Huang Qiang

Hey,@BigRedT Thanks for your excellent job.Can you tell me the code will use how many GPU memory? I have only one NVIDIA TITAN XP 12GB.

Thanks to the good work, the position embedding initialization is multiplied by a scaling factor, which is not initialized in the original VIT. It is also mentioned in the paper...

Hi, I have a question about obtain word type. Why use perception to get the word's type not Stanford parser?

The reproduced performances on unc/unc+

19

Hi， The AP I reproduced on unc and unc+ is only 30%，but it is normal on referit and gref.I tried to download the dataset and code again, but it still...