Classic Regression model regularisation error

[stanton119]

Hi,

The L1 and L2 regularisation for both the linear and logistic regression models includes the bias, rather than just the weights.
e.g. as seen here:
https://github.com/PyTorchLightning/lightning-bolts/blob/master/pl_bolts/models/regression/linear_regression.py#L61

The regularisation is commonly only applies to the weights and the bias remains unaffected. More discussion:
https://stats.stackexchange.com/questions/86991/reason-for-not-shrinking-the-bias-intercept-term-in-regression?rq=1

[github-actions]

Hi! thanks for your contribution!, great first issue!