Forum Discussion

RubenPadial's avatar
RubenPadial
Icon for Contributor rankContributor
2 years ago

Intel FPGA AI suite accuracy drop

Hello, I'm using Intel FPGA AI 2023.2 on ubuntu 20.04 host computer and trying to infer a custom CNN in a Intel Arria 10 SoC FPGA. The CNN was trained with TensorFlow and the accuracy is 98.89% acr...
Acceleration
AndrewRooney's avatar
AndrewRooney
Icon for New Contributor rankNew Contributor
2 years ago

One thing to check is your channel order. By default dla_benchmark will feed input data as RGB, but the dla_benchmark option `-bgr` will reverse the channels. This could be the culprit for a performance drop of this scale.

RubenPadial's avatar
RubenPadial
Icon for Contributor rankContributor
2 years ago

Hello @AndrewRooney,

Thank you for your help but that's not the problem. The same NN was compiled to be run only in the ARM CPU by only changing the

--fplugin flag to --fplugin "HETERO:CPU" and it worked. So the same NN result in a 74.64% acc in HETERO:FPGA,CPU mode and the wcuracy goes to 98% with HETERO:,CPU.

After performing some tests, we found the problem was related to the fully connected layers. The input shape of the FullyConnected layer, due to some design constrains, was [1, 1, 512] without a flatten layer. If a flatten layer is included before the FC layer, the inference turns to be right. So the problem was related with the FC layer and the HETERO:FPGA:CPU compilation.