Given data pairs (x,y), the linear method learns the model vector w by minizing the following objective function:
where ℓ is the loss function such as logistic loss and hinge loss.
Go the root directory of wormhole, then build by make linear. Next try
a small dataset using 2 worker and 1 server:
tracker/dmlc_local.py -n 2 -s 1 bin/linear.dmlc learn/linear/guide/demo.conf