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 11:19:55 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   44C    P0              78W / 700W |      7MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   2  NVIDIA H100 80GB HBM3          On  | 00000000:69:02.0 Off |                    0 |
| N/A   45C    P0              75W / 700W |      7MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   3  NVIDIA H100 80GB HBM3          On  | 00000000:6B:02.0 Off |                    0 |
| N/A   39C    P0              77W / 700W |     17MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   4  NVIDIA H100 80GB HBM3          On  | 00000000:6F:02.0 Off |                    0 |
| N/A   39C    P0              86W / 700W |     26MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   5  NVIDIA H100 80GB HBM3          On  | 00000000:71:02.0 Off |                    0 |
| N/A   44C    P0              74W / 700W |      7MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   6  NVIDIA H100 80GB HBM3          On  | 00000000:73:02.0 Off |                    0 |
| N/A   45C    P0              77W / 700W |      7MiB / 81559MiB |      0%      Default |
|                                         |                      |             Disabled |
+-----------------------------------------+----------------------+----------------------+
|   7  NVIDIA H100 80GB HBM3          On  | 00000000:75:02.0 Off |                    0 |
| N/A   37C    P0              79W / 700W |      7MiB / 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:22965ms step_avg:nanms
step:2/1480 train_time:23052ms step_avg:nanms
step:3/1480 train_time:23191ms step_avg:nanms
step:4/1480 train_time:23331ms step_avg:nanms
step:5/1480 train_time:23472ms step_avg:nanms
step:6/1480 train_time:23612ms step_avg:nanms
step:7/1480 train_time:23754ms step_avg:nanms
step:8/1480 train_time:23895ms step_avg:nanms
step:9/1480 train_time:24040ms step_avg:nanms
step:10/1480 train_time:24184ms step_avg:nanms
step:11/1480 train_time:142ms step_avg:nanms
step:12/1480 train_time:285ms step_avg:nanms
step:13/1480 train_time:427ms step_avg:142.17ms
step:14/1480 train_time:569ms step_avg:142.13ms
step:15/1480 train_time:710ms step_avg:141.96ms
step:16/1480 train_time:852ms step_avg:142.06ms
step:17/1480 train_time:997ms step_avg:142.38ms
step:18/1480 train_time:1141ms step_avg:142.59ms
step:19/1480 train_time:1285ms step_avg:142.74ms
step:20/1480 train_time:1428ms step_avg:142.78ms
step:21/1480 train_time:1570ms step_avg:142.72ms
step:22/1480 train_time:1711ms step_avg:142.60ms
step:23/1480 train_time:1855ms step_avg:142.66ms
step:24/1480 train_time:1999ms step_avg:142.79ms
step:25/1480 train_time:2144ms step_avg:142.91ms
step:26/1480 train_time:2287ms step_avg:142.96ms
step:27/1480 train_time:2429ms step_avg:142.88ms
step:28/1480 train_time:2571ms step_avg:142.83ms
step:29/1480 train_time:2714ms step_avg:142.82ms
step:30/1480 train_time:2855ms step_avg:142.75ms
step:31/1480 train_time:2998ms step_avg:142.77ms
step:32/1480 train_time:3143ms step_avg:142.84ms
step:33/1480 train_time:3287ms step_avg:142.90ms
step:34/1480 train_time:3429ms step_avg:142.89ms
step:35/1480 train_time:3570ms step_avg:142.82ms
step:36/1480 train_time:3711ms step_avg:142.75ms
step:37/1480 train_time:3854ms step_avg:142.74ms
step:38/1480 train_time:3999ms step_avg:142.83ms
step:39/1480 train_time:4142ms step_avg:142.81ms
step:40/1480 train_time:4286ms step_avg:142.86ms
step:41/1480 train_time:4429ms step_avg:142.86ms
step:42/1480 train_time:4570ms step_avg:142.81ms
step:43/1480 train_time:4711ms step_avg:142.76ms
step:44/1480 train_time:4852ms step_avg:142.70ms
step:45/1480 train_time:4996ms step_avg:142.73ms
step:46/1480 train_time:5141ms step_avg:142.81ms
step:47/1480 train_time:5285ms step_avg:142.85ms
step:48/1480 train_time:5428ms step_avg:142.84ms
step:49/1480 train_time:5569ms step_avg:142.80ms
step:50/1480 train_time:5711ms step_avg:142.77ms
step:51/1480 train_time:5852ms step_avg:142.74ms
step:52/1480 train_time:5995ms step_avg:142.75ms
step:53/1480 train_time:6140ms step_avg:142.79ms
step:54/1480 train_time:6285ms step_avg:142.83ms
step:55/1480 train_time:6427ms step_avg:142.82ms
step:56/1480 train_time:6569ms step_avg:142.80ms
step:57/1480 train_time:6711ms step_avg:142.79ms
step:58/1480 train_time:6852ms step_avg:142.75ms
step:59/1480 train_time:6995ms step_avg:142.75ms
step:60/1480 train_time:7138ms step_avg:142.77ms
step:61/1480 train_time:7283ms step_avg:142.80ms
step:62/1480 train_time:7427ms step_avg:142.82ms
step:63/1480 train_time:7568ms step_avg:142.80ms
step:64/1480 train_time:7710ms step_avg:142.77ms
step:65/1480 train_time:7852ms step_avg:142.76ms
step:66/1480 train_time:7995ms step_avg:142.77ms
step:67/1480 train_time:8138ms step_avg:142.77ms
step:68/1480 train_time:8281ms step_avg:142.78ms
step:69/1480 train_time:8425ms step_avg:142.79ms
step:70/1480 train_time:8568ms step_avg:142.81ms
step:71/1480 train_time:8711ms step_avg:142.80ms
step:72/1480 train_time:8852ms step_avg:142.77ms
step:73/1480 train_time:8998ms step_avg:142.82ms
step:74/1480 train_time:9138ms step_avg:142.79ms
step:75/1480 train_time:9283ms step_avg:142.81ms
step:76/1480 train_time:9427ms step_avg:142.83ms
step:77/1480 train_time:9568ms step_avg:142.81ms
step:78/1480 train_time:9711ms step_avg:142.81ms
step:79/1480 train_time:9852ms step_avg:142.78ms
step:80/1480 train_time:9993ms step_avg:142.76ms
step:81/1480 train_time:10135ms step_avg:142.74ms
step:82/1480 train_time:10277ms step_avg:142.74ms
step:83/1480 train_time:10422ms step_avg:142.77ms
step:84/1480 train_time:10566ms step_avg:142.79ms
step:85/1480 train_time:10708ms step_avg:142.77ms
step:86/1480 train_time:10850ms step_avg:142.76ms
step:87/1480 train_time:10992ms step_avg:142.75ms
step:88/1480 train_time:11133ms step_avg:142.73ms
step:89/1480 train_time:11275ms step_avg:142.72ms
step:90/1480 train_time:11417ms step_avg:142.72ms
step:91/1480 train_time:11562ms step_avg:142.74ms
step:92/1480 train_time:11707ms step_avg:142.76ms
step:93/1480 train_time:11850ms step_avg:142.77ms
step:94/1480 train_time:11992ms step_avg:142.76ms
step:95/1480 train_time:12133ms step_avg:142.75ms
step:96/1480 train_time:12274ms step_avg:142.73ms
step:97/1480 train_time:12416ms step_avg:142.72ms
step:98/1480 train_time:12560ms step_avg:142.73ms
step:99/1480 train_time:12704ms step_avg:142.74ms
step:100/1480 train_time:12848ms step_avg:142.75ms
step:101/1480 train_time:12989ms step_avg:142.74ms
step:102/1480 train_time:13131ms step_avg:142.73ms
step:103/1480 train_time:13272ms step_avg:142.71ms
step:104/1480 train_time:13413ms step_avg:142.69ms
step:105/1480 train_time:13556ms step_avg:142.69ms
step:106/1480 train_time:13700ms step_avg:142.71ms
step:107/1480 train_time:13843ms step_avg:142.71ms
step:108/1480 train_time:13985ms step_avg:142.70ms
step:109/1480 train_time:14127ms step_avg:142.70ms
step:110/1480 train_time:14270ms step_avg:142.70ms
step:111/1480 train_time:14412ms step_avg:142.70ms
step:112/1480 train_time:14560ms step_avg:142.75ms
step:113/1480 train_time:14708ms step_avg:142.80ms
step:114/1480 train_time:14855ms step_avg:142.84ms
step:115/1480 train_time:15004ms step_avg:142.90ms
step:116/1480 train_time:15151ms step_avg:142.94ms
step:117/1480 train_time:15298ms step_avg:142.97ms
step:118/1480 train_time:15446ms step_avg:143.01ms
step:119/1480 train_time:15592ms step_avg:143.04ms
step:120/1480 train_time:15737ms step_avg:143.06ms
step:121/1480 train_time:15884ms step_avg:143.10ms
step:122/1480 train_time:16032ms step_avg:143.14ms
step:123/1480 train_time:16179ms step_avg:143.18ms
step:124/1480 train_time:16326ms step_avg:143.21ms
step:125/1480 train_time:16472ms step_avg:143.24ms
step:125/1480 val_loss:4.4114 train_time:16529ms step_avg:143.73ms
step:126/1480 train_time:16626ms step_avg:143.33ms
step:127/1480 train_time:16774ms step_avg:143.37ms
step:128/1480 train_time:16921ms step_avg:143.39ms
step:129/1480 train_time:17067ms step_avg:143.42ms
step:130/1480 train_time:17213ms step_avg:143.44ms
step:131/1480 train_time:17359ms step_avg:143.46ms
step:132/1480 train_time:17505ms step_avg:143.48ms
step:133/1480 train_time:17653ms step_avg:143.52ms
step:134/1480 train_time:17801ms step_avg:143.56ms
step:135/1480 train_time:17946ms step_avg:143.57ms
step:136/1480 train_time:18093ms step_avg:143.60ms
step:137/1480 train_time:18240ms step_avg:143.62ms
step:138/1480 train_time:18386ms step_avg:143.64ms
step:139/1480 train_time:18534ms step_avg:143.67ms
step:140/1480 train_time:18681ms step_avg:143.70ms
step:141/1480 train_time:18828ms step_avg:143.73ms
step:142/1480 train_time:18976ms step_avg:143.75ms
step:143/1480 train_time:19121ms step_avg:143.77ms
step:144/1480 train_time:19268ms step_avg:143.79ms
step:145/1480 train_time:19415ms step_avg:143.81ms
step:146/1480 train_time:19560ms step_avg:143.82ms
step:147/1480 train_time:19707ms step_avg:143.85ms
step:148/1480 train_time:19854ms step_avg:143.87ms
step:149/1480 train_time:20001ms step_avg:143.89ms
step:150/1480 train_time:20148ms step_avg:143.91ms
step:151/1480 train_time:20296ms step_avg:143.95ms
step:152/1480 train_time:20442ms step_avg:143.96ms
step:153/1480 train_time:20591ms step_avg:143.99ms
step:154/1480 train_time:20737ms step_avg:144.01ms
step:155/1480 train_time:20884ms step_avg:144.03ms
step:156/1480 train_time:21032ms step_avg:144.06ms
step:157/1480 train_time:21179ms step_avg:144.08ms
step:158/1480 train_time:21327ms step_avg:144.10ms
step:159/1480 train_time:21475ms step_avg:144.13ms
step:160/1480 train_time:21621ms step_avg:144.14ms
step:161/1480 train_time:21769ms step_avg:144.16ms
step:162/1480 train_time:21916ms step_avg:144.19ms
step:163/1480 train_time:22061ms step_avg:144.19ms
step:164/1480 train_time:22207ms step_avg:144.20ms
step:165/1480 train_time:22354ms step_avg:144.22ms
step:166/1480 train_time:22501ms step_avg:144.24ms
step:167/1480 train_time:22647ms step_avg:144.25ms
