Open xLuge opened 3 years ago
Did you follow the instructions at https://hangzhang.org/PyTorch-Encoding/model_zoo/segmentation.html#pascal-voc-dataset
Make sure you first train on MS-COCO, then finetune on pascal-aug, and then finetune again on pascal-vov
hi, I submitted the results of the val set and the test set to the official website for testing, but the two results differ by four points. How can I reduce this gap? Also, where can I go to test the test set for the PContexts data set. I'm not sure the result in some papers are test set results or validation set results
hi, I submitted the results of the val set and the test set to the official website for testing, but the two results differ by four points. How can I reduce this gap? Also, where can I go to test the test set for the PContexts data set. I'm not sure the result in some papers are test set results or validation set results
hi,may I ask about the issue you encountered on the Pascal VOC dataset's test set at the time? [Error while evaluating challenge 6 .Results were not generated properly.]The evaluation log is as follow.
hi, I submitted the results of the val set and the test set to the official website for testing, but the two results differ by four points. How can I reduce this gap? Also, where can I go to test the test set for the PContexts data set. I'm not sure the result in some papers are test set results or validation set results