mAP function for TORCHVISION OBJECT DETECTION FINETUNING TUTORIAL

1

Hi I follow the TORCHVISION OBJECT DETECTION FINETUNING TUTORIAL:https://pytorch.org/tutorials/intermediate/torchvision_tutorial.html to finish prediction. I want to compute the box mAP of my test data(about 100). How to compute the mAP for test data? Is there a function to compute it?

2

The tutorial shows the IoU metric at the end using the CocoEvaluator. Could you just reuse the evaluate method or the linked class?

1 Like
3

thanks for your answer. I want to predict my a new test data, and get prediction box data, and mAP result of prediction.