step:168/1480 train_time:22795ms step_avg:144.27ms
step:169/1480 train_time:22941ms step_avg:144.29ms
step:170/1480 train_time:23088ms step_avg:144.30ms
step:171/1480 train_time:23236ms step_avg:144.32ms
step:172/1480 train_time:23381ms step_avg:144.33ms
step:173/1480 train_time:23529ms step_avg:144.35ms
step:174/1480 train_time:23676ms step_avg:144.37ms
step:175/1480 train_time:23822ms step_avg:144.38ms
step:176/1480 train_time:23969ms step_avg:144.39ms
step:177/1480 train_time:24118ms step_avg:144.42ms
step:178/1480 train_time:24263ms step_avg:144.42ms
step:179/1480 train_time:24410ms step_avg:144.44ms
step:180/1480 train_time:24557ms step_avg:144.45ms
step:181/1480 train_time:24703ms step_avg:144.46ms
step:182/1480 train_time:24849ms step_avg:144.47ms
step:183/1480 train_time:24996ms step_avg:144.48ms
step:184/1480 train_time:25141ms step_avg:144.49ms
step:185/1480 train_time:25288ms step_avg:144.50ms
step:186/1480 train_time:25438ms step_avg:144.54ms
step:187/1480 train_time:25580ms step_avg:144.52ms
step:188/1480 train_time:25728ms step_avg:144.54ms
step:189/1480 train_time:25876ms step_avg:144.56ms
step:190/1480 train_time:26023ms step_avg:144.57ms
step:191/1480 train_time:26170ms step_avg:144.59ms
step:192/1480 train_time:26317ms step_avg:144.60ms
step:193/1480 train_time:26462ms step_avg:144.60ms
step:194/1480 train_time:26610ms step_avg:144.62ms
step:195/1480 train_time:26757ms step_avg:144.63ms
step:196/1480 train_time:26903ms step_avg:144.64ms
step:197/1480 train_time:27050ms step_avg:144.65ms
step:198/1480 train_time:27197ms step_avg:144.66ms
step:199/1480 train_time:27343ms step_avg:144.67ms
step:200/1480 train_time:27490ms step_avg:144.68ms
step:201/1480 train_time:27637ms step_avg:144.70ms
step:202/1480 train_time:27783ms step_avg:144.70ms
step:203/1480 train_time:27931ms step_avg:144.72ms
step:204/1480 train_time:28078ms step_avg:144.73ms
step:205/1480 train_time:28223ms step_avg:144.73ms
step:206/1480 train_time:28368ms step_avg:144.74ms
step:207/1480 train_time:28516ms step_avg:144.75ms
step:208/1480 train_time:28661ms step_avg:144.75ms
step:209/1480 train_time:28809ms step_avg:144.77ms
step:210/1480 train_time:28956ms step_avg:144.78ms
step:211/1480 train_time:29102ms step_avg:144.79ms
step:212/1480 train_time:29249ms step_avg:144.80ms
step:213/1480 train_time:29396ms step_avg:144.81ms
step:214/1480 train_time:29542ms step_avg:144.81ms
step:215/1480 train_time:29689ms step_avg:144.82ms
step:216/1480 train_time:29836ms step_avg:144.83ms
step:217/1480 train_time:29981ms step_avg:144.84ms
step:218/1480 train_time:30129ms step_avg:144.85ms
step:219/1480 train_time:30276ms step_avg:144.86ms
step:220/1480 train_time:30422ms step_avg:144.87ms
step:221/1480 train_time:30571ms step_avg:144.88ms
step:222/1480 train_time:30720ms step_avg:144.91ms
step:223/1480 train_time:30872ms step_avg:144.94ms
step:224/1480 train_time:31021ms step_avg:144.96ms
step:225/1480 train_time:31173ms step_avg:144.99ms
step:226/1480 train_time:31322ms step_avg:145.01ms
step:227/1480 train_time:31473ms step_avg:145.04ms
step:228/1480 train_time:31622ms step_avg:145.06ms
step:229/1480 train_time:31773ms step_avg:145.08ms
step:230/1480 train_time:31923ms step_avg:145.10ms
step:231/1480 train_time:32073ms step_avg:145.13ms
step:232/1480 train_time:32224ms step_avg:145.15ms
step:233/1480 train_time:32375ms step_avg:145.18ms
step:234/1480 train_time:32525ms step_avg:145.20ms
step:235/1480 train_time:32676ms step_avg:145.23ms
step:236/1480 train_time:32825ms step_avg:145.24ms
step:237/1480 train_time:32976ms step_avg:145.27ms
step:238/1480 train_time:33126ms step_avg:145.29ms
step:239/1480 train_time:33276ms step_avg:145.31ms
step:240/1480 train_time:33426ms step_avg:145.33ms
step:241/1480 train_time:33577ms step_avg:145.35ms
step:242/1480 train_time:33728ms step_avg:145.38ms
step:243/1480 train_time:33878ms step_avg:145.40ms
step:244/1480 train_time:34028ms step_avg:145.42ms
step:245/1480 train_time:34179ms step_avg:145.44ms
step:246/1480 train_time:34330ms step_avg:145.47ms
step:247/1480 train_time:34480ms step_avg:145.48ms
step:248/1480 train_time:34630ms step_avg:145.51ms
step:249/1480 train_time:34780ms step_avg:145.52ms
step:250/1480 train_time:34930ms step_avg:145.54ms
step:250/1480 val_loss:3.9872 train_time:34989ms step_avg:145.79ms
step:251/1480 train_time:35087ms step_avg:145.59ms
step:252/1480 train_time:35238ms step_avg:145.61ms
step:253/1480 train_time:35387ms step_avg:145.63ms
step:254/1480 train_time:35537ms step_avg:145.64ms
step:255/1480 train_time:35687ms step_avg:145.66ms
step:256/1480 train_time:35837ms step_avg:145.68ms
step:257/1480 train_time:35988ms step_avg:145.70ms
step:258/1480 train_time:36140ms step_avg:145.73ms
step:259/1480 train_time:36292ms step_avg:145.75ms
step:260/1480 train_time:36441ms step_avg:145.77ms
step:261/1480 train_time:36591ms step_avg:145.78ms
step:262/1480 train_time:36740ms step_avg:145.79ms
step:263/1480 train_time:36891ms step_avg:145.81ms
step:264/1480 train_time:37045ms step_avg:145.85ms
step:265/1480 train_time:37194ms step_avg:145.86ms
step:266/1480 train_time:37345ms step_avg:145.88ms
step:267/1480 train_time:37496ms step_avg:145.90ms
step:268/1480 train_time:37647ms step_avg:145.92ms
step:269/1480 train_time:37796ms step_avg:145.93ms
step:270/1480 train_time:37950ms step_avg:145.96ms
step:271/1480 train_time:38098ms step_avg:145.97ms
step:272/1480 train_time:38249ms step_avg:145.99ms
step:273/1480 train_time:38399ms step_avg:146.00ms
step:274/1480 train_time:38551ms step_avg:146.03ms
step:275/1480 train_time:38702ms step_avg:146.04ms
step:276/1480 train_time:38852ms step_avg:146.06ms
step:277/1480 train_time:39001ms step_avg:146.07ms
step:278/1480 train_time:39152ms step_avg:146.09ms
step:279/1480 train_time:39303ms step_avg:146.11ms
step:280/1480 train_time:39454ms step_avg:146.13ms
step:281/1480 train_time:39605ms step_avg:146.14ms
step:282/1480 train_time:39756ms step_avg:146.16ms
step:283/1480 train_time:39906ms step_avg:146.18ms
step:284/1480 train_time:40056ms step_avg:146.19ms
step:285/1480 train_time:40205ms step_avg:146.20ms
step:286/1480 train_time:40356ms step_avg:146.22ms
step:287/1480 train_time:40506ms step_avg:146.23ms
step:288/1480 train_time:40656ms step_avg:146.25ms
step:289/1480 train_time:40807ms step_avg:146.26ms
step:290/1480 train_time:40957ms step_avg:146.28ms
step:291/1480 train_time:41109ms step_avg:146.30ms
step:292/1480 train_time:41259ms step_avg:146.31ms
step:293/1480 train_time:41410ms step_avg:146.32ms
step:294/1480 train_time:41559ms step_avg:146.33ms
step:295/1480 train_time:41710ms step_avg:146.35ms
step:296/1480 train_time:41860ms step_avg:146.36ms
step:297/1480 train_time:42012ms step_avg:146.38ms
step:298/1480 train_time:42161ms step_avg:146.39ms
step:299/1480 train_time:42313ms step_avg:146.41ms
step:300/1480 train_time:42463ms step_avg:146.42ms
step:301/1480 train_time:42614ms step_avg:146.44ms
step:302/1480 train_time:42763ms step_avg:146.45ms
step:303/1480 train_time:42914ms step_avg:146.46ms
step:304/1480 train_time:43064ms step_avg:146.48ms
step:305/1480 train_time:43215ms step_avg:146.49ms
step:306/1480 train_time:43365ms step_avg:146.50ms
step:307/1480 train_time:43516ms step_avg:146.52ms
step:308/1480 train_time:43667ms step_avg:146.53ms
step:309/1480 train_time:43817ms step_avg:146.55ms
step:310/1480 train_time:43967ms step_avg:146.56ms
step:311/1480 train_time:44119ms step_avg:146.57ms
step:312/1480 train_time:44270ms step_avg:146.59ms
step:313/1480 train_time:44420ms step_avg:146.60ms
step:314/1480 train_time:44570ms step_avg:146.61ms
step:315/1480 train_time:44719ms step_avg:146.62ms
step:316/1480 train_time:44869ms step_avg:146.63ms
step:317/1480 train_time:45020ms step_avg:146.64ms
step:318/1480 train_time:45171ms step_avg:146.66ms
step:319/1480 train_time:45321ms step_avg:146.67ms
step:320/1480 train_time:45471ms step_avg:146.68ms
step:321/1480 train_time:45621ms step_avg:146.69ms
step:322/1480 train_time:45772ms step_avg:146.71ms
step:323/1480 train_time:45922ms step_avg:146.72ms
step:324/1480 train_time:46073ms step_avg:146.73ms
step:325/1480 train_time:46222ms step_avg:146.74ms
step:326/1480 train_time:46373ms step_avg:146.75ms
step:327/1480 train_time:46523ms step_avg:146.76ms
step:328/1480 train_time:46674ms step_avg:146.77ms
step:329/1480 train_time:46825ms step_avg:146.79ms
step:330/1480 train_time:46977ms step_avg:146.80ms
step:331/1480 train_time:47131ms step_avg:146.82ms
step:332/1480 train_time:47286ms step_avg:146.85ms
step:333/1480 train_time:47440ms step_avg:146.87ms
step:334/1480 train_time:47594ms step_avg:146.89ms
step:335/1480 train_time:47748ms step_avg:146.92ms
step:336/1480 train_time:47902ms step_avg:146.94ms
step:337/1480 train_time:48056ms step_avg:146.96ms
step:338/1480 train_time:48209ms step_avg:146.98ms
step:339/1480 train_time:48362ms step_avg:147.00ms
step:340/1480 train_time:48516ms step_avg:147.02ms
step:341/1480 train_time:48670ms step_avg:147.04ms
step:342/1480 train_time:48825ms step_avg:147.06ms
step:343/1480 train_time:48979ms step_avg:147.08ms
step:344/1480 train_time:49132ms step_avg:147.10ms
step:345/1480 train_time:49287ms step_avg:147.12ms
step:346/1480 train_time:49442ms step_avg:147.15ms
step:347/1480 train_time:49595ms step_avg:147.17ms
step:348/1480 train_time:49748ms step_avg:147.18ms
step:349/1480 train_time:49902ms step_avg:147.20ms
step:350/1480 train_time:50056ms step_avg:147.22ms
step:351/1480 train_time:50209ms step_avg:147.24ms
step:352/1480 train_time:50364ms step_avg:147.26ms
step:353/1480 train_time:50518ms step_avg:147.28ms
step:354/1480 train_time:50670ms step_avg:147.30ms
step:355/1480 train_time:50826ms step_avg:147.32ms
step:356/1480 train_time:50979ms step_avg:147.34ms
step:357/1480 train_time:51133ms step_avg:147.36ms
step:358/1480 train_time:51287ms step_avg:147.38ms
step:359/1480 train_time:51442ms step_avg:147.40ms
