This week, I worked with Sid to achieve and verify our design requirements for classification accuracy and latency using the completed PCB.<\/p>\n
I improved the convolutional neural network training by adding data augmentation to the training set. Since the camera occasionally shifts between the days, we added data augmentations that choose randomly-shifted crops of the input image to model those physical movements. I applied this data augmentation only to the training set. <\/p>\n
Sid and I captured and labelled 2724 new captures with the LED’s from the PCB. Combined with the new data augmentation and a learning rate scheduler, we achieved the best validation accuracy of 99.1%. This model achieved 98.1% accuracy on the test set<\/strong>. This is the best validation accuracy we have achieved, so we are using it for the MVP. <\/p>\n We also measured the classification latency. The design requirements specify 2 second latency from capture to web update. We dealt 52 cards over a 52 second period and achieved the following latency statistics for each card: I have now completed the imaging and machine learning system. We have some final testing to carry out, but so far, we have achieved every design requirement that we have tested.<\/p>\n","protected":false},"excerpt":{"rendered":" This week, I worked with Sid to achieve and verify our design requirements for classification accuracy and latency using the completed PCB. I improved the convolutional neural network training by adding data augmentation to the training set. Since the camera occasionally shifts between the days, we added data augmentations that choose randomly-shifted crops of the … <\/p>\n
\nmin: 0.162s
\nmean: 0.184s
\nmedian: 0.177s
\nmax: 0.224s
\nThe maximum latency of 0.224s is far less than our design requirement of 2 seconds.<\/strong><\/p>\n