def __init__(self, creator: TensorCreator, network: nn.Module,
optimizer: Optimizer, storage: ObservationStorage,
params: NNetParams):
"""Unit constructor.
Args:
creator: creator of this node
network: pytorch neural network module
optimizer: pytorch optimizer object
storage: baselines observation storage object
params: baselines parameter object
"""
super().__init__(creator.device)
self.params = params
self.network = network
self.optimizer = optimizer
self.storage = storage
self.device = creator.device
self.output = creator.zeros(
SpaceEngineersConnectorConfig().task_to_agent_buffer_size,
dtype=self._float_dtype,
device=self._device)
self.label = creator.zeros(
SpaceEngineersConnectorConfig().task_to_agent_buffer_size,
dtype=self._float_dtype,
device=self._device)
self.task_control = creator.zeros(4, device=self._device)
self.cur_train_step = 0
self.last_train_loss = 0
def __init__(self, use_dataset: bool = True):
super().__init__("Task 0 - Base topology",
inputs=Task0BaseGroupInputs(self),
outputs=Task0BaseGroupOutputs(self))
if use_dataset:
params = DatasetSeObjectsParams(
dataset_config=DatasetConfig.TRAIN_TEST, save_gpu_memory=True)
self.se_node = DatasetSeObjectsNode(params)
else:
se_config = SpaceEngineersConnectorConfig()
se_config.curriculum = list((0, -1))
actions_descriptor = SpaceEngineersActionsDescriptor()
self.se_node = SpaceEngineersConnectorNode(actions_descriptor,
se_config)
def __init__(self):
super().__init__(device='cuda')
self._actions_descriptor = SpaceEngineersActionsDescriptor()
self._se_config = SpaceEngineersConnectorConfig()
self._node_se_connector = SpaceEngineersConnectorNode(
self._actions_descriptor, self._se_config)
self._node_action_monitor = ActionMonitorNode(self._actions_descriptor)
self._blank_action = ConstantNode(
shape=self._actions_descriptor.ACTION_COUNT, constant=0)
self._blank_task_data = ConstantNode(
shape=self._se_config.agent_to_task_buffer_size, constant=0)
self._blank_task_control = ConstantNode(
shape=self._se_config.TASK_CONTROL_SIZE, constant=0)
self.add_node(self._node_se_connector)
self.add_node(self._node_action_monitor)
self.add_node(self._blank_action)
self.add_node(self._blank_task_data)
self.add_node(self._blank_task_control)
Connector.connect(self._blank_action.outputs.output,
self._node_action_monitor.inputs.action_in)
Connector.connect(self._node_action_monitor.outputs.action_out,
self._node_se_connector.inputs.agent_action)
Connector.connect(self._blank_task_data.outputs.output,
self._node_se_connector.inputs.agent_to_task_label)
Connector.connect(self._blank_task_control.outputs.output,
self._node_se_connector.inputs.task_control)
def __init__(self):
super().__init__(device='cuda')
self._se_config = SpaceEngineersConnectorConfig()
self.SX = self._se_config.render_width
self.SY = self._se_config.render_height
# setup the params
# SeToyArchDebugTopology.config_se_communication()
expert_params = self.get_expert_params()
# create the nodes
self._actions_descriptor = SpaceEngineersActionsDescriptor()
self._node_se_connector = SpaceEngineersConnectorNode(
self._actions_descriptor, self._se_config)
self._node_action_monitor = ActionMonitorNode(self._actions_descriptor)
self._blank_action = ConstantNode(
shape=self._actions_descriptor.ACTION_COUNT, constant=0)
self._blank_task_data = ConstantNode(
shape=self._se_config.agent_to_task_buffer_size, constant=0)
self._blank_task_control = ConstantNode(
shape=self._se_config.TASK_CONTROL_SIZE, constant=0)
# parent_rf_dims = (self.lrf_width, self.lrf_height, 3)
self._node_lrf = ReceptiveFieldNode((self.SY, self.SX, 3),
Size2D(self.SX // EOX,
self.SY // EOY))
self._node_flock = ExpertFlockNode(expert_params)
