MyCaffe 0.10.2.309-beta1

MyCaffe AI Platform and Test Application (CUDA 10.2.89, cuDNN 7.6.5) with ONNX AI Model Support (onnx.ai).

CUDA 10.2.89, cuDNN 7.6.5, nvapi 440, Native Caffe up to 10/24/2018, Windows 10-1909, Driver 442.18 and 446.14

MyCaffe[1] (a complete C# re-write of CAFFE[2]) now supports ONNX AI Models! Easily import ONNX models into MyCaffe and export MyCaffe models to ONNX using the new MyCaffeConversionControl.

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.

REQUIRED SOFTWARE: 1.) Install NVIDIA CUDA 10.2.89 which you can download from https://developer.nvidia.com/cuda-downloads 2.) Install NVIDIA cuDNN 7.6.5 which you can download from https://developer.nvidia.com/cudnn 3.) Download and install Microsoft SQL Express 2016 (or later).

This release of the MyCaffe AI Platform and Test Applications has the following new additions: • CUDA 10.2.89/cuDNN 7.6.5 supported (with driver 442.18 or above). • Windows 1909, OS Build 18363.836 now supported. • Added new MyCaffeConversionControl to easily load and save *.onnx model files. • Added new ImageData layer for easy data input without SQL. • Upgraded to EntityFramework 6.4.4 • Upgraded to Google.ProtoBuf 3.12.1 • Improved low-level cuddnn handle sharing to now support very large networks such as ResNet152. • Added support for model building of ResNet152, ResNet101 and Siamese versions of each. • Added DISABLED snapshot update method to disable OnSnapshot event when not forced. • Added low-level protections to CudaDnnDll. • Expanded low-level host buffer support in CudaDnnDll. • Optimized low-level return values. • Improved samples with specific SQL checks when SQL is required. • Improved MyCaffe Test Application to run the GYM host even when SQL does not exist. • Added MNIST file export support to MyCaffe Test Application. • Added cuDnn Tensor Core support to Convolution, RNN and LSTM layers. • Added cuDnn Dilation support to Convolution layer. • Added MultiBox support to MyCaffeControl.TestAll. • Optimized memory use in the DataLayer.

The following bug fixes are in this release: • Fixed bugs in LoadLite where RUN net is now created. • Fixed bugs in Clone after using LoadLite to open a project. • Fixed bugs in MyCaffeControl when RUN net not loaded. • Fixed bugs in VOCDataLoader where images with height greater than width are loaded correctly now. • Fixed bugs in VOCDataLoader where labels are now saved to the database.

Easily import/export ONNX AI Models, run Siamese Nets[3][4], Neural Style, train Deep Q-Learning or Policy Gradient[1] models to beat Pong or Cart-Pole, or 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 Siamese Net[3][4], Deep Q-Learning with NoisyNet and Experienced Replay, Policy Gradient[1], Neural Style Transfer, 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 keeps your GPU's cool as you train, but also gives you detailed information on each of your GPU's (such as temperature, fan speed, overclock, and usage), and 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] MyCaffe: A Complete C# Re-Write of Caffe with Reinforcement Learning by D. Brown, 2018.

[2] Caffe: Convolutional Architecture for Fast Feature Embedding by Yangqing Jia, Evan Shelhamer, Jeff Donahue, Sergey Karayev, Jonathan Long, Ross Girshick, Sergio Guadarrama, Trevor Darrell, 2014, arXiv:1408.5093