step:360/1480 train_time:51597ms step_avg:147.42ms
step:361/1480 train_time:51752ms step_avg:147.44ms
step:362/1480 train_time:51906ms step_avg:147.46ms
step:363/1480 train_time:52060ms step_avg:147.48ms
step:364/1480 train_time:52214ms step_avg:147.50ms
step:365/1480 train_time:52368ms step_avg:147.51ms
step:366/1480 train_time:52522ms step_avg:147.53ms
step:367/1480 train_time:52675ms step_avg:147.55ms
step:368/1480 train_time:52828ms step_avg:147.57ms
step:369/1480 train_time:52982ms step_avg:147.58ms
step:370/1480 train_time:53135ms step_avg:147.60ms
step:371/1480 train_time:53289ms step_avg:147.61ms
step:372/1480 train_time:53442ms step_avg:147.63ms
step:373/1480 train_time:53596ms step_avg:147.65ms
step:374/1480 train_time:53749ms step_avg:147.66ms
step:375/1480 train_time:53903ms step_avg:147.68ms
step:375/1480 val_loss:3.8065 train_time:53964ms step_avg:147.85ms
step:376/1480 train_time:54061ms step_avg:147.71ms
step:377/1480 train_time:54216ms step_avg:147.73ms
step:378/1480 train_time:54369ms step_avg:147.74ms
step:379/1480 train_time:54522ms step_avg:147.76ms
step:380/1480 train_time:54674ms step_avg:147.77ms
step:381/1480 train_time:54826ms step_avg:147.78ms
step:382/1480 train_time:54979ms step_avg:147.79ms
step:383/1480 train_time:55135ms step_avg:147.81ms
step:384/1480 train_time:55289ms step_avg:147.83ms
step:385/1480 train_time:55442ms step_avg:147.85ms
step:386/1480 train_time:55595ms step_avg:147.86ms
step:387/1480 train_time:55749ms step_avg:147.88ms
step:388/1480 train_time:55902ms step_avg:147.89ms
step:389/1480 train_time:56055ms step_avg:147.90ms
step:390/1480 train_time:56209ms step_avg:147.92ms
step:391/1480 train_time:56363ms step_avg:147.93ms
step:392/1480 train_time:56516ms step_avg:147.95ms
step:393/1480 train_time:56669ms step_avg:147.96ms
step:394/1480 train_time:56822ms step_avg:147.97ms
step:395/1480 train_time:56975ms step_avg:147.99ms
step:396/1480 train_time:57129ms step_avg:148.00ms
step:397/1480 train_time:57284ms step_avg:148.02ms
step:398/1480 train_time:57438ms step_avg:148.04ms
step:399/1480 train_time:57593ms step_avg:148.05ms
step:400/1480 train_time:57749ms step_avg:148.07ms
step:401/1480 train_time:57902ms step_avg:148.09ms
step:402/1480 train_time:58055ms step_avg:148.10ms
step:403/1480 train_time:58209ms step_avg:148.11ms
step:404/1480 train_time:58363ms step_avg:148.13ms
step:405/1480 train_time:58517ms step_avg:148.14ms
step:406/1480 train_time:58671ms step_avg:148.16ms
step:407/1480 train_time:58826ms step_avg:148.18ms
step:408/1480 train_time:58979ms step_avg:148.19ms
step:409/1480 train_time:59133ms step_avg:148.20ms
step:410/1480 train_time:59286ms step_avg:148.21ms
step:411/1480 train_time:59441ms step_avg:148.23ms
step:412/1480 train_time:59595ms step_avg:148.25ms
step:413/1480 train_time:59752ms step_avg:148.27ms
step:414/1480 train_time:59903ms step_avg:148.28ms
step:415/1480 train_time:60057ms step_avg:148.29ms
step:416/1480 train_time:60212ms step_avg:148.30ms
step:417/1480 train_time:60366ms step_avg:148.32ms
step:418/1480 train_time:60519ms step_avg:148.33ms
step:419/1480 train_time:60674ms step_avg:148.35ms
step:420/1480 train_time:60827ms step_avg:148.36ms
step:421/1480 train_time:60981ms step_avg:148.37ms
step:422/1480 train_time:61134ms step_avg:148.38ms
step:423/1480 train_time:61288ms step_avg:148.40ms
step:424/1480 train_time:61443ms step_avg:148.41ms
step:425/1480 train_time:61597ms step_avg:148.43ms
step:426/1480 train_time:61751ms step_avg:148.44ms
step:427/1480 train_time:61905ms step_avg:148.45ms
step:428/1480 train_time:62058ms step_avg:148.46ms
step:429/1480 train_time:62211ms step_avg:148.48ms
step:430/1480 train_time:62365ms step_avg:148.49ms
step:431/1480 train_time:62520ms step_avg:148.50ms
step:432/1480 train_time:62674ms step_avg:148.52ms
step:433/1480 train_time:62829ms step_avg:148.53ms
step:434/1480 train_time:62983ms step_avg:148.54ms
step:435/1480 train_time:63135ms step_avg:148.55ms
step:436/1480 train_time:63289ms step_avg:148.57ms
step:437/1480 train_time:63443ms step_avg:148.58ms
step:438/1480 train_time:63596ms step_avg:148.59ms
step:439/1480 train_time:63751ms step_avg:148.60ms
step:440/1480 train_time:63906ms step_avg:148.62ms
step:441/1480 train_time:64061ms step_avg:148.63ms
step:442/1480 train_time:64218ms step_avg:148.65ms
step:443/1480 train_time:64374ms step_avg:148.67ms
step:444/1480 train_time:64531ms step_avg:148.69ms
step:445/1480 train_time:64688ms step_avg:148.71ms
step:446/1480 train_time:64844ms step_avg:148.72ms
step:447/1480 train_time:64999ms step_avg:148.74ms
step:448/1480 train_time:65156ms step_avg:148.76ms
step:449/1480 train_time:65314ms step_avg:148.78ms
step:450/1480 train_time:65472ms step_avg:148.80ms
step:451/1480 train_time:65630ms step_avg:148.82ms
step:452/1480 train_time:65787ms step_avg:148.84ms
step:453/1480 train_time:65943ms step_avg:148.85ms
step:454/1480 train_time:66098ms step_avg:148.87ms
step:455/1480 train_time:66254ms step_avg:148.89ms
step:456/1480 train_time:66410ms step_avg:148.90ms
step:457/1480 train_time:66566ms step_avg:148.92ms
step:458/1480 train_time:66722ms step_avg:148.93ms
step:459/1480 train_time:66879ms step_avg:148.95ms
step:460/1480 train_time:67036ms step_avg:148.97ms
step:461/1480 train_time:67193ms step_avg:148.99ms
step:462/1480 train_time:67351ms step_avg:149.01ms
step:463/1480 train_time:67511ms step_avg:149.03ms
step:464/1480 train_time:67669ms step_avg:149.05ms
step:465/1480 train_time:67826ms step_avg:149.07ms
step:466/1480 train_time:67981ms step_avg:149.08ms
step:467/1480 train_time:68139ms step_avg:149.10ms
step:468/1480 train_time:68294ms step_avg:149.11ms
step:469/1480 train_time:68450ms step_avg:149.13ms
step:470/1480 train_time:68608ms step_avg:149.15ms
step:471/1480 train_time:68765ms step_avg:149.16ms
step:472/1480 train_time:68924ms step_avg:149.19ms
step:473/1480 train_time:69080ms step_avg:149.20ms
step:474/1480 train_time:69237ms step_avg:149.22ms
step:475/1480 train_time:69394ms step_avg:149.23ms
step:476/1480 train_time:69552ms step_avg:149.25ms
step:477/1480 train_time:69710ms step_avg:149.27ms
step:478/1480 train_time:69866ms step_avg:149.29ms
step:479/1480 train_time:70024ms step_avg:149.30ms
step:480/1480 train_time:70180ms step_avg:149.32ms
step:481/1480 train_time:70336ms step_avg:149.33ms
step:482/1480 train_time:70492ms step_avg:149.35ms
step:483/1480 train_time:70651ms step_avg:149.37ms
step:484/1480 train_time:70809ms step_avg:149.39ms
step:485/1480 train_time:70966ms step_avg:149.40ms
step:486/1480 train_time:71125ms step_avg:149.42ms
step:487/1480 train_time:71281ms step_avg:149.44ms
step:488/1480 train_time:71437ms step_avg:149.45ms
step:489/1480 train_time:71593ms step_avg:149.46ms
step:490/1480 train_time:71749ms step_avg:149.48ms
step:491/1480 train_time:71906ms step_avg:149.49ms
step:492/1480 train_time:72063ms step_avg:149.51ms
step:493/1480 train_time:72219ms step_avg:149.52ms
step:494/1480 train_time:72375ms step_avg:149.54ms
step:495/1480 train_time:72533ms step_avg:149.55ms
step:496/1480 train_time:72691ms step_avg:149.57ms
step:497/1480 train_time:72848ms step_avg:149.58ms
step:498/1480 train_time:73006ms step_avg:149.60ms
step:499/1480 train_time:73163ms step_avg:149.62ms
step:500/1480 train_time:73320ms step_avg:149.63ms
step:500/1480 val_loss:3.6856 train_time:73382ms step_avg:149.76ms
step:501/1480 train_time:73479ms step_avg:149.65ms
step:502/1480 train_time:73637ms step_avg:149.67ms
step:503/1480 train_time:73794ms step_avg:149.68ms
step:504/1480 train_time:73950ms step_avg:149.70ms
step:505/1480 train_time:74105ms step_avg:149.71ms
step:506/1480 train_time:74260ms step_avg:149.72ms
step:507/1480 train_time:74416ms step_avg:149.73ms
step:508/1480 train_time:74574ms step_avg:149.75ms
step:509/1480 train_time:74733ms step_avg:149.77ms
step:510/1480 train_time:74890ms step_avg:149.78ms
step:511/1480 train_time:75048ms step_avg:149.80ms
step:512/1480 train_time:75205ms step_avg:149.81ms
step:513/1480 train_time:75359ms step_avg:149.82ms
step:514/1480 train_time:75515ms step_avg:149.83ms
step:515/1480 train_time:75673ms step_avg:149.85ms
step:516/1480 train_time:75833ms step_avg:149.87ms
step:517/1480 train_time:75992ms step_avg:149.88ms
step:518/1480 train_time:76150ms step_avg:149.90ms
step:519/1480 train_time:76307ms step_avg:149.92ms
step:520/1480 train_time:76464ms step_avg:149.93ms
step:521/1480 train_time:76621ms step_avg:149.94ms
step:522/1480 train_time:76777ms step_avg:149.95ms
step:523/1480 train_time:76934ms step_avg:149.97ms
step:524/1480 train_time:77091ms step_avg:149.98ms
step:525/1480 train_time:77249ms step_avg:150.00ms
step:526/1480 train_time:77407ms step_avg:150.01ms
step:527/1480 train_time:77562ms step_avg:150.02ms
step:528/1480 train_time:77718ms step_avg:150.04ms
step:529/1480 train_time:77875ms step_avg:150.05ms
step:530/1480 train_time:78034ms step_avg:150.06ms
step:531/1480 train_time:78191ms step_avg:150.08ms
step:532/1480 train_time:78351ms step_avg:150.10ms
step:533/1480 train_time:78510ms step_avg:150.11ms
step:534/1480 train_time:78668ms step_avg:150.13ms
step:535/1480 train_time:78824ms step_avg:150.14ms
step:536/1480 train_time:78980ms step_avg:150.15ms
step:537/1480 train_time:79135ms step_avg:150.16ms
step:538/1480 train_time:79293ms step_avg:150.18ms
step:539/1480 train_time:79454ms step_avg:150.20ms
step:540/1480 train_time:79612ms step_avg:150.21ms
step:541/1480 train_time:79768ms step_avg:150.22ms
step:542/1480 train_time:79924ms step_avg:150.23ms
step:543/1480 train_time:80079ms step_avg:150.24ms
step:544/1480 train_time:80235ms step_avg:150.25ms
step:545/1480 train_time:80393ms step_avg:150.27ms
step:546/1480 train_time:80551ms step_avg:150.28ms
step:547/1480 train_time:80709ms step_avg:150.30ms
step:548/1480 train_time:80866ms step_avg:150.31ms
step:549/1480 train_time:81022ms step_avg:150.32ms
