AutoViML / Auto_ViML

Automatically Build Multiple ML Models with a Single Line of Code. Created by Ram Seshadri. Collaborators Welcome. Permission Granted upon Request.
Apache License 2.0
522 stars 102 forks source link

ModuleNotFoundError: No module named 'lightgbm' #25

Closed emsi closed 2 years ago

emsi commented 2 years ago

I've installed from pip. I'm using py 3.9.

When attempting:

train_x, test_x, final, predicted= Auto_NLP(input_feature, train, test,target,score_type="balanced_accuracy",top_num_features=100,modeltype="Classification", verbose=2, build_model=True)

I got:

[nltk_data]    | 
[nltk_data]  Done downloading collection popular

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
Input In [35], in <cell line: 1>()
----> 1 train_x, test_x, final, predicted= Auto_NLP(input_feature, train, test,target,score_type="balanced_accuracy",top_num_features=100,modeltype="Classification", verbose=2, build_model=True)

File /usr/local/lib/python3.9/site-packages/autoviml/Auto_NLP.py:1334, in Auto_NLP(nlp_column, train, test, target, score_type, modeltype, top_num_features, verbose, build_model)
   1332 nltk.download("popular")
   1333 calibrator_flag = False
-> 1334 from lightgbm import LGBMClassifier, LGBMRegressor
   1335 seed = 99
   1336 train = copy.deepcopy(train)

ModuleNotFoundError: No module named 'lightgbm'

A missing dependency?

AutoViML commented 2 years ago

Hi @emsi 👍 Thanks for testing out AutoViML. I will add the lightGBM dependency when I roll out the next version. In the meantime, try pip installing lightgbm as a workaround in your machine. Thanks AutoViML

emsi commented 2 years ago

Hi @emsi +1 Thanks for testing out AutoViML. I will add the lightGBM dependency when I roll out the next version. In the meantime, try pip installing lightgbm as a workaround in your machine. Thanks AutoViML

That's precisely what I did :)

AutoViML commented 2 years ago

@emsi 👍 LightGBM has been added to requirements.txt and setup.py - this should work now. You can also create a pull request in these files in the future if you feel anything is missing. Thanks AutoViML Team