perform training on custom datasets in format of ava

[PedestrianZQZ]

I have looked through the architecture of dataset.md, and wanted to prepare my own datasets in ava style to perform training on pretrained model. I really wounder how did the ava_train_predicted_boxes.csv and ava_val_predicted_boxes.csv produced? The coordinates in the two file are different from the ava official annotation files such as ava_train_v2.1.csv. BTW, if I use my own datasets, is the [official AVA annotation files] needed? Thanks in advance

ava
|_ frames
|  |_ [video name 0]
|  |  |_ [video name 0]_000001.jpg
|  |  |_ [video name 0]_000002.jpg
|  |  |_ ...
|  |_ [video name 1]
|     |_ [video name 1]_000001.jpg
|     |_ [video name 1]_000002.jpg
|     |_ ...
|_ frame_lists
|  |_ train.csv
|  |_ val.csv
|_ annotations
   |_ [official AVA annotation files]
   |_ ava_train_predicted_boxes.csv
   |_ ava_val_predicted_boxes.csv