nninit
NN init
Provides parameter initialisation schemes for neural network modules in Torch7
Weight initialisation schemes for Torch7 neural network modules
100 stars
12 watching
12 forks
Language: Lua
last commit: over 9 years agoLinked from 1 awesome list
deep-learning
carpedm20/awesome-torchRelated projects:
| Repository | Description | Stars |
|---|---|---|
 alykhantejani/nninit | Provides weight initialization schemes for PyTorch neural networks | 70 |
 torch/nn | An open-source neural network package providing a modular framework for building and training neural networks. | 1,346 |
 kaishengtai/torch-ntm | An implementation of a Neural Turing Machine model using Torch | 279 |
| torch/nngraph | Graphical computation library for building neural network architectures | 299 |
 gpleiss/nnlr | Adds layer-wise learning rate schemes to a deep neural network implementation | 47 |
 majianjia/nnom | A high-level inference Neural Network library specifically designed for microcontrollers. | 965 |
 jhkim89/pyramidnet | A Torch implementation of a novel neural network architecture designed to improve the generalization ability of deep image classification models. | 129 |
 clementfarabet/lua---nnx | An extension to Torch7's neural network package with experimental modules and optimizations. | 97 |
| kaixhin/grokking-pytorch | An introduction to using PyTorch for deep learning tasks | 1,194 |
 nagadomi/kaggle-cifar10-torch7 | A deep neural network architecture optimized for image classification on the CIFAR-10 dataset | 246 |
 soumith/inception.torch | A Torch implementation of a pre-trained neural network architecture used in image recognition tasks | 66 |
 kaiminghe/resnet-1k-layers | Represents a neural network architecture with 1K layers, designed for image recognition tasks. | 909 |
 chncwang/insnet | A C++ library for building instance-dependent neural networks with padding-free dynamic batching. | 66 |
| soumith/imagenet-multigpu.torch | A toolkit for training neural networks on the ImageNet dataset using multiple GPUs. | 402 |
 nngen/nngen | Generates hardware-specific accelerator designs for neural networks | 340 |