Creating a Custom Heuristic Function Optimizer and Loss

Registration, Initialization, and Forward

@heuristic_factory.register_class("resnet_fc_asym")
class ResnetFCHeur(HeurNNet[FlatIn]):
    @staticmethod
    def nnet_input_type() -> Type[FlatIn]:
        return FlatIn

    def __init__(self, nnet_input: FlatIn, out_dim: int, q_fix: bool, res_dim: int = 1000, num_blocks: int = 4, batch_norm: bool = False,
                 lr: float = 0.001, momentum: float = 0.5, over_w: float = 1.0):
        super().__init__(nnet_input, out_dim, q_fix)
        self.lr = lr
        self.momentum: float = momentum
        self.lr_curr: float = self.lr
        self.over_w: float = over_w
        # one hots
        self.one_hots: nn.ModuleList = nn.ModuleList()
        input_dim_tot: int = 0
        input_dims, one_hot_depths = self.nnet_input.get_input_info()
        for input_dim, one_hot_depth in zip(input_dims, one_hot_depths, strict=True):
            assert one_hot_depth >= 1
            self.one_hots.append(OneHot(one_hot_depth, True))
            input_dim_tot += input_dim * one_hot_depth

        # res net
        self.res_dim: int = res_dim

        def res_block_init() -> nn.Module:
            return FullyConnectedModel(res_dim, [res_dim] * 2, ["RELU", "LINEAR"], batch_norms=[batch_norm] * 2)

        self.heur = nn.Sequential(
            nn.Linear(input_dim_tot, res_dim),
            ResnetModel(res_block_init, num_blocks, "RELU"),
            nn.Linear(res_dim, self.out_dim)
        )

    # end init

    # start forward
    def _forward(self, inputs: List[Tensor]) -> Tensor:
        inputs_oh: List[Tensor] = [one_hot(input_i) for input_i, one_hot in zip(inputs, self.one_hots)]
        x: Tensor = self.heur(torch.cat(inputs_oh, dim=1))
        return x

Important

DeepXube expects the first three arguments, nnet_input: FlatIn, out_dim: int, q_fix: bool to have these exact names so the neural network can be properly initialized.

Custom Optimizer

Instead of using the default Adam optimizer, we implement stochastic gradient descent with momentum. Instead of using the default exponential learning rate decay, we implement a step decay.

    def get_optimizer(self) -> Optimizer:
        return optim.SGD(self.parameters(), lr=self.lr, momentum=self.momentum)

    def update_optimizer(self, optimizer: Optimizer, train_itr: int) -> None:
        self.lr_curr: float = self.lr / ((train_itr // 100) + 1)
        for param_group in optimizer.param_groups:
            param_group['lr'] = self.lr_curr

Custom Loss

    def get_loss_and_info(self, fwd_tr_tensors: List[Tensor], get_info: bool) -> Tuple[Tensor, Optional[str]]:
        ctgs_nnet: Tensor = fwd_tr_tensors[0]
        ctgs_targ: Tensor = fwd_tr_tensors[1]

        err = ctgs_nnet - ctgs_targ
        sq_err = err ** 2
        sq_err_w = (self.over_w * sq_err * (err > 0.0)) + (sq_err * (err <= 0.0))
        loss = sq_err_w.mean()

        info: Optional[str] = None
        if get_info:
            info = f"targ_ctg: {ctgs_targ.mean().item():.2f}, nnet_ctg: {ctgs_nnet.mean().item():.2f}, lr: {self.lr_curr:.2E}"

        return loss, info

Representation

    def __repr__(self) -> str:
        repr_str: str = super().__repr__()
        repr_str = f"{repr_str}\nOver est weight: {self.over_w}"

        return repr_str

Parser

@heuristic_factory.register_parser("resnet_fc_asym")
class ResnetFCParser(DelimParser):
    def __init__(self) -> None:
        super().__init__()
        self.add_argument("H", "res_dim", int, "dimensionality of hidden layers in residual blocks")
        self.add_argument("B", "num_blocks", int, "number of residual blocks")
        self.add_argument("bn", "batch_norm", None, "Batch normalization")
        self.add_argument("O", "over_w", float, "weight for overestimation")

    @property
    def delim(self) -> str:
        return "_"

Training

deepxube train --domain cube3 --heur resnet_fc_asym.200H_2B_bn_20o --heur_type V --pathfind graph_v --step_max 100 --up_itrs 100 --search_itrs 50 --backup -1 --procs 2 --batch_size 200 --max_itrs 5000 --dir tutorial/heur_asym_loss/models/ --display 50

--display 50: to display the neural network training information every 50 iterations.

device: cpu, devices: [], on_gpu: False
ResnetFCHeur(
  (one_hots): ModuleList(
    (0): OneHot()
  )
  (heur): Sequential(
    (0): Linear(in_features=324, out_features=200, bias=True)
    (1): ResnetModel(
      (blocks): ModuleList(
        (0-1): 2 x ModuleList(
          (0): FullyConnectedModel(
            (layers): ModuleList(
              (0): ModuleList(
                (0): Linear(in_features=200, out_features=200, bias=True)
                (1): BatchNorm1d(200, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                (2): ReLU()
              )
              (1): ModuleList(
                (0): Linear(in_features=200, out_features=200, bias=True)
                (1): BatchNorm1d(200, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
                (2): LinearAct()
              )
            )
          )
        )
      )
      (act_fns): ModuleList(
        (0-1): 2 x ReLU()
      )
    )
    (2): Linear(in_features=200, out_features=1, bias=True)
  )
)
Over est weight: 20.0
Number of trainable parameters: 227,601
Initializing data buffer with max size 20,000
Input array sizes:
index: 0, dtype: uint8, shape: (54,)
index: 1, dtype: float64, shape: ()
Data buffer initialized. Time: 0.0001201629638671875
UpdateHeurVRLKeepGoal(UpArgs(procs=2, up_itrs=100, step_max=100, search_itrs=50, ub_heur_solns=False, backup=-1, policy_rand_prob=0.0, up_gen_itrs=None, up_batch_size=100, nnet_batch_size=20000, sync_main=False, v=False))
GraphSearchHeurNodeActsEnum(batch_size=1, weight=1.0, eps=0.0)
TrainArgs(batch_size=200, max_itrs=5000, balance_steps=False, rb=0, loss_thresh=inf, targ_up_searches=0, skip_heur=False, skip_policy=False, checkpoint=0, grad_accum=1, display=50)
Cube3()

