resolved conflicts

docs/Learning-Environment-Design-Brains.md

    *  `Graph Model` : This must be the `bytes` file corresponding to the pretrained Tensorflow graph. (You must first drag this file into your Resources folder and then from the Resources folder into the inspector)
    *  `Graph Scope` : If you set a scope while training your TensorFlow model, all your placeholder name will have a prefix. You must specify that prefix here.
    *  `Batch Size Node Name` : If the batch size is one of the inputs of your graph, you must specify the name if the placeholder here. The brain will make the batch size equal to the number of agents connected to the brain automatically.
-   *  `State Node Name` : If your graph uses the state as an input, you must specify the name if the placeholder here.
+   *  `Vector Observation Node Name` : If your graph uses a vector observation as an input, you must specify the name if the placeholder here.
    *  `Recurrent Input Node Name` : If your graph uses a recurrent input / memory as input and outputs new recurrent input / memory, you must specify the name if the input placeholder here.
    *  `Recurrent Output Node Name` : If your graph uses a recurrent input / memory as input and outputs new recurrent input / memory, you must specify the name if the output placeholder here.
    * `Visual Observation Placeholder Name` : If your graph uses observations as input, you must specify it here. Note that the number of observations is equal to the length of `Camera Resolutions` in the brain parameters.

docs/Using-Docker.md

 Unity environment **has** to be built for the **linux platform**. From the Build Settings Window, please select the architecture to be `x86_64` and choose the build to be `headless` (_This is important because we are running it in a container that does not have graphics drivers installed_). 
 Save the generated environment in the directory to be mounted (e.g. we have conveniently created an empty directory called at the top level `unity-volume`). 
 
-![Build Settings For Docker](../images/docker_build_settings.png)
+![Build Settings For Docker](images/docker_build_settings.png)
 
 - Ensure that `unity-volume/<environment-name>.x86_64` and `unity-volume/environment-name_Data`. So for example, `<environment_name>` might be `3Dball` and you might want to ensure that `unity-volume/3Dball.x86_64` and `unity-volume/3Dball_Data` are both present in the directory `unity-volume`.
 

python/requirements.txt

 matplotlib
 numpy>=1.11.0
 jupyter
-mock>=2.0.0
 pytest>=3.2.2
 docopt
 pyyaml

python/tests/test_bc.py

-import mock
+import unittest.mock as mock
 import pytest
 
 import numpy as np
                         run_list = [model.sample_action, model.policy]
                         feed_dict = {model.batch_size: 2,
                                      model.sequence_length: 1,
-                                     model.state_in: np.array([[1, 2, 3, 1, 2, 3],
+                                     model.vector_in: np.array([[1, 2, 3, 1, 2, 3],
                                                                [3, 4, 5, 3, 4, 5]])}
                         sess.run(run_list, feed_dict=feed_dict)
                         env.close()
                         feed_dict = {model.batch_size: 2,
                                      model.dropout_rate: 1.0,
                                      model.sequence_length: 1,
-                                     model.state_in: np.array([[1, 2, 3, 1, 2, 3],
+                                     model.vector_in: np.array([[1, 2, 3, 1, 2, 3],
-                                     model.observation_in[0]: np.ones([2, 40, 30, 3])}
+                                     model.visual_in[0]: np.ones([2, 40, 30, 3])}
                         sess.run(run_list, feed_dict=feed_dict)
                         env.close()
 

python/tests/test_ppo.py

-import mock
+import unittest.mock as mock
 import pytest
 
 import numpy as np
                                     model.learning_rate]
                         feed_dict = {model.batch_size: 2,
                                      model.sequence_length: 1,
-                                     model.state_in: np.array([[1, 2, 3, 1, 2, 3],
+                                     model.vector_in: np.array([[1, 2, 3, 1, 2, 3],
                                                                [3, 4, 5, 3, 4, 5]])}
                         sess.run(run_list, feed_dict=feed_dict)
                         env.close()
                                     model.learning_rate]
                         feed_dict = {model.batch_size: 2,
                                      model.sequence_length: 1,
-                                     model.state_in: np.array([[1, 2, 3, 1, 2, 3],
+                                     model.vector_in: np.array([[1, 2, 3, 1, 2, 3],
-                                     model.observation_in[0]: np.ones([2, 40, 30, 3])
+                                     model.visual_in[0]: np.ones([2, 40, 30, 3])
                                      }
                         sess.run(run_list, feed_dict=feed_dict)
                         env.close()

python/tests/test_unityagents.py

 import json
-import mock
+import unittest.mock as mock
 import pytest
 import struct
 
           "agents": [1,2],
           "vectorObservations": [1,2,3,4,5,6,1,2,3,4,5,6],
           "rewards": [1,2],
-          "vectorActions": [1,2,3,4],
+          "previousVectorActions": [1,2,3,4],
+          "previousTextActions":["",""],
           "memories": [],
           "dones": [false, false],
           "maxes": [false, false],
   "agents": [1,2,3],
   "vectorObservations": [1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9],
   "rewards": [1,2,3],
-  "vectorActions": [1,2,3,4,5,6],
+  "previousVectorActions": [1,2,3,4,5,6],
+  "previousTextActions":["","",""],
   "memories": [],
   "dones": [false, false, false],
   "maxes": [false, false, false],
   "agents": [1,2,3],
   "vectorObservations": [1,2,3,4,5,6,7,8,9,1,2,3,4,5,6,7,8,9],
   "rewards": [1,2,3],
-  "vectorActions": [1,2,3,4,5,6],
+  "previousVectorActions": [1,2,3,4,5,6],
+  "previousTextActions":["","",""],
   "memories": [],
   "dones": [false, false, true],
   "maxes": [false, false, false],

