DLtrain platform development  took few years of innovation and coding.  DLtrain is designed to  make their silicon EDGE computing ready for AI work load. DLtrain is a perfect tool to handle issues in porting trained AI models in Edge computers with ease.

Silicon Vendor  team can take advantage of the above infrastructure and move their GPU silicon into the IoT edge market.  DLtrain licence with source code  ( non exclusive ) can be provided to customer  teams  such that they can port DLtrain on their CPU+GPU silicon,  Porting PyTorch, Tensor flow models on to Embedded device is one of the challenging problem and DLtrain is solving the same issue and provide option for Silicon vendor to move in to OEM ready solution.   For example TI has their own inference engine (TIDL), Qualcomm has their own  (SPNE) and ST Micro and many other silicon vendors. DLtrain provides C and C++ code along with a licence for the Customer  team to quickly move to market.

 

Open-CE

An Open Cognitive Environment

Open Cognitive Environment (Open-CE), based on IBM’s PowerAI to enable improved consumability of AI and deep learning frameworks

A2O POWER processor core, an out-of-order follow-up to the A2I core, and associated FPGA environment. 

“SKY130 is the industry’s first open source foundry process design kit,

A2O POWER processor core RTL and associated FPGA implementation (using ADM-PCIE-9V3 FPGA)

OpenPOWER Foundation Introduces IBM Hardware and Software Contributions at OpenPOWER Summit 2020