Chang-Bin Zhang

I remember that we use 8x 1080ti to train the model on the ADE20K, but I'm not sure the batch size. @schuy1er

Hi, @HieuPhan33 @Ze-Yang I'm sorry for the late reply. We have fixed bugs in the repository, and you can find the re-implemented results on VOC and ADE20K. If there is...

Do you use the script in the current commit? And the results in the [link](https://github.com/zhangchbin/RCIL/blob/main/results/2022-08-26_voc_15-5s_RCIL_overlap.csv).

I add the `--overlap` option (step 0) to the scripts for overlap setting just now.

Hi, @HieuPhan33 I trained the model on the setting of 15-1 overlapped and got slightly higher performance (59.9) than our reported results (59.4). I have uploaded the detailed results in...

Hi, @HieuPhan33 I also clone this repository and re-train the 15-1 disjoint setting. The experimental results are in the ```results``` and ```logs```, respectively. You can refer to them.

Hi, @XiaorongLi-95 We have added scripts for Cityscapes.

@XiaorongLi-95 We usually use pytorch 1.3.1 on 4 2080ti gpus. For the difference on cityscapes, I have fixed the initialization for Cityscapes dataset, and you can refer to [link](https://github.com/zhangchbin/RCIL/blob/main/segmentation_module.py#L239). If...

> I got the same problem Is there any solution?

> > > I got the same problem > > > > > > Is there any solution? > > Not yet but I use the configuration of torch1.2.0_cuda11.6, cuda11.6...