python/tests/test_unitytrainers.py

 import yaml
-import mock
+import unittest.mock as mock
 import pytest
 
 from unitytrainers.trainer_controller import TrainerController
                 mock_socket.return_value.accept.return_value = (mock_socket, 0)
                 mock_socket.recv.return_value.decode.return_value = dummy_start
                 tc = TrainerController(' ', ' ', 1, None, True, True, False, 1,
-                                       1, 1, 1)
+                                       1, 1, 1, '', "tests/test_unitytrainers.py")
-    open_name = '%s.open' % __name__
+    open_name = 'unitytrainers.trainer_controller' + '.open'
-        with mock.patch(open_name, create=True) as mock_open:
-            mock_open.return_value = 0
-            mock_load.return_value = dummy_config
-            config = TrainerController._load_config("tests/test_unitytrainers.py")
-            assert(len(config) == 1)
-            assert(config['default']['trainer'] == "ppo")
+        with mock.patch(open_name, create=True) as _:
+            with mock.patch('subprocess.Popen'):
+                with mock.patch('socket.socket') as mock_socket:
+                    with mock.patch('glob.glob') as mock_glob:
+                        mock_load.return_value = dummy_config
+                        mock_glob.return_value = ['FakeLaunchPath']
+                        mock_socket.return_value.accept.return_value = (mock_socket, 0)
+                        mock_socket.recv.return_value.decode.return_value = dummy_start
+                        mock_load.return_value = dummy_config
+                        tc = TrainerController(' ', ' ', 1, None, True, True, False, 1,
+                                                   1, 1, 1, '','')
+                        config = tc._load_config()
+                        assert(len(config) == 1)
+                        assert(config['default']['trainer'] == "ppo")
-    open_name = '%s.open' % __name__
+    open_name = 'unitytrainers.trainer_controller' + '.open'
-        with mock.patch(open_name, create=True) as mock_open:
-            mock_open.return_value = 0
+        with mock.patch(open_name, create=True) as _:
             with mock.patch('subprocess.Popen'):
                 with mock.patch('socket.socket') as mock_socket:
                     with mock.patch('glob.glob') as mock_glob:
                         tc = TrainerController(' ', ' ', 1, None, True, True, False, 1,
-                                               1, 1, 1)
+                                               1, 1, 1, '', "tests/test_unitytrainers.py")
-                        config = tc._load_config("tests/test_unitytrainers.py")
+                        config = tc._load_config()
                         tf.reset_default_graph()
                         with tf.Session() as sess:
                             tc._initialize_trainers(config, sess)
                         # Test for Behavior Cloning Trainer
                         mock_load.return_value = dummy_bc_config
-                        config = tc._load_config("tests/test_unitytrainers.py")
+                        config = tc._load_config()
                         tf.reset_default_graph()
                         with tf.Session() as sess:
                             tc._initialize_trainers(config, sess)
                         mock_load.return_value = dummy_bad_config
-                        config = tc._load_config("tests/test_unitytrainers.py")
+                        config = tc._load_config()
                         tf.reset_default_graph()
                         with tf.Session() as sess:
                             with pytest.raises(UnityEnvironmentException):
     b = Buffer()
     for fake_agent_id in range(4):
         for step in range(9):
-            b[fake_agent_id]['state'].append(
+            b[fake_agent_id]['vector_observation'].append(
                 [100 * fake_agent_id + 10 * step + 1,
                  100 * fake_agent_id + 10 * step + 2,
                  100 * fake_agent_id + 10 * step + 3]
-    a = b[1]['state'].get_batch(batch_size=2, training_length=None, sequential=True)
+    a = b[1]['vector_observation'].get_batch(batch_size=2, training_length=None, sequential=True)
-    a = b[2]['state'].get_batch(batch_size=2, training_length=3, sequential=True)
+    a = b[2]['vector_observation'].get_batch(batch_size=2, training_length=3, sequential=True)
-    a = b[2]['state'].get_batch(batch_size=2, training_length=3, sequential=False)
+    a = b[2]['vector_observation'].get_batch(batch_size=2, training_length=3, sequential=False)
     assert_array(a, np.array([
         [[251, 252, 253], [261, 262, 263], [271, 272, 273]],
         [[261, 262, 263], [271, 272, 273], [281, 282, 283]]

python/trainer_config.yaml

     buffer_size: 20240
     gamma: 0.995
     max_steps: 1e6
-    summary_freq: 5000
+    summary_freq: 3000
+
+ReacherBrain:
+    normalize: true
+    num_epoch: 3
+    time_horizon: 1000
+    batch_size: 2024
+    buffer_size: 20240
+    gamma: 0.995
+    max_steps: 1e6
+    summary_freq: 3000
 
 BrainWallJumpCC:
     max_steps: 2.0e5
     trainer: imitation
     max_steps: 10000
     summary_freq: 1000
-    brain_to_imitate: ExpertBrain
+    brain_to_imitate: TeacherBrain
     batch_size: 16
     batches_per_epoch: 5
     num_layers: 4

python/unitytrainers/bc/models.py

         if brain.vector_action_space_type == "discrete":
             self.action_probs = tf.nn.softmax(self.policy)
             self.sample_action = tf.cast(tf.multinomial(self.policy, 1, name="action"), tf.int32)
-            self.true_action = tf.placeholder(shape=[None], dtype=tf.int32, name="expert_action")
+            self.true_action = tf.placeholder(shape=[None], dtype=tf.int32, name="teacher_action")
             self.action_oh = tf.one_hot(self.true_action, self.a_size)
             self.loss = tf.reduce_sum(-tf.log(self.action_probs + 1e-10) * self.action_oh)
             self.action_percent = tf.reduce_mean(tf.cast(
-            self.true_action = tf.placeholder(shape=[None, self.a_size], dtype=tf.float32, name="expert_action")
+            self.true_action = tf.placeholder(shape=[None, self.a_size], dtype=tf.float32, name="teacher_action")
             self.loss = tf.reduce_sum(tf.squared_difference(self.true_action, self.sample_action))
 
         optimizer = tf.train.AdamOptimizer(learning_rate=lr)

python/unitytrainers/bc/trainer.py

         run_list = [self.model.sample_action]
         if self.use_observations:
             for i, _ in enumerate(agent_brain.visual_observations):
-                feed_dict[self.model.observation_in[i]] = agent_brain.visual_observations[i]
+                feed_dict[self.model.visual_in[i]] = agent_brain.visual_observations[i]
-            feed_dict[self.model.state_in] = agent_brain.vector_observations
+            feed_dict[self.model.vector_in] = agent_brain.vector_observations
         if self.use_recurrent:
             if agent_brain.memories.shape[1] == 0:
                 agent_brain.memories = np.zeros((len(agent_brain.agents), self.m_size))
         :param take_action_outputs: The outputs of the take action method.
         """
 
