LSUVinit
Initialization method
Implementation of a method to initialize neural network layers in a deep learning framework.
Reference caffe implementation of LSUV initialization
112 stars
10 watching
33 forks
Language: C++
last commit: almost 9 years agoLinked from 1 awesome list
caffeconvolutional-networksconvolutional-neural-networksdeep-learninginitializationlsuvlsuv-initialization
Related projects:
| Repository | Description | Stars |
|---|---|---|
| A C++ implementation of a deep learning framework designed for speed and modularity. | 59 | |
| An OpenCL implementation of Caffe, a mainstream DNN framework. | 518 | |
| A framework for deep learning using C++ and modularity to speed up development. | 6 | |
| An implementation of the Xception neural network in Caffe, a deep learning framework. | 45 | |
| This code provides an initialization method for convolutional neural networks based on data-dependent parameters. | 138 | |
| A collection of simple JavaScript implementations of deep learning concepts | 35 | |
| A C++ implementation of the deep learning framework used for rapid prototyping and building of image recognition models. | 220 | |
| An implementation of a deep learning framework with video features | 175 | |
| A C++ interface to the caffe deep learning framework, providing a convenient entry point for developers to build and deploy neural networks. | 15 | |
| Generates class visualizations from deep neural networks using Caffe | 358 | |
| An implementation of ShuffleNet V2 as a deep learning model in Caffe2 for computer vision tasks | 25 | |
| A fast and modular deep learning framework for computer vision tasks. | 169 | |
| Provides pre-trained neural network models based on DenseNet architecture, converted from Torch format to Caffe model for use in deep learning applications. | 357 | |
| Re-implementation of a neural network model with batch normalization and customized training parameters. | 131 | |
| An implementation of Deep Q-Network using Caffe to train and test reinforcement learning algorithms. | 212 |