Fix: convert functional to model

fix: Functional added to layers

Fix: serve action uses new tensor dimensions

Readme updated

Freeze sklearn version

Releasing v1.13.1

Merge pull request #15 from evilsocket/strategy

Fix: implementation of multi gpu strategies

test

fix

added training strategy

Releasing v1.13

Merge pull request #14 from evilsocket/tf-2

import fixed

to-fdeep modified