Joseph K J

flush_network flushes out 'concat_block' 'fadein_block' modules from model definition. Kindly read through the paper to understand more.

> Hi, yaoyao. I directly followed the instructions and ran the code "python main.py --nb_cl_fg=50 --nb_cl=10 --gpu=0 --random_seed=1993 --baseline=lucir --branch_mode=dual --branch_1=ss --branch_2=free --dataset=imagenet". I just changed the dataset option and...

Hi @yaoyao-liu : Thanks again for your work. I tried to replicate ImageNet results, but the numbers for the baselines are far from what we have on paper (~10%) or...

HI @yaoyao-liu, Many thanks for your swift response. You are correct, I am indeed trying to replicate the baseline results. I will also try to fix this by CVPR, I...

I believe @yaoyao-liu has imagenet included with the AANet work. Thank you @yaoyao-liu!

How many GPUs are you using? I wonder whether the inconsistency in training is due to that. Kindly reopen if the issue persists.

Hi @ColinTr, Thank you very much for sharing your analysis. Kindly allow me to share my thoughts on the same: As you have rightly pointed it out, NCD is nascent...

Looking forward @gidariss. Thanks in advance!

You need to convert the .mat annotations into images, before converting to tf_records You can do this by: ``` from tf_image_segmentation.utils.pascal_voc import convert_pascal_berkeley_augmented_mat_annotations_to_png pascal_berkeley_root = '/home/joseph/Dataset/BerkleySegmentationData/benchmark_RELEASE' convert_pascal_berkeley_augmented_mat_annotations_to_png(pascal_berkeley_root) ```

Please submit a PR, thanks!