zju-openks / openks Goto Github PK

我下载了百科的预训练中文词向量，是一个bz2文件，我直接解压改名为baike_word_ngram_300d，运行text_keyphrase.py文件，说无法解析这个词向量。

Markdown is a plain text format for writing structured document.

It's easy for read and write in anywhere.

For pdf and docx , you need a pdf reader or a word to read it.

In the apis folder, some documents can be rewrite to markdown files for better reading.

Traceback (most recent call last):
File "/home/huawei_share/.conda/envs/OpenKs/lib/python3.7/runpy.py", line 183, in _run_module_as_main
mod_name, mod_spec, code = _get_module_details(mod_name, _Error)
File "/home/huawei_share/.conda/envs/OpenKs/lib/python3.7/runpy.py", line 109, in _get_module_details
import(pkg_name)
File "/data02/OpenData/chx/OpenKS/examples/kg_dist_learn.py", line 45, in
kglearn.run(architect='ps', num_workers=2)
File "/data02/OpenData/chx/OpenKS/openks/models/pytorch/kg_learn_dist.py", line 191, in run
margin=self.args['margin']
File "/data02/OpenData/chx/OpenKS/openks/models/pytorch/kg_modules/TransE.py", line 20, in init
self.epsilon = kwargs['epsilon']
KeyError: 'epsilon'

export your dependencies to requirements.txt

pip freeze > requirements.txt

Then we can install it by

pip install -r requirements.txt

请问gen_module文件在哪里，market接口什么时候开源？

python -m examples.text_keyphrase
请问如何更换自己的数据集，我直接在后面加数据是不行的吗，还是需要标注或者关键词分数？