step:550/1480 train_time:81179ms step_avg:150.33ms
step:551/1480 train_time:81336ms step_avg:150.34ms
step:552/1480 train_time:81495ms step_avg:150.36ms
step:553/1480 train_time:81655ms step_avg:150.38ms
step:554/1480 train_time:81816ms step_avg:150.40ms
step:555/1480 train_time:81975ms step_avg:150.41ms
step:556/1480 train_time:82135ms step_avg:150.43ms
step:557/1480 train_time:82295ms step_avg:150.45ms
step:558/1480 train_time:82455ms step_avg:150.46ms
step:559/1480 train_time:82614ms step_avg:150.48ms
step:560/1480 train_time:82774ms step_avg:150.50ms
step:561/1480 train_time:82933ms step_avg:150.51ms
step:562/1480 train_time:83093ms step_avg:150.53ms
step:563/1480 train_time:83252ms step_avg:150.55ms
step:564/1480 train_time:83411ms step_avg:150.56ms
step:565/1480 train_time:83570ms step_avg:150.58ms
step:566/1480 train_time:83731ms step_avg:150.59ms
step:567/1480 train_time:83890ms step_avg:150.61ms
step:568/1480 train_time:84048ms step_avg:150.62ms
step:569/1480 train_time:84208ms step_avg:150.64ms
step:570/1480 train_time:84367ms step_avg:150.66ms
step:571/1480 train_time:84525ms step_avg:150.67ms
step:572/1480 train_time:84682ms step_avg:150.68ms
step:573/1480 train_time:84842ms step_avg:150.70ms
step:574/1480 train_time:85002ms step_avg:150.71ms
step:575/1480 train_time:85161ms step_avg:150.73ms
step:576/1480 train_time:85319ms step_avg:150.74ms
step:577/1480 train_time:85478ms step_avg:150.75ms
step:578/1480 train_time:85638ms step_avg:150.77ms
step:579/1480 train_time:85797ms step_avg:150.78ms
step:580/1480 train_time:85955ms step_avg:150.80ms
step:581/1480 train_time:86115ms step_avg:150.81ms
step:582/1480 train_time:86274ms step_avg:150.83ms
step:583/1480 train_time:86435ms step_avg:150.85ms
step:584/1480 train_time:86595ms step_avg:150.86ms
step:585/1480 train_time:86753ms step_avg:150.88ms
step:586/1480 train_time:86915ms step_avg:150.89ms
step:587/1480 train_time:87075ms step_avg:150.91ms
step:588/1480 train_time:87235ms step_avg:150.92ms
step:589/1480 train_time:87395ms step_avg:150.94ms
step:590/1480 train_time:87554ms step_avg:150.96ms
step:591/1480 train_time:87713ms step_avg:150.97ms
step:592/1480 train_time:87873ms step_avg:150.98ms
step:593/1480 train_time:88035ms step_avg:151.00ms
step:594/1480 train_time:88195ms step_avg:151.02ms
step:595/1480 train_time:88356ms step_avg:151.04ms
step:596/1480 train_time:88517ms step_avg:151.05ms
step:597/1480 train_time:88675ms step_avg:151.07ms
step:598/1480 train_time:88834ms step_avg:151.08ms
step:599/1480 train_time:88993ms step_avg:151.09ms
step:600/1480 train_time:89154ms step_avg:151.11ms
step:601/1480 train_time:89314ms step_avg:151.12ms
step:602/1480 train_time:89474ms step_avg:151.14ms
step:603/1480 train_time:89636ms step_avg:151.16ms
step:604/1480 train_time:89794ms step_avg:151.17ms
step:605/1480 train_time:89955ms step_avg:151.18ms
step:606/1480 train_time:90117ms step_avg:151.20ms
step:607/1480 train_time:90277ms step_avg:151.22ms
step:608/1480 train_time:90436ms step_avg:151.23ms
step:609/1480 train_time:90596ms step_avg:151.24ms
step:610/1480 train_time:90755ms step_avg:151.26ms
step:611/1480 train_time:90916ms step_avg:151.27ms
step:612/1480 train_time:91075ms step_avg:151.29ms
step:613/1480 train_time:91237ms step_avg:151.30ms
step:614/1480 train_time:91396ms step_avg:151.32ms
step:615/1480 train_time:91555ms step_avg:151.33ms
step:616/1480 train_time:91714ms step_avg:151.34ms
step:617/1480 train_time:91873ms step_avg:151.36ms
step:618/1480 train_time:92033ms step_avg:151.37ms
step:619/1480 train_time:92193ms step_avg:151.38ms
step:620/1480 train_time:92354ms step_avg:151.40ms
step:621/1480 train_time:92513ms step_avg:151.41ms
step:622/1480 train_time:92673ms step_avg:151.43ms
step:623/1480 train_time:92834ms step_avg:151.44ms
step:624/1480 train_time:92994ms step_avg:151.46ms
step:625/1480 train_time:93153ms step_avg:151.47ms
step:625/1480 val_loss:3.6047 train_time:93217ms step_avg:151.57ms
step:626/1480 train_time:93315ms step_avg:151.49ms
step:627/1480 train_time:93475ms step_avg:151.50ms
step:628/1480 train_time:93632ms step_avg:151.51ms
step:629/1480 train_time:93790ms step_avg:151.52ms
step:630/1480 train_time:93948ms step_avg:151.53ms
step:631/1480 train_time:94107ms step_avg:151.54ms
step:632/1480 train_time:94267ms step_avg:151.56ms
step:633/1480 train_time:94427ms step_avg:151.57ms
step:634/1480 train_time:94587ms step_avg:151.58ms
step:635/1480 train_time:94746ms step_avg:151.59ms
step:636/1480 train_time:94906ms step_avg:151.61ms
step:637/1480 train_time:95066ms step_avg:151.62ms
step:638/1480 train_time:95226ms step_avg:151.63ms
step:639/1480 train_time:95386ms step_avg:151.65ms
step:640/1480 train_time:95546ms step_avg:151.66ms
step:641/1480 train_time:95706ms step_avg:151.67ms
step:642/1480 train_time:95866ms step_avg:151.69ms
step:643/1480 train_time:96027ms step_avg:151.70ms
step:644/1480 train_time:96186ms step_avg:151.71ms
step:645/1480 train_time:96346ms step_avg:151.73ms
step:646/1480 train_time:96507ms step_avg:151.74ms
step:647/1480 train_time:96667ms step_avg:151.75ms
step:648/1480 train_time:96827ms step_avg:151.77ms
step:649/1480 train_time:96986ms step_avg:151.78ms
step:650/1480 train_time:97145ms step_avg:151.79ms
step:651/1480 train_time:97306ms step_avg:151.80ms
step:652/1480 train_time:97466ms step_avg:151.82ms
step:653/1480 train_time:97626ms step_avg:151.83ms
step:654/1480 train_time:97787ms step_avg:151.84ms
step:655/1480 train_time:97947ms step_avg:151.86ms
step:656/1480 train_time:98108ms step_avg:151.87ms
step:657/1480 train_time:98268ms step_avg:151.88ms
step:658/1480 train_time:98428ms step_avg:151.89ms
step:659/1480 train_time:98590ms step_avg:151.91ms
step:660/1480 train_time:98751ms step_avg:151.92ms
step:661/1480 train_time:98913ms step_avg:151.94ms
step:662/1480 train_time:99073ms step_avg:151.95ms
step:663/1480 train_time:99232ms step_avg:151.96ms
step:664/1480 train_time:99394ms step_avg:151.98ms
step:665/1480 train_time:99556ms step_avg:151.99ms
step:666/1480 train_time:99715ms step_avg:152.00ms
step:667/1480 train_time:99877ms step_avg:152.02ms
step:668/1480 train_time:100038ms step_avg:152.03ms
step:669/1480 train_time:100201ms step_avg:152.05ms
step:670/1480 train_time:100362ms step_avg:152.06ms
step:671/1480 train_time:100523ms step_avg:152.08ms
step:672/1480 train_time:100687ms step_avg:152.09ms
step:673/1480 train_time:100849ms step_avg:152.11ms
step:674/1480 train_time:101011ms step_avg:152.12ms
step:675/1480 train_time:101172ms step_avg:152.14ms
step:676/1480 train_time:101334ms step_avg:152.15ms
step:677/1480 train_time:101495ms step_avg:152.17ms
step:678/1480 train_time:101655ms step_avg:152.18ms
step:679/1480 train_time:101816ms step_avg:152.19ms
step:680/1480 train_time:101979ms step_avg:152.21ms
step:681/1480 train_time:102139ms step_avg:152.22ms
step:682/1480 train_time:102301ms step_avg:152.23ms
step:683/1480 train_time:102464ms step_avg:152.25ms
step:684/1480 train_time:102625ms step_avg:152.26ms
step:685/1480 train_time:102789ms step_avg:152.28ms
step:686/1480 train_time:102950ms step_avg:152.29ms
step:687/1480 train_time:103110ms step_avg:152.30ms
step:688/1480 train_time:103273ms step_avg:152.32ms
step:689/1480 train_time:103435ms step_avg:152.33ms
step:690/1480 train_time:103598ms step_avg:152.35ms
step:691/1480 train_time:103759ms step_avg:152.36ms
step:692/1480 train_time:103921ms step_avg:152.38ms
step:693/1480 train_time:104084ms step_avg:152.39ms
step:694/1480 train_time:104247ms step_avg:152.41ms
step:695/1480 train_time:104409ms step_avg:152.42ms
step:696/1480 train_time:104570ms step_avg:152.43ms
step:697/1480 train_time:104734ms step_avg:152.45ms
step:698/1480 train_time:104894ms step_avg:152.46ms
step:699/1480 train_time:105056ms step_avg:152.48ms
step:700/1480 train_time:105217ms step_avg:152.49ms
step:701/1480 train_time:105378ms step_avg:152.50ms
step:702/1480 train_time:105537ms step_avg:152.51ms
step:703/1480 train_time:105697ms step_avg:152.52ms
step:704/1480 train_time:105858ms step_avg:152.53ms
step:705/1480 train_time:106021ms step_avg:152.55ms
step:706/1480 train_time:106187ms step_avg:152.57ms
step:707/1480 train_time:106349ms step_avg:152.58ms
step:708/1480 train_time:106511ms step_avg:152.59ms
step:709/1480 train_time:106672ms step_avg:152.61ms
step:710/1480 train_time:106832ms step_avg:152.62ms
step:711/1480 train_time:106994ms step_avg:152.63ms
step:712/1480 train_time:107162ms step_avg:152.65ms
step:713/1480 train_time:107325ms step_avg:152.67ms
step:714/1480 train_time:107488ms step_avg:152.68ms
step:715/1480 train_time:107648ms step_avg:152.69ms
step:716/1480 train_time:107808ms step_avg:152.70ms
step:717/1480 train_time:107970ms step_avg:152.72ms
step:718/1480 train_time:108130ms step_avg:152.73ms
step:719/1480 train_time:108290ms step_avg:152.74ms
step:720/1480 train_time:108454ms step_avg:152.75ms
step:721/1480 train_time:108614ms step_avg:152.76ms
step:722/1480 train_time:108775ms step_avg:152.77ms
step:723/1480 train_time:108934ms step_avg:152.78ms
step:724/1480 train_time:109094ms step_avg:152.79ms
step:725/1480 train_time:109256ms step_avg:152.81ms
step:726/1480 train_time:109421ms step_avg:152.82ms
step:727/1480 train_time:109586ms step_avg:152.84ms
step:728/1480 train_time:109747ms step_avg:152.85ms
step:729/1480 train_time:109910ms step_avg:152.87ms
step:730/1480 train_time:110072ms step_avg:152.88ms
step:731/1480 train_time:110232ms step_avg:152.89ms
step:732/1480 train_time:110392ms step_avg:152.90ms
step:733/1480 train_time:110554ms step_avg:152.91ms
step:734/1480 train_time:110716ms step_avg:152.92ms
step:735/1480 train_time:110876ms step_avg:152.93ms
step:736/1480 train_time:111037ms step_avg:152.94ms
step:737/1480 train_time:111197ms step_avg:152.95ms
step:738/1480 train_time:111359ms step_avg:152.97ms
step:739/1480 train_time:111520ms step_avg:152.98ms
step:740/1480 train_time:111687ms step_avg:153.00ms