self.add_node(self._node_se_connector)
self.add_node(self._node_action_monitor)
self.add_node(self._blank_action)
self.add_node(self._blank_task_data)
self.add_node(self._blank_task_control)
self.add_node(self._node_flock)
self.add_node(self._node_lrf)
Connector.connect(self._blank_action.outputs.output,
self._node_action_monitor.inputs.action_in)
Connector.connect(self._node_action_monitor.outputs.action_out,
self._node_se_connector.inputs.agent_action)
Connector.connect(self._blank_task_data.outputs.output,
self._node_se_connector.inputs.agent_to_task_label)
Connector.connect(self._blank_task_control.outputs.output,
self._node_se_connector.inputs.task_control)
# SE -> Flock (no LRF)
# Connector.connect(self._node_se_connector.outputs.image_output,
# self._node_flock.inputs.sp.data_input)
# Image -> LRF -> Flock
Connector.connect(self._node_se_connector.outputs.image_output,
self._node_lrf.inputs[0])
Connector.connect(self._node_lrf.outputs[0],
self._node_flock.inputs.sp.data_input)
def __init__(self):
super().__init__(device='cuda')
self._actions_descriptor = SpaceEngineersActionsDescriptor()
self._se_config = SpaceEngineersConnectorConfig()
self._se_config.skip_frames = 1
self._se_config.render_width = self.SIZE.value
self._se_config.render_height = self.SIZE.value
num_training_trajectories = 3000
num_testing_trajectories = 100
overseer = SampleCollectionOverseer(self._se_config.render_width,
self._se_config.render_height,
num_training_trajectories,
num_testing_trajectories)
self._node_se_connector = SpaceEngineersConnectorNode(
self._actions_descriptor,
self._se_config,
sample_collection_overseer=overseer)
self._node_action_monitor = ActionMonitorNode(self._actions_descriptor)
self._blank_action = ConstantNode(
shape=self._actions_descriptor.ACTION_COUNT, constant=0)
self._blank_task_data = ConstantNode(
shape=self._se_config.agent_to_task_buffer_size, constant=0)
self._blank_task_control = ConstantNode(
shape=self._se_config.TASK_CONTROL_SIZE, constant=0)
self.add_node(self._node_se_connector)
self.add_node(self._node_action_monitor)
self.add_node(self._blank_action)
self.add_node(self._blank_task_data)
self.add_node(self._blank_task_control)
Connector.connect(self._blank_action.outputs.output,
self._node_action_monitor.inputs.action_in)
Connector.connect(self._node_action_monitor.outputs.action_out,
self._node_se_connector.inputs.agent_action)
Connector.connect(self._blank_task_data.outputs.output,
self._node_se_connector.inputs.agent_to_task_label)
Connector.connect(self._blank_task_control.outputs.output,
self._node_se_connector.inputs.task_control)
def __init__(self, curriculum: tuple = (0, -1)):
self._curriculum = curriculum
self.se_config = SpaceEngineersConnectorConfig()
self.se_config.curriculum = list(self._curriculum)
def __init__(self,
seed: int = 0,
device: str = 'cuda',
eox: int = 2,
eoy: int = 2,
num_cc: int = 100,
batch_s=300,
tp_learn_period=50,
tp_max_enc_seq=1000,
se_skip_frames=9):
super().__init__(eox, eoy)
self._se_config = SpaceEngineersConnectorConfig()
self._se_config.skip_frames = se_skip_frames
self._se_config.curriculum = [0, -1]
# compute/setup parameters of the model
_, self._sy, self._sx, self._no_channels = init_se_dataset_world_params(
random_order=False)
flock_size, input_size = compute_flock_sizes(self._sy, self._sx,
self._no_channels,
self._eoy, self._eox)
expert_params = setup_flock_params(no_clusters=num_cc,
buffer_size=batch_s * 2,
batch_size=batch_s,
tp_learn_period=tp_learn_period,
max_enc_seq=tp_max_enc_seq,
flock_size=flock_size,
input_size=input_size)
flock_input_size, flock_output_size = compute_lrf_params(
