xxxiaol / gci Goto Github PK

The following warning appears when running the gci.py file
WARNING:root:No stemmer available in pke for 'zh' language -> falling back to porter stemmer.
WARNING:root:No stoplist available in pke for 'zh' language.

Without stopword lists and stemmers, what are the subsequent requirements for the dataset this project handles?

多谢分享！
有一个问题，我使用data.json文件发现案件的数量和论文中的似乎不一致，例如下图所示，在data.json文件中 故意伤害 有5006个，但是在论文中有6377个，是不是我哪里没有考虑到？多谢
我已经尝试过了将训练集验证集测试集都计算在内，但是这样案件数量就会多于论文的数量。同时我注意到论文中好像要去掉多项罪名，但是在data.json中仍然存在多项罪名的数据，我尝试将这些数据过滤掉后，总的数量又会小于论文所示的数量。以上的数量都只是说 故意伤害 这一个罪名

I just ran this project in Colab and I got this problem:

File "/content/GCI-main/yake_modified.py", line 97, in candidate_selection
self.candidate_filtering(stoplist=list(string.punctuation), minimum_length=1, minimum_word_size=1)
TypeError: candidate_filtering() got an unexpected keyword argument 'stoplist'

How can I fix it?

ERROR:root:No stemmer available for 'zh' language -> fall back to porter.
Thank you for your amazing work! Could you help me figure out the possible cause of this error. Thx.

I run the version boudinfl/pke@0729bca in colab.
And I get this error:

LookupError: Can't find any language named 'hinglish'

I didn't encounter this problem when use latest PKE. How can I fix it?

How long will it take to 'Run preprocess.py to prepare data for GCI'?

The warning occurs when I run the gci.py.
No stemmer for 'zh' language.
Stemming will not be applied.

pip install javabridge doesn't work:

This is my solution when using pycausal:

install javabridge before pycausal:

sudo apt install openjdk-8-jdk gcc

pip install git+https://github.com/LeeKamentsky/python-javabridge

download pycausal and open setup.py

python setup.py install --user

您好，我想请问一下去哪里下载Py-Causal呀？
如果是这个链接的：https://github.com/bd2kccd/py-causal，
他要求的是python=3.6,而GCI要求的是3.7以上，冲突了。

如题，我现在使用浏览器登入该网站会出现如下问题：

无法正常进入网站。想请问这一现象是暂时的还是该网站已经失效？如果该网站已经失效的话，请问对应的文件您这边还能否提供呢？
麻烦您了，谢谢！