import os import sys with open(sys.argv[0]) as f: code = f.read() # read the code of this file ASAP, for logging import copy import time import uuid from dataclasses import dataclass from functools import lru_cache from pathlib import Path os.environ["PYTORCH_ALLOC_CONF"] = "expandable_segments:True" import torch torch.empty( 1, device="cuda", requires_grad=True ).backward() # prevents a bug on some systems import torch.distributed as dist import torch.nn.functional as F from torch import nn, Tensor # use of FlexAttention contributed by @KoszarskyB from torch.nn.attention.flex_attention import BlockMask, flex_attention torch._inductor.config.coordinate_descent_tuning = ( True # we allow this flag for medium track ) # torch._dynamo.config.compiled_autograd = True class Snoo: """ @DominikKallusky, @vishal9-team, @vinaysrao Sparse Nesterov Outer Optimizer (Snoo) is a momentum-based wrapper to any optimizer that can improve the stability and smoothness of the optimization process and thus the quality of large language models (LLM) and other models. Snoo implicitly adds temporal regularization to the parameters, thus smoothing the training trajectory and instilling a bias towards flatter minima and lower parameter norms. Snoo is computationally efficient, incurring minimal overhead in compute and moderate memory usage. """ @torch.no_grad() def __init__(self, model: nn.Module, lr: float, momentum: float, k: int) -> None: self.model = model self.lr = lr self.momentum = momentum self.k = k self.current_step = 0 self.outer_buf = [p.clone() for p in model.parameters()] self.model_params = list(self.model.parameters()) self.optimizer = torch.optim.SGD( self.model.parameters(), lr=lr, momentum=momentum, nesterov=True, fused=True, ) @torch.no_grad() def step( self, ) -> None: if self.current_step % self.k == 0: for p_new, p_old in zip(self.model_params, self.outer_buf): p_new.grad = p_old.data - p_new.data p_new.copy_(p_old, non_blocking=True) self.optimizer.step() for p_new, p_old in zip(self.model_params, self.outer_buf): p_old.copy_(p_new, non_blocking=True) self.current_step += 1 def state_dict(self): state_dict = { "current_step": self.current_step, "lr": self.lr, "momentum": self.momentum, "k": self.k, "outer_buf": [p.clone() for p in self.outer_buf], "optimizer_state_dict": self.optimizer.state_dict(), } return state_dict def load_state_dict(self, state_dict): self.current_step = state_dict["current_step"] self.lr = state_dict["lr"] self.momentum = state_dict["momentum"] self.k = state_dict["k"] for p_src, p_dst in zip(state_dict["outer_buf"], self.outer_buf): p_dst.copy_(p_src) self.optimizer.load_state_dict(state_dict["optimizer_state_dict"]) # ----------------------------------------------------------------------------- # Muon optimizer def zeropower_via_newtonschulz5(G: Tensor) -> Tensor: """ Newton-Schulz iteration to compute the zeroth power / orthogonalization of G. We opt to use a quintic iteration whose coefficients are selected to maximize the slope at zero. For the purpose of minimizing steps, it turns out to be empirically effective to keep increasing the slope at zero even beyond the point where the iteration no longer converges all the way to one everywhere on the interval. This iteration therefore does not produce UV^T but rather something like US'V^T where S' is diagonal with S_{ii}' ∈ [1 - l, 1 + r], which turns out not to hurt model performance at all relative to UV^T, where USV^T = G is the SVD. """ assert ( G.ndim >= 2 ) # batched Muon implementation by @scottjmaddox, and put into practice in the record by @YouJiacheng X = G.bfloat16() if G.size(-2) > G.size(-1): X = X.mT # Ensure spectral norm is at most 1 X = X / (X.norm(dim=(-2, -1), keepdim=True) + 1e-7) # Perform the NS iterations for a, b, c in [ (4.0848, -6.8946, 2.9270), (3.9505, -6.3029, 2.6377), (3.7418, -5.5913, 2.3037), (2.8769, -3.1427, 1.2046), (2.8366, -3.0525, 1.2012), ]: A = X @ X.mT B = ( b * A + c * A @ A ) # quintic computation strategy adapted from suggestion by @jxbz, @leloykun, and @YouJiacheng X = a * X + B @ X if G.size(-2) > G.size(-1): X = X.mT return X @torch.compile def update( acc_bf16_view_u16: Tensor, mantissa: Tensor, momentum_buffer: Tensor, update_smoothing_buffer: Tensor, grad: Tensor, momentum: Tensor, update_smoothing: Tensor, eff_lr: Tensor, eff_weight_decay: Tensor, ): assert acc_bf16_view_u16.dtype == mantissa.dtype == torch.uint16 grad = grad.float() momentum_buffer.copy_(momentum * momentum_buffer + (1 - momentum) * grad) v = zeropower_via_newtonschulz5(momentum * momentum_buffer + (1 - momentum) * grad) update_smoothing_buffer.copy_(update_smoothing * update_smoothing_buffer + (1 - update_smoothing) * v) v = update_smoothing_buffer acc_m_u32 = (acc_bf16_view_u16.to(torch.uint32) << 16) | mantissa.to(torch.uint32) acc_m_u32.view(torch.float32).mul_(1 - eff_weight_decay) acc_m_u32.view(torch.float32).add_(other=v, alpha=-eff_lr) acc_bf16_view_u16.copy_((acc_m_u32 >> 16).to(torch.uint16)) mantissa.copy_(acc_m_u32.to(torch.uint16)) class Muon(torch.optim.Optimizer): """ Muon - MomentUm Orthogonalized by Newton-schulz https://kellerjordan.github.io/posts/muon/ Muon internally runs standard SGD-momentum, and then performs an orthogonalization post- processing step, in which each 2D parameter's update is replaced with the nearest orthogonal matrix. To efficiently orthogonalize each update, we use a Newton-Schulz iteration, which has the advantage that it can be stably run in bfloat16 on the GPU. Warning: This optimizer should not be used for the embedding layer, the final fully connected layer, or any {0,1}-D parameters; those should all be optimized by a standard method (e.g., AdamW). """ def __init__( self, params, lr=0.02, weight_decay=0.01, momentum=0.95, update_smoothing=0.0, rank=0, world_size=1 ): self.rank = rank self.world_size = world_size defaults = dict(lr=lr, weight_decay=weight_decay, momentum=momentum, update_smoothing=update_smoothing) super().__init__(params, defaults) assert all( p.dtype == torch.bfloat16 for group in self.param_groups for p in group["params"] ) @torch.no_grad() def step(self): futures: list[torch.Future] = [] for group in self.param_groups: params: list[Tensor] = group["params"] params_pad = params + [torch.empty_like(params[-1])] * self.world_size momentum = torch._as_tensor_fullprec(group["momentum"]) update_smoothing = torch._as_tensor_fullprec(group["update_smoothing"]) for base_i in range(len(params))[:: self.world_size]: if base_i + self.rank < len(params): p = params[base_i + self.rank] state = self.state[p] if len(state) == 0: state["mantissa"] = torch.zeros_like(p, dtype=torch.uint16) state["momentum_buffer"] = torch.zeros_like( p, dtype=torch.float32 ) state["update_smoothing_buffer"] = torch.zeros_like( p, dtype=torch.bfloat16 ) update( p.view(torch.uint16), state["mantissa"], state["momentum_buffer"], state["update_smoothing_buffer"], p.grad, momentum, update_smoothing, eff_lr=torch._as_tensor_fullprec( group["lr"] * max(1, p.size(-2) / p.size(-1)) ** 0.5 ), eff_weight_decay=torch._as_tensor_fullprec( group["lr"] * group["weight_decay"] * getattr(p, "wd_mul", 1.0) ), ) futures.append( dist.all_gather( params_pad[base_i : base_i + self.world_size], params_pad[base_i + self.rank], async_op=True, ).get_future() ) torch.futures.collect_all(futures).wait() # ----------------------------------------------------------------------------- # PyTorch nn.Module definitions for the model def norm(x: Tensor): return F.rms_norm(x, (x.size(-1),)) @torch.no_grad() def init_linear(w: Tensor): std = 0.5 * (w.size(-1) ** -0.5) # 0.5 is a bit better than the default 1/sqrt(3) bound = (3**0.5) * std return w.uniform_(-bound, bound) class Rotary(nn.Module): def __init__(self, dim: int, max_seq_len: int): super().__init__() # half-truncate RoPE by @YouJiacheng (w/ base freq tuning) angular_freq = (1 / 1024) ** torch.linspace( 0, 1, steps=dim // 4, dtype=torch.float32 ) angular_freq = torch.cat([angular_freq, angular_freq.new_zeros(dim // 4)]) t = torch.arange(max_seq_len, dtype=torch.float32) theta = torch.einsum("i,j -> ij", t, angular_freq) self.cos = nn.Buffer(theta.cos(), persistent=False) self.sin = nn.Buffer(theta.sin(), persistent=False) def forward(self, x_BTHD: Tensor): assert self.cos.size(0) >= x_BTHD.size(-3) cos, sin = ( self.cos[None, : x_BTHD.size(-3), None, :], self.sin[None, : x_BTHD.size(-3), None, :], ) x1, x2 = x_BTHD.to(dtype=torch.float32).chunk(2, dim=-1) y1 = x1 * cos + x2 * sin y2 = x1 * (-sin) + x2 * cos return torch.cat((y1, y2), 3).type_as(x_BTHD) class CausalSelfAttention(nn.Module): def __init__(self, dim: int, num_heads: int, max_seq_len: int, head_dim=128): super().__init__() self.num_heads = num_heads self.head_dim = head_dim hdim = num_heads * head_dim # merged QKV weights: suggested by many, implemented by @fernbear.bsky.social, and further improved by @YouJiacheng # https://x.com/hi_tysam/status/1879699187107033311 self.qkvo_w = nn.Parameter(init_linear(torch.empty(4, hdim, dim)).bfloat16()) self.qkvo_w.detach()[3].zero_() # out zero init suggested by @Grad62304977 self.rotary = Rotary(head_dim, max_seq_len) # scale the attention logits by given constant, instead of the default head_dim**-0.5, by @leloykun # inspired by learnable scalars used by @brendanh0gan https://x.com/hi_tysam/status/1879693583898591283 self.attn_scale = 0.12 def forward( self, x: Tensor, ve: Tensor | None, block_mask: BlockMask, lambdas: Tensor ): B, T = x.size(0), x.size(1) # batch size, sequence length assert B == 1, "Must use batch size = 1 for FlexAttention" q, k, v = ( F.linear(x, self.qkvo_w[:3].flatten(end_dim=1)) .view(B, T, 3 * self.num_heads, self.head_dim) .chunk(3, dim=-2) ) q, k = norm(q), norm(k) # QK norm @Grad62304977 q, k = self.rotary(q), self.rotary(k) v = norm(v) if ve is not None: v = lambdas[0] * v + lambdas[1] * ve.view_as( v ) # @KoszarskyB & @Grad62304977 else: # skip mid-layers token value embeddings by @YouJiacheng v = lambdas[0] * v y = flex_attention( q.transpose(1, 2), k.transpose(1, 2), v.transpose(1, 2), block_mask=block_mask, scale=self.attn_scale, ).transpose(1, 2) y = y.contiguous().view( B, T, self.num_heads * self.head_dim ) # re-assemble all head outputs side by side y = F.linear(y, self.qkvo_w[3]) return y class MLP(nn.Module): def __init__(self, dim: int): super().__init__() hdim = 4 * dim self.fc_w = nn.Parameter(init_linear(torch.empty(hdim, dim)).bfloat16()) self.proj_w = nn.Parameter(torch.zeros(dim, hdim).bfloat16()) self.fc_w.wd_mul = 2.0 self.proj_w.wd_mul = 2.0 def forward(self, x: Tensor): x = F.linear(x, self.fc_w) x = F.relu( x ).square() # https://arxiv.org/abs/2109.08668v2; ~1-2% better than GELU; suggested by @SKYLINEZ007 and @Grad62304977 x = F.linear(x, self.proj_w) return x class Block(nn.Module): def __init__(self, dim: int, num_heads: int, max_seq_len: int, layer_idx: int): super().__init__() # skip attention of blocks.7 (the 8th layer) by @YouJiacheng self.attn = ( CausalSelfAttention(dim, num_heads, max_seq_len) if layer_idx != 7 else None ) self.mlp = MLP(dim) def forward( self, x: Tensor, ve: Tensor | None, x00: Tensor, x01: Tensor, block_mask: BlockMask, lambdas: Tensor, sa_lambdas: Tensor, ): x = lambdas[0] * x + lambdas[1] * x00 + lambdas[2] * x01 if self.attn is not None: x = x + self.attn(x, ve, block_mask, sa_lambdas) x = x + self.mlp(norm(x)) return x # ----------------------------------------------------------------------------- # The main model def next_multiple_of_n(v: float | int, *, n: int): return next(x for x in range(n, int(v) + 1 + n, n) if x >= v) class GPT(nn.Module): def __init__( self, vocab_size: int, num_layers: int, num_heads: int, model_dim: int, max_seq_len: int, ): super().__init__() self.embed1 = nn.Embedding(vocab_size, model_dim) self.embed2 = nn.Embedding(vocab_size, model_dim) # token value embeddings by @KoszarskyB - inspired by @Grad62304977's value residual implementation following https://arxiv.org/abs/2410.17897 # value embedding code simplification inspired by @ragulpr https://github.com/KellerJordan/modded-nanogpt/pull/78 self.value_embeds = nn.ModuleList( [nn.Embedding(vocab_size, model_dim) for _ in range(5)] ) self.blocks = nn.ModuleList( [Block(model_dim, num_heads, max_seq_len, i) for i in range(num_layers)] ) # there are only 50257 unique GPT-2 tokens; we extend to nearest multiple of 128 for efficiency. # suggested to me by @Grad62304977. this originates from Karpathy's experiments. self.lm_head_w = nn.Parameter( torch.zeros(next_multiple_of_n(vocab_size, n=128), model_dim) ) # Add learnable skip connection weights for decoder layers assert num_layers % 2 == 0 self.scalars = nn.Parameter( torch.cat( [ torch.ones(num_layers), # skip_weights *[ torch.tensor([1.0, 0.0, 0.0]) for _ in range(num_layers) ], # block lambdas *[ torch.tensor([0.5, 0.5]) for _ in range(num_layers) ], # SA lambdas torch.zeros(1), # Smear-MTP lambda torch.tensor([1.0, 0.0]), # backout lambdas ] ) ) for m in self.modules(): if isinstance(m, nn.Embedding): m.bfloat16() def create_blockmasks(self, input_seq: Tensor, sliding_window_num_blocks: Tensor): BLOCK_SIZE = 128 docs = (input_seq == 50256).cumsum(0) def document_causal(b, h, q_idx, kv_idx): causal_mask = q_idx >= kv_idx document_mask = docs[q_idx] == docs[kv_idx] return causal_mask & document_mask def dense_to_ordered(dense_blockmask: Tensor): num_blocks = dense_blockmask.sum(dim=-1, dtype=torch.int32) indices = ( dense_blockmask.argsort(dim=-1, descending=False, stable=True) .flip(-1) .to(torch.int32) ) return num_blocks[None, None].contiguous(), indices[None, None].contiguous() # manual block mask creation by @YouJiacheng assert len(input_seq) % BLOCK_SIZE == 0 NUM_BLOCKS = len(input_seq) // BLOCK_SIZE block_idx = torch.arange(NUM_BLOCKS, dtype=torch.int32, device="cuda") causal_blockmask_any = block_idx[:, None] >= block_idx causal_blockmask_all = block_idx[:, None] > block_idx docs_low = docs.view(-1, BLOCK_SIZE)[:, 0].contiguous() docs_high = docs.view(-1, BLOCK_SIZE)[:, -1].contiguous() document_blockmask_any = (docs_low[:, None] <= docs_high) & ( docs_high[:, None] >= docs_low ) document_blockmask_all = (docs_low[:, None] == docs_high) & ( docs_high[:, None] == docs_low ) blockmask_any = causal_blockmask_any & document_blockmask_any blockmask_all = causal_blockmask_all & document_blockmask_all partial_kv_num_blocks, partial_kv_indices = dense_to_ordered( blockmask_any & ~blockmask_all ) full_kv_num_blocks, full_kv_indices = dense_to_ordered(blockmask_all) def build_bm(window_size_blocks: Tensor) -> BlockMask: return BlockMask.from_kv_blocks( torch.clamp_max( partial_kv_num_blocks, torch.clamp_min(window_size_blocks - full_kv_num_blocks, 1), ), partial_kv_indices, torch.clamp_max(full_kv_num_blocks, window_size_blocks - 1), full_kv_indices, BLOCK_SIZE=BLOCK_SIZE, mask_mod=document_causal, ) # Long-short SWA block masks by @leloykun & @YouJiacheng, adapated from suggestion by @Grad62304977, following Gemma 2 paper return build_bm(sliding_window_num_blocks), build_bm( sliding_window_num_blocks // 2 ) def forward( self, input_seq: Tensor, target_seq: Tensor, sliding_window_num_blocks: Tensor ): assert input_seq.ndim == 1 ve = [value_embed(input_seq) for value_embed in self.value_embeds] # 012 ... 012 structure on token value embeddings by @YouJiacheng, improved on @leloykun's U-net structure ve = ( [ve[0], ve[1], ve[2], ve[3], ve[4]] + [None] * (len(self.blocks) - 10) + [ve[0], ve[1], ve[2], ve[3], ve[4]] ) assert len(ve) == len(self.blocks) long_bm, short_bm = self.create_blockmasks(input_seq, sliding_window_num_blocks) block_masks = [ long_bm, short_bm, short_bm, short_bm, long_bm, short_bm, short_bm, short_bm, short_bm, short_bm, short_bm, long_bm, short_bm, short_bm, short_bm, long_bm, ] assert len(block_masks) == len(self.blocks) x = x00 = norm( self.embed1(input_seq)[None] ) # use of norm here by @Grad62304977 x01 = norm(self.embed2(input_seq)[None]) skip_connections = [] skip_map = { 9: 6, 10: 4, 11: 2, } skip_weights = self.scalars[: len(self.blocks)] lambdas = self.scalars[1 * len(self.blocks) : 4 * len(self.blocks)].view(-1, 3) sa_lambdas = self.scalars[4 * len(self.blocks) : 6 * len(self.blocks)].view( -1, 2 ) for i in range(len(self.blocks)): if i in skip_map: x = x + skip_weights[skip_map[i]] * skip_connections[skip_map[i]] x = self.blocks[i]( x, ve[i], x00, x01, block_masks[i], lambdas[i], sa_lambdas[i] ) skip_connections.append(x) skip_lambdas = self.scalars[-2:] x = norm(x) * skip_lambdas[0] + norm(skip_connections[11]) * skip_lambdas[1] # Smear-MTP by snimu x = torch.cat( [x[:, :1], x[:, 1:] + self.scalars[-3] * x[:, :-1]], dim=1, ) if self.training: logits: Tensor = F.linear( x.flatten(end_dim=1), self.lm_head_w.bfloat16() ).float() loss = F.cross_entropy( 15 * logits * torch.rsqrt(logits.square() + 225), target_seq ) return loss loss = 0 for i in range(4): logits: Tensor = F.linear( x.flatten(end_dim=1).chunk(4)[i], self.lm_head_w.bfloat16() ).float() loss += ( F.cross_entropy( 15 * logits * torch.rsqrt(logits.square() + 225), target_seq.chunk(4)[i], ) / 4 ) return loss # ----------------------------------------------------------------------------- # Our own simple Distributed Data Loader def _load_data_shard(file: Path): header = torch.from_file( str(file), False, 256, dtype=torch.int32 ) # header is 256 int32 assert header[0] == 20240520, "magic number mismatch in the data .bin file" assert header[1] == 1, "unsupported version" num_tokens = int(header[2]) # number of tokens (claimed) with file.open("rb", buffering=0) as f: tokens = torch.empty( num_tokens, dtype=torch.uint16, pin_memory=True ) # avoid pin_memory copy by @YouJiacheng f.seek(256 * 4) nbytes = f.readinto(tokens.numpy()) # avoid bytes->array copy by @YouJiacheng assert nbytes == 2 * num_tokens, "number of tokens read does not match header" return tokens def distributed_data_generator( filename_pattern: str, batch_size: int, rank: int, world_size: int ): files = sorted(Path.cwd().glob(filename_pattern)) assert batch_size % world_size == 0 local_batch_size = batch_size // world_size file_iter = iter( files ) # use itertools.cycle(files) instead if you want to do multi-epoch training tokens, pos = _load_data_shard(next(file_iter)), 0 while True: if pos + batch_size + 1 >= len(tokens): tokens, pos = _load_data_shard(next(file_iter)), 0 buf = tokens[pos + rank * local_batch_size :][: local_batch_size + 1] inputs = buf[:-1].to( device="cuda", dtype=torch.int32, non_blocking=True ) # no sync on host side; targets = buf[1:].to( device="cuda", dtype=torch.int64, non_blocking=True ) # H2D in another stream isn't helpful. pos += batch_size yield inputs, targets # ----------------------------------------------------------------------------- # int main @dataclass class Hyperparameters: # data train_files = "data/fineweb10B/fineweb_train_*.bin" # input .bin to train on val_files = ( "data/fineweb10B/fineweb_val_*.bin" # input .bin to eval validation loss on ) val_tokens = 10485760 # how many tokens of validation data? it's important to keep this fixed for consistent comparisons train_seq_len = 64 * 1024 # FlexAttention sequence length val_seq_len = 4 * 64 * 1024 # FlexAttention sequence length for validation # optimization num_iterations = 5535 # number of iterations to run cooldown_frac = 0.7 # fraction of training spent cooling down the learning rate final_lr_scale = 0.01 # architecture vocab_size = 50257 # evaluation and logging val_loss_every = ( 125 # every how many steps to evaluate val loss? 0 for only at the end ) save_checkpoint = False args = Hyperparameters() run_id = int(os.environ.get("RUN_ID", 0)) # torchrun sets these env variables rank = int(os.environ["RANK"]) world_size = int(os.environ["WORLD_SIZE"]) assert world_size == 8 # this code is designed for 8xH100 assert torch.cuda.is_available() device = torch.device("cuda", int(os.environ["LOCAL_RANK"])) torch.cuda.set_device(device) dist.init_process_group(backend="nccl", device_id=device) dist.barrier() master_process = rank == 0 # this process will do logging, checkpointing etc. # begin logging if master_process: run_id_full = f"{run_id:03d}_{uuid.uuid4()}" path = f"logs" os.makedirs(path, exist_ok=True) logfile = f"{path}/{run_id_full}.txt" print(logfile) def print0(s, console=False): if master_process: with open(logfile, "a") as f: if console: print(s) print(s, file=f) # begin by printing this file (the Python code) print0(code) print0("=" * 100) # log information about the hardware/software environment this is running on print0(f"Running Python {sys.version}") print0( f"Running PyTorch {torch.version.__version__} compiled for CUDA {torch.version.cuda}" ) def nvidia_smi(): import subprocess # avoid top level import try: return subprocess.run( ["nvidia-smi"], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True ).stdout except FileNotFoundError: print0("nvidia-smi not found. Using library functions instead.", console=True) import pynvml pynvml.nvmlInit() device_count = pynvml.nvmlDeviceGetCount() print0( f"{'GPU':<3} {'Name':<25} {'Temp':<5} {'Power':<10} {'Mem Used':<15} {'Mem Total':<15} {'Util':<6}", console=True, ) for i in range(device_count): handle = pynvml.nvmlDeviceGetHandleByIndex(i) name = pynvml.nvmlDeviceGetName(handle) temp = pynvml.nvmlDeviceGetTemperature(handle, pynvml.NVML_TEMPERATURE_GPU) power = pynvml.nvmlDeviceGetPowerUsage(handle) // 1000 # mW to W mem_info = pynvml.nvmlDeviceGetMemoryInfo(handle) mem_used = mem_info.used // (1024 * 1024) mem_total = mem_info.total // (1024 * 1024) util = pynvml.nvmlDeviceGetUtilizationRates(handle).gpu print0( f"{i:<3} {name:<25} {temp:<5} {power:<10} {mem_used:<15} {mem_total:<15} {util:<6}%", console=True, ) pynvml.nvmlShutdown() print0(nvidia_smi()) print0("=" * 100) ######################################## # Construct model and optimizer # ######################################## model: nn.Module = GPT( vocab_size=args.vocab_size, num_layers=16, num_heads=8, model_dim=1024, max_seq_len=max(args.train_seq_len, args.val_seq_len), ).cuda() for m in model.modules(): if isinstance(m, nn.Embedding): m.bfloat16() for param in model.parameters(): dist.broadcast(param.detach(), 0) # collect the parameters to optimize hidden_matrix_params = sorted( (p for p in model.blocks.parameters() if p.ndim >= 2), key=lambda x: x.size(), reverse=True, ) embed_params = [ *model.embed1.parameters(), *model.embed2.parameters(), *model.value_embeds.parameters(), ] scalar_params = [model.scalars] head_params: list[nn.Parameter] = [model.lm_head_w] # sanity check params_collections = [hidden_matrix_params, embed_params, scalar_params, head_params] optimized_parameters_set = {p for params in params_collections for p in params} assert optimized_parameters_set == {*model.parameters()} assert len(optimized_parameters_set) == sum(len(lst) for lst in params_collections) # init the optimizer(s) adam_param_groups = [ dict(params=head_params, lr=1 / 320), dict(params=embed_params, lr=0.3), dict(params=scalar_params, lr=0.015), ] # small adam epsilon by @YouJiacheng. this is an alternate method of fixing the world_size dependence # discovered by @fernbear.bsky.social https://x.com/hi_tysam/status/1879692937589875094 inner_optimizers = [ torch.optim.AdamW( adam_param_groups, betas=(0.8, 0.95), eps=1e-10, weight_decay=0.0, fused=True ) ] inner_hidden_optim = Muon( hidden_matrix_params, lr=0.03, momentum=0.95, update_smoothing=0.2, rank=rank, world_size=world_size ) inner_optimizers += [inner_hidden_optim] outer_optim = Snoo(model, lr=0.68, momentum=0.37, k=28) all_optimizers: list[torch.optim.Optimizer] = [outer_optim] + inner_optimizers def opt_params(opt: torch.optim.Optimizer) -> list[nn.Parameter]: return [p for group in opt.param_groups for p in group["params"]] opt2params = {opt: opt_params(opt) for opt in inner_optimizers} for opt in inner_optimizers: for group in opt.param_groups: group["initial_lr"] = group["lr"] # learning rate schedule: stable then decay def get_lr(step: int): x = step / args.num_iterations # progress in training assert 0 <= x < 1 if x < 1 - args.cooldown_frac: return 1.0 else: return (1 - x) / args.cooldown_frac * (1 - args.final_lr_scale) + args.final_lr_scale # attention window size schedule: linearly increase @lru_cache(1) def get_window_size_blocks_helper(window_size: int): return torch.tensor(window_size // 128, dtype=torch.int32, pin_memory=True).cuda( non_blocking=True ) def get_window_size_blocks(step: int): x = step / args.num_iterations # progress in training assert 0 <= x <= 1 # Linearly increase the block-wise sliding window size over training 128 -> 1792 # increase by @fernbear.bsky.social; block-wise by @YouJiacheng factor = 4 * x**3 - 6 * x**2 + 3 * x # cubic schedule by @jadenj3o window_size = next_multiple_of_n(3456 * factor, n=128) return get_window_size_blocks_helper(window_size) model: nn.Module = torch.compile(model, dynamic=False) ######################################## # Warmup kernels # ######################################## # Warmup the training kernels, then re-initialize the state so we aren't cheating warmup_steps = 10 initial_state = copy.deepcopy( dict( model=model.state_dict(), optimizers=[opt.state_dict() for opt in all_optimizers], ) ) for _ in range(warmup_steps): inputs = targets = torch.randint( 0, args.vocab_size, size=(args.train_seq_len,), device="cuda" ) model(inputs.to(torch.int32), targets, get_window_size_blocks(0)).backward() for param in model.parameters(): dist.all_reduce(param.grad, op=dist.ReduceOp.AVG) for opt in inner_optimizers: opt.step() outer_optim.step() model.zero_grad(set_to_none=True) model.load_state_dict(initial_state["model"]) for opt, opt_state in zip(all_optimizers, initial_state["optimizers"]): opt.load_state_dict(opt_state) del initial_state ######################################## # Training and validation # ######################################## torch.cuda.reset_peak_memory_stats() train_loader = distributed_data_generator( args.train_files, world_size * args.train_seq_len, rank, world_size ) training_time_ms = 0 # start the clock dist.barrier() t0 = time.perf_counter() # begin training train_steps = args.num_iterations for step in range(train_steps + 1): last_step = step == train_steps # --------------- VALIDATION SECTION ----------------- if last_step or (args.val_loss_every > 0 and step % args.val_loss_every == 0): # stop the clock dist.barrier() training_time_ms += 1000 * (time.perf_counter() - t0) model.eval() val_batch_size = world_size * args.val_seq_len assert args.val_tokens % val_batch_size == 0 val_steps = args.val_tokens // val_batch_size val_loader = distributed_data_generator( args.val_files, val_batch_size, rank, world_size ) val_loss = 0 with torch.no_grad(): for _ in range(val_steps): inputs, targets = next(val_loader) val_loss += model(inputs, targets, get_window_size_blocks(step)) val_loss /= val_steps del val_loader dist.all_reduce(val_loss, op=dist.ReduceOp.AVG) print0( f"step:{step}/{train_steps} val_loss:{val_loss:.6f} train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms/max(step, 1):.2f}ms", console=True, ) model.train() # start the clock again dist.barrier() t0 = time.perf_counter() if last_step: if master_process and args.save_checkpoint: log = dict( step=step, code=code, model=model.state_dict(), optimizers=[opt.state_dict() for opt in all_optimizers], ) os.makedirs(f"logs/{run_id_full}", exist_ok=True) torch.save(log, f"logs/{run_id_full}/state_step{step:06d}.pt") # the last step only has the validation loop, so break to avoid training break # --------------- TRAINING SECTION ----------------- inputs, targets = next(train_loader) model(inputs, targets, get_window_size_blocks(step)).backward() opt2futures = { opt: [ dist.all_reduce(p.grad, op=dist.ReduceOp.AVG, async_op=True).get_future() for p in params ] for opt, params in opt2params.items() } # set optimization hyperparameters for opt in inner_optimizers: for group in opt.param_groups: group["lr"] = group["initial_lr"] * get_lr(step) for group in inner_hidden_optim.param_groups: frac = min(step / 300, 1) # momentum warmup for muon group["momentum"] = (1 - frac) * 0.85 + frac * 0.95 # step the optimizers for opt in inner_optimizers: torch.futures.collect_all(opt2futures[opt]).wait() opt.step() outer_optim.step() # null the gradients model.zero_grad(set_to_none=True) # logging approx_training_time_ms = training_time_ms + 1000 * (time.perf_counter() - t0) print0( f"step:{step+1}/{train_steps} train_time:{approx_training_time_ms:.0f}ms step_avg:{approx_training_time_ms/(step + 1):.2f}ms", console=True, ) print0( f"peak memory allocated: {torch.cuda.max_memory_allocated() // 1024 // 1024} MiB " f"reserved: {torch.cuda.max_memory_reserved() // 1024 // 1024} MiB", console=True, ) dist.destroy_process_group() ==================================================================================================== Running Python 3.10.12 (main, Aug 15 2025, 14:32:43) [GCC 11.4.0] Running PyTorch 2.10.0.dev20251103+cu126 compiled for CUDA 12.6 Mon Nov 3 12:53:02 2025 +---------------------------------------------------------------------------------------+ | NVIDIA-SMI 535.183.06 Driver Version: 535.183.06 CUDA Version: 12.2 | |-----------------------------------------+----------------------+----------------------+ | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |=========================================+======================+======================| | 0 NVIDIA H100 80GB HBM3 On | 00000000:00:0B.0 Off | 0 | | N/A 27C P0 111W / 700W | 5825MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 1 NVIDIA H100 80GB HBM3 On | 00000000:00:0C.0 Off | 0 | | N/A 30C P0 112W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 2 NVIDIA H100 80GB HBM3 On | 00000000:00:0D.0 Off | 0 | | N/A 29C P0 108W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 3 NVIDIA H100 80GB HBM3 On | 00000000:00:0E.0 Off | 0 | | N/A 28C P0 110W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 4 NVIDIA H100 80GB HBM3 On | 00000000:00:0F.0 Off | 0 | | N/A 27C P0 109W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 5 NVIDIA H100 80GB HBM3 On | 00000000:00:10.0 Off | 0 | | N/A 29C P0 109W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 6 NVIDIA H100 80GB HBM3 On | 00000000:00:11.0 Off | 0 | | N/A 30C P0 108W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ | 7 NVIDIA H100 80GB HBM3 On | 00000000:00:12.0 Off | 0 | | N/A 28C P0 109W / 700W | 1515MiB / 81559MiB | 0% Default | | | | Disabled | +-----------------------------------------+----------------------+----------------------+ +---------------------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=======================================================================================| | 0 N/A N/A 313827 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 0 N/A N/A 313828 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 0 N/A N/A 313829 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 0 N/A N/A 313830 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 0 N/A N/A 313831 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 0 N/A N/A 313832 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 0 N/A N/A 313833 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 0 N/A N/A 313834 C ...tu/modded-nanogpt/.venv/bin/python3 610MiB | | 1 N/A N/A 313828 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 2 N/A N/A 313829 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 3 N/A N/A 313830 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 4 N/A N/A 313831 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 5 N/A N/A 313832 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 6 N/A N/A 313833 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | | 7 N/A N/A 313834 C ...tu/modded-nanogpt/.venv/bin/python3 1506MiB | +---------------------------------------------------------------------------------------+ ==================================================================================================== step:0/5535 val_loss:10.825840 train_time:0ms step_avg:0.13ms step:1/5535 train_time:90ms step_avg:89.75ms step:2/5535 train_time:268ms step_avg:134.08ms step:3/5535 train_time:478ms step_avg:159.45ms step:4/5535 train_time:700ms step_avg:174.89ms step:5/5535 train_time:924ms step_avg:184.78ms step:6/5535 train_time:1150ms step_avg:191.62ms step:7/5535 train_time:1384ms step_avg:197.68ms step:8/5535 train_time:1612ms step_avg:201.46ms step:9/5535 train_time:1838ms step_avg:204.18ms step:10/5535 train_time:2061ms step_avg:206.15ms step:11/5535 train_time:2291ms step_avg:208.28ms step:12/5535 train_time:2522ms step_avg:210.16ms step:13/5535 train_time:2749ms step_avg:211.49ms step:14/5535 train_time:2976ms step_avg:212.56ms step:15/5535 train_time:3203ms step_avg:213.51ms step:16/5535 train_time:3434ms step_avg:214.63ms step:17/5535 train_time:3664ms step_avg:215.53ms step:18/5535 train_time:3890ms step_avg:216.09ms step:19/5535 train_time:4117ms step_avg:216.66ms step:20/5535 train_time:4345ms step_avg:217.27ms step:21/5535 train_time:4576ms step_avg:217.89ms step:22/5535 train_time:4802ms step_avg:218.28ms step:23/5535 train_time:5029ms step_avg:218.67ms step:24/5535 train_time:5256ms step_avg:219.01ms step:25/5535 train_time:5485ms step_avg:219.40ms step:26/5535 train_time:5714ms step_avg:219.77ms step:27/5535 train_time:5941ms step_avg:220.03ms step:28/5535 train_time:6168ms step_avg:220.29ms step:29/5535 train_time:6423ms step_avg:221.47ms step:30/5535 train_time:6631ms step_avg:221.02ms step:31/5535 train_time:6857ms step_avg:221.20ms step:32/5535 train_time:7084ms step_avg:221.37ms step:33/5535 train_time:7313ms step_avg:221.60ms step:34/5535 train_time:7540ms step_avg:221.77ms step:35/5535 train_time:7770ms step_avg:222.00ms step:36/5535 train_time:7996ms step_avg:222.12ms step:37/5535 train_time:8225ms step_avg:222.29ms step:38/5535 train_time:8452ms step_avg:222.43ms step:39/5535 train_time:8680ms step_avg:222.58ms step:40/5535 train_time:8909ms step_avg:222.72ms step:41/5535 train_time:9136ms step_avg:222.84ms step:42/5535 train_time:9364ms step_avg:222.95ms step:43/5535 train_time:9592ms step_avg:223.07ms step:44/5535 train_time:9820ms step_avg:223.18ms step:45/5535 train_time:10049ms step_avg:223.31ms step:46/5535 train_time:10277ms step_avg:223.41ms step:47/5535 train_time:10503ms step_avg:223.47ms step:48/5535 train_time:10732ms step_avg:223.58ms step:49/5535 train_time:10960ms step_avg:223.68ms step:50/5535 train_time:11188ms step_avg:223.77ms step:51/5535 train_time:11414ms step_avg:223.81ms step:52/5535 train_time:11641ms step_avg:223.87ms step:53/5535 train_time:11872ms step_avg:224.00ms step:54/5535 train_time:12099ms step_avg:224.06ms step:55/5535 train_time:12326ms step_avg:224.11ms step:56/5535 train_time:12554ms step_avg:224.17ms step:57/5535 train_time:12808ms step_avg:224.70ms step:58/5535 train_time:13018ms step_avg:224.44ms step:59/5535 train_time:13244ms step_avg:224.47ms step:60/5535 train_time:13471ms step_avg:224.52ms step:61/5535 train_time:13699ms step_avg:224.58ms step:62/5535 train_time:13927ms step_avg:224.64ms step:63/5535 train_time:14155ms step_avg:224.68ms step:64/5535 train_time:14383ms step_avg:224.73ms step:65/5535 train_time:14610ms step_avg:224.77ms step:66/5535 train_time:14837ms step_avg:224.80ms step:67/5535 train_time:15065ms step_avg:224.85ms step:68/5535 train_time:15292ms step_avg:224.88ms step:69/5535 train_time:15520ms step_avg:224.92ms step:70/5535 train_time:15746ms step_avg:224.94ms step:71/5535 train_time:15974ms step_avg:224.98ms step:72/5535 train_time:16201ms step_avg:225.02ms step:73/5535 train_time:16431ms step_avg:225.09ms step:74/5535 train_time:16659ms step_avg:225.12ms step:75/5535 train_time:16888ms step_avg:225.18ms step:76/5535 train_time:17118ms step_avg:225.23ms step:77/5535 train_time:17346ms step_avg:225.28ms step:78/5535 train_time:17576ms step_avg:225.33ms step:79/5535 train_time:17803ms step_avg:225.36ms step:80/5535 train_time:18032ms step_avg:225.40ms step:81/5535 train_time:18262ms step_avg:225.45ms step:82/5535 train_time:18492ms step_avg:225.52ms step:83/5535 train_time:18720ms step_avg:225.55ms step:84/5535 train_time:18949ms step_avg:225.58ms step:85/5535 train_time:19206ms step_avg:225.95ms step:86/5535 train_time:19414ms step_avg:225.74ms step:87/5535 train_time:19644ms step_avg:225.80ms step:88/5535 train_time:19874ms step_avg:225.84ms step:89/5535 train_time:20103ms step_avg:225.87ms step:90/5535 train_time:20332ms step_avg:225.91ms step:91/5535 train_time:20560ms step_avg:225.93ms step:92/5535 train_time:20788ms step_avg:225.95ms step:93/5535 train_time:21015ms step_avg:225.97ms step:94/5535 train_time:21242ms step_avg:225.98ms step:95/5535 train_time:21471ms step_avg:226.01ms step:96/5535 train_time:21701ms step_avg:226.05ms step:97/5535 train_time:21927ms step_avg:226.05ms step:98/5535 train_time:22156ms step_avg:226.09ms step:99/5535 train_time:22383ms step_avg:226.10ms step:100/5535 train_time:22613ms step_avg:226.13ms step:101/5535 train_time:22840ms step_avg:226.14ms step:102/5535 train_time:23069ms step_avg:226.17ms step:103/5535 train_time:23297ms step_avg:226.18ms step:104/5535 train_time:23526ms step_avg:226.21ms step:105/5535 train_time:23755ms step_avg:226.23ms step:106/5535 train_time:23983ms step_avg:226.25ms step:107/5535 train_time:24211ms step_avg:226.27ms step:108/5535 train_time:24439ms step_avg:226.28ms step:109/5535 train_time:24668ms step_avg:226.31ms step:110/5535 train_time:24895ms step_avg:226.32ms step:111/5535 train_time:25123ms step_avg:226.34ms step:112/5535 train_time:25353ms step_avg:226.36ms step:113/5535 train_time:25608ms step_avg:226.62ms step:114/5535 train_time:25816ms step_avg:226.46ms step:115/5535 train_time:26043ms step_avg:226.46ms step:116/5535 train_time:26271ms step_avg:226.48ms step:117/5535 train_time:26499ms step_avg:226.49ms step:118/5535 train_time:26728ms step_avg:226.51ms step:119/5535 train_time:26956ms step_avg:226.52ms step:120/5535 train_time:27183ms step_avg:226.53ms step:121/5535 train_time:27413ms step_avg:226.55ms step:122/5535 train_time:27639ms step_avg:226.55ms step:123/5535 train_time:27867ms step_avg:226.56ms step:124/5535 train_time:28096ms step_avg:226.58ms step:125/5535 train_time:28325ms step_avg:226.60ms step:125/5535 val_loss:4.263118 train_time:28554ms step_avg:228.43ms step:126/5535 train_time:28581ms step_avg:226.83ms step:127/5535 train_time:28780ms step_avg:226.61ms step:128/5535 train_time:29012ms step_avg:226.65ms step:129/5535 train_time:29246ms step_avg:226.72ms step:130/5535 train_time:29471ms step_avg:226.70ms step:131/5535 train_time:29699ms step_avg:226.71ms step:132/5535 train_time:29925ms step_avg:226.71ms step:133/5535 train_time:30158ms step_avg:226.75ms step:134/5535 train_time:30386ms step_avg:226.76ms step:135/5535 train_time:30612ms step_avg:226.75ms step:136/5535 train_time:30839ms step_avg:226.76ms step:137/5535 train_time:31069ms step_avg:226.78ms step:138/5535 train_time:31300ms step_avg:226.81ms step:139/5535 train_time:31526ms step_avg:226.81ms step:140/5535 train_time:31751ms step_avg:226.79ms step:141/5535 train_time:32006ms step_avg:226.99ms step:142/5535 train_time:32215ms step_avg:226.87ms step:143/5535 train_time:32442ms step_avg:226.87ms step:144/5535 train_time:32667ms step_avg:226.85ms step:145/5535 train_time:32893ms step_avg:226.85ms step:146/5535 train_time:33122ms step_avg:226.86ms step:147/5535 train_time:33353ms step_avg:226.89ms step:148/5535 train_time:33582ms step_avg:226.90ms step:149/5535 train_time:33808ms step_avg:226.90ms step:150/5535 train_time:34038ms step_avg:226.92ms step:151/5535 train_time:34268ms step_avg:226.94ms step:152/5535 train_time:34500ms step_avg:226.97ms step:153/5535 train_time:34727ms step_avg:226.97ms step:154/5535 train_time:34955ms step_avg:226.98ms step:155/5535 train_time:35183ms step_avg:226.99ms step:156/5535 train_time:35411ms step_avg:226.99ms step:157/5535 train_time:35640ms step_avg:227.00ms step:158/5535 train_time:35867ms step_avg:227.00ms step:159/5535 train_time:36097ms step_avg:227.02ms step:160/5535 train_time:36325ms step_avg:227.03ms step:161/5535 train_time:36553ms step_avg:227.04ms step:162/5535 train_time:36783ms step_avg:227.05ms step:163/5535 train_time:37009ms step_avg:227.05ms step:164/5535 train_time:37240ms step_avg:227.07ms step:165/5535 train_time:37467ms step_avg:227.07ms step:166/5535 train_time:37697ms step_avg:227.09ms step:167/5535 train_time:37924ms step_avg:227.09ms step:168/5535 train_time:38153ms step_avg:227.10ms step:169/5535 train_time:38408ms step_avg:227.27ms step:170/5535 train_time:38619ms step_avg:227.17ms step:171/5535 train_time:38843ms step_avg:227.15ms step:172/5535 train_time:39072ms step_avg:227.16ms step:173/5535 train_time:39301ms step_avg:227.17ms step:174/5535 train_time:39528ms step_avg:227.17ms step:175/5535 train_time:39759ms step_avg:227.19ms step:176/5535 train_time:39985ms step_avg:227.19ms step:177/5535 train_time:40212ms step_avg:227.19ms step:178/5535 train_time:40442ms step_avg:227.20ms step:179/5535 train_time:40670ms