diff --git a/pytorch/.gitignore b/pytorch/.gitignore new file mode 100644 index 00000000..fbca2253 --- /dev/null +++ b/pytorch/.gitignore @@ -0,0 +1 @@ +results/ diff --git a/torch/__init__.py b/pytorch/__init__.py similarity index 100% rename from torch/__init__.py rename to pytorch/__init__.py diff --git a/torch/main.py b/pytorch/main.py similarity index 92% rename from torch/main.py rename to pytorch/main.py index fc46d6b6..d864a82a 100644 --- a/torch/main.py +++ b/pytorch/main.py @@ -14,7 +14,8 @@ import torch.multiprocessing as mp import torch.utils.data import torch.utils.data.distributed -from model import cnnscore +#from torch.utils.tensorboard import SummaryWriter +from model import cnnscore, vggnet from dataset import CustomDataset parser = argparse.ArgumentParser(description='PyTorch DLSCORE-CNN Training') @@ -63,7 +64,7 @@ 'multi node data parallel training') best_loss = np.inf - +#writer = SummaryWriter('runs/') def main(): args = parser.parse_args() @@ -94,7 +95,8 @@ def main(): args.world_size = ngpus_per_node * args.world_size # Use torch.multiprocessing.spawn to launch distributed processes: the # main_worker process function - mp.spawn(main_worker, nprocs=ngpus_per_node, args=(ngpus_per_node, args)) + mp.spawn(main_worker, nprocs=ngpus_per_node, + args=(ngpus_per_node, args)) else: # Simply call main_worker function main_worker(args.gpu, ngpus_per_node, args) @@ -118,7 +120,10 @@ def main_worker(gpu, ngpus_per_node, args): world_size=args.world_size, rank=args.rank) # create model print("=> creating model '{}'".format('cnnscore')) - model = cnnscore() + #model = cnnscore() + model = vggnet() + print(model) + if args.distributed: # For multiprocessing distributed, DistributedDataParallel constructor @@ -131,8 +136,10 @@ def main_worker(gpu, ngpus_per_node, args): # DistributedDataParallel, we need to divide the batch size # ourselves based on the total number of GPUs we have args.batch_size = int(args.batch_size / ngpus_per_node) - args.workers = int((args.workers + ngpus_per_node - 1) / ngpus_per_node) - model = torch.nn.parallel.DistributedDataParallel(model, device_ids=[args.gpu]) + args.workers = int( + (args.workers + ngpus_per_node - 1) / ngpus_per_node) + model = torch.nn.parallel.DistributedDataParallel( + model, device_ids=[args.gpu]) else: model.cuda() # DistributedDataParallel will divide and allocate batch_size to all @@ -148,7 +155,8 @@ def main_worker(gpu, ngpus_per_node, args): # define loss function (criterion) and optimizer criterion = nn.MSELoss().cuda(args.gpu) - optimizer = torch.optim.Adam(model.parameters(), args.lr, betas=(0.9, 0.999)) + optimizer = torch.optim.Adam( + model.parameters(), args.lr, betas=(0.9, 0.999)) # optionally resume from a checkpoint if args.resume: @@ -163,6 +171,7 @@ def main_worker(gpu, ngpus_per_node, args): args.start_epoch = checkpoint['epoch'] best_loss = checkpoint['best_loss'] if args.gpu is not None: + print(args.gpu) # best_loss1 may be from a checkpoint from a different GPU best_loss = best_loss.to(args.gpu) model.load_state_dict(checkpoint['state_dict']) @@ -181,14 +190,15 @@ def main_worker(gpu, ngpus_per_node, args): train_dataset = CustomDataset(train_dir) test_dataset = CustomDataset(val_dir) - if args.distributed: - train_sampler = torch.utils.data.distributed.DistributedSampler(train_dataset) + train_sampler = torch.utils.data.distributed.DistributedSampler( + train_dataset) else: train_sampler = None train_loader = torch.utils.data.DataLoader( - train_dataset, batch_size=args.batch_size, shuffle=(train_sampler is None), + train_dataset, batch_size=args.batch_size, shuffle=( + train_sampler is None), num_workers=args.workers, pin_memory=True, sampler=train_sampler) val_loader = torch.utils.data.DataLoader( @@ -216,7 +226,7 @@ def main_worker(gpu, ngpus_per_node, args): best_loss = min(loss, best_loss) if not args.multiprocessing_distributed or (args.multiprocessing_distributed - and args.rank % ngpus_per_node == 0): + and args.rank % ngpus_per_node == 0): save_checkpoint({ 'epoch': epoch + 1, 'state_dict': model.state_dict(), @@ -247,7 +257,7 @@ def train(train_loader, model, criterion, optimizer, epoch, args): target = target.cuda(args.gpu, non_blocking=True) # compute output - output = model(input) + output, _ = model(input) loss = criterion(output, target) # measure and record loss @@ -264,7 +274,7 @@ def train(train_loader, model, criterion, optimizer, epoch, args): if i % args.print_freq == 0: progress.display(i) - + def pearsonr(x, y): """ @@ -311,9 +321,11 @@ def validate(val_loader, model, criterion, args): target = target.cuda(args.gpu, non_blocking=True) # compute output - output = model(input) + output, last_layer_features = model(input) + #print(last_layer_features.shape) + np.save('l_features_' + str(i), last_layer_features) loss = criterion(output, target) - + # measure pearsonr and record loss actual_values[i] = target.mean() predicted_values[i] = output.mean() @@ -329,7 +341,8 @@ def validate(val_loader, model, criterion, args): pr = pearsonr(actual_values, predicted_values).item() mseloss = criterion(predicted_values, actual_values).item() - print('Test: [{0}/{0}]\t Pearson R: {1:.4f}\t MSE loss: {2:.4f}'.format(len(val_loader), pr, mseloss)) + print('Test: [{0}/{0}]\t Pearson R: {1:.4f}\t MSE loss: {2:.4f}'.format( + len(val_loader), pr, mseloss)) return mseloss @@ -342,6 +355,7 @@ def save_checkpoint(state, is_best, filename='checkpoint.pth.tar'): class AverageMeter(object): """Computes and stores the average and current value""" + def __init__(self, name, fmt=':f'): self.name = name self.fmt = fmt diff --git a/torch/model.py b/pytorch/model.py similarity index 55% rename from torch/model.py rename to pytorch/model.py index 5dea5d3a..4ced1865 100644 --- a/torch/model.py +++ b/pytorch/model.py @@ -2,7 +2,54 @@ import torch.nn as nn import torch.nn.functional as F -__all__ = ['cnnscore'] +__all__ = ['cnnscore', 'vggnet'] + +# Example VGGNet https://github.com/pytorch/vision/blob/master/torchvision/models/vgg.py + + +class VGGNet(nn.Module): + def __init__(self, features, init_weights=False): + super(VGGNet, self).__init__() + + self.features = features + # TODO: Following + #self.avgpool = nn.AdaptiveAvgPool2d((7, 7)) + self.lastblock = nn.Sequential( + nn.Linear(512, 4096), + nn.ReLU(True), + # nn.Dropout(), + nn.Linear(4096, 4096), + nn.ReLU(True) + #nn.Linear(4096, 1) + ) + self.dense = nn.Linear(4096, 1) + if init_weights: + self._initialize_weights() + + # TODO: Use the following initialization + def _initialize_weights(self): + for m in self.modules(): + if isinstance(m, nn.Conv3d): + nn.init.kaiming_normal_( + m.weight, mode='fan_out', nonlinearity='relu') + if m.bias is not None: + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.BatchNorm3d): + nn.init.constant_(m.weight, 1) + nn.init.constant_(m.bias, 0) + elif isinstance(m, nn.Linear): + nn.init.normal_(m.weight, 0, 0.01) + nn.init.constant_(m.bias, 0) + + def forward(self, x): + x = self.features(x) + x = torch.flatten(x, 1) + x = self.lastblock(x) + last_layer_features = x.cpu().detach().numpy() + x = self.dense(x) + + return x, last_layer_features + class CNNScore(nn.Module): def __init__(self): @@ -10,85 +57,120 @@ def __init__(self): self.conv1 = nn.Conv3d(16, 96, kernel_size=1, stride=2) self.fire2_squeeze = nn.Conv3d(96, 16, kernel_size=1) self.fire2_expand1 = nn.Conv3d(16, 64, kernel_size=1) - self.fire2_expand2 = nn.Conv3d(16, 64, kernel_size=3, padding=1) # Padding = (k-1)/2 where k is the kernel size - + # Padding = (k-1)/2 where k is the kernel size + self.fire2_expand2 = nn.Conv3d(16, 64, kernel_size=3, padding=1) + self.fire3_squeeze = nn.Conv3d(128, 16, kernel_size=1) self.fire3_expand1 = nn.Conv3d(16, 64, kernel_size=1) self.fire3_expand2 = nn.Conv3d(16, 64, kernel_size=3, padding=1) - + self.fire4_squeeze = nn.Conv3d(128, 32, kernel_size=1) self.fire4_expand1 = nn.Conv3d(32, 128, kernel_size=1) self.fire4_expand2 = nn.Conv3d(32, 128, kernel_size=3, padding=1) - + self.pool = nn.MaxPool3d(kernel_size=3, stride=2) self.fire5_squeeze = nn.Conv3d(256, 32, kernel_size=1) self.fire5_expand1 = nn.Conv3d(32, 128, kernel_size=1) self.fire5_expand2 = nn.Conv3d(32, 128, kernel_size=3, padding=1) - + self.fire6_squeeze = nn.Conv3d(256, 48, kernel_size=1) self.fire6_expand1 = nn.Conv3d(48, 192, kernel_size=1) self.fire6_expand2 = nn.Conv3d(48, 192, kernel_size=3, padding=1) - + self.fire7_squeeze = nn.Conv3d(384, 48, kernel_size=1) self.fire7_expand1 = nn.Conv3d(48, 192, kernel_size=1) self.fire7_expand2 = nn.Conv3d(48, 192, kernel_size=3, padding=1) - + self.fire8_squeeze = nn.Conv3d(384, 64, kernel_size=1) self.fire8_expand1 = nn.Conv3d(64, 256, kernel_size=1) self.fire8_expand2 = nn.Conv3d(64, 256, kernel_size=3, padding=1) - + self.avg_pool = nn.AvgPool3d(kernel_size=3, padding=1) - + self.dense1 = nn.Linear(512*2*2*2, 1) - - + def forward(self, x): x = F.relu(self.conv1(x)) x = F.relu(self.fire2_squeeze(x)) expand1 = F.relu(self.fire2_expand1(x)) expand2 = F.relu(self.fire2_expand2(x)) merge1 = torch.cat((expand1, expand2), 1) - + x = F.relu(self.fire3_squeeze(merge1)) expand1 = F.relu(self.fire3_expand1(x)) expand2 = F.relu(self.fire3_expand2(x)) merge2 = torch.cat((expand1, expand2), 1) - + x = F.relu(self.fire4_squeeze(merge2)) expand1 = F.relu(self.fire4_expand1(x)) expand2 = F.relu(self.fire4_expand2(x)) merge3 = torch.cat((expand1, expand2), 1) pool1 = self.pool(merge3) - + x = F.relu(self.fire5_squeeze(pool1)) expand1 = F.relu(self.fire5_expand1(x)) expand2 = F.relu(self.fire5_expand2(x)) merge4 = torch.cat((expand1, expand2), 1) - + x = F.relu(self.fire6_squeeze(merge4)) expand1 = F.relu(self.fire6_expand1(x)) expand2 = F.relu(self.fire6_expand2(x)) merge5 = torch.cat((expand1, expand2), 1) - + x = F.relu(self.fire7_squeeze(merge5)) expand1 = F.relu(self.fire7_expand1(x)) expand2 = F.relu(self.fire7_expand2(x)) merge6 = torch.cat((expand1, expand2), 1) - + x = F.relu(self.fire8_squeeze(merge6)) expand1 = F.relu(self.fire8_expand1(x)) expand2 = F.relu(self.fire8_expand2(x)) merge7 = torch.cat((expand1, expand2), 1) - + pool2 = self.avg_pool(merge7) x = pool2.view(-1, 512*2*2*2) x = self.dense1(x) #x = x.view(-1) - + return x def cnnscore(**kwargs): model = CNNScore(**kwargs) - return model \ No newline at end of file + return model + + +cfgs = { + 'A': [64, 64, 64, 'M', 128, 128, 128, 'M', 256, 256, 256, 256, 'M', 512, 512, 512, 512, 'M'] +} + + +def make_layers(cfg, batch_norm=True): + layers = [] + in_channels = 16 + + for v in cfg: + if v == 'M': + layers += [nn.MaxPool3d(kernel_size=2, stride=2)] + else: + conv = nn.Conv3d(in_channels, v, kernel_size=3, padding=1) # padding = (k-1)//2 + if batch_norm: + layers += [conv, nn.BatchNorm3d(v), nn.ReLU(inplace=True)] + else: + layers += [conv, nn.ReLU(inplace=True)] + in_channels = v + + return nn.Sequential(*layers) + + +def vggnet(**kwargs): + + # TODO: The last maxpool layer is (512, 3, 3) to (512, 1, 1). Fix it + # TODO: The last dense layers are 4096 to 1. There are two such layers. + # TODO: Use dropout to regularize. Looks like training loss is very less compared to val loss + + model = VGGNet(make_layers( + cfgs['A'], batch_norm=True), init_weights=True, **kwargs) + #model = VGGNet(**kwargs) + return model diff --git a/pytorch/model_test.py b/pytorch/model_test.py new file mode 100644 index 00000000..9ece6a45 --- /dev/null +++ b/pytorch/model_test.py @@ -0,0 +1,7 @@ +import torch +from model import vggnet +from torchsummary import summary + +device = torch.device("cuda" if torch.cuda.is_available() else "cpu") +model = vggnet().to(device) +summary(model, input_size=(16, 24, 24, 24)) \ No newline at end of file diff --git a/torch/test.py b/pytorch/test.py similarity index 100% rename from torch/test.py rename to pytorch/test.py diff --git a/pytorch/test.sh b/pytorch/test.sh new file mode 100755 index 00000000..d59968df --- /dev/null +++ b/pytorch/test.sh @@ -0,0 +1 @@ +python main.py ../dataset --multiprocessing-distributed -j 8 --resume model_best.pth.tar --evaluate diff --git a/pytorch/train.sh b/pytorch/train.sh new file mode 100755 index 00000000..684980ce --- /dev/null +++ b/pytorch/train.sh @@ -0,0 +1,2 @@ +#python main.py ../dataset --dist-url 'tcp://chanti00.utep.edu:12356' --world-size 2 --multiprocessing-distributed -j 8 --batch-size 128 --epochs 2 +python main.py ../dataset --multiprocessing-distributed -j 8 --batch-size 128 --epochs 100 diff --git a/torch/dataset.py b/torch/dataset.py deleted file mode 100644 index 5c942d5b..00000000 --- a/torch/dataset.py +++ /dev/null @@ -1,28 +0,0 @@ -import torch -from torch.utils.data import DataLoader, Dataset -import json -import os -import numpy as np - -class CustomDataset(Dataset): - - def __init__(self, data_dir): - self.data_dir = data_dir - self.pdbmap = json.load(open(os.path.join(data_dir, 'pdbmap.json'))) - self.labels = json.load(open(os.path.join(data_dir, 'labels.json'))) - - def __getitem__(self, idx): - if torch.is_tensor(idx): - idx = idx.tolist() - - samples = torch.from_numpy(np.load(os.path.join(self.data_dir, 'id_'+str(idx)+'.npy')).astype(np.float32))#[np.newaxis, :] - # TODO: Check if the following permuation preserves the data - samples = samples.permute(3, 0, 1, 2) # Get the channels at the beginning - targets = torch.from_numpy(np.array(self.labels['id_'+str(idx)]).astype(np.float32)) - targets = targets.view(-1) - - return samples, targets - - def __len__(self): - return len(self.labels) - diff --git a/torch/log.txt b/torch/log.txt deleted file mode 100644 index 59622ec9..00000000 --- a/torch/log.txt +++ /dev/null @@ -1,1872 +0,0 @@ -Use GPU: 5 for training -=> creating model 'cnnscore' -Use GPU: 4 for training -=> creating model 'cnnscore' -Use GPU: 2 for training -=> creating model 'cnnscore' -Use GPU: 6 for training -=> creating model 'cnnscore' -Use GPU: 0 for training -=> creating model 'cnnscore' -Use GPU: 7 for training -=> creating model 'cnnscore' -Use GPU: 3 for training -=> creating model 'cnnscore' -Use GPU: 1 for training -=> creating model 'cnnscore' -Epoch: [0][0/706] Time 9.273 (9.273) Data 8.296 (8.296) Loss 43.7812 (43.7812) -Epoch: [0][10/706] Time 0.182 (0.993) Data 0.024 (0.757) Loss 36.6127 (41.0027) -Epoch: [0][20/706] Time 0.149 (0.604) Data 0.000 (0.397) Loss 46.7483 (42.5662) -Epoch: [0][30/706] Time 0.190 (0.456) Data 0.000 (0.269) Loss 23.6645 (40.5746) -Epoch: [0][40/706] Time 0.253 (0.386) Data 0.000 (0.206) Loss 3.3825 (35.1845) -Epoch: [0][50/706] Time 0.139 (0.341) Data 0.000 (0.167) Loss 4.0140 (29.6901) -Epoch: [0][60/706] Time 0.038 (0.309) Data 0.000 (0.140) Loss 5.6525 (25.4991) -Epoch: [0][70/706] Time 0.155 (0.288) Data 0.024 (0.121) Loss 4.5481 (22.4706) -Epoch: [0][80/706] Time 0.181 (0.273) Data 0.000 (0.106) Loss 5.1761 (20.1880) -Epoch: [0][90/706] Time 0.147 (0.263) Data 0.000 (0.096) Loss 5.1384 (18.4905) -Epoch: [0][100/706] Time 0.152 (0.251) Data 0.002 (0.087) Loss 4.5511 (17.0494) -Epoch: [0][110/706] Time 0.193 (0.247) Data 0.005 (0.080) Loss 4.4273 (15.8497) -Epoch: [0][120/706] Time 0.175 (0.240) Data 0.000 (0.073) Loss 4.1081 (14.8156) -Epoch: [0][130/706] Time 0.166 (0.236) Data 0.013 (0.068) Loss 3.1548 (13.9584) -Epoch: [0][140/706] Time 0.267 (0.231) Data 0.000 (0.063) Loss 4.1572 (13.3199) -Epoch: [0][150/706] Time 0.196 (0.227) Data 0.000 (0.059) Loss 2.3899 (12.6904) -Epoch: [0][160/706] Time 0.107 (0.223) Data 0.000 (0.056) Loss 2.1265 (12.1773) -Epoch: [0][170/706] Time 0.150 (0.219) Data 0.043 (0.054) Loss 3.4673 (11.6809) -Epoch: [0][180/706] Time 0.162 (0.216) Data 0.000 (0.051) Loss 3.0005 (11.2717) -Epoch: [0][190/706] Time 0.182 (0.213) Data 0.000 (0.049) Loss 4.5859 (10.8871) -Epoch: [0][200/706] Time 0.126 (0.210) Data 0.013 (0.047) Loss 3.0499 (10.5359) -Epoch: [0][210/706] Time 0.147 (0.209) Data 0.000 (0.045) Loss 3.4410 (10.2281) -Epoch: [0][220/706] Time 0.150 (0.206) Data 0.009 (0.043) Loss 4.7609 (9.9595) -Epoch: [0][230/706] Time 0.099 (0.205) Data 0.000 (0.042) Loss 6.3430 (9.6948) -Epoch: [0][240/706] Time 0.182 (0.204) Data 0.106 (0.041) Loss 4.1766 (9.4667) -Epoch: [0][250/706] Time 0.186 (0.203) Data 0.000 (0.039) Loss 3.1245 (9.2372) -Epoch: [0][260/706] Time 0.190 (0.202) Data 0.035 (0.038) Loss 2.5587 (9.0283) -Epoch: [0][270/706] Time 0.165 (0.200) Data 0.000 (0.037) Loss 4.0169 (8.8330) -Epoch: [0][280/706] Time 0.191 (0.200) Data 0.000 (0.036) Loss 3.6347 (8.6631) -Epoch: [0][290/706] Time 0.150 (0.199) Data 0.000 (0.035) Loss 5.8254 (8.5198) -Epoch: [0][300/706] Time 0.220 (0.198) Data 0.000 (0.033) Loss 4.8305 (8.3710) -Epoch: [0][310/706] Time 0.272 (0.197) Data 0.002 (0.032) Loss 1.9660 (8.2156) -Epoch: [0][320/706] Time 0.147 (0.197) Data 0.000 (0.032) Loss 2.7627 (8.0928) -Epoch: [0][330/706] Time 0.117 (0.195) Data 0.000 (0.031) Loss 2.8903 (7.9359) -Epoch: [0][340/706] Time 0.079 (0.195) Data 0.000 (0.030) Loss 3.7515 (7.7976) -Epoch: [0][350/706] Time 0.226 (0.194) Data 0.000 (0.029) Loss 3.2747 (7.6631) -Epoch: [0][360/706] Time 0.298 (0.193) Data 0.000 (0.028) Loss 3.6034 (7.5383) -Epoch: [0][370/706] Time 0.195 (0.192) Data 0.061 (0.028) Loss 3.3247 (7.4077) -Epoch: [0][380/706] Time 0.096 (0.192) Data 0.000 (0.027) Loss 4.3023 (7.2990) -Epoch: [0][390/706] Time 0.134 (0.191) Data 0.000 (0.027) Loss 1.6932 (7.1717) -Epoch: [0][400/706] Time 0.293 (0.190) Data 0.002 (0.026) Loss 3.4855 (7.0601) -Epoch: [0][410/706] Time 0.165 (0.189) Data 0.000 (0.026) Loss 3.1472 (6.9529) -Epoch: [0][420/706] Time 0.237 (0.189) Data 0.004 (0.025) Loss 1.5653 (6.8612) -Epoch: [0][430/706] Time 0.165 (0.188) Data 0.026 (0.025) Loss 2.9950 (6.7592) -Epoch: [0][440/706] Time 0.157 (0.187) Data 0.000 (0.024) Loss 2.9650 (6.6754) -Epoch: [0][450/706] Time 0.085 (0.187) Data 0.022 (0.024) Loss 3.5954 (6.5926) -Epoch: [0][460/706] Time 0.036 (0.186) Data 0.000 (0.023) Loss 2.4112 (6.5126) -Epoch: [0][470/706] Time 0.111 (0.186) Data 0.000 (0.023) Loss 3.6040 (6.4353) -Epoch: [0][480/706] Time 0.149 (0.186) Data 0.001 (0.022) Loss 1.8813 (6.3562) -Epoch: [0][490/706] Time 0.221 (0.186) Data 0.010 (0.022) Loss 4.6093 (6.2903) -Epoch: [0][500/706] Time 0.134 (0.185) Data 0.000 (0.022) Loss 1.4926 (6.2248) -Epoch: [0][510/706] Time 0.255 (0.185) Data 0.005 (0.022) Loss 1.9583 (6.1587) -Epoch: [0][520/706] Time 0.190 (0.185) Data 0.039 (0.022) Loss 2.1274 (6.0921) -Epoch: [0][530/706] Time 0.122 (0.184) Data 0.000 (0.021) Loss 1.8860 (6.0271) -Epoch: [0][540/706] Time 0.122 (0.184) Data 0.000 (0.021) Loss 1.7688 (5.9639) -Epoch: [0][550/706] Time 0.143 (0.183) Data 0.000 (0.021) Loss 6.0982 (5.9113) -Epoch: [0][560/706] Time 0.057 (0.183) Data 0.000 (0.020) Loss 2.6791 (5.8684) -Epoch: [0][570/706] Time 0.180 (0.182) Data 0.000 (0.020) Loss 2.0202 (5.8205) -Epoch: [0][580/706] Time 0.141 (0.182) Data 0.000 (0.020) Loss 2.7016 (5.7693) -Epoch: [0][590/706] Time 0.210 (0.181) Data 0.000 (0.020) Loss 1.5937 (5.7044) -Epoch: [0][600/706] Time 0.237 (0.181) Data 0.014 (0.019) Loss 1.6962 (5.6546) -Epoch: [0][610/706] Time 0.168 (0.181) Data 0.000 (0.019) Loss 3.4061 (5.6036) -Epoch: [0][620/706] Time 0.170 (0.181) Data 0.000 (0.019) Loss 2.9352 (5.5639) -Epoch: [0][630/706] Time 0.139 (0.180) Data 0.000 (0.019) Loss 4.2232 (5.5267) -Epoch: [0][640/706] Time 0.196 (0.180) Data 0.000 (0.018) Loss 6.2598 (5.4956) -Epoch: [0][650/706] Time 0.112 (0.179) Data 0.000 (0.018) Loss 1.7745 (5.4476) -Epoch: [0][660/706] Time 0.195 (0.179) Data 0.000 (0.018) Loss 1.6714 (5.4090) -Epoch: [0][670/706] Time 0.105 (0.179) Data 0.000 (0.018) Loss 5.1122 (5.3751) -Epoch: [0][680/706] Time 0.160 (0.178) Data 0.000 (0.017) Loss 2.9198 (5.3436) -Epoch: [0][690/706] Time 0.160 (0.178) Data 0.000 (0.017) Loss 1.5070 (5.3062) -Epoch: [0][700/706] Time 0.025 (0.176) Data 0.000 (0.017) Loss 2.7606 (5.2740) -inside validate -Epoch: [0][0/706] Time 9.306 (9.306) Data 7.457 (7.457) Loss 43.9778 (43.9778) -Epoch: [0][10/706] Time 0.112 (0.990) Data 0.000 (0.684) Loss 32.0457 (44.9971) -Epoch: [0][20/706] Time 0.224 (0.597) Data 0.000 (0.361) Loss 46.2877 (43.4616) -Epoch: [0][30/706] Time 0.083 (0.455) Data 0.000 (0.246) Loss 32.1011 (40.5160) -Epoch: [0][40/706] Time 0.290 (0.386) Data 0.035 (0.188) Loss 2.9689 (34.5271) -Epoch: [0][50/706] Time 0.222 (0.341) Data 0.026 (0.153) Loss 3.9701 (28.8828) -Epoch: [0][60/706] Time 0.157 (0.313) Data 0.000 (0.130) Loss 5.1752 (24.9473) -Epoch: [0][70/706] Time 0.132 (0.290) Data 0.005 (0.112) Loss 4.3321 (22.0870) -Epoch: [0][80/706] Time 0.060 (0.274) Data 0.000 (0.100) Loss 5.6243 (19.9272) -Epoch: [0][90/706] Time 0.166 (0.262) Data 0.008 (0.089) Loss 3.2691 (18.1939) -Epoch: [0][100/706] Time 0.216 (0.252) Data 0.000 (0.081) Loss 3.9228 (16.8295) -Epoch: [0][110/706] Time 0.128 (0.245) Data 0.000 (0.074) Loss 4.8470 (15.6568) -Epoch: [0][120/706] Time 0.102 (0.241) Data 0.000 (0.069) Loss 4.5921 (14.7488) -Epoch: [0][130/706] Time 0.189 (0.237) Data 0.000 (0.065) Loss 2.9446 (13.9030) -Epoch: [0][140/706] Time 0.344 (0.232) Data 0.000 (0.060) Loss 2.9774 (13.1927) -Epoch: [0][150/706] Time 0.322 (0.227) Data 0.017 (0.056) Loss 5.4703 (12.5638) -Epoch: [0][160/706] Time 0.174 (0.223) Data 0.014 (0.053) Loss 4.3027 (12.0206) -Epoch: [0][170/706] Time 0.144 (0.218) Data 0.000 (0.050) Loss 6.1502 (11.5494) -Epoch: [0][180/706] Time 0.133 (0.216) Data 0.020 (0.048) Loss 4.7193 (11.0810) -Epoch: [0][190/706] Time 0.141 (0.212) Data 0.000 (0.045) Loss 4.1568 (10.7050) -Epoch: [0][200/706] Time 0.144 (0.210) Data 0.002 (0.043) Loss 6.2228 (10.3571) -Epoch: [0][210/706] Time 0.103 (0.208) Data 0.015 (0.042) Loss 4.2638 (10.0363) -Epoch: [0][220/706] Time 0.053 (0.206) Data 0.000 (0.040) Loss 5.5579 (9.7489) -Epoch: [0][230/706] Time 0.288 (0.205) Data 0.001 (0.038) Loss 3.7308 (9.5253) -Epoch: [0][240/706] Time 0.117 (0.203) Data 0.000 (0.037) Loss 2.9624 (9.2970) -Epoch: [0][250/706] Time 0.189 (0.203) Data 0.029 (0.036) Loss 5.9777 (9.1140) -Epoch: [0][260/706] Time 0.142 (0.201) Data 0.008 (0.035) Loss 2.3561 (8.9474) -Epoch: [0][270/706] Time 0.161 (0.200) Data 0.000 (0.034) Loss 2.1498 (8.7602) -Epoch: [0][280/706] Time 0.162 (0.200) Data 0.002 (0.033) Loss 5.1034 (8.5955) -Epoch: [0][290/706] Time 0.116 (0.199) Data 0.000 (0.032) Loss 2.2567 (8.4094) -Epoch: [0][300/706] Time 0.144 (0.198) Data 0.000 (0.031) Loss 3.6908 (8.2615) -Epoch: [0][310/706] Time 0.288 (0.197) Data 0.004 (0.030) Loss 2.7694 (8.0950) -Epoch: [0][320/706] Time 0.178 (0.196) Data 0.000 (0.030) Loss 2.0372 (7.9412) -Epoch: [0][330/706] Time 0.134 (0.196) Data 0.000 (0.029) Loss 2.9480 (7.7864) -Epoch: [0][340/706] Time 0.096 (0.195) Data 0.000 (0.028) Loss 5.0005 (7.6601) -Epoch: [0][350/706] Time 0.140 (0.194) Data 0.007 (0.027) Loss 3.5564 (7.5483) -Epoch: [0][360/706] Time 0.184 (0.193) Data 0.000 (0.027) Loss 2.5042 (7.4312) -Epoch: [0][370/706] Time 0.155 (0.192) Data 0.000 (0.026) Loss 3.1139 (7.3125) -Epoch: [0][380/706] Time 0.198 (0.192) Data 0.000 (0.026) Loss 3.0692 (7.2150) -Epoch: [0][390/706] Time 0.125 (0.191) Data 0.000 (0.025) Loss 2.3913 (7.1062) -Epoch: [0][400/706] Time 0.377 (0.190) Data 0.008 (0.025) Loss 2.7726 (6.9963) -Epoch: [0][410/706] Time 0.167 (0.190) Data 0.000 (0.024) Loss 3.7323 (6.8986) -Epoch: [0][420/706] Time 0.270 (0.189) Data 0.000 (0.024) Loss 4.3164 (6.8154) -Epoch: [0][430/706] Time 0.133 (0.188) Data 0.000 (0.023) Loss 2.8270 (6.7232) -Epoch: [0][440/706] Time 0.149 (0.188) Data 0.037 (0.023) Loss 2.8069 (6.6389) -Epoch: [0][450/706] Time 0.178 (0.187) Data 0.000 (0.023) Loss 3.3156 (6.5547) -Epoch: [0][460/706] Time 0.071 (0.187) Data 0.002 (0.022) Loss 3.2223 (6.4824) -Epoch: [0][470/706] Time 0.166 (0.186) Data 0.000 (0.022) Loss 1.7287 (6.4078) -Epoch: [0][480/706] Time 0.136 (0.186) Data 0.010 (0.021) Loss 3.7276 (6.3479) -Epoch: [0][490/706] Time 0.172 (0.186) Data 0.000 (0.021) Loss 2.8970 (6.2832) -Epoch: [0][500/706] Time 0.135 (0.185) Data 0.000 (0.021) Loss 2.3491 (6.2227) -Epoch: [0][510/706] Time 0.342 (0.185) Data 0.000 (0.020) Loss 4.5122 (6.1755) -Epoch: [0][520/706] Time 0.098 (0.184) Data 0.000 (0.020) Loss 2.9735 (6.1196) -Epoch: [0][530/706] Time 0.143 (0.184) Data 0.000 (0.020) Loss 2.6567 (6.0537) -Epoch: [0][540/706] Time 0.208 (0.184) Data 0.006 (0.020) Loss 3.3075 (6.0033) -Epoch: [0][550/706] Time 0.147 (0.183) Data 0.000 (0.019) Loss 4.2869 (5.9465) -Epoch: [0][560/706] Time 0.160 (0.183) Data 0.000 (0.019) Loss 2.1219 (5.8970) -Epoch: [0][570/706] Time 0.097 (0.182) Data 0.003 (0.019) Loss 2.0763 (5.8436) -Epoch: [0][580/706] Time 0.067 (0.182) Data 0.000 (0.019) Loss 2.7619 (5.7940) -Epoch: [0][590/706] Time 0.114 (0.181) Data 0.000 (0.018) Loss 1.5099 (5.7425) -Epoch: [0][600/706] Time 0.133 (0.181) Data 0.000 (0.018) Loss 3.3808 (5.6963) -Epoch: [0][610/706] Time 0.207 (0.181) Data 0.000 (0.018) Loss 2.0204 (5.6433) -Epoch: [0][620/706] Time 0.157 (0.180) Data 0.000 (0.018) Loss 2.9018 (5.5956) -Epoch: [0][630/706] Time 0.127 (0.180) Data 0.000 (0.017) Loss 3.1907 (5.5458) -Epoch: [0][640/706] Time 0.098 (0.180) Data 0.074 (0.017) Loss 1.6034 (5.5074) -Epoch: [0][650/706] Time 0.168 (0.179) Data 0.120 (0.017) Loss 2.2920 (5.4661) -Epoch: [0][660/706] Time 0.230 (0.179) Data 0.000 (0.017) Loss 1.7432 (5.4245) -Epoch: [0][670/706] Time 0.107 (0.179) Data 0.029 (0.017) Loss 0.8288 (5.3911) -Epoch: [0][680/706] Time 0.123 (0.178) Data 0.000 (0.017) Loss 1.6368 (5.3593) -Epoch: [0][690/706] Time 0.184 (0.178) Data 0.000 (0.017) Loss 2.6060 (5.3211) -Epoch: [0][700/706] Time 0.024 (0.176) Data 0.000 (0.016) Loss 2.0462 (5.2770) -inside validate -Epoch: [0][0/706] Time 9.279 (9.279) Data 6.781 (6.781) Loss 48.4298 (48.4298) -Epoch: [0][10/706] Time 0.181 (1.005) Data 0.018 (0.619) Loss 41.0558 (46.7052) -Epoch: [0][20/706] Time 0.105 (0.601) Data 0.000 (0.331) Loss 35.1504 (43.6154) -Epoch: [0][30/706] Time 0.130 (0.460) Data 0.000 (0.225) Loss 32.2156 (41.4865) -Epoch: [0][40/706] Time 0.177 (0.389) Data 0.000 (0.171) Loss 4.5833 (35.8459) -Epoch: [0][50/706] Time 0.172 (0.341) Data 0.000 (0.138) Loss 4.2765 (29.9234) -Epoch: [0][60/706] Time 0.135 (0.311) Data 0.000 (0.116) Loss 3.7185 (25.6463) -Epoch: [0][70/706] Time 0.238 (0.291) Data 0.005 (0.100) Loss 4.5449 (22.6682) -Epoch: [0][80/706] Time 0.223 (0.274) Data 0.007 (0.088) Loss 6.7830 (20.3990) -Epoch: [0][90/706] Time 0.227 (0.262) Data 0.000 (0.079) Loss 2.6813 (18.6320) -Epoch: [0][100/706] Time 0.388 (0.254) Data 0.000 (0.072) Loss 3.5365 (17.2279) -Epoch: [0][110/706] Time 0.145 (0.247) Data 0.000 (0.066) Loss 4.2246 (16.0745) -Epoch: [0][120/706] Time 0.248 (0.242) Data 0.002 (0.060) Loss 3.1311 (15.0758) -Epoch: [0][130/706] Time 0.152 (0.237) Data 0.016 (0.056) Loss 2.0549 (14.2282) -Epoch: [0][140/706] Time 0.054 (0.231) Data 0.016 (0.053) Loss 4.2596 (13.4916) -Epoch: [0][150/706] Time 0.264 (0.226) Data 0.005 (0.049) Loss 3.2887 (12.8522) -Epoch: [0][160/706] Time 0.184 (0.223) Data 0.005 (0.047) Loss 5.2555 (12.3141) -Epoch: [0][170/706] Time 0.125 (0.219) Data 0.000 (0.044) Loss 2.4943 (11.8206) -Epoch: [0][180/706] Time 0.049 (0.216) Data 0.000 (0.042) Loss 4.7814 (11.3783) -Epoch: [0][190/706] Time 0.225 (0.213) Data 0.009 (0.040) Loss 3.2696 (10.9805) -Epoch: [0][200/706] Time 0.135 (0.210) Data 0.000 (0.038) Loss 3.8138 (10.6168) -Epoch: [0][210/706] Time 0.176 (0.208) Data 0.000 (0.037) Loss 4.0394 (10.3144) -Epoch: [0][220/706] Time 0.134 (0.207) Data 0.000 (0.036) Loss 5.7033 (10.0182) -Epoch: [0][230/706] Time 0.179 (0.205) Data 0.000 (0.035) Loss 2.3560 (9.7332) -Epoch: [0][240/706] Time 0.153 (0.204) Data 0.000 (0.033) Loss 3.0160 (9.4815) -Epoch: [0][250/706] Time 0.082 (0.203) Data 0.000 (0.032) Loss 2.2496 (9.2878) -Epoch: [0][260/706] Time 0.144 (0.202) Data 0.000 (0.031) Loss 3.5626 (9.0680) -Epoch: [0][270/706] Time 0.315 (0.201) Data 0.000 (0.030) Loss 5.5424 (8.8735) -Epoch: [0][280/706] Time 0.150 (0.200) Data 0.000 (0.029) Loss 6.4262 (8.7193) -Epoch: [0][290/706] Time 0.161 (0.199) Data 0.021 (0.028) Loss 4.1452 (8.5346) -Epoch: [0][300/706] Time 0.184 (0.198) Data 0.000 (0.028) Loss 4.0946 (8.3708) -Epoch: [0][310/706] Time 0.141 (0.197) Data 0.018 (0.027) Loss 2.8782 (8.2157) -Epoch: [0][320/706] Time 0.226 (0.196) Data 0.000 (0.026) Loss 3.4344 (8.0733) -Epoch: [0][330/706] Time 0.268 (0.195) Data 0.177 (0.026) Loss 4.0020 (7.9181) -Epoch: [0][340/706] Time 0.148 (0.195) Data 0.027 (0.025) Loss 3.5979 (7.7708) -Epoch: [0][350/706] Time 0.124 (0.194) Data 0.000 (0.025) Loss 3.5631 (7.6466) -Epoch: [0][360/706] Time 0.178 (0.193) Data 0.000 (0.024) Loss 2.2941 (7.5321) -Epoch: [0][370/706] Time 0.158 (0.192) Data 0.000 (0.024) Loss 1.8477 (7.4150) -Epoch: [0][380/706] Time 0.120 (0.192) Data 0.020 (0.023) Loss 2.2733 (7.2838) -Epoch: [0][390/706] Time 0.155 (0.191) Data 0.000 (0.023) Loss 3.6436 (7.1927) -Epoch: [0][400/706] Time 0.218 (0.190) Data 0.000 (0.022) Loss 4.3071 (7.0985) -Epoch: [0][410/706] Time 0.126 (0.189) Data 0.000 (0.022) Loss 2.5885 (6.9914) -Epoch: [0][420/706] Time 0.163 (0.189) Data 0.000 (0.022) Loss 2.9067 (6.8971) -Epoch: [0][430/706] Time 0.123 (0.188) Data 0.002 (0.021) Loss 2.0391 (6.8008) -Epoch: [0][440/706] Time 0.137 (0.187) Data 0.000 (0.021) Loss 2.9169 (6.7169) -Epoch: [0][450/706] Time 0.280 (0.188) Data 0.000 (0.020) Loss 2.4427 (6.6385) -Epoch: [0][460/706] Time 0.087 (0.187) Data 0.005 (0.020) Loss 2.6100 (6.5513) -Epoch: [0][470/706] Time 0.149 (0.186) Data 0.000 (0.020) Loss 3.3577 (6.4802) -Epoch: [0][480/706] Time 0.158 (0.186) Data 0.008 (0.020) Loss 2.9320 (6.4077) -Epoch: [0][490/706] Time 0.172 (0.186) Data 0.000 (0.019) Loss 2.0522 (6.3431) -Epoch: [0][500/706] Time 0.156 (0.185) Data 0.000 (0.019) Loss 3.4436 (6.2737) -Epoch: [0][510/706] Time 0.200 (0.185) Data 0.000 (0.019) Loss 4.9813 (6.2248) -Epoch: [0][520/706] Time 0.058 (0.184) Data 0.000 (0.018) Loss 3.5226 (6.1639) -Epoch: [0][530/706] Time 0.225 (0.184) Data 0.000 (0.018) Loss 3.7303 (6.0935) -Epoch: [0][540/706] Time 0.126 (0.184) Data 0.000 (0.018) Loss 4.7602 (6.0457) -Epoch: [0][550/706] Time 0.117 (0.183) Data 0.000 (0.017) Loss 2.9582 (5.9881) -Epoch: [0][560/706] Time 0.199 (0.183) Data 0.000 (0.017) Loss 2.9999 (5.9365) -Epoch: [0][570/706] Time 0.120 (0.182) Data 0.014 (0.017) Loss 2.5402 (5.8740) -Epoch: [0][580/706] Time 0.139 (0.182) Data 0.004 (0.017) Loss 2.8438 (5.8120) -Epoch: [0][590/706] Time 0.097 (0.182) Data 0.000 (0.016) Loss 2.0868 (5.7586) -Epoch: [0][600/706] Time 0.223 (0.181) Data 0.023 (0.016) Loss 4.7334 (5.7170) -Epoch: [0][610/706] Time 0.213 (0.181) Data 0.014 (0.016) Loss 1.6728 (5.6729) -Epoch: [0][620/706] Time 0.077 (0.180) Data 0.009 (0.016) Loss 3.1532 (5.6339) -Epoch: [0][630/706] Time 0.128 (0.180) Data 0.023 (0.016) Loss 3.3617 (5.5984) -Epoch: [0][640/706] Time 0.193 (0.180) Data 0.000 (0.016) Loss 2.6781 (5.5533) -Epoch: [0][650/706] Time 0.149 (0.179) Data 0.000 (0.015) Loss 3.4790 (5.5028) -Epoch: [0][660/706] Time 0.215 (0.179) Data 0.015 (0.015) Loss 3.0638 (5.4707) -Epoch: [0][670/706] Time 0.093 (0.179) Data 0.000 (0.015) Loss 3.6359 (5.4316) -Epoch: [0][680/706] Time 0.119 (0.178) Data 0.000 (0.015) Loss 4.8591 (5.3971) -Epoch: [0][690/706] Time 0.079 (0.178) Data 0.000 (0.015) Loss 2.8482 (5.3671) -Epoch: [0][700/706] Time 0.024 (0.176) Data 0.000 (0.015) Loss 2.5585 (5.3280) -inside validate -Epoch: [0][0/706] Time 9.270 (9.270) Data 8.424 (8.424) Loss 43.3566 (43.3566) -Epoch: [0][10/706] Time 0.122 (0.993) Data 0.000 (0.768) Loss 35.7618 (41.7033) -Epoch: [0][20/706] Time 0.233 (0.598) Data 0.003 (0.403) Loss 32.7526 (41.1986) -Epoch: [0][30/706] Time 0.186 (0.455) Data 0.037 (0.275) Loss 32.2709 (39.1498) -Epoch: [0][40/706] Time 0.261 (0.385) Data 0.000 (0.209) Loss 3.2942 (34.2699) -Epoch: [0][50/706] Time 0.146 (0.341) Data 0.003 (0.170) Loss 2.7830 (28.9196) -Epoch: [0][60/706] Time 0.164 (0.310) Data 0.020 (0.144) Loss 2.7485 (24.9066) -Epoch: [0][70/706] Time 0.116 (0.289) Data 0.000 (0.125) Loss 5.6361 (21.9928) -Epoch: [0][80/706] Time 0.151 (0.273) Data 0.000 (0.111) Loss 3.9117 (19.8556) -Epoch: [0][90/706] Time 0.143 (0.263) Data 0.000 (0.099) Loss 4.7167 (18.1228) -Epoch: [0][100/706] Time 0.105 (0.253) Data 0.000 (0.089) Loss 4.2658 (16.7479) -Epoch: [0][110/706] Time 0.244 (0.246) Data 0.000 (0.082) Loss 2.8308 (15.6269) -Epoch: [0][120/706] Time 0.185 (0.240) Data 0.000 (0.075) Loss 2.3103 (14.6822) -Epoch: [0][130/706] Time 0.153 (0.236) Data 0.000 (0.070) Loss 3.9364 (13.8503) -Epoch: [0][140/706] Time 0.135 (0.232) Data 0.000 (0.065) Loss 3.9325 (13.1329) -Epoch: [0][150/706] Time 0.242 (0.227) Data 0.000 (0.062) Loss 4.2105 (12.5137) -Epoch: [0][160/706] Time 0.038 (0.223) Data 0.000 (0.058) Loss 3.5734 (12.0188) -Epoch: [0][170/706] Time 0.184 (0.219) Data 0.020 (0.055) Loss 1.9319 (11.5411) -Epoch: [0][180/706] Time 0.146 (0.216) Data 0.000 (0.053) Loss 4.2084 (11.0900) -Epoch: [0][190/706] Time 0.264 (0.213) Data 0.000 (0.050) Loss 2.1235 (10.6788) -Epoch: [0][200/706] Time 0.194 (0.210) Data 0.000 (0.048) Loss 2.3877 (10.3285) -Epoch: [0][210/706] Time 0.099 (0.208) Data 0.000 (0.046) Loss 2.7575 (10.0247) -Epoch: [0][220/706] Time 0.128 (0.206) Data 0.000 (0.044) Loss 1.6335 (9.7433) -Epoch: [0][230/706] Time 0.115 (0.205) Data 0.000 (0.043) Loss 5.0230 (9.4753) -Epoch: [0][240/706] Time 0.094 (0.204) Data 0.000 (0.041) Loss 3.7243 (9.2507) -Epoch: [0][250/706] Time 0.158 (0.203) Data 0.000 (0.040) Loss 1.5645 (9.0307) -Epoch: [0][260/706] Time 0.080 (0.201) Data 0.002 (0.038) Loss 5.1294 (8.8266) -Epoch: [0][270/706] Time 0.198 (0.200) Data 0.000 (0.037) Loss 3.6659 (8.6333) -Epoch: [0][280/706] Time 0.235 (0.201) Data 0.000 (0.036) Loss 4.0619 (8.4421) -Epoch: [0][290/706] Time 0.237 (0.199) Data 0.000 (0.035) Loss 3.8095 (8.2987) -Epoch: [0][300/706] Time 0.218 (0.197) Data 0.000 (0.034) Loss 6.1189 (8.1404) -Epoch: [0][310/706] Time 0.126 (0.197) Data 0.000 (0.033) Loss 5.0491 (8.0157) -Epoch: [0][320/706] Time 0.167 (0.196) Data 0.000 (0.032) Loss 4.1583 (7.8717) -Epoch: [0][330/706] Time 0.174 (0.195) Data 0.018 (0.031) Loss 4.0936 (7.7352) -Epoch: [0][340/706] Time 0.185 (0.195) Data 0.012 (0.030) Loss 3.0116 (7.5944) -Epoch: [0][350/706] Time 0.133 (0.194) Data 0.000 (0.030) Loss 2.4396 (7.4699) -Epoch: [0][360/706] Time 0.162 (0.193) Data 0.000 (0.029) Loss 4.6075 (7.3511) -Epoch: [0][370/706] Time 0.222 (0.192) Data 0.000 (0.028) Loss 3.4681 (7.2290) -Epoch: [0][380/706] Time 0.103 (0.191) Data 0.008 (0.028) Loss 3.4979 (7.1107) -Epoch: [0][390/706] Time 0.160 (0.191) Data 0.000 (0.027) Loss 1.6644 (7.0090) -Epoch: [0][400/706] Time 0.190 (0.190) Data 0.005 (0.026) Loss 3.0295 (6.8922) -Epoch: [0][410/706] Time 0.431 (0.190) Data 0.165 (0.026) Loss 1.2957 (6.7842) -Epoch: [0][420/706] Time 0.174 (0.189) Data 0.007 (0.025) Loss 4.3161 (6.7046) -Epoch: [0][430/706] Time 0.146 (0.188) Data 0.000 (0.025) Loss 2.4913 (6.6180) -Epoch: [0][440/706] Time 0.174 (0.188) Data 0.011 (0.025) Loss 3.0941 (6.5297) -Epoch: [0][450/706] Time 0.151 (0.187) Data 0.000 (0.024) Loss 3.2184 (6.4602) -Epoch: [0][460/706] Time 0.161 (0.187) Data 0.000 (0.024) Loss 2.4490 (6.3883) -Epoch: [0][470/706] Time 0.200 (0.186) Data 0.000 (0.023) Loss 3.0697 (6.3221) -Epoch: [0][480/706] Time 0.186 (0.186) Data 0.000 (0.023) Loss 2.5652 (6.2637) -Epoch: [0][490/706] Time 0.186 (0.186) Data 0.001 (0.022) Loss 2.8172 (6.1856) -Epoch: [0][500/706] Time 0.179 (0.185) Data 0.008 (0.022) Loss 4.4758 (6.1191) -Epoch: [0][510/706] Time 0.228 (0.185) Data 0.000 (0.022) Loss 2.9978 (6.0481) -Epoch: [0][520/706] Time 0.144 (0.185) Data 0.014 (0.021) Loss 4.1120 (6.0063) -Epoch: [0][530/706] Time 0.092 (0.184) Data 0.000 (0.021) Loss 1.8455 (5.9376) -Epoch: [0][540/706] Time 0.100 (0.183) Data 0.001 (0.021) Loss 2.4048 (5.8759) -Epoch: [0][550/706] Time 0.236 (0.183) Data 0.000 (0.021) Loss 2.9944 (5.8157) -Epoch: [0][560/706] Time 0.205 (0.183) Data 0.122 (0.020) Loss 2.9828 (5.7697) -Epoch: [0][570/706] Time 0.131 (0.182) Data 0.001 (0.020) Loss 2.3112 (5.7164) -Epoch: [0][580/706] Time 0.116 (0.182) Data 0.000 (0.020) Loss 2.8152 (5.6645) -Epoch: [0][590/706] Time 0.074 (0.181) Data 0.000 (0.019) Loss 2.8938 (5.6202) -Epoch: [0][600/706] Time 0.093 (0.181) Data 0.005 (0.019) Loss 2.2363 (5.5737) -Epoch: [0][610/706] Time 0.157 (0.181) Data 0.000 (0.019) Loss 0.7963 (5.5377) -Epoch: [0][620/706] Time 0.068 (0.180) Data 0.023 (0.019) Loss 1.2323 (5.4924) -Epoch: [0][630/706] Time 0.133 (0.180) Data 0.003 (0.018) Loss 3.7994 (5.4598) -Epoch: [0][640/706] Time 0.234 (0.180) Data 0.043 (0.018) Loss 3.2380 (5.4174) -Epoch: [0][650/706] Time 0.138 (0.179) Data 0.000 (0.018) Loss 2.9465 (5.3773) -Epoch: [0][660/706] Time 0.209 (0.179) Data 0.008 (0.018) Loss 1.9786 (5.3397) -Epoch: [0][670/706] Time 0.136 (0.179) Data 0.000 (0.018) Loss 2.7800 (5.3051) -Epoch: [0][680/706] Time 0.106 (0.178) Data 0.004 (0.018) Loss 1.2648 (5.2555) -Epoch: [0][690/706] Time 0.119 (0.178) Data 0.000 (0.017) Loss 2.3244 (5.2110) -Epoch: [0][700/706] Time 0.025 (0.176) Data 0.000 (0.017) Loss 2.9125 (5.1780) -inside validate -Epoch: [0][0/706] Time 9.282 (9.282) Data 6.623 (6.623) Loss 44.9327 (44.9327) -Epoch: [0][10/706] Time 0.160 (1.002) Data 0.011 (0.604) Loss 43.4475 (43.2021) -Epoch: [0][20/706] Time 0.174 (0.597) Data 0.000 (0.318) Loss 36.5762 (42.3764) -Epoch: [0][30/706] Time 0.177 (0.457) Data 0.000 (0.215) Loss 31.5012 (40.7765) -Epoch: [0][40/706] Time 0.192 (0.385) Data 0.000 (0.163) Loss 3.9172 (35.0510) -Epoch: [0][50/706] Time 0.131 (0.342) Data 0.000 (0.133) Loss 4.5722 (29.1811) -Epoch: [0][60/706] Time 0.205 (0.312) Data 0.000 (0.113) Loss 1.7294 (25.1167) -Epoch: [0][70/706] Time 0.218 (0.290) Data 0.000 (0.097) Loss 4.2902 (22.3084) -Epoch: [0][80/706] Time 0.183 (0.275) Data 0.005 (0.086) Loss 3.1367 (20.1359) -Epoch: [0][90/706] Time 0.259 (0.262) Data 0.001 (0.077) Loss 5.0983 (18.4130) -Epoch: [0][100/706] Time 0.062 (0.251) Data 0.000 (0.070) Loss 4.5993 (16.9484) -Epoch: [0][110/706] Time 0.209 (0.246) Data 0.076 (0.065) Loss 2.7832 (15.7784) -Epoch: [0][120/706] Time 0.232 (0.240) Data 0.006 (0.060) Loss 3.3960 (14.7805) -Epoch: [0][130/706] Time 0.111 (0.236) Data 0.000 (0.056) Loss 2.5398 (13.9614) -Epoch: [0][140/706] Time 0.299 (0.231) Data 0.029 (0.052) Loss 1.8820 (13.2436) -Epoch: [0][150/706] Time 0.146 (0.227) Data 0.000 (0.049) Loss 3.2122 (12.6113) -Epoch: [0][160/706] Time 0.142 (0.223) Data 0.020 (0.047) Loss 4.3948 (12.0532) -Epoch: [0][170/706] Time 0.134 (0.218) Data 0.005 (0.044) Loss 5.2315 (11.6180) -Epoch: [0][180/706] Time 0.188 (0.216) Data 0.000 (0.042) Loss 6.0750 (11.2292) -Epoch: [0][190/706] Time 0.144 (0.213) Data 0.000 (0.040) Loss 4.7591 (10.8376) -Epoch: [0][200/706] Time 0.106 (0.210) Data 0.000 (0.039) Loss 3.7509 (10.5219) -Epoch: [0][210/706] Time 0.154 (0.209) Data 0.015 (0.037) Loss 1.8348 (10.2228) -Epoch: [0][220/706] Time 0.171 (0.206) Data 0.000 (0.036) Loss 4.4392 (9.9380) -Epoch: [0][230/706] Time 0.180 (0.205) Data 0.000 (0.034) Loss 3.5013 (9.7042) -Epoch: [0][240/706] Time 0.233 (0.204) Data 0.000 (0.033) Loss 2.8788 (9.4637) -Epoch: [0][250/706] Time 0.115 (0.203) Data 0.000 (0.032) Loss 3.9517 (9.2312) -Epoch: [0][260/706] Time 0.188 (0.202) Data 0.005 (0.031) Loss 3.9902 (9.0185) -Epoch: [0][270/706] Time 0.160 (0.200) Data 0.016 (0.030) Loss 4.8580 (8.8304) -Epoch: [0][280/706] Time 0.171 (0.200) Data 0.008 (0.029) Loss 5.8785 (8.6676) -Epoch: [0][290/706] Time 0.145 (0.199) Data 0.000 (0.028) Loss 2.6185 (8.4927) -Epoch: [0][300/706] Time 0.161 (0.198) Data 0.000 (0.027) Loss 1.7884 (8.3225) -Epoch: [0][310/706] Time 0.071 (0.197) Data 0.000 (0.026) Loss 4.1143 (8.1751) -Epoch: [0][320/706] Time 0.246 (0.196) Data 0.000 (0.026) Loss 2.4283 (8.0230) -Epoch: [0][330/706] Time 0.143 (0.195) Data 0.002 (0.025) Loss 2.4110 (7.8757) -Epoch: [0][340/706] Time 0.122 (0.195) Data 0.000 (0.025) Loss 3.1787 (7.7403) -Epoch: [0][350/706] Time 0.086 (0.194) Data 0.022 (0.024) Loss 3.3941 (7.6123) -Epoch: [0][360/706] Time 0.131 (0.193) Data 0.000 (0.023) Loss 2.3606 (7.4793) -Epoch: [0][370/706] Time 0.172 (0.192) Data 0.020 (0.023) Loss 3.6307 (7.3673) -Epoch: [0][380/706] Time 0.080 (0.191) Data 0.000 (0.022) Loss 2.9831 (7.2577) -Epoch: [0][390/706] Time 0.133 (0.191) Data 0.000 (0.022) Loss 4.3452 (7.1611) -Epoch: [0][400/706] Time 0.284 (0.190) Data 0.000 (0.021) Loss 3.0082 (7.0576) -Epoch: [0][410/706] Time 0.283 (0.190) Data 0.077 (0.021) Loss 3.1628 (6.9600) -Epoch: [0][420/706] Time 0.151 (0.189) Data 0.000 (0.021) Loss 2.9701 (6.8660) -Epoch: [0][430/706] Time 0.161 (0.188) Data 0.007 (0.020) Loss 2.4728 (6.7824) -Epoch: [0][440/706] Time 0.135 (0.188) Data 0.000 (0.020) Loss 2.5133 (6.7003) -Epoch: [0][450/706] Time 0.188 (0.188) Data 0.071 (0.020) Loss 3.0822 (6.6067) -Epoch: [0][460/706] Time 0.096 (0.187) Data 0.010 (0.020) Loss 3.7917 (6.5278) -Epoch: [0][470/706] Time 0.168 (0.187) Data 0.000 (0.019) Loss 3.1482 (6.4450) -Epoch: [0][480/706] Time 0.198 (0.186) Data 0.000 (0.019) Loss 3.3619 (6.3666) -Epoch: [0][490/706] Time 0.191 (0.186) Data 0.000 (0.019) Loss 3.6006 (6.2926) -Epoch: [0][500/706] Time 0.152 (0.186) Data 0.000 (0.018) Loss 2.6715 (6.2205) -Epoch: [0][510/706] Time 0.225 (0.185) Data 0.000 (0.018) Loss 2.6792 (6.1568) -Epoch: [0][520/706] Time 0.134 (0.184) Data 0.000 (0.018) Loss 2.8775 (6.0929) -Epoch: [0][530/706] Time 0.110 (0.184) Data 0.000 (0.018) Loss 1.6893 (6.0319) -Epoch: [0][540/706] Time 0.078 (0.184) Data 0.002 (0.017) Loss 1.5212 (5.9700) -Epoch: [0][550/706] Time 0.202 (0.183) Data 0.000 (0.017) Loss 3.6055 (5.9085) -Epoch: [0][560/706] Time 0.051 (0.183) Data 0.000 (0.017) Loss 2.5571 (5.8535) -Epoch: [0][570/706] Time 0.206 (0.182) Data 0.000 (0.016) Loss 1.7188 (5.7950) -Epoch: [0][580/706] Time 0.161 (0.182) Data 0.005 (0.016) Loss 4.2880 (5.7468) -Epoch: [0][590/706] Time 0.085 (0.181) Data 0.000 (0.016) Loss 3.0299 (5.7007) -Epoch: [0][600/706] Time 0.215 (0.181) Data 0.000 (0.016) Loss 2.4649 (5.6484) -Epoch: [0][610/706] Time 0.191 (0.181) Data 0.000 (0.016) Loss 3.0358 (5.6003) -Epoch: [0][620/706] Time 0.158 (0.180) Data 0.012 (0.015) Loss 3.2750 (5.5605) -Epoch: [0][630/706] Time 0.069 (0.180) Data 0.000 (0.015) Loss 3.1849 (5.5216) -Epoch: [0][640/706] Time 0.212 (0.180) Data 0.000 (0.015) Loss 1.9852 (5.4855) -Epoch: [0][650/706] Time 0.069 (0.179) Data 0.000 (0.015) Loss 3.4488 (5.4452) -Epoch: [0][660/706] Time 0.185 (0.179) Data 0.000 (0.014) Loss 3.9314 (5.4061) -Epoch: [0][670/706] Time 0.081 (0.179) Data 0.004 (0.014) Loss 0.7095 (5.3615) -Epoch: [0][680/706] Time 0.105 (0.178) Data 0.000 (0.014) Loss 2.7406 (5.3247) -Epoch: [0][690/706] Time 0.175 (0.178) Data 0.000 (0.014) Loss 4.1669 (5.2834) -Epoch: [0][700/706] Time 0.021 (0.176) Data 0.000 (0.014) Loss 3.6803 (5.2500) -inside validate -Epoch: [0][0/706] Time 9.446 (9.446) Data 6.326 (6.326) Loss 41.1825 (41.1825) -Epoch: [0][10/706] Time 0.157 (0.990) Data 0.039 (0.584) Loss 40.4952 (43.2361) -Epoch: [0][20/706] Time 0.193 (0.596) Data 0.000 (0.307) Loss 34.3756 (42.8838) -Epoch: [0][30/706] Time 0.231 (0.457) Data 0.015 (0.210) Loss 28.2018 (40.3380) -Epoch: [0][40/706] Time 0.214 (0.386) Data 0.002 (0.159) Loss 5.5274 (34.1745) -Epoch: [0][50/706] Time 0.165 (0.341) Data 0.000 (0.129) Loss 3.3931 (28.6024) -Epoch: [0][60/706] Time 0.115 (0.310) Data 0.007 (0.108) Loss 3.4990 (24.7291) -Epoch: [0][70/706] Time 0.158 (0.290) Data 0.000 (0.094) Loss 5.9395 (21.8957) -Epoch: [0][80/706] Time 0.116 (0.273) Data 0.001 (0.083) Loss 5.2083 (19.6922) -Epoch: [0][90/706] Time 0.239 (0.262) Data 0.000 (0.074) Loss 3.5325 (17.9426) -Epoch: [0][100/706] Time 0.127 (0.251) Data 0.000 (0.067) Loss 4.0950 (16.5935) -Epoch: [0][110/706] Time 0.243 (0.245) Data 0.000 (0.061) Loss 5.2329 (15.4758) -Epoch: [0][120/706] Time 0.187 (0.240) Data 0.007 (0.056) Loss 4.1025 (14.5035) -Epoch: [0][130/706] Time 0.168 (0.237) Data 0.000 (0.053) Loss 3.8608 (13.6563) -Epoch: [0][140/706] Time 0.299 (0.231) Data 0.009 (0.050) Loss 3.5996 (12.9351) -Epoch: [0][150/706] Time 0.226 (0.226) Data 0.002 (0.047) Loss 2.5297 (12.3095) -Epoch: [0][160/706] Time 0.159 (0.223) Data 0.000 (0.044) Loss 4.2293 (11.7826) -Epoch: [0][170/706] Time 0.084 (0.218) Data 0.000 (0.042) Loss 4.0215 (11.3283) -Epoch: [0][180/706] Time 0.247 (0.216) Data 0.007 (0.041) Loss 4.3051 (10.9198) -Epoch: [0][190/706] Time 0.179 (0.213) Data 0.000 (0.039) Loss 3.4869 (10.5700) -Epoch: [0][200/706] Time 0.206 (0.211) Data 0.000 (0.037) Loss 4.9523 (10.2391) -Epoch: [0][210/706] Time 0.173 (0.209) Data 0.017 (0.035) Loss 2.7213 (9.9285) -Epoch: [0][220/706] Time 0.187 (0.207) Data 0.000 (0.034) Loss 3.7719 (9.6530) -Epoch: [0][230/706] Time 0.222 (0.205) Data 0.002 (0.033) Loss 4.4169 (9.4351) -Epoch: [0][240/706] Time 0.222 (0.205) Data 0.000 (0.032) Loss 4.0282 (9.2010) -Epoch: [0][250/706] Time 0.154 (0.203) Data 0.000 (0.031) Loss 4.6461 (9.0080) -Epoch: [0][260/706] Time 0.294 (0.203) Data 0.014 (0.030) Loss 2.0598 (8.7895) -Epoch: [0][270/706] Time 0.278 (0.202) Data 0.000 (0.028) Loss 3.7385 (8.6049) -Epoch: [0][280/706] Time 0.171 (0.200) Data 0.002 (0.028) Loss 5.9370 (8.4315) -Epoch: [0][290/706] Time 0.157 (0.199) Data 0.000 (0.027) Loss 2.4879 (8.3007) -Epoch: [0][300/706] Time 0.256 (0.198) Data 0.081 (0.026) Loss 4.2397 (8.1479) -Epoch: [0][310/706] Time 0.075 (0.197) Data 0.000 (0.026) Loss 5.8137 (8.0070) -Epoch: [0][320/706] Time 0.195 (0.196) Data 0.000 (0.025) Loss 3.3306 (7.8720) -Epoch: [0][330/706] Time 0.109 (0.195) Data 0.015 (0.025) Loss 2.0692 (7.7137) -Epoch: [0][340/706] Time 0.095 (0.195) Data 0.005 (0.024) Loss 2.0057 (7.5659) -Epoch: [0][350/706] Time 0.145 (0.194) Data 0.010 (0.023) Loss 2.5474 (7.4460) -Epoch: [0][360/706] Time 0.178 (0.193) Data 0.027 (0.023) Loss 3.7421 (7.3256) -Epoch: [0][370/706] Time 0.156 (0.192) Data 0.011 (0.022) Loss 1.7487 (7.2040) -Epoch: [0][380/706] Time 0.142 (0.192) Data 0.000 (0.022) Loss 2.1930 (7.0927) -Epoch: [0][390/706] Time 0.123 (0.191) Data 0.000 (0.021) Loss 5.4662 (6.9925) -Epoch: [0][400/706] Time 0.108 (0.190) Data 0.000 (0.021) Loss 2.7798 (6.8850) -Epoch: [0][410/706] Time 0.154 (0.190) Data 0.000 (0.021) Loss 3.3253 (6.7847) -Epoch: [0][420/706] Time 0.142 (0.189) Data 0.003 (0.020) Loss 2.4046 (6.6879) -Epoch: [0][430/706] Time 0.202 (0.188) Data 0.000 (0.020) Loss 1.4413 (6.5966) -Epoch: [0][440/706] Time 0.177 (0.188) Data 0.000 (0.019) Loss 2.0765 (6.5136) -Epoch: [0][450/706] Time 0.140 (0.187) Data 0.000 (0.019) Loss 3.8305 (6.4341) -Epoch: [0][460/706] Time 0.193 (0.187) Data 0.000 (0.019) Loss 2.3949 (6.3625) -Epoch: [0][470/706] Time 0.117 (0.186) Data 0.000 (0.018) Loss 1.5495 (6.3012) -Epoch: [0][480/706] Time 0.156 (0.186) Data 0.000 (0.018) Loss 4.1282 (6.2341) -Epoch: [0][490/706] Time 0.189 (0.186) Data 0.006 (0.018) Loss 3.4204 (6.1583) -Epoch: [0][500/706] Time 0.124 (0.185) Data 0.000 (0.018) Loss 1.8471 (6.0819) -Epoch: [0][510/706] Time 0.136 (0.185) Data 0.000 (0.018) Loss 3.4830 (6.0234) -Epoch: [0][520/706] Time 0.228 (0.185) Data 0.000 (0.017) Loss 4.4536 (5.9572) -Epoch: [0][530/706] Time 0.135 (0.184) Data 0.000 (0.017) Loss 2.6193 (5.9031) -Epoch: [0][540/706] Time 0.140 (0.184) Data 0.000 (0.017) Loss 2.6970 (5.8381) -Epoch: [0][550/706] Time 0.096 (0.183) Data 0.010 (0.017) Loss 4.1912 (5.7880) -Epoch: [0][560/706] Time 0.127 (0.183) Data 0.000 (0.016) Loss 3.5133 (5.7427) -Epoch: [0][570/706] Time 0.291 (0.182) Data 0.012 (0.016) Loss 2.8835 (5.6894) -Epoch: [0][580/706] Time 0.164 (0.182) Data 0.005 (0.016) Loss 3.9494 (5.6424) -Epoch: [0][590/706] Time 0.051 (0.181) Data 0.000 (0.016) Loss 2.2715 (5.6015) -Epoch: [0][600/706] Time 0.213 (0.181) Data 0.019 (0.016) Loss 2.7813 (5.5518) -Epoch: [0][610/706] Time 0.257 (0.181) Data 0.000 (0.016) Loss 2.8138 (5.5041) -Epoch: [0][620/706] Time 0.103 (0.180) Data 0.007 (0.015) Loss 1.1138 (5.4682) -Epoch: [0][630/706] Time 0.100 (0.180) Data 0.016 (0.015) Loss 1.8424 (5.4247) -Epoch: [0][640/706] Time 0.263 (0.180) Data 0.000 (0.015) Loss 3.5471 (5.3850) -Epoch: [0][650/706] Time 0.120 (0.179) Data 0.000 (0.015) Loss 1.7864 (5.3393) -Epoch: [0][660/706] Time 0.206 (0.179) Data 0.005 (0.015) Loss 2.9983 (5.3001) -Epoch: [0][670/706] Time 0.171 (0.179) Data 0.000 (0.015) Loss 2.8595 (5.2583) -Epoch: [0][680/706] Time 0.179 (0.179) Data 0.000 (0.014) Loss 1.3420 (5.2239) -Epoch: [0][690/706] Time 0.255 (0.178) Data 0.000 (0.014) Loss 1.1615 (5.1894) -Epoch: [0][700/706] Time 0.025 (0.176) Data 0.000 (0.014) Loss 2.5902 (5.1528) -inside validate -Epoch: [0][0/706] Time 9.257 (9.257) Data 6.904 (6.904) Loss 46.9392 (46.9392) -Epoch: [0][10/706] Time 0.206 (0.998) Data 0.000 (0.637) Loss 36.4863 (42.4265) -Epoch: [0][20/706] Time 0.188 (0.596) Data 0.000 (0.337) Loss 41.2150 (42.8300) -Epoch: [0][30/706] Time 0.160 (0.455) Data 0.000 (0.231) Loss 24.4870 (41.3227) -Epoch: [0][40/706] Time 0.260 (0.385) Data 0.000 (0.175) Loss 3.3713 (35.6103) -Epoch: [0][50/706] Time 0.121 (0.340) Data 0.000 (0.143) Loss 5.2313 (29.7555) -Epoch: [0][60/706] Time 0.197 (0.310) Data 0.011 (0.120) Loss 3.3830 (25.5695) -Epoch: [0][70/706] Time 0.126 (0.289) Data 0.017 (0.104) Loss 4.4183 (22.6043) -Epoch: [0][80/706] Time 0.119 (0.274) Data 0.000 (0.094) Loss 5.8141 (20.3905) -Epoch: [0][90/706] Time 0.077 (0.262) Data 0.000 (0.084) Loss 4.8870 (18.6150) -Epoch: [0][100/706] Time 0.146 (0.252) Data 0.000 (0.076) Loss 7.3811 (17.3104) -Epoch: [0][110/706] Time 0.170 (0.245) Data 0.000 (0.069) Loss 3.9041 (16.0491) -Epoch: [0][120/706] Time 0.096 (0.240) Data 0.000 (0.064) Loss 4.8351 (15.0538) -Epoch: [0][130/706] Time 0.131 (0.236) Data 0.000 (0.060) Loss 5.3418 (14.2138) -Epoch: [0][140/706] Time 0.271 (0.231) Data 0.000 (0.056) Loss 4.3386 (13.5198) -Epoch: [0][150/706] Time 0.236 (0.226) Data 0.000 (0.053) Loss 4.9112 (12.9352) -Epoch: [0][160/706] Time 0.166 (0.223) Data 0.005 (0.050) Loss 3.9539 (12.3776) -Epoch: [0][170/706] Time 0.115 (0.218) Data 0.000 (0.047) Loss 4.3524 (11.8989) -Epoch: [0][180/706] Time 0.131 (0.216) Data 0.000 (0.045) Loss 1.7779 (11.5155) -Epoch: [0][190/706] Time 0.196 (0.212) Data 0.019 (0.043) Loss 3.6577 (11.1265) -Epoch: [0][200/706] Time 0.200 (0.210) Data 0.000 (0.041) Loss 6.7872 (10.7829) -Epoch: [0][210/706] Time 0.183 (0.208) Data 0.011 (0.040) Loss 5.7115 (10.4877) -Epoch: [0][220/706] Time 0.183 (0.206) Data 0.003 (0.038) Loss 4.1582 (10.1741) -Epoch: [0][230/706] Time 0.173 (0.205) Data 0.007 (0.036) Loss 3.9689 (9.9009) -Epoch: [0][240/706] Time 0.196 (0.203) Data 0.004 (0.035) Loss 4.4565 (9.6379) -Epoch: [0][250/706] Time 0.035 (0.203) Data 0.000 (0.034) Loss 4.4626 (9.4151) -Epoch: [0][260/706] Time 0.107 (0.201) Data 0.018 (0.033) Loss 5.4872 (9.2326) -Epoch: [0][270/706] Time 0.253 (0.201) Data 0.000 (0.032) Loss 2.7825 (9.0204) -Epoch: [0][280/706] Time 0.147 (0.200) Data 0.000 (0.031) Loss 3.6315 (8.8507) -Epoch: [0][290/706] Time 0.198 (0.199) Data 0.009 (0.030) Loss 3.3470 (8.6787) -Epoch: [0][300/706] Time 0.178 (0.198) Data 0.000 (0.029) Loss 3.3525 (8.5154) -Epoch: [0][310/706] Time 0.147 (0.197) Data 0.003 (0.028) Loss 2.6387 (8.3336) -Epoch: [0][320/706] Time 0.241 (0.196) Data 0.001 (0.027) Loss 2.9629 (8.1681) -Epoch: [0][330/706] Time 0.236 (0.195) Data 0.000 (0.027) Loss 2.5483 (8.0252) -Epoch: [0][340/706] Time 0.185 (0.195) Data 0.000 (0.026) Loss 3.6524 (7.8862) -Epoch: [0][350/706] Time 0.208 (0.194) Data 0.000 (0.025) Loss 3.5321 (7.7506) -Epoch: [0][360/706] Time 0.166 (0.193) Data 0.001 (0.025) Loss 2.2244 (7.6099) -Epoch: [0][370/706] Time 0.145 (0.192) Data 0.000 (0.024) Loss 3.1572 (7.4899) -Epoch: [0][380/706] Time 0.101 (0.192) Data 0.004 (0.024) Loss 3.3744 (7.3659) -Epoch: [0][390/706] Time 0.129 (0.191) Data 0.000 (0.023) Loss 4.1872 (7.2811) -Epoch: [0][400/706] Time 0.248 (0.190) Data 0.001 (0.023) Loss 2.8128 (7.1942) -Epoch: [0][410/706] Time 0.200 (0.189) Data 0.000 (0.022) Loss 3.5235 (7.0870) -Epoch: [0][420/706] Time 0.286 (0.189) Data 0.008 (0.022) Loss 3.2610 (6.9841) -Epoch: [0][430/706] Time 0.175 (0.188) Data 0.000 (0.022) Loss 2.1329 (6.8970) -Epoch: [0][440/706] Time 0.157 (0.188) Data 0.010 (0.022) Loss 2.1469 (6.8079) -Epoch: [0][450/706] Time 0.085 (0.187) Data 0.000 (0.021) Loss 3.8129 (6.7316) -Epoch: [0][460/706] Time 0.297 (0.187) Data 0.008 (0.021) Loss 4.4565 (6.6588) -Epoch: [0][470/706] Time 0.148 (0.186) Data 0.000 (0.020) Loss 3.8682 (6.5853) -Epoch: [0][480/706] Time 0.154 (0.186) Data 0.000 (0.020) Loss 1.5659 (6.5042) -Epoch: [0][490/706] Time 0.253 (0.186) Data 0.000 (0.020) Loss 2.6194 (6.4278) -Epoch: [0][500/706] Time 0.151 (0.185) Data 0.000 (0.019) Loss 2.7362 (6.3503) -Epoch: [0][510/706] Time 0.215 (0.185) Data 0.000 (0.019) Loss 2.0649 (6.2741) -Epoch: [0][520/706] Time 0.186 (0.185) Data 0.042 (0.019) Loss 1.8007 (6.2083) -Epoch: [0][530/706] Time 0.229 (0.184) Data 0.029 (0.018) Loss 1.3606 (6.1380) -Epoch: [0][540/706] Time 0.189 (0.183) Data 0.000 (0.018) Loss 1.3262 (6.0817) -Epoch: [0][550/706] Time 0.073 (0.183) Data 0.000 (0.018) Loss 1.8562 (6.0157) -Epoch: [0][560/706] Time 0.168 (0.183) Data 0.014 (0.018) Loss 2.2618 (5.9648) -Epoch: [0][570/706] Time 0.109 (0.182) Data 0.000 (0.018) Loss 1.7563 (5.9107) -Epoch: [0][580/706] Time 