step:741/1480 train_time:111852ms step_avg:153.01ms
step:742/1480 train_time:112013ms step_avg:153.02ms
step:743/1480 train_time:112174ms step_avg:153.03ms
step:744/1480 train_time:112335ms step_avg:153.05ms
step:745/1480 train_time:112501ms step_avg:153.06ms
step:746/1480 train_time:112662ms step_avg:153.07ms
step:747/1480 train_time:112823ms step_avg:153.08ms
step:748/1480 train_time:112989ms step_avg:153.10ms
step:749/1480 train_time:113152ms step_avg:153.11ms
step:750/1480 train_time:113311ms step_avg:153.12ms
step:750/1480 val_loss:3.5475 train_time:113375ms step_avg:153.21ms
step:751/1480 train_time:113476ms step_avg:153.14ms
step:752/1480 train_time:113641ms step_avg:153.15ms
step:753/1480 train_time:113802ms step_avg:153.17ms
step:754/1480 train_time:113962ms step_avg:153.17ms
step:755/1480 train_time:114122ms step_avg:153.18ms
step:756/1480 train_time:114284ms step_avg:153.20ms
step:757/1480 train_time:114447ms step_avg:153.21ms
step:758/1480 train_time:114608ms step_avg:153.22ms
step:759/1480 train_time:114770ms step_avg:153.23ms
step:760/1480 train_time:114931ms step_avg:153.24ms
step:761/1480 train_time:115093ms step_avg:153.25ms
step:762/1480 train_time:115255ms step_avg:153.26ms
step:763/1480 train_time:115418ms step_avg:153.28ms
step:764/1480 train_time:115581ms step_avg:153.29ms
step:765/1480 train_time:115743ms step_avg:153.30ms
step:766/1480 train_time:115905ms step_avg:153.31ms
step:767/1480 train_time:116066ms step_avg:153.32ms
step:768/1480 train_time:116228ms step_avg:153.34ms
step:769/1480 train_time:116392ms step_avg:153.35ms
step:770/1480 train_time:116555ms step_avg:153.36ms
step:771/1480 train_time:116721ms step_avg:153.38ms
step:772/1480 train_time:116884ms step_avg:153.39ms
step:773/1480 train_time:117046ms step_avg:153.40ms
step:774/1480 train_time:117207ms step_avg:153.41ms
step:775/1480 train_time:117369ms step_avg:153.42ms
step:776/1480 train_time:117533ms step_avg:153.44ms
step:777/1480 train_time:117701ms step_avg:153.46ms
step:778/1480 train_time:117865ms step_avg:153.47ms
step:779/1480 train_time:118027ms step_avg:153.48ms
step:780/1480 train_time:118190ms step_avg:153.49ms
step:781/1480 train_time:118354ms step_avg:153.51ms
step:782/1480 train_time:118519ms step_avg:153.52ms
step:783/1480 train_time:118681ms step_avg:153.53ms
step:784/1480 train_time:118844ms step_avg:153.55ms
step:785/1480 train_time:119006ms step_avg:153.56ms
step:786/1480 train_time:119169ms step_avg:153.57ms
step:787/1480 train_time:119332ms step_avg:153.58ms
step:788/1480 train_time:119497ms step_avg:153.60ms
step:789/1480 train_time:119660ms step_avg:153.61ms
step:790/1480 train_time:119826ms step_avg:153.62ms
step:791/1480 train_time:119994ms step_avg:153.64ms
step:792/1480 train_time:120160ms step_avg:153.66ms
step:793/1480 train_time:120322ms step_avg:153.67ms
step:794/1480 train_time:120486ms step_avg:153.68ms
step:795/1480 train_time:120652ms step_avg:153.70ms
step:796/1480 train_time:120819ms step_avg:153.71ms
step:797/1480 train_time:120984ms step_avg:153.73ms
step:798/1480 train_time:121148ms step_avg:153.74ms
step:799/1480 train_time:121315ms step_avg:153.76ms
step:800/1480 train_time:121479ms step_avg:153.77ms
step:801/1480 train_time:121643ms step_avg:153.78ms
step:802/1480 train_time:121810ms step_avg:153.80ms
step:803/1480 train_time:121971ms step_avg:153.81ms
step:804/1480 train_time:122134ms step_avg:153.82ms
step:805/1480 train_time:122300ms step_avg:153.84ms
step:806/1480 train_time:122462ms step_avg:153.85ms
step:807/1480 train_time:122624ms step_avg:153.86ms
step:808/1480 train_time:122789ms step_avg:153.87ms
step:809/1480 train_time:122949ms step_avg:153.88ms
step:810/1480 train_time:123110ms step_avg:153.89ms
step:811/1480 train_time:123271ms step_avg:153.90ms
step:812/1480 train_time:123436ms step_avg:153.91ms
step:813/1480 train_time:123597ms step_avg:153.92ms
step:814/1480 train_time:123762ms step_avg:153.93ms
step:815/1480 train_time:123924ms step_avg:153.94ms
step:816/1480 train_time:124088ms step_avg:153.96ms
step:817/1480 train_time:124251ms step_avg:153.97ms
step:818/1480 train_time:124411ms step_avg:153.97ms
step:819/1480 train_time:124574ms step_avg:153.99ms
step:820/1480 train_time:124741ms step_avg:154.00ms
step:821/1480 train_time:124902ms step_avg:154.01ms
step:822/1480 train_time:125066ms step_avg:154.02ms
step:823/1480 train_time:125228ms step_avg:154.03ms
step:824/1480 train_time:125389ms step_avg:154.04ms
step:825/1480 train_time:125554ms step_avg:154.05ms
step:826/1480 train_time:125722ms step_avg:154.07ms
step:827/1480 train_time:125886ms step_avg:154.08ms
step:828/1480 train_time:126049ms step_avg:154.09ms
step:829/1480 train_time:126212ms step_avg:154.11ms
step:830/1480 train_time:126377ms step_avg:154.12ms
step:831/1480 train_time:126542ms step_avg:154.13ms
step:832/1480 train_time:126705ms step_avg:154.14ms
step:833/1480 train_time:126869ms step_avg:154.15ms
step:834/1480 train_time:127034ms step_avg:154.17ms
step:835/1480 train_time:127197ms step_avg:154.18ms
step:836/1480 train_time:127365ms step_avg:154.19ms
step:837/1480 train_time:127526ms step_avg:154.20ms
step:838/1480 train_time:127689ms step_avg:154.21ms
step:839/1480 train_time:127851ms step_avg:154.22ms
step:840/1480 train_time:128011ms step_avg:154.23ms
step:841/1480 train_time:128171ms step_avg:154.24ms
step:842/1480 train_time:128335ms step_avg:154.25ms
step:843/1480 train_time:128497ms step_avg:154.26ms
step:844/1480 train_time:128661ms step_avg:154.27ms
step:845/1480 train_time:128824ms step_avg:154.28ms
step:846/1480 train_time:128988ms step_avg:154.29ms
step:847/1480 train_time:129152ms step_avg:154.30ms
step:848/1480 train_time:129315ms step_avg:154.31ms
step:849/1480 train_time:129478ms step_avg:154.32ms
step:850/1480 train_time:129642ms step_avg:154.34ms
step:851/1480 train_time:129806ms step_avg:154.35ms
step:852/1480 train_time:129968ms step_avg:154.36ms
step:853/1480 train_time:130130ms step_avg:154.37ms
step:854/1480 train_time:130296ms step_avg:154.38ms
step:855/1480 train_time:130461ms step_avg:154.39ms
step:856/1480 train_time:130623ms step_avg:154.40ms
step:857/1480 train_time:130788ms step_avg:154.41ms
step:858/1480 train_time:130955ms step_avg:154.43ms
step:859/1480 train_time:131120ms step_avg:154.44ms
step:860/1480 train_time:131282ms step_avg:154.45ms
step:861/1480 train_time:131447ms step_avg:154.46ms
step:862/1480 train_time:131617ms step_avg:154.48ms
step:863/1480 train_time:131785ms step_avg:154.50ms
step:864/1480 train_time:131950ms step_avg:154.51ms
step:865/1480 train_time:132111ms step_avg:154.52ms
step:866/1480 train_time:132277ms step_avg:154.53ms
step:867/1480 train_time:132442ms step_avg:154.54ms
step:868/1480 train_time:132603ms step_avg:154.55ms
step:869/1480 train_time:132765ms step_avg:154.56ms
step:870/1480 train_time:132929ms step_avg:154.57ms
step:871/1480 train_time:133091ms step_avg:154.58ms
step:872/1480 train_time:133256ms step_avg:154.59ms
step:873/1480 train_time:133420ms step_avg:154.60ms
step:874/1480 train_time:133585ms step_avg:154.61ms
step:875/1480 train_time:133750ms step_avg:154.62ms
step:875/1480 val_loss:3.5027 train_time:133814ms step_avg:154.70ms
step:876/1480 train_time:133915ms step_avg:154.64ms
step:877/1480 train_time:134079ms step_avg:154.65ms
step:878/1480 train_time:134241ms step_avg:154.66ms
step:879/1480 train_time:134405ms step_avg:154.67ms
step:880/1480 train_time:134570ms step_avg:154.68ms
step:881/1480 train_time:134733ms step_avg:154.69ms
step:882/1480 train_time:134899ms step_avg:154.70ms
step:883/1480 train_time:135065ms step_avg:154.71ms
step:884/1480 train_time:135232ms step_avg:154.73ms
step:885/1480 train_time:135397ms step_avg:154.74ms
step:886/1480 train_time:135564ms step_avg:154.75ms
step:887/1480 train_time:135733ms step_avg:154.77ms
step:888/1480 train_time:135907ms step_avg:154.79ms
step:889/1480 train_time:136075ms step_avg:154.81ms
step:890/1480 train_time:136237ms step_avg:154.81ms
step:891/1480 train_time:136402ms step_avg:154.83ms
step:892/1480 train_time:136566ms step_avg:154.84ms
step:893/1480 train_time:136729ms step_avg:154.85ms
step:894/1480 train_time:136897ms step_avg:154.86ms
step:895/1480 train_time:137065ms step_avg:154.88ms
step:896/1480 train_time:137231ms step_avg:154.89ms
step:897/1480 train_time:137396ms step_avg:154.90ms
step:898/1480 train_time:137563ms step_avg:154.91ms
step:899/1480 train_time:137728ms step_avg:154.92ms
step:900/1480 train_time:137892ms step_avg:154.94ms
step:901/1480 train_time:138057ms step_avg:154.95ms
step:902/1480 train_time:138219ms step_avg:154.95ms
step:903/1480 train_time:138391ms step_avg:154.97ms
step:904/1480 train_time:138556ms step_avg:154.98ms
step:905/1480 train_time:138717ms step_avg:154.99ms
step:906/1480 train_time:138884ms step_avg:155.00ms
step:907/1480 train_time:139052ms step_avg:155.02ms
step:908/1480 train_time:139214ms step_avg:155.03ms
step:909/1480 train_time:139379ms step_avg:155.04ms
step:910/1480 train_time:139550ms step_avg:155.06ms
step:911/1480 train_time:139715ms step_avg:155.07ms
step:912/1480 train_time:139881ms step_avg:155.08ms
step:913/1480 train_time:140048ms step_avg:155.09ms
step:914/1480 train_time:140215ms step_avg:155.11ms
step:915/1480 train_time:140385ms step_avg:155.12ms
step:916/1480 train_time:140549ms step_avg:155.13ms
step:917/1480 train_time:140714ms step_avg:155.14ms
step:918/1480 train_time:140880ms step_avg:155.15ms
step:919/1480 train_time:141050ms step_avg:155.17ms
step:920/1480 train_time:141215ms step_avg:155.18ms
step:921/1480 train_time:141380ms step_avg:155.19ms
step:922/1480 train_time:141548ms step_avg:155.21ms
step:923/1480 train_time:141712ms step_avg:155.22ms
step:924/1480 train_time:141875ms step_avg:155.22ms
step:925/1480 train_time:142040ms step_avg:155.23ms
step:926/1480 train_time:142202ms step_avg:155.24ms
step:927/1480 train_time:142368ms step_avg:155.25ms
