search

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

memory issue while running multi label classification #19

Closed taknev83 closed 4 years ago

taknev83 commented 4 years ago

Hi Ram,

I am trying multi label classification with the following parameters & encountering "Kernal tried to allocate more memory than available" error, while running in kaggle with memory of 16GB. The predictor in train is title & abstract of the book. Is there is any way to overcome this memory issue?

model, features, trainm, testm = Auto_ViML( train=train, target=['Computer_Science', 'Physics', 'Mathematics', 'Statistics', 'Quantitative_Biology', 'Quantitative_Finance'], test=test, sample_submission=sample_submission, hyper_param="GS", feature_reduction=False, scoring_parameter="f1", KMeans_Featurizer=False, Boosting_Flag=True, Binning_Flag=True, Add_Poly=False, Stacking_Flag=False, Imbalanced_Flag=False, verbose=2, )

image

Thanks, Venkatesh

AutoViML commented 4 years ago

I have fixed the problem in Auto_NLP. Can you please combine the title and abstract columns into a single column (and then delete them) and then feed the data set to Auto_NLP. Make sure you use the latest version that's in the github by typing this command in the Colab: !pip install git+https://github.com/AutoViML/Auto_ViML.git

Then do this: from autoviml.Auto_ViML import Auto_NLP

Then run Auto_NLP with the data set. It will run the whole thing and provide you results.

Ram

On Tue, Aug 18, 2020 at 1:28 PM Venkatesh Rengarajan Muthu < notifications@github.com> wrote:

Hi Ram,

I am trying multi label classification with the following parameters & encountering "Kernal tried to allocate more memory than available" error, while running in kaggle with memory of 16GB. The predictor in train is title & abstract of the book. Is there is any way to overcome this memory issue?

model, features, trainm, testm = Auto_ViML( train=train, target=['Computer_Science', 'Physics', 'Mathematics', 'Statistics', 'Quantitative_Biology', 'Quantitative_Finance'], test=test, sample_submission=sample_submission, hyper_param="GS", feature_reduction=False, scoring_parameter="f1", KMeans_Featurizer=False, Boosting_Flag=True, Binning_Flag=True, Add_Poly=False, Stacking_Flag=False, Imbalanced_Flag=False, verbose=2, )

[image: image] https://user-images.githubusercontent.com/61316462/90545176-4c9fac00-e199-11ea-84b7-bd3dfac4a8c0.png

Thanks, Venkatesh

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/AutoViML/Auto_ViML/issues/19, or unsubscribe https://github.com/notifications/unsubscribe-auth/AMKBH6E2AB4C3SWEUFIB4ILSBK2ZXANCNFSM4QD2FULQ .