-        # Used to collect expert experience into training buffer
-        info_expert = curr_info[self.brain_to_imitate]
-        next_info_expert = next_info[self.brain_to_imitate]
-        for agent_id in info_expert.agents:
-            self.training_buffer[agent_id].last_brain_info = info_expert
+        # Used to collect teacher experience into training buffer
+        info_teacher = curr_info[self.brain_to_imitate]
+        next_info_teacher = next_info[self.brain_to_imitate]
+        for agent_id in info_teacher.agents:
+            self.training_buffer[agent_id].last_brain_info = info_teacher
-        for agent_id in next_info_expert.agents:
-            stored_info_expert = self.training_buffer[agent_id].last_brain_info
-            if stored_info_expert is None:
+        for agent_id in next_info_teacher.agents:
+            stored_info_teacher = self.training_buffer[agent_id].last_brain_info
+            if stored_info_teacher is None:
-                idx = stored_info_expert.agents.index(agent_id)
-                next_idx = next_info_expert.agents.index(agent_id)
-                info_expert_record, info_expert_reset = info_expert.text_observations[idx].lower().split(",")
-                next_info_expert_record, next_info_expert_reset = next_info_expert.text_observations[idx].\
-                    lower().split(",")
-                if next_info_expert_reset == "true":
-                    self.training_buffer.reset_update_buffer()
-
-                if info_expert_record == "true" and next_info_expert_record == "true":
-                    if not stored_info_expert.local_done[idx]:
+                idx = stored_info_teacher.agents.index(agent_id)
+                next_idx = next_info_teacher.agents.index(agent_id)
+                if info_teacher.text_observations[idx] != "":
+                    info_teacher_record, info_teacher_reset = info_teacher.text_observations[idx].lower().split(",")
+                    next_info_teacher_record, next_info_teacher_reset = next_info_teacher.text_observations[idx].\
+                        lower().split(",")
+                    if next_info_teacher_reset == "true":
+                        self.training_buffer.reset_update_buffer()
+                else:
+                    info_teacher_record, next_info_teacher_record = "true", "true"
+                if info_teacher_record == "true" and next_info_teacher_record == "true":
+                    if not stored_info_teacher.local_done[idx]:
-                            for i, _ in enumerate(stored_info_expert.visual_observations):
+                            for i, _ in enumerate(stored_info_teacher.visual_observations):
-                                    .append(stored_info_expert.visual_observations[i][idx])
+                                    .append(stored_info_teacher.visual_observations[i][idx])
-                                .append(stored_info_expert.vector_observations[idx])
+                                .append(stored_info_teacher.vector_observations[idx])
-                            if stored_info_expert.memories.shape[1] == 0:
-                                stored_info_expert.memories = np.zeros((len(stored_info_expert.agents), self.m_size))
-                            self.training_buffer[agent_id]['memory'].append(stored_info_expert.memories[idx])
-                        self.training_buffer[agent_id]['actions'].append(next_info_expert.previous_vector_actions[next_idx])
+                            if stored_info_teacher.memories.shape[1] == 0:
+                                stored_info_teacher.memories = np.zeros((len(stored_info_teacher.agents), self.m_size))
+                            self.training_buffer[agent_id]['memory'].append(stored_info_teacher.memories[idx])
+                        self.training_buffer[agent_id]['actions'].append(next_info_teacher.
+                                                                         previous_vector_actions[next_idx])
         info_student = curr_info[self.brain_name]
         next_info_student = next_info[self.brain_name]
         for agent_id in info_student.agents:
         Processing involves calculating value and advantage targets for model updating step.
         :param info: Current AllBrainInfo
         """
-        info_expert = info[self.brain_to_imitate]
-        for l in range(len(info_expert.agents)):
-            if ((info_expert.local_done[l] or
-                 len(self.training_buffer[info_expert.agents[l]]['actions']) > self.trainer_parameters[
+        info_teacher = info[self.brain_to_imitate]
+        for l in range(len(info_teacher.agents)):
+            if ((info_teacher.local_done[l] or
+                 len(self.training_buffer[info_teacher.agents[l]]['actions']) > self.trainer_parameters[
-                    and len(self.training_buffer[info_expert.agents[l]]['actions']) > 0):
-                agent_id = info_expert.agents[l]
+                    and len(self.training_buffer[info_teacher.agents[l]]['actions']) > 0):
+                agent_id = info_teacher.agents[l]
                 self.training_buffer.append_update_buffer(agent_id, batch_size=None,
                                                           training_length=self.sequence_length)
                 self.training_buffer[agent_id].reset_agent()
                          self.model.batch_size: self.n_sequences,
                          self.model.sequence_length: self.sequence_length}
             if not self.is_continuous:
-                feed_dict[self.model.state_in] = batch_states.reshape([-1, 1])
+                feed_dict[self.model.vector_in] = batch_states.reshape([-1, 1])
-                feed_dict[self.model.state_in] = batch_states.reshape([-1, self.brain.vector_observation_space_size *
+                feed_dict[self.model.vector_in] = batch_states.reshape([-1, self.brain.vector_observation_space_size *
-                for i, _ in enumerate(self.model.observation_in):
+                for i, _ in enumerate(self.model.visual_in):
-                    feed_dict[self.model.observation_in[i]] = _obs.reshape([-1, _w, _h, _c])
+                    feed_dict[self.model.visual_in[i]] = _obs.reshape([-1, _w, _h, _c])
             if self.use_recurrent:
                 feed_dict[self.model.memory_in] = np.zeros([self.n_sequences, self.m_size])
 

python/unitytrainers/models.py

 class LearningModel(object):
     def __init__(self, m_size, normalize, use_recurrent, brain):
         self.brain = brain
-        self.state_in = None
+        self.vector_in = None
-        self.observation_in = []
+        self.visual_in = []
         self.batch_size = tf.placeholder(shape=None, dtype=tf.int32, name='batch_size')
         self.sequence_length = tf.placeholder(shape=None, dtype=tf.int32, name='sequence_length')
         self.m_size = m_size
         else:
             c_channels = 3
 
-        observation_in = tf.placeholder(shape=[None, o_size_h, o_size_w, c_channels], dtype=tf.float32, name=name)
-        return observation_in
+        visual_in = tf.placeholder(shape=[None, o_size_h, o_size_w, c_channels], dtype=tf.float32, name=name)
+        return visual_in
-            self.state_in = tf.placeholder(shape=[None, s_size], dtype=tf.float32, name='state')
+            self.vector_in = tf.placeholder(shape=[None, s_size], dtype=tf.float32, name='vector_observation')
             if self.normalize:
                 self.running_mean = tf.get_variable("running_mean", [s_size], trainable=False, dtype=tf.float32,
                                                     initializer=tf.zeros_initializer())
                 self.update_mean = tf.assign(self.running_mean, self.new_mean)
                 self.update_variance = tf.assign(self.running_variance, self.new_variance)
 
-                self.normalized_state = tf.clip_by_value((self.state_in - self.running_mean) / tf.sqrt(
+                self.normalized_state = tf.clip_by_value((self.vector_in - self.running_mean) / tf.sqrt(
-                self.normalized_state = self.state_in
+                self.normalized_state = self.vector_in
-            self.state_in = tf.placeholder(shape=[None, 1], dtype=tf.int32, name='state')
+            self.vector_in = tf.placeholder(shape=[None, 1], dtype=tf.int32, name='vector_observation')
 
     def create_continuous_state_encoder(self, h_size, activation, num_layers):
         """
         :param num_layers: number of hidden layers to create.
         :return: List of hidden layer tensors.
         """
-        conv1 = tf.layers.conv2d(self.observation_in[-1], 16, kernel_size=[8, 8], strides=[4, 4],
+        conv1 = tf.layers.conv2d(self.visual_in[-1], 16, kernel_size=[8, 8], strides=[4, 4],
                                  activation=tf.nn.elu)
         conv2 = tf.layers.conv2d(conv1, 32, kernel_size=[4, 4], strides=[2, 2],
                                  activation=tf.nn.elu)
         :param num_layers: number of hidden layers to create.
         :return: List of hidden layer tensors.
         """
-        state_in = tf.reshape(self.state_in, [-1])
-        state_onehot = c_layers.one_hot_encoding(state_in, s_size)
+        vector_in = tf.reshape(self.vector_in, [-1])
+        state_onehot = c_layers.one_hot_encoding(vector_in, s_size)
         hidden = state_onehot
         for j in range(num_layers):
             hidden = tf.layers.dense(hidden, h_size, use_bias=False, activation=activation)
         else:
             activation_fn = self.swish
 
-        self.observation_in = []
+        self.visual_in = []
-            self.observation_in.append(visual_input)
+            self.visual_in.append(visual_input)
         self.create_vector_input(s_size)
 
         final_hiddens = []
         _half_point = int(m_size / 2)
         with tf.variable_scope(name):
             rnn_cell = tf.contrib.rnn.BasicLSTMCell(_half_point)
-            lstm_state_in = tf.contrib.rnn.LSTMStateTuple(memory_in[:, :_half_point], memory_in[:, _half_point:])
+            lstm_vector_in = tf.contrib.rnn.LSTMStateTuple(memory_in[:, :_half_point], memory_in[:, _half_point:])
-                                                                initial_state=lstm_state_in,
+                                                                initial_state=lstm_vector_in,
                                                                 time_major=False,
                                                                 dtype=tf.float32)
 

python/unitytrainers/ppo/models.py

         appropriate PPO agent model for the environment.
         :param brain: BrainInfo used to generate specific network graph.
         :param lr: Learning rate.
-        :param h_size: Size of hidden layers/
+        :param h_size: Size of hidden layers
         :param epsilon: Value for policy-divergence threshold.
         :param beta: Strength of entropy regularization.
         :return: a sub-class of PPOAgent tailored to the environment.