Getting Data - itr: 0, update_num: 0, targ_update: 0, num_gen: 20,000
Times - steps_gen: 0.02, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.00, gc: 0.33, ->get_states: 0.03, ->pathfinding: 2.80, Tot: 3.22
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.44, nodes: 0.57, up_inst: 0.04, heur: 1.50, filt: 0.12, cost: 0.03, pushpop: 0.05, edges_next: 0.01, set_next: 0.00, Tot: 2.80
Data - %solved: 1.82, path_costs: 0.750, search_itrs: 6.481, cost-to-go (mean/min/max): 1.02/0.00/2.00
Itr: 0, loss: 2.74E+00, targ_ctg: 1.01, nnet_ctg: -0.56, lr: 1.00E-03, Time: 1.98
Itr: 50, loss: 1.81E-01, targ_ctg: 1.01, nnet_ctg: 0.74, lr: 1.00E-03, Time: 2.08
Train - itrs: 100, loss: 1.37E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.75, up_end: 0.23, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.18

Getting Data - itr: 100, update_num: 1, targ_update: 1, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.38, ->get_states: 0.03, ->pathfinding: 3.19, Tot: 3.65
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.46, nodes: 0.48, up_inst: 0.05, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.22, edges_next: 0.01, set_next: 0.00, Tot: 3.19
Data - %solved: 3.70, path_costs: 0.943, search_itrs: 3.714, cost-to-go (mean/min/max): 1.45/0.00/2.66
Itr: 100, loss: 5.99E-01, targ_ctg: 1.46, nnet_ctg: 0.74, lr: 5.00E-04, Time: 1.98
Itr: 150, loss: 1.39E-01, targ_ctg: 1.48, nnet_ctg: 1.22, lr: 5.00E-04, Time: 2.07
Train - itrs: 100, loss: 1.41E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.75, up_end: 0.22, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.17

Getting Data - itr: 200, update_num: 2, targ_update: 2, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.23, Tot: 3.71
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.58, nodes: 0.36, up_inst: 0.18, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.12, edges_next: 0.01, set_next: 0.00, Tot: 3.23
Data - %solved: 4.29, path_costs: 1.420, search_itrs: 4.703, cost-to-go (mean/min/max): 2.09/0.00/3.31
Itr: 200, loss: 8.02E-01, targ_ctg: 2.10, nnet_ctg: 1.23, lr: 3.33E-04, Time: 2.02
Itr: 250, loss: 1.44E-01, targ_ctg: 2.11, nnet_ctg: 1.85, lr: 3.33E-04, Time: 2.12
Train - itrs: 100, loss: 1.55E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.77, up_end: 0.24, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.21

Getting Data - itr: 300, update_num: 3, targ_update: 3, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.38, ->get_states: 0.03, ->pathfinding: 3.24, Tot: 3.71
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.40, nodes: 0.56, up_inst: 0.05, heur: 1.85, filt: 0.11, cost: 0.03, pushpop: 0.21, edges_next: 0.01, set_next: 0.00, Tot: 3.24
Data - %solved: 5.48, path_costs: 1.743, search_itrs: 4.915, cost-to-go (mean/min/max): 2.63/0.00/3.98
Itr: 300, loss: 6.76E-01, targ_ctg: 2.63, nnet_ctg: 1.86, lr: 2.50E-04, Time: 2.01
Itr: 350, loss: 2.02E-01, targ_ctg: 2.63, nnet_ctg: 2.38, lr: 2.50E-04, Time: 2.11
Train - itrs: 100, loss: 1.70E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.78, up_end: 0.23, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.21

Getting Data - itr: 400, update_num: 4, targ_update: 4, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.38, ->get_states: 0.03, ->pathfinding: 3.22, Tot: 3.69
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.51, nodes: 0.35, up_inst: 0.12, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.25, edges_next: 0.01, set_next: 0.00, Tot: 3.22
Data - %solved: 6.57, path_costs: 2.392, search_itrs: 8.008, cost-to-go (mean/min/max): 3.19/0.00/4.54
Itr: 400, loss: 6.79E-01, targ_ctg: 3.13, nnet_ctg: 2.35, lr: 2.00E-04, Time: 2.00
Itr: 450, loss: 2.81E-01, targ_ctg: 3.17, nnet_ctg: 2.86, lr: 2.00E-04, Time: 2.11
Train - itrs: 100, loss: 1.75E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.77, up_end: 0.23, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.20

Getting Data - itr: 500, update_num: 5, targ_update: 5, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.22, Tot: 3.70
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.52, nodes: 0.42, up_inst: 0.05, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.25, edges_next: 0.01, set_next: 0.00, Tot: 3.22
Data - %solved: 6.41, path_costs: 2.653, search_itrs: 10.605, cost-to-go (mean/min/max): 3.69/0.00/5.07
Itr: 500, loss: 8.38E-01, targ_ctg: 3.77, nnet_ctg: 2.90, lr: 1.67E-04, Time: 2.02
Itr: 550, loss: 2.88E-01, targ_ctg: 3.73, nnet_ctg: 3.34, lr: 1.67E-04, Time: 2.15
Train - itrs: 100, loss: 3.16E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.77, up_end: 0.24, data_samp: 0.00, train: 0.23, save_net: 0.00, save_status: 0.00, Tot: 2.26

Getting Data - itr: 600, update_num: 6, targ_update: 6, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.40, ->get_states: 0.03, ->pathfinding: 3.25, Tot: 3.74
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.51, nodes: 0.35, up_inst: 0.17, heur: 1.85, filt: 0.11, cost: 0.03, pushpop: 0.19, edges_next: 0.01, set_next: 0.00, Tot: 3.25
Data - %solved: 7.15, path_costs: 2.598, search_itrs: 8.046, cost-to-go (mean/min/max): 4.05/0.00/5.53
Itr: 600, loss: 5.67E-01, targ_ctg: 3.94, nnet_ctg: 3.35, lr: 1.43E-04, Time: 2.03
Itr: 650, loss: 4.85E-01, targ_ctg: 4.05, nnet_ctg: 3.68, lr: 1.43E-04, Time: 2.12
Train - itrs: 100, loss: 4.13E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.23, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.23

Getting Data - itr: 700, update_num: 7, targ_update: 7, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.38, ->get_states: 0.03, ->pathfinding: 3.26, Tot: 3.74
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.56, nodes: 0.48, up_inst: 0.10, heur: 1.84, filt: 0.11, cost: 0.03, pushpop: 0.10, edges_next: 0.01, set_next: 0.00, Tot: 3.26
Data - %solved: 7.28, path_costs: 2.565, search_itrs: 8.276, cost-to-go (mean/min/max): 4.35/0.00/5.83
Itr: 700, loss: 5.76E-01, targ_ctg: 4.32, nnet_ctg: 3.71, lr: 1.25E-04, Time: 2.04
Itr: 750, loss: 4.96E-01, targ_ctg: 4.32, nnet_ctg: 3.99, lr: 1.25E-04, Time: 2.14
Train - itrs: 100, loss: 4.32E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.24, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.24

