6c019d60eabdd9e50c9cc5752c53e6351bc4130b,mushroom/policy/td_policy.py,Boltzmann,__call__,#Boltzmann#,164

Before Change


            return qs[action] / np.sum(qs)
        else:
            p = np.ones(qs.size)
            for i in xrange(p.size):
                p[i] = qs[i] / np.sum(qs)

            return p

    def draw_action(self, state):
        return np.array([np.random.choice(self._approximator.n_actions,

After Change



            return qs[action] / np.sum(qs)
        else:
            return qs / np.sum(qs)

    def draw_action(self, state):
        return np.array([np.random.choice(self._approximator.n_actions,
                                          p=self(state))])
Italian Trulli
In pattern: SUPERPATTERN

Frequency: 3

Non-data size: 7

Instances


Project Name: AIRLab-POLIMI/mushroom
Commit Name: 6c019d60eabdd9e50c9cc5752c53e6351bc4130b
Time: 2017-12-24
Author: carloderamo@gmail.com
File Name: mushroom/policy/td_policy.py
Class Name: Boltzmann
Method Name: __call__


Project Name: AIRLab-POLIMI/mushroom
Commit Name: ca4c92fe1b53086373cfca478560df06b03701d8
Time: 2017-08-28
Author: carlo.deramo@gmail.com
File Name: examples/atari_dqn/convnet.py
Class Name: ConvNet
Method Name: predict


Project Name: pymanopt/pymanopt
Commit Name: 4346ff930104b0d9b6735090a1ce4c485d40b981
Time: 2016-02-25
Author: jamiehntownsend@gmail.com
File Name: pymanopt/manifolds/grassmann.py
Class Name: Grassmann
Method Name: retr