houjp / kaggle-quora-question-pairs Goto Github PK

@staticmethod
def logistic(y):
assert isinstance(p, np.ndarray), 'type error'
return np.exp(y) / (1. + np.exp(y))

in this def, the p should be y or the variable p defined in other def?

请教大佬,在 PowerfulWord 类中,同样是计算正确语句对数量和比例,为什么在单侧是用0==label判断,在双侧是用1==label判断呢?

Thanks

Traceback (most recent call last):
File "/data/git/text_similarity/kaggle-quora-question-pairs/bin/feature_engineering/graph.py", line 332, in
demo()
File "/data/git/text_similarity/kaggle-quora-question-pairs/bin/feature_engineering/graph.py", line 327, in demo
GraphEdgeMaxCliqueSize(config_fp).extract('train')
File "/data/git/text_similarity/kaggle-quora-question-pairs/bin/feature_engineering/graph.py", line 135, in init
UDG.init(self, config_fp)
File "/data/git/text_similarity/kaggle-quora-question-pairs/bin/feature_engineering/graph.py", line 127, in init
reverse)
File "/data/git/text_similarity/kaggle-quora-question-pairs/bin/feature_engineering/graph.py", line 83, in generate_graph
fin = csv.reader(open('%s/train.csv' % config.get('DIRECTORY', 'origin_pt')))
File "/home/lyy/anaconda2/lib/python2.7/ConfigParser.py", line 607, in get
raise NoSectionError(section)
ConfigParser.NoSectionError: No section: 'DIRECTORY'

Did your solution have been archived ?

大佬们，你们机器配置和python版本，其他框架版本能说一下吗？我刚刚装ConfigParser失败了，改成小写的就可以了

请问两种图中 权重分别是什么值

Could you please explain more about PowerfulWord? How does it work?
PowerfulWord

Thanks very much!

Program 01

新手小白，想问问如何可以启动工程，非常感谢！

Train data: Who is Michael Jordan? and Who is Air Jordan? are one right match.
The test data is Who is Jumpman Jordan? and Who is Air Jordan?

What feature should I add for training? Which model can capture the match info for this case?

Thank you！！！
@pl8787 @HouJP

https://www.kaggle.com/c/quora-question-pairs/discussion/34355
What is the function of XGB / LGBM in this 1st solution?
Thank you very much.
@HouJP @pl8787 @faneshion @ddDragon

I can not find the topic-model-feature code.
Thank you!
@pl8787 @HouJP

小白看了大神的代码，收获颇多，但是不知道怎么运行起来生成一个submission，求大神指点。

The INPUT are two sentences.

The OUTPUT is the match score.

Am I right?

Thank you!@pl8787 @HouJP @ddDragon

where can I see the deep model? thanks!

@HouJP @pl8787 @faneshion
我现在只尝试传统方法，大概特征维数在40左右，logloss还是有0.4之多，咋办啊……
多谢！

@HouJP @pl8787 @faneshion
多谢多谢！

how do i see sent invit on linkedin if i do not have ani more pend invit ?
peopl say do not tri to pleas other . doe be nice to other mean pleas them ?
用完你的预处理之后，生成的句子变成了这样。
这个冒号``是如何出现的???

每次看test question是不是duplicate的时候，是要把所有的questions都检查一遍吗？
@pl8787 @HouJP @ddDragon 谢谢！！