thu-keg / cleve Goto Github PK

请问此时的contrast example是如何构造的？相关代码是否可以上传？谢谢。

Hey, I‘m interested in the work. When could you release your code? Thanks！

Is the fine-tuned model shared? If not, could you kindly share it? I wanted to run an inference with it to check sample outputs.

Hi, I didn't find the code about Event Structure Pre-training. Is it missing ?

您好，我正在尝试复现您的代码，在生成contrast_examples.pkl文件后我开始对模型进行预训练，这时程序utils_ee.py第197行：trigger_idx = (node[0],node[1]-1)报错：unsupported operand type(s) for -: 'str' and 'int'。我尝试使用这种方式修改：trigger_idx = (node[0],int(node[1])-1)。
之后第382行：assert arg_posi in entities抛出异常，还有一些类似的问题我无法解决。
我想请教一下应该如何解决这些问题，或是您是否有修改后的代码？

请问对于amr图的编码工作代码是如何进行的？使用GCC应当如何调整输入？作者，这部分的代码可以发布一下吗？

(1) As the title expresses, the evaluation of EAE uses the golden trigger or the trigger predicted by the model.
(2) And does the EAE metric include the evaluation of role type?

Please confirm about the publish of the code

I used your model but cannot reproduce the same performance. Can you provide the processing data format?
The calculation process of the metrics uses the command "from sklearn.metrics import f1_score, precision_score," tha library recognizes events one token for one trigger. However, the definition of event detection is to identify and class the span of a trigger when the trigger consists of multi-tokens . Is it reasonable to use this library in this case?

不知道该项目是否可以在中文的数据集上面也有比较好的效果 这块是否有对于的预训练模型