Open lsj1111 opened 4 months ago
I would like to ask whether the results on DOTA in your paper were divided on the test set, and how you divided the data set. Was it divided according to the original DOTA data?
I see that your data set configuration seems to use the validation set as the test set. Is the final result measured in this way?
thank you for your reply.
use /tools/data/dota/split/img_split.py to split dota dataset
python tools/data/dota/split/img_split.py --base-json tools/data/dota/split/split_configs/ms_trainval.json
# Total images number: 138883
python tools/data/dota/split/img_split.py --base-json tools/data/dota/split/split_configs/ms_test.json
# Total images number: 71888
As for the base config file of dataset with ms split, see this.
I would like to ask whether the results on DOTA in your paper were divided on the test set, and how you divided the data set. Was it divided according to the original DOTA data?
I see that your data set configuration seems to use the validation set as the test set. Is the final result measured in this way?
thank you for your reply.