sized avatar
📚
Working from everywhere
📚
Working from everywhere

Highlights

  • Pro

Block or report kalfazed

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
kalfazed/README.md


🎉 Welcome to my GitHub profile!

Never leave that till tomorrow which you can do today.

-- Benjamin Franklin

In 2024, I focused on contributing to the development of several software frameworks, including:

  • Model training framework for autonomous driving
  • Model deployment framework for edge computing
  • MLOps framework
  • Optimization toolkit

The Model Training Framework for Autonomous Driving I developed supports a variety of tasks, such as 2D/3D object detection, depth estimation, instance and semantic segmentation, and optical flow estimation. This framework is designed to facilitate efficient model training for autonomous driving applications, covering critical computer vision tasks.

On the deployment side, I worked on the Model Deployment Framework for Edge Computing, optimizing models to meet the constraints of edge devices. To achieve this, I applied techniques like quantization, pruning, and re-parameterization, all aimed at reducing model size and improving inference performance. These optimizations ensure that the models can be efficiently deployed on resource-constrained devices, making them suitable for real-time edge applications.

Additionally, I developed an Optimization Toolkit, which offers a set of tools and libraries designed to enhance program performance. The toolkit includes customized PyTorch operations, a lightweight visualizer, and a high-performance parser, and so on. All aimed at improving the efficiency of both model training and deployment workflows.

For more details about my work and projects, please visit my profile.

📘 My favorite open-source projects

Pinned Loading

  1. tensorrt_starter tensorrt_starter Public

    This repository give a guidline to learn CUDA and TensorRT from the beginning.

    C++ 165 44

  2. multi-thread-programming multi-thread-programming Public

    This is a repository to practice multi-thread programming in C++

    C++ 18 7

  3. vit-pytorch vit-pytorch Public

    Forked from lucidrains/vit-pytorch

    Implementation of Vision Transformer, a simple way to achieve SOTA in vision classification with only a single transformer encoder, in Pytorch

    Python

  4. my_dot_files my_dot_files Public

    There are some useful dotfiles that I used for everyday :)

    Shell 5 2