xmu-xiaoma666
Student at Xiamen University
Xiamen City, Fujian, China
Summary
🤩
Rockstarxmu-xiaoma666 is a Ph.D. student at Xiamen University's MAC Lab and an intern at Alibaba’s MinD Lab, bringing eight years of hands-on experience in machine learning and computer vision research. They specialize in implementing and dissecting attention mechanisms, MLP architectures, and re-parameterization techniques in PyTorch, with a public repo that consolidates many state-of-the-art modules for visual tasks. Their work bridges research and reproducible engineering—translating recent papers into usable code like CoTAttention, RepMLP, and gMLP while providing practical demos. Based in Xiamen, China, they combine academic rigor with industry exposure at a leading AI lab, making them adept at rapid prototyping and experimental evaluation. An interesting detail: their contributions emphasize not just model implementations but also instructional clarity, aimed at helping others understand and experiment with complex attention patterns.
8 years of coding experience
Github Skills (7)
neural-network10
attention-mechanism10
pytorch10
machine-learning10
convolutional-neural-networks10
ps9
computer-vision9
Programming languages (2)
Github contributions (5)
xmu-xiaoma666/External-Attention-pytorch
May 2021 - Jan 2023
🍀 Pytorch implementation of various Attention Mechanisms, MLP, Re-parameter, Convolution, which is helpful to further understand papers.⭐⭐⭐
Role in this project:
ML Engineer
ML EngineerContributions:266 commits, 29 PRs, 231 pushes in 1 year 8 months
Contributions summary:Xmu-xiaoma666 contributed various PyTorch implementations of attention mechanisms, MLPs, and convolution techniques, focusing on understanding and implementing papers related to visual tasks. Their commits involved implementing modules like SimplifiedScaledDotProductAttention, SEAttention, SKAttention, CBAMBlock, BAMBlock, ECAAttention, gMLP, DANet, PSA, RepMLP, ResMLP, mlp_mixer, CoTAttention, and UFOAttention. They also uploaded and demonstrated the usage of implemented attention mechanisms in the main program.
pytorchmlpvisual-tasksdeep-learningattention-mechanism
xmu-xiaoma666/xmu-xiaoma666.github.io
Jun 2022 - Feb 2024
Contributions:111 pushes, 1 branch, 1 comment in 1 year 9 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.