public SGD(learningRate as Double = 0.01)
Configures a finite, positive learning rate.
learningRate
optimizer = SGD<Float>(0.1)