SkBlaz / rakun

Rank-based Unsupervised Keyword Extraction via Metavertex Aggregation
GNU General Public License v3.0
99 stars 22 forks source link

Problem while running mrakun/__init__.py -- lemmatizer = self.hyperparameters['lemmatizer'] KeyError: 'lemmatizer' #2

Closed aayushisanghi closed 5 years ago

aayushisanghi commented 5 years ago

I tried to run the sample code provided, but am getting the above error. Unsure how I could debug it, please do help!

SkBlaz commented 5 years ago

Hello, thanks for the issue. Have you tried with the examples in the README first? If the lemmatizer is giving you problems, you can simply comment that line, it is not at all necessary for the algorithm's function.

Could you please provide the full example of what you tried to run?

aayushisanghi commented 5 years ago
Screenshot 2019-09-30 at 11 47 20 PM

The above is a screenshot of the code I tried to run. (Basically the example provided in the README). Also, the problem exists in the lib/python3.7/site-packages/mrakun/init.py, which means it probably is an issue with the initial setup of Rakun. Is it still okay to remove the entire lemmatizer bit?

SkBlaz commented 5 years ago

Hey, yes, it is perfectly fine. Both stemming, as well as lemmatization are optional -> RaKUn should work smoothly without too.

Yesterday I was updating the repo, its possible you used some of the intermediary commits, which perhaps did not work as planned. Could you go through the README again (I've updated it). Also. The first thing to try are probably examples in the ./examples folder. Especially the Brexit example is independent of paths, if that is the problem.

On Mon, Sep 30, 2019 at 8:21 PM aayushisanghi notifications@github.com wrote:

[image: Screenshot 2019-09-30 at 11 47 20 PM] https://user-images.githubusercontent.com/49807208/65904787-c772a380-e3dc-11e9-807b-5a56a2f995c1.png

The above is a screenshot of the code I tried to run. (Basically the example provided in the README). Also, the problem exists in the lib/python3.7/site-packages/mrakun/init.py, which means it probably is an issue with the initial setup of Rakun. Is it still okay to remove the entire lemmatizer bit?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/SkBlaz/rakun/issues/2?email_source=notifications&email_token=ACMSERAOOIB7FVWFDKA3AETQMI7Q5A5CNFSM4I36PF3KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD76TFYY#issuecomment-536687331, or mute the thread https://github.com/notifications/unsubscribe-auth/ACMSERBWW7X6PY2UXMJF6BTQMI7Q5ANCNFSM4I36PF3A .

SkBlaz commented 5 years ago

Another thing, could you try to install via python3 setup.py install directly? We will update the pip version soon though.

SkBlaz commented 5 years ago

The pip version is online and seems to work fine with this example. Please, report if that is the case with your example also.