Training model with L1 regularization

epfml / sent2vec

General purpose unsupervised sentence representations

Other

1.19k stars 256 forks source link

Training model with L1 regularization #77

Closed patrickfrank1 closed 4 years ago

patrickfrank1 commented 5 years ago

Hi,

in your 2018 paper you also train models on a modified loss function that includes L1 regularization. How can I train a new model with L1 regularization? Is this functionality also implemented in this library here?

Cheers, Patrick

mpagli commented 5 years ago

Not sure if this was pushed in the final repo. Any idea @guptaprkhr ?

guptaprkhr commented 4 years ago

Hello, the functionality was not added in the repo as it gave only marginal gains.