Synthetic neural network training fits on Gigabyte servers
Mindtech is on a path to introduce integrated AI processors for visual processing in a variety of target markets but initially it is offering its Chameleon simulator to create synthetic vision datasets for training neural networks.
The Chameleon platform creates and “films” virtual worlds to produce synthetic data for neural network training and Mindtech said the platform is a good fit with Gigabyte’s servers.
The ability to scale from one to many Gigabyte servers has allowed Mindtech to create so-called “ready-to-train” data packs, where images and annotation is supplied enabling customers to create AI applications for virtual smart cities, smart machines and smart vision. This can be for retail, premise management, robotics and transportation and once developed the AI routines can be redeployed in the real world.
“Creating and processing data for training AI takes significant compute resource, and GIGABYTE’s servers are an excellent platform upon which to build a solution for synthetic data creation,” said Mindtech’s vice president of product management Chris Longstaff, in a statement.
Mindtech’s Chameleon platform and Data Packs are available for immediate licensing.
Related links and articles:
News articles:
AI processor startup emerges from Imagination
AI processor market to exceed $20 billion in 2019
Intel scales up self-learning neuromorphic computing
Cerebras Wafer Scale Engine: An Introduction
SambaNova heads to ‘unicorn’ status, raises $150 million
Israeli AI hopeful raises $27 million
Startup proposes special SRAM for AI
Gyrfalcon launches second AI accelerator
NovuMind benchmarks tensor processor
Flex Logix tips move into NN processing
Indo-US startup preps agent-based AI processor
Alibaba forms chip subsidiary Pingtouge
Kneron launches improved line of NPUs
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
