Yilun Xu

Results 6 comments of Yilun Xu

关于pre-computed image features

> 请问VLP 里有全局的feature下载 和regional的feature下载，是应该下载哪个？ > > # data > wget -O COCO00 "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212019&authkey=ACn4bwZ0nmZ0nik" > wget -O COCO01 "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212018&authkey=AHoTGG-7-6kwoAY" > wget -O flickr30k.tar.gz "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212015&authkey=AFZ2iehPM8HREeA" > wget -O CC.tar.gz "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%213781&authkey=ANA--esfJnWIKIE" 请问这个怎么比COCO本身还大呀？

关于pre-computed image features

> > > 请问VLP 里有全局的feature下载 和regional的feature下载，是应该下载哪个？ > > > # data > > > wget -O COCO00 "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212019&authkey=ACn4bwZ0nmZ0nik" > > > wget -O COCO01 "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212018&authkey=AHoTGG-7-6kwoAY" > > > wget -O...

关于pre-computed image features

> 请问VLP 里有全局的feature下载 和regional的feature下载，是应该下载哪个？ > > # data > wget -O COCO00 "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212019&authkey=ACn4bwZ0nmZ0nik" > wget -O COCO01 "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212018&authkey=AHoTGG-7-6kwoAY" > wget -O flickr30k.tar.gz "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%212015&authkey=AFZ2iehPM8HREeA" > wget -O CC.tar.gz "https://onedrive.live.com/download?cid=E5364FD183A1F5BB&resid=E5364FD183A1F5BB%213781&authkey=ANA--esfJnWIKIE" 您好，请问您是用什么方法下载这么大的数据的？

关于model.py

请问是否需要在训练部分对代码进行调整，我在训练完2个epoch后就会报 CUDA out of memory 的错误，但是我是11G的2080，不知道是什么原因出现这样的问题.

Does CTUNet 's masked self-attention mechanism in the graph attention network USE the neighbours relation in Structural Graph Construction part?

From the codes opensource in CTUNet, I only see this: `multimodal_context, batched_img_label, batched_img_bieo_label, bert_token_embeddings = \ self.infor_context_module(info_feat_list, pos_feat=gt_bboxes[0], img_metas=img_metas, info_labels=None, bieo_labels=None, gt_texts=gt_texts[0], char_nums=char_nums)` I can't see any param about graph...

您好 如果使用SciTSR进行训练 也是转换成davar格式么？

> 是的，需要先理解SciTSR的格式和Davar格式表达的含义之后，脚本实现很简单，所有的信息都是可以从原来的推导出来的。对于davar的json格式中如果有字段含义不理解，这里可以帮你解答一下。 请问您是如何处理SciTSR数据集里chunk内bbox与img无法对应的问题的