search

sdv-dev / TGAN

Generative adversarial training for generating synthetic tabular data.
MIT License
281 stars 91 forks source link

Standardation needed? #35

Closed bbb801 closed 5 years ago

bbb801 commented 5 years ago

Dear contributor,

May I know is it necessary to standard or normalize the data before using the TGAN? Is there a big difference between the results with and without such standardization/normalization? Thanks.

Beato

leix28 commented 5 years ago

Yes. Value with arbitrary range makes the training difficult. Without normalization, the model can easily run into numerical issues or gradient issues.

csala commented 5 years ago

Closing this, as the question was already answered.