MyCaffe 0.10.0.122-beta1

A complete C# re-write of Berkeley's open source Convolutional Architecture for Fast Feature Encoding (CAFFE) for Windows C# Developers, now with Policy Gradient reinforcement learning support!

This is a prerelease version of MyCaffe.
There is a newer prerelease version of this package available.
See the version list below for details.
Install-Package MyCaffe -Version 0.10.0.122-beta1
dotnet add package MyCaffe --version 0.10.0.122-beta1
<PackageReference Include="MyCaffe" Version="0.10.0.122-beta1" />
For projects that support PackageReference, copy this XML node into the project file to reference the package.
paket add MyCaffe --version 0.10.0.122-beta1
The NuGet Team does not provide support for this client. Please contact its maintainers for support.

MyCaffe AI Platform and Test Application (CUDA 10.0.130, cuDNN 7.4.1) Release with LSTM Recurrent Learning Support and Char-RNN

CUDA 10.0.130, cuDNN 7.4.1, nvapi 410, Native Caffe up to 10/24/2018, Windows 10-1803, Driver 416.94

IMPORTANT NOTE: When using TCC mode, we recommend that ALL headless GPU’s are placed in TCC mode for we have experienced stability issues when using a mix of TCC and WDM modes with headless GPU’s.

MyCaffe now supports Recurrent Learning using CUDA 10 and cuDNN 7.4.1 to implement the Char-RNN described by A. Karpathy [1] and originally implemented on Caffe [2] by adepierre [3].

This release of the MyCaffe AI Platform and Test Applications has the following new additions:

  • CUDA 10.0.130/cuDNN 7.4.1 now supported (with driver 416.94).
  • RTX 2080 ti supported.
  • Added support for custom gyms.
  • Added new DataGeneral gym.
  • Added new streaming database support.
  • Added new data preprocessing support with CUDA plugins.
  • Added new CUDA level extension support.
  • Added support for prototxt comments.
  • Added new TrainerRNN for Recurrent Learning training.
  • Added new Char-RNN model that learns Shakespeare.
  • Added new WAV-RNN model for learning audio.

The following bug fixes are in this release:

  • Fixed bug in Input layer not loading shape from prototxt.
  • Fixed bug related to mixed types in saved weights.
  • Improved blob sharing in Bias, Embed, LSTMSimple, PreLU and Scale layers.
  • Improved AutoTest sub-item status reporting.

Easily create the CIFAR-10 and MNIST datasets using the MyCaffe Test Application which you can download from the MyCaffe GitHub site.

Create and train the Policy Gradient Reinforcement Learning, Auto-Encoder, DANN and ResNet models by following step-by-step instructions in the SignalPop Tutorials. And, to see other cool examples that show what MyCaffe can do, see the SignalPop Examples.

If you would like to visually design, develop, test and debug your models, see the SignalPop AI Designer specifically designed to enhance your MyCaffe deep learning.

Also, check out the SignalPop Universal Miner that not only gives you detailed information on each of your GPU's (such as temperature, fan speed, overclock, and usage), it allows you to easily start mining Ethereum. When not training AI, put those GPU's to use making some Ether - never let a good GPU go to waste!

Happy ‘deep’ learning!

[1] A. Karpathy, The Unreasonable Effectiveness of Recurrent Neural Networks, Andrej Karpathy blog, May 21, 2015.

[2] Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama and T. Darrell, Caffe: Convolutional Architecture for Fast Feature Embedding, June 20, 2014.

[3] adepierre, adepierre/caffe-char-rnn Github, Github.com, January 25, 2017.

MyCaffe AI Platform and Test Application (CUDA 10.0.130, cuDNN 7.4.1) Release with LSTM Recurrent Learning Support and Char-RNN

CUDA 10.0.130, cuDNN 7.4.1, nvapi 410, Native Caffe up to 10/24/2018, Windows 10-1803, Driver 416.94

IMPORTANT NOTE: When using TCC mode, we recommend that ALL headless GPU’s are placed in TCC mode for we have experienced stability issues when using a mix of TCC and WDM modes with headless GPU’s.

MyCaffe now supports Recurrent Learning using CUDA 10 and cuDNN 7.4.1 to implement the Char-RNN described by A. Karpathy [1] and originally implemented on Caffe [2] by adepierre [3].

This release of the MyCaffe AI Platform and Test Applications has the following new additions:

  • CUDA 10.0.130/cuDNN 7.4.1 now supported (with driver 416.94).
  • RTX 2080 ti supported.
  • Added support for custom gyms.
  • Added new DataGeneral gym.
  • Added new streaming database support.
  • Added new data preprocessing support with CUDA plugins.
  • Added new CUDA level extension support.
  • Added support for prototxt comments.
  • Added new TrainerRNN for Recurrent Learning training.
  • Added new Char-RNN model that learns Shakespeare.
  • Added new WAV-RNN model for learning audio.

The following bug fixes are in this release:

  • Fixed bug in Input layer not loading shape from prototxt.
  • Fixed bug related to mixed types in saved weights.
  • Improved blob sharing in Bias, Embed, LSTMSimple, PreLU and Scale layers.
  • Improved AutoTest sub-item status reporting.

Easily create the CIFAR-10 and MNIST datasets using the MyCaffe Test Application which you can download from the MyCaffe GitHub site.

Create and train the Policy Gradient Reinforcement Learning, Auto-Encoder, DANN and ResNet models by following step-by-step instructions in the SignalPop Tutorials. And, to see other cool examples that show what MyCaffe can do, see the SignalPop Examples.

If you would like to visually design, develop, test and debug your models, see the SignalPop AI Designer specifically designed to enhance your MyCaffe deep learning.

Also, check out the SignalPop Universal Miner that not only gives you detailed information on each of your GPU's (such as temperature, fan speed, overclock, and usage), it allows you to easily start mining Ethereum. When not training AI, put those GPU's to use making some Ether - never let a good GPU go to waste!

Happy ‘deep’ learning!

[1] A. Karpathy, The Unreasonable Effectiveness of Recurrent Neural Networks, Andrej Karpathy blog, May 21, 2015.

[2] Y. Jia, E. Shelhamer, J. Donahue, S. Karayev, J. Long, R. Girshick, S. Guadarrama and T. Darrell, Caffe: Convolutional Architecture for Fast Feature Embedding, June 20, 2014.

[3] adepierre, adepierre/caffe-char-rnn Github, Github.com, January 25, 2017.

Release Notes

MyCaffe AI Platform

This package is not used by any popular GitHub repositories.

Version History

Version Downloads Last updated
0.10.1.169-beta1 62 7/8/2019
0.10.1.145-beta1 82 5/31/2019
0.10.1.48-beta1 94 4/18/2019
0.10.1.21-beta1 89 3/5/2019
0.10.0.190-beta1 166 1/15/2019
0.10.0.140-beta1 110 11/29/2018
0.10.0.122-beta1 134 11/15/2018
0.10.0.75-beta1 130 10/7/2018
Show less