step_avg:227.21ms step:180/5535 train_time:40899ms step_avg:227.22ms step:181/5535 train_time:41126ms step_avg:227.21ms step:182/5535 train_time:41353ms step_avg:227.21ms step:183/5535 train_time:41583ms step_avg:227.23ms step:184/5535 train_time:41811ms step_avg:227.23ms step:185/5535 train_time:42040ms step_avg:227.24ms step:186/5535 train_time:42267ms step_avg:227.24ms step:187/5535 train_time:42497ms step_avg:227.26ms step:188/5535 train_time:42726ms step_avg:227.26ms step:189/5535 train_time:42953ms step_avg:227.26ms step:190/5535 train_time:43180ms step_avg:227.26ms step:191/5535 train_time:43409ms step_avg:227.27ms step:192/5535 train_time:43638ms step_avg:227.28ms step:193/5535 train_time:43867ms step_avg:227.29ms step:194/5535 train_time:44095ms step_avg:227.29ms step:195/5535 train_time:44323ms step_avg:227.30ms step:196/5535 train_time:44551ms step_avg:227.30ms step:197/5535 train_time:44807ms step_avg:227.45ms step:198/5535 train_time:45014ms step_avg:227.34ms step:199/5535 train_time:45241ms step_avg:227.34ms step:200/5535 train_time:45470ms step_avg:227.35ms step:201/5535 train_time:45697ms step_avg:227.35ms step:202/5535 train_time:45926ms step_avg:227.36ms step:203/5535 train_time:46155ms step_avg:227.36ms step:204/5535 train_time:46382ms step_avg:227.36ms step:205/5535 train_time:46610ms step_avg:227.37ms step:206/5535 train_time:46839ms step_avg:227.37ms step:207/5535 train_time:47066ms step_avg:227.37ms step:208/5535 train_time:47294ms step_avg:227.38ms step:209/5535 train_time:47523ms step_avg:227.38ms step:210/5535 train_time:47751ms step_avg:227.39ms step:211/5535 train_time:47979ms step_avg:227.39ms step:212/5535 train_time:48206ms step_avg:227.39ms step:213/5535 train_time:48435ms step_avg:227.39ms step:214/5535 train_time:48664ms step_avg:227.40ms step:215/5535 train_time:48890ms step_avg:227.40ms step:216/5535 train_time:49118ms step_avg:227.40ms step:217/5535 train_time:49345ms step_avg:227.40ms step:218/5535 train_time:49574ms step_avg:227.40ms step:219/5535 train_time:49803ms step_avg:227.41ms step:220/5535 train_time:50030ms step_avg:227.41ms step:221/5535 train_time:50259ms step_avg:227.42ms step:222/5535 train_time:50488ms step_avg:227.42ms step:223/5535 train_time:50716ms step_avg:227.43ms step:224/5535 train_time:50944ms step_avg:227.43ms step:225/5535 train_time:51205ms step_avg:227.58ms step:226/5535 train_time:51418ms step_avg:227.52ms step:227/5535 train_time:51648ms step_avg:227.52ms step:228/5535 train_time:51880ms step_avg:227.54ms step:229/5535 train_time:52112ms step_avg:227.56ms step:230/5535 train_time:52344ms step_avg:227.58ms step:231/5535 train_time:52578ms step_avg:227.61ms step:232/5535 train_time:52809ms step_avg:227.62ms step:233/5535 train_time:53041ms step_avg:227.64ms step:234/5535 train_time:53272ms step_avg:227.66ms step:235/5535 train_time:53504ms step_avg:227.68ms step:236/5535 train_time:53737ms step_avg:227.70ms step:237/5535 train_time:53968ms step_avg:227.71ms step:238/5535 train_time:54202ms step_avg:227.74ms step:239/5535 train_time:54433ms step_avg:227.75ms step:240/5535 train_time:54665ms step_avg:227.77ms step:241/5535 train_time:54898ms step_avg:227.79ms step:242/5535 train_time:55130ms step_avg:227.81ms step:243/5535 train_time:55362ms step_avg:227.83ms step:244/5535 train_time:55595ms step_avg:227.85ms step:245/5535 train_time:55827ms step_avg:227.87ms step:246/5535 train_time:56060ms step_avg:227.88ms step:247/5535 train_time:56293ms step_avg:227.90ms step:248/5535 train_time:56525ms step_avg:227.92ms step:249/5535 train_time:56756ms step_avg:227.94ms step:250/5535 train_time:56989ms step_avg:227.96ms step:250/5535 val_loss:3.845987 train_time:57223ms step_avg:228.89ms step:251/5535 train_time:57248ms step_avg:228.08ms step:252/5535 train_time:57454ms step_avg:227.99ms step:253/5535 train_time:57721ms step_avg:228.15ms step:254/5535 train_time:57936ms step_avg:228.09ms step:255/5535 train_time:58163ms step_avg:228.09ms step:256/5535 train_time:58392ms step_avg:228.10ms step:257/5535 train_time:58627ms step_avg:228.12ms step:258/5535 train_time:58862ms step_avg:228.15ms step:259/5535 train_time:59093ms step_avg:228.16ms step:260/5535 train_time:59322ms step_avg:228.16ms step:261/5535 train_time:59554ms step_avg:228.17ms step:262/5535 train_time:59789ms step_avg:228.20ms step:263/5535 train_time:60022ms step_avg:228.22ms step:264/5535 train_time:60253ms step_avg:228.23ms step:265/5535 train_time:60485ms step_avg:228.25ms step:266/5535 train_time:60717ms step_avg:228.26ms step:267/5535 train_time:60949ms step_avg:228.28ms step:268/5535 train_time:61181ms step_avg:228.29ms step:269/5535 train_time:61411ms step_avg:228.29ms step:270/5535 train_time:61643ms step_avg:228.31ms step:271/5535 train_time:61875ms step_avg:228.32ms step:272/5535 train_time:62108ms step_avg:228.34ms step:273/5535 train_time:62341ms step_avg:228.35ms step:274/5535 train_time:62572ms step_avg:228.36ms step:275/5535 train_time:62805ms step_avg:228.38ms step:276/5535 train_time:63039ms step_avg:228.40ms step:277/5535 train_time:63272ms step_avg:228.42ms step:278/5535 train_time:63502ms step_avg:228.43ms step:279/5535 train_time:63735ms step_avg:228.44ms step:280/5535 train_time:63968ms step_avg:228.46ms step:281/5535 train_time:64227ms step_avg:228.57ms step:282/5535 train_time:64440ms step_avg:228.51ms step:283/5535 train_time:64669ms step_avg:228.51ms step:284/5535 train_time:64901ms step_avg:228.53ms step:285/5535 train_time:65132ms step_avg:228.53ms step:286/5535 train_time:65364ms step_avg:228.54ms step:287/5535 train_time:65595ms step_avg:228.55ms step:288/5535 train_time:65826ms step_avg:228.56ms step:289/5535 train_time:66059ms step_avg:228.58ms step:290/5535 train_time:66291ms step_avg:228.59ms step:291/5535 train_time:66523ms step_avg:228.60ms step:292/5535 train_time:66756ms step_avg:228.62ms step:293/5535 train_time:66988ms step_avg:228.63ms step:294/5535 train_time:67221ms step_avg:228.64ms step:295/5535 train_time:67452ms step_avg:228.65ms step:296/5535 train_time:67685ms step_avg:228.67ms step:297/5535 train_time:67918ms step_avg:228.68ms step:298/5535 train_time:68151ms step_avg:228.69ms step:299/5535 train_time:68382ms step_avg:228.70ms step:300/5535 train_time:68616ms step_avg:228.72ms step:301/5535 train_time:68847ms step_avg:228.73ms step:302/5535 train_time:69080ms step_avg:228.74ms step:303/5535 train_time:69311ms step_avg:228.75ms step:304/5535 train_time:69543ms step_avg:228.76ms step:305/5535 train_time:69776ms step_avg:228.77ms step:306/5535 train_time:70010ms step_avg:228.79ms step:307/5535 train_time:70243ms step_avg:228.80ms step:308/5535 train_time:70476ms step_avg:228.82ms step:309/5535 train_time:70736ms step_avg:228.92ms step:310/5535 train_time:70950ms step_avg:228.87ms step:311/5535 train_time:71183ms step_avg:228.88ms step:312/5535 train_time:71415ms step_avg:228.90ms step:313/5535 train_time:71647ms step_avg:228.91ms step:314/5535 train_time:71881ms step_avg:228.92ms step:315/5535 train_time:72114ms step_avg:228.93ms step:316/5535 train_time:72347ms step_avg:228.95ms step:317/5535 train_time:72580ms step_avg:228.96ms step:318/5535 train_time:72812ms step_avg:228.97ms step:319/5535 train_time:73047ms step_avg:228.99ms step:320/5535 train_time:73280ms step_avg:229.00ms step:321/5535 train_time:73511ms step_avg:229.01ms step:322/5535 train_time:73744ms step_avg:229.02ms step:323/5535 train_time:73977ms step_avg:229.03ms step:324/5535 train_time:74211ms step_avg:229.05ms step:325/5535 train_time:74444ms step_avg:229.06ms step:326/5535 train_time:74678ms step_avg:229.07ms step:327/5535 train_time:74910ms step_avg:229.08ms step:328/5535 train_time:75142ms step_avg:229.09ms step:329/5535 train_time:75376ms step_avg:229.10ms step:330/5535 train_time:75608ms step_avg:229.12ms step:331/5535 train_time:75842ms step_avg:229.13ms step:332/5535 train_time:76075ms step_avg:229.14ms step:333/5535 train_time:76307ms step_avg:229.15ms step:334/5535 train_time:76539ms step_avg:229.16ms step:335/5535 train_time:76772ms step_avg:229.17ms step:336/5535 train_time:77004ms step_avg:229.18ms step:337/5535 train_time:77264ms step_avg:229.27ms step:338/5535 train_time:77476ms step_avg:229.22ms step:339/5535 train_time:77708ms step_avg:229.23ms step:340/5535 train_time:77941ms step_avg:229.24ms step:341/5535 train_time:78174ms step_avg:229.25ms step:342/5535 train_time:78407ms step_avg:229.26ms step:343/5535 train_time:78640ms step_avg:229.27ms step:344/5535 train_time:78872ms step_avg:229.28ms step:345/5535 train_time:79105ms step_avg:229.29ms step:346/5535 train_time:79338ms step_avg:229.30ms step:347/5535 train_time:79571ms step_avg:229.31ms step:348/5535 train_time:79804ms step_avg:229.32ms step:349/5535 train_time:80036ms step_avg:229.33ms step:350/5535 train_time:80269ms step_avg:229.34ms step:351/5535 train_time:80501ms step_avg:229.35ms step:352/5535 train_time:80732ms step_avg:229.35ms step:353/5535 train_time:80965ms step_avg:229.36ms step:354/5535 train_time:81197ms step_avg:229.37ms step:355/5535 train_time:81430ms step_avg:229.38ms step:356/5535 train_time:81664ms step_avg:229.39ms step:357/5535 train_time:81895ms step_avg:229.40ms step:358/5535 train_time:82127ms step_avg:229.41ms step:359/5535 train_time:82361ms step_avg:229.42ms step:360/5535 train_time:82593ms step_avg:229.42ms step:361/5535 train_time:82826ms step_avg:229.44ms step:362/5535 train_time:83058ms step_avg:229.44ms step:363/5535 train_time:83291ms step_avg:229.45ms step:364/5535 train_time:83524ms step_avg:229.46ms step:365/5535 train_time:83783ms step_avg:229.54ms step:366/5535 train_time:83996ms step_avg:229.50ms step:367/5535 train_time:84228ms step_avg:229.50ms step:368/5535 train_time:84460ms step_avg:229.51ms step:369/5535 train_time:84692ms step_avg:229.52ms step:370/5535 train_time:84925ms step_avg:229.53ms step:371/5535 train_time:85157ms step_avg:229.53ms step:372/5535 train_time:85390ms step_avg:229.54ms step:373/5535 train_time:85623ms step_avg:229.55ms step:374/5535 train_time:85855ms step_avg:229.56ms step:375/5535 train_time:86088ms step_avg:229.57ms step:375/5535 val_loss:3.670916 train_time:86321ms step_avg:230.19ms step:376/5535 train_time:86348ms step_avg:229.65ms step:377/5535 train_time:86550ms step_avg:229.58ms step:378/5535 train_time:86791ms step_avg:229.61ms step:379/5535 train_time:87027ms step_avg:229.62ms step:380/5535 train_time:87257ms step_avg:229.62ms step:381/5535 train_time:87487ms step_avg:229.62ms step:382/5535 train_time:87722ms step_avg:229.64ms step:383/5535 train_time:87957ms step_avg:229.65ms step:384/5535 train_time:88189ms step_avg:229.66ms step:385/5535 train_time:88420ms step_avg:229.66ms step:386/5535 train_time:88653ms step_avg:229.67ms step:387/5535 train_time:88887ms step_avg:229.68ms step:388/5535 train_time:89122ms step_avg:229.69ms step:389/5535 train_time:89353ms step_avg:229.70ms step:390/5535 train_time:89585ms step_avg:229.70ms step:391/5535 train_time:89818ms step_avg:229.71ms step:392/5535 train_time:90051ms step_avg:229.72ms step:393/5535 train_time:90312ms step_avg:229.80ms step:394/5535 train_time:90523ms step_avg:229.75ms step:395/5535 train_time:90755ms step_avg:229.76ms step:396/5535 train_time:90988ms step_avg:229.77ms step:397/5535 train_time:91221ms step_avg:229.77ms step:398/5535 train_time:91457ms step_avg:229.79ms step:399/5535 train_time:91692ms step_avg:229.80ms step:400/5535 train_time:91930ms step_avg:229.82ms step:401/5535 train_time:92166ms step_avg:229.84ms step:402/5535 train_time:92402ms step_avg:229.86ms step:403/5535 train_time:92639ms step_avg:229.87ms step:404/5535 train_time:92875ms step_avg:229.89ms step:405/5535 train_time:93112ms step_avg:229.91ms step:406/5535 train_time:93350ms step_avg:229.93ms step:407/5535 train_time:93588ms step_avg:229.95ms step:408/5535 train_time:93822ms step_avg:229.96ms step:409/5535 train_time:94057ms step_avg:229.97ms step:410/5535 train_time:94294ms step_avg:229.98ms step:411/5535 train_time:94530ms step_avg:230.00ms step:412/5535 train_time:94766ms step_avg:230.01ms step:413/5535 train_time:95002ms step_avg:230.03ms step:414/5535 train_time:95240ms step_avg:230.05ms step:415/5535 train_time:95476ms step_avg:230.06ms step:416/5535 train_time:95712ms step_avg:230.08ms step:417/5535 train_time:95949ms step_avg:230.09ms step:418/5535 train_time:96186ms step_avg:230.11ms step:419/5535 train_time:96423ms step_avg:230.13ms step:420/5535 train_time:96660ms step_avg:230.14ms step:421/5535 train_time:96924ms step_avg:230.22ms step:422/5535 train_time:97141ms step_avg:230.19ms step:423/5535 train_time:97377ms step_avg:230.21ms step:424/5535 train_time:97611ms step_avg:230.21ms step:425/5535 train_time:97846ms step_avg:230.23ms step:426/5535 train_time:98084ms step_avg:230.24ms step:427/5535 train_time:98320ms step_avg:230.26ms step:428/5535 train_time:98557ms step_avg:230.27ms step:429/5535 train_time:98792ms step_avg:230.29ms step:430/5535 train_time:99029ms step_avg:230.30ms step:431/5535 train_time:99264ms step_avg:230.31ms step:432/5535 train_time:99501ms step_avg:230.33ms step:433/5535 train_time:99737ms step_avg:230.34ms step:434/5535 train_time:99972ms step_avg:230.35ms step:435/5535 train_time:100209ms step_avg:230.36ms step:436/5535 train_time:100443ms step_avg:230.37ms step:437/5535 train_time:100680ms step_avg:230.39ms step:438/5535 train_time:100916ms step_avg:230.40ms step:439/5535 train_time:101153ms step_avg:230.42ms step:440/5535 train_time:101389ms step_avg:230.43ms step:441/5535 train_time:101625ms step_avg:230.44ms step:442/5535 train_time:101862ms step_avg:230.46ms step:443/5535 train_time:102097ms step_avg:230.47ms step:444/5535 train_time:102335ms step_avg:230.48ms step:445/5535 train_time:102570ms step_avg:230.49ms step:446/5535 train_time:102808ms step_avg:230.51ms step:447/5535 train_time:103045ms step_avg:230.53ms step:448/5535 train_time:103281ms step_avg:230.54ms step:449/5535 train_time:103544ms step_avg:230.61ms step:450/5535 train_time:103760ms step_avg:230.58ms step:451/5535 train_time:103996ms step_avg:230.59ms step:452/5535 train_time:104232ms step_avg:230.60ms step:453/5535 train_time:104466ms step_avg:230.61ms step:454/5535 train_time:104703ms step_avg:230.62ms step:455/5535 train_time:104941ms step_avg:230.64ms step:456/5535 train_time:105176ms step_avg:230.65ms step:457/5535 train_time:105411ms step_avg:230.66ms step:458/5535 train_time:105648ms step_avg:230.67ms step:459/5535 train_time:105887ms step_avg:230.69ms step:460/5535 train_time:106124ms step_avg:230.70ms step:461/5535 train_time:106360ms step_avg:230.72ms step:462/5535 train_time:106597ms step_avg:230.73ms step:463/5535 train_time:106834ms step_avg:230.74ms step:464/5535 train_time:107070ms step_avg:230.75ms step:465/5535 train_time:107307ms step_avg:230.77ms step:466/5535 train_time:107544ms step_avg:230.78ms step:467/5535 train_time:107779ms step_avg:230.79ms step:468/5535 train_time:108014ms step_avg:230.80ms step:469/5535 train_time:108251ms step_avg:230.81ms step:470/5535 train_time:108489ms step_avg:230.83ms step:471/5535 train_time:108724ms step_avg:230.84ms step:472/5535 train_time:108960ms step_avg:230.85ms step:473/5535 train_time:109196ms step_avg:230.86ms step:474/5535 train_time:109431ms step_avg:230.87ms step:475/5535 train_time:109667ms step_avg:230.88ms step:476/5535 train_time:109904ms step_avg:230.89ms step:477/5535 train_time:110169ms step_avg:230.96ms step:478/5535 train_time:110387ms step_avg:230.93ms step:479/5535 train_time:110620ms step_avg:230.94ms step:480/5535 train_time:110855ms step_avg:230.95ms step:481/5535 train_time:111091ms step_avg:230.96ms step:482/5535 train_time:111328ms step_avg:230.97ms step:483/5535 train_time:111565ms step_avg:230.98ms step:484/5535 train_time:111800ms step_avg:230.99ms step:485/5535 train_time:112036ms step_avg:231.00ms step:486/5535 train_time:112272ms step_avg:231.01ms step:487/5535 train_time:112508ms step_avg:231.02ms step:488/5535 train_time:112744ms step_avg:231.03ms step:489/5535 train_time:112980ms step_avg:231.04ms step:490/5535 train_time:113216ms step_avg:231.05ms step:491/5535 train_time:113452ms step_avg:231.06ms step:492/5535 train_time:113690ms step_avg:231.08ms step:493/5535 train_time:113926ms step_avg:231.09ms step:494/5535 train_time:114163ms step_avg:231.10ms step:495/5535 train_time:114399ms step_avg:231.11ms step:496/5535 train_time:114637ms step_avg:231.12ms step:497/5535 train_time:114873ms step_avg:231.13ms step:498/5535 train_time:115111ms step_avg:231.15ms step:499/5535 train_time:115348ms step_avg:231.16ms step:500/5535 train_time:115584ms step_avg:231.17ms step:500/5535 val_loss:3.556445 train_time:115821ms step_avg:231.64ms step:501/5535 train_time:115849ms step_avg:231.23ms step:502/5535 train_time:116056ms step_avg:231.19ms step:503/5535 train_time:116301ms step_avg:231.22ms step:504/5535 train_time:116538ms step_avg:231.23ms step:505/5535 train_time:116800ms step_avg:231.29ms step:506/5535 train_time:117013ms step_avg:231.25ms step:507/5535 train_time:117253ms step_avg:231.27ms step:508/5535 train_time:117491ms step_avg:231.28ms step:509/5535 train_time:117726ms step_avg:231.29ms step:510/5535 train_time:117961ms step_avg:231.30ms step:511/5535 train_time:118198ms step_avg:231.31ms step:512/5535 train_time:118438ms step_avg:231.32ms step:513/5535 train_time:118674ms step_avg:231.33ms step:514/5535 train_time:118909ms step_avg:231.34ms step:515/5535 train_time:119146ms step_avg:231.35ms step:516/5535 train_time:119385ms step_avg:231.37ms step:517/5535 train_time:119621ms step_avg:231.38ms step:518/5535 train_time:119857ms step_avg:231.38ms step:519/5535 train_time:120093ms step_avg:231.39ms step:520/5535 train_time:120330ms step_avg:231.40ms step:521/5535 train_time:120567ms step_avg:231.42ms step:522/5535 train_time:120805ms step_avg:231.43ms step:523/5535 train_time:121040ms step_avg:231.43ms step:524/5535 train_time:121276ms step_avg:231.44ms step:525/5535 train_time:121513ms step_avg:231.45ms step:526/5535 train_time:121751ms step_avg:231.47ms step:527/5535 train_time:121989ms step_avg:231.48ms step:528/5535 train_time:122225ms step_avg:231.49ms step:529/5535 train_time:122462ms step_avg:231.50ms step:530/5535 train_time:122699ms step_avg:231.51ms step:531/5535 train_time:122937ms step_avg:231.52ms step:532/5535 train_time:123174ms step_avg:231.53ms step:533/5535 train_time:123438ms step_avg:231.59ms step:534/5535 train_time:123654ms step_avg:231.56ms step:535/5535 train_time:123889ms step_avg:231.57ms step:536/5535 train_time:124126ms step_avg:231.58ms step:537/5535 train_time:124362ms step_avg:231.59ms step:538/5535 train_time:124598ms step_avg:231.60ms step:539/5535 train_time:124834ms step_avg:231.60ms step:540/5535 train_time:125072ms step_avg:231.61ms step:541/5535 train_time:125310ms step_avg:231.63ms step:542/5535 train_time:125546ms step_avg:231.63ms step:543/5535 train_time:125783ms step_avg:231.64ms step:544/5535 train_time:126019ms step_avg:231.65ms step:545/5535 train_time:126258ms step_avg:231.67ms step:546/5535 train_time:126496ms step_avg:231.68ms step:547/5535 train_time:126732ms step_avg:231.69ms step:548/5535 train_time:126969ms step_avg:231.70ms step:549/5535 train_time:127206ms step_avg:231.71ms step:550/5535 train_time:127443ms step_avg:231.71ms step:551/5535 train_time:127678ms step_avg:231.72ms step:552/5535 train_time:127917ms step_avg:231.73ms step:553/5535 train_time:128154ms step_avg:231.74ms step:554/5535 train_time:128391ms step_avg:231.75ms step:555/5535 train_time:128629ms step_avg:231.76ms step:556/5535 train_time:128867ms step_avg:231.78ms step:557/5535 train_time:129104ms step_avg:231.79ms step:558/5535 train_time:129340ms step_avg:231.79ms step:559/5535 train_time:129576ms step_avg:231.80ms step:560/5535 train_time:129814ms step_avg:231.81ms step:561/5535 train_time:130079ms step_avg:231.87ms step:562/5535 train_time:130294ms step_avg:231.84ms step:563/5535 train_time:130530ms step_avg:231.85ms step:564/5535 train_time:130766ms step_avg:231.86ms step:565/5535 train_time:131003ms step_avg:231.86ms step:566/5535 train_time:131239ms step_avg:231.87ms step:567/5535 train_time:131476ms step_avg:231.88ms step:568/5535 train_time:131713ms step_avg:231.89ms step:569/5535 train_time:131950ms step_avg:231.90ms step:570/5535 train_time:132184ms step_avg:231.90ms step:571/5535 train_time:132422ms step_avg:231.91ms step:572/5535 train_time:132657ms step_avg:231.92ms step:573/5535 train_time:132894ms step_avg:231.93ms step:574/5535 train_time:133133ms step_avg:231.94ms step:575/5535 train_time:133370ms step_avg:231.95ms step:576/5535 train_time:133608ms step_avg:231.96ms step:577/5535 train_time:133845ms step_avg:231.97ms step:578/5535 train_time:134082ms step_avg:231.98ms step:579/5535 train_time:134319ms step_avg:231.98ms step:580/5535 train_time:134558ms step_avg:232.00ms step:581/5535 train_time:134793ms step_avg:232.00ms step:582/5535 train_time:135030ms step_avg:232.01ms step:583/5535 train_time:135268ms step_avg:232.02ms step:584/5535 train_time:135505ms step_avg:232.03ms step:585/5535 train_time:135742ms step_avg:232.04ms step:586/5535 train_time:135977ms step_avg:232.04ms step:587/5535 train_time:136214ms step_avg:232.05ms step:588/5535 train_time:136453ms step_avg:232.06ms step:589/5535 train_time:136718ms step_avg:232.12ms step:590/5535 train_time:136933ms step_avg:232.09ms step:591/5535 train_time:137169ms step_avg:232.10ms step:592/5535 train_time:137406ms step_avg:232.11ms step:593/5535 train_time:137644ms step_avg:232.11ms step:594/5535 train_time:137880ms step_avg:232.12ms step:595/5535 train_time:138116ms step_avg:232.13ms step:596/5535 train_time:138353ms step_avg:232.14ms step:597/5535 train_time:138593ms step_avg:232.15ms step:598/5535 train_time:138830ms step_avg:232.16ms step:599/5535 train_time:139067ms step_avg:232.17ms step:600/5535 train_time:139303ms step_avg:232.17ms step:601/5535 train_time:139541ms step_avg:232.18ms step:602/5535 train_time:139779ms step_avg:232.19ms step:603/5535 train_time:140022ms step_avg:232.21ms step:604/5535 train_time:140260ms step_avg:232.22ms step:605/5535 train_time:140498ms step_avg:232.23ms step:606/5535 train_time:140737ms step_avg:232.24ms step:607/5535 train_time:140979ms step_avg:232.26ms step:608/5535 train_time:141219ms step_avg:232.27ms step:609/5535 train_time:141458ms step_avg:232.28ms step:610/5535 train_time:141696ms step_avg:232.29ms step:611/5535 train_time:141936ms step_avg:232.30ms step:612/5535 train_time:142176ms step_avg:232.31ms step:613/5535 train_time:142418ms step_avg:232.33ms step:614/5535 train_time:142657ms step_avg:232.34ms step:615/5535 train_time:142895ms step_avg:232.35ms step:616/5535 train_time:143135ms step_avg:232.36ms step:617/5535 train_time:143402ms step_avg:232.42ms step:618/5535 train_time:143624ms step_avg:232.40ms step:619/5535 train_time:143861ms step_avg:232.41ms step:620/5535 train_time:144100ms step_avg:232.42ms step:621/5535 train_time:144340ms step_avg:232.43ms step:622/5535 train_time:144580ms step_avg:232.44ms step:623/5535 train_time:144819ms step_avg:232.45ms step:624/5535 train_time:145060ms step_avg:232.47ms step:625/5535 train_time:145299ms step_avg:232.48ms step:625/5535 val_loss:3.478980 train_time:145540ms step_avg:232.86ms step:626/5535 train_time:145567ms step_avg:232.54ms step:627/5535 train_time:145779ms step_avg:232.50ms step:628/5535 train_time:146029ms step_avg:232.53ms step:629/5535 train_time:146269ms step_avg:232.54ms step:630/5535 train_time:146505ms step_avg:232.55ms step:631/5535 train_time:146742ms step_avg:232.55ms step:632/5535 train_time:146984ms step_avg:232.57ms step:633/5535 train_time:147225ms step_avg:232.58ms step:634/5535 train_time:147463ms step_avg:232.59ms step:635/5535 train_time:147700ms step_avg:232.60ms step:636/5535 train_time:147940ms step_avg:232.61ms step:637/5535 train_time:148182ms step_avg:232.62ms step:638/5535 train_time:148420ms step_avg:232.63ms step:639/5535 train_time:148660ms step_avg:232.64ms step:640/5535 train_time:148900ms step_avg:232.66ms step:641/5535 train_time:149142ms step_avg:232.67ms step:642/5535 train_time:149381ms step_avg:232.68ms step:643/5535 train_time:149620ms step_avg:232.69ms step:644/5535 train_time:149860ms step_avg:232.70ms step:645/5535 train_time:150130ms step_avg:232.76ms step:646/5535 train_time:150351ms step_avg:232.74ms step:647/5535 train_time:150586ms step_avg:232.75ms step:648/5535 train_time:150826ms step_avg:232.76ms step:649/5535 train_time:151069ms step_avg:232.77ms step:650/5535 train_time:151310ms step_avg:232.78ms step:651/5535 train_time:151549ms step_avg:232.79ms step:652/5535 train_time:151789ms step_avg:232.81ms step:653/5535 train_time:152029ms step_avg:232.82ms step:654/5535 train_time:152268ms step_avg:232.83ms step:655/5535 train_time:152510ms step_avg:232.84ms step:656/5535 train_time:152747ms step_avg:232.85ms step:657/5535 train_time:152986ms step_avg:232.86ms step:658/5535 train_time:153227ms step_avg:232.87ms step:659/5535 train_time:153467ms step_avg:232.88ms step:660/5535 train_time:153707ms step_avg:232.89ms step:661/5535 train_time:153945ms step_avg:232.90ms step:662/5535 train_time:154184ms step_avg:232.91ms step:663/5535 train_time:154424ms step_avg:232.92ms step:664/5535 train_time:154664ms step_avg:232.93ms step:665/5535 train_time:154905ms step_avg:232.94ms step:666/5535 train_time:155144ms step_avg:232.95ms step:667/5535 train_time:155384ms step_avg:232.96ms step:668/5535 train_time:155624ms step_avg:232.97ms step:669/5535 train_time:155865ms step_avg:232.98ms step:670/5535 train_time:156104ms step_avg:232.99ms step:671/5535 train_time:156343ms step_avg:233.00ms step:672/5535 train_time:156583ms step_avg:233.01ms step:673/5535 train_time:156850ms step_avg:233.06ms step:674/5535 train_time:157070ms step_avg:233.04ms step:675/5535 train_time:157309ms step_avg:233.05ms step:676/5535 train_time:157549ms step_avg:233.06ms step:677/5535 train_time:157788ms step_avg:233.07ms step:678/5535 train_time:158027ms step_avg:233.08ms step:679/5535 train_time:158267ms step_avg:233.09ms step:680/5535 train_time:158507ms step_avg:233.10ms step:681/5535 train_time:158747ms step_avg:233.11ms step:682/5535 train_time:158986ms step_avg:233.12ms step:683/5535 train_time:159227ms step_avg:233.13ms step:684/5535 train_time:159466ms step_avg:233.14ms step:685/5535 train_time:159705ms step_avg:233.15ms step:686/5535 train_time:159947ms step_avg:233.16ms step:687/5535 train_time:160186ms step_avg:233.17ms step:688/5535 train_time:160426ms step_avg:233.18ms step:689/5535 train_time:160666ms step_avg:233.19ms step:690/5535 train_time:160908ms step_avg:233.20ms step:691/5535 train_time:161147ms step_avg:233.21ms step:692/5535 train_time:161387ms step_avg:233.22ms step:693/5535 train_time:161628ms step_avg:233.23ms step:694/5535 train_time:161868ms step_avg:233.24ms step:695/5535 train_time:162108ms step_avg:233.25ms step:696/5535 train_time:162346ms step_avg:233.26ms step:697/5535 train_time:162588ms step_avg:233.27ms step:698/5535 train_time:162828ms step_avg:233.28ms step:699/5535 train_time:163067ms step_avg:233.29ms step:700/5535 train_time:163309ms step_avg:233.30ms step:701/5535 train_time:163575ms step_avg:233.35ms step:702/5535 train_time:163793ms step_avg:233.32ms step:703/5535 train_time:164032ms step_avg:233.33ms step:704/5535 train_time:164271ms step_avg:233.34ms step:705/5535 train_time:164513ms step_avg:233.35ms step:706/5535 train_time:164756ms step_avg:233.37ms step:707/5535 train_time:164996ms step_avg:233.38ms step:708/5535 train_time:165237ms step_avg:233.39ms step:709/5535 train_time:165476ms step_avg:233.39ms step:710/5535 train_time:165715ms step_avg:233.40ms step:711/5535 train_time:165957ms step_avg:233.41ms step:712/5535 train_time:166198ms step_avg:233.42ms step:713/5535 train_time:166438ms step_avg:233.43ms step:714/5535 train_time:166678ms step_avg:233.44ms step:715/5535 train_time:166918ms step_avg:233.45ms step:716/5535 train_time:167158ms step_avg:233.46ms step:717/5535 train_time:167396ms step_avg:233.47ms step:718/5535 train_time:167636ms step_avg:233.48ms step:719/5535 train_time:167875ms step_avg:233.48ms step:720/5535 train_time:168115ms step_avg:233.49ms step:721/5535 train_time:168358ms step_avg:233.51ms step:722/5535 train_time:168598ms step_avg:233.52ms step:723/5535 train_time:168837ms step_avg:233.52ms step:724/5535 train_time:169078ms step_avg:233.53ms step:725/5535 train_time:169321ms step_avg:233.55ms step:726/5535 train_time:169561ms step_avg:233.56ms step:727/5535 train_time:169801ms step_avg:233.56ms step:728/5535 train_time:170042ms step_avg:233.57ms step:729/5535 train_time:170310ms step_avg:233.62ms step:730/5535 train_time:170532ms step_avg:233.61ms step:731/5535 train_time:170770ms step_avg:233.61ms step:732/5535 train_time:171009ms step_avg:233.62ms step:733/5535 train_time:171248ms step_avg:233.63ms step:734/5535 train_time:171487ms step_avg:233.63ms step:735/5535 train_time:171727ms step_avg:233.64ms step:736/5535 train_time:171968ms step_avg:233.65ms step:737/5535 train_time:172208ms step_avg:233.66ms step:738/5535 train_time:172447ms step_avg:233.67ms step:739/5535 train_time:172687ms step_avg:233.68ms step:740/5535 train_time:172927ms step_avg:233.68ms step:741/5535 train_time:173169ms step_avg:233.70ms step:742/5535 train_time:173408ms step_avg:233.70ms step:743/5535 train_time:173647ms step_avg:233.71ms step:744/5535 train_time:173890ms step_avg:233.72ms step:745/5535 train_time:174132ms step_avg:233.73ms step:746/5535 train_time:174372ms step_avg:233.74ms step:747/5535 train_time:174612ms step_avg:233.75ms step:748/5535 train_time:174852ms step_avg:233.76ms step:749/5535 train_time:175095ms step_avg:233.77ms step:750/5535 train_time:175334ms step_avg:233.78ms step:750/5535 val_loss:3.425456 train_time:175574ms step_avg:234.10ms step:751/5535 train_time:175595ms step_avg:233.82ms step:752/5535 train_time:175816ms step_avg:233.80ms step:753/5535 train_time:176063ms step_avg:233.82ms step:754/5535 train_time:176304ms step_avg:233.83ms step:755/5535 train_time:176541ms step_avg:233.83ms step:756/5535 train_time:176780ms step_avg:233.84ms step:757/5535 train_time:177055ms step_avg:233.89ms step:758/5535 train_time:177276ms step_avg:233.87ms step:759/5535 train_time:177513ms step_avg:233.88ms step:760/5535 train_time:177750ms step_avg:233.88ms step:761/5535 train_time:177992ms step_avg:233.89ms step:762/5535 train_time:178233ms step_avg:233.90ms step:763/5535 train_time:178471ms step_avg:233.91ms step:764/5535 train_time:178709ms step_avg:233.91ms step:765/5535 train_time:178949ms step_avg:233.92ms step:766/5535 train_time:179194ms step_avg:233.93ms step:767/5535 train_time:179434ms step_avg:233.94ms step:768/5535 train_time:179673ms step_avg:233.95ms step:769/5535 train_time:179913ms step_avg:233.96ms step:770/5535 train_time:180155ms step_avg:233.97ms step:771/5535 train_time:180393ms step_avg:233.97ms step:772/5535 train_time:180632ms step_avg:233.98ms step:773/5535 train_time:180871ms step_avg:233.99ms step:774/5535 train_time:181110ms step_avg:233.99ms step:775/5535 train_time:181351ms step_avg:234.00ms step:776/5535 train_time:181591ms step_avg:234.01ms step:777/5535 train_time:181831ms step_avg:234.02ms step:778/5535 train_time:182071ms step_avg:234.02ms step:779/5535 train_time:182311ms step_avg:234.03ms step:780/5535 train_time:182555ms step_avg:234.04ms step:781/5535 train_time:182793ms step_avg:234.05ms step:782/5535 train_time:183034ms step_avg:234.06ms step:783/5535 train_time:183272ms step_avg:234.06ms step:784/5535 train_time:183512ms step_avg:234.07ms step:785/5535 train_time:183780ms step_avg:234.11ms step:786/5535 train_time:183999ms step_avg:234.10ms step:787/5535 train_time:184237ms step_avg:234.10ms step:788/5535 train_time:184479ms step_avg:234.11ms step:789/5535 train_time:184720ms step_avg:234.12ms step:790/5535 train_time:184960ms step_avg:234.13ms step:791/5535 train_time:185199ms step_avg:234.13ms step:792/5535 train_time:185442ms step_avg:234.14ms step:793/5535 train_time:185682ms step_avg:234.15ms step:794/5535 train_time:185923ms step_avg:234.16ms step:795/5535 train_time:186165ms step_avg:234.17ms step:796/5535 train_time:186405ms step_avg:234.18ms step:797/5535 train_time:186646ms step_avg:234.19ms step:798/5535 train_time:186887ms step_avg:234.19ms step:799/5535 train_time:187131ms step_avg:234.21ms step:800/5535 train_time:187369ms step_avg:234.21ms step:801/5535 train_time:187609ms step_avg:234.22ms step:802/5535 train_time:187849ms step_avg:234.23ms step:803/5535 train_time:188089ms step_avg:234.23ms step:804/5535 train_time:188329ms step_avg:234.24ms step:805/5535 train_time:188570ms step_avg:234.25ms step:806/5535 train_time:188811ms step_avg:234.26ms step:807/5535 train_time:189051ms step_avg:234.26ms step:808/5535 train_time:189290ms step_avg:234.27ms step:809/5535 train_time:189529ms step_avg:234.28ms step:810/5535 train_time:189769ms step_avg:234.28ms step:811/5535 train_time:190009ms step_avg:234.29ms step:812/5535 train_time:190250ms step_avg:234.30ms step:813/5535 train_time:190516ms step_avg:234.34ms step:814/5535 train_time:190735ms step_avg:234.32ms step:815/5535 train_time:190974ms step_avg:234.32ms step:816/5535 train_time:191216ms step_avg:234.33ms step:817/5535 train_time:191454ms step_avg:234.34ms step:818/5535 train_time:191692ms step_avg:234.34ms step:819/5535 train_time:191932ms step_avg:234.35ms step:820/5535 train_time:192172ms step_avg:234.36ms step:821/5535 train_time:192412ms step_avg:234.36ms step:822/5535 train_time:192650ms step_avg:234.37ms step:823/5535 train_time:192890ms step_avg:234.37ms step:824/5535 train_time:193129ms step_avg:234.38ms step:825/5535 train_time:193371ms step_avg:234.39ms step:826/5535 train_time:193610ms step_avg:234.39ms step:827/5535 train_time:193850ms step_avg:234.40ms step:828/5535 train_time:194089ms step_avg:234.41ms step:829/5535 train_time:194327ms step_avg:234.41ms step:830/5535 train_time:194569ms step_avg:234.42ms step:831/5535 train_time:194809ms step_avg:234.43ms step:832/5535 train_time:195049ms step_avg:234.43ms step:833/5535 train_time:195288ms step_avg:234.44ms step:834/5535 train_time:195530ms step_avg:234.45ms step:835/5535 train_time:195770ms step_avg:234.46ms step:836/5535 train_time:196011ms step_avg:234.46ms step:837/5535 train_time:196252ms step_avg:234.47ms step:838/5535 train_time:196492ms step_avg:234.48ms step:839/5535 train_time:196731ms step_avg:234.48ms step:840/5535 train_time:196970ms step_avg:234.49ms step:841/5535 train_time:197237ms step_avg:234.53ms step:842/5535 train_time:197458ms step_avg:234.51ms step:843/5535 train_time:197697ms step_avg:234.52ms step:844/5535 train_time:197936ms step_avg:234.52ms step:845/5535 train_time:198177ms step_avg:234.53ms step:846/5535 train_time:198417ms step_avg:234.54ms step:847/5535 train_time:198658ms step_avg:234.54ms step:848/5535 train_time:198899ms step_avg:234.55ms step:849/5535 train_time:199139ms step_avg:234.56ms step:850/5535 train_time:199378ms step_avg:234.56ms step:851/5535 train_time:199623ms step_avg:234.57ms step:852/5535 train_time:199865ms step_avg:234.58ms step:853/5535 train_time:200107ms step_avg:234.59ms step:854/5535 train_time:200350ms step_avg:234.60ms step:855/5535 train_time:200593ms step_avg:234.61ms step:856/5535 train_time:200834ms step_avg:234.62ms step:857/5535 train_time:201078ms step_avg:234.63ms step:858/5535 train_time:201323ms step_avg:234.64ms step:859/5535 train_time:201564ms step_avg:234.65ms step:860/5535 train_time:201808ms step_avg:234.66ms step:861/5535 train_time:202050ms step_avg:234.67ms step:862/5535 train_time:202292ms step_avg:234.68ms step:863/5535 train_time:202538ms step_avg:234.69ms step:864/5535 train_time:202783ms step_avg:234.70ms step:865/5535 train_time:203024ms step_avg:234.71ms step:866/5535 train_time:203266ms step_avg:234.72ms step:867/5535 train_time:203508ms step_avg:234.73ms step:868/5535 train_time:203750ms step_avg:234.74ms step:869/5535 train_time:204019ms step_avg:234.77ms step:870/5535 train_time:204240ms step_avg:234.76ms step:871/5535 train_time:204483ms step_avg:234.77ms step:872/5535 train_time:204726ms step_avg:234.78ms step:873/5535 train_time:204967ms step_avg:234.79ms step:874/5535 train_time:205211ms step_avg:234.79ms step:875/5535 train_time:205456ms step_avg:234.81ms step:875/5535 val_loss:3.380815 train_time:205697ms step_avg:235.08ms step:876/5535 train_time:205718ms step_avg:234.84ms step:877/5535 train_time:205941ms step_avg:234.82ms step:878/5535 train_time:206193ms step_avg:234.84ms step:879/5535 train_time:206435ms step_avg:234.85ms step:880/5535 train_time:206674ms step_avg:234.86ms step:881/5535 train_time:206916ms step_avg:234.86ms step:882/5535 train_time:207163ms step_avg:234.88ms step:883/5535 train_time:207406ms step_avg:234.89ms step:884/5535 train_time:207645ms step_avg:234.89ms step:885/5535 train_time:207887ms step_avg:234.90ms step:886/5535 train_time:208133ms step_avg:234.91ms step:887/5535 train_time:208378ms step_avg:234.92ms step:888/5535 train_time:208619ms step_avg:234.93ms step:889/5535 train_time:208865ms step_avg:234.94ms step:890/5535 train_time:209106ms step_avg:234.95ms step:891/5535 train_time:209352ms step_avg:234.96ms step:892/5535 train_time:209595ms step_avg:234.97ms step:893/5535 train_time:209837ms step_avg:234.98ms step:894/5535 train_time:210080ms step_avg:234.99ms step:895/5535 train_time:210325ms step_avg:235.00ms step:896/5535 train_time:210568ms step_avg:235.01ms step:897/5535 train_time:210839ms step_avg:235.05ms step:898/5535 train_time:211061ms step_avg:235.03ms step:899/5535 train_time:211305ms step_avg:235.04ms step:900/5535 train_time:211548ms step_avg:235.05ms step:901/5535 train_time:211790ms step_avg:235.06ms step:902/5535 train_time:212032ms step_avg:235.07ms step:903/5535 train_time:212280ms step_avg:235.08ms step:904/5535 train_time:212521ms step_avg:235.09ms step:905/5535 train_time:212762ms step_avg:235.10ms step:906/5535 train_time:213004ms step_avg:235.10ms step:907/5535 train_time:213251ms step_avg:235.12ms step:908/5535 train_time:213493ms step_avg:235.12ms step:909/5535 train_time:213734ms step_avg:235.13ms step:910/5535 train_time:213980ms step_avg:235.14ms step:911/5535 train_time:214222ms step_avg:235.15ms step:912/5535 train_time:214465ms step_avg:235.16ms step:913/5535 train_time:214708ms step_avg:235.17ms step:914/5535 train_time:214951ms step_avg:235.18ms step:915/5535 train_time:215195ms step_avg:235.19ms step:916/5535 train_time:215437ms step_avg:235.19ms step:917/5535 train_time:215681ms step_avg:235.20ms step:918/5535 train_time:215924ms step_avg:235.21ms step:919/5535 train_time:216170ms step_avg:235.22ms step:920/5535 train_time:216412ms step_avg:235.23ms step:921/5535 train_time:216655ms step_avg:235.24ms step:922/5535 train_time:216898ms step_avg:235.25ms step:923/5535 train_time:217141ms step_avg:235.26ms step:924/5535 train_time:217384ms step_avg:235.26ms step:925/5535 train_time:217655ms step_avg:235.30ms step:926/5535 train_time:217875ms step_avg:235.29ms step:927/5535 train_time:218118ms step_avg:235.29ms step:928/5535 train_time:218361ms step_avg:235.30ms step:929/5535 train_time:218608ms step_avg:235.31ms step:930/5535 train_time:218850ms step_avg:235.32ms step:931/5535 train_time:219092ms step_avg:235.33ms step:932/5535 train_time:219334ms step_avg:235.34ms step:933/5535 train_time:219579ms step_avg:235.35ms step:934/5535 train_time:219821ms step_avg:235.35ms step:935/5535 train_time:220064ms step_avg:235.36ms step:936/5535 train_time:220308ms step_avg:235.37ms step:937/5535 train_time:220553ms step_avg:235.38ms step:938/5535 train_time:220794ms step_avg:235.39ms step:939/5535 train_time:221037ms step_avg:235.40ms step:940/5535 train_time:221280ms step_avg:235.40ms step:941/5535 train_time:221522ms step_avg:235.41ms step:942/5535 train_time:221764ms step_avg:235.42ms step:943/5535 train_time:222006ms step_avg:235.43ms step:944/5535 train_time:222251ms step_avg:235.44ms step:945/5535 train_time:222497ms step_avg:235.45ms step:946/5535 train_time:222738ms step_avg:235.45ms step:947/5535 train_time:222982ms step_avg:235.46ms step:948/5535 train_time:223224ms step_avg:235.47ms step:949/5535 train_time:223469ms step_avg:235.48ms step:950/5535 train_time:223711ms step_avg:235.49ms step:951/5535 train_time:223957ms step_avg:235.50ms step:952/5535 train_time:224196ms step_avg:235.50ms step:953/5535 train_time:224469ms step_avg:235.54ms step:954/5535 train_time:224689ms step_avg:235.52ms step:955/5535 train_time:224931ms step_avg:235.53ms step:956/5535 train_time:225174ms step_avg:235.54ms step:957/5535 train_time:225415ms step_avg:235.54ms step:958/5535 train_time:225661ms step_avg:235.55ms step:959/5535 train_time:225902ms step_avg:235.56ms step:960/5535 train_time:226146ms step_avg:235.57ms step:961/5535 train_time:226387ms step_avg:235.57ms step:962/5535 train_time:226631ms step_avg:235.58ms step:963/5535 train_time:226876ms step_avg:235.59ms step:964/5535 train_time:227117ms step_avg:235.60ms step:965/5535 train_time:227359ms step_avg:235.60ms step:966/5535 