GLM¶

Generalized Linear Model.

This serves as a base class for linear and logistic regression.

Parameters¶

• optimizer (mutable)

  Type → optim.base.Optimizer

  The sequential optimizer used for updating the weights. Note that the intercept updates are handled separately.

• loss (mutable)

  Type → optim.losses.Loss

  The loss function to optimize for.

• l2 (mutable)

  Type → float

  Amount of L2 regularization used to push weights towards 0. For now, only one type of penalty can be used. The joint use of L1 and L2 is not explicitly supported.

• l1 (mutable)

  Type → float

  Amount of L1 regularization used to push weights towards 0. For now, only one type of penalty can be used. The joint use of L1 and L2 is not explicitly supported.

• intercept_init

  Type → float

  Initial intercept value.

• intercept_lr (mutable)

  Type → optim.base.Scheduler | float

  Learning rate scheduler used for updating the intercept. A optim.schedulers.Constant is used if a float is provided. The intercept is not updated when this is set to 0.

• clip_gradient (mutable)

  Type → float

  Clips the absolute value of each gradient value.

• initializer (mutable)

  Type → optim.base.Initializer

  Weights initialization scheme.

Attributes¶

• weights

Methods¶