Have been always a happy programmer:D
Repositories
Mingxue-Xu/llm-profiler
A fine-grained LLM profiler that calculates the algorithmic operations (e.g. ADD, MULT), parameters, cache (buffers) and activations (intermediate output between the layers).
Mingxue-Xu/goldcrest
Full-Pipeline LLM Compression Framework for Low-Rank Factorization
Mingxue-Xu/omnipies
Research Argument Prototype.
Mingxue-Xu/economic-index
Mingxue-Xu/TransformerCompression
Mingxue-Xu/map-anything
MapAnything: Universal Feed-Forward Metric 3D Reconstruction
Mingxue-Xu/sparsegpt
Code for the ICML 2023 paper "SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot".
Mingxue-Xu/mingxue-xu.github.io
A beautiful, simple, clean, and responsive Jekyll theme for academics
Mingxue-Xu/ori
The code of the paper "Data Origin Inference in Machine Learning".
Mingxue-Xu/server-kit
Common functions for GPU/CPU central/edge servers.
Mingxue-Xu/PaddleSpeech
A Speech Toolkit based on PaddlePaddle.
Mingxue-Xu/PaddleHub
Awesome pre-trained models toolkit based on PaddlePaddle.(300+ models including Image, Text, Audio and Video with Easy Inference & Serving deployment)
Mingxue-Xu/PaddleOCR
Awesome multilingual OCR toolkits based on PaddlePaddle （practical ultra lightweight OCR system, support 80+ languages recognition, provide data annotation and synthesis tools, support training and deployment among server, mobile, embedded and IoT devices）