-
这段想把单条多个情感极性的样本,转为多条的代码有bug
![image](https://user-images.githubusercontent.com/3937341/159827972-74441f0f-2827-4730-b461-de2a19688a47.png)
提供一个badcase:
[-999, -999, -999, Positive]
这种样本,会被过滤掉…
-
I try to run [`demos/aspect_polarity_classification/sentiment_inference.py`](https://github.com/yangheng95/PyABSA/blob/release/demos/aspect_polarity_classification/sentiment_inference.py) by replacing…
-
我标注了一套自己的数据集,单独跑apc(用你们的APC模型跑的)任务指标正常。APC acc为91,f1为91.但是我跑多任务的时候,用你们的ATEPC的时候,ATE指标倒是正常,可是APC的指标很低,F1为37.37(max:37.37)。。。这是为啥。。。
是不是我想做多任务的时候只能先方面抽取ATE再情感极性分类?
-
hi, yangheng!
As i was trying to use my own dataset to extract aspect, i followed your code as below, trying to load my data from local using absolute path:
```
from pyabsa import available_ch…
-
hi @yangheng95 I am not able to load the model sometimes, Not sure if this is the right way to do it. Please check the screen shot for details
![Screenshot from 2021-12-09 01-19-02](https://user-ima…
-
你好,请问中英文数据集的处理方式是一样的吗?在代码中哪里有所体现
-
我查看了作者的demo文件,我发现这些都需要从goole云盘下载自己的预训练模型(先要上传至谷歌云盘)
然而,我想加载本地的模型。例如:
![image](https://user-images.githubusercontent.com/40120224/151295824-0412e49a-9ad2-47ea-8aa6-a2228aaba2cf.png)
![image](https:/…
-
May I know how to make my own dataset? The steps shown are quite confusing.
-
在我使用checkpoint进行测试的时候,我使用了在训练时相同的文件夹名作为target_file,并且在文件夹中建立了.train.txt.inference的文件,可是出现了如下错误。
![1635161741(1)](https://user-images.githubusercontent.com/92669083/138688152-936acd8b-2d24-4025-b333-4…
-
**Hello dear, I hope you are fine. First of all, thank you so much for your great effort in this repository. I am following the quickstart of Aspect Polarity Classification in which I have trained the…