bp20200202

如果方便的话，我也想要一份bi-lstm的代码，邮箱[email protected],万分谢谢

> @Magicat128 > 您好 谢谢你的回覆 > 请问您有尝试将textING做在中文的数据上吗? > 因为我一样将中文多标签文本跟标签用成你要的输入格式(已经断过词) > 但是用textING train时，完全train不起来(有画train loss ,val loss ,图形十分奇怪) > 有将文本down sampling成balanced data,但还是没有好的结果! > > 但是用相同的前处理,textGCN可以跑! > 我亦有将textGCN , textING跑过 mr的英文数据集,可以跑. > 但中文只有textGCN可以跑....

> > > @ Magicat128 > > > 您好谢谢你的回覆 > > > 请问您有尝试将texting做在中文的数据上吗？ > > > 因为我一样将中文多标签文本跟标签用成你要的输入格式（已经断过词） > > > 但是用texting train时，完全火车不起来（有画火车损失，val损失，图形十分奇怪） > > > 有将文本向下采样成平衡数据，但还是没有好的结果！ > > > 但是用相同的前处理，textGCN可以跑！...

> Hello, > I have modified the trainer to process data batch by batch.. > Is this a right way? > https://github.com/CRIPAC-DIG/TextING/blob/c2492c276a6b59ca88337e582dfd2f3616f3988d/train.py#L124 > > b_train_adj, b_train_mask = preprocess_adj(train_adj[idx]) > b_train_feature...

> But if we really need to include unlabeled documents that won't be used for testing either, we need to change codes on model/code/utils.py and perhaps build_data.py(if we use this...

> > > But if we really need to include unlabeled documents that won't be used for testing either, we need to change codes on model/code/utils.py and perhaps build_data.py(if we...

> @bp20200202 > I just remember that we also made one mistake of only using the **Titles** instead of **Description** for the **Agnews** Dataset. Maybe you are also using the...

@Chengmeng94 May I ask if you have solved this problem? I also encountered the same problem. I would like to know how you solved it