Getting Data - itr: 800, update_num: 8, targ_update: 8, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.00, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.85, Tot: 4.34
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.56, nodes: 0.35, up_inst: 0.18, heur: 2.36, filt: 0.13, cost: 0.03, pushpop: 0.20, edges_next: 0.01, set_next: 0.00, Tot: 3.85
Data - %solved: 6.71, path_costs: 2.831, search_itrs: 11.631, cost-to-go (mean/min/max): 4.67/0.00/6.31
Itr: 800, loss: 8.84E-01, targ_ctg: 4.80, nnet_ctg: 3.94, lr: 1.11E-04, Time: 2.33
Itr: 850, loss: 4.55E-01, targ_ctg: 4.74, nnet_ctg: 4.29, lr: 1.11E-04, Time: 2.43
Train - itrs: 100, loss: 5.33E-01, targ_updated: True
Times - up_start: 0.00, up_data: 2.09, up_end: 0.24, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.53

Getting Data - itr: 900, update_num: 9, targ_update: 9, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.38, ->get_states: 0.04, ->pathfinding: 3.24, Tot: 3.71
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.50, nodes: 0.63, up_inst: 0.04, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.07, edges_next: 0.01, set_next: 0.00, Tot: 3.24
Data - %solved: 8.69, path_costs: 3.082, search_itrs: 11.200, cost-to-go (mean/min/max): 4.89/0.00/6.59
Itr: 900, loss: 5.66E-01, targ_ctg: 4.90, nnet_ctg: 4.25, lr: 1.00E-04, Time: 2.01
Itr: 950, loss: 5.73E-01, targ_ctg: 4.90, nnet_ctg: 4.46, lr: 1.00E-04, Time: 2.11
Train - itrs: 100, loss: 4.70E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.78, up_end: 0.23, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.20

Getting Data - itr: 1000, update_num: 10, targ_update: 10, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.38, ->get_states: 0.03, ->pathfinding: 3.32, Tot: 3.80
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.36, nodes: 0.70, up_inst: 0.10, heur: 1.88, filt: 0.11, cost: 0.03, pushpop: 0.11, edges_next: 0.01, set_next: 0.00, Tot: 3.32
Data - %solved: 8.56, path_costs: 2.990, search_itrs: 7.696, cost-to-go (mean/min/max): 5.08/0.00/7.04
Itr: 1000, loss: 6.22E-01, targ_ctg: 5.00, nnet_ctg: 4.45, lr: 9.09E-05, Time: 2.05
Itr: 1050, loss: 6.53E-01, targ_ctg: 5.16, nnet_ctg: 4.62, lr: 9.09E-05, Time: 2.15
Train - itrs: 100, loss: 6.79E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.82, up_end: 0.23, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.26

Getting Data - itr: 1100, update_num: 11, targ_update: 11, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.24, Tot: 3.73
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.47, nodes: 0.58, up_inst: 0.05, heur: 1.83, filt: 0.11, cost: 0.03, pushpop: 0.14, edges_next: 0.01, set_next: 0.00, Tot: 3.24
Data - %solved: 8.51, path_costs: 3.419, search_itrs: 10.349, cost-to-go (mean/min/max): 5.37/0.00/7.18
Itr: 1100, loss: 7.49E-01, targ_ctg: 5.36, nnet_ctg: 4.63, lr: 8.33E-05, Time: 2.02
Itr: 1150, loss: 5.89E-01, targ_ctg: 5.43, nnet_ctg: 4.91, lr: 8.33E-05, Time: 2.12
Train - itrs: 100, loss: 6.99E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.23, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.21

Getting Data - itr: 1200, update_num: 12, targ_update: 12, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.41, ->get_states: 0.03, ->pathfinding: 3.27, Tot: 3.77
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.60, nodes: 0.46, up_inst: 0.04, heur: 1.84, filt: 0.11, cost: 0.03, pushpop: 0.15, edges_next: 0.01, set_next: 0.00, Tot: 3.27
Data - %solved: 7.98, path_costs: 3.341, search_itrs: 8.891, cost-to-go (mean/min/max): 5.60/0.00/7.59
Itr: 1200, loss: 7.18E-01, targ_ctg: 5.62, nnet_ctg: 4.95, lr: 7.69E-05, Time: 2.04
Itr: 1250, loss: 4.72E-01, targ_ctg: 5.61, nnet_ctg: 5.11, lr: 7.69E-05, Time: 2.14
Train - itrs: 100, loss: 8.20E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.80, up_end: 0.24, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.24

Getting Data - itr: 1300, update_num: 13, targ_update: 13, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.41, ->get_states: 0.04, ->pathfinding: 3.25, Tot: 3.75
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.49, nodes: 0.47, up_inst: 0.05, heur: 1.84, filt: 0.11, cost: 0.03, pushpop: 0.23, edges_next: 0.01, set_next: 0.00, Tot: 3.25
Data - %solved: 8.51, path_costs: 3.413, search_itrs: 11.265, cost-to-go (mean/min/max): 5.76/0.00/7.85
Itr: 1300, loss: 1.01E+00, targ_ctg: 5.69, nnet_ctg: 5.18, lr: 7.14E-05, Time: 2.05
Itr: 1350, loss: 9.86E-01, targ_ctg: 5.71, nnet_ctg: 5.24, lr: 7.14E-05, Time: 2.15
Train - itrs: 100, loss: 7.12E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.26, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.25

Getting Data - itr: 1400, update_num: 14, targ_update: 14, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.47, ->get_states: 0.04, ->pathfinding: 3.49, Tot: 4.06
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.58, nodes: 0.48, up_inst: 0.05, heur: 2.00, filt: 0.12, cost: 0.03, pushpop: 0.12, edges_next: 0.09, set_next: 0.00, Tot: 3.49
Data - %solved: 9.04, path_costs: 3.311, search_itrs: 7.993, cost-to-go (mean/min/max): 5.88/0.00/8.01
Itr: 1400, loss: 7.06E-01, targ_ctg: 5.92, nnet_ctg: 5.25, lr: 6.67E-05, Time: 2.20
Itr: 1450, loss: 6.82E-01, targ_ctg: 5.84, nnet_ctg: 5.40, lr: 6.67E-05, Time: 2.31
Train - itrs: 100, loss: 6.77E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.93, up_end: 0.26, data_samp: 0.00, train: 0.22, save_net: 0.00, save_status: 0.00, Tot: 2.42