python/unitytrainers/ppo/trainer.py

             feed_dict[self.model.prev_action] = np.reshape(curr_brain_info.previous_vector_actions, [-1])
         if self.use_observations:
             for i, _ in enumerate(curr_brain_info.visual_observations):
-                feed_dict[self.model.observation_in[i]] = curr_brain_info.visual_observations[i]
+                feed_dict[self.model.visual_in[i]] = curr_brain_info.visual_observations[i]
-            feed_dict[self.model.state_in] = curr_brain_info.vector_observations
+            feed_dict[self.model.vector_in] = curr_brain_info.vector_observations
         if self.use_recurrent:
             if curr_brain_info.memories.shape[1] == 0:
                 curr_brain_info.memories = np.zeros((len(curr_brain_info.agents), self.m_size))
                     feed_dict = {self.model.batch_size: len(info.vector_observations), self.model.sequence_length: 1}
                     if self.use_observations:
                         for i in range(len(info.visual_observations)):
-                            feed_dict[self.model.observation_in[i]] = info.visual_observations[i]
+                            feed_dict[self.model.visual_in[i]] = info.visual_observations[i]
-                        feed_dict[self.model.state_in] = info.vector_observations
+                        feed_dict[self.model.vector_in] = info.vector_observations
                     if self.use_recurrent:
                         if info.memories.shape[1] == 0:
                             info.memories = np.zeros((len(info.vector_observations), self.m_size))
                             _buffer['prev_action'][start:end]).reshape([-1])
                 if self.use_states:
                     if self.brain.vector_observation_space_type == "continuous":
-                        feed_dict[self.model.state_in] = np.array(
+                        feed_dict[self.model.vector_in] = np.array(
-                        feed_dict[self.model.state_in] = np.array(
+                        feed_dict[self.model.vector_in] = np.array(
-                    for i, _ in enumerate(self.model.observation_in):
+                    for i, _ in enumerate(self.model.visual_in):
-                        feed_dict[self.model.observation_in[i]] = _obs.reshape([-1, _w, _h, _c])
+                        feed_dict[self.model.visual_in[i]] = _obs.reshape([-1, _w, _h, _c])
                 # Memories are zeros
                 if self.use_recurrent:
                     # feed_dict[self.model.memory_in] = np.zeros([batch_size, self.m_size])

unity-environment/Assets/ML-Agents/Examples/3DBall/Scripts/Ball3DAgent.cs

 
     public override void AgentAction(float[] act)
     {
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             float action_z = 2f * Mathf.Clamp(act[0], -1f, 1f);
             if ((gameObject.transform.rotation.z < 0.25f && action_z > 0f) ||

unity-environment/Assets/ML-Agents/Examples/3DBall/Scripts/Ball3DDecision.cs

 
     public float[] Decide(List<float> state, List<Texture2D> observation, float reward, bool done, List<float> memory)
     {
-        if (gameObject.GetComponent<Brain>().brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (gameObject.GetComponent<Brain>().brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             List<float> act = new List<float>();
 

unity-environment/Assets/ML-Agents/Examples/3DBall/Scripts/Ball3DHardAgent.cs

 
     public override void AgentAction(float[] act)
     {
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             float action_z = 2f * Mathf.Clamp(act[0], -1f, 1f);
             if ((gameObject.transform.rotation.z < 0.25f && action_z > 0f) ||

unity-environment/Assets/ML-Agents/Examples/Area/Scripts/AreaAgent.cs

         float directionZ = 0;
         float directionY = 0;
 
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             directionX = Mathf.Clamp(act[0], -1f, 1f);
             directionZ = Mathf.Clamp(act[1], -1f, 1f);

