https://github.com/FranckZibi/SharpNet

Open-source Deep Learning library in C# with CUDA and BLAS support
https://github.com/FranckZibi/SharpNet

Last synced: 2 months ago
JSON representation

Open-source Deep Learning library in C# with CUDA and BLAS support

• Host: GitHub
• URL: https://github.com/FranckZibi/SharpNet
• Owner: FranckZibi
• License: gpl-3.0
• Created: 2019-02-28T21:57:38.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2024-08-12T07:38:38.000Z (5 months ago)
• Last Synced: 2024-08-12T09:03:26.857Z (5 months ago)
• Language: C#
• Homepage:
• Size: 4.8 MB
• Stars: 12
• Watchers: 8
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-dotnet-machine-learning - SharpNet - source Deep Learning library in C# with CUDA and BLAS support. (Uncategorized / Uncategorized)

README

        
SharpNet is an Open-source Deep Learning library written in C# 10.0.

It supports:

 - Residual Networks [v1](https://arxiv.org/pdf/1512.03385.pdf), [v2](https://arxiv.org/pdf/1603.05027.pdf), [WideResNet](https://arxiv.org/pdf/1605.07146.pdf) and [EfficientNet](https://arxiv.org/pdf/1905.11946.pdf)

 - [DenseNet](https://arxiv.org/pdf/1608.06993.pdf)

 - BatchNorm / Conv1D / Conv2D / Dense / Dropout / Embedding / GRU / LSTM / Pooling / RNN / Shortcut / SimpleRNN layers

 - Elu / Relu / Leaky Relu / Sigmoid / Softmax / Swish / Tanh activations

 - SGD & Adam optimizers

 - Image Data Augmentation (with Cutout/CutMix/Mixup)

 - Ensemble Learning

 

It can be run both on GPU (using NVIDIA cuDNN) and on the CPU (using MKL Blas).

It is targeted to make a good use of the GPU (even if it is not currently as fast as MxNet) :

 - on ResNet18 v1, it is between 1.5x (batch size = 128) and 3x time (batch size = 32) faster then TensorFlow 1.x

It requires:

- [Visual Studio 2022](https://visualstudio.microsoft.com/downloads/)

- [DotNetCore 6.0](https://dotnet.microsoft.com/download/dotnet-core/6.0)

- [CUDA Toolkit 10.1,  10.2 or 11.0](https://developer.nvidia.com/cuda-downloads)

- [CuDNN 8.0](https://developer.nvidia.com/rdp/cudnn-download)

- [Intel MKL](https://software.intel.com/en-us/mkl)