self._sy, self._sx, self._no_channels, self._eoy, self._eox)
# SE nodes
self._actions_descriptor = SpaceEngineersActionsDescriptor()
self._node_se_connector = SpaceEngineersConnectorNode(
self._actions_descriptor, self._se_config)
self._node_action_monitor = ActionMonitorNode(self._actions_descriptor)
self._blank_action = ConstantNode(
shape=self._actions_descriptor.ACTION_COUNT, constant=0)
self._blank_task_data = ConstantNode(
shape=self._se_config.agent_to_task_buffer_size, constant=0)
# flock-related nodes
self._lrf_node = ReceptiveFieldNode(flock_input_size,
flock_output_size)
self._flock_node = ExpertFlockNode(expert_params, seed=seed)
self._zero_context = ConstantNode(
shape=(expert_params.flock_size, NUMBER_OF_CONTEXT_TYPES,
expert_params.temporal.incoming_context_size),
constant=0)
self._blank_task_control = ConstantNode(
shape=self._se_config.TASK_CONTROL_SIZE, constant=0)
# add nodes to the graph
self.add_node(self._lrf_node)
self.add_node(self._flock_node)
self.add_node(self._zero_context)
self.add_node(self._node_se_connector)
self.add_node(self._node_action_monitor)
self.add_node(self._blank_action)
self.add_node(self._blank_task_data)
self.add_node(self._blank_task_control)
# connect SE -> LRF -> SP
Connector.connect(self._node_se_connector.outputs.image_output,
self._lrf_node.inputs[0])
Connector.connect(self._lrf_node.outputs[0],
self._flock_node.inputs.sp.data_input)
Connector.connect(self._zero_context.outputs.output,
self._flock_node.inputs.tp.context_input)
# connect NOOP -> action_override
Connector.connect(self._blank_action.outputs.output,
self._node_action_monitor.inputs.action_in)
Connector.connect(self._node_action_monitor.outputs.action_out,
self._node_se_connector.inputs.agent_action)
# connect blank_task_data -> SE aux input
Connector.connect(self._blank_task_data.outputs.output,
self._node_se_connector.inputs.agent_to_task_label)
Connector.connect(self._blank_task_control.outputs.output,
self._node_se_connector.inputs.task_control)
# prepare for run
set_global_seeds(seed)
self._last_step_duration = 0
def __init__(self, curriculum: tuple = (1, -1)):
super().__init__()
se_config = SpaceEngineersConnectorConfig()
se_config.render_width = 16
se_config.render_height = 16
se_config.curriculum = list(curriculum)
base_expert_params = ExpertParams()
base_expert_params.flock_size = 1
base_expert_params.n_cluster_centers = 100
base_expert_params.compute_reconstruction = False
base_expert_params.spatial.cluster_boost_threshold = 1000
base_expert_params.spatial.learning_rate = 0.2
base_expert_params.spatial.batch_size = 1000
base_expert_params.spatial.buffer_size = 1010
base_expert_params.spatial.learning_period = 100
base_expert_params.temporal.batch_size = 1000
base_expert_params.temporal.buffer_size = 1010
base_expert_params.temporal.learning_period = 200
base_expert_params.temporal.forgetting_limit = 20000
# parent_expert_params = ExpertParams()
# parent_expert_params.flock_size = 1
# parent_expert_params.n_cluster_centers = 20
# parent_expert_params.compute_reconstruction = True
# parent_expert_params.temporal.exploration_probability = 0.9
# parent_expert_params.spatial.cluster_boost_threshold = 1000
# parent_expert_params.spatial.learning_rate = 0.2
# parent_expert_params.spatial.batch_size = 1000
# parent_expert_params.spatial.buffer_size = 1010
# parent_expert_params.spatial.learning_period = 100
# parent_expert_params.temporal.context_without_rewards_size = se_config.LOCATION_SIZE_ONE_HOT
# SE nodes
actions_descriptor = SpaceEngineersActionsDescriptor()
node_se_connector = SpaceEngineersConnectorNode(
actions_descriptor, se_config)
node_action_monitor = ActionMonitorNode(actions_descriptor)
