ResNet-50 - a misleading machine learning inference benchmark for megapixel images: Page 5 of 5

July 01, 2019 //By Geoff Tate, CEO of Flex Logix Technologies Inc.
ResNet-50 source: Mahmood et al.
Geoff Tate looks at the shortcomings of ResNet-50 as an inference benchmark in machine learning and considers the importance of image size, batch size and throughput for assessing inference performance.

All models that process megapixel images will use memory very differently than tiny models like ResNet-50’s 224x224.  The ratio of weights correspond to activation flips for large images. To really get a sense for how well an inference accelerator will perform for any CNN with megapixel images, a vendor needs to look at a benchmark that uses megapixel images. The clear industry trend is to larger models and larger images so YOLOv3 is more representative of the future of inference acceleration.  Using on-chip memory effectively will be critical for low cost/low power inference.

Geoff Tate is the CEO of Flex Logix Technologies Inc. (Mountain View, Calif.), a licensor of FPGA fabric and neural network acceleration cores. Prior to his current position Tate was a co-founder and CEO of Rambus Inc.

Related links and articles:

www.flex-logix.com

News articles:

Flex revamps NN fabric, announces edge AI processor

Eta adds spiking neural network support to MCU

Microsoft, Alexa, Bosch join Intel by investing in Syntiant

NovuMind benchmarks tensor processor


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.