Do any of the models allow for L2 Regularization?

[mg10011]

Basic question to start a conversation.

[manujosephv]

Typically, in PyTorch L2 Regularisation is applied using weight decay
Use weight_decay > 0 for L2 regularization:

optimizer = torch.optim.Adam(model.parameters(), lr=1e-4, weight_decay=1e-5)

Although this doesn't work like a Lasso Regression and necessarily gives you a sparse usage of features...