Binomial logistic regression is a linear classification algorithm, each sample is labeled as
positive(+1) or negtive(-1). In this algorithm, the probability describing a single sample
being drawn from the positive class is modeled using a logistic function:
P(Y=+1|X) = 1 / [1+ exp(-dot(w,x))]. This task learns a binomial logistic regression model
with mini-batch gradient descent.
Linear Supertypes
TrainTask[LongWritable, Text], BaseTask[LongWritable, Text, LabeledData], BaseTaskInterface[LongWritable, Text, LabeledData], AnyRef, Any
Binomial logistic regression is a linear classification algorithm, each sample is labeled as positive(+1) or negtive(-1). In this algorithm, the probability describing a single sample being drawn from the positive class is modeled using a logistic function: P(Y=+1|X) = 1 / [1+ exp(-dot(w,x))]. This task learns a binomial logistic regression model with mini-batch gradient descent.