fossology / atarashi

Atarashi scans for license statements in open source software, focusing on text statistics. Designed to work stand-alone and with FOSSology.
http://fossology.github.io/atarashi
GNU General Public License v2.0
26 stars 23 forks source link

feat(model): Add linearsvc agent #102

Open its-sushant opened 2 years ago

its-sushant commented 2 years ago

Description

Added a agent for Linear Support vector machine model.

Files

How to use

Accuracy

The accuracy score that has been tested using evaluator.py is 63%

CC: @hastagAB @Kaushl2208 @GMishx @ag4ums @vasudevmaduri

amanjain97 commented 2 years ago

@hastagAB I think all PRs should have sphinx updates in them included or have we added this to github workflows ?

Kaushl2208 commented 2 years ago

Tested the PR, Changes seems to be working :)

image

A small thing that we can look at a warning that popped up due to the change in version for unpickling the file with different versions:

image

You can read about it from here: https://scikit-learn.org/stable/model_persistence.html#security-maintainability-limitations

Comments? @its-sushant @hastagAB @GMishx

Kaushl2208 commented 2 years ago

Tested the PR, Changes seems to be working :)

A small thing that we can look at a warning that popped up due to the change in version for unpickling the file with different versions:

You can read about it from here: https://scikit-learn.org/stable/model_persistence.html#security-maintainability-limitations

Comments? @its-sushant @hastagAB @GMishx

Is it happening because of old version of linearsvc being used in linearsvc>=0.1.1 ??

Please try fixing it to a working version using linearsvc==0.1.1 so pickling and unpickling does not cause issues.

its-sushant commented 2 years ago

Tested the PR, Changes seems to be working :) A small thing that we can look at a warning that popped up due to the change in version for unpickling the file with different versions: You can read about it from here: https://scikit-learn.org/stable/model_persistence.html#security-maintainability-limitations Comments? @its-sushant @hastagAB @GMishx

Is it happening because of old version of linearsvc being used in linearsvc>=0.1.1 ??

Please try fixing it to a working version using linearsvc==0.1.1 so pickling and unpickling does not cause issues.

@Kaushl2208 Actually the model has been trained on sklearn==1.0.2 and atarashi uses the updated version, that's why it is showing some warning. I need to train the model again with sklearn==1.1.1 then I guess the warning will be gone.