Getting Data - itr: 1500, update_num: 15, targ_update: 15, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.41, ->get_states: 0.04, ->pathfinding: 3.31, Tot: 3.82
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.40, nodes: 0.46, up_inst: 0.24, heur: 1.87, filt: 0.11, cost: 0.03, pushpop: 0.17, edges_next: 0.01, set_next: 0.00, Tot: 3.31
Data - %solved: 9.83, path_costs: 3.625, search_itrs: 9.881, cost-to-go (mean/min/max): 6.08/0.00/8.38
Itr: 1500, loss: 9.17E-01, targ_ctg: 5.92, nnet_ctg: 5.39, lr: 6.25E-05, Time: 2.07
Itr: 1550, loss: 6.52E-01, targ_ctg: 6.06, nnet_ctg: 5.58, lr: 6.25E-05, Time: 2.17
Train - itrs: 100, loss: 4.82E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.82, up_end: 0.24, data_samp: 0.00, train: 0.20, save_net: 0.01, save_status: 0.00, Tot: 2.28

Getting Data - itr: 1600, update_num: 16, targ_update: 16, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.38, ->get_states: 0.04, ->pathfinding: 3.29, Tot: 3.77
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.42, nodes: 0.51, up_inst: 0.12, heur: 1.87, filt: 0.11, cost: 0.03, pushpop: 0.20, edges_next: 0.01, set_next: 0.00, Tot: 3.29
Data - %solved: 9.02, path_costs: 3.536, search_itrs: 8.710, cost-to-go (mean/min/max): 6.10/0.00/8.49
Itr: 1600, loss: 9.26E-01, targ_ctg: 6.10, nnet_ctg: 5.63, lr: 5.88E-05, Time: 2.05
Itr: 1650, loss: 9.12E-01, targ_ctg: 6.14, nnet_ctg: 5.55, lr: 5.88E-05, Time: 2.15
Train - itrs: 100, loss: 5.15E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.81, up_end: 0.23, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.26

Getting Data - itr: 1700, update_num: 17, targ_update: 17, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.40, ->get_states: 0.03, ->pathfinding: 3.29, Tot: 3.78
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.27, nodes: 0.75, up_inst: 0.08, heur: 1.86, filt: 0.11, cost: 0.03, pushpop: 0.15, edges_next: 0.01, set_next: 0.00, Tot: 3.29
Data - %solved: 10.20, path_costs: 4.018, search_itrs: 12.119, cost-to-go (mean/min/max): 6.32/0.00/8.64
Itr: 1700, loss: 6.89E-01, targ_ctg: 6.29, nnet_ctg: 5.55, lr: 5.56E-05, Time: 2.05
Itr: 1750, loss: 6.79E-01, targ_ctg: 6.38, nnet_ctg: 5.78, lr: 5.56E-05, Time: 2.14
Train - itrs: 100, loss: 9.86E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.81, up_end: 0.24, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.24

Getting Data - itr: 1800, update_num: 18, targ_update: 18, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.05, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.56, ->get_states: 0.04, ->pathfinding: 3.60, Tot: 4.27
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.55, nodes: 0.37, up_inst: 0.10, heur: 1.96, filt: 0.12, cost: 0.03, pushpop: 0.43, edges_next: 0.01, set_next: 0.00, Tot: 3.60
Data - %solved: 9.66, path_costs: 3.199, search_itrs: 9.721, cost-to-go (mean/min/max): 6.36/0.00/8.91
Itr: 1800, loss: 1.04E+00, targ_ctg: 6.27, nnet_ctg: 5.77, lr: 5.26E-05, Time: 2.35
Itr: 1850, loss: 8.97E-01, targ_ctg: 6.58, nnet_ctg: 5.82, lr: 5.26E-05, Time: 2.45
Train - itrs: 100, loss: 6.84E-01, targ_updated: True
Times - up_start: 0.00, up_data: 2.00, up_end: 0.34, data_samp: 0.00, train: 0.22, save_net: 0.00, save_status: 0.00, Tot: 2.57

Getting Data - itr: 1900, update_num: 19, targ_update: 19, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.05, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.45, ->get_states: 0.04, ->pathfinding: 3.75, Tot: 4.31
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.55, nodes: 0.58, up_inst: 0.09, heur: 2.16, filt: 0.12, cost: 0.03, pushpop: 0.17, edges_next: 0.01, set_next: 0.00, Tot: 3.75
Data - %solved: 9.27, path_costs: 3.477, search_itrs: 10.395, cost-to-go (mean/min/max): 6.43/0.00/8.99
Itr: 1900, loss: 8.69E-01, targ_ctg: 6.40, nnet_ctg: 5.82, lr: 5.00E-05, Time: 2.34
Itr: 1950, loss: 8.33E-01, targ_ctg: 6.55, nnet_ctg: 5.85, lr: 5.00E-05, Time: 2.45
Train - itrs: 100, loss: 1.22E+00, targ_updated: True
Times - up_start: 0.00, up_data: 2.07, up_end: 0.27, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.55

Getting Data - itr: 2000, update_num: 20, targ_update: 20, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.33, Tot: 3.81
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.36, nodes: 0.52, up_inst: 0.17, heur: 1.90, filt: 0.11, cost: 0.03, pushpop: 0.20, edges_next: 0.01, set_next: 0.00, Tot: 3.33
Data - %solved: 8.80, path_costs: 2.890, search_itrs: 5.955, cost-to-go (mean/min/max): 6.42/0.00/8.99
Itr: 2000, loss: 6.18E-01, targ_ctg: 6.37, nnet_ctg: 5.85, lr: 4.76E-05, Time: 2.06
Itr: 2050, loss: 1.16E+00, targ_ctg: 6.23, nnet_ctg: 5.91, lr: 4.76E-05, Time: 2.17
Train - itrs: 100, loss: 7.65E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.83, up_end: 0.23, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.27

Getting Data - itr: 2100, update_num: 21, targ_update: 21, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.40, ->get_states: 0.04, ->pathfinding: 3.24, Tot: 3.74
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.51, nodes: 0.49, up_inst: 0.10, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.15, edges_next: 0.01, set_next: 0.00, Tot: 3.24
Data - %solved: 10.33, path_costs: 3.683, search_itrs: 13.318, cost-to-go (mean/min/max): 6.57/0.00/9.23
Itr: 2100, loss: 9.75E-01, targ_ctg: 6.61, nnet_ctg: 5.90, lr: 4.55E-05, Time: 2.03
Itr: 2150, loss: 2.61E+00, targ_ctg: 6.55, nnet_ctg: 5.96, lr: 4.55E-05, Time: 2.13
Train - itrs: 100, loss: 7.16E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.24, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.22

