SWS2023-Home-Work-Traffic

NUS SWS2023 Cloud computing with big data project

CV model

model

• use yolov5 to find the cars in the road
• use the VisDrone datasets and self-labeled datasets to train our model

performance

details are in file: CVresults

deployment

EC2

• upload our model to the EC2
• use flask to run our model on the server, and receive the request
• use boto3 to send the result to the AWS S3

EMR

• configure enviroment using bootstrap action script
• upload files to cluster from S3 bucket
• deploy pyspark program