unity-environment/Assets/ML-Agents/Examples/Banana/BananaImitation.unity

   - component: {fileID: 7561199}
   - component: {fileID: 7561200}
   m_Layer: 0
-  m_Name: ExpertBrain
+  m_Name: TeacherBrain
   m_TagString: Untagged
   m_Icon: {fileID: 0}
   m_NavMeshLayer: 0
     vectorObservationSpaceType: 1
   brainType: 0
   CoreBrains:
-  - {fileID: 29328243}
-  - {fileID: 687699474}
-  - {fileID: 2082760005}
-  - {fileID: 728415826}
-  instanceID: 53712
+  - {fileID: 732354759}
+  - {fileID: 1117344192}
+  - {fileID: 714235400}
+  - {fileID: 2091631578}
+  instanceID: 12494
 --- !u!114 &7561200
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_Script: {fileID: 11500000, guid: c0205d4e23c794363ba1d78b1e3d7339, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
---- !u!114 &29328243
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 41e9bda8f3cf1492fa74926a530f6f70, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  broadcast: 1
-  continuousPlayerActions:
-  - key: 119
-    index: 0
-    value: 1
-  - key: 97
-    index: 1
-    value: -1
-  - key: 100
-    index: 1
-    value: 1
-  - key: 32
-    index: 2
-    value: 1
-  - key: 115
-    index: 0
-    value: -1
-  discretePlayerActions:
-  - key: 119
-    value: 1
-  - key: 32
-    value: 2
-  - key: 97
-    value: 3
-  - key: 100
-    value: 4
-  defaultAction: 0
-  brain: {fileID: 7561199}
 --- !u!1001 &85500286
 Prefab:
   m_ObjectHideFlags: 0
   m_PrefabParentObject: {fileID: 4978301839030222, guid: c7684b8f068694edb9b432c4a427e36d,
     type: 2}
   m_PrefabInternal: {fileID: 93417038}
+--- !u!114 &154237818
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 35813a1be64e144f887d7d5f15b963fa, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  brain: {fileID: 2015024680}
 --- !u!1 &192430538
 GameObject:
   m_ObjectHideFlags: 0
   m_Father: {fileID: 0}
   m_RootOrder: 3
   m_LocalEulerAnglesHint: {x: 0, y: 0, z: 0}
---- !u!114 &545810891
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 943466ab374444748a364f9d6c3e2fe2, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  broadcast: 1
-  brain: {fileID: 7561199}
---- !u!114 &687699474
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 943466ab374444748a364f9d6c3e2fe2, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  broadcast: 1
-  brain: {fileID: 7561199}
---- !u!114 &728415826
+--- !u!114 &523165459
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_PrefabParentObject: {fileID: 0}
   m_EditorHideFlags: 0
   m_Script: {fileID: 11500000, guid: 8b23992c8eb17439887f5e944bf04a40, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
   m_EditorClassIdentifier: 
   broadcast: 1
   graphModel: {fileID: 4900000, guid: 2671f35d7452742cb9f019fb5e30d682, type: 3}
     minValue: -0.5
     maxValue: 0.5
   BatchSizePlaceholderName: batch_size
-  StatePlacholderName: state
+  VectorObservationPlacholderName: state
-  ObservationPlaceholderName: []
+  VisualObservationPlaceholderName: []
---- !u!114 &753967989
+--- !u!114 &714235400
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_PrefabParentObject: {fileID: 0}
   m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 8b23992c8eb17439887f5e944bf04a40, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_Script: {fileID: 11500000, guid: 35813a1be64e144f887d7d5f15b963fa, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  brain: {fileID: 7561199}
+--- !u!114 &732354759
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 41e9bda8f3cf1492fa74926a530f6f70, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  graphModel: {fileID: 4900000, guid: 2671f35d7452742cb9f019fb5e30d682, type: 3}
-  graphScope: 
-  graphPlaceholders:
-  - name: epsilon
-    valueType: 1
-    minValue: -0.5
-    maxValue: 0.5
-  BatchSizePlaceholderName: batch_size
-  StatePlacholderName: state
-  RecurrentInPlaceholderName: recurrent_in
-  RecurrentOutPlaceholderName: recurrent_out
-  ObservationPlaceholderName: []
-  ActionPlaceholderName: action
-  PreviousActionPlaceholderName: prev_action
+  continuousPlayerActions:
+  - key: 119
+    index: 0
+    value: 1
+  - key: 97
+    index: 1
+    value: -1
+  - key: 100
+    index: 1
+    value: 1
+  - key: 32
+    index: 2
+    value: 1
+  - key: 115
+    index: 0
+    value: -1
+  discretePlayerActions:
+  - key: 119
+    value: 1
+  - key: 32
+    value: 2
+  - key: 97
+    value: 3
+  - key: 100
+    value: 4
+  defaultAction: 0
   brain: {fileID: 7561199}
 --- !u!1 &762086410
 GameObject:
   m_Father: {fileID: 0}
   m_RootOrder: 0
   m_LocalEulerAnglesHint: {x: 45, y: 45, z: 0}
+--- !u!114 &829097546
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 943466ab374444748a364f9d6c3e2fe2, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  broadcast: 1
+  brain: {fileID: 7561199}
 --- !u!1 &844811898
 GameObject:
   m_ObjectHideFlags: 0
   m_Father: {fileID: 1812954964}
   m_RootOrder: 0
   m_LocalEulerAnglesHint: {x: 0, y: 90, z: 0}
+--- !u!114 &963921654
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 41e9bda8f3cf1492fa74926a530f6f70, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  broadcast: 1
+  continuousPlayerActions:
+  - key: 119
+    index: 0
+    value: 1
+  - key: 97
+    index: 1
+    value: -1
+  - key: 100
+    index: 1
+    value: 1
+  - key: 32
+    index: 2
+    value: 1
+  - key: 115
+    index: 0
+    value: -1
+  discretePlayerActions:
+  - key: 119
+    value: 1
+  - key: 32
+    value: 2
+  - key: 97
+    value: 3
+  - key: 100
+    value: 4
+  defaultAction: 0
+  brain: {fileID: 7561199}
 --- !u!1001 &991769960
 Prefab:
   m_ObjectHideFlags: 0
       propertyPath: agentParameters.numberOfActionsBetweenDecisions
       value: 1
       objectReference: {fileID: 0}
+    - target: {fileID: 1023049942000704, guid: de088c15c5e19472981198726aca0baa, type: 2}
+      propertyPath: m_Name
+      value: TeacherAgent
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: de088c15c5e19472981198726aca0baa, type: 2}
   m_IsPrefabParent: 0
   m_Script: {fileID: 11500000, guid: d1cf16abc39fb4d6ca81222fc73d1bb5, type: 3}
   m_Name: 
   m_EditorClassIdentifier: 
+--- !u!114 &1117344192
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 943466ab374444748a364f9d6c3e2fe2, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  broadcast: 1
+  brain: {fileID: 7561199}
 --- !u!1 &1196437247
 GameObject:
   m_ObjectHideFlags: 0
   m_Father: {fileID: 0}
   m_RootOrder: 2
   m_LocalEulerAnglesHint: {x: 0, y: 0, z: 0}
---- !u!114 &1669781038
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 41e9bda8f3cf1492fa74926a530f6f70, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  broadcast: 1
-  continuousPlayerActions:
-  - key: 119
-    index: 0
-    value: 1
-  - key: 97
-    index: 1
-    value: -1
-  - key: 100
-    index: 1
-    value: 1
-  - key: 32
-    index: 2
-    value: 1
-  - key: 115
-    index: 0
-    value: -1
-  discretePlayerActions:
-  - key: 119
-    value: 1
-  - key: 32
-    value: 2
-  - key: 97
-    value: 3
-  - key: 100
-    value: 4
-  defaultAction: 0
-  brain: {fileID: 7561199}
 --- !u!1 &1799584680
 GameObject:
   m_ObjectHideFlags: 0
     vectorObservationSpaceType: 1
   brainType: 2
   CoreBrains:
-  - {fileID: 1669781038}
-  - {fileID: 545810891}
-  - {fileID: 2144941368}
-  - {fileID: 753967989}
-  instanceID: 54598
+  - {fileID: 963921654}
+  - {fileID: 829097546}
+  - {fileID: 154237818}
+  - {fileID: 523165459}
+  instanceID: 12662
 --- !u!1 &2038164702
 GameObject:
   m_ObjectHideFlags: 0
   m_PrefabInternal: {fileID: 0}
   m_GameObject: {fileID: 2038164702}
   m_Mesh: {fileID: 10202, guid: 0000000000000000e000000000000000, type: 0}
---- !u!114 &2082760005
+--- !u!114 &2091631578
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_PrefabParentObject: {fileID: 0}
   m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 35813a1be64e144f887d7d5f15b963fa, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_Script: {fileID: 11500000, guid: 8b23992c8eb17439887f5e944bf04a40, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  broadcast: 1
+  graphModel: {fileID: 4900000, guid: 2671f35d7452742cb9f019fb5e30d682, type: 3}
+  graphScope: 
+  graphPlaceholders:
+  - name: epsilon
+    valueType: 1
+    minValue: -0.5
+    maxValue: 0.5
+  BatchSizePlaceholderName: batch_size
+  VectorObservationPlacholderName: state
+  RecurrentInPlaceholderName: recurrent_in
+  RecurrentOutPlaceholderName: recurrent_out
+  VisualObservationPlaceholderName: []
+  ActionPlaceholderName: action
+  PreviousActionPlaceholderName: prev_action
---- !u!114 &2144941368
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 35813a1be64e144f887d7d5f15b963fa, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  brain: {fileID: 2015024680}

unity-environment/Assets/ML-Agents/Examples/Hallway/Scripts/HallwayAgent.cs

         Vector3 dirToGo = Vector3.zero;
         Vector3 rotateDir = Vector3.zero;
 
+
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             dirToGo = transform.forward * Mathf.Clamp(act[0], -1f, 1f);
             rotateDir = transform.up * Mathf.Clamp(act[1], -1f, 1f);

