立秋

In the paper table2, why compare Hamming-distance+counting-sort with Euclidean-distance+quick-sort? If choose Euclidean-distance+count-sort, the difference between them will be much small.

How to find suitable lrs with FastaiLRFinder when the optimizer has Multiple groups ?

5

```python optimizer = optim.SGD([ {'params': model.conv.parameters(), 'lr': 1}, {'params': model.linear.parameters(), 'lr': 0.1}, ], lr=3e-4, momentum=0.9) ```` Such as this, the optimizer has two different groups. Can anyone give an example?

在下载数据时报错

1

下载数据集时报错 运行下面代码： ``` from fastNLP.io import WeiboNERPipe data_bundle = WeiboNERPipe().process_from_file() ``` 抛出异常： File is not a zip file  系统是Ubuntu20.04

Nice work! Where can I get the pretrained model of SYSU?

How to get the visualization result like Fig.3?

```python for id in sorted(self.ids): self.all_files.extend(self.files_rgb[id]) #self.rgb_files.extend(self.files_rgb[id]) self.all_files.extend(self.files_rgb[id]) #self.ir_files.extend(self.files_ir[id]) ``` Why both step add files_rgb?

release code

3

Nice work. When do you plan release code?