train_time:227602ms step_avg:235.61ms step:967/5535 train_time:227843ms step_avg:235.62ms step:968/5535 train_time:228085ms step_avg:235.62ms step:969/5535 train_time:228329ms step_avg:235.63ms step:970/5535 train_time:228572ms step_avg:235.64ms step:971/5535 train_time:228813ms step_avg:235.65ms step:972/5535 train_time:229055ms step_avg:235.65ms step:973/5535 train_time:229299ms step_avg:235.66ms step:974/5535 train_time:229544ms step_avg:235.67ms step:975/5535 train_time:229789ms step_avg:235.68ms step:976/5535 train_time:230031ms step_avg:235.69ms step:977/5535 train_time:230273ms step_avg:235.69ms step:978/5535 train_time:230515ms step_avg:235.70ms step:979/5535 train_time:230759ms step_avg:235.71ms step:980/5535 train_time:231003ms step_avg:235.72ms step:981/5535 train_time:231272ms step_avg:235.75ms step:982/5535 train_time:231492ms step_avg:235.74ms step:983/5535 train_time:231733ms step_avg:235.74ms step:984/5535 train_time:231976ms step_avg:235.75ms step:985/5535 train_time:232218ms step_avg:235.75ms step:986/5535 train_time:232463ms step_avg:235.76ms step:987/5535 train_time:232705ms step_avg:235.77ms step:988/5535 train_time:232948ms step_avg:235.78ms step:989/5535 train_time:233193ms step_avg:235.79ms step:990/5535 train_time:233435ms step_avg:235.79ms step:991/5535 train_time:233677ms step_avg:235.80ms step:992/5535 train_time:233922ms step_avg:235.81ms step:993/5535 train_time:234168ms step_avg:235.82ms step:994/5535 train_time:234409ms step_avg:235.82ms step:995/5535 train_time:234654ms step_avg:235.83ms step:996/5535 train_time:234894ms step_avg:235.84ms step:997/5535 train_time:235137ms step_avg:235.84ms step:998/5535 train_time:235379ms step_avg:235.85ms step:999/5535 train_time:235623ms step_avg:235.86ms step:1000/5535 train_time:235863ms step_avg:235.86ms step:1000/5535 val_loss:3.347032 train_time:236107ms step_avg:236.11ms step:1001/5535 train_time:236130ms step_avg:235.89ms step:1002/5535 train_time:236351ms step_avg:235.88ms step:1003/5535 train_time:236603ms step_avg:235.90ms step:1004/5535 train_time:236846ms step_avg:235.90ms step:1005/5535 train_time:237087ms step_avg:235.91ms step:1006/5535 train_time:237328ms step_avg:235.91ms step:1007/5535 train_time:237576ms step_avg:235.93ms step:1008/5535 train_time:237820ms step_avg:235.93ms step:1009/5535 train_time:238094ms step_avg:235.97ms step:1010/5535 train_time:238313ms step_avg:235.95ms step:1011/5535 train_time:238556ms step_avg:235.96ms step:1012/5535 train_time:238799ms step_avg:235.97ms step:1013/5535 train_time:239044ms step_avg:235.98ms step:1014/5535 train_time:239284ms step_avg:235.98ms step:1015/5535 train_time:239528ms step_avg:235.99ms step:1016/5535 train_time:239773ms step_avg:236.00ms step:1017/5535 train_time:240018ms step_avg:236.01ms step:1018/5535 train_time:240260ms step_avg:236.01ms step:1019/5535 train_time:240507ms step_avg:236.02ms step:1020/5535 train_time:240749ms step_avg:236.03ms step:1021/5535 train_time:240992ms step_avg:236.04ms step:1022/5535 train_time:241234ms step_avg:236.04ms step:1023/5535 train_time:241478ms step_avg:236.05ms step:1024/5535 train_time:241720ms step_avg:236.06ms step:1025/5535 train_time:241966ms step_avg:236.06ms step:1026/5535 train_time:242208ms step_avg:236.07ms step:1027/5535 train_time:242450ms step_avg:236.08ms step:1028/5535 train_time:242693ms step_avg:236.08ms step:1029/5535 train_time:242941ms step_avg:236.09ms step:1030/5535 train_time:243185ms step_avg:236.10ms step:1031/5535 train_time:243428ms step_avg:236.11ms step:1032/5535 train_time:243669ms step_avg:236.11ms step:1033/5535 train_time:243912ms step_avg:236.12ms step:1034/5535 train_time:244157ms step_avg:236.13ms step:1035/5535 train_time:244400ms step_avg:236.14ms step:1036/5535 train_time:244642ms step_avg:236.14ms step:1037/5535 train_time:244914ms step_avg:236.18ms step:1038/5535 train_time:245134ms step_avg:236.16ms step:1039/5535 train_time:245375ms step_avg:236.16ms step:1040/5535 train_time:245618ms step_avg:236.17ms step:1041/5535 train_time:245860ms step_avg:236.18ms step:1042/5535 train_time:246103ms step_avg:236.18ms step:1043/5535 train_time:246346ms step_avg:236.19ms step:1044/5535 train_time:246589ms step_avg:236.20ms step:1045/5535 train_time:246832ms step_avg:236.20ms step:1046/5535 train_time:247075ms step_avg:236.21ms step:1047/5535 train_time:247319ms step_avg:236.22ms step:1048/5535 train_time:247564ms step_avg:236.22ms step:1049/5535 train_time:247805ms step_avg:236.23ms step:1050/5535 train_time:248048ms step_avg:236.24ms step:1051/5535 train_time:248296ms step_avg:236.25ms step:1052/5535 train_time:248538ms step_avg:236.25ms step:1053/5535 train_time:248779ms step_avg:236.26ms step:1054/5535 train_time:249021ms step_avg:236.26ms step:1055/5535 train_time:249265ms step_avg:236.27ms step:1056/5535 train_time:249509ms step_avg:236.28ms step:1057/5535 train_time:249750ms step_avg:236.28ms step:1058/5535 train_time:249994ms step_avg:236.29ms step:1059/5535 train_time:250239ms step_avg:236.30ms step:1060/5535 train_time:250487ms step_avg:236.31ms step:1061/5535 train_time:250729ms step_avg:236.31ms step:1062/5535 train_time:250971ms step_avg:236.32ms step:1063/5535 train_time:251216ms step_avg:236.33ms step:1064/5535 train_time:251458ms step_avg:236.33ms step:1065/5535 train_time:251729ms step_avg:236.37ms step:1066/5535 train_time:251950ms step_avg:236.35ms step:1067/5535 train_time:252194ms step_avg:236.36ms step:1068/5535 train_time:252439ms step_avg:236.37ms step:1069/5535 train_time:252688ms step_avg:236.38ms step:1070/5535 train_time:252929ms step_avg:236.38ms step:1071/5535 train_time:253175ms step_avg:236.39ms step:1072/5535 train_time:253415ms step_avg:236.40ms step:1073/5535 train_time:253660ms step_avg:236.40ms step:1074/5535 train_time:253902ms step_avg:236.41ms step:1075/5535 train_time:254149ms step_avg:236.42ms step:1076/5535 train_time:254389ms step_avg:236.42ms step:1077/5535 train_time:254634ms step_avg:236.43ms step:1078/5535 train_time:254876ms step_avg:236.43ms step:1079/5535 train_time:255127ms step_avg:236.45ms step:1080/5535 train_time:255368ms step_avg:236.45ms step:1081/5535 train_time:255611ms step_avg:236.46ms step:1082/5535 train_time:255854ms step_avg:236.46ms step:1083/5535 train_time:256099ms step_avg:236.47ms step:1084/5535 train_time:256341ms step_avg:236.48ms step:1085/5535 train_time:256584ms step_avg:236.48ms step:1086/5535 train_time:256827ms step_avg:236.49ms step:1087/5535 train_time:257072ms step_avg:236.50ms step:1088/5535 train_time:257316ms step_avg:236.50ms step:1089/5535 train_time:257560ms step_avg:236.51ms step:1090/5535 train_time:257805ms step_avg:236.52ms step:1091/5535 train_time:258051ms step_avg:236.53ms step:1092/5535 train_time:258292ms step_avg:236.53ms step:1093/5535 train_time:258563ms step_avg:236.56ms step:1094/5535 train_time:258784ms step_avg:236.55ms step:1095/5535 train_time:259026ms step_avg:236.55ms step:1096/5535 train_time:259272ms step_avg:236.56ms step:1097/5535 train_time:259513ms step_avg:236.57ms step:1098/5535 train_time:259759ms step_avg:236.57ms step:1099/5535 train_time:260001ms step_avg:236.58ms step:1100/5535 train_time:260248ms step_avg:236.59ms step:1101/5535 train_time:260489ms step_avg:236.59ms step:1102/5535 train_time:260734ms step_avg:236.60ms step:1103/5535 train_time:260979ms step_avg:236.61ms step:1104/5535 train_time:261222ms step_avg:236.61ms step:1105/5535 train_time:261464ms step_avg:236.62ms step:1106/5535 train_time:261707ms step_avg:236.62ms step:1107/5535 train_time:261951ms step_avg:236.63ms step:1108/5535 train_time:262197ms step_avg:236.64ms step:1109/5535 train_time:262438ms step_avg:236.64ms step:1110/5535 train_time:262681ms step_avg:236.65ms step:1111/5535 train_time:262925ms step_avg:236.66ms step:1112/5535 train_time:263169ms step_avg:236.66ms step:1113/5535 train_time:263412ms step_avg:236.67ms step:1114/5535 train_time:263654ms step_avg:236.67ms step:1115/5535 train_time:263896ms step_avg:236.68ms step:1116/5535 train_time:264139ms step_avg:236.68ms step:1117/5535 train_time:264383ms step_avg:236.69ms step:1118/5535 train_time:264629ms step_avg:236.70ms step:1119/5535 train_time:264874ms step_avg:236.71ms step:1120/5535 train_time:265117ms step_avg:236.71ms step:1121/5535 train_time:265389ms step_avg:236.74ms step:1122/5535 train_time:265610ms step_avg:236.73ms step:1123/5535 train_time:265852ms step_avg:236.73ms step:1124/5535 train_time:266098ms step_avg:236.74ms step:1125/5535 train_time:266338ms step_avg:236.75ms step:1125/5535 val_loss:3.316779 train_time:266583ms step_avg:236.96ms step:1126/5535 train_time:266606ms step_avg:236.77ms step:1127/5535 train_time:266825ms step_avg:236.76ms step:1128/5535 train_time:267077ms step_avg:236.77ms step:1129/5535 train_time:267320ms step_avg:236.78ms step:1130/5535 train_time:267560ms step_avg:236.78ms step:1131/5535 train_time:267803ms step_avg:236.78ms step:1132/5535 train_time:268048ms step_avg:236.79ms step:1133/5535 train_time:268292ms step_avg:236.80ms step:1134/5535 train_time:268536ms step_avg:236.80ms step:1135/5535 train_time:268780ms step_avg:236.81ms step:1136/5535 train_time:269027ms step_avg:236.82ms step:1137/5535 train_time:269270ms step_avg:236.83ms step:1138/5535 train_time:269513ms step_avg:236.83ms step:1139/5535 train_time:269754ms step_avg:236.83ms step:1140/5535 train_time:269998ms step_avg:236.84ms step:1141/5535 train_time:270243ms step_avg:236.85ms step:1142/5535 train_time:270485ms step_avg:236.85ms step:1143/5535 train_time:270729ms step_avg:236.86ms step:1144/5535 train_time:270972ms step_avg:236.86ms step:1145/5535 train_time:271217ms step_avg:236.87ms step:1146/5535 train_time:271459ms step_avg:236.88ms step:1147/5535 train_time:271701ms step_avg:236.88ms step:1148/5535 train_time:271945ms step_avg:236.89ms step:1149/5535 train_time:272217ms step_avg:236.92ms step:1150/5535 train_time:272439ms step_avg:236.90ms step:1151/5535 train_time:272683ms step_avg:236.91ms step:1152/5535 train_time:272926ms step_avg:236.91ms step:1153/5535 train_time:273172ms step_avg:236.92ms step:1154/5535 train_time:273417ms step_avg:236.93ms step:1155/5535 train_time:273658ms step_avg:236.93ms step:1156/5535 train_time:273903ms step_avg:236.94ms step:1157/5535 train_time:274146ms step_avg:236.95ms step:1158/5535 train_time:274389ms step_avg:236.95ms step:1159/5535 train_time:274632ms step_avg:236.96ms step:1160/5535 train_time:274875ms step_avg:236.96ms step:1161/5535 train_time:275117ms step_avg:236.97ms step:1162/5535 train_time:275363ms step_avg:236.97ms step:1163/5535 train_time:275606ms step_avg:236.98ms step:1164/5535 train_time:275849ms step_avg:236.98ms step:1165/5535 train_time:276090ms step_avg:236.99ms step:1166/5535 train_time:276333ms step_avg:236.99ms step:1167/5535 train_time:276578ms step_avg:237.00ms step:1168/5535 train_time:276822ms step_avg:237.00ms step:1169/5535 train_time:277064ms step_avg:237.01ms step:1170/5535 train_time:277306ms step_avg:237.01ms step:1171/5535 train_time:277548ms step_avg:237.02ms step:1172/5535 train_time:277792ms step_avg:237.02ms step:1173/5535 train_time:278035ms step_avg:237.03ms step:1174/5535 train_time:278285ms step_avg:237.04ms step:1175/5535 train_time:278528ms step_avg:237.04ms step:1176/5535 train_time:278771ms step_avg:237.05ms step:1177/5535 train_time:279046ms step_avg:237.08ms step:1178/5535 train_time:279269ms step_avg:237.07ms step:1179/5535 train_time:279510ms step_avg:237.07ms step:1180/5535 train_time:279756ms step_avg:237.08ms step:1181/5535 train_time:280002ms step_avg:237.09ms step:1182/5535 train_time:280245ms step_avg:237.09ms step:1183/5535 train_time:280487ms step_avg:237.10ms step:1184/5535 train_time:280729ms step_avg:237.10ms step:1185/5535 train_time:280974ms step_avg:237.11ms step:1186/5535 train_time:281218ms step_avg:237.11ms step:1187/5535 train_time:281466ms step_avg:237.12ms step:1188/5535 train_time:281709ms step_avg:237.13ms step:1189/5535 train_time:281952ms step_avg:237.13ms step:1190/5535 train_time:282197ms step_avg:237.14ms step:1191/5535 train_time:282441ms step_avg:237.15ms step:1192/5535 train_time:282682ms step_avg:237.15ms step:1193/5535 train_time:282926ms step_avg:237.16ms step:1194/5535 train_time:283170ms step_avg:237.16ms step:1195/5535 train_time:283418ms step_avg:237.17ms step:1196/5535 train_time:283665ms step_avg:237.18ms step:1197/5535 train_time:283910ms step_avg:237.18ms step:1198/5535 train_time:284160ms step_avg:237.20ms step:1199/5535 train_time:284408ms step_avg:237.20ms step:1200/5535 train_time:284655ms step_avg:237.21ms step:1201/5535 train_time:284900ms step_avg:237.22ms step:1202/5535 train_time:285147ms step_avg:237.23ms step:1203/5535 train_time:285397ms step_avg:237.24ms step:1204/5535 train_time:285641ms step_avg:237.24ms step:1205/5535 train_time:285915ms step_avg:237.27ms step:1206/5535 train_time:286141ms step_avg:237.26ms step:1207/5535 train_time:286385ms step_avg:237.27ms step:1208/5535 train_time:286628ms step_avg:237.28ms step:1209/5535 train_time:286873ms step_avg:237.28ms step:1210/5535 train_time:287123ms step_avg:237.29ms step:1211/5535 train_time:287370ms step_avg:237.30ms step:1212/5535 train_time:287615ms step_avg:237.31ms step:1213/5535 train_time:287861ms step_avg:237.31ms step:1214/5535 train_time:288109ms step_avg:237.32ms step:1215/5535 train_time:288359ms step_avg:237.33ms step:1216/5535 train_time:288602ms step_avg:237.34ms step:1217/5535 train_time:288855ms step_avg:237.35ms step:1218/5535 train_time:289101ms step_avg:237.36ms step:1219/5535 train_time:289344ms step_avg:237.36ms step:1220/5535 train_time:289588ms step_avg:237.37ms step:1221/5535 train_time:289837ms step_avg:237.38ms step:1222/5535 train_time:290082ms step_avg:237.38ms step:1223/5535 train_time:290326ms step_avg:237.39ms step:1224/5535 train_time:290569ms step_avg:237.39ms step:1225/5535 train_time:290819ms step_avg:237.40ms step:1226/5535 train_time:291063ms step_avg:237.41ms step:1227/5535 train_time:291308ms step_avg:237.41ms step:1228/5535 train_time:291554ms step_avg:237.42ms step:1229/5535 train_time:291799ms step_avg:237.43ms step:1230/5535 train_time:292046ms step_avg:237.44ms step:1231/5535 train_time:292294ms step_avg:237.44ms step:1232/5535 train_time:292542ms step_avg:237.45ms step:1233/5535 train_time:292816ms step_avg:237.48ms step:1234/5535 train_time:293038ms step_avg:237.47ms step:1235/5535 train_time:293283ms step_avg:237.48ms step:1236/5535 train_time:293529ms step_avg:237.48ms step:1237/5535 train_time:293775ms step_avg:237.49ms step:1238/5535 train_time:294027ms step_avg:237.50ms step:1239/5535 train_time:294270ms step_avg:237.51ms step:1240/5535 train_time:294514ms step_avg:237.51ms step:1241/5535 train_time:294763ms step_avg:237.52ms step:1242/5535 train_time:295008ms step_avg:237.53ms step:1243/5535 train_time:295252ms step_avg:237.53ms step:1244/5535 train_time:295499ms step_avg:237.54ms step:1245/5535 train_time:295744ms step_avg:237.55ms step:1246/5535 train_time:295990ms step_avg:237.55ms step:1247/5535 train_time:296237ms step_avg:237.56ms step:1248/5535 train_time:296484ms step_avg:237.57ms step:1249/5535 train_time:296726ms step_avg:237.57ms step:1250/5535 train_time:296968ms step_avg:237.57ms step:1250/5535 val_loss:3.291978 train_time:297215ms step_avg:237.77ms step:1251/5535 train_time:297236ms step_avg:237.60ms step:1252/5535 train_time:297462ms step_avg:237.59ms step:1253/5535 train_time:297717ms step_avg:237.60ms step:1254/5535 train_time:297960ms step_avg:237.61ms step:1255/5535 train_time:298210ms step_avg:237.62ms step:1256/5535 train_time:298451ms step_avg:237.62ms step:1257/5535 train_time:298702ms step_avg:237.63ms step:1258/5535 train_time:298946ms step_avg:237.64ms step:1259/5535 train_time:299191ms step_avg:237.64ms step:1260/5535 train_time:299434ms step_avg:237.65ms step:1261/5535 train_time:299709ms step_avg:237.68ms step:1262/5535 train_time:299935ms step_avg:237.67ms step:1263/5535 train_time:300181ms step_avg:237.67ms step:1264/5535 train_time:300423ms step_avg:237.68ms step:1265/5535 train_time:300669ms step_avg:237.68ms step:1266/5535 train_time:300917ms step_avg:237.69ms step:1267/5535 train_time:301162ms step_avg:237.70ms step:1268/5535 train_time:301407ms step_avg:237.70ms step:1269/5535 train_time:301653ms step_avg:237.71ms step:1270/5535 train_time:301900ms step_avg:237.72ms step:1271/5535 train_time:302146ms step_avg:237.72ms step:1272/5535 train_time:302390ms step_avg:237.73ms step:1273/5535 train_time:302636ms step_avg:237.73ms step:1274/5535 train_time:302882ms step_avg:237.74ms step:1275/5535 train_time:303127ms step_avg:237.75ms step:1276/5535 train_time:303371ms step_avg:237.75ms step:1277/5535 train_time:303615ms step_avg:237.76ms step:1278/5535 train_time:303861ms step_avg:237.76ms step:1279/5535 train_time:304108ms step_avg:237.77ms step:1280/5535 train_time:304357ms step_avg:237.78ms step:1281/5535 train_time:304602ms step_avg:237.78ms step:1282/5535 train_time:304848ms step_avg:237.79ms step:1283/5535 train_time:305093ms step_avg:237.80ms step:1284/5535 train_time:305340ms step_avg:237.80ms step:1285/5535 train_time:305583ms step_avg:237.81ms step:1286/5535 train_time:305829ms step_avg:237.81ms step:1287/5535 train_time:306078ms step_avg:237.82ms step:1288/5535 train_time:306324ms step_avg:237.83ms step:1289/5535 train_time:306602ms step_avg:237.86ms step:1290/5535 train_time:306826ms step_avg:237.85ms step:1291/5535 train_time:307074ms step_avg:237.86ms step:1292/5535 train_time:307317ms step_avg:237.86ms step:1293/5535 train_time:307567ms step_avg:237.87ms step:1294/5535 train_time:307811ms step_avg:237.88ms step:1295/5535 train_time:308059ms step_avg:237.88ms step:1296/5535 train_time:308303ms step_avg:237.89ms step:1297/5535 train_time:308551ms step_avg:237.90ms step:1298/5535 train_time:308796ms step_avg:237.90ms step:1299/5535 train_time:309041ms step_avg:237.91ms step:1300/5535 train_time:309286ms step_avg:237.91ms step:1301/5535 train_time:309531ms step_avg:237.92ms step:1302/5535 train_time:309775ms step_avg:237.92ms step:1303/5535 train_time:310023ms step_avg:237.93ms step:1304/5535 train_time:310272ms step_avg:237.94ms step:1305/5535 train_time:310518ms step_avg:237.94ms step:1306/5535 train_time:310764ms step_avg:237.95ms step:1307/5535 train_time:311010ms step_avg:237.96ms step:1308/5535 train_time:311256ms step_avg:237.96ms step:1309/5535 train_time:311501ms step_avg:237.97ms step:1310/5535 train_time:311746ms step_avg:237.97ms step:1311/5535 train_time:311991ms step_avg:237.98ms step:1312/5535 train_time:312235ms step_avg:237.98ms step:1313/5535 train_time:312480ms step_avg:237.99ms step:1314/5535 train_time:312724ms step_avg:237.99ms step:1315/5535 train_time:312970ms step_avg:238.00ms step:1316/5535 train_time:313215ms step_avg:238.01ms step:1317/5535 train_time:313489ms step_avg:238.03ms step:1318/5535 train_time:313711ms step_avg:238.02ms step:1319/5535 train_time:313960ms step_avg:238.03ms step:1320/5535 train_time:314205ms step_avg:238.03ms step:1321/5535 train_time:314452ms step_avg:238.04ms step:1322/5535 train_time:314700ms step_avg:238.05ms step:1323/5535 train_time:314947ms step_avg:238.06ms step:1324/5535 train_time:315194ms step_avg:238.06ms step:1325/5535 train_time:315439ms step_avg:238.07ms step:1326/5535 train_time:315685ms step_avg:238.07ms step:1327/5535 train_time:315931ms step_avg:238.08ms step:1328/5535 train_time:316177ms step_avg:238.08ms step:1329/5535 train_time:316430ms step_avg:238.10ms step:1330/5535 train_time:316676ms step_avg:238.10ms step:1331/5535 train_time:316924ms step_avg:238.11ms step:1332/5535 train_time:317175ms step_avg:238.12ms step:1333/5535 train_time:317422ms step_avg:238.13ms step:1334/5535 train_time:317665ms step_avg:238.13ms step:1335/5535 train_time:317909ms step_avg:238.13ms step:1336/5535 train_time:318158ms step_avg:238.14ms step:1337/5535 train_time:318405ms step_avg:238.15ms step:1338/5535 train_time:318647ms step_avg:238.15ms step:1339/5535 train_time:318893ms step_avg:238.16ms step:1340/5535 train_time:319141ms step_avg:238.16ms step:1341/5535 train_time:319385ms step_avg:238.17ms step:1342/5535 train_time:319628ms step_avg:238.17ms step:1343/5535 train_time:319872ms step_avg:238.18ms step:1344/5535 train_time:320117ms step_avg:238.18ms step:1345/5535 train_time:320393ms step_avg:238.21ms step:1346/5535 train_time:320613ms step_avg:238.20ms step:1347/5535 train_time:320860ms step_avg:238.20ms step:1348/5535 train_time:321104ms step_avg:238.21ms step:1349/5535 train_time:321351ms step_avg:238.21ms step:1350/5535 train_time:321595ms step_avg:238.22ms step:1351/5535 train_time:321840ms step_avg:238.22ms step:1352/5535 train_time:322084ms step_avg:238.23ms step:1353/5535 train_time:322332ms step_avg:238.24ms step:1354/5535 train_time:322578ms step_avg:238.24ms step:1355/5535 train_time:322821ms step_avg:238.24ms step:1356/5535 train_time:323065ms step_avg:238.25ms step:1357/5535 train_time:323310ms step_avg:238.25ms step:1358/5535 train_time:323556ms step_avg:238.26ms step:1359/5535 train_time:323803ms step_avg:238.27ms step:1360/5535 train_time:324048ms step_avg:238.27ms step:1361/5535 train_time:324295ms step_avg:238.28ms step:1362/5535 train_time:324542ms step_avg:238.28ms step:1363/5535 train_time:324789ms step_avg:238.29ms step:1364/5535 train_time:325033ms step_avg:238.29ms step:1365/5535 train_time:325276ms step_avg:238.30ms step:1366/5535 train_time:325522ms step_avg:238.30ms step:1367/5535 train_time:325767ms step_avg:238.31ms step:1368/5535 train_time:326013ms step_avg:238.31ms step:1369/5535 train_time:326261ms step_avg:238.32ms step:1370/5535 train_time:326510ms step_avg:238.33ms step:1371/5535 train_time:326752ms step_avg:238.33ms step:1372/5535 train_time:327002ms step_avg:238.34ms step:1373/5535 train_time:327272ms step_avg:238.36ms step:1374/5535 train_time:327496ms step_avg:238.35ms step:1375/5535 train_time:327740ms step_avg:238.36ms step:1375/5535 val_loss:3.272248 train_time:327986ms step_avg:238.53ms step:1376/5535 train_time:328009ms step_avg:238.38ms step:1377/5535 train_time:328231ms step_avg:238.37ms step:1378/5535 train_time:328482ms step_avg:238.38ms step:1379/5535 train_time:328728ms step_avg:238.38ms step:1380/5535 train_time:328973ms step_avg:238.39ms step:1381/5535 train_time:329219ms step_avg:238.39ms step:1382/5535 train_time:329470ms step_avg:238.40ms step:1383/5535 train_time:329713ms step_avg:238.40ms step:1384/5535 train_time:329960ms step_avg:238.41ms step:1385/5535 train_time:330203ms step_avg:238.41ms step:1386/5535 train_time:330449ms step_avg:238.42ms step:1387/5535 train_time:330695ms step_avg:238.42ms step:1388/5535 train_time:330940ms step_avg:238.43ms step:1389/5535 train_time:331185ms step_avg:238.43ms step:1390/5535 train_time:331429ms step_avg:238.44ms step:1391/5535 train_time:331677ms step_avg:238.44ms step:1392/5535 train_time:331926ms step_avg:238.45ms step:1393/5535 train_time:332168ms step_avg:238.46ms step:1394/5535 train_time:332412ms step_avg:238.46ms step:1395/5535 train_time:332656ms step_avg:238.46ms step:1396/5535 train_time:332902ms step_avg:238.47ms step:1397/5535 train_time:333146ms step_avg:238.47ms step:1398/5535 train_time:333389ms step_avg:238.48ms step:1399/5535 train_time:333634ms step_avg:238.48ms step:1400/5535 train_time:333881ms step_avg:238.49ms step:1401/5535 train_time:334154ms step_avg:238.51ms step:1402/5535 train_time:334376ms step_avg:238.50ms step:1403/5535 train_time:334621ms step_avg:238.50ms step:1404/5535 train_time:334866ms step_avg:238.51ms step:1405/5535 train_time:335112ms step_avg:238.51ms step:1406/5535 train_time:335361ms step_avg:238.52ms step:1407/5535 train_time:335605ms step_avg:238.53ms step:1408/5535 train_time:335849ms step_avg:238.53ms step:1409/5535 train_time:336100ms step_avg:238.54ms step:1410/5535 train_time:336346ms step_avg:238.54ms step:1411/5535 train_time:336592ms step_avg:238.55ms step:1412/5535 train_time:336835ms step_avg:238.55ms step:1413/5535 train_time:337082ms step_avg:238.56ms step:1414/5535 train_time:337327ms step_avg:238.56ms step:1415/5535 train_time:337570ms step_avg:238.57ms step:1416/5535 train_time:337823ms step_avg:238.58ms step:1417/5535 train_time:338067ms step_avg:238.58ms step:1418/5535 train_time:338312ms step_avg:238.58ms step:1419/5535 train_time:338556ms step_avg:238.59ms step:1420/5535 train_time:338804ms step_avg:238.59ms step:1421/5535 train_time:339050ms step_avg:238.60ms step:1422/5535 train_time:339294ms step_avg:238.60ms step:1423/5535 train_time:339536ms step_avg:238.61ms step:1424/5535 train_time:339782ms step_avg:238.61ms step:1425/5535 train_time:340032ms step_avg:238.62ms step:1426/5535 train_time:340280ms step_avg:238.63ms step:1427/5535 train_time:340523ms step_avg:238.63ms step:1428/5535 train_time:340767ms step_avg:238.63ms step:1429/5535 train_time:341040ms step_avg:238.66ms step:1430/5535 train_time:341263ms step_avg:238.65ms step:1431/5535 train_time:341510ms step_avg:238.65ms step:1432/5535 train_time:341753ms step_avg:238.65ms step:1433/5535 train_time:342001ms step_avg:238.66ms step:1434/5535 train_time:342249ms step_avg:238.67ms step:1435/5535 train_time:342496ms step_avg:238.67ms step:1436/5535 train_time:342739ms step_avg:238.68ms step:1437/5535 train_time:342985ms step_avg:238.68ms step:1438/5535 train_time:343228ms step_avg:238.68ms step:1439/5535 train_time:343473ms step_avg:238.69ms step:1440/5535 train_time:343719ms step_avg:238.69ms step:1441/5535 train_time:343966ms step_avg:238.70ms step:1442/5535 train_time:344211ms step_avg:238.70ms step:1443/5535 train_time:344465ms step_avg:238.71ms step:1444/5535 train_time:344711ms step_avg:238.72ms step:1445/5535 train_time:344954ms step_avg:238.72ms step:1446/5535 train_time:345200ms step_avg:238.73ms step:1447/5535 train_time:345449ms step_avg:238.73ms step:1448/5535 train_time:345693ms step_avg:238.74ms step:1449/5535 train_time:345939ms step_avg:238.74ms step:1450/5535 train_time:346183ms step_avg:238.75ms step:1451/5535 train_time:346428ms step_avg:238.75ms step:1452/5535 train_time:346672ms step_avg:238.76ms step:1453/5535 train_time:346919ms step_avg:238.76ms step:1454/5535 train_time:347162ms step_avg:238.76ms step:1455/5535 train_time:347410ms step_avg:238.77ms step:1456/5535 train_time:347658ms step_avg:238.78ms step:1457/5535 train_time:347932ms step_avg:238.80ms step:1458/5535 train_time:348157ms step_avg:238.79ms step:1459/5535 train_time:348402ms step_avg:238.80ms step:1460/5535 train_time:348651ms step_avg:238.80ms step:1461/5535 train_time:348896ms step_avg:238.81ms step:1462/5535 train_time:349141ms step_avg:238.81ms step:1463/5535 train_time:349385ms step_avg:238.81ms step:1464/5535 train_time:349633ms step_avg:238.82ms step:1465/5535 train_time:349879ms step_avg:238.83ms step:1466/5535 train_time:350124ms step_avg:238.83ms step:1467/5535 train_time:350370ms step_avg:238.83ms step:1468/5535 train_time:350616ms step_avg:238.84ms step:1469/5535 train_time:350863ms step_avg:238.84ms step:1470/5535 train_time:351110ms step_avg:238.85ms step:1471/5535 train_time:351359ms step_avg:238.86ms step:1472/5535 train_time:351605ms step_avg:238.86ms step:1473/5535 train_time:351850ms step_avg:238.87ms step:1474/5535 train_time:352095ms step_avg:238.87ms step:1475/5535 train_time:352343ms step_avg:238.88ms step:1476/5535 train_time:352587ms step_avg:238.88ms step:1477/5535 train_time:352838ms step_avg:238.89ms step:1478/5535 train_time:353089ms step_avg:238.90ms step:1479/5535 train_time:353332ms step_avg:238.90ms step:1480/5535 train_time:353579ms step_avg:238.90ms step:1481/5535 train_time:353826ms step_avg:238.91ms step:1482/5535 train_time:354074ms step_avg:238.92ms step:1483/5535 train_time:354330ms step_avg:238.93ms step:1484/5535 train_time:354571ms step_avg:238.93ms step:1485/5535 train_time:354842ms step_avg:238.95ms step:1486/5535 train_time:355066ms step_avg:238.94ms step:1487/5535 train_time:355315ms step_avg:238.95ms step:1488/5535 train_time:355560ms step_avg:238.95ms step:1489/5535 train_time:355806ms step_avg:238.96ms step:1490/5535 train_time:356051ms step_avg:238.96ms step:1491/5535 train_time:356299ms step_avg:238.97ms step:1492/5535 train_time:356544ms step_avg:238.97ms step:1493/5535 train_time:356788ms step_avg:238.97ms step:1494/5535 train_time:357035ms step_avg:238.98ms step:1495/5535 train_time:357283ms step_avg:238.99ms step:1496/5535 train_time:357526ms step_avg:238.99ms step:1497/5535 train_time:357769ms step_avg:238.99ms step:1498/5535 train_time:358012ms step_avg:238.99ms step:1499/5535 train_time:358257ms step_avg:239.00ms step:1500/5535 train_time:358508ms step_avg:239.01ms step:1500/5535 val_loss:3.252971 train_time:358757ms step_avg:239.17ms step:1501/5535 train_time:358778ms step_avg:239.03ms step:1502/5535 train_time:359013ms step_avg:239.02ms step:1503/5535 train_time:359269ms step_avg:239.03ms step:1504/5535 train_time:359514ms step_avg:239.04ms step:1505/5535 train_time:359757ms step_avg:239.04ms step:1506/5535 train_time:360001ms step_avg:239.04ms step:1507/5535 train_time:360254ms step_avg:239.05ms step:1508/5535 train_time:360502ms step_avg:239.06ms step:1509/5535 train_time:360745ms step_avg:239.06ms step:1510/5535 train_time:360987ms step_avg:239.06ms step:1511/5535 train_time:361235ms step_avg:239.07ms step:1512/5535 train_time:361484ms step_avg:239.08ms step:1513/5535 train_time:361758ms step_avg:239.10ms step:1514/5535 train_time:361979ms step_avg:239.09ms step:1515/5535 train_time:362227ms step_avg:239.09ms step:1516/5535 train_time:362477ms step_avg:239.10ms step:1517/5535 train_time:362724ms step_avg:239.11ms step:1518/5535 train_time:362969ms step_avg:239.11ms step:1519/5535 train_time:363213ms step_avg:239.11ms step:1520/5535 train_time:363460ms step_avg:239.12ms step:1521/5535 train_time:363707ms step_avg:239.12ms step:1522/5535 train_time:363950ms step_avg:239.13ms step:1523/5535 train_time:364197ms step_avg:239.13ms step:1524/5535 train_time:364441ms step_avg:239.13ms step:1525/5535 train_time:364690ms step_avg:239.14ms step:1526/5535 train_time:364937ms step_avg:239.15ms step:1527/5535 train_time:365182ms step_avg:239.15ms step:1528/5535 train_time:365430ms step_avg:239.16ms step:1529/5535 train_time:365675ms step_avg:239.16ms step:1530/5535 train_time:365920ms step_avg:239.16ms step:1531/5535 train_time:366168ms step_avg:239.17ms step:1532/5535 train_time:366414ms step_avg:239.17ms step:1533/5535 train_time:366662ms step_avg:239.18ms step:1534/5535 train_time:366914ms step_avg:239.19ms step:1535/5535 train_time:367157ms step_avg:239.19ms step:1536/5535 train_time:367404ms step_avg:239.20ms step:1537/5535 train_time:367648ms step_avg:239.20ms step:1538/5535 train_time:367896ms step_avg:239.20ms step:1539/5535 train_time:368144ms step_avg:239.21ms step:1540/5535 train_time:368388ms step_avg:239.21ms step:1541/5535 train_time:368662ms step_avg:239.24ms step:1542/5535 train_time:368890ms step_avg:239.23ms step:1543/5535 train_time:369135ms step_avg:239.23ms step:1544/5535 train_time:369379ms step_avg:239.24ms step:1545/5535 train_time:369624ms step_avg:239.24ms step:1546/5535 train_time:369876ms step_avg:239.25ms step:1547/5535 train_time:370121ms step_avg:239.25ms step:1548/5535 train_time:370371ms step_avg:239.26ms step:1549/5535 train_time:370616ms step_avg:239.26ms step:1550/5535 train_time:370861ms step_avg:239.26ms step:1551/5535 train_time:371109ms step_avg:239.27ms step:1552/5535 train_time:371353ms step_avg:239.27ms step:1553/5535 train_time:371597ms step_avg:239.28ms step:1554/5535 train_time:371845ms step_avg:239.28ms step:1555/5535 train_time:372091ms step_avg:239.29ms step:1556/5535 train_time:372336ms step_avg:239.29ms step:1557/5535 train_time:372584ms step_avg:239.30ms step:1558/5535 train_time:372828ms step_avg:239.30ms step:1559/5535 train_time:373076ms step_avg:239.30ms step:1560/5535 train_time:373322ms step_avg:239.31ms step:1561/5535 train_time:373568ms step_avg:239.31ms step:1562/5535 train_time:373812ms step_avg:239.32ms step:1563/5535 train_time:374055ms step_avg:239.32ms step:1564/5535 train_time:374301ms step_avg:239.32ms step:1565/5535 train_time:374546ms step_avg:239.33ms step:1566/5535 train_time:374793ms step_avg:239.33ms step:1567/5535 train_time:375039ms step_avg:239.34ms step:1568/5535 train_time:375282ms step_avg:239.34ms step:1569/5535 train_time:375556ms step_avg:239.36ms step:1570/5535 train_time:375781ms step_avg:239.35ms step:1571/5535 train_time:376026ms step_avg:239.35ms step:1572/5535 train_time:376272ms step_avg:239.36ms step:1573/5535 train_time:376517ms step_avg:239.36ms step:1574/5535 train_time:376764ms step_avg:239.37ms step:1575/5535 train_time:377009ms step_avg:239.37ms step:1576/5535 train_time:377254ms step_avg:239.37ms step:1577/5535 train_time:377506ms step_avg:239.38ms step:1578/5535 train_time:377751ms step_avg:239.39ms step:1579/5535 train_time:377997ms step_avg:239.39ms step:1580/5535 train_time:378244ms step_avg:239.39ms step:1581/5535 train_time:378488ms step_avg:239.40ms step:1582/5535 train_time:378740ms step_avg:239.41ms step:1583/5535 train_time:378985ms step_avg:239.41ms step:1584/5535 train_time:379232ms step_avg:239.41ms step:1585/5535 train_time:379479ms step_avg:239.42ms step:1586/5535 train_time:379726ms step_avg:239.42ms step:1587/5535 train_time:379970ms step_avg:239.43ms step:1588/5535 train_time:380213ms step_avg:239.43ms step:1589/5535 train_time:380462ms step_avg:239.43ms step:1590/5535 train_time:380708ms step_avg:239.44ms step:1591/5535 train_time:380956ms step_avg:239.44ms step:1592/5535 train_time:381202ms step_avg:239.45ms step:1593/5535 train_time:381453ms step_avg:239.46ms step:1594/5535 train_time:381699ms step_avg:239.46ms step:1595/5535 train_time:381943ms step_avg:239.46ms step:1596/5535 train_time:382189ms step_avg:239.47ms step:1597/5535 train_time:382468ms step_avg:239.49ms step:1598/5535 train_time:382689ms step_avg:239.48ms step:1599/5535 train_time:382935ms step_avg:239.48ms step:1600/5535 train_time:383179ms step_avg:239.49ms step:1601/5535 train_time:383425ms step_avg:239.49ms step:1602/5535 train_time:383672ms step_avg:239.50ms step:1603/5535 train_time:383923ms step_avg:239.50ms step:1604/5535 train_time:384166ms step_avg:239.50ms step:1605/5535 train_time:384415ms step_avg:239.51ms step:1606/5535 train_time:384663ms step_avg:239.52ms step:1607/5535 train_time:384912ms step_avg:239.52ms step:1608/5535 train_time:385153ms step_avg:239.52ms step:1609/5535 train_time:385400ms step_avg:239.53ms step:1610/5535 train_time:385645ms step_avg:239.53ms step:1611/5535 train_time:385899ms step_avg:239.54ms step:1612/5535 train_time:386144ms step_avg:239.54ms step:1613/5535 train_time:386395ms step_avg:239.55ms step:1614/5535 train_time:386638ms step_avg:239.55ms step:1615/5535 train_time:386887ms step_avg:239.56ms step:1616/5535 train_time:387139ms step_avg:239.57ms step:1617/5535 train_time:387383ms step_avg:239.57ms step:1618/5535 train_time:387626ms step_avg:239.57ms step:1619/5535 train_time:387871ms step_avg:239.57ms step:1620/5535 train_time:388120ms step_avg:239.58ms step:1621/5535 train_time:388368ms step_avg:239.59ms step:1622/5535 train_time:388613ms step_avg:239.59ms step:1623/5535 train_time:388859ms step_avg:239.59ms step:1624/5535 train_time:389108ms step_avg:239.60ms step:1625/5535 train_time:389387ms step_avg:239.62ms step:1625/5535 val_loss:3.236582 train_time:389610ms step_avg:239.76ms step:1626/5535 train_time:389631ms step_avg:239.63ms step:1627/5535 train_time:389855ms step_avg:239.62ms step:1628/5535 train_time:390107ms step_avg:239.62ms step:1629/5535 train_time:390350ms step_avg:239.63ms step:1630/5535 train_time:390592ms step_avg:239.63ms step:1631/5535 train_time:390835ms step_avg:239.63ms step:1632/5535 train_time:391088ms step_avg:239.64ms step:1633/5535 train_time:391331ms step_avg:239.64ms step:1634/5535 train_time:391575ms step_avg:239.64ms step:1635/5535 train_time:391822ms step_avg:239.65ms step:1636/5535 train_time:392073ms step_avg:239.65ms step:1637/5535 train_time:392319ms step_avg:239.66ms step:1638/5535 train_time:392563ms step_avg:239.66ms step:1639/5535 train_time:392814ms step_avg:239.67ms step:1640/5535 train_time:393066ms step_avg:239.67ms step:1641/5535 train_time:393311ms step_avg:239.68ms step:1642/5535 train_time:393558ms step_avg:239.68ms step:1643/5535 train_time:393804ms step_avg:239.69ms step:1644/5535 train_time:394053ms step_avg:239.69ms step:1645/5535 train_time:394297ms step_avg:239.69ms step:1646/5535 train_time:394544ms step_avg:239.70ms step:1647/5535 train_time:394789ms step_avg:239.70ms step:1648/5535 train_time:395037ms step_avg:239.71ms step:1649/5535 train_time:395282ms step_avg:239.71ms step:1650/5535 train_time:395527ms step_avg:239.71ms step:1651/5535 train_time:395775ms step_avg:239.72ms step:1652/5535 train_time:396021ms step_avg:239.72ms step:1653/5535 train_time:396298ms step_avg:239.74ms step:1654/5535 train_time:396522ms step_avg:239.74ms step:1655/5535 train_time:396767ms step_avg:239.74ms step:1656/5535 train_time:397015ms step_avg:239.74ms step:1657/5535 train_time:397264ms step_avg:239.75ms step:1658/5535 train_time:397508ms step_avg:239.75ms step:1659/5535 train_time:397752ms step_avg:239.75ms step:1660/5535 train_time:397999ms step_avg:239.76ms step:1661/5535 train_time:398245ms step_avg:239.76ms step:1662/5535 train_time:398491ms step_avg:239.77ms step:1663/5535 train_time:398735ms step_avg:239.77ms step:1664/5535 train_time:398983ms step_avg:239.77ms step:1665/5535 train_time:399228ms step_avg:239.78ms step:1666/5535 train_time:399475ms step_avg:239.78ms step:1667/5535 train_time:399721ms step_avg:239.78ms step:1668/5535 train_time:399971ms step_avg:239.79ms step:1669/5535 train_time:400220ms step_avg:239.80ms step:1670/5535 train_time:400465ms step_avg:239.80ms step:1671/5535 train_time:400711ms step_avg:239.80ms step:1672/5535 train_time:400961ms step_avg:239.81ms step:1673/5535 train_time:401205ms step_avg:239.81ms step:1674/5535 train_time:401449ms step_avg:239.81ms step:1675/5535 train_time:401694ms step_avg:239.82ms step:1676/5535 train_time:401938ms step_avg:239.82ms step:1677/5535 train_time:402184ms step_avg:239.82ms step:1678/5535 train_time:402431ms step_avg:239.83ms step:1679/5535 train_time:402675ms step_avg:239.83ms step:1680/5535 train_time:402920ms step_avg:239.83ms step:1681/5535 train_time:403196ms step_avg:239.85ms step:1682/5535 train_time:403419ms step_avg:239.84ms step:1683/5535 train_time:403663ms step_avg:239.85ms step:1684/5535 train_time:403911ms step_avg:239.85ms step:1685/5535 train_time:404158ms step_avg:239.86ms step:1686/5535 train_time:404403ms step_avg:239.86ms step:1687/5535 train_time:404649ms step_avg:239.86ms step:1688/5535 train_time:404895ms step_avg:239.87ms step:1689/5535 train_time:405149ms step_avg:239.88ms step:1690/5535 train_time:405393ms step_avg:239.88ms step:1691/5535 train_time:405637ms step_avg:239.88ms step:1692/5535 train_time:405881ms step_avg:239.88ms step:1693/5535 train_time:406127ms step_avg:239.89ms step:1694/5535 train_time:406374ms step_avg:239.89ms step:1695/5535 train_time:406619ms step_avg:239.89ms step:1696/5535 train_time:406865ms step_avg:239.90ms step:1697/5535 train_time:407110ms step_avg:239.90ms step:1698/5535 train_time:407355ms step_avg:239.90ms step:1699/5535 train_time:407601ms step_avg:239.91ms step:1700/5535 train_time:407846ms step_avg:239.91ms step:1701/5535 train_time:408090ms step_avg:239.91ms step:1702/5535 train_time:408336ms step_avg:239.92ms step:1703/5535 train_time:408582ms step_avg:239.92ms step:1704/5535 train_time:408825ms step_avg:239.92ms step:1705/5535 train_time:409072ms step_avg:239.92ms step:1706/5535 train_time:409316ms step_avg:239.93ms step:1707/5535 train_time:409562ms step_avg:239.93ms step:1708/5535 train_time:409808ms step_avg:239.93ms step:1709/5535 train_time:410081ms step_avg:239.95ms step:1710/5535 train_time:410307ms step_avg:239.95ms step:1711/5535 train_time:410553ms step_avg:239.95ms step:1712/5535 train_time:410796ms step_avg:239.95ms step:1713/5535 train_time:411042ms step_avg:239.95ms step:1714/5535 train_time:411290ms step_avg:239.96ms step:1715/5535 train_time:411544ms step_avg:239.97ms step:1716/5535 train_time:411790ms step_avg:239.97ms step:1717/5535 train_time:412033ms step_avg:239.97ms step:1718/5535 train_time:412277ms step_avg:239.98ms step:1719/5535 train_time:412525ms step_avg:239.98ms step:1720/5535 train_time:412772ms step_avg:239.98ms step:1721/5535 train_time:413022ms step_avg:239.99ms step:1722/5535 train_time:413266ms step_avg:239.99ms step:1723/5535 train_time:413516ms step_avg:240.00ms step:1724/5535 train_time:413761ms step_avg:240.00ms step:1725/5535 train_time:414012ms step_avg:240.01ms step:1726/5535 train_time:414258ms step_avg:240.01ms step:1727/5535 train_time:414503ms step_avg:240.01ms step:1728/5535 train_time:414749ms step_avg:240.02ms step:1729/5535 train_time:414995ms step_avg:240.02ms step:1730/5535 train_time:415240ms step_avg:240.02ms step:1731/5535 train_time:415489ms step_avg:240.03ms step:1732/5535 train_time:415733ms step_avg:240.03ms step:1733/5535 train_time:415976ms