TokenFusion

TokenFusion copied to clipboard

Hi,Thanks for your creative work！ I am trying to reproduce the results for SUN RGBD on semantic segmentation task. The readme only give instructions about how to prepare for NYUDv2....

rzhevcherkasy

sunrgbd_pc_bbox_votes_50k_v1_train

3

Hi, /sunrgbd/sunrgbd_pc_bbox_votes_50k_v1_train How can I get this file

XT-1997

l1_lamda value for SUN RGBD

4

Hi, I am trying to reproduce the results for SUN RGBD. The paper mentions 1e-3 whereas the ReadMe mentions 1e-6. Thanks!

harshm121

Thanks for making your code public! I was wondering if you would also release the sun rgb-d data (preprocesses, if any) you used.

harshm121

checkpoint model architrcture

1

The model architecture of the checkpoint you provided for sematic segmentation is different from the model architecture you provided in the code ?

Norooa

loss2d_class_error 100.0000

1

Hi! Is loss2d_class_error 100.0000 normal? Thanks [12/21 19:28:05] group-free INFO: Train: [2][30/95] 0head_neg_ratio 0.6538 0head_pos_ratio 0.3462 1head_neg_ratio 0.6538 1head_pos_ratio 0.3462 2head_neg_ratio 0.6538 2head_pos_ratio 0.3462 3head_neg_ratio 0.6538 3head_pos_ratio 0.3462 4head_neg_ratio 0.6538...

keithAND2020

How long does it take to train the semantic_segmentation once?

Xinghetayue

Training error on single GPU

2

Hi, first of all this is excellent work! I am trying to train some data on the segmentation portion. When I training using the command“ python main.py --backbone mit_b3 -c...

happyencoding

在图像翻译任务中，为什么每个block后要使用LN？block的第一层就是LN，现在的话，是连续使用了两个LN层。

guoliangq

No pytorch models (.pth) saved during training?

5

Hi, first of all this is excellent work! I am trying to test some data on the segmentation portion. When I train it with pretained segformer weight, it saves data.pkl...

Rifahaziz