Fix bug in DDPG parameter space noise adaptation (#306)

The training loop used the rollout step variable `t` rather than the
training step variable `t_train` to decide when to adapt the scale of
the parameter space noise.

baselines/ddpg/training.py

@@ -109,7 +109,7 @@ def train(env, nb_epochs, nb_epoch_cycles, render_eval, reward_scale, render, pa
                 epoch_adaptive_distances = []
                 for t_train in range(nb_train_steps):
                     # Adapt param noise, if necessary.
-                    if memory.nb_entries >= batch_size and t % param_noise_adaption_interval == 0:
+                    if memory.nb_entries >= batch_size and t_train % param_noise_adaption_interval == 0:
                         distance = agent.adapt_param_noise()
                         epoch_adaptive_distances.append(distance)