step:928/1480 train_time:142533ms step_avg:155.26ms
step:929/1480 train_time:142698ms step_avg:155.28ms
step:930/1480 train_time:142864ms step_avg:155.29ms
step:931/1480 train_time:143027ms step_avg:155.30ms
step:932/1480 train_time:143193ms step_avg:155.31ms
step:933/1480 train_time:143360ms step_avg:155.32ms
step:934/1480 train_time:143526ms step_avg:155.33ms
step:935/1480 train_time:143698ms step_avg:155.35ms
step:936/1480 train_time:143865ms step_avg:155.36ms
step:937/1480 train_time:144034ms step_avg:155.38ms
step:938/1480 train_time:144197ms step_avg:155.38ms
step:939/1480 train_time:144366ms step_avg:155.40ms
step:940/1480 train_time:144533ms step_avg:155.41ms
step:941/1480 train_time:144697ms step_avg:155.42ms
step:942/1480 train_time:144862ms step_avg:155.43ms
step:943/1480 train_time:145033ms step_avg:155.45ms
step:944/1480 train_time:145204ms step_avg:155.47ms
step:945/1480 train_time:145369ms step_avg:155.47ms
step:946/1480 train_time:145537ms step_avg:155.49ms
step:947/1480 train_time:145704ms step_avg:155.50ms
step:948/1480 train_time:145870ms step_avg:155.51ms
step:949/1480 train_time:146035ms step_avg:155.52ms
step:950/1480 train_time:146199ms step_avg:155.53ms
step:951/1480 train_time:146369ms step_avg:155.55ms
step:952/1480 train_time:146534ms step_avg:155.56ms
step:953/1480 train_time:146703ms step_avg:155.57ms
step:954/1480 train_time:146872ms step_avg:155.58ms
step:955/1480 train_time:147035ms step_avg:155.59ms
step:956/1480 train_time:147201ms step_avg:155.60ms
step:957/1480 train_time:147370ms step_avg:155.62ms
step:958/1480 train_time:147539ms step_avg:155.63ms
step:959/1480 train_time:147705ms step_avg:155.64ms
step:960/1480 train_time:147874ms step_avg:155.66ms
step:961/1480 train_time:148039ms step_avg:155.67ms
step:962/1480 train_time:148202ms step_avg:155.67ms
step:963/1480 train_time:148368ms step_avg:155.69ms
step:964/1480 train_time:148535ms step_avg:155.70ms
step:965/1480 train_time:148698ms step_avg:155.70ms
step:966/1480 train_time:148863ms step_avg:155.71ms
step:967/1480 train_time:149027ms step_avg:155.72ms
step:968/1480 train_time:149192ms step_avg:155.73ms
step:969/1480 train_time:149356ms step_avg:155.74ms
step:970/1480 train_time:149518ms step_avg:155.75ms
step:971/1480 train_time:149683ms step_avg:155.76ms
step:972/1480 train_time:149848ms step_avg:155.77ms
step:973/1480 train_time:150013ms step_avg:155.78ms
step:974/1480 train_time:150181ms step_avg:155.79ms
step:975/1480 train_time:150347ms step_avg:155.80ms
step:976/1480 train_time:150512ms step_avg:155.81ms
step:977/1480 train_time:150676ms step_avg:155.82ms
step:978/1480 train_time:150841ms step_avg:155.83ms
step:979/1480 train_time:151007ms step_avg:155.84ms
step:980/1480 train_time:151172ms step_avg:155.85ms
step:981/1480 train_time:151341ms step_avg:155.86ms
step:982/1480 train_time:151504ms step_avg:155.87ms
step:983/1480 train_time:151669ms step_avg:155.88ms
step:984/1480 train_time:151834ms step_avg:155.89ms
step:985/1480 train_time:152001ms step_avg:155.90ms
step:986/1480 train_time:152165ms step_avg:155.91ms
step:987/1480 train_time:152330ms step_avg:155.92ms
step:988/1480 train_time:152497ms step_avg:155.93ms
step:989/1480 train_time:152661ms step_avg:155.94ms
step:990/1480 train_time:152831ms step_avg:155.95ms
step:991/1480 train_time:152998ms step_avg:155.96ms
step:992/1480 train_time:153173ms step_avg:155.98ms
step:993/1480 train_time:153350ms step_avg:156.00ms
step:994/1480 train_time:153515ms step_avg:156.01ms
step:995/1480 train_time:153679ms step_avg:156.02ms
step:996/1480 train_time:153840ms step_avg:156.02ms
step:997/1480 train_time:154004ms step_avg:156.03ms
step:998/1480 train_time:154168ms step_avg:156.04ms
step:999/1480 train_time:154333ms step_avg:156.05ms
step:1000/1480 train_time:154503ms step_avg:156.06ms
step:1000/1480 val_loss:3.4412 train_time:154570ms step_avg:156.13ms
step:1001/1480 train_time:154670ms step_avg:156.07ms
step:1002/1480 train_time:154838ms step_avg:156.09ms
step:1003/1480 train_time:155009ms step_avg:156.10ms
step:1004/1480 train_time:155176ms step_avg:156.11ms
step:1005/1480 train_time:155345ms step_avg:156.13ms
step:1006/1480 train_time:155511ms step_avg:156.14ms
step:1007/1480 train_time:155678ms step_avg:156.15ms
step:1008/1480 train_time:155847ms step_avg:156.16ms
step:1009/1480 train_time:156022ms step_avg:156.18ms
step:1010/1480 train_time:156188ms step_avg:156.19ms
step:1011/1480 train_time:156353ms step_avg:156.20ms
step:1012/1480 train_time:156520ms step_avg:156.21ms
step:1013/1480 train_time:156689ms step_avg:156.22ms
step:1014/1480 train_time:156856ms step_avg:156.23ms
step:1015/1480 train_time:157026ms step_avg:156.24ms
step:1016/1480 train_time:157193ms step_avg:156.26ms
step:1017/1480 train_time:157366ms step_avg:156.27ms
step:1018/1480 train_time:157535ms step_avg:156.28ms
step:1019/1480 train_time:157704ms step_avg:156.30ms
step:1020/1480 train_time:157873ms step_avg:156.31ms
step:1021/1480 train_time:158038ms step_avg:156.32ms
step:1022/1480 train_time:158206ms step_avg:156.33ms
step:1023/1480 train_time:158373ms step_avg:156.34ms
step:1024/1480 train_time:158539ms step_avg:156.35ms
step:1025/1480 train_time:158710ms step_avg:156.36ms
step:1026/1480 train_time:158875ms step_avg:156.37ms
step:1027/1480 train_time:159041ms step_avg:156.38ms
step:1028/1480 train_time:159213ms step_avg:156.40ms
step:1029/1480 train_time:159386ms step_avg:156.41ms
step:1030/1480 train_time:159554ms step_avg:156.43ms
step:1031/1480 train_time:159718ms step_avg:156.43ms
step:1032/1480 train_time:159890ms step_avg:156.45ms
step:1033/1480 train_time:160055ms step_avg:156.46ms
step:1034/1480 train_time:160224ms step_avg:156.47ms
step:1035/1480 train_time:160390ms step_avg:156.48ms
step:1036/1480 train_time:160556ms step_avg:156.49ms
step:1037/1480 train_time:160724ms step_avg:156.50ms
step:1038/1480 train_time:160891ms step_avg:156.51ms
step:1039/1480 train_time:161063ms step_avg:156.52ms
step:1040/1480 train_time:161231ms step_avg:156.53ms
step:1041/1480 train_time:161397ms step_avg:156.54ms
step:1042/1480 train_time:161563ms step_avg:156.55ms
step:1043/1480 train_time:161729ms step_avg:156.56ms
step:1044/1480 train_time:161893ms step_avg:156.57ms
step:1045/1480 train_time:162065ms step_avg:156.58ms
step:1046/1480 train_time:162233ms step_avg:156.60ms
step:1047/1480 train_time:162400ms step_avg:156.61ms
step:1048/1480 train_time:162566ms step_avg:156.61ms
step:1049/1480 train_time:162731ms step_avg:156.62ms
step:1050/1480 train_time:162901ms step_avg:156.64ms
step:1051/1480 train_time:163070ms step_avg:156.65ms
step:1052/1480 train_time:163238ms step_avg:156.66ms
step:1053/1480 train_time:163405ms step_avg:156.67ms
step:1054/1480 train_time:163572ms step_avg:156.68ms
step:1055/1480 train_time:163738ms step_avg:156.69ms
step:1056/1480 train_time:163903ms step_avg:156.70ms
step:1057/1480 train_time:164069ms step_avg:156.70ms
step:1058/1480 train_time:164239ms step_avg:156.72ms
step:1059/1480 train_time:164411ms step_avg:156.73ms
step:1060/1480 train_time:164579ms step_avg:156.74ms
step:1061/1480 train_time:164744ms step_avg:156.75ms
step:1062/1480 train_time:164909ms step_avg:156.76ms
step:1063/1480 train_time:165072ms step_avg:156.76ms
step:1064/1480 train_time:165235ms step_avg:156.77ms
step:1065/1480 train_time:165404ms step_avg:156.78ms
step:1066/1480 train_time:165571ms step_avg:156.79ms
step:1067/1480 train_time:165743ms step_avg:156.81ms
step:1068/1480 train_time:165910ms step_avg:156.81ms
step:1069/1480 train_time:166082ms step_avg:156.83ms
step:1070/1480 train_time:166248ms step_avg:156.84ms
step:1071/1480 train_time:166421ms step_avg:156.85ms
step:1072/1480 train_time:166587ms step_avg:156.86ms
step:1073/1480 train_time:166750ms step_avg:156.87ms
step:1074/1480 train_time:166916ms step_avg:156.88ms
step:1075/1480 train_time:167088ms step_avg:156.89ms
step:1076/1480 train_time:167254ms step_avg:156.90ms
step:1077/1480 train_time:167419ms step_avg:156.91ms
step:1078/1480 train_time:167593ms step_avg:156.92ms
step:1079/1480 train_time:167766ms step_avg:156.94ms
step:1080/1480 train_time:167936ms step_avg:156.95ms
step:1081/1480 train_time:168104ms step_avg:156.96ms
step:1082/1480 train_time:168270ms step_avg:156.97ms
step:1083/1480 train_time:168436ms step_avg:156.98ms
step:1084/1480 train_time:168602ms step_avg:156.99ms
step:1085/1480 train_time:168772ms step_avg:157.00ms
step:1086/1480 train_time:168942ms step_avg:157.01ms
step:1087/1480 train_time:169109ms step_avg:157.02ms
step:1088/1480 train_time:169278ms step_avg:157.03ms
step:1089/1480 train_time:169452ms step_avg:157.05ms
step:1090/1480 train_time:169623ms step_avg:157.06ms
step:1091/1480 train_time:169790ms step_avg:157.07ms
step:1092/1480 train_time:169958ms step_avg:157.08ms
step:1093/1480 train_time:170127ms step_avg:157.09ms
step:1094/1480 train_time:170292ms step_avg:157.10ms
step:1095/1480 train_time:170458ms step_avg:157.10ms
step:1096/1480 train_time:170627ms step_avg:157.12ms
step:1097/1480 train_time:170796ms step_avg:157.13ms
step:1098/1480 train_time:170967ms step_avg:157.14ms
step:1099/1480 train_time:171137ms step_avg:157.15ms
step:1100/1480 train_time:171309ms step_avg:157.16ms
step:1101/1480 train_time:171481ms step_avg:157.18ms
step:1102/1480 train_time:171652ms step_avg:157.19ms
step:1103/1480 train_time:171829ms step_avg:157.21ms
step:1104/1480 train_time:171997ms step_avg:157.22ms
step:1105/1480 train_time:172167ms step_avg:157.23ms
step:1106/1480 train_time:172334ms step_avg:157.24ms
step:1107/1480 train_time:172504ms step_avg:157.25ms
step:1108/1480 train_time:172670ms step_avg:157.26ms
step:1109/1480 train_time:172836ms step_avg:157.27ms
step:1110/1480 train_time:173002ms step_avg:157.27ms
step:1111/1480 train_time:173168ms step_avg:157.28ms
step:1112/1480 train_time:173338ms step_avg:157.29ms
step:1113/1480 train_time:173516ms step_avg:157.31ms
