alibaba / alibaba/EasyRec

PDN 中的 PATH 分数计算与论文中的公式不符

Open
#509 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Python
Stars
2.4k
Forks
383
PR merge metrics
No merged PRs in 30d

Description

https://github.com/alibaba/EasyRec/blob/d965fb796ca834285b82dec071128aec3a82c584/easy_rec/python/model/pdn.py#L43-L45

代码中实现的是:

```
trigger_out = tf.exp()
sim_out = tf.exp()
logits = tf.multiplay(sim_out, trigger_out)
```

相当于是:

$$ PATH_{uji} = e^{t_{uj}} e^{s_{ji}} $$

论文中的合并公式(公式(10))是:

$$
PATH_{uji} = MEG(t_{uj}, s_{ji}) = \ln(1 + e^{t_{uj}} e^{s_{ji}})
$$

Contributor guide

No contributing guide indexed for this repository

Research direction

Start by reading easy_rec/python/model/pdn.py lines 43-45 and compare the calculation with formula (10) in the referenced paper. Look for existing PDN model tests or validation around this calculation; done means the PATH score follows the paper’s formula and the relevant behavior is validated.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Clearly specified
Newbie friendliness
48/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.