ducha-aiki/LSUVinit
Reference caffe implementation of LSUV initialization
This project provides an initialization technique for deep learning models, specifically within the Caffe framework, that helps neural networks learn more effectively from data. It takes your existing Caffe model definition as input and helps set up the initial weights of the network layers. This is for machine learning engineers and researchers who are building and training deep neural networks.
114 stars. No commits in the last 6 months.
Use this if you are working with Caffe and need to improve the initial weight setup of your neural networks to achieve faster or more stable training.
Not ideal if you are not using the Caffe deep learning framework, as this implementation is specifically tied to Caffe (though re-implementations exist for other frameworks).
Stars
114
Forks
33
Language
C++
License
—
Category
Last pushed
Oct 31, 2017
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/ducha-aiki/LSUVinit"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
apache/tvm
Open Machine Learning Compiler Framework
uxlfoundation/oneDNN
oneAPI Deep Neural Network Library (oneDNN)
Tencent/ncnn
ncnn is a high-performance neural network inference framework optimized for the mobile platform
OpenMined/TenSEAL
A library for doing homomorphic encryption operations on tensors
iree-org/iree-turbine
IREE's PyTorch Frontend, based on Torch Dynamo.