Getting Data - itr: 2200, update_num: 22, targ_update: 22, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.42, ->get_states: 0.03, ->pathfinding: 3.26, Tot: 3.77
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.40, nodes: 0.39, up_inst: 0.24, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.24, edges_next: 0.01, set_next: 0.00, Tot: 3.26
Data - %solved: 8.43, path_costs: 4.154, search_itrs: 13.783, cost-to-go (mean/min/max): 6.78/0.00/9.45
Itr: 2200, loss: 1.22E+00, targ_ctg: 6.81, nnet_ctg: 6.02, lr: 4.35E-05, Time: 2.05
Itr: 2250, loss: 5.70E-01, targ_ctg: 6.75, nnet_ctg: 6.19, lr: 4.35E-05, Time: 2.16
Train - itrs: 100, loss: 7.46E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.26, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.26

Getting Data - itr: 2300, update_num: 23, targ_update: 23, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.46, ->get_states: 0.04, ->pathfinding: 3.41, Tot: 3.97
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.61, nodes: 0.38, up_inst: 0.09, heur: 1.91, filt: 0.11, cost: 0.03, pushpop: 0.24, edges_next: 0.01, set_next: 0.00, Tot: 3.41
Data - %solved: 7.94, path_costs: 3.440, search_itrs: 11.115, cost-to-go (mean/min/max): 6.59/0.00/9.50
Itr: 2300, loss: 2.48E+00, targ_ctg: 6.49, nnet_ctg: 6.22, lr: 4.17E-05, Time: 2.15
Itr: 2350, loss: 8.97E-01, targ_ctg: 6.68, nnet_ctg: 5.99, lr: 4.17E-05, Time: 2.25
Train - itrs: 100, loss: 8.14E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.89, up_end: 0.25, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.35

Getting Data - itr: 2400, update_num: 24, targ_update: 24, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.44, ->get_states: 0.04, ->pathfinding: 3.48, Tot: 4.02
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.49, nodes: 0.66, up_inst: 0.05, heur: 1.98, filt: 0.11, cost: 0.03, pushpop: 0.12, edges_next: 0.01, set_next: 0.00, Tot: 3.48
Data - %solved: 9.73, path_costs: 3.770, search_itrs: 10.869, cost-to-go (mean/min/max): 6.78/0.00/9.50
Itr: 2400, loss: 1.19E+00, targ_ctg: 6.96, nnet_ctg: 6.00, lr: 4.00E-05, Time: 2.17
Itr: 2450, loss: 1.23E+00, targ_ctg: 7.02, nnet_ctg: 6.08, lr: 4.00E-05, Time: 2.28
Train - itrs: 100, loss: 1.10E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.92, up_end: 0.25, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.38

Getting Data - itr: 2500, update_num: 25, targ_update: 25, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.45, ->get_states: 0.04, ->pathfinding: 3.42, Tot: 3.96
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.31, nodes: 0.52, up_inst: 0.22, heur: 1.92, filt: 0.11, cost: 0.03, pushpop: 0.25, edges_next: 0.01, set_next: 0.00, Tot: 3.42
Data - %solved: 10.00, path_costs: 4.284, search_itrs: 16.653, cost-to-go (mean/min/max): 6.65/0.00/9.45
Itr: 2500, loss: 1.18E+00, targ_ctg: 6.59, nnet_ctg: 6.19, lr: 3.85E-05, Time: 2.15
Itr: 2550, loss: 8.30E-01, targ_ctg: 6.53, nnet_ctg: 6.04, lr: 3.85E-05, Time: 2.25
Train - itrs: 100, loss: 1.03E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.89, up_end: 0.25, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.35

Getting Data - itr: 2600, update_num: 26, targ_update: 26, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.40, ->get_states: 0.04, ->pathfinding: 3.33, Tot: 3.83
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.35, nodes: 0.63, up_inst: 0.08, heur: 1.89, filt: 0.11, cost: 0.03, pushpop: 0.20, edges_next: 0.01, set_next: 0.00, Tot: 3.33
Data - %solved: 8.29, path_costs: 3.569, search_itrs: 10.379, cost-to-go (mean/min/max): 6.69/0.00/9.58
Itr: 2600, loss: 1.11E+00, targ_ctg: 6.51, nnet_ctg: 6.08, lr: 3.70E-05, Time: 2.07
Itr: 2650, loss: 8.42E-01, targ_ctg: 6.80, nnet_ctg: 6.08, lr: 3.70E-05, Time: 2.17
Train - itrs: 100, loss: 1.34E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.83, up_end: 0.24, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.27

Getting Data - itr: 2700, update_num: 27, targ_update: 27, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.44, ->get_states: 0.04, ->pathfinding: 3.41, Tot: 3.96
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.27, nodes: 0.80, up_inst: 0.05, heur: 1.93, filt: 0.11, cost: 0.03, pushpop: 0.20, edges_next: 0.01, set_next: 0.00, Tot: 3.41
Data - %solved: 9.19, path_costs: 3.242, search_itrs: 6.738, cost-to-go (mean/min/max): 6.66/0.00/9.62
Itr: 2700, loss: 1.26E+00, targ_ctg: 6.48, nnet_ctg: 6.14, lr: 3.57E-05, Time: 2.15
Itr: 2750, loss: 8.68E-01, targ_ctg: 6.68, nnet_ctg: 6.07, lr: 3.57E-05, Time: 2.26
Train - itrs: 100, loss: 1.10E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.88, up_end: 0.26, data_samp: 0.00, train: 0.22, save_net: 0.00, save_status: 0.00, Tot: 2.37

Getting Data - itr: 2800, update_num: 28, targ_update: 28, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.43, ->get_states: 0.04, ->pathfinding: 3.42, Tot: 3.94
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.40, nodes: 0.63, up_inst: 0.10, heur: 1.95, filt: 0.11, cost: 0.03, pushpop: 0.16, edges_next: 0.01, set_next: 0.00, Tot: 3.42
Data - %solved: 9.14, path_costs: 4.069, search_itrs: 14.251, cost-to-go (mean/min/max): 6.76/0.00/10.16
Itr: 2800, loss: 1.20E+00, targ_ctg: 6.72, nnet_ctg: 6.06, lr: 3.45E-05, Time: 2.15
Itr: 2850, loss: 1.23E+00, targ_ctg: 6.83, nnet_ctg: 6.11, lr: 3.45E-05, Time: 2.25
Train - itrs: 100, loss: 1.22E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.89, up_end: 0.25, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.35

Getting Data - itr: 2900, update_num: 29, targ_update: 29, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.44, ->get_states: 0.03, ->pathfinding: 3.44, Tot: 3.98
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.67, nodes: 0.44, up_inst: 0.09, heur: 1.94, filt: 0.11, cost: 0.03, pushpop: 0.12, edges_next: 0.01, set_next: 0.00, Tot: 3.44
Data - %solved: 8.16, path_costs: 3.375, search_itrs: 9.731, cost-to-go (mean/min/max): 6.81/0.00/9.65
Itr: 2900, loss: 9.12E-01, targ_ctg: 6.76, nnet_ctg: 6.18, lr: 3.33E-05, Time: 2.16
Itr: 2950, loss: 1.48E+00, targ_ctg: 6.58, nnet_ctg: 6.23, lr: 3.33E-05, Time: 2.28
Train - itrs: 100, loss: 1.63E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.89, up_end: 0.27, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.38

