Sam comments

Results 16 comments of

Sam

请问比赛的数据集能分享一下吗？找不到下载的地址，想重现一下，谢谢

过了比赛时间，官网的数据下载不了。可以提供一份或者邮箱[email protected]，非常感谢！

KeyError: 'BEiT is not in the backbone registry'

i follow the steps using https://github.com/SamXiaosheng/unilm/tree/master/beit/semantic_segmentation, how to solve it?thanks advance.

KeyError: 'BEiT is not in the backbone registry'

thanks for your answer, i check the version of mmseg lib(mmcv-full==1.3.0 & mmsegmentation==0.14.1) ,mmsegmentation is higher verison , is it possible the problem?

the training data of Meta fintuning how to extract?

请问训练好了，要怎么推理？可以有inferrence文件吗 ------------------ 原始邮件 ------------------ 发件人: "Gongjie ***@***.***>; 发送时间: 2022年4月18日(星期一) 下午3:25 收件人: ***@***.***>; 抄送: "生second ***@***.***>; ***@***.***>; 主题: Re: [ZhangGongjie/Meta-DETR] the training data of Meta fintuning how to extract? (Issue #19) Sorry, I...

i have one small question about that the backbone from the file "Base-FSOD-C4.yaml" is set empty?

if not NAME，than its deault backbone is ResNET? 发自我的iPhone ------------------ Original ------------------ From: Guangxing Han ***@***.***> Date: Thu,Nov 24,2022 0:43 AM To: GuangxingHan/Meta-Faster-R-CNN ***@***.***> Cc: Sam ***@***.***>, Author ***@***.***> Subject:...

i have one small question about that the backbone from the file "Base-FSOD-C4.yaml" is set empty?

Hello, i have some confusion about your project, the second step "sh scripts/faster_rcnn_with_fpn_coco_base_classes_branch.sh"  and third step "scripts/few_shot_finetune_coco_resnet101.sh" have Superlative relationship?  ------------------ 原始邮件 ------------------ 发件人: "Guangxing ***@***.***>; 发送时间: 2022年11月24日(星期四) 上午7:54 收件人: ***@***.***>; 抄送: "生second...

i have one small question about that the backbone from the file "Base-FSOD-C4.yaml" is set empty?

Hi, i found the second step use build_resnet_fpn_backbone and generated the final model with FPN module, but not used for third step. The third step use the pretraind model from the...

could you help me this problem?

in the first stage , and when testing the few shot. 发自我的iPhone ------------------ Original ------------------ From: Guangxing Han ***@***.***> Date: Fri,Dec 9,2022 4:21 AM To: GuangxingHan/Meta-Faster-R-CNN ***@***.***> Cc: Sam ***@***.***>,...

could you help me this problem?

[12/08 17:53:30] d2.data.datasets.coco INFO: Loading datasets/coco/annotations/instances_val2014.json takes 4.73 seconds. [12/08 17:53:30] d2.data.datasets.coco INFO: Loaded 40504 images in COCO format from datasets/coco/annotations/instances_val2014.json [12/08 17:53:33] d2.data.build INFO: Distribution of instances among all...

could you help me this problem?

i modify cls_num = pred_cls.shape[0]#pred_cls.unique().shape[0] in fsod_fast_rcnn.py file line 108 ,and it can work.