step:1114/1480 train_time:173690ms step_avg:157.33ms
step:1115/1480 train_time:173862ms step_avg:157.34ms
step:1116/1480 train_time:174029ms step_avg:157.35ms
step:1117/1480 train_time:174202ms step_avg:157.36ms
step:1118/1480 train_time:174377ms step_avg:157.38ms
step:1119/1480 train_time:174544ms step_avg:157.39ms
step:1120/1480 train_time:174712ms step_avg:157.40ms
step:1121/1480 train_time:174882ms step_avg:157.41ms
step:1122/1480 train_time:175049ms step_avg:157.42ms
step:1123/1480 train_time:175216ms step_avg:157.43ms
step:1124/1480 train_time:175385ms step_avg:157.44ms
step:1125/1480 train_time:175552ms step_avg:157.45ms
step:1125/1480 val_loss:3.3851 train_time:175619ms step_avg:157.51ms
step:1126/1480 train_time:175720ms step_avg:157.46ms
step:1127/1480 train_time:175890ms step_avg:157.47ms
step:1128/1480 train_time:176062ms step_avg:157.48ms
step:1129/1480 train_time:176236ms step_avg:157.49ms
step:1130/1480 train_time:176405ms step_avg:157.50ms
step:1131/1480 train_time:176581ms step_avg:157.52ms
step:1132/1480 train_time:176746ms step_avg:157.53ms
step:1133/1480 train_time:176918ms step_avg:157.54ms
step:1134/1480 train_time:177088ms step_avg:157.55ms
step:1135/1480 train_time:177256ms step_avg:157.56ms
step:1136/1480 train_time:177426ms step_avg:157.57ms
step:1137/1480 train_time:177596ms step_avg:157.58ms
step:1138/1480 train_time:177769ms step_avg:157.60ms
step:1139/1480 train_time:177937ms step_avg:157.61ms
step:1140/1480 train_time:178105ms step_avg:157.62ms
step:1141/1480 train_time:178278ms step_avg:157.63ms
step:1142/1480 train_time:178445ms step_avg:157.64ms
step:1143/1480 train_time:178616ms step_avg:157.65ms
step:1144/1480 train_time:178783ms step_avg:157.66ms
step:1145/1480 train_time:178949ms step_avg:157.66ms
step:1146/1480 train_time:179119ms step_avg:157.68ms
step:1147/1480 train_time:179288ms step_avg:157.68ms
step:1148/1480 train_time:179457ms step_avg:157.70ms
step:1149/1480 train_time:179629ms step_avg:157.71ms
step:1150/1480 train_time:179798ms step_avg:157.72ms
step:1151/1480 train_time:179973ms step_avg:157.73ms
step:1152/1480 train_time:180144ms step_avg:157.74ms
step:1153/1480 train_time:180316ms step_avg:157.76ms
step:1154/1480 train_time:180482ms step_avg:157.76ms
step:1155/1480 train_time:180655ms step_avg:157.78ms
step:1156/1480 train_time:180834ms step_avg:157.80ms
step:1157/1480 train_time:181004ms step_avg:157.81ms
step:1158/1480 train_time:181173ms step_avg:157.82ms
step:1159/1480 train_time:181340ms step_avg:157.82ms
step:1160/1480 train_time:181507ms step_avg:157.83ms
step:1161/1480 train_time:181677ms step_avg:157.84ms
step:1162/1480 train_time:181847ms step_avg:157.85ms
step:1163/1480 train_time:182016ms step_avg:157.86ms
step:1164/1480 train_time:182183ms step_avg:157.87ms
step:1165/1480 train_time:182349ms step_avg:157.88ms
step:1166/1480 train_time:182518ms step_avg:157.89ms
step:1167/1480 train_time:182687ms step_avg:157.90ms
step:1168/1480 train_time:182855ms step_avg:157.91ms
step:1169/1480 train_time:183023ms step_avg:157.91ms
step:1170/1480 train_time:183193ms step_avg:157.92ms
step:1171/1480 train_time:183359ms step_avg:157.93ms
step:1172/1480 train_time:183524ms step_avg:157.94ms
step:1173/1480 train_time:183696ms step_avg:157.95ms
step:1174/1480 train_time:183877ms step_avg:157.97ms
step:1175/1480 train_time:184048ms step_avg:157.98ms
step:1176/1480 train_time:184218ms step_avg:157.99ms
step:1177/1480 train_time:184395ms step_avg:158.01ms
step:1178/1480 train_time:184563ms step_avg:158.02ms
step:1179/1480 train_time:184729ms step_avg:158.02ms
step:1180/1480 train_time:184910ms step_avg:158.04ms
step:1181/1480 train_time:185080ms step_avg:158.05ms
step:1182/1480 train_time:185250ms step_avg:158.06ms
step:1183/1480 train_time:185420ms step_avg:158.07ms
step:1184/1480 train_time:185590ms step_avg:158.08ms
step:1185/1480 train_time:185763ms step_avg:158.10ms
step:1186/1480 train_time:185934ms step_avg:158.11ms
step:1187/1480 train_time:186115ms step_avg:158.13ms
step:1188/1480 train_time:186281ms step_avg:158.13ms
step:1189/1480 train_time:186454ms step_avg:158.15ms
step:1190/1480 train_time:186621ms step_avg:158.15ms
step:1191/1480 train_time:186794ms step_avg:158.17ms
step:1192/1480 train_time:186960ms step_avg:158.17ms
step:1193/1480 train_time:187126ms step_avg:158.18ms
step:1194/1480 train_time:187296ms step_avg:158.19ms
step:1195/1480 train_time:187470ms step_avg:158.20ms
step:1196/1480 train_time:187652ms step_avg:158.22ms
step:1197/1480 train_time:187824ms step_avg:158.23ms
step:1198/1480 train_time:188006ms step_avg:158.25ms
step:1199/1480 train_time:188176ms step_avg:158.26ms
step:1200/1480 train_time:188344ms step_avg:158.27ms
step:1201/1480 train_time:188511ms step_avg:158.28ms
step:1202/1480 train_time:188690ms step_avg:158.30ms
step:1203/1480 train_time:188865ms step_avg:158.31ms
step:1204/1480 train_time:189039ms step_avg:158.32ms
step:1205/1480 train_time:189206ms step_avg:158.33ms
step:1206/1480 train_time:189375ms step_avg:158.34ms
step:1207/1480 train_time:189544ms step_avg:158.35ms
step:1208/1480 train_time:189711ms step_avg:158.36ms
step:1209/1480 train_time:189884ms step_avg:158.37ms
step:1210/1480 train_time:190059ms step_avg:158.38ms
step:1211/1480 train_time:190234ms step_avg:158.40ms
step:1212/1480 train_time:190406ms step_avg:158.41ms
step:1213/1480 train_time:190580ms step_avg:158.42ms
step:1214/1480 train_time:190757ms step_avg:158.44ms
step:1215/1480 train_time:190931ms step_avg:158.45ms
step:1216/1480 train_time:191100ms step_avg:158.46ms
step:1217/1480 train_time:191275ms step_avg:158.47ms
step:1218/1480 train_time:191444ms step_avg:158.48ms
step:1219/1480 train_time:191623ms step_avg:158.50ms
step:1220/1480 train_time:191794ms step_avg:158.51ms
step:1221/1480 train_time:191963ms step_avg:158.52ms
step:1222/1480 train_time:192130ms step_avg:158.52ms
step:1223/1480 train_time:192299ms step_avg:158.53ms
step:1224/1480 train_time:192479ms step_avg:158.55ms
step:1225/1480 train_time:192651ms step_avg:158.56ms
step:1226/1480 train_time:192823ms step_avg:158.57ms
step:1227/1480 train_time:192995ms step_avg:158.58ms
step:1228/1480 train_time:193164ms step_avg:158.59ms
step:1229/1480 train_time:193337ms step_avg:158.60ms
step:1230/1480 train_time:193516ms step_avg:158.62ms
step:1231/1480 train_time:193692ms step_avg:158.63ms
step:1232/1480 train_time:193868ms step_avg:158.65ms
step:1233/1480 train_time:194038ms step_avg:158.66ms
step:1234/1480 train_time:194208ms step_avg:158.67ms
step:1235/1480 train_time:194381ms step_avg:158.68ms
step:1236/1480 train_time:194550ms step_avg:158.69ms
step:1237/1480 train_time:194720ms step_avg:158.70ms
step:1238/1480 train_time:194906ms step_avg:158.72ms
step:1239/1480 train_time:195078ms step_avg:158.73ms
step:1240/1480 train_time:195250ms step_avg:158.74ms
step:1241/1480 train_time:195422ms step_avg:158.75ms
step:1242/1480 train_time:195593ms step_avg:158.76ms
step:1243/1480 train_time:195767ms step_avg:158.77ms
step:1244/1480 train_time:195933ms step_avg:158.78ms
step:1245/1480 train_time:196101ms step_avg:158.79ms
step:1246/1480 train_time:196272ms step_avg:158.80ms
step:1247/1480 train_time:196440ms step_avg:158.80ms
step:1248/1480 train_time:196610ms step_avg:158.81ms
step:1249/1480 train_time:196778ms step_avg:158.82ms
step:1250/1480 train_time:196945ms step_avg:158.83ms
step:1250/1480 val_loss:3.3349 train_time:197017ms step_avg:158.88ms
step:1251/1480 train_time:197127ms step_avg:158.85ms
step:1252/1480 train_time:197296ms step_avg:158.85ms
step:1253/1480 train_time:197465ms step_avg:158.86ms
step:1254/1480 train_time:197636ms step_avg:158.87ms
step:1255/1480 train_time:197821ms step_avg:158.89ms
step:1256/1480 train_time:197994ms step_avg:158.90ms
step:1257/1480 train_time:198164ms step_avg:158.91ms
step:1258/1480 train_time:198338ms step_avg:158.92ms
step:1259/1480 train_time:198510ms step_avg:158.94ms
step:1260/1480 train_time:198677ms step_avg:158.94ms
step:1261/1480 train_time:198849ms step_avg:158.95ms
step:1262/1480 train_time:199026ms step_avg:158.97ms
step:1263/1480 train_time:199199ms step_avg:158.98ms
step:1264/1480 train_time:199368ms step_avg:158.99ms
step:1265/1480 train_time:199535ms step_avg:158.99ms
step:1266/1480 train_time:199707ms step_avg:159.00ms
step:1267/1480 train_time:199877ms step_avg:159.01ms
step:1268/1480 train_time:200050ms step_avg:159.02ms
step:1269/1480 train_time:200225ms step_avg:159.03ms
step:1270/1480 train_time:200395ms step_avg:159.04ms
step:1271/1480 train_time:200566ms step_avg:159.05ms
step:1272/1480 train_time:200732ms step_avg:159.06ms
step:1273/1480 train_time:200905ms step_avg:159.07ms
step:1274/1480 train_time:201076ms step_avg:159.08ms
step:1275/1480 train_time:201243ms step_avg:159.09ms
step:1276/1480 train_time:201410ms step_avg:159.09ms
step:1277/1480 train_time:201583ms step_avg:159.10ms
step:1278/1480 train_time:201752ms step_avg:159.11ms
step:1279/1480 train_time:201923ms step_avg:159.12ms
step:1280/1480 train_time:202104ms step_avg:159.14ms
step:1281/1480 train_time:202273ms step_avg:159.14ms
step:1282/1480 train_time:202438ms step_avg:159.15ms
step:1283/1480 train_time:202608ms step_avg:159.16ms
step:1284/1480 train_time:202778ms step_avg:159.17ms
step:1285/1480 train_time:202946ms step_avg:159.17ms
step:1286/1480 train_time:203115ms step_avg:159.18ms
step:1287/1480 train_time:203287ms step_avg:159.19ms
step:1288/1480 train_time:203458ms step_avg:159.20ms
step:1289/1480 train_time:203640ms step_avg:159.22ms
step:1290/1480 train_time:203821ms step_avg:159.24ms
step:1291/1480 train_time:203994ms step_avg:159.25ms
step:1292/1480 train_time:204168ms step_avg:159.26ms
step:1293/1480 train_time:204342ms step_avg:159.27ms
step:1294/1480 train_time:204513ms step_avg:159.28ms
step:1295/1480 train_time:204685ms step_avg:159.29ms
step:1296/1480 train_time:204860ms step_avg:159.30ms
step:1297/1480 train_time:205032ms step_avg:159.31ms
