Merge remote-tracking branch 'upstream/develop' into develop-flat-code-restructure

python/mlagents/mlagents/envs/environment.py

                 )
                 if b in value:
                     if value[b] is not None:
-                        action.value = value[b][i]
+                        action.value = float(value[b][i])
                 rl_in.agent_actions[b].value.extend([action])
                 rl_in.command = 0
         return self.wrap_unity_input(rl_in)

python/mlagents/mlagents/learn.py

         docker_target_name = run_options['--docker-target-name']
 
     # General parameters
+    env_path = run_options['<env>']
     run_id = run_options['--run-id']
     load_model = run_options['--load']
     train_model = run_options['--train']
     logger.info(options)
     num_runs = int(options['--num-runs'])
     seed = int(options['--seed'])
-    env_path = options['<env>']
-    if env_path is None and num_runs > 1:
+    if options['<env>'] is None and num_runs > 1:
         raise TrainerError("It is not possible to launch more than one concurrent training session "
                            "when training from the editor")