https://github.com/layumi/batch-normal-for-caffe
Extend batch normalization layer for caffe
https://github.com/layumi/batch-normal-for-caffe
Last synced: 6 months ago
JSON representation
Extend batch normalization layer for caffe
- Host: GitHub
- URL: https://github.com/layumi/batch-normal-for-caffe
- Owner: layumi
- Created: 2015-08-10T06:32:37.000Z (about 10 years ago)
- Default Branch: master
- Last Pushed: 2015-08-10T08:10:53.000Z (about 10 years ago)
- Last Synced: 2025-02-11T17:59:15.030Z (8 months ago)
- Language: C++
- Homepage:
- Size: 412 KB
- Stars: 3
- Watchers: 3
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Batch-Normal-For-Caffe
There are some codes for new vision Caffe to extend Batch Normalization Layer.
To add this layer, you only have to modify 'common_layers.hpp', 'upgrade_proto.cpp' and 'caffe.proto'.
And add 'bn_layer.cpp', 'bn_layer.cu'. It isn't necessary to modify 'layer_factory.cpp' any more.
The original code is https://github.com/ChenglongChen/batch_normalization, which I modified from.
And all their rights belong to the orignial author.
In the tool dir, I include some codes to convert the model generalized from MatConvnet to Caffemodel by using matcaffe.