Getting Data - itr: 3000, update_num: 30, targ_update: 30, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.42, ->get_states: 0.04, ->pathfinding: 3.55, Tot: 4.07
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.40, nodes: 0.75, up_inst: 0.05, heur: 1.99, filt: 0.11, cost: 0.03, pushpop: 0.19, edges_next: 0.01, set_next: 0.00, Tot: 3.55
Data - %solved: 8.55, path_costs: 4.032, search_itrs: 12.454, cost-to-go (mean/min/max): 6.80/0.00/9.94
Itr: 3000, loss: 1.29E+00, targ_ctg: 6.71, nnet_ctg: 6.19, lr: 3.23E-05, Time: 2.20
Itr: 3050, loss: 9.50E-01, targ_ctg: 6.81, nnet_ctg: 6.20, lr: 3.23E-05, Time: 2.30
Train - itrs: 100, loss: 1.11E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.95, up_end: 0.24, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.40

Getting Data - itr: 3100, update_num: 31, targ_update: 31, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.46, ->get_states: 0.04, ->pathfinding: 3.48, Tot: 4.04
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.32, nodes: 0.47, up_inst: 0.13, heur: 1.96, filt: 0.11, cost: 0.03, pushpop: 0.42, edges_next: 0.01, set_next: 0.00, Tot: 3.48
Data - %solved: 9.43, path_costs: 3.311, search_itrs: 7.451, cost-to-go (mean/min/max): 6.68/0.00/9.75
Itr: 3100, loss: 7.02E-01, targ_ctg: 6.84, nnet_ctg: 6.19, lr: 3.13E-05, Time: 2.19
Itr: 3150, loss: 1.02E+00, targ_ctg: 6.38, nnet_ctg: 6.11, lr: 3.13E-05, Time: 2.29
Train - itrs: 100, loss: 1.66E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.93, up_end: 0.26, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.40

Getting Data - itr: 3200, update_num: 32, targ_update: 32, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.05, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.46, ->get_states: 0.03, ->pathfinding: 3.49, Tot: 4.05
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.40, nodes: 0.63, up_inst: 0.11, heur: 1.96, filt: 0.11, cost: 0.03, pushpop: 0.20, edges_next: 0.01, set_next: 0.00, Tot: 3.49
Data - %solved: 8.39, path_costs: 3.463, search_itrs: 8.173, cost-to-go (mean/min/max): 6.92/0.00/9.98
Itr: 3200, loss: 9.77E-01, targ_ctg: 6.88, nnet_ctg: 6.06, lr: 3.03E-05, Time: 2.20
Itr: 3250, loss: 1.02E+00, targ_ctg: 7.01, nnet_ctg: 6.26, lr: 3.03E-05, Time: 2.30
Train - itrs: 100, loss: 9.16E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.92, up_end: 0.28, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.40

Getting Data - itr: 3300, update_num: 33, targ_update: 33, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.41, ->get_states: 0.04, ->pathfinding: 3.49, Tot: 3.99
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.56, nodes: 0.65, up_inst: 0.05, heur: 1.93, filt: 0.11, cost: 0.03, pushpop: 0.11, edges_next: 0.01, set_next: 0.00, Tot: 3.49
Data - %solved: 9.28, path_costs: 3.919, search_itrs: 9.557, cost-to-go (mean/min/max): 6.73/0.00/10.04
Itr: 3300, loss: 1.59E+00, targ_ctg: 6.82, nnet_ctg: 6.29, lr: 2.94E-05, Time: 2.15
Itr: 3350, loss: 6.31E-01, targ_ctg: 6.78, nnet_ctg: 6.15, lr: 2.94E-05, Time: 2.25
Train - itrs: 100, loss: 8.74E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.92, up_end: 0.23, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.35

Getting Data - itr: 3400, update_num: 34, targ_update: 34, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.25, Tot: 3.73
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.43, nodes: 0.58, up_inst: 0.12, heur: 1.83, filt: 0.11, cost: 0.03, pushpop: 0.11, edges_next: 0.01, set_next: 0.00, Tot: 3.25
Data - %solved: 9.37, path_costs: 4.089, search_itrs: 15.284, cost-to-go (mean/min/max): 6.74/0.00/10.06
Itr: 3400, loss: 1.72E+00, targ_ctg: 6.72, nnet_ctg: 6.15, lr: 2.86E-05, Time: 2.03
Itr: 3450, loss: 9.60E-01, targ_ctg: 6.91, nnet_ctg: 6.12, lr: 2.86E-05, Time: 2.12
Train - itrs: 100, loss: 1.21E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.23, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.22

Getting Data - itr: 3500, update_num: 35, targ_update: 35, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.43, ->get_states: 0.04, ->pathfinding: 3.31, Tot: 3.84
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.53, nodes: 0.47, up_inst: 0.17, heur: 1.85, filt: 0.11, cost: 0.03, pushpop: 0.11, edges_next: 0.01, set_next: 0.00, Tot: 3.31
Data - %solved: 9.51, path_costs: 4.034, search_itrs: 13.736, cost-to-go (mean/min/max): 6.89/0.00/10.32
Itr: 3500, loss: 1.21E+00, targ_ctg: 7.11, nnet_ctg: 6.11, lr: 2.78E-05, Time: 2.08
Itr: 3550, loss: 1.25E+00, targ_ctg: 6.67, nnet_ctg: 6.23, lr: 2.78E-05, Time: 2.19
Train - itrs: 100, loss: 1.18E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.83, up_end: 0.25, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.29

Getting Data - itr: 3600, update_num: 36, targ_update: 36, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.05, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.48, ->get_states: 0.04, ->pathfinding: 3.42, Tot: 4.00
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.51, nodes: 0.62, up_inst: 0.05, heur: 1.92, filt: 0.11, cost: 0.03, pushpop: 0.16, edges_next: 0.01, set_next: 0.00, Tot: 3.42
Data - %solved: 9.15, path_costs: 3.494, search_itrs: 8.745, cost-to-go (mean/min/max): 6.71/0.00/10.32
Itr: 3600, loss: 8.71E-01, targ_ctg: 6.77, nnet_ctg: 6.24, lr: 2.70E-05, Time: 2.17
Itr: 3650, loss: 1.16E+00, targ_ctg: 6.43, nnet_ctg: 6.16, lr: 2.70E-05, Time: 2.27
Train - itrs: 100, loss: 6.98E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.91, up_end: 0.26, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.37

