After previously revealing they would be making hardware products available familiar machine learning at the edge, Google has this week launched its new Google Coral range of AI platform components the form of development board, USB accelerator and camera. With a PCI-E accelerator and system-on-module (SOM) launching very soon.
The Google Coral development board looks similar to a Raspberry Pi and offers a single-board computer with a removable system-on-module (SOM) featuring the Edge TPU. The board has been designed to help o quickly prototype on-device ML products as well as scale from prototype to production using the removable system-on-module (SOM).
“The Edge TPU is a small ASIC designed by Google that provides high performance ML inferencing for low-power devices. For example, it can execute state-of-the-art mobile vision models such as MobileNet V2 at 100+ fps, in a power efficient manner.”
Specifications of the Edge TPU Module include a NXP i.MX 8M SOC (quad Cortex-A53, Cortex-M4F) CPU supported by Integrated GC7000 Lite Graphics with a ML accelerator in the form of a Google Edge TPU coprocessor. Other features include 1 GB LPDDR4 RAM, 8GB of eMMC Flash memory as well as connectivity via wireless 2×2 MIMO 802.11b/g/n/ac 2.4/5GHz and Bluetooth 4.1.
Features and benefits of the Google Coral development board include :
– No need to build models from the ground up. TensorFlow Lite models can be compiled to run on the Coral Dev Board.
– Considers your manufacturing needs. The SOM can be removed from the baseboard, ordered in bulk, and integrated into your hardware.
– SOC + ML + Connectivity all on the board running a derivative of Debian Linux we call Mendel, so you can run your favourite Linux tools with this board.
The Google Coral development Board is now available to purchase priced at $150 and supports Python with C++support coming soon. To learn more about all the new components launched by Google this week jump over to the official Coral website by following the link below.
Source: Google
