Qinkai's Projects
AITemplate is a Python framework which renders neural network into high performance CUDA/HIP C++ code. Specialized for FP16 TensorCore (NVIDIA GPU) and MatrixCore (AMD GPU) inference.
This is the repository of the project of AML2021: Graph injection attack & defense
A repository of AsciiDoc source files for testing AsciiDoc rendering on GitHub.
😎 A curated list of awesome GitHub Profile READMEs 📝
Repository for benchmarking graph neural networks
LM Studio JSON configuration file format and a collection of example config files.
Resources for Data Centric AI
learn in datamining
The official implementation of the paper "DC Coefficients Recovery from AC Coefficients in the JPEG Compression Scenario"
Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising (TIP, 2017)
The official implementation of the paper "Deep Residual Learning based Enhanced JPEG Compression in the Internet of Things"
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
Transformer related optimization, including BERT, GPT
GLM-130B: An Open Bilingual Pre-Trained Model
Graph Robustness Benchmark
Code for the paper "Evaluating Large Language Models Trained on Code"
🙌
KDD CUP 2020 ML Track 2 "Adversarial Attacks and Defense on Academic Graph": 1st Place Solution
Tips for releasing research code in Machine Learning (with official NeurIPS 2020 recommendations)
Concise, consistent, and legible badges in SVG and raster format
上海交通大学 XeLaTeX 学位论文及课程论文模板 Shanghai Jiao Tong University XeLaTeX Thesis Template
The official implementation of the paper "Towards Secure and Efficient Deep Learning Inference in Dependable IoT Systems"