quantisation 开源项目推荐（25 个）

strath-sdr /dsp_notebooks

一套以无线通信为主题的数字信号处理笔记本。

yoshakami /plt0

Wii/GC texture encoder with palette support (png, jpeg, gif, bmp, tiff, tpl, tex0, plt0, bti)

martinferianc /quantised-bayesian-nets

On the Effects of Quantisation on Model Uncertainty in Bayesian Neural Networks

sayannath /TensorFlow-Notebooks

My Tensorflow Notebook. In this notebooks I have implemented various kind of model optimisation techniques.

NvMayMay /nvfp4-lora-spark

LoRA fine-tune and serve NVFP4 models on one DGX Spark (GB10, 128 GB UMA): text backbones via generic-family onboarding, plus VLMs (vision tower, or LLM+tower jointly via --train-target both) validated end-to-end on Pixtral and Nemotron-Omni. Fused Triton dequant; runtime-LoRA and merge serving.

A-Sathvik /JPEG-compression

Applying SDLC for JPEG compression project

WillKirkmanM /jpeg

JPEG Encoding with Discrete Cosine Transform, Quantisation & Run-Length Compression

JollyGagging /Llama3-Runtime-Quantization

This Repository Stores an empirical study conducted on Meta-Llama-3-8B-Instruct Model , Running on a Nvidia T4 16Gb . Quantizing it to 4-bit NF4 to reclaim 54% VRAM and accelerate Llama-3-8B generation up to 8.2x on Edge Hardware . with Negligible Loss in Model Quality . Benchmarks Showcase the Results of the Research .

jounaidr /JRpeg

Image compression codec based on JPEG implemented in python

Sameeh07 /Meeting-audio-summarisation

Transcribe meeting audio and summarises.

ayushk12 /image-compression-using-graph-coloring

A research based project to optimize the graph coloring problem using the quantisation part of image compression and come up with a solution which is better than conventional algorithms for graph coloring.

drsamonline /art-kit

The art primitives Forge was missing: an indexed palette, a pixel canvas, draw ops, a delta undo stack, dithering, median-cut quantisation and sprite sheets.

kyaiooiayk /Cheap-ML-models

Optimising train, inference and throughput of expensive ML models

hanaml-lab /vsr-mobile

On-device video super-resolution, 270p to 1080p: a 1.9M-parameter recurrent student trained with DRIFT's terms and QAT, ONNX, TensorRT INT8 and Core ML exports, and a Rust playback pipeline.

evanpacini /dither

Simple dithering library in Rust, based on image-rs

aronmandrella /quant-studio

Source code: C++, SFML, MATLAB. The aim of the project was to implement various quantization algorithms in C ++ and to compare them in terms of the quality of the obtained results and the time needed to perform the calculations. On the way I've created an app with very basic GUI.

shahoismael /quantization-answer-fidelity

Do compressed classroom LLMs keep their answers, not just their scores? Fidelity, confident error, semantic entropy and measured inference energy across three model families at three quantisation levels.

Varun0157 /quantisation

Experiments in quantisation consisting of quantisation from scratch, bitsandbytes, and llama.cpp. [Assignment 4 of Advanced Natural Language Processing, IIIT-H Monsoon '24]

hanaml-lab /ondevice-video-sr

Reference implementation of DRIFT: an INT8 recurrent video super-resolution student whose quantisation drift is bounded by a contraction penalty, with long-rollout QAT.

aqilmarwan /attentionDM

Pytorch implementation of Self Attention Enhanced Post Training Quantisation for Diffusion Models