unity-environment/Assets/ML-Agents/Examples/Reacher/Scene.unity

 --- !u!104 &2
 RenderSettings:
   m_ObjectHideFlags: 0
-  serializedVersion: 8
+  serializedVersion: 9
   m_Fog: 0
   m_FogColor: {r: 0.5, g: 0.5, b: 0.5, a: 1}
   m_FogMode: 3
   m_ReflectionIntensity: 1
   m_CustomReflection: {fileID: 0}
   m_Sun: {fileID: 762086411}
-  m_IndirectSpecularColor: {r: 0.4465785, g: 0.49641252, b: 0.574817, a: 1}
+  m_IndirectSpecularColor: {r: 0.4465934, g: 0.49642956, b: 0.5748249, a: 1}
+  m_UseRadianceAmbientProbe: 0
 --- !u!157 &3
 LightmapSettings:
   m_ObjectHideFlags: 0
     m_EnableBakedLightmaps: 1
     m_EnableRealtimeLightmaps: 1
   m_LightmapEditorSettings:
-    serializedVersion: 9
+    serializedVersion: 10
-    m_TextureWidth: 1024
-    m_TextureHeight: 1024
+    m_AtlasSize: 1024
     m_AO: 0
     m_AOMaxDistance: 1
     m_CompAOExponent: 1
     m_PVRFilteringAtrousPositionSigmaDirect: 0.5
     m_PVRFilteringAtrousPositionSigmaIndirect: 2
     m_PVRFilteringAtrousPositionSigmaAO: 1
+    m_ShowResolutionOverlay: 1
   m_LightingDataAsset: {fileID: 0}
   m_UseShadowmask: 1
 --- !u!196 &4
       propertyPath: m_Name
       value: Agent (1)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   m_PrefabParentObject: {fileID: 1644872085946016, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
     type: 2}
   m_PrefabInternal: {fileID: 103720292}
+--- !u!114 &105498462
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 41e9bda8f3cf1492fa74926a530f6f70, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  broadcast: 1
+  continuousPlayerActions: []
+  discretePlayerActions: []
+  defaultAction: -1
+  brain: {fileID: 846768605}
+--- !u!114 &169591833
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 8b23992c8eb17439887f5e944bf04a40, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  broadcast: 1
+  graphModel: {fileID: 4900000, guid: 8db6173148a6f4e7fa654ed627c88d7a, type: 3}
+  graphScope: 
+  graphPlaceholders: []
+  BatchSizePlaceholderName: batch_size
+  VectorObservationPlacholderName: state
+  RecurrentInPlaceholderName: recurrent_in
+  RecurrentOutPlaceholderName: recurrent_out
+  VisualObservationPlaceholderName: []
+  ActionPlaceholderName: action
+  PreviousActionPlaceholderName: prev_action
+  brain: {fileID: 846768605}
 --- !u!1001 &201192304
 Prefab:
   m_ObjectHideFlags: 0
       propertyPath: brain
       value: 
       objectReference: {fileID: 846768605}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (13)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (22)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   m_PrefabParentObject: {fileID: 1644872085946016, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
     type: 2}
   m_PrefabInternal: {fileID: 269521201}
---- !u!114 &327113840
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 41e9bda8f3cf1492fa74926a530f6f70, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  broadcast: 1
-  continuousPlayerActions: []
-  discretePlayerActions: []
-  defaultAction: -1
-  brain: {fileID: 846768605}
 --- !u!1 &330647278
 GameObject:
   m_ObjectHideFlags: 0
   m_MotionVectors: 1
   m_LightProbeUsage: 1
   m_ReflectionProbeUsage: 1
+  m_RenderingLayerMask: 4294967295
   m_Materials:
   - {fileID: 2100000, guid: 3736de91af62e4be7a3d8752592c6c61, type: 2}
   m_StaticBatchInfo:
       propertyPath: m_Name
       value: Agent (20)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (31)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (24)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (12)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (5)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (9)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   m_PrefabParentObject: {fileID: 1644872085946016, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
     type: 2}
   m_PrefabInternal: {fileID: 547117383}
---- !u!114 &551611617
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 8b23992c8eb17439887f5e944bf04a40, type: 3}
-  m_Name: (Clone)(Clone)
-  m_EditorClassIdentifier: 
-  broadcast: 1
-  graphModel: {fileID: 4900000, guid: 8db6173148a6f4e7fa654ed627c88d7a, type: 3}
-  graphScope: 
-  graphPlaceholders:
-  - name: epsilon
-    valueType: 1
-    minValue: -1
-    maxValue: 1
-  BatchSizePlaceholderName: batch_size
-  StatePlacholderName: state
-  RecurrentInPlaceholderName: recurrent_in
-  RecurrentOutPlaceholderName: recurrent_out
-  ObservationPlaceholderName: []
-  ActionPlaceholderName: action
-  brain: {fileID: 846768605}
 --- !u!1001 &591860098
 Prefab:
   m_ObjectHideFlags: 0
     - target: {fileID: 1395682910799436, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
       propertyPath: m_Name
       value: Agent (3)
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
       objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
       propertyPath: m_Name
       value: Agent (18)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (8)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (15)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (19)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   - component: {fileID: 846768604}
   - component: {fileID: 846768605}
   m_Layer: 0
-  m_Name: Brain
+  m_Name: ReacherBrain
   m_TagString: Untagged
   m_Icon: {fileID: 0}
   m_NavMeshLayer: 0
   m_Name: 
   m_EditorClassIdentifier: 
   brainParameters:
-    vectorObservationSize: 26
-    numStackedVectorObservations: 1
+    vectorObservationSize: 24
+    numStackedVectorObservations: 3
     vectorActionSize: 4
     cameraResolutions: []
     vectorActionDescriptions:
     - 
     vectorActionSpaceType: 1
     vectorObservationSpaceType: 1
-  brainType: 0
+  brainType: 3
-  - {fileID: 327113840}
-  - {fileID: 1028133595}
-  - {fileID: 1146467767}
-  - {fileID: 551611617}
-  instanceID: 95334
+  - {fileID: 105498462}
+  - {fileID: 949391681}
+  - {fileID: 1042778401}
+  - {fileID: 169591833}
+  instanceID: 3972
 --- !u!1001 &915173805
 Prefab:
   m_ObjectHideFlags: 0
       propertyPath: m_Name
       value: Agent (14)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   m_PrefabParentObject: {fileID: 1644872085946016, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
     type: 2}
   m_PrefabInternal: {fileID: 915173805}
---- !u!114 &1028133595
+--- !u!114 &949391681
 MonoBehaviour:
   m_ObjectHideFlags: 0
   m_PrefabParentObject: {fileID: 0}
   m_EditorHideFlags: 0
   m_Script: {fileID: 11500000, guid: 943466ab374444748a364f9d6c3e2fe2, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+--- !u!114 &1042778401
+MonoBehaviour:
+  m_ObjectHideFlags: 0
+  m_PrefabParentObject: {fileID: 0}
+  m_PrefabInternal: {fileID: 0}
+  m_GameObject: {fileID: 0}
+  m_Enabled: 1
+  m_EditorHideFlags: 0
+  m_Script: {fileID: 11500000, guid: 35813a1be64e144f887d7d5f15b963fa, type: 3}
+  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
+  m_EditorClassIdentifier: 
+  brain: {fileID: 846768605}
 --- !u!1001 &1065008966
 Prefab:
   m_ObjectHideFlags: 0
       propertyPath: m_Name
       value: Agent (11)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   m_PrefabParentObject: {fileID: 1644872085946016, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
     type: 2}
   m_PrefabInternal: {fileID: 1065008966}
---- !u!114 &1146467767
-MonoBehaviour:
-  m_ObjectHideFlags: 0
-  m_PrefabParentObject: {fileID: 0}
-  m_PrefabInternal: {fileID: 0}
-  m_GameObject: {fileID: 0}
-  m_Enabled: 1
-  m_EditorHideFlags: 0
-  m_Script: {fileID: 11500000, guid: 35813a1be64e144f887d7d5f15b963fa, type: 3}
-  m_Name: (Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)(Clone)
-  m_EditorClassIdentifier: 
-  brain: {fileID: 846768605}
 --- !u!1001 &1147289133
 Prefab:
   m_ObjectHideFlags: 0
       propertyPath: m_Name
       value: Agent (7)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (26)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (16)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (27)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (17)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (28)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (6)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
   trainingConfiguration:
     width: 80
     height: 80
-    qualityLevel: 1
+    qualityLevel: 0
     timeScale: 100
     targetFrameRate: 60
   inferenceConfiguration:
       propertyPath: m_Name
       value: Agent (23)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
     - target: {fileID: 1395682910799436, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
       propertyPath: m_Name
       value: Agent (21)
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
       objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
       propertyPath: m_Name
       value: Agent (2)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (10)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (30)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (29)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
       propertyPath: m_Name
       value: Agent (4)
       objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
+      objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
   m_IsPrefabParent: 0
     - target: {fileID: 1395682910799436, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}
       propertyPath: m_Name
       value: Agent (25)
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.maxStep
+      value: 4000
+      objectReference: {fileID: 0}
+    - target: {fileID: 114955921823023820, guid: 2f13abef2db804f96bdc7692a1dcf2b2,
+        type: 2}
+      propertyPath: agentParameters.numberOfActionsBetweenDecisions
+      value: 4
       objectReference: {fileID: 0}
     m_RemovedComponents: []
   m_ParentPrefab: {fileID: 100100000, guid: 2f13abef2db804f96bdc7692a1dcf2b2, type: 2}