# flock-related nodes
flock_node = ExpertFlockNode(base_expert_params)
blank_task_control = ConstantNode((se_config.TASK_CONTROL_SIZE, ))
blank_task_labels = ConstantNode((20, ))
# parent_flock_node = ExpertFlockNode(parent_expert_params)
join_node = JoinNode(flatten=True)
actions = ['FORWARD', 'BACKWARD', 'LEFT', 'RIGHT']
action_count = len(actions)
pass_actions_node = PassNode(output_shape=(action_count, ),
name="pass actions")
fork_node = ForkNode(
0, [base_expert_params.n_cluster_centers, action_count])
def squeeze(inputs, outputs):
outputs[0].copy_(inputs[0].squeeze())
squeeze_node = LambdaNode(
squeeze,
1, [(base_expert_params.n_cluster_centers + action_count, )],
name="squeeze lambda node")
def stack_and_unsqueeze(inputs, outputs):
outputs[0].copy_(torch.stack([inputs[0], inputs[1]]).unsqueeze(0))
stack_unsqueeze_node = LambdaNode(
stack_and_unsqueeze,
2, [(1, 2, se_config.LOCATION_SIZE_ONE_HOT)],
name="stack and unsqueeze node")
to_one_hot_node = ToOneHotNode()
action_parser_node = AgentActionsParserNode(actions_descriptor,
actions)
random_node = RandomNumberNode(0,
action_count,
name="random action generator",
generate_new_every_n=5,
randomize_intervals=True)
switch_node = SwitchNode(2)
# add nodes to the graph
self.add_node(flock_node)
# self.add_node(parent_flock_node)
self.add_node(node_se_connector)
self.add_node(node_action_monitor)
self.add_node(blank_task_control)
self.add_node(blank_task_labels)
# self.add_node(join_node)
# self.add_node(fork_node)
# self.add_node(pass_actions_node)
# self.add_node(squeeze_node)
# self.add_node(to_one_hot_node)
# self.add_node(stack_unsqueeze_node)
self.add_node(action_parser_node)
self.add_node(random_node)
# self.add_node(switch_node)
# first layer
Connector.connect(node_se_connector.outputs.image_output,
flock_node.inputs.sp.data_input)
# Connector.connect(
# flock_node.outputs.tp.projection_outputs,
# join_node.inputs[0]
# )
# Connector.connect(
# pass_actions_node.outputs.output,
# join_node.inputs[1]
# )
# # second layer
# Connector.connect(
# join_node.outputs.output,
# parent_flock_node.inputs.sp.data_input
# )
# Connector.connect(
# node_se_connector.outputs.task_to_agent_location_one_hot,
# stack_unsqueeze_node.inputs[0]
# )
# Connector.connect(
# node_se_connector.outputs.task_to_agent_location_target_one_hot,
# stack_unsqueeze_node.inputs[1]
# )
# Connector.connect(
# stack_unsqueeze_node.outputs[0],
# parent_flock_node.inputs.tp.context_input
# )
#
# # actions
# Connector.connect(
# parent_flock_node.outputs.sp.predicted_reconstructed_input,
# squeeze_node.inputs[0]
# )
# Connector.connect(
# squeeze_node.outputs[0],
# fork_node.inputs.input
# )
# Connector.connect(
# fork_node.outputs[1],
# to_one_hot_node.inputs.input
# )
# Connector.connect(
# random_node.outputs.one_hot_output,
# switch_node.inputs[0]
# )
# Connector.connect(
# to_one_hot_node.outputs.output,
# switch_node.inputs[1]
# )
# Connector.connect(
# switch_node.outputs.output,
# action_parser_node.inputs.input
# )
# directly use random exploration
Connector.connect(random_node.outputs.one_hot_output,
action_parser_node.inputs.input)
Connector.connect(action_parser_node.outputs.output,
node_action_monitor.inputs.action_in)
# Connector.connect(
# switch_node.outputs.output,
# pass_actions_node.inputs.input,
# is_low_priority=True
# )
Connector.connect(
node_action_monitor.outputs.action_out,
node_se_connector.inputs.agent_action,
# is_low_priority=True
is_backward=False)
# blank connection
Connector.connect(blank_task_control.outputs.output,
node_se_connector.inputs.task_control)
Connector.connect(blank_task_labels.outputs.output,
node_se_connector.inputs.agent_to_task_label)
