MyCaffe 0.10.2.38-beta1

MyCaffe AI Platform and Test Application (CUDA 10.2.89, cuDNN 7.6.5) with Siamese Net and One-Shot Learning CUDA 10.2.89, cuDNN 7.6.5, nvapi 435, Native Caffe up to 10/24/2018, Windows 10-1903, Driver 441.41 and 441.28

MyCaffe[1] (a complete C# re-write of CAFFE[2]) now supports the Siamese Net[3][4] with Contrastive Loss and runs on the newly released CUDA 10.2.89, CuDNN 7.6.5 for easy One-Shot learning!

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 441.28 or above). • Windows 1903, OS Build 18363.476 now supported. • Added new AccuracyEncoding Layer for Siamese Nets. • Added new Decode Layer for Siamese Nets. • Added new Siamese Net support to Data Layer with multiple images per channel. • Added active GPU display to automated testing. • Added testing priorities to automated testing. • Added improved label balancing to In-Memory Database. • Added ability to activate/deactivate Dropout Layer for staged training. • Improved automated testing error information. • Improved pinned memory usage. • Optimized cuDnn handle usage.

The following bug fixes are in this release: • Fixed bugs in InferBlobShape which now takes into account SSD resize parameter. • Fixed bugs in DataTransformer which now properly resizes images based on resize parameter. • Fixed bugs in DataTransformer.CropImage. • Fixed bugs in SSD parameters which are now properly persisted to prototxt. • Fixed bugs in ImageTransforms.ApplyDistort. • Fixed bugs in AnnotatedDataParameter.Copy and Load Batch. • Fixed bugs in VOC0712 Dataset Loader - annotation data now created at correct scale. • Fixed bugs in MultiBoxLoss Layer forward and backward passes. • Fixed bugs in GetDeviceID automated test. • Fixed bugs in Weight sharing now only occur on original owner net. • Fixed bugs in DataLayer synchronization which caused issues on slower machines. • Fixed bugs in numerous automated tests by optimizing memory use.

Easily 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

[3] Siamese Network Training with Caffe by Berkeley Artificial Intelligence (BAIR)

[4] Siamese Neural Network for One-shot Image Recognition by G. Koch, R. Zemel and R. Salakhutdinov, ICML 2015 Deep Learning Workshop, 2015.