unity-environment/Assets/ML-Agents/Examples/Reacher/Scripts/FlyCamera.cs

 {
 
     /*
-    Writen by Windexglow 11-13-10.  Use it, edit it, steal it I don't care.  
-    Converted to C# 27-02-13 - no credit wanted.
-    Simple flycam I made, since I couldn't find any others made public.  
-    Made simple to use (drag and drop, done) for regular keyboard layout  
     wasd : basic movement
     shift : Makes camera accelerate
     space : Moves camera on X and Z axis only.  So camera doesn't gain any height*/

unity-environment/Assets/ML-Agents/Examples/Reacher/Scripts/ReacherAgent.cs

-using System.Collections;
-using System.Collections.Generic;
-using UnityEngine;
+using UnityEngine;
 
 public class ReacherAgent : Agent {
 
     Rigidbody rbB;
     float goalSpeed;
 
+    /// <summary>
+    /// Collect the rigidbodies of the reacher in order to resue them for 
+    /// observations and actions.
+    /// </summary>
     public override void InitializeAgent()
     {
         rbA = pendulumA.GetComponent<Rigidbody>();
+    /// <summary>
+    /// We collect the normalized rotations, angularal velocities, and velocities of both
+    /// limbs of the reacher as well as the relative position of the target and hand.
+    /// </summary>
-        AddVectorObs(pendulumA.transform.rotation.x);
-        AddVectorObs(pendulumA.transform.rotation.y);
-        AddVectorObs(pendulumA.transform.rotation.z);
-        AddVectorObs(pendulumA.transform.rotation.w);
+        AddVectorObs(pendulumA.transform.rotation.eulerAngles.x / 180.0f - 1.0f);
+        AddVectorObs(pendulumA.transform.rotation.eulerAngles.y / 180.0f - 1.0f);
+        AddVectorObs(pendulumA.transform.rotation.eulerAngles.z / 180.0f - 1.0f);
         AddVectorObs(rbA.angularVelocity.x);
         AddVectorObs(rbA.angularVelocity.y);
         AddVectorObs(rbA.angularVelocity.z);
 
-        AddVectorObs(pendulumB.transform.rotation.x);
-        AddVectorObs(pendulumB.transform.rotation.y);
-        AddVectorObs(pendulumB.transform.rotation.z);
-        AddVectorObs(pendulumB.transform.rotation.w);
+        AddVectorObs(pendulumB.transform.rotation.eulerAngles.x / 180.0f - 1.0f);
+        AddVectorObs(pendulumB.transform.rotation.eulerAngles.y / 180.0f - 1.0f);
+        AddVectorObs(pendulumB.transform.rotation.eulerAngles.z / 180.0f - 1.0f);
         AddVectorObs(rbB.angularVelocity.x);
         AddVectorObs(rbB.angularVelocity.y);
         AddVectorObs(rbB.angularVelocity.z);
 
-        AddVectorObs(goal.transform.position.x - transform.position.x);
-        AddVectorObs(goal.transform.position.y - transform.position.y);
-        AddVectorObs(goal.transform.position.z - transform.position.z);
-
-        AddVectorObs(hand.transform.position.x - transform.position.x);
-        AddVectorObs(hand.transform.position.y - transform.position.y);
-        AddVectorObs(hand.transform.position.z - transform.position.z);
-
+        Vector3 localGoalPosition = goal.transform.position - transform.position;
+        AddVectorObs(localGoalPosition.x);
+        AddVectorObs(localGoalPosition.y);
+        AddVectorObs(localGoalPosition.z);
-    }
+        Vector3 localHandPosition = hand.transform.position - transform.position;
+        AddVectorObs(localHandPosition.x);
+        AddVectorObs(localHandPosition.y);
+        AddVectorObs(localHandPosition.z);
+	}
+    /// <summary>
+    /// The agent's four actions correspond to torques on each of the two joints.
+    /// </summary>
     public override void AgentAction(float[] act)
     {
         goalDegree += goalSpeed;
         torque_x = Mathf.Clamp(act[2], -1, 1) * 100f;
         torque_z = Mathf.Clamp(act[3], -1, 1) * 100f;
         rbB.AddTorque(new Vector3(torque_x, 0f, torque_z));
-
-    }
+	}
+    /// <summary>
+    /// Used to move the position of the target goal around the agent.
+    /// </summary>
     void UpdateGoalPosition() {
         float radians = (goalDegree * Mathf.PI) / 180f;
         float goalX = 8f * Mathf.Cos(radians);
     }
 
-
+    /// <summary>
+    /// Resets the position and velocity of the agent and the goal.
+    /// </summary>
     public override void AgentReset()
     {
         pendulumA.transform.position = new Vector3(0f, -4f, 0f) + transform.position;
         goalSpeed = academy.goalSpeed;
 
         goal.transform.localScale = new Vector3(goalSize, goalSize, goalSize);
-    }
-
-    public override void AgentOnDone()
-    {
-
     }
 }

unity-environment/Assets/ML-Agents/Examples/Reacher/Scripts/ReacherGoal.cs

         }
     }
 
-
-    private void OnTriggerStay(Collider other)
+    private void OnTriggerStay(Collider other) 
-            agent.GetComponent<ReacherAgent>().SetReward(  0.1f);
-            //agent.GetComponent<PendulumAgent>().done = true;
+            agent.GetComponent<ReacherAgent>().AddReward(0.01f);
         }
     }
 }

unity-environment/Assets/ML-Agents/Examples/Reacher/TFModels/Reacher.bytes.meta

 fileFormatVersion: 2
 guid: 8db6173148a6f4e7fa654ed627c88d7a
-timeCreated: 1508690694
-licenseType: Pro
+  externalObjects: {}
   userData: 
   assetBundleName: 
   assetBundleVariant: 

unity-environment/Assets/ML-Agents/Scripts/Agent.cs

             return;
 
         BrainParameters param = brain.brainParameters;
-        if (param.vectorActionSpaceType == StateType.continuous)
+        if (param.vectorActionSpaceType == SpaceType.continuous)
         {
             _action.vectorActions = new float[param.vectorActionSize];
             _info.StoredVectorActions = new float[param.vectorActionSize];
         _action.textActions = "";
         _info.memories = new List<float>();
         _action.memories = new List<float>();
-        if (param.vectorObservationSpaceType == StateType.continuous)
+        if (param.vectorObservationSpaceType == SpaceType.continuous)
         {
             _info.vectorObservation =
                 new List<float>(param.vectorObservationSize);
         CollectObservations();
 
         BrainParameters param = brain.brainParameters;
-        if (param.vectorObservationSpaceType == StateType.continuous)
+        if (param.vectorObservationSpaceType == SpaceType.continuous)
         {
             if (_info.vectorObservation.Count != param.vectorObservationSize)
             {

unity-environment/Assets/ML-Agents/Scripts/Brain.cs

     Internal
 }
 
-public enum StateType
+public enum SpaceType
 {
     discrete,
     continuous
     public string[] vectorActionDescriptions;
     /**< \brief The list of strings describing what the actions correpond to */
 
