Zhaochong An

syncBN 'super' object has no attribute '_specify_ddp_gpu_num'

5

Hi, I notice that you fixed this bug at https://github.com/open-mmlab/mmcv/pull/1138. But I still encounter this when I run my code. I install pytorch and mmcv-full through ``` pip install torch==1.9.0+cu111...

Hi, I suffer from performance variance by rerunning the model on different settings. It seems you use different parameters for different dataset and few-shot setting that you did not mention...

Train code

2

Hi, I notice that the repo does not contain the training code. Can you upload the code used for training the model please? Thank you so much!

train dataset

1

Hi, can you explain more on the ground truth file (.csv in your preprocessed training dataset)? From my understanding, you divide the whole image into 32*32 patches and corresponding count...

Hi, I saw that you are not using the background class during training and testing. I am wondering how would you use the model to predict the binary mask for...