Getting Data - itr: 3700, update_num: 37, targ_update: 37, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.45, ->get_states: 0.03, ->pathfinding: 3.43, Tot: 3.98
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.41, nodes: 0.42, up_inst: 0.17, heur: 1.94, filt: 0.11, cost: 0.03, pushpop: 0.31, edges_next: 0.01, set_next: 0.00, Tot: 3.43
Data - %solved: 8.01, path_costs: 3.696, search_itrs: 7.747, cost-to-go (mean/min/max): 7.09/0.00/10.15
Itr: 3700, loss: 1.33E+00, targ_ctg: 7.09, nnet_ctg: 6.08, lr: 2.63E-05, Time: 2.16
Itr: 3750, loss: 1.58E+00, targ_ctg: 7.31, nnet_ctg: 6.32, lr: 2.63E-05, Time: 2.26
Train - itrs: 100, loss: 7.45E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.89, up_end: 0.26, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.37

Getting Data - itr: 3800, update_num: 38, targ_update: 38, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.41, ->get_states: 0.03, ->pathfinding: 3.35, Tot: 3.86
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.47, nodes: 0.45, up_inst: 0.18, heur: 1.91, filt: 0.11, cost: 0.03, pushpop: 0.16, edges_next: 0.01, set_next: 0.00, Tot: 3.35
Data - %solved: 9.21, path_costs: 3.591, search_itrs: 10.051, cost-to-go (mean/min/max): 6.75/0.00/10.23
Itr: 3800, loss: 1.25E+00, targ_ctg: 6.85, nnet_ctg: 6.43, lr: 2.56E-05, Time: 2.10
Itr: 3850, loss: 1.20E+00, targ_ctg: 6.79, nnet_ctg: 6.16, lr: 2.56E-05, Time: 2.21
Train - itrs: 100, loss: 8.41E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.84, up_end: 0.26, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.32

Getting Data - itr: 3900, update_num: 39, targ_update: 39, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.43, ->get_states: 0.03, ->pathfinding: 3.37, Tot: 3.89
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.33, nodes: 0.59, up_inst: 0.16, heur: 1.91, filt: 0.11, cost: 0.03, pushpop: 0.21, edges_next: 0.01, set_next: 0.00, Tot: 3.37
Data - %solved: 8.99, path_costs: 4.219, search_itrs: 9.448, cost-to-go (mean/min/max): 6.89/0.00/10.40
Itr: 3900, loss: 1.04E+00, targ_ctg: 6.91, nnet_ctg: 6.19, lr: 2.50E-05, Time: 2.11
Itr: 3950, loss: 7.66E-01, targ_ctg: 6.92, nnet_ctg: 6.27, lr: 2.50E-05, Time: 2.21
Train - itrs: 100, loss: 1.26E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.87, up_end: 0.23, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.32

Getting Data - itr: 4000, update_num: 40, targ_update: 40, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.05, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.51, ->get_states: 0.04, ->pathfinding: 3.43, Tot: 4.04
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.48, nodes: 0.48, up_inst: 0.09, heur: 1.93, filt: 0.11, cost: 0.03, pushpop: 0.28, edges_next: 0.01, set_next: 0.00, Tot: 3.43
Data - %solved: 8.82, path_costs: 3.614, search_itrs: 6.939, cost-to-go (mean/min/max): 6.94/0.00/10.23
Itr: 4000, loss: 1.44E+00, targ_ctg: 6.92, nnet_ctg: 6.29, lr: 2.44E-05, Time: 2.18
Itr: 4050, loss: 1.26E+00, targ_ctg: 7.13, nnet_ctg: 6.30, lr: 2.44E-05, Time: 2.29
Train - itrs: 100, loss: 1.24E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.91, up_end: 0.27, data_samp: 0.00, train: 0.22, save_net: 0.00, save_status: 0.00, Tot: 2.41

Getting Data - itr: 4100, update_num: 41, targ_update: 41, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.42, ->get_states: 0.04, ->pathfinding: 3.42, Tot: 3.93
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.45, nodes: 0.56, up_inst: 0.17, heur: 1.93, filt: 0.11, cost: 0.03, pushpop: 0.13, edges_next: 0.01, set_next: 0.00, Tot: 3.42
Data - %solved: 8.83, path_costs: 3.483, search_itrs: 11.132, cost-to-go (mean/min/max): 6.86/0.00/10.36
Itr: 4100, loss: 1.35E+00, targ_ctg: 6.63, nnet_ctg: 6.35, lr: 2.38E-05, Time: 2.13
Itr: 4150, loss: 7.12E-01, targ_ctg: 6.97, nnet_ctg: 6.31, lr: 2.38E-05, Time: 2.23
Train - itrs: 100, loss: 1.27E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.88, up_end: 0.24, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.33

Getting Data - itr: 4200, update_num: 42, targ_update: 42, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.46, ->get_states: 0.03, ->pathfinding: 3.31, Tot: 3.86
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.53, nodes: 0.27, up_inst: 0.19, heur: 1.85, filt: 0.11, cost: 0.03, pushpop: 0.29, edges_next: 0.01, set_next: 0.00, Tot: 3.31
Data - %solved: 8.90, path_costs: 3.395, search_itrs: 7.719, cost-to-go (mean/min/max): 7.07/0.00/10.48
Itr: 4200, loss: 1.00E+00, targ_ctg: 7.00, nnet_ctg: 6.28, lr: 2.33E-05, Time: 2.09
Itr: 4250, loss: 7.76E-01, targ_ctg: 6.88, nnet_ctg: 6.43, lr: 2.33E-05, Time: 2.19
Train - itrs: 100, loss: 1.51E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.84, up_end: 0.25, data_samp: 0.00, train: 0.22, save_net: 0.00, save_status: 0.00, Tot: 2.31

Getting Data - itr: 4300, update_num: 43, targ_update: 43, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.41, ->get_states: 0.04, ->pathfinding: 3.35, Tot: 3.85
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.41, nodes: 0.64, up_inst: 0.05, heur: 1.90, filt: 0.11, cost: 0.03, pushpop: 0.19, edges_next: 0.01, set_next: 0.00, Tot: 3.35
Data - %solved: 8.05, path_costs: 3.355, search_itrs: 10.598, cost-to-go (mean/min/max): 6.94/0.00/10.32
Itr: 4300, loss: 6.67E-01, targ_ctg: 6.94, nnet_ctg: 6.40, lr: 2.27E-05, Time: 2.10
Itr: 4350, loss: 1.42E+00, targ_ctg: 6.74, nnet_ctg: 6.36, lr: 2.27E-05, Time: 2.20
Train - itrs: 100, loss: 9.92E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.84, up_end: 0.25, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.31

