Recurrence

I trained COCO+LVIS SegFormerB0-S1 using the trainval_scripts/train_focalclickB0_S1_cclvs.sh with nothing changed.
And the val result is as follows:

compared with the model published：

I notice that the BS is 32 in paper, and it's 64 in the bash script. Is it the reason for the accuracy gap？

Also, there are two ways of data augmentation in the code and the paper says "During training, we only use flip and random resize with the scale from 0.75 to 1.4 as data augmentation." which one is used in the model published?

Start with trainval_scripts/train_focalclickB0_S1_cclvs.sh and compare its batch size with the paper's BS 32. Then inspect the two data-augmentation paths mentioned in the issue and determine which matches the published model. Done means explaining the validation gap and identifying the configuration needed to reproduce the published result.

Written by the indexing model from the issue text.