MyCaffe 0.10.0.140-beta1

CUDA 10.0.130, cuDNN 7.4.1, nvapi 410, Native Caffe up to 10/24/2018, Windows 10-1803, Driver 417.01

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 the CUDA 10/cuDNN 7.4.1 LSTM implementation (which is 5x times faster than the CAFFE [2] version!) to implement the Char-RNN described by A. Karpathy [1] and originally implemented in CAFFE 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 supported (with driver 417.01). • Added cuDNN LSTM engine to LSTM Layer. • Added new Parameter Layer. • Added new GramLayer. • Added new TVLossLayer. • Added new LBFGSSolver.

The following bug fixes are in this release: • Fixed lock-up bug in Automated Testing Application.

To read more about cuDNN LSTM in MyCaffe, see the SignalPop Blog.

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 Recurrent Learning, 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), but allows you to easily mine 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.