step_avg:240.03ms step:1734/5535 train_time:416222ms step_avg:240.04ms step:1735/5535 train_time:416471ms step_avg:240.04ms step:1736/5535 train_time:416720ms step_avg:240.05ms step:1737/5535 train_time:416994ms step_avg:240.07ms step:1738/5535 train_time:417221ms step_avg:240.06ms step:1739/5535 train_time:417470ms step_avg:240.06ms step:1740/5535 train_time:417718ms step_avg:240.07ms step:1741/5535 train_time:417964ms step_avg:240.07ms step:1742/5535 train_time:418207ms step_avg:240.07ms step:1743/5535 train_time:418458ms step_avg:240.08ms step:1744/5535 train_time:418705ms step_avg:240.08ms step:1745/5535 train_time:418949ms step_avg:240.09ms step:1746/5535 train_time:419191ms step_avg:240.09ms step:1747/5535 train_time:419435ms step_avg:240.09ms step:1748/5535 train_time:419690ms step_avg:240.10ms step:1749/5535 train_time:419934ms step_avg:240.10ms step:1750/5535 train_time:420176ms step_avg:240.10ms step:1750/5535 val_loss:3.220371 train_time:420426ms step_avg:240.24ms step:1751/5535 train_time:420447ms step_avg:240.12ms step:1752/5535 train_time:420671ms step_avg:240.11ms step:1753/5535 train_time:420925ms step_avg:240.12ms step:1754/5535 train_time:421169ms step_avg:240.12ms step:1755/5535 train_time:421418ms step_avg:240.12ms step:1756/5535 train_time:421663ms step_avg:240.13ms step:1757/5535 train_time:421913ms step_avg:240.13ms step:1758/5535 train_time:422157ms step_avg:240.13ms step:1759/5535 train_time:422402ms step_avg:240.14ms step:1760/5535 train_time:422646ms step_avg:240.14ms step:1761/5535 train_time:422892ms step_avg:240.14ms step:1762/5535 train_time:423137ms step_avg:240.15ms step:1763/5535 train_time:423384ms step_avg:240.15ms step:1764/5535 train_time:423632ms step_avg:240.15ms step:1765/5535 train_time:423905ms step_avg:240.17ms step:1766/5535 train_time:424129ms step_avg:240.16ms step:1767/5535 train_time:424374ms step_avg:240.17ms step:1768/5535 train_time:424618ms step_avg:240.17ms step:1769/5535 train_time:424868ms step_avg:240.17ms step:1770/5535 train_time:425117ms step_avg:240.18ms step:1771/5535 train_time:425367ms step_avg:240.18ms step:1772/5535 train_time:425611ms step_avg:240.19ms step:1773/5535 train_time:425858ms step_avg:240.19ms step:1774/5535 train_time:426107ms step_avg:240.20ms step:1775/5535 train_time:426358ms step_avg:240.20ms step:1776/5535 train_time:426603ms step_avg:240.20ms step:1777/5535 train_time:426848ms step_avg:240.21ms step:1778/5535 train_time:427094ms step_avg:240.21ms step:1779/5535 train_time:427340ms step_avg:240.21ms step:1780/5535 train_time:427586ms step_avg:240.22ms step:1781/5535 train_time:427832ms step_avg:240.22ms step:1782/5535 train_time:428079ms step_avg:240.22ms step:1783/5535 train_time:428323ms step_avg:240.23ms step:1784/5535 train_time:428567ms step_avg:240.23ms step:1785/5535 train_time:428813ms step_avg:240.23ms step:1786/5535 train_time:429058ms step_avg:240.23ms step:1787/5535 train_time:429305ms step_avg:240.24ms step:1788/5535 train_time:429552ms step_avg:240.24ms step:1789/5535 train_time:429797ms step_avg:240.24ms step:1790/5535 train_time:430043ms step_avg:240.25ms step:1791/5535 train_time:430290ms step_avg:240.25ms step:1792/5535 train_time:430538ms step_avg:240.26ms step:1793/5535 train_time:430809ms step_avg:240.27ms step:1794/5535 train_time:431030ms step_avg:240.26ms step:1795/5535 train_time:431278ms step_avg:240.27ms step:1796/5535 train_time:431527ms step_avg:240.27ms step:1797/5535 train_time:431771ms step_avg:240.27ms step:1798/5535 train_time:432019ms step_avg:240.28ms step:1799/5535 train_time:432264ms step_avg:240.28ms step:1800/5535 train_time:432510ms step_avg:240.28ms step:1801/5535 train_time:432754ms step_avg:240.29ms step:1802/5535 train_time:433000ms step_avg:240.29ms step:1803/5535 train_time:433249ms step_avg:240.29ms step:1804/5535 train_time:433500ms step_avg:240.30ms step:1805/5535 train_time:433744ms step_avg:240.30ms step:1806/5535 train_time:433992ms step_avg:240.31ms step:1807/5535 train_time:434242ms step_avg:240.31ms step:1808/5535 train_time:434485ms step_avg:240.31ms step:1809/5535 train_time:434730ms step_avg:240.31ms step:1810/5535 train_time:434974ms step_avg:240.32ms step:1811/5535 train_time:435222ms step_avg:240.32ms step:1812/5535 train_time:435467ms step_avg:240.32ms step:1813/5535 train_time:435712ms step_avg:240.33ms step:1814/5535 train_time:435957ms step_avg:240.33ms step:1815/5535 train_time:436203ms step_avg:240.33ms step:1816/5535 train_time:436453ms step_avg:240.34ms step:1817/5535 train_time:436699ms step_avg:240.34ms step:1818/5535 train_time:436949ms step_avg:240.35ms step:1819/5535 train_time:437195ms step_avg:240.35ms step:1820/5535 train_time:437440ms step_avg:240.35ms step:1821/5535 train_time:437719ms step_avg:240.37ms step:1822/5535 train_time:437940ms step_avg:240.36ms step:1823/5535 train_time:438183ms step_avg:240.36ms step:1824/5535 train_time:438431ms step_avg:240.37ms step:1825/5535 train_time:438678ms step_avg:240.37ms step:1826/5535 train_time:438927ms step_avg:240.38ms step:1827/5535 train_time:439178ms step_avg:240.38ms step:1828/5535 train_time:439426ms step_avg:240.39ms step:1829/5535 train_time:439681ms step_avg:240.39ms step:1830/5535 train_time:439924ms step_avg:240.40ms step:1831/5535 train_time:440168ms step_avg:240.40ms step:1832/5535 train_time:440415ms step_avg:240.40ms step:1833/5535 train_time:440659ms step_avg:240.40ms step:1834/5535 train_time:440904ms step_avg:240.41ms step:1835/5535 train_time:441149ms step_avg:240.41ms step:1836/5535 train_time:441395ms step_avg:240.41ms step:1837/5535 train_time:441642ms step_avg:240.41ms step:1838/5535 train_time:441886ms step_avg:240.42ms step:1839/5535 train_time:442130ms step_avg:240.42ms step:1840/5535 train_time:442376ms step_avg:240.42ms step:1841/5535 train_time:442621ms step_avg:240.42ms step:1842/5535 train_time:442867ms step_avg:240.43ms step:1843/5535 train_time:443117ms step_avg:240.43ms step:1844/5535 train_time:443363ms step_avg:240.44ms step:1845/5535 train_time:443608ms step_avg:240.44ms step:1846/5535 train_time:443854ms step_avg:240.44ms step:1847/5535 train_time:444099ms step_avg:240.44ms step:1848/5535 train_time:444351ms step_avg:240.45ms step:1849/5535 train_time:444628ms step_avg:240.47ms step:1850/5535 train_time:444853ms step_avg:240.46ms step:1851/5535 train_time:445100ms step_avg:240.46ms step:1852/5535 train_time:445352ms step_avg:240.47ms step:1853/5535 train_time:445599ms step_avg:240.47ms step:1854/5535 train_time:445848ms step_avg:240.48ms step:1855/5535 train_time:446094ms step_avg:240.48ms step:1856/5535 train_time:446341ms step_avg:240.49ms step:1857/5535 train_time:446586ms step_avg:240.49ms step:1858/5535 train_time:446836ms step_avg:240.49ms step:1859/5535 train_time:447083ms step_avg:240.50ms step:1860/5535 train_time:447328ms step_avg:240.50ms step:1861/5535 train_time:447577ms step_avg:240.50ms step:1862/5535 train_time:447827ms step_avg:240.51ms step:1863/5535 train_time:448078ms step_avg:240.51ms step:1864/5535 train_time:448325ms step_avg:240.52ms step:1865/5535 train_time:448573ms step_avg:240.52ms step:1866/5535 train_time:448824ms step_avg:240.53ms step:1867/5535 train_time:449070ms step_avg:240.53ms step:1868/5535 train_time:449317ms step_avg:240.53ms step:1869/5535 train_time:449566ms step_avg:240.54ms step:1870/5535 train_time:449814ms step_avg:240.54ms step:1871/5535 train_time:450060ms step_avg:240.55ms step:1872/5535 train_time:450307ms step_avg:240.55ms step:1873/5535 train_time:450557ms step_avg:240.55ms step:1874/5535 train_time:450802ms step_avg:240.56ms step:1875/5535 train_time:451049ms step_avg:240.56ms step:1875/5535 val_loss:3.202163 train_time:451298ms step_avg:240.69ms step:1876/5535 train_time:451319ms step_avg:240.58ms step:1877/5535 train_time:451575ms step_avg:240.58ms step:1878/5535 train_time:451809ms step_avg:240.58ms step:1879/5535 train_time:452061ms step_avg:240.59ms step:1880/5535 train_time:452305ms step_avg:240.59ms step:1881/5535 train_time:452551ms step_avg:240.59ms step:1882/5535 train_time:452801ms step_avg:240.60ms step:1883/5535 train_time:453053ms step_avg:240.60ms step:1884/5535 train_time:453301ms step_avg:240.61ms step:1885/5535 train_time:453551ms step_avg:240.61ms step:1886/5535 train_time:453797ms step_avg:240.61ms step:1887/5535 train_time:454046ms step_avg:240.62ms step:1888/5535 train_time:454292ms step_avg:240.62ms step:1889/5535 train_time:454538ms step_avg:240.62ms step:1890/5535 train_time:454787ms step_avg:240.63ms step:1891/5535 train_time:455038ms step_avg:240.63ms step:1892/5535 train_time:455284ms step_avg:240.64ms step:1893/5535 train_time:455533ms step_avg:240.64ms step:1894/5535 train_time:455778ms step_avg:240.64ms step:1895/5535 train_time:456026ms step_avg:240.65ms step:1896/5535 train_time:456274ms step_avg:240.65ms step:1897/5535 train_time:456522ms step_avg:240.65ms step:1898/5535 train_time:456771ms step_avg:240.66ms step:1899/5535 train_time:457019ms step_avg:240.66ms step:1900/5535 train_time:457266ms step_avg:240.67ms step:1901/5535 train_time:457515ms step_avg:240.67ms step:1902/5535 train_time:457764ms step_avg:240.68ms step:1903/5535 train_time:458016ms step_avg:240.68ms step:1904/5535 train_time:458261ms step_avg:240.68ms step:1905/5535 train_time:458537ms step_avg:240.70ms step:1906/5535 train_time:458759ms step_avg:240.69ms step:1907/5535 train_time:459006ms step_avg:240.70ms step:1908/5535 train_time:459252ms step_avg:240.70ms step:1909/5535 train_time:459502ms step_avg:240.70ms step:1910/5535 train_time:459755ms step_avg:240.71ms step:1911/5535 train_time:460002ms step_avg:240.71ms step:1912/5535 train_time:460247ms step_avg:240.72ms step:1913/5535 train_time:460493ms step_avg:240.72ms step:1914/5535 train_time:460747ms step_avg:240.72ms step:1915/5535 train_time:461000ms step_avg:240.73ms step:1916/5535 train_time:461250ms step_avg:240.74ms step:1917/5535 train_time:461500ms step_avg:240.74ms step:1918/5535 train_time:461750ms step_avg:240.75ms step:1919/5535 train_time:462000ms step_avg:240.75ms step:1920/5535 train_time:462249ms step_avg:240.75ms step:1921/5535 train_time:462495ms step_avg:240.76ms step:1922/5535 train_time:462743ms step_avg:240.76ms step:1923/5535 train_time:462993ms step_avg:240.77ms step:1924/5535 train_time:463240ms step_avg:240.77ms step:1925/5535 train_time:463489ms step_avg:240.77ms step:1926/5535 train_time:463734ms step_avg:240.78ms step:1927/5535 train_time:463983ms step_avg:240.78ms step:1928/5535 train_time:464235ms step_avg:240.79ms step:1929/5535 train_time:464487ms step_avg:240.79ms step:1930/5535 train_time:464730ms step_avg:240.79ms step:1931/5535 train_time:464978ms step_avg:240.80ms step:1932/5535 train_time:465226ms step_avg:240.80ms step:1933/5535 train_time:465514ms step_avg:240.82ms step:1934/5535 train_time:465738ms step_avg:240.82ms step:1935/5535 train_time:465983ms step_avg:240.82ms step:1936/5535 train_time:466237ms step_avg:240.82ms step:1937/5535 train_time:466490ms step_avg:240.83ms step:1938/5535 train_time:466738ms step_avg:240.84ms step:1939/5535 train_time:466983ms step_avg:240.84ms step:1940/5535 train_time:467229ms step_avg:240.84ms step:1941/5535 train_time:467477ms step_avg:240.84ms step:1942/5535 train_time:467723ms step_avg:240.85ms step:1943/5535 train_time:467970ms step_avg:240.85ms step:1944/5535 train_time:468216ms step_avg:240.85ms step:1945/5535 train_time:468463ms step_avg:240.85ms step:1946/5535 train_time:468719ms step_avg:240.86ms step:1947/5535 train_time:468967ms step_avg:240.87ms step:1948/5535 train_time:469224ms step_avg:240.87ms step:1949/5535 train_time:469470ms step_avg:240.88ms step:1950/5535 train_time:469723ms step_avg:240.88ms step:1951/5535 train_time:469972ms step_avg:240.89ms step:1952/5535 train_time:470219ms step_avg:240.89ms step:1953/5535 train_time:470477ms step_avg:240.90ms step:1954/5535 train_time:470728ms step_avg:240.90ms step:1955/5535 train_time:470974ms step_avg:240.91ms step:1956/5535 train_time:471223ms step_avg:240.91ms step:1957/5535 train_time:471470ms step_avg:240.91ms step:1958/5535 train_time:471716ms step_avg:240.92ms step:1959/5535 train_time:471962ms step_avg:240.92ms step:1960/5535 train_time:472216ms step_avg:240.93ms step:1961/5535 train_time:472492ms step_avg:240.94ms step:1962/5535 train_time:472715ms step_avg:240.94ms step:1963/5535 train_time:472962ms step_avg:240.94ms step:1964/5535 train_time:473210ms step_avg:240.94ms step:1965/5535 train_time:473456ms step_avg:240.94ms step:1966/5535 train_time:473707ms step_avg:240.95ms step:1967/5535 train_time:473955ms step_avg:240.95ms step:1968/5535 train_time:474206ms step_avg:240.96ms step:1969/5535 train_time:474453ms step_avg:240.96ms step:1970/5535 train_time:474704ms step_avg:240.97ms step:1971/5535 train_time:474950ms step_avg:240.97ms step:1972/5535 train_time:475199ms step_avg:240.97ms step:1973/5535 train_time:475444ms step_avg:240.98ms step:1974/5535 train_time:475691ms step_avg:240.98ms step:1975/5535 train_time:475941ms step_avg:240.98ms step:1976/5535 train_time:476189ms step_avg:240.99ms step:1977/5535 train_time:476441ms step_avg:240.99ms step:1978/5535 train_time:476689ms step_avg:241.00ms step:1979/5535 train_time:476935ms step_avg:241.00ms step:1980/5535 train_time:477181ms step_avg:241.00ms step:1981/5535 train_time:477427ms step_avg:241.00ms step:1982/5535 train_time:477676ms step_avg:241.01ms step:1983/5535 train_time:477921ms step_avg:241.01ms step:1984/5535 train_time:478172ms step_avg:241.01ms step:1985/5535 train_time:478422ms step_avg:241.02ms step:1986/5535 train_time:478668ms step_avg:241.02ms step:1987/5535 train_time:478915ms step_avg:241.02ms step:1988/5535 train_time:479163ms step_avg:241.03ms step:1989/5535 train_time:479443ms step_avg:241.05ms step:1990/5535 train_time:479668ms step_avg:241.04ms step:1991/5535 train_time:479917ms step_avg:241.04ms step:1992/5535 train_time:480165ms step_avg:241.05ms step:1993/5535 train_time:480416ms step_avg:241.05ms step:1994/5535 train_time:480664ms step_avg:241.05ms step:1995/5535 train_time:480921ms step_avg:241.06ms step:1996/5535 train_time:481166ms step_avg:241.07ms step:1997/5535 train_time:481413ms step_avg:241.07ms step:1998/5535 train_time:481660ms step_avg:241.07ms step:1999/5535 train_time:481909ms step_avg:241.07ms step:2000/5535 train_time:482160ms step_avg:241.08ms step:2000/5535 val_loss:3.187255 train_time:482409ms step_avg:241.20ms step:2001/5535 train_time:482430ms step_avg:241.09ms step:2002/5535 train_time:482656ms step_avg:241.09ms step:2003/5535 train_time:482914ms step_avg:241.10ms step:2004/5535 train_time:483158ms step_avg:241.10ms step:2005/5535 train_time:483404ms step_avg:241.10ms step:2006/5535 train_time:483650ms step_avg:241.10ms step:2007/5535 train_time:483903ms step_avg:241.11ms step:2008/5535 train_time:484147ms step_avg:241.11ms step:2009/5535 train_time:484405ms step_avg:241.12ms step:2010/5535 train_time:484651ms step_avg:241.12ms step:2011/5535 train_time:484896ms step_avg:241.12ms step:2012/5535 train_time:485143ms step_avg:241.12ms step:2013/5535 train_time:485394ms step_avg:241.13ms step:2014/5535 train_time:485639ms step_avg:241.13ms step:2015/5535 train_time:485887ms step_avg:241.13ms step:2016/5535 train_time:486134ms step_avg:241.14ms step:2017/5535 train_time:486412ms step_avg:241.16ms step:2018/5535 train_time:486641ms step_avg:241.15ms step:2019/5535 train_time:486889ms step_avg:241.15ms step:2020/5535 train_time:487136ms step_avg:241.16ms step:2021/5535 train_time:487384ms step_avg:241.16ms step:2022/5535 train_time:487634ms step_avg:241.16ms step:2023/5535 train_time:487879ms step_avg:241.17ms step:2024/5535 train_time:488128ms step_avg:241.17ms step:2025/5535 train_time:488377ms step_avg:241.17ms step:2026/5535 train_time:488628ms step_avg:241.18ms step:2027/5535 train_time:488877ms step_avg:241.18ms step:2028/5535 train_time:489127ms step_avg:241.19ms step:2029/5535 train_time:489379ms step_avg:241.19ms step:2030/5535 train_time:489630ms step_avg:241.20ms step:2031/5535 train_time:489878ms step_avg:241.20ms step:2032/5535 train_time:490129ms step_avg:241.21ms step:2033/5535 train_time:490381ms step_avg:241.21ms step:2034/5535 train_time:490626ms step_avg:241.21ms step:2035/5535 train_time:490873ms step_avg:241.22ms step:2036/5535 train_time:491123ms step_avg:241.22ms step:2037/5535 train_time:491368ms step_avg:241.22ms step:2038/5535 train_time:491623ms step_avg:241.23ms step:2039/5535 train_time:491868ms step_avg:241.23ms step:2040/5535 train_time:492114ms step_avg:241.23ms step:2041/5535 train_time:492361ms step_avg:241.24ms step:2042/5535 train_time:492610ms step_avg:241.24ms step:2043/5535 train_time:492867ms step_avg:241.25ms step:2044/5535 train_time:493111ms step_avg:241.25ms step:2045/5535 train_time:493389ms step_avg:241.27ms step:2046/5535 train_time:493616ms step_avg:241.26ms step:2047/5535 train_time:493867ms step_avg:241.26ms step:2048/5535 train_time:494122ms step_avg:241.27ms step:2049/5535 train_time:494379ms step_avg:241.28ms step:2050/5535 train_time:494625ms step_avg:241.28ms step:2051/5535 train_time:494872ms step_avg:241.28ms step:2052/5535 train_time:495121ms step_avg:241.29ms step:2053/5535 train_time:495368ms step_avg:241.29ms step:2054/5535 train_time:495619ms step_avg:241.29ms step:2055/5535 train_time:495870ms step_avg:241.30ms step:2056/5535 train_time:496118ms step_avg:241.30ms step:2057/5535 train_time:496366ms step_avg:241.31ms step:2058/5535 train_time:496616ms step_avg:241.31ms step:2059/5535 train_time:496865ms step_avg:241.31ms step:2060/5535 train_time:497113ms step_avg:241.32ms step:2061/5535 train_time:497364ms step_avg:241.32ms step:2062/5535 train_time:497606ms step_avg:241.32ms step:2063/5535 train_time:497857ms step_avg:241.33ms step:2064/5535 train_time:498104ms step_avg:241.33ms step:2065/5535 train_time:498352ms step_avg:241.33ms step:2066/5535 train_time:498607ms step_avg:241.34ms step:2067/5535 train_time:498851ms step_avg:241.34ms step:2068/5535 train_time:499099ms step_avg:241.34ms step:2069/5535 train_time:499356ms step_avg:241.35ms step:2070/5535 train_time:499604ms step_avg:241.35ms step:2071/5535 train_time:499849ms step_avg:241.36ms step:2072/5535 train_time:500097ms step_avg:241.36ms step:2073/5535 train_time:500375ms step_avg:241.38ms step:2074/5535 train_time:500603ms step_avg:241.37ms step:2075/5535 train_time:500847ms step_avg:241.37ms step:2076/5535 train_time:501097ms step_avg:241.38ms step:2077/5535 train_time:501347ms step_avg:241.38ms step:2078/5535 train_time:501594ms step_avg:241.38ms step:2079/5535 train_time:501843ms step_avg:241.39ms step:2080/5535 train_time:502088ms step_avg:241.39ms step:2081/5535 train_time:502339ms step_avg:241.39ms step:2082/5535 train_time:502589ms step_avg:241.40ms step:2083/5535 train_time:502838ms step_avg:241.40ms step:2084/5535 train_time:503082ms step_avg:241.40ms step:2085/5535 train_time:503327ms step_avg:241.40ms step:2086/5535 train_time:503575ms step_avg:241.41ms step:2087/5535 train_time:503824ms step_avg:241.41ms step:2088/5535 train_time:504068ms step_avg:241.41ms step:2089/5535 train_time:504320ms step_avg:241.42ms step:2090/5535 train_time:504566ms step_avg:241.42ms step:2091/5535 train_time:504812ms step_avg:241.42ms step:2092/5535 train_time:505059ms step_avg:241.42ms step:2093/5535 train_time:505306ms step_avg:241.43ms step:2094/5535 train_time:505555ms step_avg:241.43ms step:2095/5535 train_time:505807ms step_avg:241.44ms step:2096/5535 train_time:506060ms step_avg:241.44ms step:2097/5535 train_time:506307ms step_avg:241.44ms step:2098/5535 train_time:506553ms step_avg:241.45ms step:2099/5535 train_time:506802ms step_avg:241.45ms step:2100/5535 train_time:507053ms step_avg:241.45ms step:2101/5535 train_time:507331ms step_avg:241.47ms step:2102/5535 train_time:507561ms step_avg:241.47ms step:2103/5535 train_time:507809ms step_avg:241.47ms step:2104/5535 train_time:508058ms step_avg:241.47ms step:2105/5535 train_time:508303ms step_avg:241.47ms step:2106/5535 train_time:508558ms step_avg:241.48ms step:2107/5535 train_time:508803ms step_avg:241.48ms step:2108/5535 train_time:509049ms step_avg:241.48ms step:2109/5535 train_time:509301ms step_avg:241.49ms step:2110/5535 train_time:509551ms step_avg:241.49ms step:2111/5535 train_time:509800ms step_avg:241.50ms step:2112/5535 train_time:510049ms step_avg:241.50ms step:2113/5535 train_time:510298ms step_avg:241.50ms step:2114/5535 train_time:510547ms step_avg:241.51ms step:2115/5535 train_time:510796ms step_avg:241.51ms step:2116/5535 train_time:511041ms step_avg:241.51ms step:2117/5535 train_time:511294ms step_avg:241.52ms step:2118/5535 train_time:511542ms step_avg:241.52ms step:2119/5535 train_time:511791ms step_avg:241.52ms step:2120/5535 train_time:512038ms step_avg:241.53ms step:2121/5535 train_time:512288ms step_avg:241.53ms step:2122/5535 train_time:512537ms step_avg:241.53ms step:2123/5535 train_time:512783ms step_avg:241.54ms step:2124/5535 train_time:513031ms step_avg:241.54ms step:2125/5535 train_time:513282ms step_avg:241.54ms step:2125/5535 val_loss:3.173715 train_time:513533ms step_avg:241.66ms step:2126/5535 train_time:513554ms step_avg:241.56ms step:2127/5535 train_time:513788ms step_avg:241.56ms step:2128/5535 train_time:514042ms step_avg:241.56ms step:2129/5535 train_time:514317ms step_avg:241.58ms step:2130/5535 train_time:514536ms step_avg:241.57ms step:2131/5535 train_time:514784ms step_avg:241.57ms step:2132/5535 train_time:515036ms step_avg:241.57ms step:2133/5535 train_time:515284ms step_avg:241.58ms step:2134/5535 train_time:515533ms step_avg:241.58ms step:2135/5535 train_time:515779ms step_avg:241.58ms step:2136/5535 train_time:516030ms step_avg:241.59ms step:2137/5535 train_time:516276ms step_avg:241.59ms step:2138/5535 train_time:516521ms step_avg:241.59ms step:2139/5535 train_time:516769ms step_avg:241.59ms step:2140/5535 train_time:517014ms step_avg:241.60ms step:2141/5535 train_time:517260ms step_avg:241.60ms step:2142/5535 train_time:517507ms step_avg:241.60ms step:2143/5535 train_time:517752ms step_avg:241.60ms step:2144/5535 train_time:517999ms step_avg:241.60ms step:2145/5535 train_time:518247ms step_avg:241.61ms step:2146/5535 train_time:518500ms step_avg:241.61ms step:2147/5535 train_time:518749ms step_avg:241.62ms step:2148/5535 train_time:518997ms step_avg:241.62ms step:2149/5535 train_time:519244ms step_avg:241.62ms step:2150/5535 train_time:519496ms step_avg:241.63ms step:2151/5535 train_time:519741ms step_avg:241.63ms step:2152/5535 train_time:519990ms step_avg:241.63ms step:2153/5535 train_time:520237ms step_avg:241.63ms step:2154/5535 train_time:520487ms step_avg:241.64ms step:2155/5535 train_time:520742ms step_avg:241.64ms step:2156/5535 train_time:520990ms step_avg:241.65ms step:2157/5535 train_time:521264ms step_avg:241.66ms step:2158/5535 train_time:521486ms step_avg:241.65ms step:2159/5535 train_time:521732ms step_avg:241.65ms step:2160/5535 train_time:521982ms step_avg:241.66ms step:2161/5535 train_time:522231ms step_avg:241.66ms step:2162/5535 train_time:522478ms step_avg:241.66ms step:2163/5535 train_time:522725ms step_avg:241.67ms step:2164/5535 train_time:522978ms step_avg:241.67ms step:2165/5535 train_time:523230ms step_avg:241.68ms step:2166/5535 train_time:523473ms step_avg:241.68ms step:2167/5535 train_time:523717ms step_avg:241.68ms step:2168/5535 train_time:523968ms step_avg:241.68ms step:2169/5535 train_time:524213ms step_avg:241.68ms step:2170/5535 train_time:524461ms step_avg:241.69ms step:2171/5535 train_time:524710ms step_avg:241.69ms step:2172/5535 train_time:524956ms step_avg:241.69ms step:2173/5535 train_time:525204ms step_avg:241.70ms step:2174/5535 train_time:525457ms step_avg:241.70ms step:2175/5535 train_time:525704ms step_avg:241.70ms step:2176/5535 train_time:525952ms step_avg:241.71ms step:2177/5535 train_time:526198ms step_avg:241.71ms step:2178/5535 train_time:526445ms step_avg:241.71ms step:2179/5535 train_time:526698ms step_avg:241.72ms step:2180/5535 train_time:526950ms step_avg:241.72ms step:2181/5535 train_time:527198ms step_avg:241.72ms step:2182/5535 train_time:527445ms step_avg:241.73ms step:2183/5535 train_time:527691ms step_avg:241.73ms step:2184/5535 train_time:527943ms step_avg:241.73ms step:2185/5535 train_time:528218ms step_avg:241.75ms step:2186/5535 train_time:528442ms step_avg:241.74ms step:2187/5535 train_time:528693ms step_avg:241.74ms step:2188/5535 train_time:528939ms step_avg:241.75ms step:2189/5535 train_time:529190ms step_avg:241.75ms step:2190/5535 train_time:529439ms step_avg:241.75ms step:2191/5535 train_time:529684ms step_avg:241.75ms step:2192/5535 train_time:529933ms step_avg:241.76ms step:2193/5535 train_time:530181ms step_avg:241.76ms step:2194/5535 train_time:530429ms step_avg:241.76ms step:2195/5535 train_time:530676ms step_avg:241.77ms step:2196/5535 train_time:530923ms step_avg:241.77ms step:2197/5535 train_time:531169ms step_avg:241.77ms step:2198/5535 train_time:531414ms step_avg:241.77ms step:2199/5535 train_time:531661ms step_avg:241.77ms step:2200/5535 train_time:531912ms step_avg:241.78ms step:2201/5535 train_time:532159ms step_avg:241.78ms step:2202/5535 train_time:532407ms step_avg:241.78ms step:2203/5535 train_time:532656ms step_avg:241.79ms step:2204/5535 train_time:532906ms step_avg:241.79ms step:2205/5535 train_time:533163ms step_avg:241.80ms step:2206/5535 train_time:533408ms step_avg:241.80ms step:2207/5535 train_time:533664ms step_avg:241.80ms step:2208/5535 train_time:533909ms step_avg:241.81ms step:2209/5535 train_time:534156ms step_avg:241.81ms step:2210/5535 train_time:534405ms step_avg:241.81ms step:2211/5535 train_time:534651ms step_avg:241.81ms step:2212/5535 train_time:534899ms step_avg:241.82ms step:2213/5535 train_time:535175ms step_avg:241.83ms step:2214/5535 train_time:535403ms step_avg:241.83ms step:2215/5535 train_time:535651ms step_avg:241.83ms step:2216/5535 train_time:535903ms step_avg:241.83ms step:2217/5535 train_time:536159ms step_avg:241.84ms step:2218/5535 train_time:536409ms step_avg:241.84ms step:2219/5535 train_time:536657ms step_avg:241.85ms step:2220/5535 train_time:536910ms step_avg:241.85ms step:2221/5535 train_time:537161ms step_avg:241.86ms step:2222/5535 train_time:537409ms step_avg:241.86ms step:2223/5535 train_time:537655ms step_avg:241.86ms step:2224/5535 train_time:537909ms step_avg:241.87ms step:2225/5535 train_time:538155ms step_avg:241.87ms step:2226/5535 train_time:538410ms step_avg:241.87ms step:2227/5535 train_time:538655ms step_avg:241.87ms step:2228/5535 train_time:538905ms step_avg:241.88ms step:2229/5535 train_time:539155ms step_avg:241.88ms step:2230/5535 train_time:539403ms step_avg:241.88ms step:2231/5535 train_time:539651ms step_avg:241.89ms step:2232/5535 train_time:539903ms step_avg:241.89ms step:2233/5535 train_time:540157ms step_avg:241.90ms step:2234/5535 train_time:540404ms step_avg:241.90ms step:2235/5535 train_time:540650ms step_avg:241.90ms step:2236/5535 train_time:540901ms step_avg:241.91ms step:2237/5535 train_time:541156ms step_avg:241.91ms step:2238/5535 train_time:541401ms step_avg:241.91ms step:2239/5535 train_time:541653ms step_avg:241.92ms step:2240/5535 train_time:541903ms step_avg:241.92ms step:2241/5535 train_time:542181ms step_avg:241.94ms step:2242/5535 train_time:542407ms step_avg:241.93ms step:2243/5535 train_time:542654ms step_avg:241.93ms step:2244/5535 train_time:542900ms step_avg:241.93ms step:2245/5535 train_time:543149ms step_avg:241.94ms step:2246/5535 train_time:543398ms step_avg:241.94ms step:2247/5535 train_time:543646ms step_avg:241.94ms step:2248/5535 train_time:543897ms step_avg:241.95ms step:2249/5535 train_time:544146ms step_avg:241.95ms step:2250/5535 train_time:544393ms step_avg:241.95ms step:2250/5535 val_loss:3.159035 train_time:544643ms step_avg:242.06ms step:2251/5535 train_time:544664ms step_avg:241.97ms step:2252/5535 train_time:544891ms step_avg:241.96ms step:2253/5535 train_time:545148ms step_avg:241.97ms step:2254/5535 train_time:545393ms step_avg:241.97ms step:2255/5535 train_time:545637ms step_avg:241.97ms step:2256/5535 train_time:545883ms step_avg:241.97ms step:2257/5535 train_time:546134ms step_avg:241.97ms step:2258/5535 train_time:546386ms step_avg:241.98ms step:2259/5535 train_time:546633ms step_avg:241.98ms step:2260/5535 train_time:546880ms step_avg:241.98ms step:2261/5535 train_time:547129ms step_avg:241.99ms step:2262/5535 train_time:547376ms step_avg:241.99ms step:2263/5535 train_time:547622ms step_avg:241.99ms step:2264/5535 train_time:547871ms step_avg:241.99ms step:2265/5535 train_time:548122ms step_avg:242.00ms step:2266/5535 train_time:548368ms step_avg:242.00ms step:2267/5535 train_time:548614ms step_avg:242.00ms step:2268/5535 train_time:548861ms step_avg:242.00ms step:2269/5535 train_time:549139ms step_avg:242.02ms step:2270/5535 train_time:549363ms step_avg:242.01ms step:2271/5535 train_time:549612ms step_avg:242.01ms step:2272/5535 train_time:549865ms step_avg:242.02ms step:2273/5535 train_time:550111ms step_avg:242.02ms step:2274/5535 train_time:550362ms step_avg:242.02ms step:2275/5535 train_time:550615ms step_avg:242.03ms step:2276/5535 train_time:550859ms step_avg:242.03ms step:2277/5535 train_time:551104ms step_avg:242.03ms step:2278/5535 train_time:551350ms step_avg:242.03ms step:2279/5535 train_time:551603ms step_avg:242.04ms step:2280/5535 train_time:551848ms step_avg:242.04ms step:2281/5535 train_time:552095ms step_avg:242.04ms step:2282/5535 train_time:552344ms step_avg:242.04ms step:2283/5535 train_time:552590ms step_avg:242.05ms step:2284/5535 train_time:552839ms step_avg:242.05ms step:2285/5535 train_time:553090ms step_avg:242.05ms step:2286/5535 train_time:553337ms step_avg:242.05ms step:2287/5535 train_time:553586ms step_avg:242.06ms step:2288/5535 train_time:553837ms step_avg:242.06ms step:2289/5535 train_time:554087ms step_avg:242.06ms step:2290/5535 train_time:554334ms step_avg:242.07ms step:2291/5535 train_time:554579ms step_avg:242.07ms step:2292/5535 train_time:554827ms step_avg:242.07ms step:2293/5535 train_time:555079ms step_avg:242.08ms step:2294/5535 train_time:555325ms step_avg:242.08ms step:2295/5535 train_time:555572ms step_avg:242.08ms step:2296/5535 train_time:555823ms step_avg:242.08ms step:2297/5535 train_time:556107ms step_avg:242.10ms step:2298/5535 train_time:556331ms step_avg:242.09ms step:2299/5535 train_time:556580ms step_avg:242.10ms step:2300/5535 train_time:556825ms step_avg:242.10ms step:2301/5535 train_time:557076ms step_avg:242.10ms step:2302/5535 train_time:557324ms step_avg:242.10ms step:2303/5535 train_time:557574ms step_avg:242.11ms step:2304/5535 train_time:557819ms step_avg:242.11ms step:2305/5535 train_time:558070ms step_avg:242.11ms step:2306/5535 train_time:558322ms step_avg:242.12ms step:2307/5535 train_time:558570ms step_avg:242.12ms step:2308/5535 train_time:558819ms step_avg:242.12ms step:2309/5535 train_time:559070ms step_avg:242.13ms step:2310/5535 train_time:559321ms step_avg:242.13ms step:2311/5535 train_time:559568ms step_avg:242.13ms step:2312/5535 train_time:559815ms step_avg:242.13ms step:2313/5535 train_time:560064ms step_avg:242.14ms step:2314/5535 train_time:560314ms step_avg:242.14ms step:2315/5535 train_time:560565ms step_avg:242.14ms step:2316/5535 train_time:560815ms step_avg:242.15ms step:2317/5535 train_time:561060ms step_avg:242.15ms step:2318/5535 train_time:561309ms step_avg:242.15ms step:2319/5535 train_time:561561ms step_avg:242.16ms step:2320/5535 train_time:561810ms step_avg:242.16ms step:2321/5535 train_time:562064ms step_avg:242.16ms step:2322/5535 train_time:562317ms step_avg:242.17ms step:2323/5535 train_time:562565ms step_avg:242.17ms step:2324/5535 train_time:562814ms step_avg:242.17ms step:2325/5535 train_time:563090ms step_avg:242.19ms step:2326/5535 train_time:563317ms step_avg:242.18ms step:2327/5535 train_time:563565ms step_avg:242.19ms step:2328/5535 train_time:563813ms step_avg:242.19ms step:2329/5535 train_time:564060ms step_avg:242.19ms step:2330/5535 train_time:564313ms step_avg:242.19ms step:2331/5535 train_time:564559ms step_avg:242.20ms step:2332/5535 train_time:564811ms step_avg:242.20ms step:2333/5535 train_time:565057ms step_avg:242.20ms step:2334/5535 train_time:565306ms step_avg:242.20ms step:2335/5535 train_time:565551ms step_avg:242.21ms step:2336/5535 train_time:565801ms step_avg:242.21ms step:2337/5535 train_time:566049ms step_avg:242.21ms step:2338/5535 train_time:566299ms step_avg:242.22ms step:2339/5535 train_time:566543ms step_avg:242.22ms step:2340/5535 train_time:566793ms step_avg:242.22ms step:2341/5535 train_time:567042ms step_avg:242.22ms step:2342/5535 train_time:567297ms step_avg:242.23ms step:2343/5535 train_time:567543ms step_avg:242.23ms step:2344/5535 train_time:567792ms step_avg:242.23ms step:2345/5535 train_time:568049ms step_avg:242.24ms step:2346/5535 train_time:568298ms step_avg:242.24ms step:2347/5535 train_time:568548ms step_avg:242.24ms step:2348/5535 train_time:568797ms step_avg:242.25ms step:2349/5535 train_time:569047ms step_avg:242.25ms step:2350/5535 train_time:569295ms step_avg:242.25ms step:2351/5535 train_time:569543ms step_avg:242.26ms step:2352/5535 train_time:569791ms step_avg:242.26ms step:2353/5535 train_time:570070ms step_avg:242.27ms step:2354/5535 train_time:570297ms step_avg:242.27ms step:2355/5535 train_time:570545ms step_avg:242.27ms step:2356/5535 train_time:570804ms step_avg:242.28ms step:2357/5535 train_time:571052ms step_avg:242.28ms step:2358/5535 train_time:571301ms step_avg:242.28ms step:2359/5535 train_time:571548ms step_avg:242.28ms step:2360/5535 train_time:571793ms step_avg:242.29ms step:2361/5535 train_time:572040ms step_avg:242.29ms step:2362/5535 train_time:572286ms step_avg:242.29ms step:2363/5535 train_time:572541ms step_avg:242.29ms step:2364/5535 train_time:572786ms step_avg:242.30ms step:2365/5535 train_time:573034ms step_avg:242.30ms step:2366/5535 train_time:573285ms step_avg:242.30ms step:2367/5535 train_time:573537ms step_avg:242.31ms step:2368/5535 train_time:573789ms step_avg:242.31ms step:2369/5535 train_time:574036ms step_avg:242.31ms step:2370/5535 train_time:574285ms step_avg:242.31ms step:2371/5535 train_time:574533ms step_avg:242.32ms step:2372/5535 train_time:574782ms step_avg:242.32ms step:2373/5535 train_time:575027ms step_avg:242.32ms step:2374/5535 train_time:575274ms step_avg:242.32ms step:2375/5535 train_time:575525ms step_avg:242.33ms step:2375/5535 val_loss:3.147044 train_time:575770ms step_avg:242.43ms step:2376/5535 train_time:575791ms step_avg:242.34ms step:2377/5535 train_time:576021ms step_avg:242.33ms step:2378/5535 train_time:576278ms step_avg:242.34ms step:2379/5535 train_time:576522ms step_avg:242.34ms step:2380/5535 train_time:576764ms step_avg:242.34ms step:2381/5535 train_time:577039ms step_avg:242.35ms step:2382/5535 train_time:577272ms step_avg:242.35ms step:2383/5535 train_time:577523ms step_avg:242.35ms step:2384/5535 train_time:577768ms step_avg:242.35ms step:2385/5535 train_time:578013ms step_avg:242.35ms step:2386/5535 train_time:578263ms step_avg:242.36ms step:2387/5535 train_time:578511ms step_avg:242.36ms step:2388/5535 train_time:578760ms step_avg:242.36ms step:2389/5535 train_time:579007ms step_avg:242.36ms step:2390/5535 train_time:579259ms step_avg:242.37ms step:2391/5535 train_time:579505ms step_avg:242.37ms step:2392/5535 train_time:579752ms step_avg:242.37ms step:2393/5535 train_time:579997ms step_avg:242.37ms step:2394/5535 train_time:580245ms step_avg:242.37ms step:2395/5535 train_time:580494ms step_avg:242.38ms step:2396/5535 train_time:580744ms step_avg:242.38ms step:2397/5535 train_time:580995ms step_avg:242.38ms step:2398/5535 train_time:581246ms step_avg:242.39ms step:2399/5535 train_time:581494ms step_avg:242.39ms step:2400/5535 train_time:581751ms step_avg:242.40ms step:2401/5535 train_time:581998ms step_avg:242.40ms step:2402/5535 train_time:582248ms step_avg:242.40ms step:2403/5535 train_time:582496ms step_avg:242.40ms step:2404/5535 train_time:582746ms step_avg:242.41ms step:2405/5535 train_time:582992ms step_avg:242.41ms step:2406/5535 train_time:583245ms step_avg:242.41ms step:2407/5535 train_time:583500ms step_avg:242.42ms step:2408/5535 train_time:583748ms step_avg:242.42ms step:2409/5535 train_time:584024ms step_avg:242.43ms step:2410/5535 train_time:584254ms step_avg:242.43ms step:2411/5535 train_time:584503ms step_avg:242.43ms step:2412/5535 train_time:584755ms step_avg:242.44ms step:2413/5535 train_time:585002ms step_avg:242.44ms step:2414/5535 train_time:585249ms step_avg:242.44ms step:2415/5535 train_time:585507ms step_avg:242.45ms step:2416/5535 train_time:585752ms step_avg:242.45ms step:2417/5535 train_time:585999ms step_avg:242.45ms step:2418/5535 train_time:586245ms step_avg:242.45ms step:2419/5535 train_time:586496ms step_avg:242.45ms step:2420/5535 train_time:586744ms step_avg:242.46ms step:2421/5535 train_time:586992ms step_avg:242.46ms step:2422/5535 train_time:587238ms step_avg:242.46ms step:2423/5535 train_time:587492ms step_avg:242.46ms step:2424/5535 train_time:587742ms step_avg:242.47ms step:2425/5535 train_time:587989ms step_avg:242.47ms step:2426/5535 train_time:588240ms step_avg:242.47ms step:2427/5535 train_time:588488ms step_avg:242.48ms step:2428/5535 train_time:588735ms step_avg:242.48ms step:2429/5535 train_time:588984ms step_avg:242.48ms step:2430/5535 train_time:589230ms step_avg:242.48ms step:2431/5535 train_time:589476ms step_avg:242.48ms step:2432/5535 train_time:589728ms step_avg:242.49ms step:2433/5535 train_time:589978ms step_avg:242.49ms step:2434/5535 train_time:590225ms step_avg:242.49ms step:2435/5535 train_time:590477ms step_avg:242.50ms step:2436/5535 train_time:590722ms step_avg:242.50ms step:2437/5535 train_time:590998ms step_avg:242.51ms step:2438/5535 train_time:591223ms step_avg:242.50ms step:2439/5535 train_time:591472ms step_avg:242.51ms step:2440/5535 train_time:591718ms step_avg:242.51ms step:2441/5535 train_time:591967ms step_avg:242.51ms step:2442/5535 train_time:592214ms step_avg:242.51ms step:2443/5535 train_time:592463ms step_avg:242.51ms step:2444/5535 train_time:592714ms step_avg:242.52ms step:2445/5535 train_time:592965ms step_avg:242.52ms step:2446/5535 train_time:593214ms step_avg:242.52ms step:2447/5535 train_time:593461ms step_avg:242.53ms step:2448/5535 train_time:593706ms step_avg:242.53ms step:2449/5535 train_time:593953ms step_avg:242.53ms step:2450/5535 train_time:594200ms step_avg:242.53ms step:2451/5535 train_time:594448ms step_avg:242.53ms step:2452/5535 train_time:594699ms step_avg:242.54ms step:2453/5535 train_time:594948ms step_avg:242.54ms step:2454/5535 train_time:595195ms step_avg:242.54ms step:2455/5535 train_time:595453ms step_avg:242.55ms step:2456/5535 train_time:595699ms step_avg:242.55ms step:2457/5535 train_time:595950ms step_avg:242.55ms step:2458/5535 train_time:596196ms step_avg:242.55ms step:2459/5535 train_time:596451ms step_avg:242.56ms step:2460/5535 train_time:596702ms step_avg:242.56ms step:2461/5535 train_time:596947ms step_avg:242.56ms step:2462/5535 train_time:597196ms step_avg:242.57ms step:2463/5535 train_time:597444ms step_avg:242.57ms step:2464/5535 train_time:597691ms step_avg:242.57ms step:2465/5535 train_time:597975ms step_avg:242.59ms step:2466/5535 train_time:598210ms step_avg:242.58ms step:2467/5535 train_time:598459ms step_avg:242.59ms step:2468/5535 train_time:598706ms step_avg:242.59ms step:2469/5535 train_time:598953ms step_avg:242.59ms step:2470/5535 train_time:599207ms step_avg:242.59ms step:2471/5535 train_time:599457ms step_avg:242.60ms step:2472/5535 train_time:599702ms step_avg:242.60ms step:2473/5535 train_time:599948ms step_avg:242.60ms step:2474/5535 train_time:600203ms step_avg:242.60ms step:2475/5535 train_time:600452ms step_avg:242.61ms step:2476/5535 train_time:600698ms step_avg:242.61ms step:2477/5535 train_time:600944ms step_avg:242.61ms step:2478/5535 train_time:601196ms step_avg:242.61ms step:2479/5535 train_time:601444ms step_avg:242.62ms step:2480/5535 train_time:601692ms step_avg:242.62ms step:2481/5535 train_time:601939ms step_avg:242.62ms step:2482/5535 train_time:602186ms step_avg:242.62ms step:2483/5535 train_time:602435ms step_avg:242.62ms step:2484/5535 train_time:602684ms step_avg:242.63ms step:2485/5535 train_time:602930ms step_avg:242.63ms step:2486/5535 train_time:603180ms step_avg:242.63ms step:2487/5535 train_time:603429ms step_avg:242.63ms step:2488/5535 train_time:603677ms step_avg:242.64ms step:2489/5535 train_time:603924ms step_avg:242.64ms step:2490/5535 train_time:604171ms step_avg:242.64ms step:2491/5535 train_time:604422ms step_avg:242.64ms step:2492/5535 train_time:604667ms step_avg:242.64ms step:2493/5535 train_time:604945ms step_avg:242.66ms step:2494/5535 train_time:605181ms step_avg:242.65ms step:2495/5535 train_time:605430ms step_avg:242.66ms step:2496/5535 train_time:605678ms step_avg:242.66ms step:2497/5535 train_time:605923ms step_avg:242.66ms step:2498/5535 train_time:606174ms step_avg:242.66ms step:2499/5535 train_time:606420ms step_avg:242.67ms step:2500/5535 train_time:606666ms step_avg:242.67ms step:2500/5535 val_loss:3.136047 train_time:606919ms step_avg:242.77ms step:2501/5535 train_time:606940ms step_avg:242.68ms step:2502/5535 train_time:607165ms step_avg:242.67ms step:2503/5535 train_time:607421ms step_avg:242.68ms step:2504/5535 train_time:607667ms step_avg:242.68ms step:2505/5535 train_time:607911ms step_avg:242.68ms step:2506/5535 train_time:608158ms step_avg:242.68ms step:2507/5535 train_time:608410ms step_avg:242.68ms step:2508/5535 train_time:608666ms step_avg:242.69ms step:2509/5535 train_time:608914ms step_avg:242.69ms step:2510/5535 train_time:609171ms step_avg:242.70ms step:2511/5535 train_time:609417ms step_avg:242.70ms step:2512/5535 train_time:609667ms step_avg:242.70ms step:2513/5535 train_time:609915ms step_avg:242.70ms step:2514/5535 train_time:610164ms step_avg:242.71ms step:2515/5535 train_time:610416ms step_avg:242.71ms step:2516/5535 train_time:610662ms step_avg:242.71ms step:2517/5535 train_time:610912ms step_avg:242.71ms step:2518/5535 train_time:611156ms step_avg:242.72ms step:2519/5535 train_time:611409ms step_avg:242.72ms step:2520/5535 train_time:611658ms step_avg:242.72ms step:2521/5535 train_time:611939ms step_avg:242.74ms step:2522/5535 train_time:612169ms step_avg:242.73ms step:2523/5535 train_time:612412ms step_avg:242.73ms step:2524/5535 train_time:612659ms step_avg:242.73ms step:2525/5535 train_time:612910ms step_avg:242.74ms step:2526/5535 train_time:613165ms step_avg:242.74ms step:2527/5535 train_time:613410ms step_avg:242.74ms step:2528/5535 train_time:613660ms step_avg:242.75ms step:2529/5535 train_time:613910ms step_avg:242.75ms step:2530/5535 train_time:614161ms step_avg:242.75ms step:2531/5535 train_time:614406ms step_avg:242.75ms step:2532/5535 train_time:614657ms step_avg:242.76ms step:2533/5535 train_time:614905ms step_avg:242.76ms step:2534/5535 train_time:615152ms step_avg:242.76ms step:2535/5535 train_time:615402ms step_avg:242.76ms step:2536/5535 train_time:615655ms step_avg:242.77ms step:2537/5535 train_time:615902ms step_avg:242.77ms step:2538/5535 train_time:616150ms step_avg:242.77ms step:2539/5535 train_time:616399ms step_avg:242.77ms step:2540/5535 train_time:616652ms step_avg:242.78ms step:2541/5535 train_time:616898ms step_avg:242.78ms step:2542/5535 train_time:617148ms step_avg:242.78ms step:2543/5535 train_time:617401ms step_avg:242.78ms step:2544/5535 train_time:617656ms step_avg:242.79ms step:2545/5535 train_time:617907ms step_avg:242.79ms step:2546/5535 train_time:618152ms step_avg:242.79ms step:2547/5535 train_time:618406ms step_avg:242.80ms step:2548/5535 train_time:618653ms step_avg:242.80ms step:2549/5535 train_time:618928ms step_avg:242.81ms step:2550/5535 train_time:619154ms step_avg:242.81ms step:2551/5535 train_time:619402ms step_avg:242.81ms step:2552/5535 train_time:619646ms step_avg:242.81ms step:2553/5535 train_time:619897ms step_avg:242.81ms step:2554/5535 train_time:620144ms step_avg:242.81ms step:2555/5535 train_time:620389ms step_avg:242.81ms step:2556/5535 train_time:620640ms step_avg:242.82ms step:2557/5535 train_time:620890ms step_avg:242.82ms step:2558/5535 train_time:621137ms step_avg:242.82ms step:2559/5535 train_time:621384ms step_avg:242.82ms step:2560/5535 train_time:621629ms step_avg:242.82ms step:2561/5535 train_time:621878ms step_avg:242.83ms step:2562/5535 train_time:622126ms step_avg:242.83ms step:2563/5535 train_time:622376ms step_avg:242.83ms step:2564/5535 train_time:622621ms step_avg:242.83ms step:2565/5535 train_time:622866ms step_avg:242.83ms step:2566/5535 train_time:623118ms step_avg:242.84ms step:2567/5535 train_time:623366ms step_avg:242.84ms step:2568/5535 train_time:623612ms step_avg:242.84ms step:2569/5535 train_time:623861ms step_avg:242.84ms step:2570/5535 train_time:624108ms step_avg:242.84ms step:2571/5535 train_time:624357ms step_avg:242.85ms step:2572/5535 train_time:624606ms step_avg:242.85ms step:2573/5535 train_time:624853ms step_avg:242.85ms step:2574/5535 train_time:625107ms step_avg:242.85ms step:2575/5535 train_time:625354ms step_avg:242.86ms step:2576/5535 train_time:625601ms step_avg:242.86ms step:2577/5535 train_time:625876ms step_avg:242.87ms step:2578/5535 train_time:626101ms step_avg:242.86ms step:2579/5535 train_time:626348ms step_avg:242.86ms step:2580/5535 train_time:626595ms step_avg:242.87ms step:2581/5535 train_time:626843ms step_avg:242.87ms step:2582/5535 train_time:627092ms step_avg:242.87ms step:2583/5535 train_time:627340ms step_avg:242.87ms step:2584/5535 train_time:627594ms step_avg:242.88ms step:2585/5535 train_time:627839ms step_avg:242.88ms step:2586/5535 train_time:628086ms step_avg:242.88ms step:2587/5535 train_time:628334ms step_avg:242.88ms step:2588/5535 train_time:628584ms step_avg:242.88ms step:2589/5535 train_time:628831ms step_avg:242.89ms step:2590/5535 train_time:629081ms step_avg:242.89ms step:2591/5535 train_time:629330ms step_avg:242.89ms step:2592/5535 train_time:629580ms step_avg:242.89ms step:2593/5535 train_time:629831ms step_avg:242.90ms step:2594/5535 train_time:630081ms step_avg:242.90ms step:2595/5535 train_time:630327ms step_avg:242.90ms step:2596/5535 train_time:630577ms step_avg:242.90ms step:2597/5535 train_time:630826ms step_avg:242.91ms step:2598/5535 train_time:631075ms step_avg:242.91ms step:2599/5535 train_time:631324ms step_avg:242.91ms step:2600/5535 train_time:631568ms step_avg:242.91ms step:2601/5535 train_time:631817ms step_avg:242.91ms step:2602/5535 train_time:632065ms step_avg:242.92ms step:2603/5535 train_time:632315ms step_avg:242.92ms step:2604/5535 train_time:632562ms step_avg:242.92ms step:2605/5535 train_time:632843ms step_avg:242.93ms step:2606/5535 train_time:633070ms step_avg:242.93ms step:2607/5535 train_time:633318ms step_avg:242.93ms step:2608/5535 train_time:633564ms step_avg:242.93ms step:2609/5535 train_time:633813ms step_avg:242.93ms step:2610/5535 train_time:634060ms step_avg:242.93ms step:2611/5535 train_time:634310ms step_avg:242.94ms step:2612/5535 train_time:634557ms step_avg:242.94ms step:2613/5535 train_time:634802ms step_avg:242.94ms step:2614/5535 train_time:635048ms step_avg:242.94ms step:2615/5535 train_time:635295ms step_avg:242.94ms step:2616/5535 train_time:635544ms step_avg:242.94ms step:2617/5535 train_time:635794ms step_avg:242.95ms step:2618/5535 train_time:636042ms step_avg:242.95ms step:2619/5535 train_time:636287ms step_avg:242.95ms step:2620/5535 train_time:636541ms step_avg:242.95ms step:2621/5535 train_time:636788ms step_avg:242.96ms step:2622/5535 train_time:637034ms step_avg:242.96ms step:2623/5535 train_time:637281ms step_avg:242.96ms step:2624/5535 train_time:637527ms step_avg:242.96ms step:2625/5535 train_time:637776ms step_avg:242.96ms step:2625/5535 val_loss:3.124198 train_time:638027ms step_avg:243.06ms step:2626/5535 train_time:638048ms step_avg:242.97ms step:2627/5535 train_time:638277ms step_avg:242.97ms step:2628/5535 train_time:638536ms step_avg:242.97ms step:2629/5535 train_time:638782ms step_avg:242.98ms step:2630/5535 train_time:639024ms step_avg:242.97ms step:2631/5535 train_time:639276ms step_avg:242.98ms step:2632/5535 train_time:639528ms step_avg:242.98ms step:2633/5535 train_time:639804ms step_avg:242.99ms step:2634/5535 train_time:640027ms step_avg:242.99ms step:2635/5535 train_time:640276ms step_avg:242.99ms step:2636/5535 train_time:640525ms step_avg:242.99ms step:2637/5535 train_time:640774ms step_avg:242.99ms step:2638/5535 train_time:641022ms step_avg:243.00ms step:2639/5535 train_time:641270ms step_avg:243.00ms step:2640/5535 train_time:641519ms step_avg:243.00ms step:2641/5535 train_time:641768ms step_avg:243.00ms step:2642/5535 train_time:642013ms step_avg:243.00ms step:2643/5535 train_time:642261ms step_avg:243.00ms step:2644/5535 train_time:642512ms step_avg:243.01ms step:2645/5535 train_time:642760ms step_avg:243.01ms step:2646/5535 train_time:643009ms step_avg:243.01ms step:2647/5535 train_time:643260ms step_avg:243.01ms step:2648/5535 train_time:643505ms step_avg:243.02ms step:2649/5535 train_time:643767ms step_avg:243.02ms step:2650/5535 train_time:644011ms step_avg:243.02ms step:2651/5535 train_time:644259ms step_avg:243.03ms step:2652/5535 train_time:644507ms step_avg:243.03ms step:2653/5535 train_time:644751ms step_avg:243.03ms step:2654/5535 train_time:645000ms step_avg:243.03ms step:2655/5535 train_time:645248ms step_avg:243.03ms step:2656/5535 train_time:645498ms step_avg:243.03ms step:2657/5535 train_time:645743ms step_avg:243.03ms step:2658/5535 train_time:645991ms step_avg:243.04ms step:2659/5535 train_time:646237ms step_avg:243.04ms step:2660/5535 train_time:646485ms step_avg:243.04ms step:2661/5535 train_time:646759ms step_avg:243.05ms step:2662/5535 train_time:646984ms step_avg:243.04ms step:2663/5535 train_time:647231ms step_avg:243.05ms step:2664/5535 train_time:647478ms step_avg:243.05ms step:2665/5535 train_time:647724ms step_avg:243.05ms step:2666/5535 train_time:647970ms step_avg:243.05ms step:2667/5535 train_time:648215ms step_avg:243.05ms step:2668/5535 train_time:648469ms step_avg:243.05ms step:2669/5535 train_time:648715ms step_avg:243.06ms step:2670/5535 train_time:648967ms step_avg:243.06ms step:2671/5535 train_time:649217ms step_avg:243.06ms step:2672/5535 train_time:649468ms step_avg:243.06ms step:2673/5535 train_time:649717ms step_avg:243.07ms step:2674/5535 train_time:649964ms step_avg:243.07ms step:2675/5535 train_time:650211ms step_avg:243.07ms step:2676/5535 train_time:650460ms step_avg:243.07ms step:2677/5535 train_time:650712ms step_avg:243.07ms step:2678/5535 train_time:650957ms step_avg:243.08ms step:2679/5535 train_time:651209ms step_avg:243.08ms step:2680/5535 train_time:651457ms step_avg:243.08ms step:2681/5535 train_time:651705ms step_avg:243.08ms step:2682/5535 train_time:651951ms step_avg:243.08ms step:2683/5535 train_time:652198ms step_avg:243.09ms step:2684/5535 train_time:652445ms step_avg:243.09ms step:2685/5535 train_time:652695ms step_avg:243.09ms step:2686/5535 train_time:652944ms step_avg:243.09ms step:2687/5535 train_time:653194ms step_avg:243.09ms step:2688/5535 train_time:653442ms step_avg:243.10ms step:2689/5535 train_time:653720ms step_avg:243.11ms step:2690/5535 train_time:653945ms step_avg:243.10ms step:2691/5535 train_time:654191ms step_avg:243.10ms step:2692/5535 train_time:654442ms step_avg:243.11ms step:2693/5535 train_time:654689ms step_avg:243.11ms step:2694/5535 train_time:654940ms step_avg:243.11ms step:2695/5535 train_time:655186ms step_avg:243.11ms step:2696/5535 train_time:655438ms step_avg:243.12ms step:2697/5535 train_time:655688ms step_avg:243.12ms step:2698/5535 train_time:655938ms step_avg:243.12ms step:2699/5535 train_time:656189ms step_avg:243.12ms step:2700/5535 train_time:656442ms step_avg:243.13ms step:2701/5535 train_time:656688ms step_avg:243.13ms step:2702/5535 train_time:656937ms step_avg:243.13ms step:2703/5535 train_time:657186ms step_avg:243.13ms step:2704/5535 train_time:657435ms step_avg:243.13ms step:2705/5535 train_time:657682ms step_avg:243.14ms step:2706/5535 train_time:657929ms step_avg:243.14ms step:2707/5535 train_time:658175ms step_avg:243.14ms step:2708/5535 train_time:658422ms step_avg:243.14ms step:2709/5535 train_time:658672ms step_avg:243.14ms step:2710/5535 train_time:658917ms step_avg:243.14ms step:2711/5535 train_time:659162ms step_avg:243.14ms step:2712/5535 train_time:659414ms step_avg:243.15ms step:2713/5535 train_time:659662ms step_avg:243.15ms step:2714/5535 train_time:659907ms step_avg:243.15ms step:2715/5535 train_time:660153ms step_avg:243.15ms step:2716/5535 train_time:660399ms step_avg:243.15ms step:2717/5535 train_time:660675ms step_avg:243.16ms step:2718/5535 train_time:660898ms step_avg:243.16ms step:2719/5535 train_time:661145ms step_avg:243.16ms step:2720/5535 train_time:661391ms step_avg:243.16ms step:2721/5535 train_time:661640ms step_avg:243.16ms step:2722/5535 train_time:661890ms step_avg:243.16ms step:2723/5535 train_time:662138ms step_avg:243.16ms step:2724/5535 train_time:662386ms step_avg:243.17ms step:2725/5535 train_time:662634ms step_avg:243.17ms step:2726/5535 train_time:662888ms step_avg:243.17ms step:2727/5535 train_time:663137ms step_avg:243.17ms step:2728/5535 train_time:663381ms step_avg:243.17ms step:2729/5535 train_time:663629ms step_avg:243.18ms step:2730/5535 train_time:663880ms step_avg:243.18ms step:2731/5535 train_time:664128ms step_avg:243.18ms step:2732/5535 train_time:664374ms step_avg:243.18ms step:2733/5535 train_time:664618ms step_avg:243.18ms step:2734/5535 train_time:664878ms step_avg:243.19ms step:2735/5535 train_time:665123ms step_avg:243.19ms step:2736/5535 train_time:665372ms step_avg:243.19ms step:2737/5535 train_time:665623ms step_avg:243.19ms step:2738/5535 train_time:665872ms step_avg:243.20ms step:2739/5535 train_time:666118ms step_avg:243.20ms step:2740/5535 train_time:666364ms step_avg:243.20ms step:2741/5535 train_time:666609ms step_avg:243.20ms step:2742/5535 train_time:666857ms step_avg:243.20ms step:2743/5535 train_time:667105ms step_avg:243.20ms step:2744/5535 train_time:667349ms step_avg:243.20ms step:2745/5535 train_time:667629ms step_avg:243.22ms step:2746/5535 train_time:667855ms step_avg:243.21ms step:2747/5535 train_time:668103ms step_avg:243.21ms step:2748/5535 train_time:668350ms step_avg:243.21ms step:2749/5535 train_time:668600ms step_avg:243.22ms step:2750/5535 train_time:668848ms step_avg:243.22ms step:2750/5535 val_loss:3.113449 train_time:669098ms step_avg:243.31ms step:2751/5535 train_time:669119ms step_avg:243.23ms step:2752/5535 train_time:669343ms step_avg:243.22ms step:2753/5535 train_time:669601ms step_avg:243.23ms step:2754/5535 train_time:669850ms step_avg:243.23ms step:2755/5535 train_time:670096ms step_avg:243.23ms step:2756/5535 train_time:670342ms step_avg:243.23ms step:2757/5535 train_time:670592ms step_avg:243.23ms step:2758/5535 train_time:670841ms step_avg:243.23ms step:2759/5535 train_time:671085ms step_avg:243.23ms step:2760/5535 train_time:671329ms step_avg:243.24ms step:2761/5535 train_time:671583ms step_avg:243.24ms step:2762/5535 train_time:671832ms step_avg:243.24ms step:2763/5535 train_time:672080ms step_avg:243.24ms step:2764/5535 train_time:672327ms step_avg:243.24ms step:2765/5535 train_time:672573ms step_avg:243.25ms step:2766/5535 train_time:672822ms step_avg:243.25ms step:2767/5535 train_time:673070ms step_avg:243.25ms step:2768/5535 train_time:673316ms step_avg:243.25ms step:2769/5535 train_time:673569ms step_avg:243.25ms step:2770/5535 train_time:673816ms step_avg:243.25ms step:2771/5535 train_time:674064ms step_avg:243.26ms step:2772/5535 train_time:674313ms step_avg:243.26ms step:2773/5535 train_time:674589ms step_avg:243.27ms step:2774/5535 train_time:674812ms step_avg:243.26ms step:2775/5535 train_time:675059ms step_avg:243.26ms step:2776/5535 train_time:675305ms step_avg:243.27ms step:2777/5535 train_time:675555ms step_avg:243.27ms step:2778/5535 train_time:675802ms step_avg:243.27ms step:2779/5535 train_time:676051ms step_avg:243.27ms step:2780/5535 train_time:676298ms step_avg:243.27ms step:2781/5535 train_time:676545ms step_avg:243.27ms step:2782/5535 train_time:676793ms step_avg:243.28ms step:2783/5535 train_time:677046ms step_avg:243.28ms step:2784/5535 train_time:677290ms step_avg:243.28ms step:2785/5535 train_time:677537ms step_avg:243.28ms step:2786/5535 train_time:677786ms step_avg:243.28ms step:2787/5535 train_time:678032ms step_avg:243.28ms step:2788/5535 train_time:678279ms step_avg:243.29ms step:2789/5535 train_time:678524ms step_avg:243.29ms step:2790/5535 train_time:678772ms step_avg:243.29ms step:2791/5535 train_time:679022ms step_avg:243.29ms step:2792/5535 train_time:679275ms step_avg:243.29ms step:2793/5535 train_time:679520ms step_avg:243.29ms step:2794/5535 train_time:679766ms step_avg:243.29ms step:2795/5535 train_time:680013ms step_avg:243.30ms step:2796/5535 train_time:680266ms step_avg:243.30ms step:2797/5535 train_time:680512ms step_avg:243.30ms step:2798/5535 train_time:680761ms step_avg:243.30ms step:2799/5535 train_time:681009ms step_avg:243.30ms step:2800/5535 train_time:681256ms step_avg:243.31ms step:2801/5535 train_time:681535ms step_avg:243.32ms step:2802/5535 train_time:681759ms step_avg:243.31ms step:2803/5535 train_time:682003ms step_avg:243.31ms step:2804/5535 train_time:682253ms step_avg:243.31ms step:2805/5535 train_time:682502ms step_avg:243.32ms step:2806/5535 train_time:682750ms step_avg:243.32ms step:2807/5535 train_time:682997ms step_avg:243.32ms step:2808/5535 train_time:683243ms step_avg:243.32ms step:2809/5535 train_time:683489ms step_avg:243.32ms step:2810/5535 train_time:683740ms step_avg:243.32ms step:2811/5535 train_time:683988ms step_avg:243.33ms step:2812/5535 train_time:684240ms step_avg:243.33ms step:2813/5535 train_time:684487ms step_avg:243.33ms step:2814/5535 train_time:684734ms step_avg:243.33ms step:2815/5535 train_time:684981ms step_avg:243.33ms step:2816/5535 train_time:685238ms step_avg:243.34ms step:2817/5535 train_time:685494ms step_avg:243.34ms step:2818/5535 train_time:685738ms step_avg:243.34ms step:2819/5535 train_time:685984ms step_avg:243.34ms step:2820/5535 train_time:686236ms step_avg:243.35ms step:2821/5535 train_time:686487ms step_avg:243.35ms step:2822/5535 train_time:686732ms step_avg:243.35ms step:2823/5535 train_time:686980ms step_avg:243.35ms step:2824/5535 train_time:687231ms step_avg:243.35ms step:2825/5535 train_time:687483ms step_avg:243.36ms step:2826/5535 train_time:687733ms step_avg:243.36ms step:2827/5535 train_time:687977ms step_avg:243.36ms step:2828/5535 train_time:688228ms step_avg:243.36ms step:2829/5535 train_time:688507ms step_avg:243.37ms step:2830/5535 train_time:688732ms step_avg:243.37ms step:2831/5535 train_time:688982ms step_avg:243.37ms step:2832/5535 train_time:689237ms step_avg:243.37ms step:2833/5535 train_time:689483ms step_avg:243.38ms step:2834/5535 train_time:689733ms step_avg:243.38ms step:2835/5535 train_time:689980ms step_avg:243.38ms step:2836/5535 train_time:690228ms step_avg:243.38ms step:2837/5535 train_time:690474ms step_avg:243.38ms step:2838/5535 train_time:690721ms step_avg:243.38ms step:2839/5535 train_time:690970ms step_avg:243.39ms step:2840/5535 train_time:691224ms step_avg:243.39ms step:2841/5535 train_time:691469ms step_avg:243.39ms step:2842/5535 train_time:691720ms step_avg:243.39ms step:2843/5535 train_time:691966ms step_avg:243.39ms step:2844/5535 train_time:692215ms step_avg:243.39ms step:2845/5535 train_time:692466ms step_avg:243.40ms step:2846/5535 train_time:692717ms step_avg:243.40ms step:2847/5535 train_time:692961ms step_avg:243.40ms step:2848/5535 train_time:693212ms step_avg:243.40ms step:2849/5535 train_time:693461ms step_avg:243.41ms step:2850/5535 train_time:693708ms step_avg:243.41ms step:2851/5535 train_time:693960ms step_avg:243.41ms step:2852/5535 train_time:694210ms step_avg:243.41ms step:2853/5535 train_time:694465ms step_avg:243.42ms step:2854/5535 train_time:694711ms step_avg:243.42ms step:2855/5535 train_time:694961ms step_avg:243.42ms step:2856/5535 train_time:695212ms step_avg:243.42ms step:2857/5535 train_time:695488ms step_avg:243.43ms step:2858/5535 train_time:695713ms step_avg:243.43ms step:2859/5535 train_time:695965ms step_avg:243.43ms step:2860/5535 train_time:696211ms step_avg:243.43ms step:2861/5535 train_time:696458ms step_avg:243.43ms step:2862/5535 train_time:696710ms step_avg:243.43ms step:2863/5535 train_time:696958ms step_avg:243.44ms step:2864/5535 train_time:697207ms step_avg:243.44ms step:2865/5535 train_time:697454ms step_avg:243.44ms step:2866/5535 train_time:697701ms step_avg:243.44ms step:2867/5535 train_time:697957ms step_avg:243.44ms step:2868/5535 train_time:698214ms step_avg:243.45ms step:2869/5535 train_time:698459ms step_avg:243.45ms step:2870/5535 train_time:698710ms step_avg:243.45ms step:2871/5535 train_time:698962ms step_avg:243.46ms step:2872/5535 train_time:699209ms step_avg:243.46ms step:2873/5535 train_time:699461ms step_avg:243.46ms step:2874/5535 train_time:699709ms step_avg:243.46ms step:2875/5535 train_time:699955ms step_avg:243.46ms step:2875/5535 val_loss:3.104164 train_time:700204ms step_avg:243.55ms step:2876/5535 train_time:700225ms step_avg:243.47ms step:2877/5535 train_time:700451ms step_avg:243.47ms step:2878/5535 train_time:700707ms step_avg:243.47ms step:2879/5535 train_time:700953ms step_avg:243.47ms step:2880/5535 train_time:701198ms step_avg:243.47ms step:2881/5535 train_time:701447ms step_avg:243.47ms step:2882/5535 train_time:701699ms step_avg:243.48ms step:2883/5535 train_time:701945ms step_avg:243.48ms step:2884/5535 train_time:702194ms step_avg:243.48ms step:2885/5535 train_time:702468ms step_avg:243.49ms step:2886/5535 train_time:702693ms step_avg:243.48ms step:2887/5535 train_time:702940ms step_avg:243.48ms step:2888/5535 train_time:703202ms step_avg:243.49ms step:2889/5535 train_time:703451ms step_avg:243.49ms step:2890/5535 train_time:703702ms step_avg:243.50ms step:2891/5535 train_time:703951ms step_avg:243.50ms step:2892/5535 train_time:704199ms step_avg:243.50ms step:2893/5535 train_time:704449ms step_avg:243.50ms step:2894/5535 train_time:704699ms step_avg:243.50ms step:2895/5535 train_time:704945ms step_avg:243.50ms step:2896/5535 train_time:705194ms step_avg:243.51ms step:2897/5535 train_time:705444ms step_avg:243.51ms step:2898/5535 train_time:705694ms step_avg:243.51ms step:2899/5535 train_time:705953ms step_avg:243.52ms step:2900/5535 train_time:706200ms step_avg:243.52ms step:2901/5535 train_time:706449ms step_avg:243.52ms step:2902/5535 train_time:706705ms step_avg:243.52ms step:2903/5535 train_time:706951ms step_avg:243.52ms step:2904/5535 train_time:707202ms step_avg:243.53ms step:2905/5535 train_time:707448ms step_avg:243.53ms step:2906/5535 train_time:707697ms step_avg:243.53ms step:2907/5535 train_time:707947ms step_avg:243.53ms step:2908/5535 train_time:708193ms step_avg:243.53ms step:2909/5535 train_time:708443ms step_avg:243.53ms step:2910/5535 train_time:708690ms step_avg:243.54ms step:2911/5535 train_time:708939ms step_avg:243.54ms step:2912/5535 train_time:709192ms step_avg:243.54ms step:2913/5535 train_time:709466ms step_avg:243.55ms step:2914/5535 train_time:709691ms step_avg:243.55ms step:2915/5535 train_time:709938ms step_avg:243.55ms step:2916/5535 train_time:710189ms step_avg:243.55ms step:2917/5535 train_time:710435ms step_avg:243.55ms step:2918/5535 train_time:710681ms step_avg:243.55ms step:2919/5535 train_time:710932ms step_avg:243.55ms step:2920/5535 train_time:711185ms step_avg:243.56ms step:2921/5535 train_time:711434ms step_avg:243.56ms step:2922/5535 train_time:711681ms step_avg:243.56ms step:2923/5535 train_time:711932ms step_avg:243.56ms step:2924/5535 train_time:712180ms step_avg:243.56ms step:2925/5535 train_time:712428ms step_avg:243.57ms step:2926/5535 train_time:712676ms step_avg:243.57ms step:2927/5535 train_time:712933ms step_avg:243.57ms step:2928/5535 train_time:713181ms step_avg:243.57ms step:2929/5535 train_time:713428ms step_avg:243.57ms step:2930/5535 train_time:713675ms step_avg:243.58ms step:2931/5535 train_time:713924ms step_avg:243.58ms step:2932/5535 train_time:714172ms step_avg:243.58ms step:2933/5535 train_time:714418ms step_avg:243.58ms step:2934/5535 train_time:714671ms step_avg:243.58ms step:2935/5535 train_time:714916ms step_avg:243.58ms step:2936/5535 train_time:715165ms step_avg:243.58ms step:2937/5535 train_time:715412ms step_avg:243.59ms step:2938/5535 train_time:715661ms step_avg:243.59ms step:2939/5535 train_time:715905ms step_avg:243.59ms step:2940/5535 train_time:716152ms step_avg:243.59ms step:2941/5535 train_time:716428ms step_avg:243.60ms step:2942/5535 train_time:716654ms step_avg:243.59ms step:2943/5535 train_time:716902ms step_avg:243.60ms step:2944/5535 train_time:717151ms step_avg:243.60ms step:2945/5535 train_time:717398ms step_avg:243.60ms step:2946/5535 train_time:717646ms step_avg:243.60ms step:2947/5535 train_time:717893ms step_avg:243.60ms step:2948/5535 train_time:718140ms step_avg:243.60ms step:2949/5535 train_time:718392ms step_avg:243.61ms step:2950/5535 train_time:718641ms step_avg:243.61ms step:2951/5535 train_time:718894ms step_avg:243.61ms step:2952/5535 train_time:719141ms step_avg:243.61ms step:2953/5535 train_time:719390ms step_avg:243.61ms step:2954/5535 train_time:719634ms step_avg:243.61ms step:2955/5535 train_time:719879ms step_avg:243.61ms step:2956/5535 train_time:720131ms step_avg:243.62ms step:2957/5535 train_time:720382ms step_avg:243.62ms step:2958/5535 train_time:720629ms step_avg:243.62ms step:2959/5535 train_time:720875ms step_avg:243.62ms step:2960/5535 train_time:721127ms step_avg:243.62ms step:2961/5535 train_time:721377ms step_avg:243.63ms step:2962/5535 train_time:721628ms step_avg:243.63ms step:2963/5535 train_time:721871ms step_avg:243.63ms step:2964/5535 train_time:722120ms step_avg:243.63ms step:2965/5535 train_time:722367ms step_avg:243.63ms step:2966/5535 train_time:722617ms step_avg:243.63ms step:2967/5535 train_time:722867ms step_avg:243.64ms step:2968/5535 train_time:723113ms step_avg:243.64ms step:2969/5535 train_time:723387ms step_avg:243.65ms step:2970/5535 train_time:723614ms step_avg:243.64ms step:2971/5535 train_time:723866ms step_avg:243.64ms step:2972/5535 train_time:724116ms step_avg:243.65ms step:2973/5535 train_time:724364ms step_avg:243.65ms step:2974/5535 train_time:724613ms step_avg:243.65ms step:2975/5535 train_time:724863ms step_avg:243.65ms step:2976/5535 train_time:725111ms step_avg:243.65ms step:2977/5535 train_time:725356ms step_avg:243.65ms step:2978/5535 train_time:725611ms step_avg:243.66ms step:2979/5535 train_time:725859ms step_avg:243.66ms step:2980/5535 train_time:726104ms step_avg:243.66ms step:2981/5535 train_time:726351ms step_avg:243.66ms step:2982/5535 train_time:726601ms step_avg:243.66ms step:2983/5535 train_time:726849ms step_avg:243.66ms step:2984/5535 train_time:727093ms step_avg:243.66ms step:2985/5535 train_time:727339ms step_avg:243.66ms step:2986/5535 train_time:727589ms step_avg:243.67ms step:2987/5535 train_time:727836ms step_avg:243.67ms step:2988/5535 train_time:728081ms step_avg:243.67ms step:2989/5535 train_time:728331ms step_avg:243.67ms step:2990/5535 train_time:728579ms step_avg:243.67ms step:2991/5535 train_time:728839ms step_avg:243.68ms step:2992/5535 train_time:729090ms step_avg:243.68ms step:2993/5535 train_time:729342ms step_avg:243.68ms step:2994/5535 train_time:729594ms step_avg:243.69ms step:2995/5535 train_time:729840ms step_avg:243.69ms step:2996/5535 train_time:730090ms step_avg:243.69ms step:2997/5535 train_time:730368ms step_avg:243.70ms step:2998/5535 train_time:730592ms step_avg:243.69ms step:2999/5535 train_time:730844ms step_avg:243.70ms step:3000/5535 train_time:731095ms step_avg:243.70ms step:3000/5535 val_loss:3.093141 train_time:731342ms step_avg:243.78ms step:3001/5535 train_time:731362ms step_avg:243.71ms step:3002/5535 train_time:731591ms step_avg:243.70ms step:3003/5535 train_time:731848ms step_avg:243.71ms step:3004/5535 train_time:732092ms step_avg:243.71ms step:3005/5535 train_time:732344ms step_avg:243.71ms step:3006/5535 train_time:732592ms step_avg:243.71ms step:3007/5535 train_time:732845ms step_avg:243.71ms step:3008/5535 train_time:733093ms step_avg:243.71ms step:3009/5535 train_time:733340ms step_avg:243.72ms step:3010/5535 train_time:733587ms step_avg:243.72ms step:3011/5535 train_time:733843ms step_avg:243.72ms step:3012/5535 train_time:734090ms step_avg:243.72ms step:3013/5535 train_time:734335ms step_avg:243.72ms step:3014/5535 train_time:734581ms step_avg:243.72ms step:3015/5535 train_time:734831ms step_avg:243.72ms step:3016/5535 train_time:735081ms step_avg:243.73ms step:3017/5535 train_time:735329ms step_avg:243.73ms step:3018/5535 train_time:735578ms step_avg:243.73ms step:3019/5535 train_time:735828ms step_avg:243.73ms step:3020/5535 train_time:736083ms step_avg:243.74ms step:3021/5535 train_time:736332ms step_avg:243.74ms step:3022/5535 train_time:736579ms step_avg:243.74ms step:3023/5535 train_time:736825ms step_avg:243.74ms step:3024/5535 train_time:737073ms step_avg:243.74ms step:3025/5535 train_time:737350ms step_avg:243.75ms step:3026/5535 train_time:737572ms step_avg:243.74ms step:3027/5535 train_time:737820ms step_avg:243.75ms step:3028/5535 train_time:738066ms step_avg:243.75ms step:3029/5535 train_time:738320ms step_avg:243.75ms step:3030/5535 train_time:738564ms step_avg:243.75ms step:3031/5535 train_time:738812ms step_avg:243.75ms step:3032/5535 train_time:739061ms step_avg:243.75ms step:3033/5535 train_time:739313ms step_avg:243.76ms step:3034/5535 train_time:739563ms step_avg:243.76ms step:3035/5535 train_time:739808ms step_avg:243.76ms step:3036/5535 train_time:740054ms step_avg:243.76ms step:3037/5535 train_time:740302ms step_avg:243.76ms step:3038/5535 train_time:740549ms step_avg:243.76ms step:3039/5535 train_time:740797ms step_avg:243.76ms step:3040/5535 train_time:741042ms step_avg:243.76ms step:3041/5535 train_time:741291ms step_avg:243.77ms step:3042/5535 train_time:741535ms step_avg:243.77ms step:3043/5535 train_time:741785ms step_avg:243.77ms step:3044/5535 train_time:742033ms step_avg:243.77ms step:3045/5535 train_time:742282ms step_avg:243.77ms step:3046/5535 train_time:742533ms step_avg:243.77ms step:3047/5535 train_time:742781ms step_avg:243.77ms step:3048/5535 train_time:743030ms step_avg:243.78ms step:3049/5535 train_time:743278ms step_avg:243.78ms step:3050/5535 train_time:743530ms step_avg:243.78ms step:3051/5535 train_time:743783ms step_avg:243.78ms step:3052/5535 train_time:744030ms step_avg:243.78ms step:3053/5535 train_time:744304ms step_avg:243.79ms step:3054/5535 train_time:744529ms step_avg:243.79ms step:3055/5535 train_time:744789ms step_avg:243.79ms step:3056/5535 train_time:745036ms step_avg:243.79ms step:3057/5535 train_time:745288ms step_avg:243.80ms step:3058/5535 train_time:745534ms step_avg:243.80ms step:3059/5535 train_time:745784ms step_avg:243.80ms step:3060/5535 train_time:746031ms step_avg:243.80ms step:3061/5535 train_time:746278ms step_avg:243.80ms step:3062/5535 train_time:746526ms step_avg:243.80ms step:3063/5535 train_time:746775ms step_avg:243.81ms step:3064/5535 train_time:747021ms step_avg:243.81ms step:3065/5535 train_time:747267ms step_avg:243.81ms step:3066/5535 train_time:747516ms step_avg:243.81ms step:3067/5535 train_time:747762ms step_avg:243.81ms step:3068/5535 train_time:748014ms step_avg:243.81ms step:3069/5535 train_time:748264ms step_avg:243.81ms step:3070/5535 train_time:748514ms step_avg:243.82ms step:3071/5535 train_time:748760ms step_avg:243.82ms step:3072/5535 train_time:749016ms step_avg:243.82ms step:3073/5535 train_time:749261ms step_avg:243.82ms step:3074/5535 train_time:749507ms step_avg:243.82ms step:3075/5535 train_time:749757ms step_avg:243.82ms step:3076/5535 train_time:750003ms step_avg:243.82ms step:3077/5535 train_time:750252ms step_avg:243.83ms step:3078/5535 train_time:750498ms step_avg:243.83ms step:3079/5535 train_time:750747ms step_avg:243.83ms step:3080/5535 train_time:750995ms step_avg:243.83ms step:3081/5535 train_time:751273ms step_avg:243.84ms step:3082/5535 train_time:751503ms step_avg:243.84ms step:3083/5535 train_time:751753ms step_avg:243.84ms step:3084/5535 train_time:752006ms step_avg:243.84ms step:3085/5535 train_time:752257ms step_avg:243.84ms step:3086/5535 train_time:752517ms step_avg:243.85ms step:3087/5535 train_time:752763ms step_avg:243.85ms step:3088/5535 train_time:753012ms step_avg:243.85ms step:3089/5535 train_time:753261ms step_avg:243.85ms step:3090/5535 train_time:753512ms step_avg:243.85ms step:3091/5535 train_time:753763ms step_avg:243.86ms step:3092/5535 train_time:754014ms step_avg:243.86ms step:3093/5535 train_time:754260ms step_avg:243.86ms step:3094/5535 train_time:754507ms step_avg:243.86ms step:3095/5535 train_time:754767ms step_avg:243.87ms step:3096/5535 train_time:755011ms step_avg:243.87ms step:3097/5535 train_time:755260ms step_avg:243.87ms step:3098/5535 train_time:755508ms step_avg:243.87ms step:3099/5535 train_time:755757ms step_avg:243.87ms step:3100/5535 train_time:756006ms step_avg:243.87ms step:3101/5535 train_time:756254ms step_avg:243.87ms step:3102/5535 train_time:756500ms step_avg:243.87ms step:3103/5535 train_time:756754ms step_avg:243.88ms step:3104/5535 train_time:757000ms step_avg:243.88ms step:3105/5535 train_time:757247ms step_avg:243.88ms step:3106/5535 train_time:757495ms step_avg:243.88ms step:3107/5535 train_time:757746ms step_avg:243.88ms step:3108/5535 train_time:758006ms step_avg:243.89ms step:3109/5535 train_time:758284ms step_avg:243.90ms step:3110/5535 train_time:758514ms step_avg:243.90ms step:3111/5535 train_time:758761ms step_avg:243.90ms step:3112/5535 train_time:759008ms step_avg:243.90ms step:3113/5535 train_time:759257ms step_avg:243.90ms step:3114/5535 train_time:759503ms step_avg:243.90ms step:3115/5535 train_time:759749ms step_avg:243.90ms step:3116/5535 train_time:759996ms step_avg:243.90ms step:3117/5535 train_time:760250ms step_avg:243.90ms step:3118/5535 train_time:760500ms step_avg:243.91ms step:3119/5535 train_time:760749ms step_avg:243.91ms step:3120/5535 train_time:760997ms step_avg:243.91ms step:3121/5535 train_time:761246ms step_avg:243.91ms step:3122/5535 train_time:761491ms step_avg:243.91ms step:3123/5535 train_time:761740ms step_avg:243.91ms step:3124/5535 train_time:761987ms step_avg:243.91ms step:3125/5535 train_time:762238ms step_avg:243.92ms step:3125/5535 val_loss:3.081749 train_time:762488ms step_avg:244.00ms step:3126/5535 train_time:762509ms step_avg:243.92ms step:3127/5535 train_time:762733ms step_avg:243.92ms step:3128/5535 train_time:762990ms step_avg:243.92ms step:3129/5535 train_time:763239ms step_avg:243.92ms step:3130/5535 train_time:763481ms step_avg:243.92ms step:3131/5535 train_time:763726ms step_avg:243.92ms step:3132/5535 train_time:763979ms step_avg:243.93ms step:3133/5535 train_time:764226ms step_avg:243.93ms step:3134/5535 train_time:764479ms step_avg:243.93ms step:3135/5535 train_time:764724ms step_avg:243.93ms step:3136/5535 train_time:764975ms step_avg:243.93ms step:3137/5535 train_time:765253ms step_avg:243.94ms step:3138/5535 train_time:765477ms step_avg:243.94ms step:3139/5535 train_time:765725ms step_avg:243.94ms step:3140/5535 train_time:765973ms step_avg:243.94ms step:3141/5535 train_time:766222ms step_avg:243.94ms step:3142/5535 train_time:766471ms step_avg:243.94ms step:3143/5535 train_time:766717ms step_avg:243.94ms step:3144/5535 train_time:766971ms step_avg:243.95ms step:3145/5535 train_time:767221ms step_avg:243.95ms step:3146/5535 train_time:767471ms step_avg:243.95ms step:3147/5535 train_time:767717ms step_avg:243.95ms step:3148/5535 train_time:767963ms step_avg:243.95ms step:3149/5535 train_time:768211ms step_avg:243.95ms step:3150/5535 train_time:768459ms step_avg:243.96ms step:3151/5535 train_time:768703ms step_avg:243.96ms step:3152/5535 train_time:768958ms step_avg:243.96ms step:3153/5535 train_time:769206ms step_avg:243.96ms step:3154/5535 train_time:769452ms step_avg:243.96ms step:3155/5535 train_time:769698ms step_avg:243.96ms step:3156/5535 train_time:769945ms step_avg:243.96ms step:3157/5535 train_time:770198ms step_avg:243.97ms step:3158/5535 train_time:770445ms step_avg:243.97ms step:3159/5535 train_time:770692ms step_avg:243.97ms step:3160/5535 train_time:770940ms step_avg:243.97ms step:3161/5535 train_time:771187ms step_avg:243.97ms step:3162/5535 train_time:771434ms step_avg:243.97ms step:3163/5535 train_time:771682ms step_avg:243.97ms step:3164/5535 train_time:771939ms step_avg:243.98ms step:3165/5535 train_time:772220ms step_avg:243.99ms step:3166/5535 train_time:772442ms step_avg:243.98ms step:3167/5535 train_time:772690ms step_avg:243.98ms step:3168/5535 train_time:772935ms step_avg:243.98ms step:3169/5535 train_time:773183ms step_avg:243.98ms step:3170/5535 train_time:773429ms step_avg:243.98ms step:3171/5535 train_time:773677ms step_avg:243.99ms step:3172/5535 train_time:773929ms step_avg:243.99ms step:3173/5535 train_time:774176ms step_avg:243.99ms step:3174/5535 train_time:774426ms step_avg:243.99ms step:3175/5535 train_time:774674ms step_avg:243.99ms step:3176/5535 train_time:774924ms step_avg:243.99ms step:3177/5535 train_time:775173ms step_avg:244.00ms step:3178/5535 train_time:775419ms step_avg:244.00ms step:3179/5535 train_time:775674ms step_avg:244.00ms step:3180/5535 train_time:775920ms step_avg:244.00ms step:3181/5535 train_time:776168ms step_avg:244.00ms step:3182/5535 train_time:776412ms step_avg:244.00ms step:3183/5535 train_time:776659ms step_avg:244.00ms step:3184/5535 train_time:776909ms step_avg:244.00ms step:3185/5535 train_time:777155ms step_avg:244.00ms step:3186/5535 train_time:777405ms step_avg:244.01ms step:3187/5535 train_time:777650ms step_avg:244.01ms step:3188/5535 train_time:777899ms step_avg:244.01ms step:3189/5535 train_time:778147ms step_avg:244.01ms step:3190/5535 train_time:778394ms step_avg:244.01ms step:3191/5535 train_time:778644ms step_avg:244.01ms step:3192/5535 train_time:778894ms step_avg:244.01ms step:3193/5535 train_time:779180ms step_avg:244.03ms step:3194/5535 train_time:779406ms step_avg:244.02ms step:3195/5535 train_time:779661ms step_avg:244.03ms step:3196/5535 train_time:779906ms step_avg:244.03ms step:3197/5535 train_time:780158ms step_avg:244.03ms step:3198/5535 train_time:780405ms step_avg:244.03ms step:3199/5535 train_time:780656ms step_avg:244.03ms step:3200/5535 train_time:780903ms step_avg:244.03ms step:3201/5535 train_time:781150ms step_avg:244.03ms step:3202/5535 train_time:781406ms step_avg:244.04ms step:3203/5535 train_time:781653ms step_avg:244.04ms step:3204/5535 train_time:781902ms step_avg:244.04ms step:3205/5535 train_time:782149ms step_avg:244.04ms step:3206/5535 train_time:782400ms step_avg:244.04ms step:3207/5535 train_time:782652ms step_avg:244.04ms step:3208/5535 train_time:782897ms step_avg:244.05ms step:3209/5535 train_time:783144ms step_avg:244.05ms step:3210/5535 train_time:783394ms step_avg:244.05ms step:3211/5535 train_time:783644ms step_avg:244.05ms step:3212/5535 train_time:783891ms step_avg:244.05ms step:3213/5535 train_time:784139ms step_avg:244.05ms step:3214/5535 train_time:784386ms step_avg:244.05ms step:3215/5535 train_time:784633ms step_avg:244.05ms step:3216/5535 train_time:784880ms step_avg:244.05ms step:3217/5535 train_time:785128ms step_avg:244.06ms step:3218/5535 train_time:785374ms step_avg:244.06ms step:3219/5535 train_time:785621ms step_avg:244.06ms step:3220/5535 train_time:785868ms step_avg:244.06ms step:3221/5535 train_time:786145ms step_avg:244.07ms step:3222/5535 train_time:786371ms step_avg:244.06ms step:3223/5535 train_time:786625ms step_avg:244.07ms step:3224/5535 train_time:786875ms step_avg:244.07ms step:3225/5535 train_time:787124ms step_avg:244.07ms step:3226/5535 train_time:787374ms step_avg:244.07ms step:3227/5535 train_time:787627ms step_avg:244.07ms step:3228/5535 train_time:787877ms step_avg:244.08ms step:3229/5535 train_time:788127ms step_avg:244.08ms step:3230/5535 train_time:788383ms step_avg:244.08ms step:3231/5535 train_time:788637ms step_avg:244.08ms step:3232/5535 train_time:788892ms step_avg:244.09ms step:3233/5535 train_time:789136ms step_avg:244.09ms step:3234/5535 train_time:789384ms step_avg:244.09ms step:3235/5535 train_time:789636ms step_avg:244.09ms step:3236/5535 train_time:789885ms step_avg:244.09ms step:3237/5535 train_time:790129ms step_avg:244.09ms step:3238/5535 train_time:790382ms step_avg:244.10ms step:3239/5535 train_time:790631ms step_avg:244.10ms step:3240/5535 train_time:790883ms step_avg:244.10ms step:3241/5535 train_time:791125ms step_avg:244.10ms step:3242/5535 train_time:791376ms step_avg:244.10ms step:3243/5535 train_time:791625ms step_avg:244.10ms step:3244/5535 train_time:791872ms step_avg:244.10ms step:3245/5535 train_time:792119ms step_avg:244.10ms step:3246/5535 train_time:792368ms step_avg:244.11ms step:3247/5535 train_time:792616ms step_avg:244.11ms step:3248/5535 train_time:792864ms step_avg:244.11ms step:3249/5535 train_time:793147ms step_avg:244.12ms step:3250/5535 train_time:793369ms step_avg:244.11ms step:3250/5535 val_loss:3.070642 train_time:793618ms step_avg:244.19ms step:3251/5535 train_time:793639ms step_avg:244.12ms step:3252/5535 train_time:793867ms step_avg:244.12ms step:3253/5535 train_time:794128ms step_avg:244.12ms step:3254/5535 train_time:794370ms step_avg:244.12ms step:3255/5535 train_time:794615ms step_avg:244.12ms step:3256/5535 train_time:794861ms step_avg:244.12ms step:3257/5535 train_time:795112ms step_avg:244.12ms step:3258/5535 train_time:795360ms step_avg:244.13ms step:3259/5535 train_time:795606ms step_avg:244.13ms step:3260/5535 train_time:795853ms step_avg:244.13ms step:3261/5535 train_time:796100ms step_avg:244.13ms step:3262/5535 train_time:796363ms step_avg:244.13ms step:3263/5535 train_time:796606ms step_avg:244.13ms step:3264/5535 train_time:796857ms step_avg:244.14ms step:3265/5535 train_time:797103ms step_avg:244.14ms step:3266/5535 train_time:797355ms step_avg:244.14ms step:3267/5535 train_time:797600ms step_avg:244.14ms step:3268/5535 train_time:797850ms step_avg:244.14ms step:3269/5535 train_time:798098ms step_avg:244.14ms step:3270/5535 train_time:798345ms step_avg:244.14ms step:3271/5535 train_time:798593ms step_avg:244.14ms step:3272/5535 train_time:798840ms step_avg:244.14ms step:3273/5535 train_time:799087ms step_avg:244.15ms step:3274/5535 train_time:799334ms step_avg:244.15ms step:3275/5535 train_time:799584ms step_avg:244.15ms step:3276/5535 train_time:799831ms step_avg:244.15ms step:3277/5535 train_time:800107ms step_avg:244.16ms step:3278/5535 train_time:800333ms step_avg:244.15ms step:3279/5535 train_time:800582ms step_avg:244.15ms step:3280/5535 train_time:800830ms step_avg:244.16ms step:3281/5535 train_time:801084ms step_avg:244.16ms step:3282/5535 train_time:801337ms step_avg:244.16ms step:3283/5535 train_time:801581ms step_avg:244.16ms step:3284/5535 train_time:801829ms step_avg:244.16ms step:3285/5535 train_time:802081ms step_avg:244.16ms step:3286/5535 train_time:802331ms step_avg:244.17ms step:3287/5535 train_time:802578ms step_avg:244.17ms step:3288/5535 train_time:802824ms step_avg:244.17ms step:3289/5535 train_time:803082ms step_avg:244.17ms step:3290/5535 train_time:803328ms step_avg:244.17ms step:3291/5535 train_time:803576ms step_avg:244.17ms step:3292/5535 train_time:803822ms step_avg:244.17ms step:3293/5535 train_time:804084ms step_avg:244.18ms step:3294/5535 train_time:804336ms step_avg:244.18ms step:3295/5535 train_time:804583ms step_avg:244.18ms step:3296/5535 train_time:804831ms step_avg:244.18ms step:3297/5535 train_time:805080ms step_avg:244.19ms step:3298/5535 train_time:805331ms step_avg:244.19ms step:3299/5535 train_time:805581ms step_avg:244.19ms step:3300/5535 train_time:805829ms step_avg:244.19ms step:3301/5535 train_time:806083ms step_avg:244.19ms step:3302/5535 train_time:806331ms step_avg:244.19ms step:3303/5535 train_time:806580ms step_avg:244.20ms step:3304/5535 train_time:806836ms step_avg:244.20ms step:3305/5535 train_time:807114ms step_avg:244.21ms step:3306/5535 train_time:807347ms step_avg:244.21ms step:3307/5535 train_time:807596ms step_avg:244.21ms step:3308/5535 train_time:807841ms step_avg:244.21ms step:3309/5535 train_time:808091ms step_avg:244.21ms step:3310/5535 train_time:808341ms step_avg:244.21ms step:3311/5535 train_time:808589ms step_avg:244.21ms step:3312/5535 train_time:808837ms step_avg:244.21ms step:3313/5535 train_time:809087ms step_avg:244.22ms step:3314/5535 train_time:809336ms step_avg:244.22ms step:3315/5535 train_time:809589ms step_avg:244.22ms step:3316/5535 train_time:809841ms step_avg:244.22ms step:3317/5535 train_time:810088ms step_avg:244.22ms step:3318/5535 train_time:810332ms step_avg:244.22ms step:3319/5535 train_time:810582ms step_avg:244.22ms step:3320/5535 train_time:810834ms step_avg:244.23ms step:3321/5535 train_time:811080ms step_avg:244.23ms step:3322/5535 train_time:811327ms step_avg:244.23ms step:3323/5535 train_time:811574ms step_avg:244.23ms step:3324/5535 train_time:811822ms step_avg:244.23ms step:3325/5535 train_time:812067ms step_avg:244.23ms step:3326/5535 train_time:812322ms step_avg:244.23ms step:3327/5535 train_time:812568ms step_avg:244.23ms step:3328/5535 train_time:812820ms step_avg:244.24ms step:3329/5535 train_time:813068ms step_avg:244.24ms step:3330/5535 train_time:813314ms step_avg:244.24ms step:3331/5535 train_time:813560ms step_avg:244.24ms step:3332/5535 train_time:813813ms step_avg:244.24ms step:3333/5535 train_time:814088ms step_avg:244.25ms step:3334/5535 train_time:814310ms step_avg:244.24ms step:3335/5535 train_time:814556ms step_avg:244.24ms step:3336/5535 train_time:814805ms step_avg:244.25ms step:3337/5535 train_time:815050ms step_avg:244.25ms step:3338/5535 train_time:815299ms step_avg:244.25ms step:3339/5535 train_time:815547ms step_avg:244.25ms step:3340/5535 train_time:815792ms step_avg:244.25ms step:3341/5535 train_time:816040ms step_avg:244.25ms step:3342/5535 train_time:816290ms step_avg:244.25ms step:3343/5535 train_time:816537ms step_avg:244.25ms step:3344/5535 train_time:816782ms step_avg:244.25ms step:3345/5535 train_time:817031ms step_avg:244.25ms step:3346/5535 train_time:817280ms step_avg:244.26ms step:3347/5535 train_time:817530ms step_avg:244.26ms step:3348/5535 train_time:817774ms step_avg:244.26ms step:3349/5535 train_time:818022ms step_avg:244.26ms step:3350/5535 train_time:818270ms step_avg:244.26ms step:3351/5535 train_time:818523ms step_avg:244.26ms step:3352/5535 train_time:818771ms step_avg:244.26ms step:3353/5535 train_time:819021ms step_avg:244.27ms step:3354/5535 train_time:819267ms step_avg:244.27ms step:3355/5535 train_time:819517ms step_avg:244.27ms step:3356/5535 train_time:819766ms step_avg:244.27ms step:3357/5535 train_time:820013ms step_avg:244.27ms step:3358/5535 train_time:820259ms step_avg:244.27ms step:3359/5535 train_time:820506ms step_avg:244.27ms step:3360/5535 train_time:820753ms step_avg:244.27ms step:3361/5535 train_time:821032ms step_avg:244.28ms step:3362/5535 train_time:821256ms step_avg:244.28ms step:3363/5535 train_time:821502ms step_avg:244.28ms step:3364/5535 train_time:821753ms step_avg:244.28ms step:3365/5535 train_time:822001ms step_avg:244.28ms step:3366/5535 train_time:822254ms step_avg:244.28ms step:3367/5535 train_time:822501ms step_avg:244.28ms step:3368/5535 train_time:822748ms step_avg:244.28ms step:3369/5535 train_time:822992ms step_avg:244.28ms step:3370/5535 train_time:823238ms step_avg:244.28ms step:3371/5535 train_time:823485ms step_avg:244.28ms step:3372/5535 train_time:823730ms step_avg:244.29ms step:3373/5535 train_time:823975ms step_avg:244.29ms step:3374/5535 train_time:824224ms step_avg:244.29ms step:3375/5535 train_time:824471ms step_avg:244.29ms step:3375/5535 val_loss:3.062100 train_time:824719ms step_avg:244.36ms step:3376/5535 train_time:824740ms step_avg:244.30ms step:3377/5535 train_time:824967ms step_avg:244.29ms step:3378/5535 train_time:825229ms step_avg:244.30ms step:3379/5535 train_time:825481ms step_avg:244.30ms step:3380/5535 train_time:825726ms step_avg:244.30ms step:3381/5535 train_time:825973ms step_avg:244.30ms step:3382/5535 train_time:826225ms step_avg:244.30ms step:3383/5535 train_time:826469ms step_avg:244.30ms step:3384/5535 train_time:826717ms step_avg:244.30ms step:3385/5535 train_time:826963ms step_avg:244.30ms step:3386/5535 train_time:827216ms step_avg:244.30ms step:3387/5535 train_time:827462ms step_avg:244.31ms step:3388/5535 train_time:827708ms step_avg:244.31ms step:3389/5535 train_time:827985ms step_avg:244.32ms step:3390/5535 train_time:828212ms step_avg:244.31ms step:3391/5535 train_time:828460ms step_avg:244.31ms step:3392/5535 train_time:828707ms step_avg:244.31ms step:3393/5535 train_time:828955ms step_avg:244.31ms step:3394/5535 train_time:829210ms step_avg:244.32ms step:3395/5535 train_time:829461ms step_avg:244.32ms step:3396/5535 train_time:829710ms step_avg:244.32ms step:3397/5535 train_time:829961ms step_avg:244.32ms step:3398/5535 train_time:830208ms step_avg:244.32ms step:3399/5535 train_time:830458ms step_avg:244.32ms step:3400/5535 train_time:830702ms step_avg:244.32ms step:3401/5535 train_time:830954ms step_avg:244.33ms step:3402/5535 train_time:831213ms step_avg:244.33ms step:3403/5535 train_time:831470ms step_avg:244.33ms step:3404/5535 train_time:831715ms step_avg:244.33ms step:3405/5535 train_time:831960ms step_avg:244.33ms step:3406/5535 train_time:832208ms step_avg:244.34ms step:3407/5535 train_time:832461ms step_avg:244.34ms step:3408/5535 train_time:832705ms step_avg:244.34ms step:3409/5535 train_time:832951ms step_avg:244.34ms step:3410/5535 train_time:833201ms step_avg:244.34ms step:3411/5535 train_time:833453ms step_avg:244.34ms step:3412/5535 train_time:833702ms step_avg:244.34ms step:3413/5535 train_time:833947ms step_avg:244.34ms step:3414/5535 train_time:834195ms step_avg:244.35ms step:3415/5535 train_time:834441ms step_avg:244.35ms step:3416/5535 train_time:834688ms step_avg:244.35ms step:3417/5535 train_time:834966ms step_avg:244.36ms step:3418/5535 train_time:835191ms step_avg:244.35ms step:3419/5535 train_time:835437ms step_avg:244.35ms step:3420/5535 train_time:835691ms step_avg:244.35ms step:3421/5535 train_time:835938ms step_avg:244.36ms step:3422/5535 train_time:836187ms step_avg:244.36ms step:3423/5535 train_time:836435ms step_avg:244.36ms step:3424/5535 train_time:836689ms step_avg:244.36ms step:3425/5535 train_time:836936ms step_avg:244.36ms step:3426/5535 train_time:837188ms step_avg:244.36ms step:3427/5535 train_time:837437ms step_avg:244.36ms step:3428/5535 train_time:837685ms step_avg:244.37ms step:3429/5535 train_time:837943ms step_avg:244.37ms step:3430/5535 train_time:838190ms step_avg:244.37ms step:3431/5535 train_time:838440ms step_avg:244.37ms step:3432/5535 train_time:838687ms step_avg:244.37ms step:3433/5535 train_time:838931ms step_avg:244.37ms step:3434/5535 train_time:839179ms step_avg:244.37ms step:3435/5535 train_time:839425ms step_avg:244.37ms step:3436/5535 train_time:839675ms step_avg:244.38ms step:3437/5535 train_time:839923ms step_avg:244.38ms step:3438/5535 train_time:840174ms step_avg:244.38ms step:3439/5535 train_time:840422ms step_avg:244.38ms step:3440/5535 train_time:840671ms step_avg:244.38ms step:3441/5535 train_time:840918ms step_avg:244.38ms step:3442/5535 train_time:841170ms step_avg:244.38ms step:3443/5535 train_time:841417ms step_avg:244.38ms step:3444/5535 train_time:841675ms step_avg:244.39ms step:3445/5535 train_time:841954ms step_avg:244.40ms step:3446/5535 train_time:842181ms step_avg:244.39ms step:3447/5535 train_time:842426ms step_avg:244.39ms step:3448/5535 train_time:842680ms step_avg:244.40ms step:3449/5535 train_time:842930ms step_avg:244.40ms step:3450/5535 train_time:843179ms step_avg:244.40ms step:3451/5535 train_time:843426ms step_avg:244.40ms step:3452/5535 train_time:843674ms step_avg:244.40ms step:3453/5535 train_time:843928ms step_avg:244.40ms step:3454/5535 train_time:844176ms step_avg:244.41ms step:3455/5535 train_time:844423ms step_avg:244.41ms step:3456/5535 train_time:844672ms step_avg:244.41ms step:3457/5535 train_time:844925ms step_avg:244.41ms step:3458/5535 train_time:845171ms step_avg:244.41ms step:3459/5535 train_time:845416ms step_avg:244.41ms step:3460/5535 train_time:845669ms step_avg:244.41ms step:3461/5535 train_time:845916ms step_avg:244.41ms step:3462/5535 train_time:846167ms step_avg:244.42ms step:3463/5535 train_time:846419ms step_avg:244.42ms step:3464/5535 train_time:846666ms step_avg:244.42ms step:3465/5535 train_time:846912ms step_avg:244.42ms step:3466/5535 train_time:847159ms step_avg:244.42ms step:3467/5535 train_time:847407ms step_avg:244.42ms step:3468/5535 train_time:847655ms step_avg:244.42ms step:3469/5535 train_time:847904ms step_avg:244.42ms step:3470/5535 train_time:848149ms step_avg:244.42ms step:3471/5535 train_time:848400ms step_avg:244.43ms step:3472/5535 train_time:848647ms step_avg:244.43ms step:3473/5535 train_time:848928ms step_avg:244.44ms step:3474/5535 train_time:849152ms step_avg:244.43ms step:3475/5535 train_time:849400ms step_avg:244.43ms step:3476/5535 train_time:849656ms step_avg:244.43ms step:3477/5535 train_time:849903ms step_avg:244.44ms step:3478/5535 train_time:850150ms step_avg:244.44ms step:3479/5535 train_time:850394ms step_avg:244.44ms step:3480/5535 train_time:850643ms step_avg:244.44ms step:3481/5535 train_time:850891ms step_avg:244.44ms step:3482/5535 train_time:851142ms step_avg:244.44ms step:3483/5535 train_time:851387ms step_avg:244.44ms step:3484/5535 train_time:851638ms step_avg:244.44ms step:3485/5535 train_time:851884ms step_avg:244.44ms step:3486/5535 train_time:852137ms step_avg:244.45ms step:3487/5535 train_time:852390ms step_avg:244.45ms step:3488/5535 train_time:852638ms step_avg:244.45ms step:3489/5535 train_time:852886ms step_avg:244.45ms step:3490/5535 train_time:853134ms step_avg:244.45ms step:3491/5535 train_time:853379ms step_avg:244.45ms step:3492/5535 train_time:853631ms step_avg:244.45ms step:3493/5535 train_time:853881ms step_avg:244.45ms step:3494/5535 train_time:854128ms step_avg:244.46ms step:3495/5535 train_time:854377ms step_avg:244.46ms step:3496/5535 train_time:854624ms step_avg:244.46ms step:3497/5535 train_time:854870ms step_avg:244.46ms step:3498/5535 train_time:855117ms step_avg:244.46ms step:3499/5535 train_time:855367ms step_avg:244.46ms step:3500/5535 train_time:855613ms step_avg:244.46ms step:3500/5535 val_loss:3.053267 train_time:855864ms step_avg:244.53ms step:3501/5535 train_time:855895ms step_avg:244.47ms step:3502/5535 train_time:856126ms step_avg:244.47ms step:3503/5535 train_time:856380ms step_avg:244.47ms step:3504/5535 train_time:856624ms step_avg:244.47ms step:3505/5535 train_time:856878ms step_avg:244.47ms step:3506/5535 train_time:857125ms step_avg:244.47ms step:3507/5535 train_time:857374ms step_avg:244.48ms step:3508/5535 train_time:857620ms step_avg:244.48ms step:3509/5535 train_time:857866ms step_avg:244.48ms step:3510/5535 train_time:858115ms step_avg:244.48ms step:3511/5535 train_time:858364ms step_avg:244.48ms step:3512/5535 train_time:858611ms step_avg:244.48ms step:3513/5535 train_time:858856ms step_avg:244.48ms step:3514/5535 train_time:859101ms step_avg:244.48ms step:3515/5535 train_time:859349ms step_avg:244.48ms step:3516/5535 train_time:859597ms step_avg:244.48ms step:3517/5535 train_time:859844ms step_avg:244.48ms step:3518/5535 train_time:860091ms step_avg:244.48ms step:3519/5535 train_time:860344ms step_avg:244.49ms step:3520/5535 train_time:860593ms step_avg:244.49ms step:3521/5535 train_time:860841ms step_avg:244.49ms step:3522/5535 train_time:861090ms step_avg:244.49ms step:3523/5535 train_time:861335ms step_avg:244.49ms step:3524/5535 train_time:861584ms step_avg:244.49ms step:3525/5535 train_time:861830ms step_avg:244.49ms step:3526/5535 train_time:862077ms step_avg:244.49ms step:3527/5535 train_time:862327ms step_avg:244.49ms step:3528/5535 train_time:862574ms step_avg:244.49ms step:3529/5535 train_time:862851ms step_avg:244.50ms step:3530/5535 train_time:863076ms step_avg:244.50ms step:3531/5535 train_time:863323ms step_avg:244.50ms step:3532/5535 train_time:863574ms step_avg:244.50ms step:3533/5535 train_time:863823ms step_avg:244.50ms step:3534/5535 train_time:864077ms step_avg:244.50ms step:3535/5535 train_time:864323ms step_avg:244.50ms step:3536/5535 train_time:864574ms step_avg:244.51ms step:3537/5535 train_time:864826ms step_avg:244.51ms step:3538/5535 train_time:865071ms step_avg:244.51ms step:3539/5535 train_time:865321ms step_avg:244.51ms step:3540/5535 train_time:865569ms step_avg:244.51ms step:3541/5535 train_time:865815ms step_avg:244.51ms step:3542/5535 train_time:866068ms step_avg:244.51ms step:3543/5535 train_time:866314ms step_avg:244.51ms step:3544/5535 train_time:866563ms step_avg:244.52ms step:3545/5535 train_time:866811ms step_avg:244.52ms step:3546/5535 train_time:867060ms step_avg:244.52ms step:3547/5535 train_time:867307ms step_avg:244.52ms step:3548/5535 train_time:867555ms step_avg:244.52ms step:3549/5535 train_time:867803ms step_avg:244.52ms step:3550/5535 train_time:868055ms step_avg:244.52ms step:3551/5535 train_time:868303ms step_avg:244.52ms step:3552/5535 train_time:868558ms step_avg:244.53ms step:3553/5535 train_time:868821ms step_avg:244.53ms step:3554/5535 train_time:869067ms step_avg:244.53ms step:3555/5535 train_time:869316ms step_avg:244.53ms step:3556/5535 train_time:869563ms step_avg:244.53ms step:3557/5535 train_time:869848ms step_avg:244.55ms step:3558/5535 train_time:870073ms step_avg:244.54ms step:3559/5535 train_time:870318ms step_avg:244.54ms step:3560/5535 train_time:870566ms step_avg:244.54ms step:3561/5535 train_time:870815ms step_avg:244.54ms step:3562/5535 train_time:871070ms step_avg:244.55ms step:3563/5535 train_time:871315ms step_avg:244.55ms step:3564/5535 train_time:871572ms step_avg:244.55ms step:3565/5535 train_time:871820ms step_avg:244.55ms step:3566/5535 train_time:872069ms step_avg:244.55ms step:3567/5535 train_time:872314ms step_avg:244.55ms step:3568/5535 train_time:872563ms step_avg:244.55ms step:3569/5535 train_time:872809ms step_avg:244.55ms step:3570/5535 train_time:873058ms step_avg:244.55ms step:3571/5535 train_time:873304ms step_avg:244.55ms step:3572/5535 train_time:873549ms step_avg:244.55ms step:3573/5535 train_time:873799ms step_avg:244.56ms step:3574/5535 train_time:874049ms step_avg:244.56ms step:3575/5535 train_time:874298ms step_avg:244.56ms step:3576/5535 train_time:874545ms step_avg:244.56ms step:3577/5535 train_time:874793ms step_avg:244.56ms step:3578/5535 train_time:875039ms step_avg:244.56ms step:3579/5535 train_time:875286ms step_avg:244.56ms step:3580/5535 train_time:875534ms step_avg:244.56ms step:3581/5535 train_time:875782ms step_avg:244.56ms step:3582/5535 train_time:876031ms step_avg:244.56ms step:3583/5535 train_time:876278ms step_avg:244.57ms step:3584/5535 train_time:876522ms step_avg:244.57ms step:3585/5535 train_time:876803ms step_avg:244.58ms step:3586/5535 train_time:877030ms step_avg:244.57ms step:3587/5535 train_time:877277ms step_avg:244.57ms step:3588/5535 train_time:877524ms step_avg:244.57ms step:3589/5535 train_time:877771ms step_avg:244.57ms step:3590/5535 train_time:878015ms step_avg:244.57ms step:3591/5535 train_time:878264ms step_avg:244.57ms step:3592/5535 train_time:878511ms step_avg:244.57ms step:3593/5535 train_time:878761ms step_avg:244.58ms step:3594/5535 train_time:879006ms step_avg:244.58ms step:3595/5535 train_time:879261ms step_avg:244.58ms step:3596/5535 train_time:879510ms step_avg:244.58ms step:3597/5535 train_time:879758ms step_avg:244.58ms step:3598/5535 train_time:880013ms step_avg:244.58ms step:3599/5535 train_time:880265ms step_avg:244.59ms step:3600/5535 train_time:880514ms step_avg:244.59ms step:3601/5535 train_time:880758ms step_avg:244.59ms step:3602/5535 train_time:881007ms step_avg:244.59ms step:3603/5535 train_time:881252ms step_avg:244.59ms step:3604/5535 train_time:881500ms step_avg:244.59ms step:3605/5535 train_time:881750ms step_avg:244.59ms step:3606/5535 train_time:882008ms step_avg:244.59ms step:3607/5535 train_time:882253ms step_avg:244.59ms step:3608/5535 train_time:882503ms step_avg:244.60ms step:3609/5535 train_time:882751ms step_avg:244.60ms step:3610/5535 train_time:883002ms step_avg:244.60ms step:3611/5535 train_time:883252ms step_avg:244.60ms step:3612/5535 train_time:883500ms step_avg:244.60ms step:3613/5535 train_time:883774ms step_avg:244.61ms step:3614/5535 train_time:884001ms step_avg:244.60ms step:3615/5535 train_time:884247ms step_avg:244.61ms step:3616/5535 train_time:884495ms step_avg:244.61ms step:3617/5535 train_time:884739ms step_avg:244.61ms step:3618/5535 train_time:884993ms step_avg:244.61ms step:3619/5535 train_time:885241ms step_avg:244.61ms step:3620/5535 train_time:885487ms step_avg:244.61ms step:3621/5535 train_time:885738ms step_avg:244.61ms step:3622/5535 train_time:885991ms step_avg:244.61ms step:3623/5535 train_time:886239ms step_avg:244.61ms step:3624/5535 train_time:886487ms step_avg:244.62ms step:3625/5535 train_time:886735ms step_avg:244.62ms step:3625/5535 val_loss:3.044600 train_time:886986ms step_avg:244.69ms step:3626/5535 train_time:887007ms step_avg:244.62ms step:3627/5535 train_time:887234ms step_avg:244.62ms step:3628/5535 train_time:887493ms step_avg:244.62ms step:3629/5535 train_time:887737ms step_avg:244.62ms step:3630/5535 train_time:887983ms step_avg:244.62ms step:3631/5535 train_time:888229ms step_avg:244.62ms step:3632/5535 train_time:888486ms step_avg:244.63ms step:3633/5535 train_time:888731ms step_avg:244.63ms step:3634/5535 train_time:888984ms step_avg:244.63ms step:3635/5535 train_time:889228ms step_avg:244.63ms step:3636/5535 train_time:889476ms step_avg:244.63ms step:3637/5535 train_time:889724ms step_avg:244.63ms step:3638/5535 train_time:889973ms step_avg:244.63ms step:3639/5535 train_time:890217ms step_avg:244.63ms step:3640/5535 train_time:890467ms step_avg:244.63ms step:3641/5535 train_time:890746ms step_avg:244.64ms step:3642/5535 train_time:890981ms step_avg:244.64ms step:3643/5535 train_time:891226ms step_avg:244.64ms step:3644/5535 train_time:891474ms step_avg:244.64ms step:3645/5535 train_time:891725ms step_avg:244.64ms step:3646/5535 train_time:891972ms step_avg:244.64ms step:3647/5535 train_time:892221ms step_avg:244.65ms step:3648/5535 train_time:892466ms step_avg:244.65ms step:3649/5535 train_time:892716ms step_avg:244.65ms step:3650/5535 train_time:892966ms step_avg:244.65ms step:3651/5535 train_time:893214ms step_avg:244.65ms step:3652/5535 train_time:893459ms step_avg:244.65ms step:3653/5535 train_time:893707ms step_avg:244.65ms step:3654/5535 train_time:893955ms step_avg:244.65ms step:3655/5535 train_time:894202ms step_avg:244.65ms step:3656/5535 train_time:894449ms step_avg:244.65ms step:3657/5535 train_time:894700ms step_avg:244.65ms step:3658/5535 train_time:894949ms step_avg:244.66ms step:3659/5535 train_time:895199ms step_avg:244.66ms step:3660/5535 train_time:895444ms step_avg:244.66ms step:3661/5535 train_time:895693ms step_avg:244.66ms step:3662/5535 train_time:895945ms step_avg:244.66ms step:3663/5535 train_time:896193ms step_avg:244.66ms step:3664/5535 train_time:896439ms step_avg:244.66ms step:3665/5535 train_time:896684ms step_avg:244.66ms step:3666/5535 train_time:896931ms step_avg:244.66ms step:3667/5535 train_time:897182ms step_avg:244.66ms step:3668/5535 train_time:897429ms step_avg:244.66ms step:3669/5535 train_time:897702ms step_avg:244.67ms step:3670/5535 train_time:897933ms step_avg:244.67ms step:3671/5535 train_time:898184ms step_avg:244.67ms step:3672/5535 train_time:898440ms step_avg:244.67ms step:3673/5535 train_time:898687ms step_avg:244.67ms step:3674/5535 train_time:898938ms step_avg:244.68ms step:3675/5535 train_time:899190ms step_avg:244.68ms step:3676/5535 train_time:899438ms step_avg:244.68ms step:3677/5535 train_time:899685ms step_avg:244.68ms step:3678/5535 train_time:899930ms step_avg:244.68ms step:3679/5535 train_time:900176ms step_avg:244.68ms step:3680/5535 train_time:900426ms step_avg:244.68ms step:3681/5535 train_time:900675ms step_avg:244.68ms step:3682/5535 train_time:900927ms step_avg:244.68ms step:3683/5535 train_time:901172ms step_avg:244.68ms step:3684/5535 train_time:901420ms step_avg:244.69ms step:3685/5535 train_time:901666ms step_avg:244.69ms step:3686/5535 train_time:901913ms step_avg:244.69ms step:3687/5535 train_time:902163ms step_avg:244.69ms step:3688/5535 train_time:902410ms step_avg:244.69ms step:3689/5535 train_time:902658ms step_avg:244.69ms step:3690/5535 train_time:902905ms step_avg:244.69ms step:3691/5535 train_time:903152ms step_avg:244.69ms step:3692/5535 train_time:903398ms step_avg:244.69ms step:3693/5535 train_time:903649ms step_avg:244.69ms step:3694/5535 train_time:903898ms step_avg:244.69ms step:3695/5535 train_time:904145ms step_avg:244.69ms step:3696/5535 train_time:904395ms step_avg:244.70ms step:3697/5535 train_time:904675ms step_avg:244.71ms step:3698/5535 train_time:904902ms step_avg:244.70ms step:3699/5535 train_time:905147ms step_avg:244.70ms step:3700/5535 train_time:905398ms step_avg:244.70ms step:3701/5535 train_time:905645ms step_avg:244.70ms step:3702/5535 train_time:905894ms step_avg:244.70ms step:3703/5535 train_time:906138ms step_avg:244.70ms step:3704/5535 train_time:906389ms step_avg:244.71ms step:3705/5535 train_time:906638ms step_avg:244.71ms step:3706/5535 train_time:906889ms step_avg:244.71ms step:3707/5535 train_time:907141ms step_avg:244.71ms step:3708/5535 train_time:907391ms step_avg:244.71ms step:3709/5535 train_time:907638ms step_avg:244.71ms step:3710/5535 train_time:907888ms step_avg:244.71ms step:3711/5535 train_time:908139ms step_avg:244.72ms step:3712/5535 train_time:908386ms step_avg:244.72ms step:3713/5535 train_time:908636ms step_avg:244.72ms step:3714/5535 train_time:908885ms step_avg:244.72ms step:3715/5535 train_time:909136ms step_avg:244.72ms step:3716/5535 train_time:909385ms step_avg:244.72ms step:3717/5535 train_time:909642ms step_avg:244.72ms step:3718/5535 train_time:909891ms step_avg:244.73ms step:3719/5535 train_time:910139ms step_avg:244.73ms step:3720/5535 train_time:910385ms step_avg:244.73ms step:3721/5535 train_time:910638ms step_avg:244.73ms step:3722/5535 train_time:910893ms step_avg:244.73ms step:3723/5535 train_time:911142ms step_avg:244.73ms step:3724/5535 train_time:911390ms step_avg:244.73ms step:3725/5535 train_time:911669ms step_avg:244.74ms step:3726/5535 train_time:911896ms step_avg:244.74ms step:3727/5535 train_time:912142ms step_avg:244.74ms step:3728/5535 train_time:912389ms step_avg:244.74ms step:3729/5535 train_time:912638ms step_avg:244.74ms step:3730/5535 train_time:912896ms step_avg:244.74ms step:3731/5535 train_time:913148ms step_avg:244.75ms step:3732/5535 train_time:913395ms step_avg:244.75ms step:3733/5535 train_time:913650ms step_avg:244.75ms step:3734/5535 train_time:913901ms step_avg:244.75ms step:3735/5535 train_time:914150ms step_avg:244.75ms step:3736/5535 train_time:914397ms step_avg:244.75ms step:3737/5535 train_time:914649ms step_avg:244.75ms step:3738/5535 train_time:914897ms step_avg:244.76ms step:3739/5535 train_time:915142ms step_avg:244.76ms step:3740/5535 train_time:915396ms step_avg:244.76ms step:3741/5535 train_time:915646ms step_avg:244.76ms step:3742/5535 train_time:915894ms step_avg:244.76ms step:3743/5535 train_time:916139ms step_avg:244.76ms step:3744/5535 train_time:916392ms step_avg:244.76ms step:3745/5535 train_time:916639ms step_avg:244.76ms step:3746/5535 train_time:916885ms step_avg:244.76ms step:3747/5535 train_time:917134ms step_avg:244.77ms step:3748/5535 train_time:917384ms step_avg:244.77ms step:3749/5535 train_time:917633ms step_avg:244.77ms step:3750/5535 train_time:917883ms step_avg:244.77ms step:3750/5535 val_loss:3.034938 train_time:918132ms step_avg:244.84ms step:3751/5535 train_time:918153ms step_avg:244.78ms step:3752/5535 train_time:918381ms step_avg:244.77ms step:3753/5535 train_time:918671ms step_avg:244.78ms step:3754/5535 train_time:918897ms step_avg:244.78ms step:3755/5535 train_time:919141ms step_avg:244.78ms step:3756/5535 train_time:919387ms step_avg:244.78ms step:3757/5535 train_time:919642ms step_avg:244.78ms step:3758/5535 train_time:919898ms step_avg:244.78ms step:3759/5535 train_time:920142ms step_avg:244.78ms step:3760/5535 train_time:920395ms step_avg:244.79ms step:3761/5535 train_time:920647ms step_avg:244.79ms step:3762/5535 train_time:920897ms step_avg:244.79ms step:3763/5535 train_time:921144ms step_avg:244.79ms step:3764/5535 train_time:921394ms step_avg:244.79ms step:3765/5535 train_time:921642ms step_avg:244.79ms step:3766/5535 train_time:921889ms step_avg:244.79ms step:3767/5535 train_time:922139ms step_avg:244.79ms step:3768/5535 train_time:922385ms step_avg:244.79ms step:3769/5535 train_time:922632ms step_avg:244.79ms step:3770/5535 train_time:922881ms step_avg:244.80ms step:3771/5535 train_time:923131ms step_avg:244.80ms step:3772/5535 train_time:923379ms step_avg:244.80ms step:3773/5535 train_time:923628ms step_avg:244.80ms step:3774/5535 train_time:923875ms step_avg:244.80ms step:3775/5535 train_time:924126ms step_avg:244.80ms step:3776/5535 train_time:924374ms step_avg:244.80ms step:3777/5535 train_time:924622ms step_avg:244.80ms step:3778/5535 train_time:924880ms step_avg:244.81ms step:3779/5535 train_time:925127ms step_avg:244.81ms step:3780/5535 train_time:925375ms step_avg:244.81ms step:3781/5535 train_time:925654ms step_avg:244.82ms step:3782/5535 train_time:925881ms step_avg:244.81ms step:3783/5535 train_time:926136ms step_avg:244.82ms step:3784/5535 train_time:926387ms step_avg:244.82ms step:3785/5535 train_time:926634ms step_avg:244.82ms step:3786/5535 train_time:926881ms step_avg:244.82ms step:3787/5535 train_time:927128ms step_avg:244.82ms step:3788/5535 train_time:927378ms step_avg:244.82ms step:3789/5535 train_time:927627ms step_avg:244.82ms step:3790/5535 train_time:927877ms step_avg:244.82ms step:3791/5535 train_time:928128ms step_avg:244.82ms step:3792/5535 train_time:928376ms step_avg:244.82ms step:3793/5535 train_time:928623ms step_avg:244.83ms step:3794/5535 train_time:928873ms step_avg:244.83ms step:3795/5535 train_time:929121ms step_avg:244.83ms step:3796/5535 train_time:929368ms step_avg:244.83ms step:3797/5535 train_time:929618ms step_avg:244.83ms step:3798/5535 train_time:929869ms step_avg:244.83ms step:3799/5535 train_time:930115ms step_avg:244.83ms step:3800/5535 train_time:930364ms step_avg:244.83ms step:3801/5535 train_time:930613ms step_avg:244.83ms step:3802/5535 train_time:930860ms step_avg:244.83ms step:3803/5535 train_time:931110ms step_avg:244.84ms step:3804/5535 train_time:931358ms step_avg:244.84ms step:3805/5535 train_time:931603ms step_avg:244.84ms step:3806/5535 train_time:931851ms step_avg:244.84ms step:3807/5535 train_time:932102ms step_avg:244.84ms step:3808/5535 train_time:932354ms step_avg:244.84ms step:3809/5535 train_time:932632ms step_avg:244.85ms step:3810/5535 train_time:932861ms step_avg:244.85ms step:3811/5535 train_time:933106ms step_avg:244.85ms step:3812/5535 train_time:933355ms step_avg:244.85ms step:3813/5535 train_time:933602ms step_avg:244.85ms step:3814/5535 train_time:933854ms step_avg:244.85ms step:3815/5535 train_time:934107ms step_avg:244.85ms step:3816/5535 train_time:934355ms step_avg:244.85ms step:3817/5535 train_time:934606ms step_avg:244.85ms step:3818/5535 train_time:934854ms step_avg:244.85ms step:3819/5535 train_time:935104ms step_avg:244.86ms step:3820/5535 train_time:935353ms step_avg:244.86ms step:3821/5535 train_time:935599ms step_avg:244.86ms step:3822/5535 train_time:935845ms step_avg:244.86ms step:3823/5535 train_time:936092ms step_avg:244.86ms step:3824/5535 train_time:936340ms step_avg:244.86ms step:3825/5535 train_time:936585ms step_avg:244.86ms step:3826/5535 train_time:936833ms step_avg:244.86ms step:3827/5535 train_time:937080ms step_avg:244.86ms step:3828/5535 train_time:937332ms step_avg:244.86ms step:3829/5535 train_time:937587ms step_avg:244.86ms step:3830/5535 train_time:937832ms step_avg:244.86ms step:3831/5535 train_time:938079ms step_avg:244.87ms step:3832/5535 train_time:938323ms step_avg:244.87ms step:3833/5535 train_time:938575ms step_avg:244.87ms step:3834/5535 train_time:938823ms step_avg:244.87ms step:3835/5535 train_time:939068ms step_avg:244.87ms step:3836/5535 train_time:939320ms step_avg:244.87ms step:3837/5535 train_time:939603ms step_avg:244.88ms step:3838/5535 train_time:939828ms step_avg:244.87ms step:3839/5535 train_time:940079ms step_avg:244.88ms step:3840/5535 train_time:940337ms step_avg:244.88ms step:3841/5535 train_time:940592ms step_avg:244.88ms step:3842/5535 train_time:940835ms step_avg:244.88ms step:3843/5535 train_time:941085ms step_avg:244.88ms step:3844/5535 train_time:941334ms step_avg:244.88ms step:3845/5535 train_time:941580ms step_avg:244.88ms step:3846/5535 train_time:941830ms step_avg:244.89ms step:3847/5535 train_time:942077ms step_avg:244.89ms step:3848/5535 train_time:942338ms step_avg:244.89ms step:3849/5535 train_time:942586ms step_avg:244.89ms step:3850/5535 train_time:942835ms step_avg:244.89ms step:3851/5535 train_time:943083ms step_avg:244.89ms step:3852/5535 train_time:943334ms step_avg:244.89ms step:3853/5535 train_time:943585ms step_avg:244.90ms step:3854/5535 train_time:943836ms step_avg:244.90ms step:3855/5535 train_time:944083ms step_avg:244.90ms step:3856/5535 train_time:944332ms step_avg:244.90ms step:3857/5535 train_time:944583ms step_avg:244.90ms step:3858/5535 train_time:944837ms step_avg:244.90ms step:3859/5535 train_time:945086ms step_avg:244.90ms step:3860/5535 train_time:945334ms step_avg:244.91ms step:3861/5535 train_time:945589ms step_avg:244.91ms step:3862/5535 train_time:945836ms step_avg:244.91ms step:3863/5535 train_time:946094ms step_avg:244.91ms step:3864/5535 train_time:946346ms step_avg:244.91ms step:3865/5535 train_time:946622ms step_avg:244.92ms step:3866/5535 train_time:946844ms step_avg:244.92ms step:3867/5535 train_time:947093ms step_avg:244.92ms step:3868/5535 train_time:947340ms step_avg:244.92ms step:3869/5535 train_time:947590ms step_avg:244.92ms step:3870/5535 train_time:947835ms step_avg:244.92ms step:3871/5535 train_time:948084ms step_avg:244.92ms step:3872/5535 train_time:948332ms step_avg:244.92ms step:3873/5535 train_time:948581ms step_avg:244.92ms step:3874/5535 train_time:948825ms step_avg:244.92ms step:3875/5535 train_time:949072ms step_avg:244.92ms step:3875/5535 val_loss:3.025910 train_time:949321ms step_avg:244.99ms step:3876/5535 train_time:949343ms step_avg:244.93ms step:3877/5535 train_time:949570ms step_avg:244.92ms step:3878/5535 train_time:949828ms step_avg:244.93ms step:3879/5535 train_time:950075ms step_avg:244.93ms step:3880/5535 train_time:950321ms step_avg:244.93ms step:3881/5535 train_time:950566ms step_avg:244.93ms step:3882/5535 train_time:950817ms step_avg:244.93ms step:3883/5535 train_time:951071ms step_avg:244.93ms step:3884/5535 train_time:951319ms step_avg:244.93ms step:3885/5535 train_time:951570ms step_avg:244.93ms step:3886/5535 train_time:951824ms step_avg:244.94ms step:3887/5535 train_time:952070ms step_avg:244.94ms step:3888/5535 train_time:952316ms step_avg:244.94ms step:3889/5535 train_time:952562ms step_avg:244.94ms step:3890/5535 train_time:952809ms step_avg:244.94ms step:3891/5535 train_time:953055ms step_avg:244.94ms step:3892/5535 train_time:953303ms step_avg:244.94ms step:3893/5535 train_time:953582ms step_avg:244.95ms step:3894/5535 train_time:953804ms step_avg:244.94ms step:3895/5535 train_time:954055ms step_avg:244.94ms step:3896/5535 train_time:954300ms step_avg:244.94ms step:3897/5535 train_time:954549ms step_avg:244.94ms step:3898/5535 train_time:954801ms step_avg:244.95ms step:3899/5535 train_time:955051ms step_avg:244.95ms step:3900/5535 train_time:955299ms step_avg:244.95ms step:3901/5535 train_time:955548ms step_avg:244.95ms step:3902/5535 train_time:955796ms step_avg:244.95ms step:3903/5535 train_time:956043ms step_avg:244.95ms step:3904/5535 train_time:956292ms step_avg:244.95ms step:3905/5535 train_time:956546ms step_avg:244.95ms step:3906/5535 train_time:956797ms step_avg:244.96ms step:3907/5535 train_time:957047ms step_avg:244.96ms step:3908/5535 train_time:957294ms step_avg:244.96ms step:3909/5535 train_time:957543ms step_avg:244.96ms step:3910/5535 train_time:957792ms step_avg:244.96ms step:3911/5535 train_time:958040ms step_avg:244.96ms step:3912/5535 train_time:958291ms step_avg:244.96ms step:3913/5535 train_time:958542ms step_avg:244.96ms step:3914/5535 train_time:958790ms step_avg:244.96ms step:3915/5535 train_time:959036ms step_avg:244.96ms step:3916/5535 train_time:959286ms step_avg:244.97ms step:3917/5535 train_time:959542ms step_avg:244.97ms step:3918/5535 train_time:959792ms step_avg:244.97ms step:3919/5535 train_time:960041ms step_avg:244.97ms step:3920/5535 train_time:960293ms step_avg:244.97ms step:3921/5535 train_time:960572ms step_avg:244.98ms step:3922/5535 train_time:960800ms step_avg:244.98ms step:3923/5535 train_time:961046ms step_avg:244.98ms step:3924/5535 train_time:961297ms step_avg:244.98ms step:3925/5535 train_time:961547ms step_avg:244.98ms step:3926/5535 train_time:961804ms step_avg:244.98ms step:3927/5535 train_time:962049ms step_avg:244.98ms step:3928/5535 train_time:962295ms step_avg:244.98ms step:3929/5535 train_time:962544ms step_avg:244.98ms step:3930/5535 train_time:962794ms step_avg:244.99ms step:3931/5535 train_time:963041ms step_avg:244.99ms step:3932/5535 train_time:963288ms step_avg:244.99ms step:3933/5535 train_time:963541ms step_avg:244.99ms step:3934/5535 train_time:963790ms step_avg:244.99ms step:3935/5535 train_time:964043ms step_avg:244.99ms step:3936/5535 train_time:964289ms step_avg:244.99ms step:3937/5535 train_time:964539ms step_avg:244.99ms step:3938/5535 train_time:964789ms step_avg:244.99ms step:3939/5535 train_time:965036ms step_avg:245.00ms step:3940/5535 train_time:965290ms step_avg:245.00ms step:3941/5535 train_time:965541ms step_avg:245.00ms step:3942/5535 train_time:965795ms step_avg:245.00ms step:3943/5535 train_time:966043ms step_avg:245.00ms step:3944/5535 train_time:966290ms step_avg:245.00ms step:3945/5535 train_time:966540ms step_avg:245.00ms step:3946/5535 train_time:966789ms step_avg:245.00ms step:3947/5535 train_time:967041ms step_avg:245.01ms step:3948/5535 train_time:967295ms step_avg:245.01ms step:3949/5535 train_time:967578ms step_avg:245.02ms step:3950/5535 train_time:967805ms step_avg:245.01ms step:3951/5535 train_time:968051ms step_avg:245.01ms step:3952/5535 train_time:968297ms step_avg:245.01ms step:3953/5535 train_time:968542ms step_avg:245.01ms step:3954/5535 train_time:968792ms step_avg:245.02ms step:3955/5535 train_time:969043ms step_avg:245.02ms step:3956/5535 train_time:969293ms step_avg:245.02ms step:3957/5535 train_time:969542ms step_avg:245.02ms step:3958/5535 train_time:969795ms step_avg:245.02ms step:3959/5535 train_time:970045ms step_avg:245.02ms step:3960/5535 train_time:970290ms step_avg:245.02ms step:3961/5535 train_time:970542ms step_avg:245.02ms step:3962/5535 train_time:970789ms step_avg:245.03ms step:3963/5535 train_time:971039ms step_avg:245.03ms step:3964/5535 train_time:971289ms step_avg:245.03ms step:3965/5535 train_time:971538ms step_avg:245.03ms step:3966/5535 train_time:971785ms step_avg:245.03ms step:3967/5535 train_time:972034ms step_avg:245.03ms step:3968/5535 train_time:972287ms step_avg:245.03ms step:3969/5535 train_time:972533ms step_avg:245.03ms step:3970/5535 train_time:972780ms step_avg:245.03ms step:3971/5535 train_time:973029ms step_avg:245.03ms step:3972/5535 train_time:973281ms step_avg:245.04ms step:3973/5535 train_time:973530ms step_avg:245.04ms step:3974/5535 train_time:973780ms step_avg:245.04ms step:3975/5535 train_time:974028ms step_avg:245.04ms step:3976/5535 train_time:974275ms step_avg:245.04ms step:3977/5535 train_time:974559ms step_avg:245.05ms step:3978/5535 train_time:974784ms step_avg:245.04ms step:3979/5535 train_time:975029ms step_avg:245.04ms step:3980/5535 train_time:975277ms step_avg:245.04ms step:3981/5535 train_time:975526ms step_avg:245.05ms step:3982/5535 train_time:975775ms step_avg:245.05ms step:3983/5535 train_time:976020ms step_avg:245.05ms step:3984/5535 train_time:976274ms step_avg:245.05ms step:3985/5535 train_time:976527ms step_avg:245.05ms step:3986/5535 train_time:976776ms step_avg:245.05ms step:3987/5535 train_time:977021ms step_avg:245.05ms step:3988/5535 train_time:977273ms step_avg:245.05ms step:3989/5535 train_time:977522ms step_avg:245.05ms step:3990/5535 train_time:977772ms step_avg:245.06ms step:3991/5535 train_time:978020ms step_avg:245.06ms step:3992/5535 train_time:978267ms step_avg:245.06ms step:3993/5535 train_time:978516ms step_avg:245.06ms step:3994/5535 train_time:978763ms step_avg:245.06ms step:3995/5535 train_time:979014ms step_avg:245.06ms step:3996/5535 train_time:979262ms step_avg:245.06ms step:3997/5535 train_time:979515ms step_avg:245.06ms step:3998/5535 train_time:979763ms step_avg:245.06ms step:3999/5535 train_time:980009ms step_avg:245.06ms step:4000/5535 train_time:980260ms step_avg:245.06ms step:4000/5535 val_loss:3.016735 train_time:980513ms step_avg:245.13ms step:4001/5535 train_time:980538ms step_avg:245.07ms step:4002/5535 train_time:980763ms step_avg:245.07ms step:4003/5535 train_time:981021ms step_avg:245.07ms step:4004/5535 train_time:981272ms step_avg:245.07ms step:4005/5535 train_time:981546ms step_avg:245.08ms step:4006/5535 train_time:981777ms step_avg:245.08ms step:4007/5535 train_time:982030ms step_avg:245.08ms step:4008/5535 train_time:982277ms step_avg:245.08ms step:4009/5535 train_time:982527ms step_avg:245.08ms step:4010/5535 train_time:982772ms step_avg:245.08ms step:4011/5535 train_time:983025ms step_avg:245.08ms step:4012/5535 train_time:983285ms step_avg:245.09ms step:4013/5535 train_time:983524ms step_avg:245.08ms step:4014/5535 train_time:983770ms step_avg:245.08ms step:4015/5535 train_time:984023ms step_avg:245.09ms step:4016/5535 train_time:984270ms step_avg:245.09ms step:4017/5535 train_time:984517ms step_avg:245.09ms step:4018/5535 train_time:984764ms step_avg:245.09ms step:4019/5535 train_time:985014ms step_avg:245.09ms step:4020/5535 train_time:985261ms step_avg:245.09ms step:4021/5535 train_time:985511ms step_avg:245.09ms step:4022/5535 train_time:985759ms step_avg:245.09ms step:4023/5535 train_time:986009ms step_avg:245.09ms step:4024/5535 train_time:986258ms step_avg:245.09ms step:4025/5535 train_time:986506ms step_avg:245.09ms step:4026/5535 train_time:986764ms step_avg:245.10ms step:4027/5535 train_time:987010ms step_avg:245.10ms step:4028/5535 train_time:987258ms step_avg:245.10ms step:4029/5535 train_time:987505ms step_avg:245.10ms step:4030/5535 train_time:987753ms step_avg:245.10ms step:4031/5535 train_time:988001ms step_avg:245.10ms step:4032/5535 train_time:988248ms step_avg:245.10ms step:4033/5535 train_time:988524ms step_avg:245.11ms step:4034/5535 train_time:988750ms step_avg:245.10ms step:4035/5535 train_time:988998ms step_avg:245.10ms step:4036/5535 train_time:989252ms step_avg:245.11ms step:4037/5535 train_time:989501ms step_avg:245.11ms step:4038/5535 train_time:989753ms step_avg:245.11ms step:4039/5535 train_time:989999ms step_avg:245.11ms step:4040/5535 train_time:990248ms step_avg:245.11ms step:4041/5535 train_time:990497ms step_avg:245.11ms step:4042/5535 train_time:990749ms step_avg:245.11ms step:4043/5535 train_time:990997ms step_avg:245.11ms step:4044/5535 train_time:991247ms step_avg:245.12ms step:4045/5535 train_time:991492ms step_avg:245.12ms step:4046/5535 train_time:991742ms step_avg:245.12ms step:4047/5535 train_time:991988ms step_avg:245.12ms step:4048/5535 train_time:992234ms step_avg:245.12ms step:4049/5535 train_time:992485ms step_avg:245.12ms step:4050/5535 train_time:992736ms step_avg:245.12ms step:4051/5535 train_time:992986ms step_avg:245.12ms step:4052/5535 train_time:993234ms step_avg:245.12ms step:4053/5535 train_time:993481ms step_avg:245.12ms step:4054/5535 train_time:993729ms step_avg:245.12ms step:4055/5535 train_time:993977ms step_avg:245.12ms step:4056/5535 train_time:994229ms step_avg:245.13ms step:4057/5535 train_time:994474ms step_avg:245.13ms step:4058/5535 train_time:994722ms step_avg:245.13ms step:4059/5535 train_time:994971ms step_avg:245.13ms step:4060/5535 train_time:995230ms step_avg:245.13ms step:4061/5535 train_time:995505ms step_avg:245.14ms step:4062/5535 train_time:995734ms step_avg:245.13ms step:4063/5535 train_time:995983ms step_avg:245.13ms step:4064/5535 train_time:996238ms step_avg:245.14ms step:4065/5535 train_time:996486ms step_avg:245.14ms step:4066/5535 train_time:996736ms step_avg:245.14ms step:4067/5535 train_time:996984ms step_avg:245.14ms step:4068/5535 train_time:997236ms step_avg:245.14ms step:4069/5535 train_time:997486ms step_avg:245.14ms step:4070/5535 train_time:997734ms step_avg:245.14ms step:4071/5535 train_time:997982ms step_avg:245.14ms step:4072/5535 train_time:998237ms step_avg:245.15ms step:4073/5535 train_time:998483ms step_avg:245.15ms step:4074/5535 train_time:998733ms step_avg:245.15ms step:4075/5535 train_time:998982ms step_avg:245.15ms step:4076/5535 train_time:999235ms step_avg:245.15ms step:4077/5535 train_time:999482ms step_avg:245.15ms step:4078/5535 train_time:999729ms step_avg:245.15ms step:4079/5535 train_time:999980ms step_avg:245.15ms step:4080/5535 train_time:1000226ms step_avg:245.15ms step:4081/5535 train_time:1000472ms step_avg:245.15ms step:4082/5535 train_time:1000719ms step_avg:245.15ms step:4083/5535 train_time:1000968ms step_avg:245.16ms step:4084/5535 train_time:1001213ms step_avg:245.16ms step:4085/5535 train_time:1001462ms step_avg:245.16ms step:4086/5535 train_time:1001715ms step_avg:245.16ms step:4087/5535 train_time:1001967ms step_avg:245.16ms step:4088/5535 train_time:1002212ms step_avg:245.16ms step:4089/5535 train_time:1002486ms step_avg:245.17ms step:4090/5535 train_time:1002719ms step_avg:245.16ms step:4091/5535 train_time:1002969ms step_avg:245.16ms step:4092/5535 train_time:1003216ms step_avg:245.17ms step:4093/5535 train_time:1003463ms step_avg:245.17ms step:4094/5535 train_time:1003714ms step_avg:245.17ms step:4095/5535 train_time:1003964ms step_avg:245.17ms step:4096/5535 train_time:1004213ms step_avg:245.17ms step:4097/5535 train_time:1004460ms step_avg:245.17ms step:4098/5535 train_time:1004719ms step_avg:245.17ms step:4099/5535 train_time:1004967ms step_avg:245.17ms step:4100/5535 train_time:1005217ms step_avg:245.17ms step:4101/5535 train_time:1005469ms step_avg:245.18ms step:4102/5535 train_time:1005722ms step_avg:245.18ms step:4103/5535 train_time:1005975ms step_avg:245.18ms step:4104/5535 train_time:1006235ms step_avg:245.18ms step:4105/5535 train_time:1006481ms step_avg:245.18ms step:4106/5535 train_time:1006729ms step_avg:245.18ms step:4107/5535 train_time:1006980ms step_avg:245.19ms step:4108/5535 train_time:1007236ms step_avg:245.19ms step:4109/5535 train_time:1007485ms step_avg:245.19ms step:4110/5535 train_time:1007736ms step_avg:245.19ms step:4111/5535 train_time:1007995ms step_avg:245.19ms step:4112/5535 train_time:1008247ms step_avg:245.20ms step:4113/5535 train_time:1008494ms step_avg:245.20ms step:4114/5535 train_time:1008749ms step_avg:245.20ms step:4115/5535 train_time:1009000ms step_avg:245.20ms step:4116/5535 train_time:1009249ms step_avg:245.20ms step:4117/5535 train_time:1009537ms step_avg:245.21ms step:4118/5535 train_time:1009761ms step_avg:245.21ms step:4119/5535 train_time:1010014ms step_avg:245.21ms step:4120/5535 train_time:1010266ms step_avg:245.21ms step:4121/5535 train_time:1010519ms step_avg:245.21ms step:4122/5535 train_time:1010770ms step_avg:245.21ms step:4123/5535 train_time:1011022ms step_avg:245.22ms step:4124/5535 train_time:1011272ms step_avg:245.22ms step:4125/5535 train_time:1011525ms step_avg:245.22ms step:4125/5535 val_loss:3.007094 train_time:1011780ms step_avg:245.28ms step:4126/5535 train_time:1011801ms step_avg:245.23ms step:4127/5535 train_time:1012035ms step_avg:245.22ms step:4128/5535 train_time:1012293ms step_avg:245.23ms step:4129/5535 train_time:1012540ms step_avg:245.23ms step:4130/5535 train_time:1012786ms step_avg:245.23ms step:4131/5535 train_time:1013035ms step_avg:245.23ms step:4132/5535 train_time:1013286ms step_avg:245.23ms step:4133/5535 train_time:1013538ms step_avg:245.23ms step:4134/5535 train_time:1013789ms step_avg:245.23ms step:4135/5535 train_time:1014038ms step_avg:245.23ms step:4136/5535 train_time:1014293ms step_avg:245.24ms step:4137/5535 train_time:1014551ms step_avg:245.24ms step:4138/5535 train_time:1014799ms step_avg:245.24ms step:4139/5535 train_time:1015055ms step_avg:245.24ms step:4140/5535 train_time:1015306ms step_avg:245.24ms step:4141/5535 train_time:1015556ms step_avg:245.24ms step:4142/5535 train_time:1015805ms step_avg:245.25ms step:4143/5535 train_time:1016056ms step_avg:245.25ms step:4144/5535 train_time:1016307ms step_avg:245.25ms step:4145/5535 train_time:1016586ms step_avg:245.26ms step:4146/5535 train_time:1016818ms step_avg:245.25ms step:4147/5535 train_time:1017066ms step_avg:245.25ms step:4148/5535 train_time:1017315ms step_avg:245.25ms step:4149/5535 train_time:1017567ms step_avg:245.26ms step:4150/5535 train_time:1017818ms step_avg:245.26ms step:4151/5535 train_time:1018068ms step_avg:245.26ms step:4152/5535 train_time:1018316ms step_avg:245.26ms step:4153/5535 train_time:1018569ms step_avg:245.26ms step:4154/5535 train_time:1018819ms step_avg:245.26ms step:4155/5535 train_time:1019069ms step_avg:245.26ms step:4156/5535 train_time:1019327ms step_avg:245.27ms step:4157/5535 train_time:1019587ms step_avg:245.27ms step:4158/5535 train_time:1019846ms step_avg:245.27ms step:4159/5535 train_time:1020097ms step_avg:245.27ms step:4160/5535 train_time:1020352ms step_avg:245.28ms step:4161/5535 train_time:1020604ms step_avg:245.28ms step:4162/5535 train_time:1020856ms step_avg:245.28ms step:4163/5535 train_time:1021110ms step_avg:245.28ms step:4164/5535 train_time:1021369ms step_avg:245.29ms step:4165/5535 train_time:1021620ms step_avg:245.29ms step:4166/5535 train_time:1021869ms step_avg:245.29ms step:4167/5535 train_time:1022118ms step_avg:245.29ms step:4168/5535 train_time:1022369ms step_avg:245.29ms step:4169/5535 train_time:1022619ms step_avg:245.29ms step:4170/5535 train_time:1022872ms step_avg:245.29ms step:4171/5535 train_time:1023120ms step_avg:245.29ms step:4172/5535 train_time:1023369ms step_avg:245.29ms step:4173/5535 train_time:1023651ms step_avg:245.30ms step:4174/5535 train_time:1023878ms step_avg:245.30ms step:4175/5535 train_time:1024128ms step_avg:245.30ms step:4176/5535 train_time:1024383ms step_avg:245.30ms step:4177/5535 train_time:1024634ms step_avg:245.30ms step:4178/5535 train_time:1024882ms step_avg:245.30ms step:4179/5535 train_time:1025130ms step_avg:245.31ms step:4180/5535 train_time:1025381ms step_avg:245.31ms step:4181/5535 train_time:1025631ms step_avg:245.31ms step:4182/5535 train_time:1025881ms step_avg:245.31ms step:4183/5535 train_time:1026130ms step_avg:245.31ms step:4184/5535 train_time:1026382ms step_avg:245.31ms step:4185/5535 train_time:1026629ms step_avg:245.31ms step:4186/5535 train_time:1026884ms step_avg:245.31ms step:4187/5535 train_time:1027142ms step_avg:245.32ms step:4188/5535 train_time:1027389ms step_avg:245.32ms step:4189/5535 train_time:1027641ms step_avg:245.32ms step:4190/5535 train_time:1027891ms step_avg:245.32ms step:4191/5535 train_time:1028153ms step_avg:245.32ms step:4192/5535 train_time:1028404ms step_avg:245.33ms step:4193/5535 train_time:1028647ms step_avg:245.32ms step:4194/5535 train_time:1028898ms step_avg:245.33ms step:4195/5535 train_time:1029149ms step_avg:245.33ms step:4196/5535 train_time:1029402ms step_avg:245.33ms step:4197/5535 train_time:1029648ms step_avg:245.33ms step:4198/5535 train_time:1029899ms step_avg:245.33ms step:4199/5535 train_time:1030148ms step_avg:245.33ms step:4200/5535 train_time:1030399ms step_avg:245.33ms step:4201/5535 train_time:1030674ms step_avg:245.34ms step:4202/5535 train_time:1030905ms step_avg:245.34ms step:4203/5535 train_time:1031148ms step_avg:245.34ms step:4204/5535 train_time:1031403ms step_avg:245.34ms step:4205/5535 train_time:1031649ms step_avg:245.34ms step:4206/5535 train_time:1031912ms step_avg:245.34ms step:4207/5535 train_time:1032167ms step_avg:245.35ms step:4208/5535 train_time:1032421ms step_avg:245.35ms step:4209/5535 train_time:1032672ms step_avg:245.35ms step:4210/5535 train_time:1032920ms step_avg:245.35ms step:4211/5535 train_time:1033169ms step_avg:245.35ms step:4212/5535 train_time:1033422ms step_avg:245.35ms step:4213/5535 train_time:1033670ms step_avg:245.35ms step:4214/5535 train_time:1033920ms step_avg:245.35ms step:4215/5535 train_time:1034169ms step_avg:245.35ms step:4216/5535 train_time:1034420ms step_avg:245.36ms step:4217/5535 train_time:1034668ms step_avg:245.36ms step:4218/5535 train_time:1034924ms step_avg:245.36ms step:4219/5535 train_time:1035180ms step_avg:245.36ms step:4220/5535 train_time:1035431ms step_avg:245.36ms step:4221/5535 train_time:1035681ms step_avg:245.36ms step:4222/5535 train_time:1035932ms step_avg:245.37ms step:4223/5535 train_time:1036185ms step_avg:245.37ms step:4224/5535 train_time:1036436ms step_avg:245.37ms step:4225/5535 train_time:1036681ms step_avg:245.37ms step:4226/5535 train_time:1036930ms step_avg:245.37ms step:4227/5535 train_time:1037182ms step_avg:245.37ms step:4228/5535 train_time:1037432ms step_avg:245.37ms step:4229/5535 train_time:1037715ms step_avg:245.38ms step:4230/5535 train_time:1037944ms step_avg:245.38ms step:4231/5535 train_time:1038190ms step_avg:245.38ms step:4232/5535 train_time:1038441ms step_avg:245.38ms step:4233/5535 train_time:1038690ms step_avg:245.38ms step:4234/5535 train_time:1038944ms step_avg:245.38ms step:4235/5535 train_time:1039191ms step_avg:245.38ms step:4236/5535 train_time:1039447ms step_avg:245.38ms step:4237/5535 train_time:1039696ms step_avg:245.38ms step:4238/5535 train_time:1039946ms step_avg:245.39ms step:4239/5535 train_time:1040196ms step_avg:245.39ms step:4240/5535 train_time:1040449ms step_avg:245.39ms step:4241/5535 train_time:1040701ms step_avg:245.39ms step:4242/5535 train_time:1040953ms step_avg:245.39ms step:4243/5535 train_time:1041205ms step_avg:245.39ms step:4244/5535 train_time:1041455ms step_avg:245.39ms step:4245/5535 train_time:1041704ms step_avg:245.40ms step:4246/5535 train_time:1041954ms step_avg:245.40ms step:4247/5535 train_time:1042211ms step_avg:245.40ms step:4248/5535 train_time:1042462ms step_avg:245.40ms step:4249/5535 train_time:1042723ms step_avg:245.40ms step:4250/5535 train_time:1042978ms step_avg:245.41ms step:4250/5535 val_loss:2.998751 train_time:1043228ms step_avg:245.47ms step:4251/5535 train_time:1043251ms step_avg:245.41ms step:4252/5535 train_time:1043482ms step_avg:245.41ms step:4253/5535 train_time:1043739ms step_avg:245.41ms step:4254/5535 train_time:1043988ms step_avg:245.41ms step:4255/5535 train_time:1044242ms step_avg:245.42ms step:4256/5535 train_time:1044492ms step_avg:245.42ms step:4257/5535 train_time:1044770ms step_avg:245.42ms step:4258/5535 train_time:1045003ms step_avg:245.42ms step:4259/5535 train_time:1045249ms step_avg:245.42ms step:4260/5535 train_time:1045500ms step_avg:245.42ms step:4261/5535 train_time:1045751ms step_avg:245.42ms step:4262/5535 train_time:1046003ms step_avg:245.43ms step:4263/5535 train_time:1046251ms step_avg:245.43ms step:4264/5535 train_time:1046504ms step_avg:245.43ms step:4265/5535 train_time:1046759ms step_avg:245.43ms step:4266/5535 train_time:1047010ms step_avg:245.43ms step:4267/5535 train_time:1047265ms step_avg:245.43ms step:4268/5535 train_time:1047510ms step_avg:245.43ms step:4269/5535 train_time:1047764ms step_avg:245.44ms step:4270/5535 train_time:1048021ms step_avg:245.44ms step:4271/5535 train_time:1048277ms step_avg:245.44ms step:4272/5535 train_time:1048528ms step_avg:245.44ms step:4273/5535 train_time:1048780ms step_avg:245.44ms step:4274/5535 train_time:1049035ms step_avg:245.45ms step:4275/5535 train_time:1049285ms step_avg:245.45ms step:4276/5535 train_time:1049532ms step_avg:245.45ms step:4277/5535 train_time:1049783ms step_avg:245.45ms step:4278/5535 train_time:1050031ms step_avg:245.45ms step:4279/5535 train_time:1050281ms step_avg:245.45ms step:4280/5535 train_time:1050528ms step_avg:245.45ms step:4281/5535 train_time:1050777ms step_avg:245.45ms step:4282/5535 train_time:1051033ms step_avg:245.45ms step:4283/5535 train_time:1051283ms step_avg:245.45ms step:4284/5535 train_time:1051536ms step_avg:245.46ms step:4285/5535 train_time:1051816ms step_avg:245.46ms step:4286/5535 train_time:1052042ms step_avg:245.46ms step:4287/5535 train_time:1052292ms step_avg:245.46ms step:4288/5535 train_time:1052544ms step_avg:245.46ms step:4289/5535 train_time:1052803ms step_avg:245.47ms step:4290/5535 train_time:1053057ms step_avg:245.47ms step:4291/5535 train_time:1053306ms step_avg:245.47ms step:4292/5535 train_time:1053554ms step_avg:245.47ms step:4293/5535 train_time:1053806ms step_avg:245.47ms step:4294/5535 train_time:1054055ms step_avg:245.47ms step:4295/5535 train_time:1054304ms step_avg:245.47ms step:4296/5535 train_time:1054557ms step_avg:245.47ms step:4297/5535 train_time:1054805ms step_avg:245.47ms step:4298/5535 train_time:1055056ms step_avg:245.48ms step:4299/5535 train_time:1055305ms step_avg:245.48ms step:4300/5535 train_time:1055559ms step_avg:245.48ms step:4301/5535 train_time:1055808ms step_avg:245.48ms step:4302/5535 train_time:1056063ms step_avg:245.48ms step:4303/5535 train_time:1056315ms step_avg:245.48ms step:4304/5535 train_time:1056564ms step_avg:245.48ms step:4305/5535 train_time:1056813ms step_avg:245.48ms step:4306/5535 train_time:1057063ms step_avg:245.49ms step:4307/5535 train_time:1057314ms step_avg:245.49ms step:4308/5535 train_time:1057568ms step_avg:245.49ms step:4309/5535 train_time:1057818ms step_avg:245.49ms step:4310/5535 train_time:1058068ms step_avg:245.49ms step:4311/5535 train_time:1058316ms step_avg:245.49ms step:4312/5535 train_time:1058574ms step_avg:245.49ms step:4313/5535 train_time:1058858ms step_avg:245.50ms step:4314/5535 train_time:1059085ms step_avg:245.50ms step:4315/5535 train_time:1059337ms step_avg:245.50ms step:4316/5535 train_time:1059589ms step_avg:245.50ms step:4317/5535 train_time:1059839ms step_avg:245.50ms step:4318/5535 train_time:1060091ms step_avg:245.51ms step:4319/5535 train_time:1060345ms step_avg:245.51ms step:4320/5535 train_time:1060598ms step_avg:245.51ms step:4321/5535 train_time:1060849ms step_avg:245.51ms step:4322/5535 train_time:1061101ms step_avg:245.51ms step:4323/5535 train_time:1061353ms step_avg:245.51ms step:4324/5535 train_time:1061611ms step_avg:245.52ms step:4325/5535 train_time:1061874ms step_avg:245.52ms step:4326/5535 train_time:1062125ms step_avg:245.52ms step:4327/5535 train_time:1062375ms step_avg:245.52ms step:4328/5535 train_time:1062624ms step_avg:245.52ms step:4329/5535 train_time:1062876ms step_avg:245.52ms step:4330/5535 train_time:1063132ms step_avg:245.53ms step:4331/5535 train_time:1063384ms step_avg:245.53ms step:4332/5535 train_time:1063635ms step_avg:245.53ms step:4333/5535 train_time:1063885ms step_avg:245.53ms step:4334/5535 train_time:1064138ms step_avg:245.53ms step:4335/5535 train_time:1064387ms step_avg:245.53ms step:4336/5535 train_time:1064633ms step_avg:245.53ms step:4337/5535 train_time:1064893ms step_avg:245.54ms step:4338/5535 train_time:1065142ms step_avg:245.54ms step:4339/5535 train_time:1065391ms step_avg:245.54ms step:4340/5535 train_time:1065638ms step_avg:245.54ms step:4341/5535 train_time:1065919ms step_avg:245.55ms step:4342/5535 train_time:1066149ms step_avg:245.54ms step:4343/5535 train_time:1066396ms step_avg:245.54ms step:4344/5535 train_time:1066645ms step_avg:245.54ms step:4345/5535 train_time:1066899ms step_avg:245.55ms step:4346/5535 train_time:1067147ms step_avg:245.55ms step:4347/5535 train_time:1067397ms step_avg:245.55ms step:4348/5535 train_time:1067652ms step_avg:245.55ms step:4349/5535 train_time:1067905ms step_avg:245.55ms step:4350/5535 train_time:1068164ms step_avg:245.55ms step:4351/5535 train_time:1068419ms step_avg:245.56ms step:4352/5535 train_time:1068668ms step_avg:245.56ms step:4353/5535 train_time:1068921ms step_avg:245.56ms step:4354/5535 train_time:1069169ms step_avg:245.56ms step:4355/5535 train_time:1069422ms step_avg:245.56ms step:4356/5535 train_time:1069683ms step_avg:245.57ms step:4357/5535 train_time:1069931ms step_avg:245.57ms step:4358/5535 train_time:1070184ms step_avg:245.57ms step:4359/5535 train_time:1070437ms step_avg:245.57ms step:4360/5535 train_time:1070691ms step_avg:245.57ms step:4361/5535 train_time:1070941ms step_avg:245.57ms step:4362/5535 train_time:1071188ms step_avg:245.57ms step:4363/5535 train_time:1071440ms step_avg:245.57ms step:4364/5535 train_time:1071694ms step_avg:245.58ms step:4365/5535 train_time:1071944ms step_avg:245.58ms step:4366/5535 train_time:1072195ms step_avg:245.58ms step:4367/5535 train_time:1072447ms step_avg:245.58ms step:4368/5535 train_time:1072702ms step_avg:245.58ms step:4369/5535 train_time:1072981ms step_avg:245.59ms step:4370/5535 train_time:1073209ms step_avg:245.59ms step:4371/5535 train_time:1073458ms step_avg:245.59ms step:4372/5535 train_time:1073705ms step_avg:245.59ms step:4373/5535 train_time:1073958ms step_avg:245.59ms step:4374/5535 train_time:1074213ms step_avg:245.59ms step:4375/5535 train_time:1074461ms step_avg:245.59ms step:4375/5535 val_loss:2.989622 train_time:1074719ms step_avg:245.65ms step:4376/5535 train_time:1074740ms step_avg:245.60ms step:4377/5535 train_time:1074967ms step_avg:245.59ms step:4378/5535 train_time:1075234ms step_avg:245.60ms step:4379/5535 train_time:1075481ms step_avg:245.60ms step:4380/5535 train_time:1075733ms step_avg:245.60ms step:4381/5535 train_time:1075980ms step_avg:245.60ms step:4382/5535 train_time:1076234ms step_avg:245.60ms step:4383/5535 train_time:1076483ms step_avg:245.60ms step:4384/5535 train_time:1076733ms step_avg:245.61ms step:4385/5535 train_time:1076988ms step_avg:245.61ms step:4386/5535 train_time:1077259ms step_avg:245.61ms step:4387/5535 train_time:1077506ms step_avg:245.61ms step:4388/5535 train_time:1077757ms step_avg:245.61ms step:4389/5535 train_time:1078009ms step_avg:245.62ms step:4390/5535 train_time:1078261ms step_avg:245.62ms step:4391/5535 train_time:1078511ms step_avg:245.62ms step:4392/5535 train_time:1078761ms step_avg:245.62ms step:4393/5535 train_time:1079012ms step_avg:245.62ms step:4394/5535 train_time:1079264ms step_avg:245.62ms step:4395/5535 train_time:1079513ms step_avg:245.62ms step:4396/5535 train_time:1079762ms step_avg:245.62ms step:4397/5535 train_time:1080044ms step_avg:245.63ms step:4398/5535 train_time:1080276ms step_avg:245.63ms step:4399/5535 train_time:1080528ms step_avg:245.63ms step:4400/5535 train_time:1080778ms step_avg:245.63ms step:4401/5535 train_time:1081027ms step_avg:245.63ms step:4402/5535 train_time:1081280ms step_avg:245.63ms step:4403/5535 train_time:1081535ms step_avg:245.64ms step:4404/5535 train_time:1081784ms step_avg:245.64ms step:4405/5535 train_time:1082035ms step_avg:245.64ms step:4406/5535 train_time:1082283ms step_avg:245.64ms step:4407/5535 train_time:1082537ms step_avg:245.64ms step:4408/5535 train_time:1082790ms step_avg:245.64ms step:4409/5535 train_time:1083038ms step_avg:245.64ms step:4410/5535 train_time:1083293ms step_avg:245.64ms step:4411/5535 train_time:1083547ms step_avg:245.65ms step:4412/5535 train_time:1083795ms step_avg:245.65ms step:4413/5535 train_time:1084046ms step_avg:245.65ms step:4414/5535 train_time:1084300ms step_avg:245.65ms step:4415/5535 train_time:1084556ms step_avg:245.65ms step:4416/5535 train_time:1084806ms step_avg:245.65ms step:4417/5535 train_time:1085055ms step_avg:245.65ms step:4418/5535 train_time:1085306ms step_avg:245.66ms step:4419/5535 train_time:1085558ms step_avg:245.66ms step:4420/5535 train_time:1085814ms step_avg:245.66ms step:4421/5535 train_time:1086068ms step_avg:245.66ms step:4422/5535 train_time:1086314ms step_avg:245.66ms step:4423/5535 train_time:1086564ms step_avg:245.66ms step:4424/5535 train_time:1086817ms step_avg:245.66ms step:4425/5535 train_time:1087097ms step_avg:245.67ms step:4426/5535 train_time:1087325ms step_avg:245.67ms step:4427/5535 train_time:1087578ms step_avg:245.67ms step:4428/5535 train_time:1087833ms step_avg:245.67ms step:4429/5535 train_time:1088081ms step_avg:245.67ms step:4430/5535 train_time:1088339ms step_avg:245.67ms step:4431/5535 train_time:1088589ms step_avg:245.68ms step:4432/5535 train_time:1088838ms step_avg:245.68ms step:4433/5535 train_time:1089088ms step_avg:245.68ms step:4434/5535 train_time:1089336ms step_avg:245.68ms step:4435/5535 train_time:1089589ms step_avg:245.68ms step:4436/5535 train_time:1089840ms step_avg:245.68ms step:4437/5535 train_time:1090099ms step_avg:245.68ms step:4438/5535 train_time:1090364ms step_avg:245.69ms step:4439/5535 train_time:1090615ms step_avg:245.69ms step:4440/5535 train_time:1090865ms step_avg:245.69ms step:4441/5535 train_time:1091119ms step_avg:245.69ms step:4442/5535 train_time:1091371ms step_avg:245.69ms step:4443/5535 train_time:1091621ms step_avg:245.69ms step:4444/5535 train_time:1091873ms step_avg:245.70ms step:4445/5535 train_time:1092125ms step_avg:245.70ms step:4446/5535 train_time:1092376ms step_avg:245.70ms step:4447/5535 train_time:1092626ms step_avg:245.70ms step:4448/5535 train_time:1092876ms step_avg:245.70ms step:4449/5535 train_time:1093125ms step_avg:245.70ms step:4450/5535 train_time:1093382ms step_avg:245.70ms step:4451/5535 train_time:1093634ms step_avg:245.71ms step:4452/5535 train_time:1093886ms step_avg:245.71ms step:4453/5535 train_time:1094167ms step_avg:245.71ms step:4454/5535 train_time:1094395ms step_avg:245.71ms step:4455/5535 train_time:1094651ms step_avg:245.71ms step:4456/5535 train_time:1094903ms step_avg:245.71ms step:4457/5535 train_time:1095150ms step_avg:245.71ms step:4458/5535 train_time:1095415ms step_avg:245.72ms step:4459/5535 train_time:1095664ms step_avg:245.72ms step:4460/5535 train_time:1095920ms step_avg:245.72ms step:4461/5535 train_time:1096180ms step_avg:245.73ms step:4462/5535 train_time:1096433ms step_avg:245.73ms step:4463/5535 train_time:1096683ms step_avg:245.73ms step:4464/5535 train_time:1096932ms step_avg:245.73ms step:4465/5535 train_time:1097182ms step_avg:245.73ms step:4466/5535 train_time:1097436ms step_avg:245.73ms step:4467/5535 train_time:1097686ms step_avg:245.73ms step:4468/5535 train_time:1097934ms step_avg:245.73ms step:4469/5535 train_time:1098181ms step_avg:245.73ms step:4470/5535 train_time:1098442ms step_avg:245.74ms step:4471/5535 train_time:1098689ms step_avg:245.74ms step:4472/5535 train_time:1098951ms step_avg:245.74ms step:4473/5535 train_time:1099198ms step_avg:245.74ms step:4474/5535 train_time:1099456ms step_avg:245.74ms step:4475/5535 train_time:1099712ms step_avg:245.75ms step:4476/5535 train_time:1099959ms step_avg:245.75ms step:4477/5535 train_time:1100210ms step_avg:245.75ms step:4478/5535 train_time:1100461ms step_avg:245.75ms step:4479/5535 train_time:1100716ms step_avg:245.75ms step:4480/5535 train_time:1100966ms step_avg:245.75ms step:4481/5535 train_time:1101244ms step_avg:245.76ms step:4482/5535 train_time:1101472ms step_avg:245.75ms step:4483/5535 train_time:1101725ms step_avg:245.76ms step:4484/5535 train_time:1101973ms step_avg:245.76ms step:4485/5535 train_time:1102240ms step_avg:245.76ms step:4486/5535 train_time:1102494ms step_avg:245.76ms step:4487/5535 train_time:1102749ms step_avg:245.77ms step:4488/5535 train_time:1103004ms step_avg:245.77ms step:4489/5535 train_time:1103256ms step_avg:245.77ms step:4490/5535 train_time:1103506ms step_avg:245.77ms step:4491/5535 train_time:1103762ms step_avg:245.77ms step:4492/5535 train_time:1104008ms step_avg:245.77ms step:4493/5535 train_time:1104256ms step_avg:245.77ms step:4494/5535 train_time:1104507ms step_avg:245.77ms step:4495/5535 train_time:1104755ms step_avg:245.77ms step:4496/5535 train_time:1105010ms step_avg:245.78ms step:4497/5535 train_time:1105255ms step_avg:245.78ms step:4498/5535 train_time:1105506ms step_avg:245.78ms step:4499/5535 train_time:1105762ms step_avg:245.78ms step:4500/5535 train_time:1106013ms step_avg:245.78ms step:4500/5535 val_loss:2.981176 train_time:1106268ms step_avg:245.84ms step:4501/5535 train_time:1106289ms step_avg:245.79ms step:4502/5535 train_time:1106520ms step_avg:245.78ms step:4503/5535 train_time:1106781ms step_avg:245.79ms step:4504/5535 train_time:1107030ms step_avg:245.79ms step:4505/5535 train_time:1107280ms step_avg:245.79ms step:4506/5535 train_time:1107528ms step_avg:245.79ms step:4507/5535 train_time:1107779ms step_avg:245.79ms step:4508/5535 train_time:1108034ms step_avg:245.79ms step:4509/5535 train_time:1108317ms step_avg:245.80ms step:4510/5535 train_time:1108540ms step_avg:245.80ms step:4511/5535 train_time:1108788ms step_avg:245.80ms step:4512/5535 train_time:1109041ms step_avg:245.80ms step:4513/5535 train_time:1109297ms step_avg:245.80ms step:4514/5535 train_time:1109545ms step_avg:245.80ms step:4515/5535 train_time:1109792ms step_avg:245.80ms step:4516/5535 train_time:1110045ms step_avg:245.80ms step:4517/5535 train_time:1110295ms step_avg:245.80ms step:4518/5535 train_time:1110546ms step_avg:245.80ms step:4519/5535 train_time:1110793ms step_avg:245.81ms step:4520/5535 train_time:1111043ms step_avg:245.81ms step:4521/5535 train_time:1111294ms step_avg:245.81ms step:4522/5535 train_time:1111549ms step_avg:245.81ms step:4523/5535 train_time:1111806ms step_avg:245.81ms step:4524/5535 train_time:1112058ms step_avg:245.81ms step:4525/5535 train_time:1112308ms step_avg:245.81ms step:4526/5535 train_time:1112556ms step_avg:245.81ms step:4527/5535 train_time:1112813ms step_avg:245.82ms step:4528/5535 train_time:1113061ms step_avg:245.82ms step:4529/5535 train_time:1113319ms step_avg:245.82ms step:4530/5535 train_time:1113572ms step_avg:245.82ms step:4531/5535 train_time:1113822ms step_avg:245.82ms step:4532/5535 train_time:1114084ms step_avg:245.83ms step:4533/5535 train_time:1114343ms step_avg:245.83ms step:4534/5535 train_time:1114591ms step_avg:245.83ms step:4535/5535 train_time:1114843ms step_avg:245.83ms step:4536/5535 train_time:1115091ms step_avg:245.83ms step:4537/5535 train_time:1115374ms step_avg:245.84ms step:4538/5535 train_time:1115602ms step_avg:245.84ms step:4539/5535 train_time:1115857ms step_avg:245.84ms step:4540/5535 train_time:1116105ms step_avg:245.84ms step:4541/5535 train_time:1116362ms step_avg:245.84ms step:4542/5535 train_time:1116613ms step_avg:245.84ms step:4543/5535 train_time:1116863ms step_avg:245.84ms step:4544/5535 train_time:1117114ms step_avg:245.84ms step:4545/5535 train_time:1117373ms step_avg:245.85ms step:4546/5535 train_time:1117633ms step_avg:245.85ms step:4547/5535 train_time:1117885ms step_avg:245.85ms step:4548/5535 train_time:1118138ms step_avg:245.85ms step:4549/5535 train_time:1118391ms step_avg:245.85ms step:4550/5535 train_time:1118653ms step_avg:245.86ms step:4551/5535 train_time:1118906ms step_avg:245.86ms step:4552/5535 train_time:1119154ms step_avg:245.86ms step:4553/5535 train_time:1119405ms step_avg:245.86ms step:4554/5535 train_time:1119660ms step_avg:245.86ms step:4555/5535 train_time:1119917ms step_avg:245.87ms step:4556/5535 train_time:1120172ms step_avg:245.87ms step:4557/5535 train_time:1120421ms step_avg:245.87ms step:4558/5535 train_time:1120679ms step_avg:245.87ms step:4559/5535 train_time:1120934ms step_avg:245.87ms step:4560/5535 train_time:1121190ms step_avg:245.87ms step:4561/5535 train_time:1121452ms step_avg:245.88ms step:4562/5535 train_time:1121700ms step_avg:245.88ms step:4563/5535 train_time:1121965ms step_avg:245.88ms step:4564/5535 train_time:1122216ms step_avg:245.88ms step:4565/5535 train_time:1122496ms step_avg:245.89ms step:4566/5535 train_time:1122725ms step_avg:245.89ms step:4567/5535 train_time:1122975ms step_avg:245.89ms step:4568/5535 train_time:1123226ms step_avg:245.89ms step:4569/5535 train_time:1123476ms step_avg:245.89ms step:4570/5535 train_time:1123728ms step_avg:245.89ms step:4571/5535 train_time:1123997ms step_avg:245.90ms step:4572/5535 train_time:1124249ms step_avg:245.90ms step:4573/5535 train_time:1124503ms step_avg:245.90ms step:4574/5535 train_time:1124758ms step_avg:245.90ms step:4575/5535 train_time:1125009ms step_avg:245.90ms step:4576/5535 train_time:1125263ms step_avg:245.91ms step:4577/5535 train_time:1125515ms step_avg:245.91ms step:4578/5535 train_time:1125774ms step_avg:245.91ms step:4579/5535 train_time:1126022ms step_avg:245.91ms step:4580/5535 train_time:1126275ms step_avg:245.91ms step:4581/5535 train_time:1126535ms step_avg:245.91ms step:4582/5535 train_time:1126786ms step_avg:245.92ms step:4583/5535 train_time:1127033ms step_avg:245.92ms step:4584/5535 train_time:1127283ms step_avg:245.92ms step:4585/5535 train_time:1127540ms step_avg:245.92ms step:4586/5535 train_time:1127797ms step_avg:245.92ms step:4587/5535 train_time:1128049ms step_avg:245.92ms step:4588/5535 train_time:1128305ms step_avg:245.93ms step:4589/5535 train_time:1128560ms step_avg:245.93ms step:4590/5535 train_time:1128812ms step_avg:245.93ms step:4591/5535 train_time:1129069ms step_avg:245.93ms step:4592/5535 train_time:1129316ms step_avg:245.93ms step:4593/5535 train_time:1129594ms step_avg:245.94ms step:4594/5535 train_time:1129824ms step_avg:245.93ms step:4595/5535 train_time:1130075ms step_avg:245.94ms step:4596/5535 train_time:1130337ms step_avg:245.94ms step:4597/5535 train_time:1130599ms step_avg:245.94ms step:4598/5535 train_time:1130857ms step_avg:245.95ms step:4599/5535 train_time:1131108ms step_avg:245.95ms step:4600/5535 train_time:1131359ms step_avg:245.95ms step:4601/5535 train_time:1131616ms step_avg:245.95ms step:4602/5535 train_time:1131870ms step_avg:245.95ms step:4603/5535 train_time:1132120ms step_avg:245.95ms step:4604/5535 train_time:1132371ms step_avg:245.95ms step:4605/5535 train_time:1132625ms step_avg:245.96ms step:4606/5535 train_time:1132875ms step_avg:245.96ms step:4607/5535 train_time:1133125ms step_avg:245.96ms step:4608/5535 train_time:1133386ms step_avg:245.96ms step:4609/5535 train_time:1133637ms step_avg:245.96ms step:4610/5535 train_time:1133886ms step_avg:245.96ms step:4611/5535 train_time:1134136ms step_avg:245.96ms step:4612/5535 train_time:1134394ms step_avg:245.97ms step:4613/5535 train_time:1134643ms step_avg:245.97ms step:4614/5535 train_time:1134906ms step_avg:245.97ms step:4615/5535 train_time:1135159ms step_avg:245.97ms step:4616/5535 train_time:1135412ms step_avg:245.97ms step:4617/5535 train_time:1135667ms step_avg:245.98ms step:4618/5535 train_time:1135916ms step_avg:245.98ms step:4619/5535 train_time:1136167ms step_avg:245.98ms step:4620/5535 train_time:1136417ms step_avg:245.98ms step:4621/5535 train_time:1136698ms step_avg:245.99ms step:4622/5535 train_time:1136928ms step_avg:245.98ms step:4623/5535 train_time:1137179ms step_avg:245.98ms step:4624/5535 train_time:1137428ms step_avg:245.98ms step:4625/5535 train_time:1137679ms step_avg:245.98ms step:4625/5535 val_loss:2.972050 train_time:1137937ms step_avg:246.04ms step:4626/5535 train_time:1137958ms step_avg:245.99ms step:4627/5535 train_time:1138186ms step_avg:245.99ms step:4628/5535 train_time:1138443ms step_avg:245.99ms step:4629/5535 train_time:1138693ms step_avg:245.99ms step:4630/5535 train_time:1138937ms step_avg:245.99ms step:4631/5535 train_time:1139191ms step_avg:245.99ms step:4632/5535 train_time:1139451ms step_avg:246.00ms step:4633/5535 train_time:1139706ms step_avg:246.00ms step:4634/5535 train_time:1139955ms step_avg:246.00ms step:4635/5535 train_time:1140204ms step_avg:246.00ms step:4636/5535 train_time:1140466ms step_avg:246.00ms step:4637/5535 train_time:1140727ms step_avg:246.01ms step:4638/5535 train_time:1140973ms step_avg:246.01ms step:4639/5535 train_time:1141224ms step_avg:246.01ms step:4640/5535 train_time:1141475ms step_avg:246.01ms step:4641/5535 train_time:1141732ms step_avg:246.01ms step:4642/5535 train_time:1141983ms step_avg:246.01ms step:4643/5535 train_time:1142235ms step_avg:246.01ms step:4644/5535 train_time:1142496ms step_avg:246.02ms step:4645/5535 train_time:1142747ms step_avg:246.02ms step:4646/5535 train_time:1143004ms step_avg:246.02ms step:4647/5535 train_time:1143257ms step_avg:246.02ms step:4648/5535 train_time:1143518ms step_avg:246.02ms step:4649/5535 train_time:1143797ms step_avg:246.03ms step:4650/5535 train_time:1144031ms step_avg:246.03ms step:4651/5535 train_time:1144288ms step_avg:246.03ms step:4652/5535 train_time:1144538ms step_avg:246.03ms step:4653/5535 train_time:1144788ms step_avg:246.03ms step:4654/5535 train_time:1145044ms step_avg:246.03ms step:4655/5535 train_time:1145304ms step_avg:246.04ms step:4656/5535 train_time:1145558ms step_avg:246.04ms step:4657/5535 train_time:1145807ms step_avg:246.04ms step:4658/5535 train_time:1146060ms step_avg:246.04ms step:4659/5535 train_time:1146312ms step_avg:246.04ms step:4660/5535 train_time:1146566ms step_avg:246.04ms step:4661/5535 train_time:1146817ms step_avg:246.05ms step:4662/5535 train_time:1147074ms step_avg:246.05ms step:4663/5535 train_time:1147324ms step_avg:246.05ms step:4664/5535 train_time:1147582ms step_avg:246.05ms step:4665/5535 train_time:1147835ms step_avg:246.05ms step:4666/5535 train_time:1148094ms step_avg:246.06ms step:4667/5535 train_time:1148345ms step_avg:246.06ms step:4668/5535 train_time:1148601ms step_avg:246.06ms step:4669/5535 train_time:1148856ms step_avg:246.06ms step:4670/5535 train_time:1149111ms step_avg:246.06ms step:4671/5535 train_time:1149359ms step_avg:246.06ms step:4672/5535 train_time:1149611ms step_avg:246.06ms step:4673/5535 train_time:1149870ms step_avg:246.07ms step:4674/5535 train_time:1150125ms step_avg:246.07ms step:4675/5535 train_time:1150383ms step_avg:246.07ms step:4676/5535 train_time:1150636ms step_avg:246.07ms step:4677/5535 train_time:1150919ms step_avg:246.08ms step:4678/5535 train_time:1151152ms step_avg:246.08ms step:4679/5535 train_time:1151413ms step_avg:246.08ms step:4680/5535 train_time:1151672ms step_avg:246.08ms step:4681/5535 train_time:1151922ms step_avg:246.08ms step:4682/5535 train_time:1152171ms step_avg:246.09ms step:4683/5535 train_time:1152427ms step_avg:246.09ms step:4684/5535 train_time:1152678ms step_avg:246.09ms step:4685/5535 train_time:1152934ms step_avg:246.09ms step:4686/5535 train_time:1153191ms step_avg:246.09ms step:4687/5535 train_time:1153439ms step_avg:246.09ms step:4688/5535 train_time:1153692ms step_avg:246.09ms step:4689/5535 train_time:1153953ms step_avg:246.10ms step:4690/5535 train_time:1154204ms step_avg:246.10ms step:4691/5535 train_time:1154455ms step_avg:246.10ms step:4692/5535 train_time:1154717ms step_avg:246.10ms step:4693/5535 train_time:1154981ms step_avg:246.11ms step:4694/5535 train_time:1155233ms step_avg:246.11ms step:4695/5535 train_time:1155486ms step_avg:246.11ms step:4696/5535 train_time:1155747ms step_avg:246.11ms step:4697/5535 train_time:1155998ms step_avg:246.11ms step:4698/5535 train_time:1156248ms step_avg:246.11ms step:4699/5535 train_time:1156500ms step_avg:246.12ms step:4700/5535 train_time:1156753ms step_avg:246.12ms step:4701/5535 train_time:1157005ms step_avg:246.12ms step:4702/5535 train_time:1157264ms step_avg:246.12ms step:4703/5535 train_time:1157512ms step_avg:246.12ms step:4704/5535 train_time:1157763ms step_avg:246.12ms step:4705/5535 train_time:1158042ms step_avg:246.13ms step:4706/5535 train_time:1158272ms step_avg:246.13ms step:4707/5535 train_time:1158539ms step_avg:246.13ms step:4708/5535 train_time:1158791ms step_avg:246.13ms step:4709/5535 train_time:1159047ms step_avg:246.13ms step:4710/5535 train_time:1159304ms step_avg:246.14ms step:4711/5535 train_time:1159556ms step_avg:246.14ms step:4712/5535 train_time:1159808ms step_avg:246.14ms step:4713/5535 train_time:1160061ms step_avg:246.14ms step:4714/5535 train_time:1160324ms step_avg:246.14ms step:4715/5535 train_time:1160579ms step_avg:246.15ms step:4716/5535 train_time:1160833ms step_avg:246.15ms step:4717/5535 train_time:1161084ms step_avg:246.15ms step:4718/5535 train_time:1161342ms step_avg:246.15ms step:4719/5535 train_time:1161612ms step_avg:246.16ms step:4720/5535 train_time:1161861ms step_avg:246.16ms step:4721/5535 train_time:1162110ms step_avg:246.16ms step:4722/5535 train_time:1162361ms step_avg:246.16ms step:4723/5535 train_time:1162622ms step_avg:246.16ms step:4724/5535 train_time:1162871ms step_avg:246.16ms step:4725/5535 train_time:1163122ms step_avg:246.16ms step:4726/5535 train_time:1163379ms step_avg:246.17ms step:4727/5535 train_time:1163634ms step_avg:246.17ms step:4728/5535 train_time:1163893ms step_avg:246.17ms step:4729/5535 train_time:1164141ms step_avg:246.17ms step:4730/5535 train_time:1164395ms step_avg:246.17ms step:4731/5535 train_time:1164646ms step_avg:246.17ms step:4732/5535 train_time:1164898ms step_avg:246.17ms step:4733/5535 train_time:1165182ms step_avg:246.18ms step:4734/5535 train_time:1165412ms step_avg:246.18ms step:4735/5535 train_time:1165665ms step_avg:246.18ms step:4736/5535 train_time:1165917ms step_avg:246.18ms step:4737/5535 train_time:1166168ms step_avg:246.18ms step:4738/5535 train_time:1166429ms step_avg:246.19ms step:4739/5535 train_time:1166685ms step_avg:246.19ms step:4740/5535 train_time:1166938ms step_avg:246.19ms step:4741/5535 train_time:1167186ms step_avg:246.19ms step:4742/5535 train_time:1167449ms step_avg:246.19ms step:4743/5535 train_time:1167700ms step_avg:246.19ms step:4744/5535 train_time:1167961ms step_avg:246.20ms step:4745/5535 train_time:1168213ms step_avg:246.20ms step:4746/5535 train_time:1168467ms step_avg:246.20ms step:4747/5535 train_time:1168732ms step_avg:246.20ms step:4748/5535 train_time:1168983ms step_avg:246.21ms step:4749/5535 train_time:1169239ms step_avg:246.21ms step:4750/5535 train_time:1169490ms step_avg:246.21ms step:4750/5535 val_loss:2.962718 train_time:1169745ms step_avg:246.26ms step:4751/5535 train_time:1169810ms step_avg:246.22ms step:4752/5535 train_time:1170035ms step_avg:246.22ms step:4753/5535 train_time:1170286ms step_avg:246.22ms step:4754/5535 train_time:1170535ms step_avg:246.22ms step:4755/5535 train_time:1170791ms step_avg:246.22ms step:4756/5535 train_time:1171048ms step_avg:246.23ms step:4757/5535 train_time:1171302ms step_avg:246.23ms step:4758/5535 train_time:1171555ms step_avg:246.23ms step:4759/5535 train_time:1171818ms step_avg:246.23ms step:4760/5535 train_time:1172068ms step_avg:246.23ms step:4761/5535 train_time:1172350ms step_avg:246.24ms step:4762/5535 train_time:1172579ms step_avg:246.24ms step:4763/5535 train_time:1172833ms step_avg:246.24ms step:4764/5535 train_time:1173087ms step_avg:246.24ms step:4765/5535 train_time:1173336ms step_avg:246.24ms step:4766/5535 train_time:1173593ms step_avg:246.24ms step:4767/5535 train_time:1173847ms step_avg:246.24ms step:4768/5535 train_time:1174107ms step_avg:246.25ms step:4769/5535 train_time:1174360ms step_avg:246.25ms step:4770/5535 train_time:1174611ms step_avg:246.25ms step:4771/5535 train_time:1174861ms step_avg:246.25ms step:4772/5535 train_time:1175113ms step_avg:246.25ms step:4773/5535 train_time:1175365ms step_avg:246.25ms step:4774/5535 train_time:1175622ms step_avg:246.26ms step:4775/5535 train_time:1175872ms step_avg:246.26ms step:4776/5535 train_time:1176127ms step_avg:246.26ms step:4777/5535 train_time:1176378ms step_avg:246.26ms step:4778/5535 train_time:1176630ms step_avg:246.26ms step:4779/5535 train_time:1176884ms step_avg:246.26ms step:4780/5535 train_time:1177139ms step_avg:246.26ms step:4781/5535 train_time:1177393ms step_avg:246.26ms step:4782/5535 train_time:1177647ms step_avg:246.27ms step:4783/5535 train_time:1177906ms step_avg:246.27ms step:4784/5535 train_time:1178155ms step_avg:246.27ms step:4785/5535 train_time:1178406ms step_avg:246.27ms step:4786/5535 train_time:1178660ms step_avg:246.27ms step:4787/5535 train_time:1178916ms step_avg:246.27ms step:4788/5535 train_time:1179173ms step_avg:246.28ms step:4789/5535 train_time:1179462ms step_avg:246.29ms step:4790/5535 train_time:1179693ms step_avg:246.28ms step:4791/5535 train_time:1179956ms step_avg:246.29ms step:4792/5535 train_time:1180204ms step_avg:246.29ms step:4793/5535 train_time:1180456ms step_avg:246.29ms step:4794/5535 train_time:1180714ms step_avg:246.29ms step:4795/5535 train_time:1180966ms step_avg:246.29ms step:4796/5535 train_time:1181215ms step_avg:246.29ms step:4797/5535 train_time:1181467ms step_avg:246.29ms step:4798/5535 train_time:1181721ms step_avg:246.29ms step:4799/5535 train_time:1181974ms step_avg:246.30ms step:4800/5535 train_time:1182231ms step_avg:246.30ms step:4801/5535 train_time:1182483ms step_avg:246.30ms step:4802/5535 train_time:1182736ms step_avg:246.30ms step:4803/5535 train_time:1182990ms step_avg:246.30ms step:4804/5535 train_time:1183246ms step_avg:246.30ms step:4805/5535 train_time:1183509ms step_avg:246.31ms step:4806/5535 train_time:1183770ms step_avg:246.31ms step:4807/5535 train_time:1184025ms step_avg:246.31ms step:4808/5535 train_time:1184276ms step_avg:246.31ms step:4809/5535 train_time:1184532ms step_avg:246.32ms step:4810/5535 train_time:1184784ms step_avg:246.32ms step:4811/5535 train_time:1185040ms step_avg:246.32ms step:4812/5535 train_time:1185289ms step_avg:246.32ms step:4813/5535 train_time:1185543ms step_avg:246.32ms step:4814/5535 train_time:1185794ms step_avg:246.32ms step:4815/5535 train_time:1186052ms step_avg:246.32ms step:4816/5535 train_time:1186316ms step_avg:246.33ms step:4817/5535 train_time:1186595ms step_avg:246.33ms step:4818/5535 train_time:1186824ms step_avg:246.33ms step:4819/5535 train_time:1187076ms step_avg:246.33ms step:4820/5535 train_time:1187326ms step_avg:246.33ms step:4821/5535 train_time:1187585ms step_avg:246.34ms step:4822/5535 train_time:1187851ms step_avg:246.34ms step:4823/5535 train_time:1188104ms step_avg:246.34ms step:4824/5535 train_time:1188362ms step_avg:246.34ms step:4825/5535 train_time:1188615ms step_avg:246.35ms step:4826/5535 train_time:1188875ms step_avg:246.35ms step:4827/5535 train_time:1189126ms step_avg:246.35ms step:4828/5535 train_time:1189385ms step_avg:246.35ms step:4829/5535 train_time:1189638ms step_avg:246.35ms step:4830/5535 train_time:1189894ms step_avg:246.35ms step:4831/5535 train_time:1190149ms step_avg:246.36ms step:4832/5535 train_time:1190410ms step_avg:246.36ms step:4833/5535 train_time:1190665ms step_avg:246.36ms step:4834/5535 train_time:1190915ms step_avg:246.36ms step:4835/5535 train_time:1191169ms step_avg:246.36ms step:4836/5535 train_time:1191433ms step_avg:246.37ms step:4837/5535 train_time:1191693ms step_avg:246.37ms step:4838/5535 train_time:1191951ms step_avg:246.37ms step:4839/5535 train_time:1192201ms step_avg:246.37ms step:4840/5535 train_time:1192463ms step_avg:246.38ms step:4841/5535 train_time:1192719ms step_avg:246.38ms step:4842/5535 train_time:1192971ms step_avg:246.38ms step:4843/5535 train_time:1193231ms step_avg:246.38ms step:4844/5535 train_time:1193481ms step_avg:246.38ms step:4845/5535 train_time:1193769ms step_avg:246.39ms step:4846/5535 train_time:1194001ms step_avg:246.39ms step:4847/5535 train_time:1194260ms step_avg:246.39ms step:4848/5535 train_time:1194518ms step_avg:246.39ms step:4849/5535 train_time:1194771ms step_avg:246.40ms step:4850/5535 train_time:1195031ms step_avg:246.40ms step:4851/5535 train_time:1195282ms step_avg:246.40ms step:4852/5535 train_time:1195543ms step_avg:246.40ms step:4853/5535 train_time:1195811ms step_avg:246.41ms step:4854/5535 train_time:1196063ms step_avg:246.41ms step:4855/5535 train_time:1196314ms step_avg:246.41ms step:4856/5535 train_time:1196571ms step_avg:246.41ms step:4857/5535 train_time:1196825ms step_avg:246.41ms step:4858/5535 train_time:1197079ms step_avg:246.41ms step:4859/5535 train_time:1197338ms step_avg:246.42ms step:4860/5535 train_time:1197604ms step_avg:246.42ms step:4861/5535 train_time:1197862ms step_avg:246.42ms step:4862/5535 train_time:1198114ms step_avg:246.42ms step:4863/5535 train_time:1198372ms step_avg:246.43ms step:4864/5535 train_time:1198629ms step_avg:246.43ms step:4865/5535 train_time:1198885ms step_avg:246.43ms step:4866/5535 train_time:1199135ms step_avg:246.43ms step:4867/5535 train_time:1199393ms step_avg:246.43ms step:4868/5535 train_time:1199649ms step_avg:246.44ms step:4869/5535 train_time:1199904ms step_avg:246.44ms step:4870/5535 train_time:1200165ms step_avg:246.44ms step:4871/5535 train_time:1200422ms step_avg:246.44ms step:4872/5535 train_time:1200674ms step_avg:246.44ms step:4873/5535 train_time:1200967ms step_avg:246.45ms step:4874/5535 train_time:1201197ms step_avg:246.45ms step:4875/5535 train_time:1201453ms step_avg:246.45ms step:4875/5535 val_loss:2.953963 train_time:1201706ms step_avg:246.50ms step:4876/5535 train_time:1201727ms step_avg:246.46ms step:4877/5535 train_time:1201962ms step_avg:246.46ms step:4878/5535 train_time:1202230ms step_avg:246.46ms step:4879/5535 train_time:1202476ms step_avg:246.46ms step:4880/5535 train_time:1202725ms step_avg:246.46ms step:4881/5535 train_time:1202972ms step_avg:246.46ms step:4882/5535 train_time:1203229ms step_avg:246.46ms step:4883/5535 train_time:1203483ms step_avg:246.46ms step:4884/5535 train_time:1203734ms step_avg:246.46ms step:4885/5535 train_time:1203987ms step_avg:246.47ms step:4886/5535 train_time:1204245ms step_avg:246.47ms step:4887/5535 train_time:1204503ms step_avg:246.47ms step:4888/5535 train_time:1204755ms step_avg:246.47ms step:4889/5535 train_time:1205007ms step_avg:246.47ms step:4890/5535 train_time:1205264ms step_avg:246.48ms step:4891/5535 train_time:1205517ms step_avg:246.48ms step:4892/5535 train_time:1205776ms step_avg:246.48ms step:4893/5535 train_time:1206027ms step_avg:246.48ms step:4894/5535 train_time:1206281ms step_avg:246.48ms step:4895/5535 train_time:1206534ms step_avg:246.48ms step:4896/5535 train_time:1206795ms step_avg:246.49ms step:4897/5535 train_time:1207048ms step_avg:246.49ms step:4898/5535 train_time:1207319ms step_avg:246.49ms step:4899/5535 train_time:1207569ms step_avg:246.49ms step:4900/5535 train_time:1207830ms step_avg:246.50ms step:4901/5535 train_time:1208121ms step_avg:246.51ms step:4902/5535 train_time:1208349ms step_avg:246.50ms step:4903/5535 train_time:1208604ms step_avg:246.50ms step:4904/5535 train_time:1208862ms step_avg:246.51ms step:4905/5535 train_time:1209115ms step_avg:246.51ms step:4906/5535 train_time:1209370ms step_avg:246.51ms step:4907/5535 train_time:1209628ms step_avg:246.51ms step:4908/5535 train_time:1209881ms step_avg:246.51ms step:4909/5535 train_time:1210133ms step_avg:246.51ms step:4910/5535 train_time:1210390ms step_avg:246.52ms step:4911/5535 train_time:1210646ms step_avg:246.52ms step:4912/5535 train_time:1210900ms step_avg:246.52ms step:4913/5535 train_time:1211154ms step_avg:246.52ms step:4914/5535 train_time:1211409ms step_avg:246.52ms step:4915/5535 train_time:1211669ms step_avg:246.52ms step:4916/5535 train_time:1211920ms step_avg:246.53ms step:4917/5535 train_time:1212180ms step_avg:246.53ms step:4918/5535 train_time:1212433ms step_avg:246.53ms step:4919/5535 train_time:1212694ms step_avg:246.53ms step:4920/5535 train_time:1212945ms step_avg:246.53ms step:4921/5535 train_time:1213196ms step_avg:246.53ms step:4922/5535 train_time:1213447ms step_avg:246.54ms step:4923/5535 train_time:1213703ms step_avg:246.54ms step:4924/5535 train_time:1213954ms step_avg:246.54ms step:4925/5535 train_time:1214204ms step_avg:246.54ms step:4926/5535 train_time:1214457ms step_avg:246.54ms step:4927/5535 train_time:1214717ms step_avg:246.54ms step:4928/5535 train_time:1214973ms step_avg:246.54ms step:4929/5535 train_time:1215254ms step_avg:246.55ms step:4930/5535 train_time:1215493ms step_avg:246.55ms step:4931/5535 train_time:1215751ms step_avg:246.55ms step:4932/5535 train_time:1216006ms step_avg:246.55ms step:4933/5535 train_time:1216267ms step_avg:246.56ms step:4934/5535 train_time:1216538ms step_avg:246.56ms step:4935/5535 train_time:1216788ms step_avg:246.56ms step:4936/5535 train_time:1217045ms step_avg:246.56ms step:4937/5535 train_time:1217299ms step_avg:246.57ms step:4938/5535 train_time:1217558ms step_avg:246.57ms step:4939/5535 train_time:1217812ms step_avg:246.57ms step:4940/5535 train_time:1218070ms step_avg:246.57ms step:4941/5535 train_time:1218323ms step_avg:246.57ms step:4942/5535 train_time:1218578ms step_avg:246.58ms step:4943/5535 train_time:1218833ms step_avg:246.58ms step:4944/5535 train_time:1219086ms step_avg:246.58ms step:4945/5535 train_time:1219343ms step_avg:246.58ms step:4946/5535 train_time:1219598ms step_avg:246.58ms step:4947/5535 train_time:1219853ms step_avg:246.58ms step:4948/5535 train_time:1220109ms step_avg:246.59ms step:4949/5535 train_time:1220366ms step_avg:246.59ms step:4950/5535 train_time:1220626ms step_avg:246.59ms step:4951/5535 train_time:1220876ms step_avg:246.59ms step:4952/5535 train_time:1221128ms step_avg:246.59ms step:4953/5535 train_time:1221383ms step_avg:246.59ms step:4954/5535 train_time:1221643ms step_avg:246.60ms step:4955/5535 train_time:1221913ms step_avg:246.60ms step:4956/5535 train_time:1222162ms step_avg:246.60ms step:4957/5535 train_time:1222443ms step_avg:246.61ms step:4958/5535 train_time:1222674ms step_avg:246.61ms step:4959/5535 train_time:1222934ms step_avg:246.61ms step:4960/5535 train_time:1223190ms step_avg:246.61ms step:4961/5535 train_time:1223440ms step_avg:246.61ms step:4962/5535 train_time:1223695ms step_avg:246.61ms step:4963/5535 train_time:1223951ms step_avg:246.62ms step:4964/5535 train_time:1224202ms step_avg:246.62ms step:4965/5535 train_time:1224470ms step_avg:246.62ms step:4966/5535 train_time:1224737ms step_avg:246.62ms step:4967/5535 train_time:1225002ms step_avg:246.63ms step:4968/5535 train_time:1225255ms step_avg:246.63ms step:4969/5535 train_time:1225513ms step_avg:246.63ms step:4970/5535 train_time:1225769ms step_avg:246.63ms step:4971/5535 train_time:1226028ms step_avg:246.64ms step:4972/5535 train_time:1226287ms step_avg:246.64ms step:4973/5535 train_time:1226555ms step_avg:246.64ms step:4974/5535 train_time:1226823ms step_avg:246.65ms step:4975/5535 train_time:1227076ms step_avg:246.65ms step:4976/5535 train_time:1227334ms step_avg:246.65ms step:4977/5535 train_time:1227592ms step_avg:246.65ms step:4978/5535 train_time:1227855ms step_avg:246.66ms step:4979/5535 train_time:1228113ms step_avg:246.66ms step:4980/5535 train_time:1228371ms step_avg:246.66ms step:4981/5535 train_time:1228629ms step_avg:246.66ms step:4982/5535 train_time:1228884ms step_avg:246.66ms step:4983/5535 train_time:1229143ms step_avg:246.67ms step:4984/5535 train_time:1229396ms step_avg:246.67ms step:4985/5535 train_time:1229682ms step_avg:246.68ms step:4986/5535 train_time:1229911ms step_avg:246.67ms step:4987/5535 train_time:1230169ms step_avg:246.68ms step:4988/5535 train_time:1230422ms step_avg:246.68ms step:4989/5535 train_time:1230675ms step_avg:246.68ms step:4990/5535 train_time:1230930ms step_avg:246.68ms step:4991/5535 train_time:1231186ms step_avg:246.68ms step:4992/5535 train_time:1231442ms step_avg:246.68ms step:4993/5535 train_time:1231695ms step_avg:246.68ms step:4994/5535 train_time:1231956ms step_avg:246.69ms step:4995/5535 train_time:1232208ms step_avg:246.69ms step:4996/5535 train_time:1232463ms step_avg:246.69ms step:4997/5535 train_time:1232721ms step_avg:246.69ms step:4998/5535 train_time:1232973ms step_avg:246.69ms step:4999/5535 train_time:1233231ms step_avg:246.70ms step:5000/5535 train_time:1233486ms step_avg:246.70ms step:5000/5535 val_loss:2.945866 train_time:1233758ms step_avg:246.75ms step:5001/5535 train_time:1233779ms step_avg:246.71ms step:5002/5535 train_time:1234011ms step_avg:246.70ms step:5003/5535 train_time:1234274ms step_avg:246.71ms step:5004/5535 train_time:1234525ms step_avg:246.71ms step:5005/5535 train_time:1234776ms step_avg:246.71ms step:5006/5535 train_time:1235034ms step_avg:246.71ms step:5007/5535 train_time:1235292ms step_avg:246.71ms step:5008/5535 train_time:1235550ms step_avg:246.72ms step:5009/5535 train_time:1235812ms step_avg:246.72ms step:5010/5535 train_time:1236065ms step_avg:246.72ms step:5011/5535 train_time:1236321ms step_avg:246.72ms step:5012/5535 train_time:1236570ms step_avg:246.72ms step:5013/5535 train_time:1236856ms step_avg:246.73ms step:5014/5535 train_time:1237092ms step_avg:246.73ms step:5015/5535 train_time:1237347ms step_avg:246.73ms step:5016/5535 train_time:1237602ms step_avg:246.73ms step:5017/5535 train_time:1237856ms step_avg:246.73ms step:5018/5535 train_time:1238121ms step_avg:246.74ms step:5019/5535 train_time:1238370ms step_avg:246.74ms step:5020/5535 train_time:1238626ms step_avg:246.74ms step:5021/5535 train_time:1238879ms step_avg:246.74ms step:5022/5535 train_time:1239136ms step_avg:246.74ms step:5023/5535 train_time:1239396ms step_avg:246.74ms step:5024/5535 train_time:1239654ms step_avg:246.75ms step:5025/5535 train_time:1239912ms step_avg:246.75ms step:5026/5535 train_time:1240166ms step_avg:246.75ms step:5027/5535 train_time:1240419ms step_avg:246.75ms step:5028/5535 train_time:1240677ms step_avg:246.75ms step:5029/5535 train_time:1240936ms step_avg:246.76ms step:5030/5535 train_time:1241195ms step_avg:246.76ms step:5031/5535 train_time:1241447ms step_avg:246.76ms step:5032/5535 train_time:1241698ms step_avg:246.76ms step:5033/5535 train_time:1241960ms step_avg:246.76ms step:5034/5535 train_time:1242210ms step_avg:246.76ms step:5035/5535 train_time:1242463ms step_avg:246.77ms step:5036/5535 train_time:1242716ms step_avg:246.77ms step:5037/5535 train_time:1242978ms step_avg:246.77ms step:5038/5535 train_time:1243231ms step_avg:246.77ms step:5039/5535 train_time:1243485ms step_avg:246.77ms step:5040/5535 train_time:1243742ms step_avg:246.77ms step:5041/5535 train_time:1244027ms step_avg:246.78ms step:5042/5535 train_time:1244264ms step_avg:246.78ms step:5043/5535 train_time:1244515ms step_avg:246.78ms step:5044/5535 train_time:1244769ms step_avg:246.78ms step:5045/5535 train_time:1245027ms step_avg:246.78ms step:5046/5535 train_time:1245291ms step_avg:246.79ms step:5047/5535 train_time:1245545ms step_avg:246.79ms step:5048/5535 train_time:1245811ms step_avg:246.79ms step:5049/5535 train_time:1246061ms step_avg:246.79ms step:5050/5535 train_time:1246315ms step_avg:246.80ms step:5051/5535 train_time:1246574ms step_avg:246.80ms step:5052/5535 train_time:1246830ms step_avg:246.80ms step:5053/5535 train_time:1247083ms step_avg:246.80ms step:5054/5535 train_time:1247337ms step_avg:246.80ms step:5055/5535 train_time:1247594ms step_avg:246.80ms step:5056/5535 train_time:1247848ms step_avg:246.81ms step:5057/5535 train_time:1248107ms step_avg:246.81ms step:5058/5535 train_time:1248368ms step_avg:246.81ms step:5059/5535 train_time:1248630ms step_avg:246.81ms step:5060/5535 train_time:1248890ms step_avg:246.82ms step:5061/5535 train_time:1249152ms step_avg:246.82ms step:5062/5535 train_time:1249420ms step_avg:246.82ms step:5063/5535 train_time:1249675ms step_avg:246.83ms step:5064/5535 train_time:1249932ms step_avg:246.83ms step:5065/5535 train_time:1250188ms step_avg:246.83ms step:5066/5535 train_time:1250447ms step_avg:246.83ms step:5067/5535 train_time:1250703ms step_avg:246.83ms step:5068/5535 train_time:1250953ms step_avg:246.83ms step:5069/5535 train_time:1251240ms step_avg:246.84ms step:5070/5535 train_time:1251479ms step_avg:246.84ms step:5071/5535 train_time:1251733ms step_avg:246.84ms step:5072/5535 train_time:1251986ms step_avg:246.84ms step:5073/5535 train_time:1252243ms step_avg:246.84ms step:5074/5535 train_time:1252503ms step_avg:246.85ms step:5075/5535 train_time:1252771ms step_avg:246.85ms step:5076/5535 train_time:1253021ms step_avg:246.85ms step:5077/5535 train_time:1253273ms step_avg:246.85ms step:5078/5535 train_time:1253537ms step_avg:246.86ms step:5079/5535 train_time:1253805ms step_avg:246.86ms step:5080/5535 train_time:1254062ms step_avg:246.86ms step:5081/5535 train_time:1254315ms step_avg:246.86ms step:5082/5535 train_time:1254571ms step_avg:246.87ms step:5083/5535 train_time:1254844ms step_avg:246.87ms step:5084/5535 train_time:1255100ms step_avg:246.87ms step:5085/5535 train_time:1255360ms step_avg:246.88ms step:5086/5535 train_time:1255622ms step_avg:246.88ms step:5087/5535 train_time:1255874ms step_avg:246.88ms step:5088/5535 train_time:1256150ms step_avg:246.88ms step:5089/5535 train_time:1256401ms step_avg:246.89ms step:5090/5535 train_time:1256655ms step_avg:246.89ms step:5091/5535 train_time:1256914ms step_avg:246.89ms step:5092/5535 train_time:1257168ms step_avg:246.89ms step:5093/5535 train_time:1257425ms step_avg:246.89ms step:5094/5535 train_time:1257685ms step_avg:246.90ms step:5095/5535 train_time:1257944ms step_avg:246.90ms step:5096/5535 train_time:1258200ms step_avg:246.90ms step:5097/5535 train_time:1258485ms step_avg:246.91ms step:5098/5535 train_time:1258716ms step_avg:246.90ms step:5099/5535 train_time:1258982ms step_avg:246.91ms step:5100/5535 train_time:1259243ms step_avg:246.91ms step:5101/5535 train_time:1259507ms step_avg:246.91ms step:5102/5535 train_time:1259762ms step_avg:246.92ms step:5103/5535 train_time:1260019ms step_avg:246.92ms step:5104/5535 train_time:1260272ms step_avg:246.92ms step:5105/5535 train_time:1260531ms step_avg:246.92ms step:5106/5535 train_time:1260787ms step_avg:246.92ms step:5107/5535 train_time:1261049ms step_avg:246.93ms step:5108/5535 train_time:1261300ms step_avg:246.93ms step:5109/5535 train_time:1261554ms step_avg:246.93ms step:5110/5535 train_time:1261809ms step_avg:246.93ms step:5111/5535 train_time:1262070ms step_avg:246.93ms step:5112/5535 train_time:1262332ms step_avg:246.94ms step:5113/5535 train_time:1262593ms step_avg:246.94ms step:5114/5535 train_time:1262850ms step_avg:246.94ms step:5115/5535 train_time:1263106ms step_avg:246.94ms step:5116/5535 train_time:1263361ms step_avg:246.94ms step:5117/5535 train_time:1263622ms step_avg:246.95ms step:5118/5535 train_time:1263878ms step_avg:246.95ms step:5119/5535 train_time:1264132ms step_avg:246.95ms step:5120/5535 train_time:1264388ms step_avg:246.95ms step:5121/5535 train_time:1264665ms step_avg:246.96ms step:5122/5535 train_time:1264917ms step_avg:246.96ms step:5123/5535 train_time:1265171ms step_avg:246.96ms step:5124/5535 train_time:1265430ms step_avg:246.96ms step:5125/5535 train_time:1265722ms step_avg:246.97ms step:5125/5535 val_loss:2.938193 train_time:1265963ms step_avg:247.02ms step:5126/5535 train_time:1265984ms step_avg:246.97ms step:5127/5535 train_time:1266217ms step_avg:246.97ms step:5128/5535 train_time:1266492ms step_avg:246.98ms step:5129/5535 train_time:1266748ms step_avg:246.98ms step:5130/5535 train_time:1266996ms step_avg:246.98ms step:5131/5535 train_time:1267253ms step_avg:246.98ms step:5132/5535 train_time:1267522ms step_avg:246.98ms step:5133/5535 train_time:1267783ms step_avg:246.99ms step:5134/5535 train_time:1268046ms step_avg:246.99ms step:5135/5535 train_time:1268295ms step_avg:246.99ms step:5136/5535 train_time:1268563ms step_avg:246.99ms step:5137/5535 train_time:1268822ms step_avg:247.00ms step:5138/5535 train_time:1269101ms step_avg:247.00ms step:5139/5535 train_time:1269355ms step_avg:247.00ms step:5140/5535 train_time:1269615ms step_avg:247.01ms step:5141/5535 train_time:1269873ms step_avg:247.01ms step:5142/5535 train_time:1270127ms step_avg:247.01ms step:5143/5535 train_time:1270382ms step_avg:247.01ms step:5144/5535 train_time:1270638ms step_avg:247.01ms step:5145/5535 train_time:1270907ms step_avg:247.02ms step:5146/5535 train_time:1271167ms step_avg:247.02ms step:5147/5535 train_time:1271416ms step_avg:247.02ms step:5148/5535 train_time:1271676ms step_avg:247.02ms step:5149/5535 train_time:1271939ms step_avg:247.03ms step:5150/5535 train_time:1272201ms step_avg:247.03ms step:5151/5535 train_time:1272466ms step_avg:247.03ms step:5152/5535 train_time:1272720ms step_avg:247.03ms step:5153/5535 train_time:1273008ms step_avg:247.04ms step:5154/5535 train_time:1273236ms step_avg:247.04ms step:5155/5535 train_time:1273496ms step_avg:247.04ms step:5156/5535 train_time:1273748ms step_avg:247.04ms step:5157/5535 train_time:1274003ms step_avg:247.04ms step:5158/5535 train_time:1274262ms step_avg:247.05ms step:5159/5535 train_time:1274528ms step_avg:247.05ms step:5160/5535 train_time:1274782ms step_avg:247.05ms step:5161/5535 train_time:1275035ms step_avg:247.05ms step:5162/5535 train_time:1275295ms step_avg:247.05ms step:5163/5535 train_time:1275551ms step_avg:247.06ms step:5164/5535 train_time:1275809ms step_avg:247.06ms step:5165/5535 train_time:1276070ms step_avg:247.06ms step:5166/5535 train_time:1276326ms step_avg:247.06ms step:5167/5535 train_time:1276585ms step_avg:247.06ms step:5168/5535 train_time:1276850ms step_avg:247.07ms step:5169/5535 train_time:1277106ms step_avg:247.07ms step:5170/5535 train_time:1277363ms step_avg:247.07ms step:5171/5535 train_time:1277617ms step_avg:247.07ms step:5172/5535 train_time:1277882ms step_avg:247.08ms step:5173/5535 train_time:1278135ms step_avg:247.08ms step:5174/5535 train_time:1278392ms step_avg:247.08ms step:5175/5535 train_time:1278658ms step_avg:247.08ms step:5176/5535 train_time:1278917ms step_avg:247.09ms step:5177/5535 train_time:1279181ms step_avg:247.09ms step:5178/5535 train_time:1279435ms step_avg:247.09ms step:5179/5535 train_time:1279687ms step_avg:247.09ms step:5180/5535 train_time:1279970ms step_avg:247.10ms step:5181/5535 train_time:1280255ms step_avg:247.11ms step:5182/5535 train_time:1280484ms step_avg:247.10ms step:5183/5535 train_time:1280741ms step_avg:247.10ms step:5184/5535 train_time:1280999ms step_avg:247.11ms step:5185/5535 train_time:1281259ms step_avg:247.11ms step:5186/5535 train_time:1281513ms step_avg:247.11ms step:5187/5535 train_time:1281769ms step_avg:247.11ms step:5188/5535 train_time:1282031ms step_avg:247.11ms step:5189/5535 train_time:1282295ms step_avg:247.12ms step:5190/5535 train_time:1282547ms step_avg:247.12ms step:5191/5535 train_time:1282805ms step_avg:247.12ms step:5192/5535 train_time:1283066ms step_avg:247.12ms step:5193/5535 train_time:1283351ms step_avg:247.13ms step:5194/5535 train_time:1283611ms step_avg:247.13ms step:5195/5535 train_time:1283865ms step_avg:247.13ms step:5196/5535 train_time:1284119ms step_avg:247.14ms step:5197/5535 train_time:1284374ms step_avg:247.14ms step:5198/5535 train_time:1284644ms step_avg:247.14ms step:5199/5535 train_time:1284901ms step_avg:247.14ms step:5200/5535 train_time:1285152ms step_avg:247.14ms step:5201/5535 train_time:1285416ms step_avg:247.15ms step:5202/5535 train_time:1285673ms step_avg:247.15ms step:5203/5535 train_time:1285930ms step_avg:247.15ms step:5204/5535 train_time:1286188ms step_avg:247.15ms step:5205/5535 train_time:1286442ms step_avg:247.16ms step:5206/5535 train_time:1286700ms step_avg:247.16ms step:5207/5535 train_time:1286956ms step_avg:247.16ms step:5208/5535 train_time:1287217ms step_avg:247.16ms step:5209/5535 train_time:1287500ms step_avg:247.17ms step:5210/5535 train_time:1287739ms step_avg:247.17ms step:5211/5535 train_time:1287998ms step_avg:247.17ms step:5212/5535 train_time:1288254ms step_avg:247.17ms step:5213/5535 train_time:1288515ms step_avg:247.17ms step:5214/5535 train_time:1288770ms step_avg:247.17ms step:5215/5535 train_time:1289041ms step_avg:247.18ms step:5216/5535 train_time:1289300ms step_avg:247.18ms step:5217/5535 train_time:1289551ms step_avg:247.18ms step:5218/5535 train_time:1289814ms step_avg:247.19ms step:5219/5535 train_time:1290068ms step_avg:247.19ms step:5220/5535 train_time:1290328ms step_avg:247.19ms step:5221/5535 train_time:1290603ms step_avg:247.19ms step:5222/5535 train_time:1290859ms step_avg:247.20ms step:5223/5535 train_time:1291114ms step_avg:247.20ms step:5224/5535 train_time:1291370ms step_avg:247.20ms step:5225/5535 train_time:1291628ms step_avg:247.20ms step:5226/5535 train_time:1291880ms step_avg:247.20ms step:5227/5535 train_time:1292140ms step_avg:247.20ms step:5228/5535 train_time:1292393ms step_avg:247.21ms step:5229/5535 train_time:1292653ms step_avg:247.21ms step:5230/5535 train_time:1292908ms step_avg:247.21ms step:5231/5535 train_time:1293161ms step_avg:247.21ms step:5232/5535 train_time:1293421ms step_avg:247.21ms step:5233/5535 train_time:1293682ms step_avg:247.22ms step:5234/5535 train_time:1293952ms step_avg:247.22ms step:5235/5535 train_time:1294217ms step_avg:247.22ms step:5236/5535 train_time:1294487ms step_avg:247.23ms step:5237/5535 train_time:1294776ms step_avg:247.24ms step:5238/5535 train_time:1295013ms step_avg:247.23ms step:5239/5535 train_time:1295263ms step_avg:247.23ms step:5240/5535 train_time:1295531ms step_avg:247.24ms step:5241/5535 train_time:1295790ms step_avg:247.24ms step:5242/5535 train_time:1296041ms step_avg:247.24ms step:5243/5535 train_time:1296302ms step_avg:247.24ms step:5244/5535 train_time:1296562ms step_avg:247.25ms step:5245/5535 train_time:1296824ms step_avg:247.25ms step:5246/5535 train_time:1297081ms step_avg:247.25ms step:5247/5535 train_time:1297340ms step_avg:247.25ms step:5248/5535 train_time:1297596ms step_avg:247.26ms step:5249/5535 train_time:1297859ms step_avg:247.26ms step:5250/5535 train_time:1298115ms step_avg:247.26ms step:5250/5535 val_loss:2.931067 train_time:1298371ms step_avg:247.31ms step:5251/5535 train_time:1298392ms step_avg:247.27ms step:5252/5535 train_time:1298637ms step_avg:247.27ms step:5253/5535 train_time:1298895ms step_avg:247.27ms step:5254/5535 train_time:1299155ms step_avg:247.27ms step:5255/5535 train_time:1299423ms step_avg:247.27ms step:5256/5535 train_time:1299680ms step_avg:247.28ms step:5257/5535 train_time:1299934ms step_avg:247.28ms step:5258/5535 train_time:1300191ms step_avg:247.28ms step:5259/5535 train_time:1300450ms step_avg:247.28ms step:5260/5535 train_time:1300714ms step_avg:247.28ms step:5261/5535 train_time:1300968ms step_avg:247.29ms step:5262/5535 train_time:1301217ms step_avg:247.29ms step:5263/5535 train_time:1301479ms step_avg:247.29ms step:5264/5535 train_time:1301739ms step_avg:247.29ms step:5265/5535 train_time:1302024ms step_avg:247.30ms step:5266/5535 train_time:1302258ms step_avg:247.30ms step:5267/5535 train_time:1302512ms step_avg:247.30ms step:5268/5535 train_time:1302773ms step_avg:247.30ms step:5269/5535 train_time:1303028ms step_avg:247.30ms step:5270/5535 train_time:1303296ms step_avg:247.30ms step:5271/5535 train_time:1303551ms step_avg:247.31ms step:5272/5535 train_time:1303806ms step_avg:247.31ms step:5273/5535 train_time:1304067ms step_avg:247.31ms step:5274/5535 train_time:1304329ms step_avg:247.31ms step:5275/5535 train_time:1304590ms step_avg:247.32ms step:5276/5535 train_time:1304845ms step_avg:247.32ms step:5277/5535 train_time:1305101ms step_avg:247.32ms step:5278/5535 train_time:1305359ms step_avg:247.32ms step:5279/5535 train_time:1305619ms step_avg:247.32ms step:5280/5535 train_time:1305892ms step_avg:247.33ms step:5281/5535 train_time:1306144ms step_avg:247.33ms step:5282/5535 train_time:1306406ms step_avg:247.33ms step:5283/5535 train_time:1306671ms step_avg:247.33ms step:5284/5535 train_time:1306932ms step_avg:247.34ms step:5285/5535 train_time:1307191ms step_avg:247.34ms step:5286/5535 train_time:1307459ms step_avg:247.34ms step:5287/5535 train_time:1307716ms step_avg:247.35ms step:5288/5535 train_time:1307978ms step_avg:247.35ms step:5289/5535 train_time:1308231ms step_avg:247.35ms step:5290/5535 train_time:1308499ms step_avg:247.35ms step:5291/5535 train_time:1308760ms step_avg:247.36ms step:5292/5535 train_time:1309017ms step_avg:247.36ms step:5293/5535 train_time:1309303ms step_avg:247.37ms step:5294/5535 train_time:1309563ms step_avg:247.37ms step:5295/5535 train_time:1309814ms step_avg:247.37ms step:5296/5535 train_time:1310081ms step_avg:247.37ms step:5297/5535 train_time:1310339ms step_avg:247.37ms step:5298/5535 train_time:1310600ms step_avg:247.38ms step:5299/5535 train_time:1310854ms step_avg:247.38ms step:5300/5535 train_time:1311110ms step_avg:247.38ms step:5301/5535 train_time:1311368ms step_avg:247.38ms step:5302/5535 train_time:1311626ms step_avg:247.38ms step:5303/5535 train_time:1311885ms step_avg:247.39ms step:5304/5535 train_time:1312144ms step_avg:247.39ms step:5305/5535 train_time:1312401ms step_avg:247.39ms step:5306/5535 train_time:1312655ms step_avg:247.39ms step:5307/5535 train_time:1312909ms step_avg:247.39ms step:5308/5535 train_time:1313163ms step_avg:247.39ms step:5309/5535 train_time:1313429ms step_avg:247.40ms step:5310/5535 train_time:1313686ms step_avg:247.40ms step:5311/5535 train_time:1313944ms step_avg:247.40ms step:5312/5535 train_time:1314200ms step_avg:247.40ms step:5313/5535 train_time:1314484ms step_avg:247.41ms step:5314/5535 train_time:1314747ms step_avg:247.41ms step:5315/5535 train_time:1314995ms step_avg:247.41ms step:5316/5535 train_time:1315253ms step_avg:247.41ms step:5317/5535 train_time:1315518ms step_avg:247.42ms step:5318/5535 train_time:1315775ms step_avg:247.42ms step:5319/5535 train_time:1316040ms step_avg:247.42ms step:5320/5535 train_time:1316297ms step_avg:247.42ms step:5321/5535 train_time:1316583ms step_avg:247.43ms step:5322/5535 train_time:1316826ms step_avg:247.43ms step:5323/5535 train_time:1317087ms step_avg:247.43ms step:5324/5535 train_time:1317347ms step_avg:247.44ms step:5325/5535 train_time:1317605ms step_avg:247.44ms step:5326/5535 train_time:1317862ms step_avg:247.44ms step:5327/5535 train_time:1318123ms step_avg:247.44ms step:5328/5535 train_time:1318387ms step_avg:247.44ms step:5329/5535 train_time:1318652ms step_avg:247.45ms step:5330/5535 train_time:1318912ms step_avg:247.45ms step:5331/5535 train_time:1319174ms step_avg:247.45ms step:5332/5535 train_time:1319431ms step_avg:247.46ms step:5333/5535 train_time:1319687ms step_avg:247.46ms step:5334/5535 train_time:1319948ms step_avg:247.46ms step:5335/5535 train_time:1320227ms step_avg:247.47ms step:5336/5535 train_time:1320482ms step_avg:247.47ms step:5337/5535 train_time:1320738ms step_avg:247.47ms step:5338/5535 train_time:1320993ms step_avg:247.47ms step:5339/5535 train_time:1321250ms step_avg:247.47ms step:5340/5535 train_time:1321507ms step_avg:247.47ms step:5341/5535 train_time:1321760ms step_avg:247.47ms step:5342/5535 train_time:1322012ms step_avg:247.48ms step:5343/5535 train_time:1322272ms step_avg:247.48ms step:5344/5535 train_time:1322531ms step_avg:247.48ms step:5345/5535 train_time:1322788ms step_avg:247.48ms step:5346/5535 train_time:1323049ms step_avg:247.48ms step:5347/5535 train_time:1323308ms step_avg:247.49ms step:5348/5535 train_time:1323561ms step_avg:247.49ms step:5349/5535 train_time:1323864ms step_avg:247.50ms step:5350/5535 train_time:1324108ms step_avg:247.50ms step:5351/5535 train_time:1324365ms step_avg:247.50ms step:5352/5535 train_time:1324626ms step_avg:247.50ms step:5353/5535 train_time:1324882ms step_avg:247.50ms step:5354/5535 train_time:1325143ms step_avg:247.51ms step:5355/5535 train_time:1325404ms step_avg:247.51ms step:5356/5535 train_time:1325669ms step_avg:247.51ms step:5357/5535 train_time:1325920ms step_avg:247.51ms step:5358/5535 train_time:1326175ms step_avg:247.51ms step:5359/5535 train_time:1326430ms step_avg:247.51ms step:5360/5535 train_time:1326687ms step_avg:247.52ms step:5361/5535 train_time:1326954ms step_avg:247.52ms step:5362/5535 train_time:1327205ms step_avg:247.52ms step:5363/5535 train_time:1327470ms step_avg:247.52ms step:5364/5535 train_time:1327729ms step_avg:247.53ms step:5365/5535 train_time:1327998ms step_avg:247.53ms step:5366/5535 train_time:1328252ms step_avg:247.53ms step:5367/5535 train_time:1328511ms step_avg:247.53ms step:5368/5535 train_time:1328780ms step_avg:247.54ms step:5369/5535 train_time:1329039ms step_avg:247.54ms step:5370/5535 train_time:1329291ms step_avg:247.54ms step:5371/5535 train_time:1329559ms step_avg:247.54ms step:5372/5535 train_time:1329825ms step_avg:247.55ms step:5373/5535 train_time:1330086ms step_avg:247.55ms step:5374/5535 train_time:1330356ms step_avg:247.55ms step:5375/5535 train_time:1330616ms step_avg:247.56ms step:5375/5535 val_loss:2.924670 train_time:1330890ms step_avg:247.61ms step:5376/5535 train_time:1330911ms step_avg:247.57ms step:5377/5535 train_time:1331176ms step_avg:247.57ms step:5378/5535 train_time:1331417ms step_avg:247.57ms step:5379/5535 train_time:1331668ms step_avg:247.57ms step:5380/5535 train_time:1331921ms step_avg:247.57ms step:5381/5535 train_time:1332176ms step_avg:247.57ms step:5382/5535 train_time:1332456ms step_avg:247.58ms step:5383/5535 train_time:1332712ms step_avg:247.58ms step:5384/5535 train_time:1332972ms step_avg:247.58ms step:5385/5535 train_time:1333227ms step_avg:247.58ms step:5386/5535 train_time:1333487ms step_avg:247.58ms step:5387/5535 train_time:1333749ms step_avg:247.59ms step:5388/5535 train_time:1334009ms step_avg:247.59ms step:5389/5535 train_time:1334277ms step_avg:247.59ms step:5390/5535 train_time:1334526ms step_avg:247.59ms step:5391/5535 train_time:1334789ms step_avg:247.60ms step:5392/5535 train_time:1335051ms step_avg:247.60ms step:5393/5535 train_time:1335310ms step_avg:247.60ms step:5394/5535 train_time:1335567ms step_avg:247.60ms step:5395/5535 train_time:1335829ms step_avg:247.61ms step:5396/5535 train_time:1336094ms step_avg:247.61ms step:5397/5535 train_time:1336360ms step_avg:247.61ms step:5398/5535 train_time:1336622ms step_avg:247.61ms step:5399/5535 train_time:1336895ms step_avg:247.62ms step:5400/5535 train_time:1337153ms step_avg:247.62ms step:5401/5535 train_time:1337411ms step_avg:247.62ms step:5402/5535 train_time:1337672ms step_avg:247.63ms step:5403/5535 train_time:1337932ms step_avg:247.63ms step:5404/5535 train_time:1338201ms step_avg:247.63ms step:5405/5535 train_time:1338485ms step_avg:247.64ms step:5406/5535 train_time:1338717ms step_avg:247.64ms step:5407/5535 train_time:1338980ms step_avg:247.64ms step:5408/5535 train_time:1339240ms step_avg:247.64ms step:5409/5535 train_time:1339506ms step_avg:247.64ms step:5410/5535 train_time:1339760ms step_avg:247.65ms step:5411/5535 train_time:1340018ms step_avg:247.65ms step:5412/5535 train_time:1340289ms step_avg:247.65ms step:5413/5535 train_time:1340557ms step_avg:247.66ms step:5414/5535 train_time:1340806ms step_avg:247.66ms step:5415/5535 train_time:1341065ms step_avg:247.66ms step:5416/5535 train_time:1341327ms step_avg:247.66ms step:5417/5535 train_time:1341584ms step_avg:247.66ms step:5418/5535 train_time:1341845ms step_avg:247.66ms step:5419/5535 train_time:1342104ms step_avg:247.67ms step:5420/5535 train_time:1342367ms step_avg:247.67ms step:5421/5535 train_time:1342623ms step_avg:247.67ms step:5422/5535 train_time:1342887ms step_avg:247.67ms step:5423/5535 train_time:1343145ms step_avg:247.68ms step:5424/5535 train_time:1343439ms step_avg:247.68ms step:5425/5535 train_time:1343722ms step_avg:247.69ms step:5426/5535 train_time:1343987ms step_avg:247.69ms step:5427/5535 train_time:1344248ms step_avg:247.70ms step:5428/5535 train_time:1344508ms step_avg:247.70ms step:5429/5535 train_time:1344766ms step_avg:247.70ms step:5430/5535 train_time:1345021ms step_avg:247.70ms step:5431/5535 train_time:1345277ms step_avg:247.70ms step:5432/5535 train_time:1345537ms step_avg:247.71ms step:5433/5535 train_time:1345844ms step_avg:247.72ms step:5434/5535 train_time:1346081ms step_avg:247.71ms step:5435/5535 train_time:1346338ms step_avg:247.72ms step:5436/5535 train_time:1346597ms step_avg:247.72ms step:5437/5535 train_time:1346847ms step_avg:247.72ms step:5438/5535 train_time:1347134ms step_avg:247.73ms step:5439/5535 train_time:1347416ms step_avg:247.73ms step:5440/5535 train_time:1347674ms step_avg:247.73ms step:5441/5535 train_time:1347931ms step_avg:247.74ms step:5442/5535 train_time:1348186ms step_avg:247.74ms step:5443/5535 train_time:1348453ms step_avg:247.74ms step:5444/5535 train_time:1348707ms step_avg:247.74ms step:5445/5535 train_time:1348963ms step_avg:247.74ms step:5446/5535 train_time:1349221ms step_avg:247.75ms step:5447/5535 train_time:1349485ms step_avg:247.75ms step:5448/5535 train_time:1349738ms step_avg:247.75ms step:5449/5535 train_time:1349998ms step_avg:247.75ms step:5450/5535 train_time:1350263ms step_avg:247.75ms step:5451/5535 train_time:1350522ms step_avg:247.76ms step:5452/5535 train_time:1350793ms step_avg:247.76ms step:5453/5535 train_time:1351064ms step_avg:247.77ms step:5454/5535 train_time:1351336ms step_avg:247.77ms step:5455/5535 train_time:1351593ms step_avg:247.77ms step:5456/5535 train_time:1351846ms step_avg:247.77ms step:5457/5535 train_time:1352126ms step_avg:247.78ms step:5458/5535 train_time:1352380ms step_avg:247.78ms step:5459/5535 train_time:1352638ms step_avg:247.78ms step:5460/5535 train_time:1352892ms step_avg:247.78ms step:5461/5535 train_time:1353190ms step_avg:247.79ms step:5462/5535 train_time:1353440ms step_avg:247.79ms step:5463/5535 train_time:1353699ms step_avg:247.79ms step:5464/5535 train_time:1353972ms step_avg:247.80ms step:5465/5535 train_time:1354230ms step_avg:247.80ms step:5466/5535 train_time:1354491ms step_avg:247.80ms step:5467/5535 train_time:1354743ms step_avg:247.80ms step:5468/5535 train_time:1355003ms step_avg:247.81ms step:5469/5535 train_time:1355274ms step_avg:247.81ms step:5470/5535 train_time:1355536ms step_avg:247.81ms step:5471/5535 train_time:1355797ms step_avg:247.82ms step:5472/5535 train_time:1356051ms step_avg:247.82ms step:5473/5535 train_time:1356311ms step_avg:247.82ms step:5474/5535 train_time:1356587ms step_avg:247.82ms step:5475/5535 train_time:1356846ms step_avg:247.83ms step:5476/5535 train_time:1357106ms step_avg:247.83ms step:5477/5535 train_time:1357360ms step_avg:247.83ms step:5478/5535 train_time:1357620ms step_avg:247.83ms step:5479/5535 train_time:1357880ms step_avg:247.83ms step:5480/5535 train_time:1358139ms step_avg:247.84ms step:5481/5535 train_time:1358397ms step_avg:247.84ms step:5482/5535 train_time:1358672ms step_avg:247.84ms step:5483/5535 train_time:1358929ms step_avg:247.84ms step:5484/5535 train_time:1359210ms step_avg:247.85ms step:5485/5535 train_time:1359473ms step_avg:247.85ms step:5486/5535 train_time:1359731ms step_avg:247.85ms step:5487/5535 train_time:1359996ms step_avg:247.86ms step:5488/5535 train_time:1360258ms step_avg:247.86ms step:5489/5535 train_time:1360544ms step_avg:247.87ms step:5490/5535 train_time:1360776ms step_avg:247.86ms step:5491/5535 train_time:1361041ms step_avg:247.87ms step:5492/5535 train_time:1361295ms step_avg:247.87ms step:5493/5535 train_time:1361557ms step_avg:247.87ms step:5494/5535 train_time:1361820ms step_avg:247.87ms step:5495/5535 train_time:1362087ms step_avg:247.88ms step:5496/5535 train_time:1362340ms step_avg:247.88ms step:5497/5535 train_time:1362598ms step_avg:247.88ms step:5498/5535 train_time:1362861ms step_avg:247.88ms step:5499/5535 train_time:1363119ms step_avg:247.88ms step:5500/5535 train_time:1363379ms step_avg:247.89ms step:5500/5535 val_loss:2.920043 train_time:1363639ms step_avg:247.93ms step:5501/5535 train_time:1363660ms step_avg:247.89ms step:5502/5535 train_time:1363901ms step_avg:247.89ms step:5503/5535 train_time:1364167ms step_avg:247.90ms step:5504/5535 train_time:1364438ms step_avg:247.90ms step:5505/5535 train_time:1364694ms step_avg:247.90ms step:5506/5535 train_time:1364962ms step_avg:247.90ms step:5507/5535 train_time:1365218ms step_avg:247.91ms step:5508/5535 train_time:1365485ms step_avg:247.91ms step:5509/5535 train_time:1365741ms step_avg:247.91ms step:5510/5535 train_time:1366003ms step_avg:247.91ms step:5511/5535 train_time:1366277ms step_avg:247.92ms step:5512/5535 train_time:1366538ms step_avg:247.92ms step:5513/5535 train_time:1366807ms step_avg:247.92ms step:5514/5535 train_time:1367067ms step_avg:247.93ms step:5515/5535 train_time:1367331ms step_avg:247.93ms step:5516/5535 train_time:1367592ms step_avg:247.93ms step:5517/5535 train_time:1367887ms step_avg:247.94ms step:5518/5535 train_time:1368123ms step_avg:247.94ms step:5519/5535 train_time:1368382ms step_avg:247.94ms step:5520/5535 train_time:1368646ms step_avg:247.94ms step:5521/5535 train_time:1368905ms step_avg:247.95ms step:5522/5535 train_time:1369176ms step_avg:247.95ms step:5523/5535 train_time:1369432ms step_avg:247.95ms step:5524/5535 train_time:1369685ms step_avg:247.95ms step:5525/5535 train_time:1369984ms step_avg:247.96ms step:5526/5535 train_time:1370245ms step_avg:247.96ms step:5527/5535 train_time:1370497ms step_avg:247.96ms step:5528/5535 train_time:1370763ms step_avg:247.97ms step:5529/5535 train_time:1371024ms step_avg:247.97ms step:5530/5535 train_time:1371300ms step_avg:247.97ms step:5531/5535 train_time:1371567ms step_avg:247.98ms step:5532/5535 train_time:1371822ms step_avg:247.98ms step:5533/5535 train_time:1372088ms step_avg:247.98ms step:5534/5535 train_time:1372342ms step_avg:247.98ms step:5535/5535 train_time:1372603ms step_avg:247.99ms step:5535/5535 val_loss:2.919281 train_time:1372856ms step_avg:248.03ms peak memory allocated: 56810 MiB reserved: 63776 MiB