Getting Data - itr: 4400, update_num: 44, targ_update: 44, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.45, ->get_states: 0.03, ->pathfinding: 3.40, Tot: 3.94
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.38, nodes: 0.60, up_inst: 0.13, heur: 1.93, filt: 0.11, cost: 0.03, pushpop: 0.17, edges_next: 0.01, set_next: 0.00, Tot: 3.40
Data - %solved: 7.87, path_costs: 4.212, search_itrs: 14.380, cost-to-go (mean/min/max): 7.19/0.00/10.44
Itr: 4400, loss: 1.35E+00, targ_ctg: 7.28, nnet_ctg: 6.30, lr: 2.22E-05, Time: 2.15
Itr: 4450, loss: 9.77E-01, targ_ctg: 7.06, nnet_ctg: 6.48, lr: 2.22E-05, Time: 2.25
Train - itrs: 100, loss: 8.25E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.87, up_end: 0.27, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.36

Getting Data - itr: 4500, update_num: 45, targ_update: 45, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.44, ->get_states: 0.04, ->pathfinding: 3.42, Tot: 3.96
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.56, nodes: 0.38, up_inst: 0.22, heur: 1.95, filt: 0.11, cost: 0.03, pushpop: 0.15, edges_next: 0.01, set_next: 0.00, Tot: 3.42
Data - %solved: 9.56, path_costs: 3.813, search_itrs: 8.849, cost-to-go (mean/min/max): 6.71/0.00/10.48
Itr: 4500, loss: 1.06E+00, targ_ctg: 6.77, nnet_ctg: 6.53, lr: 2.17E-05, Time: 2.15
Itr: 4550, loss: 1.25E+00, targ_ctg: 6.84, nnet_ctg: 6.10, lr: 2.17E-05, Time: 2.26
Train - itrs: 100, loss: 1.36E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.88, up_end: 0.26, data_samp: 0.00, train: 0.21, save_net: 0.00, save_status: 0.00, Tot: 2.37

Getting Data - itr: 4600, update_num: 46, targ_update: 46, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.42, ->get_states: 0.04, ->pathfinding: 3.40, Tot: 3.92
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.42, nodes: 0.52, up_inst: 0.20, heur: 1.92, filt: 0.11, cost: 0.03, pushpop: 0.16, edges_next: 0.01, set_next: 0.00, Tot: 3.40
Data - %solved: 9.35, path_costs: 3.756, search_itrs: 8.527, cost-to-go (mean/min/max): 6.89/0.00/10.34
Itr: 4600, loss: 1.29E+00, targ_ctg: 6.87, nnet_ctg: 6.11, lr: 2.13E-05, Time: 2.15
Itr: 4650, loss: 1.38E+00, targ_ctg: 6.83, nnet_ctg: 6.22, lr: 2.13E-05, Time: 2.25
Train - itrs: 100, loss: 6.50E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.87, up_end: 0.27, data_samp: 0.00, train: 0.19, save_net: 0.00, save_status: 0.00, Tot: 2.34

Getting Data - itr: 4700, update_num: 47, targ_update: 47, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.43, ->get_states: 0.04, ->pathfinding: 3.37, Tot: 3.90
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.51, nodes: 0.42, up_inst: 0.22, heur: 1.88, filt: 0.11, cost: 0.03, pushpop: 0.15, edges_next: 0.01, set_next: 0.00, Tot: 3.37
Data - %solved: 9.90, path_costs: 4.158, search_itrs: 10.645, cost-to-go (mean/min/max): 6.86/0.00/10.38
Itr: 4700, loss: 9.57E-01, targ_ctg: 6.98, nnet_ctg: 6.26, lr: 2.08E-05, Time: 2.11
Itr: 4750, loss: 1.03E+00, targ_ctg: 6.79, nnet_ctg: 6.24, lr: 2.08E-05, Time: 2.21
Train - itrs: 100, loss: 1.75E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.86, up_end: 0.25, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.32

Getting Data - itr: 4800, update_num: 48, targ_update: 48, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, update_perf: 0.00, put: 0.01, gc: 0.43, ->get_states: 0.04, ->pathfinding: 3.51, Tot: 4.04
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.04, Tot: 0.04
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.39, nodes: 0.76, up_inst: 0.10, heur: 1.95, filt: 0.12, cost: 0.03, pushpop: 0.12, edges_next: 0.01, set_next: 0.00, Tot: 3.51
Data - %solved: 10.84, path_costs: 4.437, search_itrs: 14.022, cost-to-go (mean/min/max): 6.77/0.00/10.57
Itr: 4800, loss: 1.28E+00, targ_ctg: 6.75, nnet_ctg: 6.29, lr: 2.04E-05, Time: 2.20
Itr: 4850, loss: 1.55E+00, targ_ctg: 6.40, nnet_ctg: 6.20, lr: 2.04E-05, Time: 2.30
Train - itrs: 100, loss: 9.17E-01, targ_updated: True
Times - up_start: 0.00, up_data: 1.94, up_end: 0.26, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.40

Getting Data - itr: 4900, update_num: 49, targ_update: 49, num_gen: 20,000
Times - steps_gen: 0.00, inst_info: 0.00, inst_add: 0.00, update_perf: 0.00, backup: 0.04, get_tr_data: 0.00, to_np: 0.01, put: 0.01, gc: 0.39, ->get_states: 0.03, ->pathfinding: 3.25, Tot: 3.74
	(get_states): sample_goalstate_goal_pairs: 0.00, random_walk: 0.03, Tot: 0.03
	(pathfinding): root: 0.00, pop: 0.00, is_solved: 0.02, goal: 0.00, expand: 0.27, nodes: 0.62, up_inst: 0.16, heur: 1.82, filt: 0.11, cost: 0.03, pushpop: 0.21, edges_next: 0.01, set_next: 0.00, Tot: 3.25
Data - %solved: 8.70, path_costs: 3.565, search_itrs: 11.840, cost-to-go (mean/min/max): 6.97/0.00/10.43
Itr: 4900, loss: 8.34E-01, targ_ctg: 6.88, nnet_ctg: 6.14, lr: 2.00E-05, Time: 2.02
Itr: 4950, loss: 1.07E+00, targ_ctg: 7.01, nnet_ctg: 6.30, lr: 2.00E-05, Time: 2.12
Train - itrs: 100, loss: 1.25E+00, targ_updated: True
Times - up_start: 0.00, up_data: 1.79, up_end: 0.23, data_samp: 0.00, train: 0.20, save_net: 0.00, save_status: 0.00, Tot: 2.22
Done

Using deepxube train_summary --dir tutorial/heur_asym_loss/models/ we can see that the neural network output is below the target the majority of the time.