For this week, I made the final touches to our detection model. I finalized training runs for our YOLOv5x model using the expanded dataset, and measured the benchmark accuracy as well as inference speed on the Raspberry Pi. I have also set up the cloud pipeline to integrate our model inference pipeline, in preparation for the final demo.
Overall, I am on schedule with our updated project timeline. The model has been finalized, and I am helping to complete the integration of the model with the cloud pipeline, the Raspberry Pi as well as our mobile application.
For next week, we will focus on completing the deliverables, finalizing the integration between our components in order to achieve a successful final demo.