# Save the SE connector so we can check testing/training phase.
# When the se_io interface has been added, this can be removed.
self._node_se_connector = node_se_connector
def __init__(self, curriculum: tuple = (1, -1)):
super().__init__("Task 1 - Base topology world", inputs=Task1BaseGroupWorldInputs(self),
outputs=Task1BaseGroupWorldOutputs(self))
actions_descriptor = SpaceEngineersActionsDescriptor()
actions = ['FORWARD', 'BACKWARD', 'LEFT', 'RIGHT']
action_count = len(actions)
# SE nodes
se_config = SpaceEngineersConnectorConfig()
se_config.curriculum = list(curriculum)
node_se_connector = SpaceEngineersConnectorNode(actions_descriptor, se_config)
def node_se_connector_is_learning() -> bool:
if node_se_connector.outputs.metadata_testing_phase.tensor is None:
return False
else:
return node_se_connector.outputs.metadata_testing_phase.tensor.cpu().item() == 1
node_se_connector.is_learning = node_se_connector_is_learning
self.node_se_connector = node_se_connector
blank_task_control = ConstantNode((se_config.TASK_CONTROL_SIZE,))
pass_actions_node = PassNode(output_shape=(action_count,), name="pass actions")
blank_task_labels = ConstantNode((20,))
self.add_node(node_se_connector)
self.add_node(blank_task_control)
self.add_node(pass_actions_node)
self.add_node(blank_task_labels)
Connector.connect(
self.inputs.last_actions.output,
pass_actions_node.inputs.input
)
Connector.connect(
self.inputs.current_actions.output,
node_se_connector.inputs.agent_action
)
Connector.connect(
pass_actions_node.outputs.output,
self.outputs.actions.input
)
Connector.connect(
node_se_connector.outputs.image_output,
self.outputs.image.input
)
Connector.connect(
node_se_connector.outputs.task_to_agent_location_one_hot,
self.outputs.current_location.input
)
Connector.connect(
node_se_connector.outputs.task_to_agent_location_target_one_hot,
self.outputs.target_location.input
)
# blank connection
Connector.connect(blank_task_control.outputs.output,
node_se_connector.inputs.task_control)
Connector.connect(blank_task_labels.outputs.output,
node_se_connector.inputs.agent_to_task_label)
def __init__(self, **kwargs):
super().__init__(device='cpu')
self._current_step = 0
# set topology params and configs
self._params = NNetParams(NNetParams.default_params())
self._params.set_params(_nn_node_params) # params defined in this file
self._params.set_params(kwargs) # params defined in GUI
# SE config and setup
self._se_config = SpaceEngineersConnectorConfig()
self._se_config.curriculum = list(self._params.curriculum)
self._actions_descriptor = SpaceEngineersActionsDescriptor()
# set SE specific params automatically
self._params.set_params({
'input_shape':
(3, self._se_config.render_width, self._se_config.render_height),
'output_size':
self._se_config.agent_to_task_buffer_size
})
# observation storage params
self._observation_types = {
'x': (self._params.buffer_size,
*self._params.input_shape), # observations
'y':
(self._params.buffer_size, self._params.output_size), # labels
}
# data storage
self._storage = ObservationStorage(self._params.buffer_size,
self._observation_types)
self._storage.to('cpu' if self._params.mixed_mode else self.device)
# network needs to have the global seeds to have set before creating (outside of the node in this case)
set_global_seeds(seed=self._params.seed)
# ==================================================
# NOTE: Replace here with your own architecture.
# It needs to be able to take the correct
# input and output (shape/size)
# ==================================================
# neural network setup
self._network = NNet(
input_shape=self._params.input_shape,
output_shape=self._params.output_size).to(
'cuda' if self._params.mixed_mode else self.device)
# ==================================================
# neural net optimizer
self._optimizer = optim.Adam(self._network.parameters(),
lr=self._params.lr)
# SE Node
self._se_connector = SpaceEngineersConnectorNode(
self._actions_descriptor, self._se_config)
# NNet Node
self._nnet_node = NNetNode(self._network,
self._optimizer,
self._storage,
self._params,
name='Neural Network Node')
# add nodes to the topology
self.add_node(self._nnet_node)
self.add_node(self._se_connector)
# connect it all up
Connector.connect(self._se_connector.outputs.image_output,
self._nnet_node.inputs.input)
Connector.connect(self._se_connector.outputs.task_to_agent_label,
self._nnet_node.inputs.label)
Connector.connect(self._se_connector.outputs.metadata_testing_phase,
self._nnet_node.inputs.testing_phase)
Connector.connect(self._nnet_node.outputs.output,
self._se_connector.inputs.agent_action,
is_backward=True)
Connector.connect(self._nnet_node.outputs.label,
self._se_connector.inputs.agent_to_task_label,
is_backward=True)
# necessary, but not used connector
# TODO: remove once node is not needing this
Connector.connect(self._nnet_node.outputs.task_control,
self._se_connector.inputs.task_control,
is_backward=True)