Skip to content

直接跑evaluation跑不出来论文里的数值,是我哪一步错了么 #13

@WinnieLaugh

Description

@WinnieLaugh

Hi 大佬你好,

我直接把代码和数据集拉下来然后按照evaluation里的README.md跑了一下结果,发现没法跑到论文的数值,是我哪一步做错了么?

比如Agriculture这个数据,我跑的命令是

  1. nohup python script_insert.py --cls agriculture > result_agriculture_insert.log 2>&1 &
  2. python script_hypergraphrag.py --data_source agriculture
  3. python get_generation.py --data_sources agriculture --methods HyperGraphRAG
  4. python get_generation.py --data_sources agriculture --methods HyperGraphRAG
  5. python get_generation.py --data_sources agriculture --methods HyperGraphRAG

内部代码都没改。我得到的结果是:

Data Source: agriculture, Method: HyperGraphRAG
Score Dictionary:
b
f1 11.61
rsim 10.31
gen 34.12
=====================================
n
f1 13.37
rsim 11.04
gen 36.68
=====================================
overall
overall_f1 12.49
overall_rsim 10.67
overall_gen 35.4
=====================================

多谢!

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions