import os
import sys
with open(sys.argv[0]) as f:
    code = f.read() # read the code of this file ASAP, for logging
import uuid
import time
from dataclasses import dataclass
from pathlib import Path

import torch
from torch import nn
import torch.nn.functional as F
import torch.distributed as dist
import torch._inductor.config as config
from torch.nn.parallel import DistributedDataParallel as DDP
# Use of FlexAttention contributed by @KoszarskyB
from torch.nn.attention.flex_attention import BlockMask, flex_attention

# -----------------------------------------------------------------------------
# Muon optimizer

def zeropower_via_svd(G, steps=None):
    U, S, V = G.svd()
    return U @ V.T

@torch.compile
def zeropower_via_newtonschulz5(G, steps=10, eps=1e-7):
    """
    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}' ~ Uniform(0.5, 1.5), which turns out not to hurt model
    performance at all relative to UV^T, where USV^T = G is the SVD.
    """
    assert len(G.shape) == 2
    a, b, c = (3.4445, -4.7750,  2.0315)
    X = G.bfloat16()
    X /= (X.norm() + eps) # ensure top singular value <= 1
    if G.size(0) > G.size(1):
        X = X.T
    for _ in range(steps):
        A = X @ X.T
        B = b * A + c * A @ A # adapted from suggestion by @jxbz, @leloykun, and @YouJiacheng
        X = a * X + B @ X
    if G.size(0) > G.size(1):
        X = X.T
    return X

zeropower_backends = dict(svd=zeropower_via_svd, newtonschulz5=zeropower_via_newtonschulz5)

class Muon(torch.optim.Optimizer):
    """
    Muon - MomentUm Orthogonalized by Newton-schulz

    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.

    Some warnings:
    - This optimizer assumes that all parameters passed in are 2D.
    - It 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).
    - To use it with 4D convolutional filters, it works well to just flatten their last 3 dimensions.
    - We believe it is unlikely to work well for training with small batch size.
    - We believe it may not work well for finetuning pretrained models, but we haven't tested this.
    - We have not yet tried this optimizer for training scenarios larger than NanoGPT (124M).

    Arguments:
        lr: The learning rate used by the internal SGD.
        momentum: The momentum used by the internal SGD.
        nesterov: Whether to use Nesterov-style momentum in the internal SGD. (recommended)
        backend: The chosen backend for the orthogonalization step. (recommended: 'newtonschulz5')
        backend_steps: The number of iteration steps to use in the backend, if it is iterative.
    """
    def __init__(self, params, lr=0.02, momentum=0.95, nesterov=True,
                 backend='newtonschulz5', backend_steps=5):
        self.num_process = int(os.environ['WORLD_SIZE'])
        self.rank = int(os.environ["RANK"])
        defaults = dict(lr=lr, momentum=momentum, nesterov=nesterov, backend=backend, backend_steps=backend_steps)
        params: "list[torch.Tensor]" = list(params)
        assert all(isinstance(p, torch.Tensor) for p in params)
        sizes = {p.numel() for p in params}
        param_groups = [
            {
                "params": [p for p in params if p.numel() == size],
                "update_buffer": [
                    torch.empty(size, device="cuda", dtype=torch.bfloat16)
                    for _ in range(self.num_process)
                ],
            }
            for size in sizes
        ]
        super().__init__(param_groups, defaults)

    def step(self):
        for group in self.param_groups:
            lr: float = group["lr"]
            momentum: float = group["momentum"]
            nesterov: bool = group["nesterov"]
            zeropower_backend = zeropower_backends[group["backend"]]
            backend_steps: int = group["backend_steps"]
            update_buffers: "list[torch.Tensor]" = group["update_buffer"]
            # generate weight updates in distributed fashion
            params: "list[torch.Tensor]" = group["params"]
            assert len(params) % self.num_process == 0
            handle = None
            params_world = None
            def update_prev():
                if params_world is None:
                    return
                assert handle is not None
                handle.wait()
                for p_world, g_world in zip(params_world, update_buffers):
                    p_world.data.add_(
                        g_world.view_as(p_world),
                        alpha=-lr * max(1, p_world.size(0) / p_world.size(1)) ** 0.5,
                    )
            for base_i in range(len(params))[::self.num_process]:
                p = params[base_i + self.rank]
                g = p.grad
                assert g is not None
                state = self.state[p] 
                if "momentum_buffer" not in state:
                    state["momentum_buffer"] = torch.zeros_like(g)
                buf: torch.Tensor = state["momentum_buffer"]
                buf.lerp_(g, 1 - momentum)
                g = g.lerp_(buf, momentum) if nesterov else buf
                g = zeropower_backend(g, steps=backend_steps).flatten()
                update_prev()
                handle = dist.all_gather(update_buffers, g, async_op=True)
                params_world = params[base_i : base_i + self.num_process]
            update_prev()


# -----------------------------------------------------------------------------
# PyTorch nn.Module definitions for the GPT-2 model

def norm(x):
    return F.rms_norm(x, (x.size(-1),))

class CastedLinear(nn.Linear):

    def __init__(self, in_features, out_features):
        super().__init__(in_features, out_features, bias=False)

    def forward(self, x):
        return F.linear(x, self.weight.to(x.dtype))

class Rotary(torch.nn.Module):

    def __init__(self, dim, base=10000):
        super().__init__()
        self.register_buffer('inv_freq', (1 / base) ** (torch.arange(0, dim, 2) / dim))
        self.seq_len_cached = None
        self.cos_cached = None
        self.sin_cached = None

    def forward(self, x):
        seq_len = x.shape[1]
        if seq_len != self.seq_len_cached:
            t = torch.arange(seq_len, device=x.device)
            freqs = torch.outer(t, self.inv_freq)
            self.seq_len_cached = seq_len
            self.cos_cached = freqs.cos()
            self.sin_cached = freqs.sin()
        cos, sin = self.cos_cached[None, :, None, :], self.sin_cached[None, :, None, :]
        # apply_rotary_emb(x, cos, sin)
        x1, x2 = x.chunk(2, dim=3)
        y1 = x1 * cos + x2 * sin
        y2 = x1 * (-sin) + x2 * cos
        return torch.cat((y1, y2), 3).type_as(x)

class CausalSelfAttention(nn.Module):

    def __init__(self, dim, n_head):
        super().__init__()
        assert dim % n_head == 0
        self.n_head = n_head
        self.c_q = CastedLinear(dim, dim)
        self.c_k = CastedLinear(dim, dim)
        self.c_v = CastedLinear(dim, dim)
        # value residual lambda
        self.lambdas = nn.Parameter(torch.tensor([0.5, 0.5])) # @Grad62304977
        # rotary embeddings
        self.rotary = Rotary(dim // n_head) # dim // n_head = head_dim
        # output projection
        self.c_proj = CastedLinear(dim, dim)
        self.c_proj.weight.data.zero_() # zero init suggested by @Grad62304977

    def forward(self, x: torch.Tensor, vi: torch.Tensor, block_mask: BlockMask) -> torch.Tensor:
        B, T = x.size(0), x.size(1) # batch size, sequence length
        assert B == 1, "Must use batch size = 1 for FlexAttention"
        q: torch.Tensor = self.c_q(x).view(B, T, self.n_head, -1)
        k: torch.Tensor = self.c_k(x).view(B, T, self.n_head, -1)
        v: torch.Tensor = self.c_v(x).view(B, T, self.n_head, -1)
        v = self.lambdas[0] * v + self.lambdas[1] * vi.view_as(v) # @Grad62304977
        q, k = norm(q), norm(k) # QK norm suggested by @Grad62304977
        q, k = self.rotary(q), self.rotary(k)
        y = flex_attention(q.transpose(1, 2), k.transpose(1, 2), v.transpose(1, 2), block_mask=block_mask)
        y = y.transpose(1, 2).contiguous().view_as(x) # re-assemble all head outputs side by side
        y = self.c_proj(y)
        return y

class MLP(nn.Module):

    def __init__(self, dim: int):
        super().__init__()
        self.c_fc   = CastedLinear(dim, 4 * dim)
        self.c_proj = CastedLinear(4 * dim, dim)
        self.c_proj.weight.data.zero_() # zero init suggested by @Grad62304977

    def forward(self, x: torch.Tensor) -> torch.Tensor:
        x = self.c_fc(x)
        x = F.relu(x).square() # https://arxiv.org/abs/2109.08668v2; ~1-2% better than GELU; suggested by @SKYLINEZ007 and @Grad62304977
        x = self.c_proj(x)
        return x

class Block(nn.Module):

    def __init__(self, config):
        super().__init__()
        self.attn = CausalSelfAttention(config.n_embd, config.n_head)
        self.mlp = MLP(config.n_embd)
        self.lambdas = nn.Parameter(torch.tensor([1., 0.]))

    def forward(self, x: torch.Tensor, vi: torch.Tensor, x0: torch.Tensor, block_mask: BlockMask) -> torch.Tensor:
        x = self.lambdas[0] * x + self.lambdas[1] * x0
        x = x + self.attn(norm(x), vi, block_mask)
        x = x + self.mlp(norm(x))
        return x

# -----------------------------------------------------------------------------
# The main GPT-2 model

@dataclass
class GPTConfig:
    vocab_size : int = 50304
    n_layer : int = 12
    n_head : int = 6 # head dim 128 suggested by @Grad62304977
    n_embd : int = 768
    lm_head_softcap : int = 30

class GPT(nn.Module):

    def __init__(self, config: GPTConfig):
        super().__init__()
        self.n_layer = config.n_layer
        self.lm_head_softcap = config.lm_head_softcap

        # U-net design by @brendanh0gan
        self.num_encoder_layers = config.n_layer // 2 # Half of the layers for encoder
        self.num_decoder_layers = config.n_layer - self.num_encoder_layers # Remaining for decoder
        # Add learnable skip connection weights for decoder layers
        self.skip_weights = nn.Parameter(torch.ones(self.num_decoder_layers))

        self.transformer = nn.ModuleDict(dict(
            wte = nn.Embedding(config.vocab_size, config.n_embd),
            # token value embeddings by @KoszarskyB - inspired by @Grad62304977's value residual learning
            # U-net structure on token value embeddings by @leloykun
            vte = nn.Embedding(config.vocab_size, config.n_embd*self.num_encoder_layers),
            h = nn.ModuleList([Block(config) for _ in range(config.n_layer)]),
        ))
        self.lm_head = CastedLinear(config.n_embd, config.vocab_size)
        self.lm_head.weight.data.zero_() # @Grad62304977

    def forward(self, idx: torch.Tensor, target: torch.Tensor, sliding_window: torch.Tensor) -> torch.Tensor:
        BLOCK_SIZE = 128
        assert idx.ndim == 1
        docs = (idx == 50256).cumsum(0)
        docs_low = docs.reshape(-1, BLOCK_SIZE)[:, 0].contiguous()
        docs_high = docs.reshape(-1, BLOCK_SIZE)[:, -1].contiguous()
        def document_sliding_window_causal(b, h, q_idx, kv_idx):
            causal_mask = q_idx >= kv_idx
            document_mask = docs[q_idx] == docs[kv_idx]
            window_mask = q_idx - kv_idx < sliding_window
            return causal_mask & document_mask & window_mask

        S = len(idx)
        def create_sliding_window_causal_mask(S: int, sliding_window: torch.Tensor):
            kv_idx = block_idx = torch.arange(S // BLOCK_SIZE, dtype=torch.int32, device="cuda")
            q_idx = block_idx[:, None]
            causal_mask = q_idx >= kv_idx
            document_mask = (docs_low[q_idx] <= docs_high[kv_idx]) & (docs_low[kv_idx] <= docs_high[q_idx])
            window_mask = q_idx - kv_idx < ((sliding_window + BLOCK_SIZE - 1) // BLOCK_SIZE)
            dense_mask = causal_mask & document_mask & window_mask
            dense_mask = dense_mask.to(torch.int32)
            num_blocks = dense_mask.sum(dim=-1).to(torch.int32)
            indices = torch.argsort(dense_mask, dim=-1, descending=True, stable=True).to(torch.int32)
            num_blocks = num_blocks[None, None, :].contiguous()
            indices = indices[None, None, :].contiguous()
            return BlockMask.from_kv_blocks(num_blocks, indices, BLOCK_SIZE=BLOCK_SIZE, mask_mod=document_sliding_window_causal)
        block_mask = create_sliding_window_causal_mask(S, sliding_window)

        # forward the GPT model itself
        x = self.transformer.wte(idx[None]) # token embeddings of shape (b, t, n_embd)
        x = norm(x) # @Grad62304977
        x0 = x
        vi = self.transformer.vte(idx[None]).chunk(self.num_encoder_layers, dim=-1)

        # Store outputs for U-Net skip connections
        skip_connections = []
        # Encoder pass - process only the first half of the blocks
        for i in range(self.num_encoder_layers):
            x = self.transformer.h[i](x, vi[i], x0, block_mask)
            skip_connections.append(x)
        # Decoder pass - process the remaining blocks with weighted skip connections
        for i in range(self.num_decoder_layers):
            x = x + self.skip_weights[i] * skip_connections.pop()
            # U-net structure on token value embeddings by @leloykun
            x = self.transformer.h[self.num_encoder_layers + i](x, vi[self.num_encoder_layers-1-i], x0, block_mask)

        x = norm(x)
        logits = self.lm_head(x)
        logits = self.lm_head_softcap * torch.tanh(logits / self.lm_head_softcap) # @Grad62304977
        logits = logits.float()
        loss = F.cross_entropy(logits.view(-1, logits.size(-1)), target.view(-1))
        return loss

# -----------------------------------------------------------------------------
# Our own simple Distributed Data Loader

def _peek_data_shard(file: Path):
    # only reads the header, returns header data
    # header is 256 int32
    header = torch.from_file(f"{file}", False, 256, dtype=torch.int32)
    assert header[0] == 20240520, "magic number mismatch in the data .bin file"
    assert header[1] == 1, "unsupported version"
    return int(header[2]) # number of tokens (claimed)

def _load_data_shard(file: Path, ntok: int):
    with file.open("rb") as f:
        tokens = torch.empty(ntok, dtype=torch.uint16, pin_memory=True)
        f.seek(256 * 4)
        nbytes = f.readinto(tokens.numpy())
        assert nbytes == 2 * ntok, "number of tokens read does not match header?"
    return tokens

class DistributedDataLoader:
    def __init__(self, filename_pattern, T, process_rank, num_processes):
        self.process_rank = process_rank
        self.num_processes = num_processes
        self.T = T

        # glob files that match the pattern
        self.files = sorted(Path.cwd().glob(filename_pattern))
        assert len(self.files) > 0, f"did not find any files that match the pattern {filename_pattern}"

        # load and validate all data shards, count number of tokens in total
        self.ntoks = [_peek_data_shard(file) for file in self.files]
        assert min(self.ntoks) >= num_processes * T + 1
        self.ntok_total = sum(self.ntoks)

        self.reset()

    def reset(self):
        self.current_shard = -1
        self.advance()

    def advance(self): # advance to next data shard
        self.current_shard = (self.current_shard + 1) % len(self.files)
        self.current_position = self.process_rank * self.T
        self.tokens = _load_data_shard(self.files[self.current_shard], self.ntoks[self.current_shard])

    def next_batch(self):
        batch_size = self.T * self.num_processes
        buf = self.tokens[self.current_position:self.current_position+self.T+1]
        # host side async is sufficient;
        # no performance improvement was observed when introducing a separate stream.
        x = buf[:-1].to(device="cuda", dtype=torch.int32, non_blocking=True) # inputs
        y = buf[1:].to(device="cuda", dtype=torch.int64, non_blocking=True) # targets
        # advance current position and load next shard if necessary
        self.current_position += batch_size
        if self.current_position + batch_size + 1 >= len(self.tokens):
            self.advance()
        return x, y

# -----------------------------------------------------------------------------
# int main

@dataclass
class Hyperparameters:
    # data hyperparams
    input_bin : str = 'data/fineweb10B/fineweb_train_*.bin' # input .bin to train on
    input_val_bin : str = 'data/fineweb10B/fineweb_val_*.bin' # input .bin to eval validation loss on
    # optimization hyperparams
    batch_size : int = 8 # batch size, in sequences, across all devices
    sequence_length : int = 64*1024 # sequence length, in tokens
    num_iterations : int = 1480 # number of iterations to run
    warmup_iters : int = 0
    cooldown_iters : int = 600 # number of iterations of linear warmup/cooldown for triangular or trapezoidal schedule
    weight_decay : float = 0
    # evaluation and logging hyperparams
    val_loss_every : int = 125 # every how many steps to evaluate val loss? 0 for only at the end
    val_tokens : int = 10485760 # how many tokens of validation data? it's important to keep this fixed for consistent comparisons
    save_every : int = 0 # every how many steps to save the checkpoint? 0 for only at the end
args = Hyperparameters()

# set up DDP (distributed data parallel). torchrun sets this env variable
assert torch.cuda.is_available()
dist.init_process_group(backend='nccl')
ddp_rank = int(os.environ['RANK'])
ddp_local_rank = int(os.environ['LOCAL_RANK'])
ddp_world_size = int(os.environ['WORLD_SIZE'])
device = f'cuda:{ddp_local_rank}'
torch.cuda.set_device(device)
print(f"using device: {device}")
master_process = (ddp_rank == 0) # this process will do logging, checkpointing etc.

# begin logging
logfile = None
if master_process:
    run_id = str(uuid.uuid4())
    logdir = 'logs/%s/' % run_id
    # os.makedirs(logdir, exist_ok=True)
    logfile = 'logs/%s.txt' % run_id
    # create the log file
    with open(logfile, "w") as f:
        # begin the log by printing this file (the Python code)
        f.write(code)
        f.write('='*100 + '\n')
def print0(s, logonly=False):
    if master_process:
        with open(logfile, "a") as f:
            if not logonly:
                print(s)
            f.write(s+'\n')
# log information about the hardware/software environment this is running on
# and print the full `nvidia-smi` to file
print0(f"Running pytorch {torch.version.__version__} compiled for CUDA {torch.version.cuda}\nnvidia-smi:")
import subprocess
result = subprocess.run(['nvidia-smi'], stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
print0(f'{result.stdout}', logonly=True)
print0('='*100, logonly=True)

# convenience variables
T = args.sequence_length
# calculate the number of steps to take in the val loop.
assert args.val_tokens % (T * ddp_world_size) == 0
val_steps = args.val_tokens // (T * ddp_world_size)
# calculate the steps of gradient accumulation required to attain the desired global batch size.
assert args.batch_size % (ddp_world_size) == 0
train_accumulation_steps = args.batch_size // ddp_world_size
assert train_accumulation_steps == 1

# load tokens
train_loader = DistributedDataLoader(args.input_bin, T, ddp_rank, ddp_world_size)
val_loader = DistributedDataLoader(args.input_val_bin, T, ddp_rank, ddp_world_size)
print0(f"Training DataLoader: total number of tokens: {train_loader.ntok_total} across {len(train_loader.files)} files")
print0(f"Validation DataLoader: total number of tokens: {val_loader.ntok_total} across {len(val_loader.files)} files")
print0('='*100, logonly=True)
x, y = train_loader.next_batch()

# 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.
num_vocab = 50304
model = GPT(GPTConfig(vocab_size=num_vocab, n_layer=12, n_head=6, n_embd=768))
model = model.cuda().bfloat16()
for m in model.modules():
    if isinstance(m, CastedLinear):
        m.float()
if hasattr(config, "coordinate_descent_tuning"):
    config.coordinate_descent_tuning = True # suggested by @Chillee
model = torch.compile(model)
# here we wrap model into DDP container
model = DDP(model, device_ids=[ddp_local_rank])
raw_model = model.module # always contains the "raw" unwrapped model

# init the optimizer(s)
optimizer1 = torch.optim.Adam([raw_model.transformer.wte.weight, raw_model.transformer.vte.weight], lr=0.6, betas=(0.8, 0.95), fused=True)
optimizer2 = torch.optim.Adam([raw_model.lm_head.weight], lr=0.008, betas=(0.8, 0.95), fused=True)
params = list(raw_model.transformer.h.parameters())
matrix_params = [p for p in params if p.ndim == 2]
scalar_params = [p for p in params if p.ndim < 2] + [raw_model.skip_weights]
optimizer3 = Muon(matrix_params, lr=0.05, momentum=0.95)
optimizer4 = torch.optim.Adam(scalar_params, lr=0.04, betas=(0.8, 0.95), fused=True)
optimizers = [optimizer1, optimizer2, optimizer3, optimizer4]
# learning rate decay scheduler (linear warmup and cooldown)
def get_lr(it):
    assert it <= args.num_iterations
    # 1) linear warmup for warmup_iters steps
    if it < args.warmup_iters:
        return (it+1) / args.warmup_iters
    # 2) constant lr for a while
    elif it < args.num_iterations - args.cooldown_iters:
        return 1.0
    # 3) linear cooldown
    else:
        decay_ratio = (args.num_iterations - it) / args.cooldown_iters
        return decay_ratio
schedulers = [torch.optim.lr_scheduler.LambdaLR(opt, get_lr) for opt in optimizers]

sliding_window_size = torch.tensor(64, dtype=torch.int32, device="cuda")
sw_size_prev = 64
# Start training loop
training_time_ms = 0
# start the clock
torch.cuda.synchronize()
t0 = time.perf_counter()
# begin training
for step in range(args.num_iterations + 1):
    last_step = (step == args.num_iterations)
    # This effectively ignores timing first 10 steps, which are slower for weird reasons.
    # Alternately, and slightly more correctly in terms of benchmarking, we could do 10
    # steps with dummy data first, and then re-initialize the model and reset the loader.
    if step == 10:
        training_time_ms = 0
        t0 = time.perf_counter()
    timed_steps = float('nan') if step <= 11 else (step - 10) + 1 # <= 11 to avoid bug in val

    # Set the sliding window size for the current step, in chunks of 64. By @fernbear.bsky.social
    sw_size =  64 * int((64 + (1792 - 64) * step / args.num_iterations) // 64)
    if sw_size != sw_size_prev:
        sliding_window_size.copy_(sw_size, non_blocking=True)
        sw_size_prev = sw_size

    # once in a while evaluate the validation dataset
    if (last_step or (args.val_loss_every > 0 and step % args.val_loss_every == 0)):
        # stop the clock
        torch.cuda.synchronize()
        training_time_ms += 1000 * (time.perf_counter() - t0)
        # run validation batches
        model.eval()
        val_loader.reset()
        val_loss = 0.0
        for _ in range(val_steps):
            with torch.no_grad():
                x_val, y_val = val_loader.next_batch()
                val_loss += model(x_val, y_val, sliding_window=sliding_window_size)
        dist.all_reduce(val_loss, op=dist.ReduceOp.AVG)
        val_loss /= val_steps
        # log val loss to console and to logfile
        print0(f'step:{step}/{args.num_iterations} val_loss:{val_loss:.4f} train_time:{training_time_ms:.0f}ms step_avg:{training_time_ms/(timed_steps-1):.2f}ms')
        # start the clock again
        torch.cuda.synchronize()
        t0 = time.perf_counter()

    if master_process and (last_step or (args.save_every > 0 and step % args.save_every == 0)):
        # stop the clock
        torch.cuda.synchronize()
        training_time_ms += 1000 * (time.perf_counter() - t0)
        # save the state of the training process
        log = dict(step=step, code=code, model=raw_model.state_dict(), optimizers=[opt.state_dict() for opt in optimizers])
        # torch.save(log, 'logs/%s/state_step%06d.pt' % (run_id, step))
        # start the clock again
        torch.cuda.synchronize()
        t0 = time.perf_counter()

    # bit confusing: we want to make sure to eval on 0th iteration
    # but also after the very last iteration. so we loop for step <= num_iterations
    # instead of just < num_iterations (one extra due to <=), only to do
    # the validation/sampling one last time, and then we break right here as we're done.
    if last_step:
        break

    # --------------- TRAINING SECTION BEGIN -----------------
    model.train()
    loss = model(x, y, sliding_window=sliding_window_size)
    loss.backward()
    del loss
    # advance the dataset for the next batch
    x, y = train_loader.next_batch()
    # momentum warmup for Muon
    frac = min(step/300, 1)
    for group in optimizer3.param_groups:
        group['momentum'] = (1 - frac) * 0.85 + frac * 0.95
    # step the optimizers and schedulers
    for opt, sched in zip(optimizers, schedulers):
        opt.step()
        sched.step()
    # null the gradients
    model.zero_grad(set_to_none=True)
    # --------------- TRAINING SECTION END -------------------
    # everything that follows now is just diagnostics, prints, logging, etc.
    approx_time = training_time_ms + 1000 * (time.perf_counter() - t0)
    print0(f"step:{step+1}/{args.num_iterations} train_time:{approx_time:.0f}ms step_avg:{approx_time/timed_steps:.2f}ms")

if master_process:
    print(f"peak memory consumption: {torch.cuda.max_memory_allocated() // 1024 // 1024} MiB")

# -------------------------------------------------------------------------
# clean up nice
dist.destroy_process_group()
====================================================================================================
Running pytorch 2.6.0.dev20241203+cu124 compiled for CUDA 12.4
nvidia-smi:
Sun Dec  8 09:05:28 2024       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.129.03             Driver Version: 535.129.03   CUDA Version: 12.6     |
|-----------------------------------------+----------------------+----------------------+
| 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  | 00000000:65:02.0 Off |                    0 |
| N/A   37C    P0              74W / 700W |      7MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   1  NVIDIA H100 80GB HBM3          On  | 00000000:67:02.0 Off |                    0 |
| N/A   46C    P0             131W / 700W |    533MiB / 81559MiB |      1%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   2  NVIDIA H100 80GB HBM3          On  | 00000000:69:02.0 Off |                    0 |
| N/A   45C    P0              75W / 700W |     19MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   3  NVIDIA H100 80GB HBM3          On  | 00000000:6B:02.0 Off |                    0 |
| N/A   39C    P0             119W / 700W |    533MiB / 81559MiB |      1%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   4  NVIDIA H100 80GB HBM3          On  | 00000000:6F:02.0 Off |                    0 |
| N/A   39C    P0              97W / 700W |     27MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   5  NVIDIA H100 80GB HBM3          On  | 00000000:71:02.0 Off |                    0 |
| N/A   46C    P0             117W / 700W |    533MiB / 81559MiB |      1%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   6  NVIDIA H100 80GB HBM3          On  | 00000000:73:02.0 Off |                    0 |
| N/A   46C    P0             107W / 700W |     27MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   7  NVIDIA H100 80GB HBM3          On  | 00000000:75:02.0 Off |                    0 |
| N/A   39C    P0             124W / 700W |    533MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
                                                                                         
+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
+---------------------------------------------------------------------------------------+

====================================================================================================
Training DataLoader: total number of tokens: 3200000000 across 32 files
Validation DataLoader: total number of tokens: 100000000 across 1 files
====================================================================================================
step:0/1480 val_loss:10.8258 train_time:0ms step_avg:nanms
step:1/1480 train_time:22661ms step_avg:nanms
step:2/1480 train_time:22748ms step_avg:nanms
step:3/1480 train_time:22887ms step_avg:nanms
step:4/1480 train_time:23028ms step_avg:nanms
step:5/1480 train_time:23171ms step_avg:nanms
step:6/1480 train_time:23311ms step_avg:nanms
step:7/1480 train_time:23453ms step_avg:nanms
step:8/1480 train_time:23596ms step_avg:nanms
step:9/1480 train_time:23742ms step_avg:nanms
step:10/1480 train_time:23885ms step_avg:nanms
step:11/1480 train_time:141ms step_avg:nanms
step:12/1480 train_time:282ms step_avg:nanms
step:13/1480 train_time:423ms step_avg:140.92ms
step:14/1480 train_time:565ms step_avg:141.18ms
step:15/1480 train_time:709ms step_avg:141.74ms
step:16/1480 train_time:854ms step_avg:142.32ms
step:17/1480 train_time:996ms step_avg:142.35ms
step:18/1480 train_time:1138ms step_avg:142.30ms
step:19/1480 train_time:1280ms step_avg:142.21ms
step:20/1480 train_time:1420ms step_avg:142.04ms
step:21/1480 train_time:1562ms step_avg:141.99ms
step:22/1480 train_time:1706ms step_avg:142.16ms
step:23/1480 train_time:1851ms step_avg:142.39ms
step:24/1480 train_time:1995ms step_avg:142.53ms
step:25/1480 train_time:2139ms step_avg:142.60ms
step:26/1480 train_time:2281ms step_avg:142.59ms
step:27/1480 train_time:2422ms step_avg:142.48ms
step:28/1480 train_time:2564ms step_avg:142.43ms
step:29/1480 train_time:2706ms step_avg:142.42ms
step:30/1480 train_time:2850ms step_avg:142.49ms
step:31/1480 train_time:2993ms step_avg:142.53ms
step:32/1480 train_time:3135ms step_avg:142.49ms
step:33/1480 train_time:3278ms step_avg:142.51ms
step:34/1480 train_time:3418ms step_avg:142.43ms
step:35/1480 train_time:3561ms step_avg:142.42ms
step:36/1480 train_time:3703ms step_avg:142.43ms
step:37/1480 train_time:3848ms step_avg:142.50ms
step:38/1480 train_time:3990ms step_avg:142.51ms
step:39/1480 train_time:4133ms step_avg:142.51ms
step:40/1480 train_time:4276ms step_avg:142.53ms
step:41/1480 train_time:4417ms step_avg:142.49ms
step:42/1480 train_time:4558ms step_avg:142.45ms
step:43/1480 train_time:4701ms step_avg:142.46ms
step:44/1480 train_time:4847ms step_avg:142.56ms
step:45/1480 train_time:4991ms step_avg:142.60ms
step:46/1480 train_time:5134ms step_avg:142.62ms
step:47/1480 train_time:5277ms step_avg:142.63ms
step:48/1480 train_time:5419ms step_avg:142.60ms
step:49/1480 train_time:5560ms step_avg:142.56ms
step:50/1480 train_time:5701ms step_avg:142.53ms
step:51/1480 train_time:5843ms step_avg:142.52ms
step:52/1480 train_time:5985ms step_avg:142.50ms
step:53/1480 train_time:6127ms step_avg:142.49ms
step:54/1480 train_time:6270ms step_avg:142.51ms
step:55/1480 train_time:6413ms step_avg:142.52ms
step:56/1480 train_time:6554ms step_avg:142.49ms
step:57/1480 train_time:6696ms step_avg:142.47ms
step:58/1480 train_time:6837ms step_avg:142.45ms
step:59/1480 train_time:6979ms step_avg:142.42ms
step:60/1480 train_time:7120ms step_avg:142.40ms
step:61/1480 train_time:7262ms step_avg:142.38ms
step:62/1480 train_time:7403ms step_avg:142.36ms
step:63/1480 train_time:7547ms step_avg:142.39ms
step:64/1480 train_time:7691ms step_avg:142.44ms
step:65/1480 train_time:7834ms step_avg:142.43ms
step:66/1480 train_time:7976ms step_avg:142.42ms
step:67/1480 train_time:8117ms step_avg:142.40ms
step:68/1480 train_time:8259ms step_avg:142.40ms
step:69/1480 train_time:8401ms step_avg:142.39ms
step:70/1480 train_time:8542ms step_avg:142.37ms
step:71/1480 train_time:8685ms step_avg:142.38ms
step:72/1480 train_time:8827ms step_avg:142.37ms
step:73/1480 train_time:8971ms step_avg:142.40ms
step:74/1480 train_time:9114ms step_avg:142.41ms
step:75/1480 train_time:9256ms step_avg:142.40ms
step:76/1480 train_time:9398ms step_avg:142.40ms
step:77/1480 train_time:9539ms step_avg:142.38ms
step:78/1480 train_time:9682ms step_avg:142.38ms
step:79/1480 train_time:9823ms step_avg:142.36ms
step:80/1480 train_time:9966ms step_avg:142.36ms
step:81/1480 train_time:10109ms step_avg:142.38ms
step:82/1480 train_time:10252ms step_avg:142.39ms
step:83/1480 train_time:10395ms step_avg:142.40ms
step:84/1480 train_time:10537ms step_avg:142.39ms
step:85/1480 train_time:10680ms step_avg:142.40ms
step:86/1480 train_time:10821ms step_avg:142.38ms
step:87/1480 train_time:10962ms step_avg:142.36ms
step:88/1480 train_time:11103ms step_avg:142.35ms
step:89/1480 train_time:11245ms step_avg:142.35ms
step:90/1480 train_time:11387ms step_avg:142.34ms
step:91/1480 train_time:11531ms step_avg:142.36ms
step:92/1480 train_time:11674ms step_avg:142.37ms
step:93/1480 train_time:11816ms step_avg:142.37ms
step:94/1480 train_time:11958ms step_avg:142.35ms
step:95/1480 train_time:12100ms step_avg:142.35ms
step:96/1480 train_time:12242ms step_avg:142.35ms
step:97/1480 train_time:12385ms step_avg:142.36ms
step:98/1480 train_time:12528ms step_avg:142.36ms
step:99/1480 train_time:12672ms step_avg:142.38ms
step:100/1480 train_time:12814ms step_avg:142.38ms
step:101/1480 train_time:12955ms step_avg:142.36ms
step:102/1480 train_time:13097ms step_avg:142.36ms
step:103/1480 train_time:13240ms step_avg:142.36ms
step:104/1480 train_time:13381ms step_avg:142.35ms
step:105/1480 train_time:13522ms step_avg:142.34ms
step:106/1480 train_time:13665ms step_avg:142.35ms
step:107/1480 train_time:13808ms step_avg:142.35ms
step:108/1480 train_time:13951ms step_avg:142.36ms
step:109/1480 train_time:14093ms step_avg:142.36ms
step:110/1480 train_time:14236ms step_avg:142.36ms
step:111/1480 train_time:14381ms step_avg:142.39ms
step:112/1480 train_time:14527ms step_avg:142.42ms
step:113/1480 train_time:14675ms step_avg:142.47ms
step:114/1480 train_time:14820ms step_avg:142.50ms
step:115/1480 train_time:14967ms step_avg:142.54ms
step:116/1480 train_time:15115ms step_avg:142.59ms
step:117/1480 train_time:15260ms step_avg:142.62ms
step:118/1480 train_time:15406ms step_avg:142.65ms
step:119/1480 train_time:15554ms step_avg:142.70ms
step:120/1480 train_time:15701ms step_avg:142.73ms
step:121/1480 train_time:15848ms step_avg:142.77ms
step:122/1480 train_time:15996ms step_avg:142.82ms
step:123/1480 train_time:16142ms step_avg:142.85ms
step:124/1480 train_time:16288ms step_avg:142.88ms
step:125/1480 train_time:16435ms step_avg:142.91ms
step:125/1480 val_loss:4.4112 train_time:16491ms step_avg:143.40ms
step:126/1480 train_time:16587ms step_avg:142.99ms
step:127/1480 train_time:16736ms step_avg:143.04ms
step:128/1480 train_time:16882ms step_avg:143.07ms
step:129/1480 train_time:17029ms step_avg:143.10ms
step:130/1480 train_time:17173ms step_avg:143.11ms
step:131/1480 train_time:17320ms step_avg:143.14ms
step:132/1480 train_time:17467ms step_avg:143.17ms
step:133/1480 train_time:17615ms step_avg:143.22ms
step:134/1480 train_time:17765ms step_avg:143.26ms
step:135/1480 train_time:17911ms step_avg:143.29ms
step:136/1480 train_time:18057ms step_avg:143.31ms
step:137/1480 train_time:18205ms step_avg:143.34ms
step:138/1480 train_time:18350ms step_avg:143.36ms
step:139/1480 train_time:18496ms step_avg:143.38ms
step:140/1480 train_time:18645ms step_avg:143.42ms
step:141/1480 train_time:18792ms step_avg:143.45ms
step:142/1480 train_time:18939ms step_avg:143.48ms
step:143/1480 train_time:19086ms step_avg:143.50ms
step:144/1480 train_time:19231ms step_avg:143.52ms
step:145/1480 train_time:19376ms step_avg:143.53ms
step:146/1480 train_time:19523ms step_avg:143.55ms
step:147/1480 train_time:19670ms step_avg:143.58ms
step:148/1480 train_time:19816ms step_avg:143.60ms
step:149/1480 train_time:19963ms step_avg:143.62ms
step:150/1480 train_time:20110ms step_avg:143.64ms
step:151/1480 train_time:20255ms step_avg:143.65ms
step:152/1480 train_time:20402ms step_avg:143.68ms
step:153/1480 train_time:20549ms step_avg:143.70ms
step:154/1480 train_time:20695ms step_avg:143.71ms
step:155/1480 train_time:20842ms step_avg:143.74ms
step:156/1480 train_time:20989ms step_avg:143.76ms
step:157/1480 train_time:21135ms step_avg:143.78ms
step:158/1480 train_time:21283ms step_avg:143.81ms
step:159/1480 train_time:21430ms step_avg:143.83ms
step:160/1480 train_time:21574ms step_avg:143.83ms
step:161/1480 train_time:21721ms step_avg:143.85ms
step:162/1480 train_time:21868ms step_avg:143.87ms
step:163/1480 train_time:22015ms step_avg:143.89ms
step:164/1480 train_time:22163ms step_avg:143.91ms
step:165/1480 train_time:22310ms step_avg:143.93ms
step:166/1480 train_time:22455ms step_avg:143.94ms
step:167/1480 train_time:22602ms step_avg:143.96ms
step:168/1480 train_time:22748ms step_avg:143.98ms
step:169/1480 train_time:22894ms step_avg:143.99ms
step:170/1480 train_time:23043ms step_avg:144.02ms
step:171/1480 train_time:23190ms step_avg:144.04ms
step:172/1480 train_time:23336ms step_avg:144.05ms
step:173/1480 train_time:23484ms step_avg:144.07ms
step:174/1480 train_time:23631ms step_avg:144.09ms
step:175/1480 train_time:23777ms step_avg:144.10ms
step:176/1480 train_time:23924ms step_avg:144.12ms
step:177/1480 train_time:24070ms step_avg:144.13ms
step:178/1480 train_time:24217ms step_avg:144.15ms
step:179/1480 train_time:24365ms step_avg:144.17ms
step:180/1480 train_time:24511ms step_avg:144.18ms
step:181/1480 train_time:24656ms step_avg:144.19ms
step:182/1480 train_time:24804ms step_avg:144.21ms
step:183/1480 train_time:24951ms step_avg:144.23ms
step:184/1480 train_time:25097ms step_avg:144.23ms
step:185/1480 train_time:25244ms step_avg:144.25ms
step:186/1480 train_time:25390ms step_avg:144.26ms
step:187/1480 train_time:25538ms step_avg:144.28ms
step:188/1480 train_time:25686ms step_avg:144.31ms
step:189/1480 train_time:25832ms step_avg:144.31ms
step:190/1480 train_time:25978ms step_avg:144.32ms
step:191/1480 train_time:26125ms step_avg:144.34ms
step:192/1480 train_time:26270ms step_avg:144.34ms
step:193/1480 train_time:26416ms step_avg:144.35ms
step:194/1480 train_time:26563ms step_avg:144.36ms
step:195/1480 train_time:26710ms step_avg:144.38ms
step:196/1480 train_time:26855ms step_avg:144.38ms
step:197/1480 train_time:27004ms step_avg:144.41ms
step:198/1480 train_time:27151ms step_avg:144.42ms
step:199/1480 train_time:27297ms step_avg:144.43ms
step:200/1480 train_time:27445ms step_avg:144.45ms
step:201/1480 train_time:27590ms step_avg:144.45ms
step:202/1480 train_time:27736ms step_avg:144.46ms
step:203/1480 train_time:27883ms step_avg:144.47ms
step:204/1480 train_time:28030ms step_avg:144.48ms
step:205/1480 train_time:28175ms step_avg:144.49ms
step:206/1480 train_time:28322ms step_avg:144.50ms
step:207/1480 train_time:28469ms step_avg:144.51ms
step:208/1480 train_time:28614ms step_avg:144.52ms
step:209/1480 train_time:28762ms step_avg:144.53ms
step:210/1480 train_time:28909ms step_avg:144.54ms
step:211/1480 train_time:29055ms step_avg:144.55ms
step:212/1480 train_time:29202ms step_avg:144.56ms
step:213/1480 train_time:29348ms step_avg:144.57ms
step:214/1480 train_time:29493ms step_avg:144.57ms
step:215/1480 train_time:29639ms step_avg:144.58ms
step:216/1480 train_time:29786ms step_avg:144.59ms
step:217/1480 train_time:29933ms step_avg:144.60ms
step:218/1480 train_time:30081ms step_avg:144.62ms
step:219/1480 train_time:30228ms step_avg:144.63ms
step:220/1480 train_time:30373ms step_avg:144.64ms
step:221/1480 train_time:30522ms step_avg:144.66ms
step:222/1480 train_time:30672ms step_avg:144.68ms
step:223/1480 train_time:30824ms step_avg:144.71ms
step:224/1480 train_time:30973ms step_avg:144.73ms
step:225/1480 train_time:31125ms step_avg:144.77ms
step:226/1480 train_time:31273ms step_avg:144.78ms
step:227/1480 train_time:31424ms step_avg:144.81ms
step:228/1480 train_time:31573ms step_avg:144.83ms
step:229/1480 train_time:31724ms step_avg:144.86ms
step:230/1480 train_time:31873ms step_avg:144.88ms
step:231/1480 train_time:32025ms step_avg:144.91ms
step:232/1480 train_time:32174ms step_avg:144.93ms
step:233/1480 train_time:32325ms step_avg:144.96ms
step:234/1480 train_time:32475ms step_avg:144.98ms
step:235/1480 train_time:32625ms step_avg:145.00ms
step:236/1480 train_time:32774ms step_avg:145.02ms
step:237/1480 train_time:32924ms step_avg:145.04ms
step:238/1480 train_time:33074ms step_avg:145.06ms
step:239/1480 train_time:33226ms step_avg:145.09ms
step:240/1480 train_time:33376ms step_avg:145.11ms
step:241/1480 train_time:33526ms step_avg:145.14ms
step:242/1480 train_time:33676ms step_avg:145.15ms
step:243/1480 train_time:33826ms step_avg:145.18ms
step:244/1480 train_time:33975ms step_avg:145.19ms
step:245/1480 train_time:34126ms step_avg:145.22ms
step:246/1480 train_time:34276ms step_avg:145.24ms
step:247/1480 train_time:34426ms step_avg:145.26ms
step:248/1480 train_time:34575ms step_avg:145.27ms
step:249/1480 train_time:34725ms step_avg:145.29ms
step:250/1480 train_time:34874ms step_avg:145.31ms
step:250/1480 val_loss:3.9960 train_time:34932ms step_avg:145.55ms
step:251/1480 train_time:35027ms step_avg:145.34ms
step:252/1480 train_time:35179ms step_avg:145.37ms
step:253/1480 train_time:35328ms step_avg:145.38ms
step:254/1480 train_time:35477ms step_avg:145.40ms
step:255/1480 train_time:35626ms step_avg:145.41ms
step:256/1480 train_time:35775ms step_avg:145.43ms
step:257/1480 train_time:35925ms step_avg:145.45ms
step:258/1480 train_time:36078ms step_avg:145.48ms
step:259/1480 train_time:36228ms step_avg:145.50ms
step:260/1480 train_time:36379ms step_avg:145.52ms
step:261/1480 train_time:36528ms step_avg:145.53ms
step:262/1480 train_time:36678ms step_avg:145.55ms
step:263/1480 train_time:36828ms step_avg:145.56ms
step:264/1480 train_time:36978ms step_avg:145.58ms
step:265/1480 train_time:37130ms step_avg:145.61ms
step:266/1480 train_time:37281ms step_avg:145.63ms
step:267/1480 train_time:37431ms step_avg:145.65ms
step:268/1480 train_time:37582ms step_avg:145.67ms
step:269/1480 train_time:37731ms step_avg:145.68ms
step:270/1480 train_time:37881ms step_avg:145.70ms
step:271/1480 train_time:38031ms step_avg:145.71ms
step:272/1480 train_time:38182ms step_avg:145.73ms
step:273/1480 train_time:38333ms step_avg:145.75ms
step:274/1480 train_time:38483ms step_avg:145.77ms
step:275/1480 train_time:38633ms step_avg:145.78ms
step:276/1480 train_time:38783ms step_avg:145.80ms
step:277/1480 train_time:38933ms step_avg:145.82ms
step:278/1480 train_time:39084ms step_avg:145.84ms
step:279/1480 train_time:39234ms step_avg:145.85ms
step:280/1480 train_time:39386ms step_avg:145.87ms
step:281/1480 train_time:39537ms step_avg:145.89ms
step:282/1480 train_time:39687ms step_avg:145.91ms
step:283/1480 train_time:39838ms step_avg:145.93ms
step:284/1480 train_time:39988ms step_avg:145.94ms
step:285/1480 train_time:40138ms step_avg:145.96ms
step:286/1480 train_time:40288ms step_avg:145.97ms
step:287/1480 train_time:40440ms step_avg:145.99ms
step:288/1480 train_time:40588ms step_avg:146.00ms
step:289/1480 train_time:40740ms step_avg:146.02ms
step:290/1480 train_time:40889ms step_avg:146.03ms
step:291/1480 train_time:41040ms step_avg:146.05ms
step:292/1480 train_time:41191ms step_avg:146.07ms
step:293/1480 train_time:41341ms step_avg:146.08ms
step:294/1480 train_time:41491ms step_avg:146.09ms
step:295/1480 train_time:41642ms step_avg:146.11ms
step:296/1480 train_time:41791ms step_avg:146.12ms
step:297/1480 train_time:41942ms step_avg:146.14ms
step:298/1480 train_time:42091ms step_avg:146.15ms
step:299/1480 train_time:42243ms step_avg:146.17ms
step:300/1480 train_time:42395ms step_avg:146.19ms
step:301/1480 train_time:42544ms step_avg:146.20ms
step:302/1480 train_time:42694ms step_avg:146.21ms
step:303/1480 train_time:42844ms step_avg:146.23ms
step:304/1480 train_time:42994ms step_avg:146.24ms
step:305/1480 train_time:43144ms step_avg:146.25ms
step:306/1480 train_time:43294ms step_avg:146.26ms
step:307/1480 train_time:43445ms step_avg:146.28ms
step:308/1480 train_time:43595ms step_avg:146.29ms
step:309/1480 train_time:43746ms step_avg:146.31ms
step:310/1480 train_time:43896ms step_avg:146.32ms
step:311/1480 train_time:44046ms step_avg:146.33ms
step:312/1480 train_time:44196ms step_avg:146.34ms
step:313/1480 train_time:44347ms step_avg:146.36ms
step:314/1480 train_time:44499ms step_avg:146.38ms
step:315/1480 train_time:44647ms step_avg:146.38ms
step:316/1480 train_time:44798ms step_avg:146.40ms
step:317/1480 train_time:44949ms step_avg:146.41ms
step:318/1480 train_time:45099ms step_avg:146.43ms
step:319/1480 train_time:45249ms step_avg:146.44ms
step:320/1480 train_time:45400ms step_avg:146.45ms
step:321/1480 train_time:45550ms step_avg:146.46ms
step:322/1480 train_time:45701ms step_avg:146.48ms
step:323/1480 train_time:45850ms step_avg:146.49ms
step:324/1480 train_time:46001ms step_avg:146.50ms
step:325/1480 train_time:46151ms step_avg:146.51ms
step:326/1480 train_time:46301ms step_avg:146.52ms
step:327/1480 train_time:46452ms step_avg:146.54ms
step:328/1480 train_time:46602ms step_avg:146.55ms
step:329/1480 train_time:46753ms step_avg:146.56ms
step:330/1480 train_time:46905ms step_avg:146.58ms
step:331/1480 train_time:47061ms step_avg:146.61ms
step:332/1480 train_time:47216ms step_avg:146.63ms
step:333/1480 train_time:47368ms step_avg:146.65ms
step:334/1480 train_time:47522ms step_avg:146.67ms
step:335/1480 train_time:47676ms step_avg:146.69ms
step:336/1480 train_time:47830ms step_avg:146.72ms
step:337/1480 train_time:47984ms step_avg:146.74ms
step:338/1480 train_time:48139ms step_avg:146.77ms
step:339/1480 train_time:48293ms step_avg:146.79ms
step:340/1480 train_time:48448ms step_avg:146.81ms
step:341/1480 train_time:48602ms step_avg:146.83ms
step:342/1480 train_time:48756ms step_avg:146.86ms
step:343/1480 train_time:48910ms step_avg:146.88ms
step:344/1480 train_time:49063ms step_avg:146.89ms
step:345/1480 train_time:49217ms step_avg:146.92ms
step:346/1480 train_time:49370ms step_avg:146.93ms
step:347/1480 train_time:49524ms step_avg:146.96ms
step:348/1480 train_time:49678ms step_avg:146.98ms
step:349/1480 train_time:49831ms step_avg:146.99ms
step:350/1480 train_time:49984ms step_avg:147.01ms
step:351/1480 train_time:50137ms step_avg:147.03ms
step:352/1480 train_time:50292ms step_avg:147.05ms
step:353/1480 train_time:50446ms step_avg:147.07ms
step:354/1480 train_time:50600ms step_avg:147.09ms
step:355/1480 train_time:50755ms step_avg:147.12ms
step:356/1480 train_time:50908ms step_avg:147.13ms
step:357/1480 train_time:51062ms step_avg:147.15ms
step:358/1480 train_time:51215ms step_avg:147.17ms
step:359/1480 train_time:51368ms step_avg:147.19ms
step:360/1480 train_time:51523ms step_avg:147.21ms
step:361/1480 train_time:51678ms step_avg:147.23ms
step:362/1480 train_time:51831ms step_avg:147.25ms
step:363/1480 train_time:51984ms step_avg:147.26ms
step:364/1480 train_time:52139ms step_avg:147.29ms
step:365/1480 train_time:52294ms step_avg:147.31ms
step:366/1480 train_time:52447ms step_avg:147.32ms
step:367/1480 train_time:52601ms step_avg:147.34ms
step:368/1480 train_time:52755ms step_avg:147.36ms
step:369/1480 train_time:52908ms step_avg:147.37ms
step:370/1480 train_time:53061ms step_avg:147.39ms
step:371/1480 train_time:53214ms step_avg:147.41ms
step:372/1480 train_time:53368ms step_avg:147.42ms
step:373/1480 train_time:53522ms step_avg:147.44ms
step:374/1480 train_time:53676ms step_avg:147.46ms
step:375/1480 train_time:53830ms step_avg:147.48ms
step:375/1480 val_loss:3.8150 train_time:53890ms step_avg:147.64ms
step:376/1480 train_time:53987ms step_avg:147.51ms
step:377/1480 train_time:54144ms step_avg:147.53ms
step:378/1480 train_time:54297ms step_avg:147.55ms
step:379/1480 train_time:54449ms step_avg:147.56ms
step:380/1480 train_time:54601ms step_avg:147.57ms
step:381/1480 train_time:54753ms step_avg:147.58ms
step:382/1480 train_time:54907ms step_avg:147.60ms
step:383/1480 train_time:55063ms step_avg:147.62ms
step:384/1480 train_time:55218ms step_avg:147.64ms
step:385/1480 train_time:55372ms step_avg:147.66ms
step:386/1480 train_time:55525ms step_avg:147.67ms
step:387/1480 train_time:55679ms step_avg:147.69ms
step:388/1480 train_time:55831ms step_avg:147.70ms
step:389/1480 train_time:55985ms step_avg:147.72ms
step:390/1480 train_time:56140ms step_avg:147.74ms
step:391/1480 train_time:56295ms step_avg:147.76ms
step:392/1480 train_time:56448ms step_avg:147.77ms
step:393/1480 train_time:56602ms step_avg:147.78ms
step:394/1480 train_time:56756ms step_avg:147.80ms
step:395/1480 train_time:56908ms step_avg:147.81ms
step:396/1480 train_time:57062ms step_avg:147.83ms
step:397/1480 train_time:57216ms step_avg:147.85ms
step:398/1480 train_time:57370ms step_avg:147.86ms
step:399/1480 train_time:57524ms step_avg:147.88ms
step:400/1480 train_time:57678ms step_avg:147.89ms
step:401/1480 train_time:57830ms step_avg:147.90ms
step:402/1480 train_time:57983ms step_avg:147.92ms
step:403/1480 train_time:58137ms step_avg:147.93ms
step:404/1480 train_time:58292ms step_avg:147.95ms
step:405/1480 train_time:58447ms step_avg:147.97ms
step:406/1480 train_time:58601ms step_avg:147.98ms
step:407/1480 train_time:58756ms step_avg:148.00ms
step:408/1480 train_time:58909ms step_avg:148.01ms
step:409/1480 train_time:59062ms step_avg:148.03ms
step:410/1480 train_time:59215ms step_avg:148.04ms
step:411/1480 train_time:59368ms step_avg:148.05ms
step:412/1480 train_time:59522ms step_avg:148.06ms
step:413/1480 train_time:59676ms step_avg:148.08ms
step:414/1480 train_time:59830ms step_avg:148.09ms
step:415/1480 train_time:59983ms step_avg:148.11ms
step:416/1480 train_time:60135ms step_avg:148.12ms
step:417/1480 train_time:60288ms step_avg:148.13ms
step:418/1480 train_time:60442ms step_avg:148.14ms
step:419/1480 train_time:60595ms step_avg:148.15ms
step:420/1480 train_time:60749ms step_avg:148.17ms
step:421/1480 train_time:60902ms step_avg:148.18ms
step:422/1480 train_time:61055ms step_avg:148.19ms
step:423/1480 train_time:61209ms step_avg:148.21ms
step:424/1480 train_time:61363ms step_avg:148.22ms
step:425/1480 train_time:61517ms step_avg:148.23ms
step:426/1480 train_time:61671ms step_avg:148.25ms
step:427/1480 train_time:61825ms step_avg:148.26ms
step:428/1480 train_time:61978ms step_avg:148.27ms
step:429/1480 train_time:62130ms step_avg:148.28ms
step:430/1480 train_time:62283ms step_avg:148.29ms
step:431/1480 train_time:62436ms step_avg:148.30ms
step:432/1480 train_time:62589ms step_avg:148.32ms
step:433/1480 train_time:62743ms step_avg:148.33ms
step:434/1480 train_time:62897ms step_avg:148.34ms
step:435/1480 train_time:63051ms step_avg:148.36ms
step:436/1480 train_time:63205ms step_avg:148.37ms
step:437/1480 train_time:63360ms step_avg:148.38ms
step:438/1480 train_time:63513ms step_avg:148.40ms
step:439/1480 train_time:63668ms step_avg:148.41ms
step:440/1480 train_time:63824ms step_avg:148.43ms
step:441/1480 train_time:63980ms step_avg:148.45ms
step:442/1480 train_time:64138ms step_avg:148.47ms
step:443/1480 train_time:64295ms step_avg:148.49ms
step:444/1480 train_time:64450ms step_avg:148.50ms
step:445/1480 train_time:64605ms step_avg:148.52ms
step:446/1480 train_time:64762ms step_avg:148.54ms
step:447/1480 train_time:64918ms step_avg:148.55ms
step:448/1480 train_time:65074ms step_avg:148.57ms
step:449/1480 train_time:65233ms step_avg:148.60ms
step:450/1480 train_time:65390ms step_avg:148.61ms
step:451/1480 train_time:65547ms step_avg:148.63ms
step:452/1480 train_time:65703ms step_avg:148.65ms
step:453/1480 train_time:65860ms step_avg:148.67ms
step:454/1480 train_time:66018ms step_avg:148.69ms
step:455/1480 train_time:66174ms step_avg:148.70ms
step:456/1480 train_time:66329ms step_avg:148.72ms
step:457/1480 train_time:66486ms step_avg:148.74ms
step:458/1480 train_time:66642ms step_avg:148.75ms
step:459/1480 train_time:66800ms step_avg:148.78ms
step:460/1480 train_time:66959ms step_avg:148.80ms
step:461/1480 train_time:67119ms step_avg:148.82ms
step:462/1480 train_time:67276ms step_avg:148.84ms
step:463/1480 train_time:67433ms step_avg:148.86ms
step:464/1480 train_time:67588ms step_avg:148.87ms
step:465/1480 train_time:67744ms step_avg:148.89ms
step:466/1480 train_time:67901ms step_avg:148.91ms
step:467/1480 train_time:68060ms step_avg:148.93ms
step:468/1480 train_time:68218ms step_avg:148.95ms
step:469/1480 train_time:68375ms step_avg:148.96ms
step:470/1480 train_time:68533ms step_avg:148.98ms
step:471/1480 train_time:68689ms step_avg:149.00ms
step:472/1480 train_time:68846ms step_avg:149.02ms
step:473/1480 train_time:69002ms step_avg:149.03ms
step:474/1480 train_time:69159ms step_avg:149.05ms
step:475/1480 train_time:69316ms step_avg:149.07ms
step:476/1480 train_time:69474ms step_avg:149.08ms
step:477/1480 train_time:69630ms step_avg:149.10ms
step:478/1480 train_time:69786ms step_avg:149.11ms
step:479/1480 train_time:69943ms step_avg:149.13ms
step:480/1480 train_time:70101ms step_avg:149.15ms
step:481/1480 train_time:70259ms step_avg:149.17ms
step:482/1480 train_time:70417ms step_avg:149.19ms
step:483/1480 train_time:70573ms step_avg:149.20ms
step:484/1480 train_time:70731ms step_avg:149.22ms
step:485/1480 train_time:70887ms step_avg:149.24ms
step:486/1480 train_time:71044ms step_avg:149.25ms
step:487/1480 train_time:71200ms step_avg:149.27ms
step:488/1480 train_time:71356ms step_avg:149.28ms
step:489/1480 train_time:71511ms step_avg:149.29ms
step:490/1480 train_time:71668ms step_avg:149.31ms
step:491/1480 train_time:71826ms step_avg:149.33ms
step:492/1480 train_time:71982ms step_avg:149.34ms
step:493/1480 train_time:72140ms step_avg:149.36ms
step:494/1480 train_time:72297ms step_avg:149.37ms
step:495/1480 train_time:72455ms step_avg:149.39ms
step:496/1480 train_time:72613ms step_avg:149.41ms
step:497/1480 train_time:72769ms step_avg:149.42ms
step:498/1480 train_time:72926ms step_avg:149.44ms
step:499/1480 train_time:73084ms step_avg:149.46ms
step:500/1480 train_time:73242ms step_avg:149.47ms
step:500/1480 val_loss:3.6897 train_time:73302ms step_avg:149.60ms
step:501/1480 train_time:73402ms step_avg:149.50ms
step:502/1480 train_time:73561ms step_avg:149.52ms
step:503/1480 train_time:73718ms step_avg:149.53ms
step:504/1480 train_time:73873ms step_avg:149.54ms
step:505/1480 train_time:74028ms step_avg:149.55ms
step:506/1480 train_time:74185ms step_avg:149.57ms
step:507/1480 train_time:74340ms step_avg:149.58ms
step:508/1480 train_time:74496ms step_avg:149.59ms
step:509/1480 train_time:74652ms step_avg:149.60ms
step:510/1480 train_time:74810ms step_avg:149.62ms
step:511/1480 train_time:74967ms step_avg:149.63ms
step:512/1480 train_time:75123ms step_avg:149.65ms
step:513/1480 train_time:75279ms step_avg:149.66ms
step:514/1480 train_time:75435ms step_avg:149.67ms
step:515/1480 train_time:75591ms step_avg:149.69ms
step:516/1480 train_time:75751ms step_avg:149.71ms
step:517/1480 train_time:75909ms step_avg:149.72ms
step:518/1480 train_time:76066ms step_avg:149.74ms
step:519/1480 train_time:76223ms step_avg:149.75ms
step:520/1480 train_time:76380ms step_avg:149.76ms
step:521/1480 train_time:76537ms step_avg:149.78ms
step:522/1480 train_time:76695ms step_avg:149.79ms
step:523/1480 train_time:76851ms step_avg:149.81ms
step:524/1480 train_time:77007ms step_avg:149.82ms
step:525/1480 train_time:77164ms step_avg:149.83ms
step:526/1480 train_time:77321ms step_avg:149.85ms
step:527/1480 train_time:77477ms step_avg:149.86ms
step:528/1480 train_time:77634ms step_avg:149.87ms
step:529/1480 train_time:77791ms step_avg:149.89ms
step:530/1480 train_time:77948ms step_avg:149.90ms
step:531/1480 train_time:78105ms step_avg:149.91ms
step:532/1480 train_time:78261ms step_avg:149.93ms
step:533/1480 train_time:78418ms step_avg:149.94ms
step:534/1480 train_time:78575ms step_avg:149.95ms
step:535/1480 train_time:78733ms step_avg:149.97ms
step:536/1480 train_time:78890ms step_avg:149.98ms
step:537/1480 train_time:79049ms step_avg:150.00ms
step:538/1480 train_time:79207ms step_avg:150.01ms
step:539/1480 train_time:79365ms step_avg:150.03ms
step:540/1480 train_time:79522ms step_avg:150.04ms
step:541/1480 train_time:79678ms step_avg:150.05ms
step:542/1480 train_time:79834ms step_avg:150.06ms
step:543/1480 train_time:79990ms step_avg:150.08ms
step:544/1480 train_time:80148ms step_avg:150.09ms
step:545/1480 train_time:80305ms step_avg:150.10ms
step:546/1480 train_time:80460ms step_avg:150.11ms
step:547/1480 train_time:80616ms step_avg:150.12ms
step:548/1480 train_time:80774ms step_avg:150.14ms
step:549/1480 train_time:80931ms step_avg:150.15ms
step:550/1480 train_time:81090ms step_avg:150.17ms
step:551/1480 train_time:81249ms step_avg:150.18ms
step:552/1480 train_time:81409ms step_avg:150.20ms
step:553/1480 train_time:81569ms step_avg:150.22ms
step:554/1480 train_time:81730ms step_avg:150.24ms
step:555/1480 train_time:81890ms step_avg:150.26ms
step:556/1480 train_time:82051ms step_avg:150.28ms
step:557/1480 train_time:82211ms step_avg:150.30ms
step:558/1480 train_time:82372ms step_avg:150.31ms
step:559/1480 train_time:82531ms step_avg:150.33ms
step:560/1480 train_time:82691ms step_avg:150.35ms
step:561/1480 train_time:82850ms step_avg:150.36ms
step:562/1480 train_time:83010ms step_avg:150.38ms
step:563/1480 train_time:83169ms step_avg:150.40ms
step:564/1480 train_time:83329ms step_avg:150.41ms
step:565/1480 train_time:83487ms step_avg:150.43ms
step:566/1480 train_time:83648ms step_avg:150.45ms
step:567/1480 train_time:83806ms step_avg:150.46ms
step:568/1480 train_time:83964ms step_avg:150.47ms
step:569/1480 train_time:84123ms step_avg:150.49ms
step:570/1480 train_time:84281ms step_avg:150.50ms
step:571/1480 train_time:84439ms step_avg:150.52ms
step:572/1480 train_time:84598ms step_avg:150.53ms
step:573/1480 train_time:84758ms step_avg:150.55ms
step:574/1480 train_time:84919ms step_avg:150.57ms
step:575/1480 train_time:85078ms step_avg:150.58ms
step:576/1480 train_time:85236ms step_avg:150.59ms
step:577/1480 train_time:85395ms step_avg:150.61ms
step:578/1480 train_time:85554ms step_avg:150.62ms
step:579/1480 train_time:85713ms step_avg:150.64ms
step:580/1480 train_time:85872ms step_avg:150.65ms
step:581/1480 train_time:86032ms step_avg:150.67ms
step:582/1480 train_time:86192ms step_avg:150.69ms
step:583/1480 train_time:86352ms step_avg:150.70ms
step:584/1480 train_time:86512ms step_avg:150.72ms
step:585/1480 train_time:86672ms step_avg:150.73ms
step:586/1480 train_time:86832ms step_avg:150.75ms
step:587/1480 train_time:86991ms step_avg:150.76ms
step:588/1480 train_time:87151ms step_avg:150.78ms
step:589/1480 train_time:87310ms step_avg:150.79ms
step:590/1480 train_time:87470ms step_avg:150.81ms
step:591/1480 train_time:87630ms step_avg:150.83ms
step:592/1480 train_time:87789ms step_avg:150.84ms
step:593/1480 train_time:87951ms step_avg:150.86ms
step:594/1480 train_time:88112ms step_avg:150.88ms
step:595/1480 train_time:88273ms step_avg:150.89ms
step:596/1480 train_time:88435ms step_avg:150.91ms
step:597/1480 train_time:88593ms step_avg:150.93ms
step:598/1480 train_time:88753ms step_avg:150.94ms
step:599/1480 train_time:88911ms step_avg:150.95ms
step:600/1480 train_time:89071ms step_avg:150.97ms
step:601/1480 train_time:89231ms step_avg:150.98ms
step:602/1480 train_time:89390ms step_avg:151.00ms
step:603/1480 train_time:89551ms step_avg:151.01ms
step:604/1480 train_time:89711ms step_avg:151.03ms
step:605/1480 train_time:89872ms step_avg:151.05ms
step:606/1480 train_time:90035ms step_avg:151.07ms
step:607/1480 train_time:90195ms step_avg:151.08ms
step:608/1480 train_time:90354ms step_avg:151.09ms
step:609/1480 train_time:90513ms step_avg:151.11ms
step:610/1480 train_time:90671ms step_avg:151.12ms
step:611/1480 train_time:90832ms step_avg:151.14ms
step:612/1480 train_time:90992ms step_avg:151.15ms
step:613/1480 train_time:91153ms step_avg:151.17ms
step:614/1480 train_time:91313ms step_avg:151.18ms
step:615/1480 train_time:91472ms step_avg:151.19ms
step:616/1480 train_time:91630ms step_avg:151.21ms
step:617/1480 train_time:91790ms step_avg:151.22ms
step:618/1480 train_time:91950ms step_avg:151.23ms
step:619/1480 train_time:92110ms step_avg:151.25ms
step:620/1480 train_time:92269ms step_avg:151.26ms
step:621/1480 train_time:92429ms step_avg:151.27ms
step:622/1480 train_time:92589ms step_avg:151.29ms
step:623/1480 train_time:92751ms step_avg:151.31ms
step:624/1480 train_time:92911ms step_avg:151.32ms
step:625/1480 train_time:93070ms step_avg:151.33ms
step:625/1480 val_loss:3.6063 train_time:93134ms step_avg:151.44ms
step:626/1480 train_time:93233ms step_avg:151.35ms
step:627/1480 train_time:93392ms step_avg:151.36ms
step:628/1480 train_time:93550ms step_avg:151.38ms
step:629/1480 train_time:93709ms step_avg:151.39ms
step:630/1480 train_time:93867ms step_avg:151.40ms
step:631/1480 train_time:94025ms step_avg:151.41ms
step:632/1480 train_time:94184ms step_avg:151.42ms
step:633/1480 train_time:94344ms step_avg:151.43ms
step:634/1480 train_time:94503ms step_avg:151.45ms
step:635/1480 train_time:94662ms step_avg:151.46ms
step:636/1480 train_time:94821ms step_avg:151.47ms
step:637/1480 train_time:94982ms step_avg:151.49ms
step:638/1480 train_time:95141ms step_avg:151.50ms
step:639/1480 train_time:95301ms step_avg:151.51ms
step:640/1480 train_time:95461ms step_avg:151.52ms
step:641/1480 train_time:95623ms step_avg:151.54ms
step:642/1480 train_time:95782ms step_avg:151.55ms
step:643/1480 train_time:95941ms step_avg:151.57ms
step:644/1480 train_time:96100ms step_avg:151.58ms
step:645/1480 train_time:96259ms step_avg:151.59ms
step:646/1480 train_time:96420ms step_avg:151.60ms
step:647/1480 train_time:96580ms step_avg:151.62ms
step:648/1480 train_time:96741ms step_avg:151.63ms
step:649/1480 train_time:96901ms step_avg:151.64ms
step:650/1480 train_time:97060ms step_avg:151.66ms
step:651/1480 train_time:97220ms step_avg:151.67ms
step:652/1480 train_time:97381ms step_avg:151.68ms
step:653/1480 train_time:97539ms step_avg:151.69ms
step:654/1480 train_time:97699ms step_avg:151.71ms
step:655/1480 train_time:97858ms step_avg:151.72ms
step:656/1480 train_time:98019ms step_avg:151.73ms
step:657/1480 train_time:98178ms step_avg:151.74ms
step:658/1480 train_time:98338ms step_avg:151.76ms
step:659/1480 train_time:98502ms step_avg:151.77ms
step:660/1480 train_time:98663ms step_avg:151.79ms
step:661/1480 train_time:98826ms step_avg:151.81ms
step:662/1480 train_time:98986ms step_avg:151.82ms
step:663/1480 train_time:99145ms step_avg:151.83ms
step:664/1480 train_time:99308ms step_avg:151.85ms
step:665/1480 train_time:99469ms step_avg:151.86ms
step:666/1480 train_time:99629ms step_avg:151.87ms
step:667/1480 train_time:99792ms step_avg:151.89ms
step:668/1480 train_time:99953ms step_avg:151.90ms
step:669/1480 train_time:100118ms step_avg:151.92ms
step:670/1480 train_time:100280ms step_avg:151.94ms
step:671/1480 train_time:100441ms step_avg:151.95ms
step:672/1480 train_time:100602ms step_avg:151.97ms
step:673/1480 train_time:100763ms step_avg:151.98ms
step:674/1480 train_time:100926ms step_avg:152.00ms
step:675/1480 train_time:101087ms step_avg:152.01ms
step:676/1480 train_time:101248ms step_avg:152.02ms
step:677/1480 train_time:101410ms step_avg:152.04ms
step:678/1480 train_time:101570ms step_avg:152.05ms
step:679/1480 train_time:101733ms step_avg:152.07ms
step:680/1480 train_time:101897ms step_avg:152.09ms
step:681/1480 train_time:102058ms step_avg:152.10ms
step:682/1480 train_time:102222ms step_avg:152.12ms
step:683/1480 train_time:102384ms step_avg:152.13ms
step:684/1480 train_time:102544ms step_avg:152.14ms
step:685/1480 train_time:102706ms step_avg:152.16ms
step:686/1480 train_time:102866ms step_avg:152.17ms
step:687/1480 train_time:103027ms step_avg:152.18ms
step:688/1480 train_time:103190ms step_avg:152.20ms
step:689/1480 train_time:103353ms step_avg:152.21ms
step:690/1480 train_time:103518ms step_avg:152.23ms
step:691/1480 train_time:103680ms step_avg:152.25ms
step:692/1480 train_time:103842ms step_avg:152.26ms
step:693/1480 train_time:104003ms step_avg:152.27ms
step:694/1480 train_time:104165ms step_avg:152.29ms
step:695/1480 train_time:104326ms step_avg:152.30ms
step:696/1480 train_time:104487ms step_avg:152.31ms
step:697/1480 train_time:104648ms step_avg:152.33ms
step:698/1480 train_time:104808ms step_avg:152.34ms
step:699/1480 train_time:104972ms step_avg:152.35ms
step:700/1480 train_time:105133ms step_avg:152.37ms
step:701/1480 train_time:105295ms step_avg:152.38ms
step:702/1480 train_time:105455ms step_avg:152.39ms
step:703/1480 train_time:105617ms step_avg:152.41ms
step:704/1480 train_time:105778ms step_avg:152.42ms
step:705/1480 train_time:105942ms step_avg:152.43ms
step:706/1480 train_time:106107ms step_avg:152.45ms
step:707/1480 train_time:106267ms step_avg:152.46ms
step:708/1480 train_time:106428ms step_avg:152.48ms
step:709/1480 train_time:106591ms step_avg:152.49ms
step:710/1480 train_time:106751ms step_avg:152.50ms
step:711/1480 train_time:106913ms step_avg:152.52ms
step:712/1480 train_time:107081ms step_avg:152.54ms
step:713/1480 train_time:107244ms step_avg:152.55ms
step:714/1480 train_time:107404ms step_avg:152.56ms
step:715/1480 train_time:107564ms step_avg:152.57ms
step:716/1480 train_time:107725ms step_avg:152.58ms
step:717/1480 train_time:107887ms step_avg:152.60ms
step:718/1480 train_time:108046ms step_avg:152.61ms
step:719/1480 train_time:108205ms step_avg:152.62ms
step:720/1480 train_time:108367ms step_avg:152.63ms
step:721/1480 train_time:108528ms step_avg:152.64ms
step:722/1480 train_time:108690ms step_avg:152.65ms
step:723/1480 train_time:108849ms step_avg:152.66ms
step:724/1480 train_time:109012ms step_avg:152.68ms
step:725/1480 train_time:109175ms step_avg:152.69ms
step:726/1480 train_time:109339ms step_avg:152.71ms
step:727/1480 train_time:109503ms step_avg:152.72ms
step:728/1480 train_time:109663ms step_avg:152.73ms
step:729/1480 train_time:109824ms step_avg:152.75ms
step:730/1480 train_time:109986ms step_avg:152.76ms
step:731/1480 train_time:110147ms step_avg:152.77ms
step:732/1480 train_time:110308ms step_avg:152.78ms
step:733/1480 train_time:110470ms step_avg:152.79ms
step:734/1480 train_time:110630ms step_avg:152.80ms
step:735/1480 train_time:110792ms step_avg:152.82ms
step:736/1480 train_time:110954ms step_avg:152.83ms
step:737/1480 train_time:111115ms step_avg:152.84ms
step:738/1480 train_time:111276ms step_avg:152.85ms
step:739/1480 train_time:111436ms step_avg:152.86ms
step:740/1480 train_time:111603ms step_avg:152.88ms
step:741/1480 train_time:111767ms step_avg:152.90ms
step:742/1480 train_time:111929ms step_avg:152.91ms
step:743/1480 train_time:112090ms step_avg:152.92ms
step:744/1480 train_time:112253ms step_avg:152.93ms
step:745/1480 train_time:112419ms step_avg:152.95ms
step:746/1480 train_time:112580ms step_avg:152.96ms
step:747/1480 train_time:112741ms step_avg:152.97ms
step:748/1480 train_time:112908ms step_avg:152.99ms
step:749/1480 train_time:113071ms step_avg:153.01ms
step:750/1480 train_time:113229ms step_avg:153.01ms
step:750/1480 val_loss:3.5519 train_time:113294ms step_avg:153.10ms
step:751/1480 train_time:113396ms step_avg:153.03ms
step:752/1480 train_time:113561ms step_avg:153.05ms
step:753/1480 train_time:113723ms step_avg:153.06ms
step:754/1480 train_time:113884ms step_avg:153.07ms
step:755/1480 train_time:114045ms step_avg:153.08ms
step:756/1480 train_time:114206ms step_avg:153.09ms
step:757/1480 train_time:114370ms step_avg:153.11ms
step:758/1480 train_time:114529ms step_avg:153.11ms
step:759/1480 train_time:114691ms step_avg:153.13ms
step:760/1480 train_time:114851ms step_avg:153.14ms
step:761/1480 train_time:115013ms step_avg:153.15ms
step:762/1480 train_time:115174ms step_avg:153.16ms
step:763/1480 train_time:115335ms step_avg:153.17ms
step:764/1480 train_time:115497ms step_avg:153.18ms
step:765/1480 train_time:115660ms step_avg:153.19ms
step:766/1480 train_time:115824ms step_avg:153.21ms
step:767/1480 train_time:115987ms step_avg:153.22ms
step:768/1480 train_time:116149ms step_avg:153.23ms
step:769/1480 train_time:116313ms step_avg:153.24ms
step:770/1480 train_time:116477ms step_avg:153.26ms
step:771/1480 train_time:116641ms step_avg:153.27ms
step:772/1480 train_time:116804ms step_avg:153.29ms
step:773/1480 train_time:116967ms step_avg:153.30ms
step:774/1480 train_time:117129ms step_avg:153.31ms
step:775/1480 train_time:117290ms step_avg:153.32ms
step:776/1480 train_time:117455ms step_avg:153.34ms
step:777/1480 train_time:117621ms step_avg:153.35ms
step:778/1480 train_time:117784ms step_avg:153.37ms
step:779/1480 train_time:117947ms step_avg:153.38ms
step:780/1480 train_time:118110ms step_avg:153.39ms
step:781/1480 train_time:118272ms step_avg:153.40ms
step:782/1480 train_time:118435ms step_avg:153.41ms
step:783/1480 train_time:118595ms step_avg:153.42ms
step:784/1480 train_time:118760ms step_avg:153.44ms
step:785/1480 train_time:118924ms step_avg:153.45ms
step:786/1480 train_time:119090ms step_avg:153.47ms
step:787/1480 train_time:119253ms step_avg:153.48ms
step:788/1480 train_time:119417ms step_avg:153.49ms
step:789/1480 train_time:119579ms step_avg:153.50ms
step:790/1480 train_time:119744ms step_avg:153.52ms
step:791/1480 train_time:119911ms step_avg:153.54ms
step:792/1480 train_time:120075ms step_avg:153.55ms
step:793/1480 train_time:120237ms step_avg:153.56ms
step:794/1480 train_time:120402ms step_avg:153.57ms
step:795/1480 train_time:120567ms step_avg:153.59ms
step:796/1480 train_time:120733ms step_avg:153.60ms
step:797/1480 train_time:120896ms step_avg:153.62ms
step:798/1480 train_time:121062ms step_avg:153.63ms
step:799/1480 train_time:121227ms step_avg:153.65ms
step:800/1480 train_time:121389ms step_avg:153.66ms
step:801/1480 train_time:121551ms step_avg:153.67ms
step:802/1480 train_time:121718ms step_avg:153.68ms
step:803/1480 train_time:121881ms step_avg:153.70ms
step:804/1480 train_time:122044ms step_avg:153.71ms
step:805/1480 train_time:122208ms step_avg:153.72ms
step:806/1480 train_time:122370ms step_avg:153.73ms
step:807/1480 train_time:122532ms step_avg:153.74ms
step:808/1480 train_time:122695ms step_avg:153.75ms
step:809/1480 train_time:122859ms step_avg:153.77ms
step:810/1480 train_time:123021ms step_avg:153.78ms
step:811/1480 train_time:123186ms step_avg:153.79ms
step:812/1480 train_time:123350ms step_avg:153.80ms
step:813/1480 train_time:123512ms step_avg:153.81ms
step:814/1480 train_time:123675ms step_avg:153.82ms
step:815/1480 train_time:123836ms step_avg:153.83ms
step:816/1480 train_time:124000ms step_avg:153.85ms
step:817/1480 train_time:124163ms step_avg:153.86ms
step:818/1480 train_time:124326ms step_avg:153.87ms
step:819/1480 train_time:124490ms step_avg:153.88ms
step:820/1480 train_time:124654ms step_avg:153.89ms
step:821/1480 train_time:124815ms step_avg:153.90ms
step:822/1480 train_time:124979ms step_avg:153.91ms
step:823/1480 train_time:125143ms step_avg:153.93ms
step:824/1480 train_time:125306ms step_avg:153.94ms
step:825/1480 train_time:125471ms step_avg:153.95ms
step:826/1480 train_time:125636ms step_avg:153.97ms
step:827/1480 train_time:125802ms step_avg:153.98ms
step:828/1480 train_time:125965ms step_avg:153.99ms
step:829/1480 train_time:126129ms step_avg:154.00ms
step:830/1480 train_time:126292ms step_avg:154.02ms
step:831/1480 train_time:126455ms step_avg:154.03ms
step:832/1480 train_time:126620ms step_avg:154.04ms
step:833/1480 train_time:126785ms step_avg:154.05ms
step:834/1480 train_time:126948ms step_avg:154.06ms
step:835/1480 train_time:127112ms step_avg:154.07ms
step:836/1480 train_time:127277ms step_avg:154.09ms
step:837/1480 train_time:127438ms step_avg:154.10ms
step:838/1480 train_time:127601ms step_avg:154.11ms
step:839/1480 train_time:127766ms step_avg:154.12ms
step:840/1480 train_time:127927ms step_avg:154.13ms
step:841/1480 train_time:128089ms step_avg:154.14ms
step:842/1480 train_time:128253ms step_avg:154.15ms
step:843/1480 train_time:128414ms step_avg:154.16ms
step:844/1480 train_time:128576ms step_avg:154.17ms
step:845/1480 train_time:128743ms step_avg:154.18ms
step:846/1480 train_time:128908ms step_avg:154.20ms
step:847/1480 train_time:129071ms step_avg:154.21ms
step:848/1480 train_time:129232ms step_avg:154.21ms
step:849/1480 train_time:129396ms step_avg:154.23ms
step:850/1480 train_time:129559ms step_avg:154.24ms
step:851/1480 train_time:129725ms step_avg:154.25ms
step:852/1480 train_time:129888ms step_avg:154.26ms
step:853/1480 train_time:130050ms step_avg:154.27ms
step:854/1480 train_time:130215ms step_avg:154.28ms
step:855/1480 train_time:130380ms step_avg:154.30ms
step:856/1480 train_time:130543ms step_avg:154.31ms
step:857/1480 train_time:130709ms step_avg:154.32ms
step:858/1480 train_time:130876ms step_avg:154.33ms
step:859/1480 train_time:131041ms step_avg:154.35ms
step:860/1480 train_time:131202ms step_avg:154.36ms
step:861/1480 train_time:131368ms step_avg:154.37ms
step:862/1480 train_time:131537ms step_avg:154.39ms
step:863/1480 train_time:131706ms step_avg:154.40ms
step:864/1480 train_time:131871ms step_avg:154.42ms
step:865/1480 train_time:132032ms step_avg:154.42ms
step:866/1480 train_time:132198ms step_avg:154.44ms
step:867/1480 train_time:132362ms step_avg:154.45ms
step:868/1480 train_time:132525ms step_avg:154.46ms
step:869/1480 train_time:132687ms step_avg:154.47ms
step:870/1480 train_time:132852ms step_avg:154.48ms
step:871/1480 train_time:133016ms step_avg:154.49ms
step:872/1480 train_time:133180ms step_avg:154.50ms
step:873/1480 train_time:133344ms step_avg:154.51ms
step:874/1480 train_time:133509ms step_avg:154.52ms
step:875/1480 train_time:133672ms step_avg:154.53ms
step:875/1480 val_loss:3.5022 train_time:133737ms step_avg:154.61ms
step:876/1480 train_time:133837ms step_avg:154.55ms
step:877/1480 train_time:134002ms step_avg:154.56ms
step:878/1480 train_time:134164ms step_avg:154.57ms
step:879/1480 train_time:134329ms step_avg:154.58ms
step:880/1480 train_time:134493ms step_avg:154.59ms
step:881/1480 train_time:134655ms step_avg:154.60ms
step:882/1480 train_time:134820ms step_avg:154.61ms
step:883/1480 train_time:134987ms step_avg:154.62ms
step:884/1480 train_time:135154ms step_avg:154.64ms
step:885/1480 train_time:135319ms step_avg:154.65ms
step:886/1480 train_time:135486ms step_avg:154.66ms
step:887/1480 train_time:135654ms step_avg:154.68ms
step:888/1480 train_time:135827ms step_avg:154.70ms
step:889/1480 train_time:135995ms step_avg:154.72ms
step:890/1480 train_time:136157ms step_avg:154.72ms
step:891/1480 train_time:136321ms step_avg:154.73ms
step:892/1480 train_time:136485ms step_avg:154.75ms
step:893/1480 train_time:136649ms step_avg:154.75ms
step:894/1480 train_time:136816ms step_avg:154.77ms
step:895/1480 train_time:136982ms step_avg:154.78ms
step:896/1480 train_time:137147ms step_avg:154.79ms
step:897/1480 train_time:137314ms step_avg:154.81ms
step:898/1480 train_time:137481ms step_avg:154.82ms
step:899/1480 train_time:137644ms step_avg:154.83ms
step:900/1480 train_time:137808ms step_avg:154.84ms
step:901/1480 train_time:137972ms step_avg:154.85ms
step:902/1480 train_time:138136ms step_avg:154.86ms
step:903/1480 train_time:138307ms step_avg:154.88ms
step:904/1480 train_time:138472ms step_avg:154.89ms
step:905/1480 train_time:138635ms step_avg:154.90ms
step:906/1480 train_time:138800ms step_avg:154.91ms
step:907/1480 train_time:138970ms step_avg:154.93ms
step:908/1480 train_time:139133ms step_avg:154.94ms
step:909/1480 train_time:139297ms step_avg:154.95ms
step:910/1480 train_time:139468ms step_avg:154.96ms
step:911/1480 train_time:139634ms step_avg:154.98ms
step:912/1480 train_time:139799ms step_avg:154.99ms
step:913/1480 train_time:139967ms step_avg:155.00ms
step:914/1480 train_time:140135ms step_avg:155.02ms
step:915/1480 train_time:140305ms step_avg:155.03ms
step:916/1480 train_time:140468ms step_avg:155.04ms
step:917/1480 train_time:140632ms step_avg:155.05ms
step:918/1480 train_time:140801ms step_avg:155.07ms
step:919/1480 train_time:140971ms step_avg:155.08ms
step:920/1480 train_time:141136ms step_avg:155.09ms
step:921/1480 train_time:141301ms step_avg:155.11ms
step:922/1480 train_time:141469ms step_avg:155.12ms
step:923/1480 train_time:141633ms step_avg:155.13ms
step:924/1480 train_time:141797ms step_avg:155.14ms
step:925/1480 train_time:141961ms step_avg:155.15ms
step:926/1480 train_time:142124ms step_avg:155.16ms
step:927/1480 train_time:142289ms step_avg:155.17ms
step:928/1480 train_time:142455ms step_avg:155.18ms
step:929/1480 train_time:142620ms step_avg:155.19ms
step:930/1480 train_time:142784ms step_avg:155.20ms
step:931/1480 train_time:142948ms step_avg:155.21ms
step:932/1480 train_time:143115ms step_avg:155.22ms
step:933/1480 train_time:143280ms step_avg:155.23ms
step:934/1480 train_time:143447ms step_avg:155.25ms
step:935/1480 train_time:143619ms step_avg:155.26ms
step:936/1480 train_time:143786ms step_avg:155.28ms
step:937/1480 train_time:143956ms step_avg:155.29ms
step:938/1480 train_time:144118ms step_avg:155.30ms
step:939/1480 train_time:144288ms step_avg:155.32ms
step:940/1480 train_time:144455ms step_avg:155.33ms
step:941/1480 train_time:144619ms step_avg:155.34ms
step:942/1480 train_time:144785ms step_avg:155.35ms
step:943/1480 train_time:144954ms step_avg:155.36ms
step:944/1480 train_time:145127ms step_avg:155.38ms
step:945/1480 train_time:145291ms step_avg:155.39ms
step:946/1480 train_time:145460ms step_avg:155.41ms
step:947/1480 train_time:145628ms step_avg:155.42ms
step:948/1480 train_time:145795ms step_avg:155.43ms
step:949/1480 train_time:145959ms step_avg:155.44ms
step:950/1480 train_time:146121ms step_avg:155.45ms
step:951/1480 train_time:146289ms step_avg:155.46ms
step:952/1480 train_time:146454ms step_avg:155.47ms
step:953/1480 train_time:146622ms step_avg:155.48ms
step:954/1480 train_time:146790ms step_avg:155.50ms
step:955/1480 train_time:146954ms step_avg:155.51ms
step:956/1480 train_time:147119ms step_avg:155.52ms
step:957/1480 train_time:147287ms step_avg:155.53ms
step:958/1480 train_time:147456ms step_avg:155.54ms
step:959/1480 train_time:147621ms step_avg:155.55ms
step:960/1480 train_time:147790ms step_avg:155.57ms
step:961/1480 train_time:147955ms step_avg:155.58ms
step:962/1480 train_time:148119ms step_avg:155.59ms
step:963/1480 train_time:148285ms step_avg:155.60ms
step:964/1480 train_time:148453ms step_avg:155.61ms
step:965/1480 train_time:148617ms step_avg:155.62ms
step:966/1480 train_time:148782ms step_avg:155.63ms
step:967/1480 train_time:148948ms step_avg:155.64ms
step:968/1480 train_time:149113ms step_avg:155.65ms
step:969/1480 train_time:149279ms step_avg:155.66ms
step:970/1480 train_time:149442ms step_avg:155.67ms
step:971/1480 train_time:149608ms step_avg:155.68ms
step:972/1480 train_time:149772ms step_avg:155.69ms
step:973/1480 train_time:149937ms step_avg:155.70ms
step:974/1480 train_time:150106ms step_avg:155.71ms
step:975/1480 train_time:150272ms step_avg:155.72ms
step:976/1480 train_time:150438ms step_avg:155.73ms
step:977/1480 train_time:150601ms step_avg:155.74ms
step:978/1480 train_time:150767ms step_avg:155.75ms
step:979/1480 train_time:150934ms step_avg:155.76ms
step:980/1480 train_time:151099ms step_avg:155.77ms
step:981/1480 train_time:151268ms step_avg:155.79ms
step:982/1480 train_time:151433ms step_avg:155.80ms
step:983/1480 train_time:151598ms step_avg:155.80ms
step:984/1480 train_time:151761ms step_avg:155.81ms
step:985/1480 train_time:151929ms step_avg:155.83ms
step:986/1480 train_time:152096ms step_avg:155.84ms
step:987/1480 train_time:152259ms step_avg:155.84ms
step:988/1480 train_time:152427ms step_avg:155.86ms
step:989/1480 train_time:152594ms step_avg:155.87ms
step:990/1480 train_time:152763ms step_avg:155.88ms
step:991/1480 train_time:152930ms step_avg:155.89ms
step:992/1480 train_time:153103ms step_avg:155.91ms
step:993/1480 train_time:153280ms step_avg:155.93ms
step:994/1480 train_time:153445ms step_avg:155.94ms
step:995/1480 train_time:153609ms step_avg:155.95ms
step:996/1480 train_time:153772ms step_avg:155.96ms
step:997/1480 train_time:153938ms step_avg:155.97ms
step:998/1480 train_time:154102ms step_avg:155.97ms
step:999/1480 train_time:154267ms step_avg:155.98ms
step:1000/1480 train_time:154436ms step_avg:156.00ms
step:1000/1480 val_loss:3.4408 train_time:154505ms step_avg:156.07ms
step:1001/1480 train_time:154605ms step_avg:156.01ms
step:1002/1480 train_time:154772ms step_avg:156.02ms
step:1003/1480 train_time:154945ms step_avg:156.04ms
step:1004/1480 train_time:155114ms step_avg:156.05ms
step:1005/1480 train_time:155283ms step_avg:156.06ms
step:1006/1480 train_time:155452ms step_avg:156.08ms
step:1007/1480 train_time:155618ms step_avg:156.09ms
step:1008/1480 train_time:155785ms step_avg:156.10ms
step:1009/1480 train_time:155959ms step_avg:156.12ms
step:1010/1480 train_time:156125ms step_avg:156.12ms
step:1011/1480 train_time:156289ms step_avg:156.13ms
step:1012/1480 train_time:156454ms step_avg:156.14ms
step:1013/1480 train_time:156626ms step_avg:156.16ms
step:1014/1480 train_time:156793ms step_avg:156.17ms
step:1015/1480 train_time:156964ms step_avg:156.18ms
step:1016/1480 train_time:157131ms step_avg:156.19ms
step:1017/1480 train_time:157302ms step_avg:156.21ms
step:1018/1480 train_time:157471ms step_avg:156.22ms
step:1019/1480 train_time:157641ms step_avg:156.23ms
step:1020/1480 train_time:157809ms step_avg:156.25ms
step:1021/1480 train_time:157975ms step_avg:156.26ms
step:1022/1480 train_time:158144ms step_avg:156.27ms
step:1023/1480 train_time:158310ms step_avg:156.28ms
step:1024/1480 train_time:158476ms step_avg:156.29ms
step:1025/1480 train_time:158648ms step_avg:156.30ms
step:1026/1480 train_time:158813ms step_avg:156.31ms
step:1027/1480 train_time:158980ms step_avg:156.32ms
step:1028/1480 train_time:159152ms step_avg:156.34ms
step:1029/1480 train_time:159327ms step_avg:156.36ms
step:1030/1480 train_time:159495ms step_avg:156.37ms
step:1031/1480 train_time:159660ms step_avg:156.38ms
step:1032/1480 train_time:159832ms step_avg:156.39ms
step:1033/1480 train_time:159999ms step_avg:156.40ms
step:1034/1480 train_time:160169ms step_avg:156.41ms
step:1035/1480 train_time:160337ms step_avg:156.43ms
step:1036/1480 train_time:160502ms step_avg:156.43ms
step:1037/1480 train_time:160670ms step_avg:156.45ms
step:1038/1480 train_time:160841ms step_avg:156.46ms
step:1039/1480 train_time:161010ms step_avg:156.47ms
step:1040/1480 train_time:161176ms step_avg:156.48ms
step:1041/1480 train_time:161345ms step_avg:156.49ms
step:1042/1480 train_time:161509ms step_avg:156.50ms
step:1043/1480 train_time:161674ms step_avg:156.51ms
step:1044/1480 train_time:161841ms step_avg:156.52ms
step:1045/1480 train_time:162009ms step_avg:156.53ms
step:1046/1480 train_time:162177ms step_avg:156.54ms
step:1047/1480 train_time:162344ms step_avg:156.55ms
step:1048/1480 train_time:162510ms step_avg:156.56ms
step:1049/1480 train_time:162675ms step_avg:156.57ms
step:1050/1480 train_time:162845ms step_avg:156.58ms
step:1051/1480 train_time:163014ms step_avg:156.59ms
step:1052/1480 train_time:163183ms step_avg:156.61ms
step:1053/1480 train_time:163349ms step_avg:156.61ms
step:1054/1480 train_time:163516ms step_avg:156.62ms
step:1055/1480 train_time:163683ms step_avg:156.63ms
step:1056/1480 train_time:163847ms step_avg:156.64ms
step:1057/1480 train_time:164015ms step_avg:156.65ms
step:1058/1480 train_time:164186ms step_avg:156.67ms
step:1059/1480 train_time:164362ms step_avg:156.68ms
step:1060/1480 train_time:164530ms step_avg:156.69ms
step:1061/1480 train_time:164693ms step_avg:156.70ms
step:1062/1480 train_time:164861ms step_avg:156.71ms
step:1063/1480 train_time:165026ms step_avg:156.72ms
step:1064/1480 train_time:165189ms step_avg:156.73ms
step:1065/1480 train_time:165356ms step_avg:156.74ms
step:1066/1480 train_time:165523ms step_avg:156.75ms
step:1067/1480 train_time:165691ms step_avg:156.76ms
step:1068/1480 train_time:165857ms step_avg:156.76ms
step:1069/1480 train_time:166027ms step_avg:156.78ms
step:1070/1480 train_time:166192ms step_avg:156.78ms
step:1071/1480 train_time:166366ms step_avg:156.80ms
step:1072/1480 train_time:166532ms step_avg:156.81ms
step:1073/1480 train_time:166695ms step_avg:156.82ms
step:1074/1480 train_time:166862ms step_avg:156.82ms
step:1075/1480 train_time:167032ms step_avg:156.84ms
step:1076/1480 train_time:167199ms step_avg:156.85ms
step:1077/1480 train_time:167367ms step_avg:156.86ms
step:1078/1480 train_time:167542ms step_avg:156.87ms
step:1079/1480 train_time:167715ms step_avg:156.89ms
step:1080/1480 train_time:167886ms step_avg:156.90ms
step:1081/1480 train_time:168051ms step_avg:156.91ms
step:1082/1480 train_time:168216ms step_avg:156.92ms
step:1083/1480 train_time:168384ms step_avg:156.93ms
step:1084/1480 train_time:168550ms step_avg:156.94ms
step:1085/1480 train_time:168717ms step_avg:156.95ms
step:1086/1480 train_time:168886ms step_avg:156.96ms
step:1087/1480 train_time:169053ms step_avg:156.97ms
step:1088/1480 train_time:169223ms step_avg:156.98ms
step:1089/1480 train_time:169395ms step_avg:156.99ms
step:1090/1480 train_time:169566ms step_avg:157.01ms
step:1091/1480 train_time:169733ms step_avg:157.01ms
step:1092/1480 train_time:169900ms step_avg:157.02ms
step:1093/1480 train_time:170068ms step_avg:157.03ms
step:1094/1480 train_time:170233ms step_avg:157.04ms
step:1095/1480 train_time:170398ms step_avg:157.05ms
step:1096/1480 train_time:170567ms step_avg:157.06ms
step:1097/1480 train_time:170736ms step_avg:157.07ms
step:1098/1480 train_time:170907ms step_avg:157.08ms
step:1099/1480 train_time:171079ms step_avg:157.10ms
step:1100/1480 train_time:171252ms step_avg:157.11ms
step:1101/1480 train_time:171423ms step_avg:157.12ms
step:1102/1480 train_time:171594ms step_avg:157.14ms
step:1103/1480 train_time:171770ms step_avg:157.15ms
step:1104/1480 train_time:171939ms step_avg:157.16ms
step:1105/1480 train_time:172107ms step_avg:157.17ms
step:1106/1480 train_time:172273ms step_avg:157.18ms
step:1107/1480 train_time:172443ms step_avg:157.20ms
step:1108/1480 train_time:172609ms step_avg:157.20ms
step:1109/1480 train_time:172774ms step_avg:157.21ms
step:1110/1480 train_time:172942ms step_avg:157.22ms
step:1111/1480 train_time:173108ms step_avg:157.23ms
step:1112/1480 train_time:173277ms step_avg:157.24ms
step:1113/1480 train_time:173460ms step_avg:157.26ms
step:1114/1480 train_time:173631ms step_avg:157.27ms
step:1115/1480 train_time:173806ms step_avg:157.29ms
step:1116/1480 train_time:173972ms step_avg:157.30ms
step:1117/1480 train_time:174146ms step_avg:157.31ms
step:1118/1480 train_time:174319ms step_avg:157.33ms
step:1119/1480 train_time:174484ms step_avg:157.33ms
step:1120/1480 train_time:174653ms step_avg:157.35ms
step:1121/1480 train_time:174824ms step_avg:157.36ms
step:1122/1480 train_time:174989ms step_avg:157.36ms
step:1123/1480 train_time:175157ms step_avg:157.37ms
step:1124/1480 train_time:175324ms step_avg:157.38ms
step:1125/1480 train_time:175491ms step_avg:157.39ms
step:1125/1480 val_loss:3.3848 train_time:175560ms step_avg:157.45ms
step:1126/1480 train_time:175662ms step_avg:157.40ms
step:1127/1480 train_time:175832ms step_avg:157.41ms
step:1128/1480 train_time:176001ms step_avg:157.43ms
step:1129/1480 train_time:176177ms step_avg:157.44ms
step:1130/1480 train_time:176345ms step_avg:157.45ms
step:1131/1480 train_time:176523ms step_avg:157.47ms
step:1132/1480 train_time:176689ms step_avg:157.48ms
step:1133/1480 train_time:176860ms step_avg:157.49ms
step:1134/1480 train_time:177031ms step_avg:157.50ms
step:1135/1480 train_time:177199ms step_avg:157.51ms
step:1136/1480 train_time:177371ms step_avg:157.52ms
step:1137/1480 train_time:177539ms step_avg:157.53ms
step:1138/1480 train_time:177712ms step_avg:157.55ms
step:1139/1480 train_time:177880ms step_avg:157.55ms
step:1140/1480 train_time:178048ms step_avg:157.56ms
step:1141/1480 train_time:178220ms step_avg:157.58ms
step:1142/1480 train_time:178390ms step_avg:157.59ms
step:1143/1480 train_time:178560ms step_avg:157.60ms
step:1144/1480 train_time:178729ms step_avg:157.61ms
step:1145/1480 train_time:178895ms step_avg:157.62ms
step:1146/1480 train_time:179066ms step_avg:157.63ms
step:1147/1480 train_time:179233ms step_avg:157.64ms
step:1148/1480 train_time:179401ms step_avg:157.65ms
step:1149/1480 train_time:179573ms step_avg:157.66ms
step:1150/1480 train_time:179740ms step_avg:157.67ms
step:1151/1480 train_time:179913ms step_avg:157.68ms
step:1152/1480 train_time:180085ms step_avg:157.69ms
step:1153/1480 train_time:180259ms step_avg:157.71ms
step:1154/1480 train_time:180426ms step_avg:157.71ms
step:1155/1480 train_time:180598ms step_avg:157.73ms
step:1156/1480 train_time:180776ms step_avg:157.75ms
step:1157/1480 train_time:180947ms step_avg:157.76ms
step:1158/1480 train_time:181113ms step_avg:157.76ms
step:1159/1480 train_time:181282ms step_avg:157.77ms
step:1160/1480 train_time:181449ms step_avg:157.78ms
step:1161/1480 train_time:181619ms step_avg:157.79ms
step:1162/1480 train_time:181791ms step_avg:157.80ms
step:1163/1480 train_time:181959ms step_avg:157.81ms
step:1164/1480 train_time:182129ms step_avg:157.82ms
step:1165/1480 train_time:182295ms step_avg:157.83ms
step:1166/1480 train_time:182465ms step_avg:157.84ms
step:1167/1480 train_time:182633ms step_avg:157.85ms
step:1168/1480 train_time:182800ms step_avg:157.86ms
step:1169/1480 train_time:182972ms step_avg:157.87ms
step:1170/1480 train_time:183139ms step_avg:157.88ms
step:1171/1480 train_time:183307ms step_avg:157.89ms
step:1172/1480 train_time:183475ms step_avg:157.90ms
step:1173/1480 train_time:183647ms step_avg:157.91ms
step:1174/1480 train_time:183830ms step_avg:157.93ms
step:1175/1480 train_time:184000ms step_avg:157.94ms
step:1176/1480 train_time:184173ms step_avg:157.95ms
step:1177/1480 train_time:184351ms step_avg:157.97ms
step:1178/1480 train_time:184519ms step_avg:157.98ms
step:1179/1480 train_time:184684ms step_avg:157.98ms
step:1180/1480 train_time:184865ms step_avg:158.00ms
step:1181/1480 train_time:185035ms step_avg:158.01ms
step:1182/1480 train_time:185203ms step_avg:158.02ms
step:1183/1480 train_time:185374ms step_avg:158.03ms
step:1184/1480 train_time:185542ms step_avg:158.04ms
step:1185/1480 train_time:185714ms step_avg:158.05ms
step:1186/1480 train_time:185883ms step_avg:158.06ms
step:1187/1480 train_time:186066ms step_avg:158.09ms
step:1188/1480 train_time:186232ms step_avg:158.09ms
step:1189/1480 train_time:186403ms step_avg:158.10ms
step:1190/1480 train_time:186572ms step_avg:158.11ms
step:1191/1480 train_time:186743ms step_avg:158.12ms
step:1192/1480 train_time:186910ms step_avg:158.13ms
step:1193/1480 train_time:187076ms step_avg:158.14ms
step:1194/1480 train_time:187246ms step_avg:158.15ms
step:1195/1480 train_time:187420ms step_avg:158.16ms
step:1196/1480 train_time:187603ms step_avg:158.18ms
step:1197/1480 train_time:187774ms step_avg:158.19ms
step:1198/1480 train_time:187956ms step_avg:158.21ms
step:1199/1480 train_time:188127ms step_avg:158.22ms
step:1200/1480 train_time:188295ms step_avg:158.23ms
step:1201/1480 train_time:188463ms step_avg:158.24ms
step:1202/1480 train_time:188643ms step_avg:158.26ms
step:1203/1480 train_time:188817ms step_avg:158.27ms
step:1204/1480 train_time:188993ms step_avg:158.29ms
step:1205/1480 train_time:189160ms step_avg:158.29ms
step:1206/1480 train_time:189328ms step_avg:158.30ms
step:1207/1480 train_time:189496ms step_avg:158.31ms
step:1208/1480 train_time:189663ms step_avg:158.32ms
step:1209/1480 train_time:189836ms step_avg:158.33ms
step:1210/1480 train_time:190012ms step_avg:158.34ms
step:1211/1480 train_time:190186ms step_avg:158.36ms
step:1212/1480 train_time:190359ms step_avg:158.37ms
step:1213/1480 train_time:190531ms step_avg:158.38ms
step:1214/1480 train_time:190708ms step_avg:158.40ms
step:1215/1480 train_time:190881ms step_avg:158.41ms
step:1216/1480 train_time:191051ms step_avg:158.42ms
step:1217/1480 train_time:191225ms step_avg:158.43ms
step:1218/1480 train_time:191395ms step_avg:158.44ms
step:1219/1480 train_time:191573ms step_avg:158.46ms
step:1220/1480 train_time:191743ms step_avg:158.46ms
step:1221/1480 train_time:191913ms step_avg:158.47ms
step:1222/1480 train_time:192081ms step_avg:158.48ms
step:1223/1480 train_time:192251ms step_avg:158.49ms
step:1224/1480 train_time:192428ms step_avg:158.51ms
step:1225/1480 train_time:192599ms step_avg:158.52ms
step:1226/1480 train_time:192772ms step_avg:158.53ms
step:1227/1480 train_time:192945ms step_avg:158.54ms
step:1228/1480 train_time:193113ms step_avg:158.55ms
step:1229/1480 train_time:193286ms step_avg:158.56ms
step:1230/1480 train_time:193466ms step_avg:158.58ms
step:1231/1480 train_time:193641ms step_avg:158.59ms
step:1232/1480 train_time:193816ms step_avg:158.61ms
step:1233/1480 train_time:193987ms step_avg:158.62ms
step:1234/1480 train_time:194156ms step_avg:158.62ms
step:1235/1480 train_time:194331ms step_avg:158.64ms
step:1236/1480 train_time:194499ms step_avg:158.64ms
step:1237/1480 train_time:194670ms step_avg:158.65ms
step:1238/1480 train_time:194852ms step_avg:158.67ms
step:1239/1480 train_time:195022ms step_avg:158.68ms
step:1240/1480 train_time:195194ms step_avg:158.69ms
step:1241/1480 train_time:195369ms step_avg:158.71ms
step:1242/1480 train_time:195538ms step_avg:158.72ms
step:1243/1480 train_time:195713ms step_avg:158.73ms
step:1244/1480 train_time:195881ms step_avg:158.74ms
step:1245/1480 train_time:196051ms step_avg:158.75ms
step:1246/1480 train_time:196220ms step_avg:158.75ms
step:1247/1480 train_time:196390ms step_avg:158.76ms
step:1248/1480 train_time:196558ms step_avg:158.77ms
step:1249/1480 train_time:196727ms step_avg:158.78ms
step:1250/1480 train_time:196895ms step_avg:158.79ms
step:1250/1480 val_loss:3.3354 train_time:196968ms step_avg:158.85ms
step:1251/1480 train_time:197079ms step_avg:158.81ms
step:1252/1480 train_time:197249ms step_avg:158.82ms
step:1253/1480 train_time:197417ms step_avg:158.82ms
step:1254/1480 train_time:197587ms step_avg:158.83ms
step:1255/1480 train_time:197774ms step_avg:158.86ms
step:1256/1480 train_time:197948ms step_avg:158.87ms
step:1257/1480 train_time:198119ms step_avg:158.88ms
step:1258/1480 train_time:198294ms step_avg:158.89ms
step:1259/1480 train_time:198466ms step_avg:158.90ms
step:1260/1480 train_time:198634ms step_avg:158.91ms
step:1261/1480 train_time:198807ms step_avg:158.92ms
step:1262/1480 train_time:198982ms step_avg:158.93ms
step:1263/1480 train_time:199155ms step_avg:158.94ms
step:1264/1480 train_time:199322ms step_avg:158.95ms
step:1265/1480 train_time:199490ms step_avg:158.96ms
step:1266/1480 train_time:199661ms step_avg:158.97ms
step:1267/1480 train_time:199832ms step_avg:158.98ms
step:1268/1480 train_time:200002ms step_avg:158.98ms
step:1269/1480 train_time:200178ms step_avg:159.00ms
step:1270/1480 train_time:200348ms step_avg:159.01ms
step:1271/1480 train_time:200518ms step_avg:159.01ms
step:1272/1480 train_time:200684ms step_avg:159.02ms
step:1273/1480 train_time:200855ms step_avg:159.03ms
step:1274/1480 train_time:201027ms step_avg:159.04ms
step:1275/1480 train_time:201196ms step_avg:159.05ms
step:1276/1480 train_time:201361ms step_avg:159.05ms
step:1277/1480 train_time:201534ms step_avg:159.06ms
step:1278/1480 train_time:201700ms step_avg:159.07ms
step:1279/1480 train_time:201873ms step_avg:159.08ms
step:1280/1480 train_time:202052ms step_avg:159.10ms
step:1281/1480 train_time:202221ms step_avg:159.10ms
step:1282/1480 train_time:202387ms step_avg:159.11ms
step:1283/1480 train_time:202556ms step_avg:159.12ms
step:1284/1480 train_time:202725ms step_avg:159.12ms
step:1285/1480 train_time:202895ms step_avg:159.13ms
step:1286/1480 train_time:203064ms step_avg:159.14ms
step:1287/1480 train_time:203240ms step_avg:159.15ms
step:1288/1480 train_time:203414ms step_avg:159.17ms
step:1289/1480 train_time:203596ms step_avg:159.18ms
step:1290/1480 train_time:203775ms step_avg:159.20ms
step:1291/1480 train_time:203947ms step_avg:159.21ms
step:1292/1480 train_time:204121ms step_avg:159.22ms
step:1293/1480 train_time:204296ms step_avg:159.23ms
step:1294/1480 train_time:204466ms step_avg:159.24ms
step:1295/1480 train_time:204638ms step_avg:159.25ms
step:1296/1480 train_time:204813ms step_avg:159.26ms
step:1297/1480 train_time:204983ms step_avg:159.27ms
step:1298/1480 train_time:205155ms step_avg:159.28ms
step:1299/1480 train_time:205324ms step_avg:159.29ms
step:1300/1480 train_time:205492ms step_avg:159.30ms
step:1301/1480 train_time:205662ms step_avg:159.30ms
step:1302/1480 train_time:205837ms step_avg:159.32ms
step:1303/1480 train_time:206013ms step_avg:159.33ms
step:1304/1480 train_time:206186ms step_avg:159.34ms
step:1305/1480 train_time:206355ms step_avg:159.35ms
step:1306/1480 train_time:206528ms step_avg:159.36ms
step:1307/1480 train_time:206697ms step_avg:159.37ms
step:1308/1480 train_time:206865ms step_avg:159.37ms
step:1309/1480 train_time:207039ms step_avg:159.38ms
step:1310/1480 train_time:207209ms step_avg:159.39ms
step:1311/1480 train_time:207378ms step_avg:159.40ms
step:1312/1480 train_time:207553ms step_avg:159.41ms
step:1313/1480 train_time:207720ms step_avg:159.42ms
step:1314/1480 train_time:207895ms step_avg:159.43ms
step:1315/1480 train_time:208068ms step_avg:159.44ms
step:1316/1480 train_time:208236ms step_avg:159.45ms
step:1317/1480 train_time:208407ms step_avg:159.45ms
step:1318/1480 train_time:208588ms step_avg:159.47ms
step:1319/1480 train_time:208763ms step_avg:159.48ms
step:1320/1480 train_time:208940ms step_avg:159.50ms
step:1321/1480 train_time:209114ms step_avg:159.51ms
step:1322/1480 train_time:209294ms step_avg:159.52ms
step:1323/1480 train_time:209465ms step_avg:159.53ms
step:1324/1480 train_time:209641ms step_avg:159.54ms
step:1325/1480 train_time:209823ms step_avg:159.56ms
step:1326/1480 train_time:209999ms step_avg:159.57ms
step:1327/1480 train_time:210169ms step_avg:159.58ms
step:1328/1480 train_time:210339ms step_avg:159.59ms
step:1329/1480 train_time:210536ms step_avg:159.62ms
step:1330/1480 train_time:210718ms step_avg:159.63ms
step:1331/1480 train_time:210889ms step_avg:159.64ms
step:1332/1480 train_time:211062ms step_avg:159.65ms
step:1333/1480 train_time:211237ms step_avg:159.67ms
step:1334/1480 train_time:211410ms step_avg:159.68ms
step:1335/1480 train_time:211579ms step_avg:159.68ms
step:1336/1480 train_time:211765ms step_avg:159.70ms
step:1337/1480 train_time:211941ms step_avg:159.71ms
step:1338/1480 train_time:212115ms step_avg:159.73ms
step:1339/1480 train_time:212288ms step_avg:159.74ms
step:1340/1480 train_time:212460ms step_avg:159.74ms
step:1341/1480 train_time:212629ms step_avg:159.75ms
step:1342/1480 train_time:212802ms step_avg:159.76ms
step:1343/1480 train_time:212973ms step_avg:159.77ms
step:1344/1480 train_time:213144ms step_avg:159.78ms
step:1345/1480 train_time:213322ms step_avg:159.79ms
step:1346/1480 train_time:213492ms step_avg:159.80ms
step:1347/1480 train_time:213662ms step_avg:159.81ms
step:1348/1480 train_time:213832ms step_avg:159.81ms
step:1349/1480 train_time:214002ms step_avg:159.82ms
step:1350/1480 train_time:214176ms step_avg:159.83ms
step:1351/1480 train_time:214348ms step_avg:159.84ms
step:1352/1480 train_time:214519ms step_avg:159.85ms
step:1353/1480 train_time:214696ms step_avg:159.86ms
step:1354/1480 train_time:214867ms step_avg:159.87ms
step:1355/1480 train_time:215034ms step_avg:159.88ms
step:1356/1480 train_time:215207ms step_avg:159.89ms
step:1357/1480 train_time:215380ms step_avg:159.90ms
step:1358/1480 train_time:215552ms step_avg:159.91ms
step:1359/1480 train_time:215725ms step_avg:159.92ms
step:1360/1480 train_time:215901ms step_avg:159.93ms
step:1361/1480 train_time:216079ms step_avg:159.94ms
step:1362/1480 train_time:216254ms step_avg:159.95ms
step:1363/1480 train_time:216432ms step_avg:159.96ms
step:1364/1480 train_time:216600ms step_avg:159.97ms
step:1365/1480 train_time:216766ms step_avg:159.98ms
step:1366/1480 train_time:216938ms step_avg:159.98ms
step:1367/1480 train_time:217113ms step_avg:159.99ms
step:1368/1480 train_time:217285ms step_avg:160.00ms
step:1369/1480 train_time:217465ms step_avg:160.02ms
step:1370/1480 train_time:217643ms step_avg:160.03ms
step:1371/1480 train_time:217816ms step_avg:160.04ms
step:1372/1480 train_time:217993ms step_avg:160.05ms
step:1373/1480 train_time:218162ms step_avg:160.06ms
step:1374/1480 train_time:218339ms step_avg:160.07ms
step:1375/1480 train_time:218511ms step_avg:160.08ms
step:1375/1480 val_loss:3.2970 train_time:218579ms step_avg:160.13ms
step:1376/1480 train_time:218689ms step_avg:160.09ms
step:1377/1480 train_time:218861ms step_avg:160.10ms
step:1378/1480 train_time:219030ms step_avg:160.11ms
step:1379/1480 train_time:219203ms step_avg:160.12ms
step:1380/1480 train_time:219377ms step_avg:160.13ms
step:1381/1480 train_time:219556ms step_avg:160.14ms
step:1382/1480 train_time:219727ms step_avg:160.15ms
step:1383/1480 train_time:219898ms step_avg:160.16ms
step:1384/1480 train_time:220076ms step_avg:160.17ms
step:1385/1480 train_time:220241ms step_avg:160.18ms
step:1386/1480 train_time:220413ms step_avg:160.18ms
step:1387/1480 train_time:220587ms step_avg:160.19ms
step:1388/1480 train_time:220756ms step_avg:160.20ms
step:1389/1480 train_time:220930ms step_avg:160.21ms
step:1390/1480 train_time:221097ms step_avg:160.22ms
step:1391/1480 train_time:221266ms step_avg:160.22ms
step:1392/1480 train_time:221438ms step_avg:160.23ms
step:1393/1480 train_time:221610ms step_avg:160.24ms
step:1394/1480 train_time:221779ms step_avg:160.25ms
step:1395/1480 train_time:221949ms step_avg:160.25ms
step:1396/1480 train_time:222117ms step_avg:160.26ms
step:1397/1480 train_time:222284ms step_avg:160.26ms
step:1398/1480 train_time:222451ms step_avg:160.27ms
step:1399/1480 train_time:222618ms step_avg:160.27ms
step:1400/1480 train_time:222795ms step_avg:160.28ms
step:1401/1480 train_time:222961ms step_avg:160.29ms
step:1402/1480 train_time:223132ms step_avg:160.30ms
step:1403/1480 train_time:223310ms step_avg:160.31ms
step:1404/1480 train_time:223483ms step_avg:160.32ms
step:1405/1480 train_time:223657ms step_avg:160.33ms
step:1406/1480 train_time:223831ms step_avg:160.34ms
step:1407/1480 train_time:223998ms step_avg:160.34ms
step:1408/1480 train_time:224167ms step_avg:160.35ms
step:1409/1480 train_time:224349ms step_avg:160.36ms
step:1410/1480 train_time:224518ms step_avg:160.37ms
step:1411/1480 train_time:224688ms step_avg:160.38ms
step:1412/1480 train_time:224858ms step_avg:160.38ms
step:1413/1480 train_time:225030ms step_avg:160.39ms
step:1414/1480 train_time:225201ms step_avg:160.40ms
step:1415/1480 train_time:225375ms step_avg:160.41ms
step:1416/1480 train_time:225562ms step_avg:160.43ms
step:1417/1480 train_time:225735ms step_avg:160.44ms
step:1418/1480 train_time:225907ms step_avg:160.45ms
step:1419/1480 train_time:226080ms step_avg:160.45ms
step:1420/1480 train_time:226256ms step_avg:160.47ms
step:1421/1480 train_time:226431ms step_avg:160.48ms
step:1422/1480 train_time:226602ms step_avg:160.48ms
step:1423/1480 train_time:226772ms step_avg:160.49ms
step:1424/1480 train_time:226950ms step_avg:160.50ms
step:1425/1480 train_time:227132ms step_avg:160.52ms
step:1426/1480 train_time:227304ms step_avg:160.53ms
step:1427/1480 train_time:227478ms step_avg:160.54ms
step:1428/1480 train_time:227650ms step_avg:160.54ms
step:1429/1480 train_time:227819ms step_avg:160.55ms
step:1430/1480 train_time:227993ms step_avg:160.56ms
step:1431/1480 train_time:228170ms step_avg:160.57ms
step:1432/1480 train_time:228346ms step_avg:160.58ms
step:1433/1480 train_time:228525ms step_avg:160.59ms
step:1434/1480 train_time:228706ms step_avg:160.61ms
step:1435/1480 train_time:228881ms step_avg:160.62ms
step:1436/1480 train_time:229054ms step_avg:160.63ms
step:1437/1480 train_time:229223ms step_avg:160.63ms
step:1438/1480 train_time:229392ms step_avg:160.64ms
step:1439/1480 train_time:229568ms step_avg:160.65ms
step:1440/1480 train_time:229737ms step_avg:160.66ms
step:1441/1480 train_time:229909ms step_avg:160.66ms
step:1442/1480 train_time:230086ms step_avg:160.67ms
step:1443/1480 train_time:230275ms step_avg:160.69ms
step:1444/1480 train_time:230446ms step_avg:160.70ms
step:1445/1480 train_time:230618ms step_avg:160.71ms
step:1446/1480 train_time:230793ms step_avg:160.72ms
step:1447/1480 train_time:230971ms step_avg:160.73ms
step:1448/1480 train_time:231142ms step_avg:160.74ms
step:1449/1480 train_time:231313ms step_avg:160.75ms
step:1450/1480 train_time:231487ms step_avg:160.75ms
step:1451/1480 train_time:231658ms step_avg:160.76ms
step:1452/1480 train_time:231833ms step_avg:160.77ms
step:1453/1480 train_time:232002ms step_avg:160.78ms
step:1454/1480 train_time:232173ms step_avg:160.78ms
step:1455/1480 train_time:232353ms step_avg:160.80ms
step:1456/1480 train_time:232527ms step_avg:160.81ms
step:1457/1480 train_time:232697ms step_avg:160.81ms
step:1458/1480 train_time:232869ms step_avg:160.82ms
step:1459/1480 train_time:233044ms step_avg:160.83ms
step:1460/1480 train_time:233215ms step_avg:160.84ms
step:1461/1480 train_time:233391ms step_avg:160.85ms
step:1462/1480 train_time:233561ms step_avg:160.85ms
step:1463/1480 train_time:233738ms step_avg:160.87ms
step:1464/1480 train_time:233913ms step_avg:160.88ms
step:1465/1480 train_time:234085ms step_avg:160.88ms
step:1466/1480 train_time:234255ms step_avg:160.89ms
step:1467/1480 train_time:234430ms step_avg:160.90ms
step:1468/1480 train_time:234598ms step_avg:160.90ms
step:1469/1480 train_time:234771ms step_avg:160.91ms
step:1470/1480 train_time:234953ms step_avg:160.93ms
step:1471/1480 train_time:235138ms step_avg:160.94ms
step:1472/1480 train_time:235318ms step_avg:160.96ms
step:1473/1480 train_time:235489ms step_avg:160.96ms
step:1474/1480 train_time:235667ms step_avg:160.97ms
step:1475/1480 train_time:235847ms step_avg:160.99ms
step:1476/1480 train_time:236019ms step_avg:161.00ms
step:1477/1480 train_time:236203ms step_avg:161.01ms
step:1478/1480 train_time:236387ms step_avg:161.03ms
step:1479/1480 train_time:236559ms step_avg:161.03ms
step:1480/1480 train_time:236731ms step_avg:161.04ms
step:1480/1480 val_loss:3.2783 train_time:236802ms step_avg:161.09ms