step:1298/1480 train_time:205203ms step_avg:159.32ms
step:1299/1480 train_time:205373ms step_avg:159.33ms
step:1300/1480 train_time:205540ms step_avg:159.33ms
step:1301/1480 train_time:205709ms step_avg:159.34ms
step:1302/1480 train_time:205881ms step_avg:159.35ms
step:1303/1480 train_time:206058ms step_avg:159.36ms
step:1304/1480 train_time:206232ms step_avg:159.38ms
step:1305/1480 train_time:206399ms step_avg:159.38ms
step:1306/1480 train_time:206573ms step_avg:159.39ms
step:1307/1480 train_time:206740ms step_avg:159.40ms
step:1308/1480 train_time:206909ms step_avg:159.41ms
step:1309/1480 train_time:207081ms step_avg:159.42ms
step:1310/1480 train_time:207251ms step_avg:159.42ms
step:1311/1480 train_time:207420ms step_avg:159.43ms
step:1312/1480 train_time:207593ms step_avg:159.44ms
step:1313/1480 train_time:207763ms step_avg:159.45ms
step:1314/1480 train_time:207935ms step_avg:159.46ms
step:1315/1480 train_time:208104ms step_avg:159.47ms
step:1316/1480 train_time:208272ms step_avg:159.47ms
step:1317/1480 train_time:208442ms step_avg:159.48ms
step:1318/1480 train_time:208622ms step_avg:159.50ms
step:1319/1480 train_time:208798ms step_avg:159.51ms
step:1320/1480 train_time:208975ms step_avg:159.52ms
step:1321/1480 train_time:209147ms step_avg:159.53ms
step:1322/1480 train_time:209326ms step_avg:159.55ms
step:1323/1480 train_time:209498ms step_avg:159.56ms
step:1324/1480 train_time:209674ms step_avg:159.57ms
step:1325/1480 train_time:209855ms step_avg:159.59ms
step:1326/1480 train_time:210031ms step_avg:159.60ms
step:1327/1480 train_time:210200ms step_avg:159.61ms
step:1328/1480 train_time:210372ms step_avg:159.61ms
step:1329/1480 train_time:210567ms step_avg:159.64ms
step:1330/1480 train_time:210747ms step_avg:159.66ms
step:1331/1480 train_time:210917ms step_avg:159.66ms
step:1332/1480 train_time:211092ms step_avg:159.68ms
step:1333/1480 train_time:211266ms step_avg:159.69ms
step:1334/1480 train_time:211437ms step_avg:159.70ms
step:1335/1480 train_time:211606ms step_avg:159.70ms
step:1336/1480 train_time:211791ms step_avg:159.72ms
step:1337/1480 train_time:211965ms step_avg:159.73ms
step:1338/1480 train_time:212136ms step_avg:159.74ms
step:1339/1480 train_time:212311ms step_avg:159.75ms
step:1340/1480 train_time:212483ms step_avg:159.76ms
step:1341/1480 train_time:212651ms step_avg:159.77ms
step:1342/1480 train_time:212824ms step_avg:159.78ms
step:1343/1480 train_time:212995ms step_avg:159.79ms
step:1344/1480 train_time:213166ms step_avg:159.79ms
step:1345/1480 train_time:213345ms step_avg:159.81ms
step:1346/1480 train_time:213514ms step_avg:159.82ms
step:1347/1480 train_time:213684ms step_avg:159.82ms
step:1348/1480 train_time:213854ms step_avg:159.83ms
step:1349/1480 train_time:214024ms step_avg:159.84ms
step:1350/1480 train_time:214197ms step_avg:159.85ms
step:1351/1480 train_time:214368ms step_avg:159.86ms
step:1352/1480 train_time:214539ms step_avg:159.87ms
step:1353/1480 train_time:214714ms step_avg:159.88ms
step:1354/1480 train_time:214886ms step_avg:159.89ms
step:1355/1480 train_time:215054ms step_avg:159.89ms
step:1356/1480 train_time:215228ms step_avg:159.90ms
step:1357/1480 train_time:215400ms step_avg:159.91ms
step:1358/1480 train_time:215572ms step_avg:159.92ms
step:1359/1480 train_time:215744ms step_avg:159.93ms
step:1360/1480 train_time:215918ms step_avg:159.94ms
step:1361/1480 train_time:216097ms step_avg:159.95ms
step:1362/1480 train_time:216273ms step_avg:159.97ms
step:1363/1480 train_time:216453ms step_avg:159.98ms
step:1364/1480 train_time:216623ms step_avg:159.99ms
step:1365/1480 train_time:216790ms step_avg:159.99ms
step:1366/1480 train_time:216960ms step_avg:160.00ms
step:1367/1480 train_time:217131ms step_avg:160.01ms
step:1368/1480 train_time:217305ms step_avg:160.02ms
step:1369/1480 train_time:217487ms step_avg:160.03ms
step:1370/1480 train_time:217665ms step_avg:160.05ms
step:1371/1480 train_time:217836ms step_avg:160.06ms
step:1372/1480 train_time:218013ms step_avg:160.07ms
step:1373/1480 train_time:218182ms step_avg:160.07ms
step:1374/1480 train_time:218358ms step_avg:160.09ms
step:1375/1480 train_time:218529ms step_avg:160.09ms
step:1375/1480 val_loss:3.2966 train_time:218596ms step_avg:160.14ms
step:1376/1480 train_time:218704ms step_avg:160.11ms
step:1377/1480 train_time:218876ms step_avg:160.11ms
step:1378/1480 train_time:219046ms step_avg:160.12ms
step:1379/1480 train_time:219221ms step_avg:160.13ms
step:1380/1480 train_time:219394ms step_avg:160.14ms
step:1381/1480 train_time:219575ms step_avg:160.16ms
step:1382/1480 train_time:219746ms step_avg:160.16ms
step:1383/1480 train_time:219916ms step_avg:160.17ms
step:1384/1480 train_time:220096ms step_avg:160.19ms
step:1385/1480 train_time:220262ms step_avg:160.19ms
step:1386/1480 train_time:220432ms step_avg:160.20ms
step:1387/1480 train_time:220604ms step_avg:160.21ms
step:1388/1480 train_time:220772ms step_avg:160.21ms
step:1389/1480 train_time:220948ms step_avg:160.22ms
step:1390/1480 train_time:221115ms step_avg:160.23ms
step:1391/1480 train_time:221288ms step_avg:160.24ms
step:1392/1480 train_time:221460ms step_avg:160.25ms
step:1393/1480 train_time:221631ms step_avg:160.25ms
step:1394/1480 train_time:221801ms step_avg:160.26ms
step:1395/1480 train_time:221970ms step_avg:160.27ms
step:1396/1480 train_time:222137ms step_avg:160.27ms
step:1397/1480 train_time:222305ms step_avg:160.28ms
step:1398/1480 train_time:222472ms step_avg:160.28ms
step:1399/1480 train_time:222640ms step_avg:160.29ms
step:1400/1480 train_time:222816ms step_avg:160.30ms
step:1401/1480 train_time:222983ms step_avg:160.30ms
step:1402/1480 train_time:223153ms step_avg:160.31ms
step:1403/1480 train_time:223329ms step_avg:160.32ms
step:1404/1480 train_time:223499ms step_avg:160.33ms
step:1405/1480 train_time:223672ms step_avg:160.34ms
step:1406/1480 train_time:223847ms step_avg:160.35ms
step:1407/1480 train_time:224015ms step_avg:160.35ms
step:1408/1480 train_time:224182ms step_avg:160.36ms
step:1409/1480 train_time:224368ms step_avg:160.38ms
step:1410/1480 train_time:224537ms step_avg:160.38ms
step:1411/1480 train_time:224706ms step_avg:160.39ms
step:1412/1480 train_time:224873ms step_avg:160.39ms
step:1413/1480 train_time:225043ms step_avg:160.40ms
step:1414/1480 train_time:225213ms step_avg:160.41ms
step:1415/1480 train_time:225389ms step_avg:160.42ms
step:1416/1480 train_time:225576ms step_avg:160.44ms
step:1417/1480 train_time:225750ms step_avg:160.45ms
step:1418/1480 train_time:225923ms step_avg:160.46ms
step:1419/1480 train_time:226096ms step_avg:160.47ms
step:1420/1480 train_time:226270ms step_avg:160.48ms
step:1421/1480 train_time:226443ms step_avg:160.48ms
step:1422/1480 train_time:226613ms step_avg:160.49ms
step:1423/1480 train_time:226783ms step_avg:160.50ms
step:1424/1480 train_time:226959ms step_avg:160.51ms
step:1425/1480 train_time:227138ms step_avg:160.52ms
step:1426/1480 train_time:227309ms step_avg:160.53ms
step:1427/1480 train_time:227484ms step_avg:160.54ms
step:1428/1480 train_time:227656ms step_avg:160.55ms
step:1429/1480 train_time:227825ms step_avg:160.55ms
step:1430/1480 train_time:227998ms step_avg:160.56ms
step:1431/1480 train_time:228174ms step_avg:160.57ms
step:1432/1480 train_time:228352ms step_avg:160.58ms
step:1433/1480 train_time:228531ms step_avg:160.60ms
step:1434/1480 train_time:228712ms step_avg:160.61ms
step:1435/1480 train_time:228888ms step_avg:160.62ms
step:1436/1480 train_time:229061ms step_avg:160.63ms
step:1437/1480 train_time:229231ms step_avg:160.64ms
step:1438/1480 train_time:229401ms step_avg:160.64ms
step:1439/1480 train_time:229574ms step_avg:160.65ms
step:1440/1480 train_time:229744ms step_avg:160.66ms
step:1441/1480 train_time:229914ms step_avg:160.67ms
step:1442/1480 train_time:230092ms step_avg:160.68ms
step:1443/1480 train_time:230279ms step_avg:160.70ms
step:1444/1480 train_time:230450ms step_avg:160.70ms
step:1445/1480 train_time:230619ms step_avg:160.71ms
step:1446/1480 train_time:230794ms step_avg:160.72ms
step:1447/1480 train_time:230972ms step_avg:160.73ms
step:1448/1480 train_time:231143ms step_avg:160.74ms
step:1449/1480 train_time:231316ms step_avg:160.75ms
step:1450/1480 train_time:231489ms step_avg:160.76ms
step:1451/1480 train_time:231659ms step_avg:160.76ms
step:1452/1480 train_time:231832ms step_avg:160.77ms
step:1453/1480 train_time:232005ms step_avg:160.78ms
step:1454/1480 train_time:232177ms step_avg:160.79ms
step:1455/1480 train_time:232356ms step_avg:160.80ms
step:1456/1480 train_time:232529ms step_avg:160.81ms
step:1457/1480 train_time:232701ms step_avg:160.82ms
step:1458/1480 train_time:232871ms step_avg:160.82ms
step:1459/1480 train_time:233047ms step_avg:160.83ms
step:1460/1480 train_time:233219ms step_avg:160.84ms
step:1461/1480 train_time:233394ms step_avg:160.85ms
step:1462/1480 train_time:233567ms step_avg:160.86ms
step:1463/1480 train_time:233742ms step_avg:160.87ms
step:1464/1480 train_time:233915ms step_avg:160.88ms
step:1465/1480 train_time:234089ms step_avg:160.89ms
step:1466/1480 train_time:234258ms step_avg:160.89ms
step:1467/1480 train_time:234435ms step_avg:160.90ms
step:1468/1480 train_time:234606ms step_avg:160.91ms
step:1469/1480 train_time:234779ms step_avg:160.92ms
step:1470/1480 train_time:234957ms step_avg:160.93ms
step:1471/1480 train_time:235144ms step_avg:160.95ms
step:1472/1480 train_time:235324ms step_avg:160.96ms
step:1473/1480 train_time:235495ms step_avg:160.97ms
step:1474/1480 train_time:235673ms step_avg:160.98ms
step:1475/1480 train_time:235854ms step_avg:160.99ms
step:1476/1480 train_time:236026ms step_avg:161.00ms
step:1477/1480 train_time:236210ms step_avg:161.02ms
step:1478/1480 train_time:236391ms step_avg:161.03ms
step:1479/1480 train_time:236564ms step_avg:161.04ms
step:1480/1480 train_time:236738ms step_avg:161.05ms
step:1480/1480 val_loss:3.2779 train_time:236811ms step_avg:161.10ms