yonghenglh6 / Depthwiseconvolution
A personal depthwise convolution layer implementation on caffe by liuhao.(only GPU)
Stars: ✭ 512
Projects that are alternatives of or similar to Depthwiseconvolution
Deformable Convolution Pytorch
PyTorch implementation of Deformable Convolution
Stars: ✭ 410 (-19.92%)
Mutual labels: cuda
Pytorch Mobilenet
PyTorch MobileNet Implementation of "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications"
Stars: ✭ 443 (-13.48%)
Mutual labels: mobilenet
Segmentron
Support PointRend, Fast_SCNN, HRNet, Deeplabv3_plus(xception, resnet, mobilenet), ContextNet, FPENet, DABNet, EdaNet, ENet, Espnetv2, RefineNet, UNet, DANet, HRNet, DFANet, HardNet, LedNet, OCNet, EncNet, DuNet, CGNet, CCNet, BiSeNet, PSPNet, ICNet, FCN, deeplab)
Stars: ✭ 490 (-4.3%)
Mutual labels: mobilenet
Icpcuda
Super fast implementation of ICP in CUDA for compute capable devices 3.5 or higher
Stars: ✭ 416 (-18.75%)
Mutual labels: cuda
Caer
High-performance Vision library in Python. Scale your research, not boilerplate.
Stars: ✭ 452 (-11.72%)
Mutual labels: cuda
Convnet
A GPU implementation of Convolutional Neural Nets in C++
Stars: ✭ 506 (-1.17%)
Mutual labels: cuda
Mobilenetv2 Ssdlite
Caffe implementation of SSD and SSDLite detection on MobileNetv2, converted from tensorflow.
Stars: ✭ 435 (-15.04%)
Mutual labels: mobilenet
Xray Oxygen
🌀 Oxygen Engine 2.0. [Preview] Discord: https://discord.gg/P3aMf66
Stars: ✭ 481 (-6.05%)
Mutual labels: cuda
Tsdf Fusion
Fuse multiple depth frames into a TSDF voxel volume.
Stars: ✭ 426 (-16.8%)
Mutual labels: cuda
Bitcracker
BitCracker is the first open source password cracking tool for memory units encrypted with BitLocker
Stars: ✭ 463 (-9.57%)
Mutual labels: cuda
Lightseq
LightSeq: A High Performance Inference Library for Sequence Processing and Generation
Stars: ✭ 501 (-2.15%)
Mutual labels: cuda
Tsdf Fusion Python
Python code to fuse multiple RGB-D images into a TSDF voxel volume.
Stars: ✭ 464 (-9.37%)
Mutual labels: cuda
Depthwise Convolutional Layer
Introduction
This is a personal caffe implementation of mobile convolution layer. For details, please read the original paper:
How to build
- Merge the caffe folder in the repo with your own caffe. $ cp -r $REPO/caffe/* $YOURCAFFE/
- Then make. $ cd $YOURCAFFE && make
Usage
Replacing the type of mobile convolution layer with "DepthwiseConvolution" is all. Please refer to the example/Withdw_MN_train_128_1_train.prototxt, which is altered from
GPUPerformance on example net
| GPUPerformance | Origin[^nocudnn] | Mine |
|---|---|---|
| forward_batch1 | 41 ms | 8 ms |
| backward_batch1 | 51 ms | 11 ms |
| forward_batch16 | 532 ms | 36 ms |
| backward_batch16 | 695 ms | 96 ms |
[^nocudnn]: When turn on cudnn, the memory consuming of mobilenet would increase to unbelievable level. You may try.
Transfer normal net to mobilenet
I write a script [transfer2Mobilenet.py] to convert normal net to mobilenet format. You may try too.
usage: python ./transfer2Mobilenet.py sourceprototxt targetprototxt [--midbn nobn --weight_filler msra --activation ReLU] ["--origin_type" means the depthwise convolution layer's type will be "Convolution" instead of "DepthwiseConvolution"]
The "transferTypeToDepthwiseConvolution.py" will be used for changing the depthwise convolution layer's type from "Convolution" to "DepthwiseConvolution".
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].