-    public StateType vectorActionSpaceType = StateType.discrete;
+    public SpaceType vectorActionSpaceType = SpaceType.discrete;
-    public StateType vectorObservationSpaceType = StateType.continuous;
+    public SpaceType vectorObservationSpaceType = SpaceType.continuous;
     /**< \brief Defines if the state is discrete or continuous */
 }
 

unity-environment/Assets/ML-Agents/Scripts/CoreBrainInternal.cs

     ///  Modify only in inspector : Name of the placholder of the batch size
     public string BatchSizePlaceholderName = "batch_size";
     ///  Modify only in inspector : Name of the state placeholder
-    public string VectorObservationPlacholderName = "state";
+    public string VectorObservationPlacholderName = "vector_observation";
     ///  Modify only in inspector : Name of the recurrent input
     public string RecurrentInPlaceholderName = "recurrent_in";
     ///  Modify only in inspector : Name of the recurrent output
         if (hasState)
         {
             int stateLength = 1;
-            if (brain.brainParameters.vectorObservationSpaceType == StateType.continuous)
+            if (brain.brainParameters.vectorObservationSpaceType == SpaceType.continuous)
             {
                 stateLength = brain.brainParameters.vectorObservationSize;
             }
         // Create the state tensor
         if (hasState)
         {
-            if (brain.brainParameters.vectorObservationSpaceType == StateType.discrete)
+            if (brain.brainParameters.vectorObservationSpaceType == SpaceType.discrete)
             {
                 var discreteInputState = new int[currentBatchSize, 1];
                 for (int i = 0; i < currentBatchSize; i++)
 
         }
 
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             var output = networkOutput[0].GetValue() as float[,];
             var i = 0;
                 i++;
             }
         }
-        else if (brain.brainParameters.vectorActionSpaceType == StateType.discrete)
+        else if (brain.brainParameters.vectorActionSpaceType == SpaceType.discrete)
         {
             long[,] output = networkOutput[0].GetValue() as long[,];
             var i = 0;

unity-environment/Assets/ML-Agents/Scripts/CoreBrainPlayer.cs

     /// decide action
     public void DecideAction(Dictionary<Agent, AgentInfo> agentInfo)
     {
-        if (coord != null)
-        {
-            coord.GiveBrainInfo(brain, agentInfo);
-        }
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+		if (coord != null)
+		{
+			coord.GiveBrainInfo(brain, agentInfo);
+		}
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             foreach (Agent agent in agentInfo.Keys)
             {
         broadcast = EditorGUILayout.Toggle(new GUIContent("Broadcast",
                       "If checked, the brain will broadcast states and actions to Python."), broadcast);
         var serializedBrain = new SerializedObject(this);
-        if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+        if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
         {
             GUILayout.Label("Edit the continuous inputs for your actions", EditorStyles.boldLabel);
             var chas = serializedBrain.FindProperty("continuousPlayerActions");

unity-environment/Assets/ML-Agents/Scripts/ExternalCommunicator.cs

 
                 for (int i = 0; i < current_agents[brainName].Count(); i++)
                 {
-                    if (brain.brainParameters.vectorActionSpaceType == StateType.continuous)
+                    if (brain.brainParameters.vectorActionSpaceType == SpaceType.continuous)
                     {
                         current_agents[brainName][i].UpdateVectorAction(rMessage.vector_action[brainName].GetRange(
                             i * brain.brainParameters.vectorActionSize, brain.brainParameters.vectorActionSize).ToArray());

unity-environment/Assets/ML-Agents/Scripts/Monitor.cs

             float paddingwidth = 10 * widthScaler;
 
             float scale = 1f;
-            var origin = new Vector3(0, Screen.height);
+            var origin = new Vector3(Screen.width / 2 - keyPixelWidth, Screen.height);
             if (!(target == canvas.transform))
             {
                 Vector3 cam2obj = target.position - Camera.main.transform.position;

unity-environment/Assets/ML-Agents/Scripts/BCTeacherHelper.cs

+using System.Collections;
+using System.Collections.Generic;
+using UnityEngine;
+
+
+/// <summary>
+/// Behavioral Cloning Helper script. Attach to teacher agent to enable 
+/// resetting the experience buffer, as well as toggling session recording.
+/// </summary>
+public class BCTeacherHelper : MonoBehaviour {
+
+    bool recordExperiences;
+    bool resetBuffer;
+    Agent myAgent;
+    float bufferResetTime;
+
+    public KeyCode recordKey = KeyCode.R;
+    public KeyCode resetKey = KeyCode.C;
+
+    // Use this for initialization
+    void Start () {
+        recordExperiences = true;
+        resetBuffer = false;
+        myAgent = GetComponent<Agent>();
+        bufferResetTime = Time.time;
+    }
+    
+    // Update is called once per frame
+    void Update () {
+        if (Input.GetKeyDown(recordKey))
+        {
+            recordExperiences = !recordExperiences;
+        }
+        if (Input.GetKeyDown(resetKey))
+        {
+            resetBuffer = true;
+            bufferResetTime = Time.time;
+        }
+        else
+        {
+            resetBuffer = false;
+        }
+        Monitor.Log("Recording experiences " + recordKey.ToString(), recordExperiences.ToString());
+        float timeSinceBufferReset = Time.time - bufferResetTime;
+        Monitor.Log("Seconds since buffer reset " + resetKey.ToString(), Mathf.FloorToInt(timeSinceBufferReset));
+    }
+
+    void FixedUpdate()
+    {
+        // Convert both bools into single comma separated string. Python makes
+        // assumption that this structure is preserved. 
+        myAgent.SetTextObs(recordExperiences.ToString() + "," + resetBuffer.ToString());
+    }
+}

docs/Installation-Docker.md

-# Docker Set-up _[Experimental]_
-
-**Work In Progress**

unity-environment/Assets/ML-Agents/Scripts/TeacherHelper.cs

-using System.Collections;
-using System.Collections.Generic;
-using UnityEngine;
-
-public class TeacherHelper : MonoBehaviour {
-
-    bool recordExperiences;
-    bool resetBuffer;
-    Agent myAgent;
-    float bufferResetTime;
-
-    // Use this for initialization
-    void Start () {
-        recordExperiences = true;
-        resetBuffer = false;
-        myAgent = GetComponent<Agent>();
-        bufferResetTime = Time.time;
-    }
-    
-    // Update is called once per frame
-    void Update () {
-        if (Input.GetKeyDown(KeyCode.R))
-        {
-            recordExperiences = !recordExperiences;
-        }
-        if (Input.GetKeyDown(KeyCode.C))
-        {
-            resetBuffer = true;
-            bufferResetTime = Time.time;
-        }
-        else
-        {
-            resetBuffer = false;
-        }
-        Monitor.Log("Recording experiences", recordExperiences.ToString());
-        float timeSinceBufferReset = Time.time - bufferResetTime;
-        Monitor.Log("Seconds since buffer reset", Mathf.FloorToInt(timeSinceBufferReset));
-    }
-
-    void FixedUpdate()
-    {
-        // Convert both bools into single comma separated string. Python makes
-        // assumption that this structure is preserved. 
-        myAgent.SetTextObs(recordExperiences.ToString() + "," + resetBuffer.ToString());
-    }
-}