GoGoDuck912/pytorch-vector-quantization
A Pytorch Implementations for Various Vector Quantization Methods
This library helps machine learning engineers and researchers implement various vector quantization methods within PyTorch. It takes raw feature vectors, often from images or audio, and transforms them into a more compact, discrete representation. The output includes quantized vectors, a commitment loss value, and indices pointing to the chosen codebook entries, which are crucial for tasks like high-quality generative modeling.
No commits in the last 6 months.
Use this if you are a machine learning practitioner working with PyTorch and need to apply vector quantization techniques for tasks like efficient data compression or generative model training.
Not ideal if you are looking for a high-level, out-of-the-box solution for data compression or generation without needing to delve into the underlying PyTorch implementation.
Stars
36
Forks
7
Language
Python
License
MIT
Category
Last pushed
Sep 14, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/GoGoDuck912/pytorch-vector-quantization"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
open-mmlab/mmengine
OpenMMLab Foundational Library for Training Deep Learning Models
Xilinx/brevitas
Brevitas: neural network quantization in PyTorch
google/qkeras
QKeras: a quantization deep learning library for Tensorflow Keras
fastmachinelearning/qonnx
QONNX: Arbitrary-Precision Quantized Neural Networks in ONNX
tensorflow/model-optimization
A toolkit to optimize ML models for deployment for Keras and TensorFlow, including quantization...