0.146 (0.182) Data 0.000 (0.017) Loss 3.2830 (5.8566) -Epoch: [0][590/706] Time 0.222 (0.182) Data 0.000 (0.017) Loss 2.2418 (5.8097) -Epoch: [0][600/706] Time 0.200 (0.181) Data 0.000 (0.017) Loss 2.2696 (5.7568) -Epoch: [0][610/706] Time 0.230 (0.181) Data 0.000 (0.017) Loss 5.0627 (5.7153) -Epoch: [0][620/706] Time 0.137 (0.180) Data 0.000 (0.016) Loss 3.3174 (5.6661) -Epoch: [0][630/706] Time 0.083 (0.180) Data 0.000 (0.016) Loss 2.3991 (5.6235) -Epoch: [0][640/706] Time 0.297 (0.180) Data 0.019 (0.016) Loss 3.2298 (5.5823) -Epoch: [0][650/706] Time 0.182 (0.179) Data 0.026 (0.016) Loss 2.2059 (5.5366) -Epoch: [0][660/706] Time 0.220 (0.179) Data 0.000 (0.016) Loss 3.1342 (5.4928) -Epoch: [0][670/706] Time 0.099 (0.179) Data 0.000 (0.016) Loss 2.5465 (5.4509) -Epoch: [0][680/706] Time 0.097 (0.178) Data 0.000 (0.015) Loss 2.0521 (5.4094) -Epoch: [0][690/706] Time 0.193 (0.178) Data 0.008 (0.015) Loss 2.2222 (5.3651) -Epoch: [0][700/706] Time 0.024 (0.176) Data 0.000 (0.015) Loss 3.2591 (5.3341) -inside validate -Epoch: [0][0/706] Time 9.314 (9.314) Data 7.158 (7.158) Loss 38.1948 (38.1948) -Epoch: [0][10/706] Time 0.108 (0.992) Data 0.000 (0.652) Loss 30.7854 (39.2707) -Epoch: [0][20/706] Time 0.197 (0.598) Data 0.004 (0.344) Loss 54.6684 (42.5114) -Epoch: [0][30/706] Time 0.237 (0.456) Data 0.005 (0.233) Loss 36.5165 (41.1070) -Epoch: [0][40/706] Time 0.036 (0.387) Data 0.000 (0.177) Loss 2.6175 (35.1194) -Epoch: [0][50/706] Time 0.221 (0.343) Data 0.001 (0.143) Loss 3.2051 (29.3465) -Epoch: [0][60/706] Time 0.152 (0.309) Data 0.000 (0.121) Loss 3.7700 (25.3188) -Epoch: [0][70/706] Time 0.125 (0.289) Data 0.005 (0.104) Loss 4.8583 (22.3857) -Epoch: [0][80/706] Time 0.199 (0.273) Data 0.004 (0.093) Loss 5.5569 (20.0913) -Epoch: [0][90/706] Time 0.251 (0.262) Data 0.004 (0.084) Loss 3.4442 (18.3590) -Epoch: [0][100/706] Time 0.164 (0.251) Data 0.000 (0.075) Loss 6.3650 (16.9261) -Epoch: [0][110/706] Time 0.152 (0.245) Data 0.022 (0.069) Loss 4.9548 (15.7419) -Epoch: [0][120/706] Time 0.115 (0.240) Data 0.000 (0.064) Loss 2.9369 (14.7810) -Epoch: [0][130/706] Time 0.221 (0.236) Data 0.016 (0.059) Loss 3.2714 (13.9544) -Epoch: [0][140/706] Time 0.333 (0.231) Data 0.000 (0.055) Loss 5.6468 (13.2544) -Epoch: [0][150/706] Time 0.308 (0.227) Data 0.000 (0.052) Loss 3.6223 (12.6744) -Epoch: [0][160/706] Time 0.135 (0.223) Data 0.000 (0.049) Loss 2.9398 (12.1049) -Epoch: [0][170/706] Time 0.072 (0.218) Data 0.000 (0.046) Loss 2.0558 (11.6665) -Epoch: [0][180/706] Time 0.164 (0.216) Data 0.041 (0.044) Loss 6.3391 (11.2636) -Epoch: [0][190/706] Time 0.156 (0.212) Data 0.008 (0.042) Loss 2.1807 (10.8824) -Epoch: [0][200/706] Time 0.136 (0.209) Data 0.000 (0.040) Loss 4.4586 (10.5332) -Epoch: [0][210/706] Time 0.099 (0.208) Data 0.000 (0.038) Loss 4.3520 (10.2146) -Epoch: [0][220/706] Time 0.179 (0.206) Data 0.000 (0.037) Loss 3.0127 (9.9007) -Epoch: [0][230/706] Time 0.046 (0.204) Data 0.000 (0.035) Loss 3.2982 (9.6444) -Epoch: [0][240/706] Time 0.186 (0.204) Data 0.000 (0.034) Loss 5.0238 (9.4224) -Epoch: [0][250/706] Time 0.207 (0.203) Data 0.000 (0.033) Loss 2.6750 (9.2101) -Epoch: [0][260/706] Time 0.187 (0.201) Data 0.023 (0.032) Loss 4.8456 (9.0217) -Epoch: [0][270/706] Time 0.159 (0.200) Data 0.000 (0.031) Loss 3.2206 (8.8208) -Epoch: [0][280/706] Time 0.115 (0.200) Data 0.000 (0.030) Loss 3.0691 (8.6296) -Epoch: [0][290/706] Time 0.222 (0.199) Data 0.005 (0.029) Loss 3.6303 (8.4568) -Epoch: [0][300/706] Time 0.111 (0.197) Data 0.000 (0.029) Loss 1.8436 (8.2831) -Epoch: [0][310/706] Time 0.221 (0.197) Data 0.000 (0.028) Loss 3.4481 (8.1099) -Epoch: [0][320/706] Time 0.180 (0.196) Data 0.000 (0.027) Loss 4.9176 (7.9708) -Epoch: [0][330/706] Time 0.201 (0.195) Data 0.001 (0.026) Loss 2.3200 (7.8128) -Epoch: [0][340/706] Time 0.169 (0.195) Data 0.000 (0.026) Loss 3.0642 (7.6653) -Epoch: [0][350/706] Time 0.149 (0.194) Data 0.000 (0.025) Loss 1.6174 (7.5302) -Epoch: [0][360/706] Time 0.074 (0.193) Data 0.000 (0.024) Loss 2.9764 (7.4246) -Epoch: [0][370/706] Time 0.334 (0.192) Data 0.000 (0.024) Loss 3.8505 (7.3258) -Epoch: [0][380/706] Time 0.047 (0.191) Data 0.000 (0.023) Loss 2.0031 (7.2124) -Epoch: [0][390/706] Time 0.140 (0.190) Data 0.000 (0.023) Loss 4.2562 (7.1031) -Epoch: [0][400/706] Time 0.298 (0.190) Data 0.000 (0.023) Loss 2.9798 (7.0041) -Epoch: [0][410/706] Time 0.187 (0.189) Data 0.000 (0.022) Loss 3.3963 (6.9031) -Epoch: [0][420/706] Time 0.143 (0.189) Data 0.000 (0.022) Loss 4.2049 (6.8159) -Epoch: [0][430/706] Time 0.085 (0.188) Data 0.000 (0.021) Loss 2.9680 (6.7148) -Epoch: [0][440/706] Time 0.158 (0.188) Data 0.000 (0.021) Loss 3.1386 (6.6302) -Epoch: [0][450/706] Time 0.123 (0.187) Data 0.000 (0.021) Loss 2.5326 (6.5450) -Epoch: [0][460/706] Time 0.292 (0.187) Data 0.009 (0.021) Loss 2.9561 (6.4735) -Epoch: [0][470/706] Time 0.165 (0.186) Data 0.000 (0.020) Loss 4.2343 (6.3916) -Epoch: [0][480/706] Time 0.192 (0.186) Data 0.000 (0.020) Loss 1.7140 (6.3200) -Epoch: [0][490/706] Time 0.206 (0.186) Data 0.002 (0.020) Loss 3.6381 (6.2485) -Epoch: [0][500/706] Time 0.156 (0.186) Data 0.000 (0.020) Loss 2.1828 (6.1776) -Epoch: [0][510/706] Time 0.115 (0.185) Data 0.000 (0.019) Loss 4.6371 (6.1127) -Epoch: [0][520/706] Time 0.193 (0.185) Data 0.000 (0.019) Loss 1.5458 (6.0518) -Epoch: [0][530/706] Time 0.126 (0.184) Data 0.000 (0.019) Loss 4.5534 (5.9951) -Epoch: [0][540/706] Time 0.111 (0.183) Data 0.000 (0.018) Loss 3.7050 (5.9447) -Epoch: [0][550/706] Time 0.201 (0.183) Data 0.002 (0.018) Loss 5.9172 (5.8933) -Epoch: [0][560/706] Time 0.218 (0.183) Data 0.005 (0.018) Loss 4.2525 (5.8400) -Epoch: [0][570/706] Time 0.160 (0.182) Data 0.000 (0.018) Loss 2.7027 (5.7894) -Epoch: [0][580/706] Time 0.142 (0.182) Data 0.023 (0.018) Loss 3.8773 (5.7365) -Epoch: [0][590/706] Time 0.095 (0.181) Data 0.013 (0.018) Loss 2.4959 (5.6791) -Epoch: [0][600/706] Time 0.336 (0.181) Data 0.004 (0.017) Loss 1.6744 (5.6225) -Epoch: [0][610/706] Time 0.129 (0.181) Data 0.000 (0.017) Loss 4.2712 (5.5827) -Epoch: [0][620/706] Time 0.122 (0.180) Data 0.011 (0.017) Loss 3.6395 (5.5424) -Epoch: [0][630/706] Time 0.078 (0.180) Data 0.000 (0.017) Loss 2.4458 (5.4912) -Epoch: [0][640/706] Time 0.244 (0.179) Data 0.000 (0.016) Loss 3.5233 (5.4446) -Epoch: [0][650/706] Time 0.182 (0.179) Data 0.010 (0.016) Loss 1.6506 (5.3921) -Epoch: [0][660/706] Time 0.162 (0.179) Data 0.004 (0.016) Loss 2.4824 (5.3474) -Epoch: [0][670/706] Time 0.148 (0.179) Data 0.006 (0.016) Loss 1.5915 (5.3046) -Epoch: [0][680/706] Time 0.161 (0.178) Data 0.000 (0.016) Loss 2.3357 (5.2696) -Epoch: [0][690/706] Time 0.129 (0.178) Data 0.000 (0.016) Loss 2.9099 (5.2376) -Epoch: [0][700/706] Time 0.024 (0.176) Data 0.000 (0.016) Loss 1.8607 (5.1981) -inside validate -Test: [0/435] Time 5.854 (5.854) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.016 (0.586) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.138 (0.334) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.036 (0.241) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.032 (0.193) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.019 (0.166) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.075 (0.149) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.049 (0.137) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.057 (0.128) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.040 (0.122) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.208 (0.118) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.036 (0.113) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.056 (0.108) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.010 (0.107) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.059 (0.106) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.032 (0.103) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.073 (0.101) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.011 (0.100) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.014 (0.098) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.032 (0.096) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.066 (0.096) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.022 (0.096) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.044 (0.094) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.088 (0.092) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.052 (0.090) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.056 (0.090) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.073 (0.090) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.171 (0.089) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.039 (0.088) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.078 (0.088) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.088 (0.088) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.371 (0.088) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.032 (0.087) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.030 (0.087) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.080 (0.086) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.370 (0.088) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.028 (0.087) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.523 (0.088) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.061 (0.089) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.059 (0.089) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.038 (0.089) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.223 (0.090) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.070 (0.089) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.126 (0.088) Loss 15.1994 (3.0437) -Test: [0/435] Time 5.740 (5.740) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.017 (0.554) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.035 (0.307) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.028 (0.225) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.039 (0.185) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.016 (0.162) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.034 (0.148) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.045 (0.135) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.051 (0.128) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.052 (0.121) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.050 (0.117) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.052 (0.112) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.036 (0.108) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.031 (0.104) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.065 (0.103) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.065 (0.099) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.117 (0.098) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.044 (0.097) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.071 (0.096) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.049 (0.095) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.056 (0.094) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.016 (0.092) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.012 (0.091) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.058 (0.090) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.051 (0.090) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.053 (0.089) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.059 (0.089) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.068 (0.088) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.122 (0.087) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.725 (0.089) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.035 (0.088) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.032 (0.087) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.155 (0.087) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.012 (0.087) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.044 (0.087) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.399 (0.087) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.054 (0.088) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.431 (0.088) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.040 (0.089) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.077 (0.089) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.009 (0.088) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.292 (0.089) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.025 (0.088) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.034 (0.088) Loss 15.1994 (3.0437) -Test: [0/435] Time 5.563 (5.563) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.032 (0.554) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.025 (0.306) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.108 (0.222) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.031 (0.187) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.031 (0.159) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.028 (0.143) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.054 (0.130) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.031 (0.124) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.023 (0.118) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.050 (0.112) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.057 (0.109) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.300 (0.108) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.047 (0.106) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.042 (0.106) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.029 (0.104) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.129 (0.102) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.011 (0.100) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.054 (0.098) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.057 (0.097) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.201 (0.097) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.041 (0.096) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.047 (0.094) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.063 (0.092) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.123 (0.092) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.066 (0.091) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.104 (0.091) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.346 (0.091) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.038 (0.091) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.082 (0.091) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.046 (0.091) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.369 (0.091) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.033 (0.091) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.063 (0.090) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.022 (0.090) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.260 (0.090) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.030 (0.089) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.058 (0.089) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.016 (0.089) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.045 (0.089) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.053 (0.089) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.150 (0.089) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.095 (0.089) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.008 (0.088) Loss 15.1994 (3.0437) -Test: [0/435] Time 6.870 (6.870) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.096 (0.690) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.017 (0.381) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.028 (0.275) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.126 (0.222) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.275 (0.196) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.068 (0.175) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.051 (0.160) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.067 (0.149) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.104 (0.140) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.009 (0.134) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.056 (0.128) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.056 (0.123) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.343 (0.122) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.017 (0.117) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.021 (0.112) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.077 (0.110) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.016 (0.107) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.075 (0.105) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.229 (0.104) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.078 (0.103) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.010 (0.101) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.074 (0.100) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.392 (0.101) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.031 (0.100) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.112 (0.100) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.048 (0.100) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.048 (0.100) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.043 (0.099) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.074 (0.098) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.070 (0.097) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.409 (0.097) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.018 (0.095) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.130 (0.094) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.068 (0.094) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.022 (0.094) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.056 (0.093) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.215 (0.093) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.028 (0.092) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.148 (0.092) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.214 (0.092) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.060 (0.091) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.039 (0.090) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.031 (0.089) Loss 15.1994 (3.0437) -Test: [0/435] Time 5.442 (5.442) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.036 (0.537) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.043 (0.301) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.019 (0.216) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.019 (0.174) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.263 (0.154) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.059 (0.138) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.050 (0.126) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.052 (0.119) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.165 (0.113) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.073 (0.108) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.036 (0.105) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.122 (0.103) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.053 (0.102) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.027 (0.099) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.029 (0.098) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.021 (0.096) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.008 (0.097) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.076 (0.096) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.046 (0.095) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.063 (0.094) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.012 (0.094) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.060 (0.093) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.058 (0.092) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.051 (0.091) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.018 (0.091) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.026 (0.090) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.078 (0.090) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.054 (0.089) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.007 (0.089) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.159 (0.089) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.042 (0.088) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.057 (0.088) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.007 (0.087) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.008 (0.087) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.450 (0.087) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.072 (0.088) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.440 (0.088) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.082 (0.089) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.049 (0.089) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.012 (0.088) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.206 (0.089) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.063 (0.088) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.008 (0.087) Loss 15.1994 (3.0437) -Test: [0/435] Time 5.698 (5.698) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.079 (0.564) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.037 (0.314) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.244 (0.233) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.091 (0.189) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.042 (0.165) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.060 (0.149) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.056 (0.138) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.041 (0.130) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.011 (0.124) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.038 (0.118) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.052 (0.113) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.050 (0.109) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.030 (0.107) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.049 (0.105) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.061 (0.102) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.055 (0.100) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.112 (0.099) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.037 (0.098) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.029 (0.096) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.056 (0.095) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.118 (0.094) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.042 (0.092) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.035 (0.091) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.011 (0.090) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.060 (0.089) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.091 (0.088) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.070 (0.088) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.019 (0.088) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.058 (0.088) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.047 (0.087) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.134 (0.087) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.020 (0.087) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.035 (0.087) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.020 (0.086) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.198 (0.088) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.060 (0.087) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.625 (0.088) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.045 (0.089) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.055 (0.089) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.044 (0.088) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.140 (0.089) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.039 (0.088) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.053 (0.088) Loss 15.1994 (3.0437) -Test: [0/435] Time 9.136 (9.136) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.036 (0.887) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.023 (0.498) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.053 (0.358) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.051 (0.285) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.093 (0.244) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.009 (0.219) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.035 (0.197) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.042 (0.183) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.049 (0.169) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.026 (0.161) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.039 (0.152) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.035 (0.145) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.069 (0.139) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.028 (0.135) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.019 (0.130) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.056 (0.126) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.059 (0.124) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.008 (0.120) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.048 (0.118) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.015 (0.116) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.048 (0.114) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.007 (0.113) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.047 (0.111) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.088 (0.110) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.064 (0.108) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.013 (0.107) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.047 (0.106) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.040 (0.104) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.047 (0.104) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.016 (0.103) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.104 (0.102) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.081 (0.100) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.037 (0.099) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.018 (0.097) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.040 (0.097) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.036 (0.097) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.043 (0.096) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.010 (0.096) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.041 (0.096) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.022 (0.095) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.028 (0.094) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.023 (0.092) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.021 (0.091) Loss 15.1994 (3.0437) -Test: [0/435] Time 25.538 (25.538) Loss 16.1776 (16.1776) -Test: [10/435] Time 0.054 (2.372) Loss 6.4917 (10.3521) -Test: [20/435] Time 0.057 (1.268) Loss 12.3808 (9.9254) -Test: [30/435] Time 0.038 (0.879) Loss 11.6694 (8.9681) -Test: [40/435] Time 0.183 (0.684) Loss 6.3404 (8.9616) -Test: [50/435] Time 0.043 (0.561) Loss 3.6336 (8.3316) -Test: [60/435] Time 0.046 (0.480) Loss 3.9536 (7.7392) -Test: [70/435] Time 0.019 (0.420) Loss 3.3502 (7.1193) -Test: [80/435] Time 0.026 (0.377) Loss 6.0276 (6.8229) -Test: [90/435] Time 0.036 (0.341) Loss 1.0226 (6.3190) -Test: [100/435] Time 0.073 (0.316) Loss 3.4627 (5.9706) -Test: [110/435] Time 0.012 (0.294) Loss 1.6063 (5.5589) -Test: [120/435] Time 0.037 (0.275) Loss 2.1700 (5.3224) -Test: [130/435] Time 0.051 (0.259) Loss 0.8041 (5.1280) -Test: [140/435] Time 0.112 (0.245) Loss 1.2991 (4.8480) -Test: [150/435] Time 0.028 (0.232) Loss 1.0001 (4.6611) -Test: [160/435] Time 0.258 (0.223) Loss 0.5237 (4.4133) -Test: [170/435] Time 0.092 (0.214) Loss 2.5127 (4.1924) -Test: [180/435] Time 0.051 (0.206) Loss 0.0318 (3.9884) -Test: [190/435] Time 0.047 (0.198) Loss 0.3755 (3.7934) -Test: [200/435] Time 0.017 (0.191) Loss 0.2047 (3.6944) -Test: [210/435] Time 0.039 (0.184) Loss 0.0844 (3.5449) -Test: [220/435] Time 0.009 (0.177) Loss 0.3334 (3.4047) -Test: [230/435] Time 0.031 (0.170) Loss 0.6197 (3.2783) -Test: [240/435] Time 0.029 (0.164) Loss 0.1382 (3.1502) -Test: [250/435] Time 0.015 (0.158) Loss 0.0165 (3.0375) -Test: [260/435] Time 0.007 (0.152) Loss 0.0569 (2.9494) -Test: [270/435] Time 0.009 (0.147) Loss 0.0465 (2.8582) -Test: [280/435] Time 0.008 (0.142) Loss 0.2513 (2.7741) -Test: [290/435] Time 0.008 (0.138) Loss 0.1948 (2.7013) -Test: [300/435] Time 0.021 (0.134) Loss 0.9285 (2.6284) -Test: [310/435] Time 0.010 (0.130) Loss 1.7760 (2.5962) -Test: [320/435] Time 0.016 (0.126) Loss 1.8297 (2.5530) -Test: [330/435] Time 0.008 (0.123) Loss 1.6630 (2.5133) -Test: [340/435] Time 0.010 (0.120) Loss 1.3866 (2.4980) -Test: [350/435] Time 0.007 (0.117) Loss 0.6672 (2.4928) -Test: [360/435] Time 0.027 (0.114) Loss 3.9431 (2.4898) -Test: [370/435] Time 0.013 (0.111) Loss 1.2136 (2.4881) -Test: [380/435] Time 0.030 (0.109) Loss 4.0488 (2.5172) -Test: [390/435] Time 0.015 (0.107) Loss 4.1887 (2.5334) -Test: [400/435] Time 0.035 (0.105) Loss 5.4225 (2.5933) -Test: [410/435] Time 0.007 (0.103) Loss 8.4955 (2.6541) -Test: [420/435] Time 0.008 (0.101) Loss 14.2771 (2.8044) -Test: [430/435] Time 0.010 (0.099) Loss 15.1994 (3.0437) -Epoch: [1][0/706] Time 7.839 (7.839) Data 7.798 (7.798) Loss 1.4970 (1.4970) -Epoch: [1][10/706] Time 0.125 (0.833) Data 0.000 (0.712) Loss 3.8214 (2.7582) -Epoch: [1][20/706] Time 0.198 (0.535) Data 0.000 (0.381) Loss 2.5412 (2.8027) -Epoch: [1][30/706] Time 0.172 (0.417) Data 0.017 (0.260) Loss 3.8904 (2.9484) -Epoch: [1][40/706] Time 0.142 (0.359) Data 0.011 (0.198) Loss 4.0310 (2.9444) -Epoch: [1][50/706] Time 0.114 (0.319) Data 0.004 (0.160) Loss 2.3922 (2.8472) -Epoch: [1][60/706] Time 0.231 (0.295) Data 0.039 (0.134) Loss 1.0998 (2.7565) -Epoch: [1][70/706] Time 0.208 (0.276) Data 0.000 (0.116) Loss 2.6197 (2.7470) -Epoch: [1][80/706] Time 0.144 (0.263) Data 0.000 (0.101) Loss 1.4638 (2.7022) -Epoch: [1][90/706] Time 0.190 (0.252) Data 0.000 (0.091) Loss 4.2018 (2.7500) -Epoch: [1][100/706] Time 0.131 (0.244) Data 0.010 (0.082) Loss 1.5148 (2.6978) -Epoch: [1][110/706] Time 0.319 (0.238) Data 0.000 (0.075) Loss 1.4033 (2.6559) -Epoch: [1][120/706] Time 0.122 (0.232) Data 0.000 (0.069) Loss 2.7707 (2.6695) -Epoch: [1][130/706] Time 0.130 (0.227) Data 0.000 (0.064) Loss 1.2440 (2.6385) -Epoch: [1][140/706] Time 0.169 (0.222) Data 0.000 (0.060) Loss 3.7605 (2.6685) -Epoch: [1][150/706] Time 0.147 (0.218) Data 0.000 (0.056) Loss 0.8513 (2.6784) -Epoch: [1][160/706] Time 0.127 (0.215) Data 0.000 (0.052) Loss 2.8558 (2.6817) -Epoch: [1][170/706] Time 0.298 (0.212) Data 0.006 (0.049) Loss 3.1246 (2.6956) -Epoch: [1][180/706] Time 0.303 (0.211) Data 0.021 (0.048) Loss 2.0919 (2.7123) -Epoch: [1][190/706] Time 0.141 (0.208) Data 0.000 (0.046) Loss 2.7661 (2.7332) -Epoch: [1][200/706] Time 0.175 (0.206) Data 0.000 (0.044) Loss 2.3881 (2.7239) -Epoch: [1][210/706] Time 0.067 (0.203) Data 0.005 (0.042) Loss 2.1491 (2.7134) -Epoch: [1][220/706] Time 0.111 (0.201) Data 0.015 (0.041) Loss 2.4109 (2.7112) -Epoch: [1][230/706] Time 0.154 (0.199) Data 0.000 (0.039) Loss 2.7420 (2.7465) -Epoch: [1][240/706] Time 0.175 (0.198) Data 0.000 (0.038) Loss 3.0747 (2.7379) -Epoch: [1][250/706] Time 0.190 (0.197) Data 0.000 (0.037) Loss 4.3585 (2.7545) -Epoch: [1][260/706] Time 0.133 (0.196) Data 0.000 (0.036) Loss 4.7377 (2.7393) -Epoch: [1][270/706] Time 0.161 (0.194) Data 0.000 (0.035) Loss 2.5427 (2.7409) -Epoch: [1][280/706] Time 0.041 (0.193) Data 0.012 (0.033) Loss 3.5825 (2.7395) -Epoch: [1][290/706] Time 0.167 (0.193) Data 0.000 (0.033) Loss 2.2422 (2.7504) -Epoch: [1][300/706] Time 0.289 (0.191) Data 0.000 (0.032) Loss 4.1080 (2.7401) -Epoch: [1][310/706] Time 0.065 (0.190) Data 0.005 (0.031) Loss 2.8406 (2.7394) -Epoch: [1][320/706] Time 0.134 (0.189) Data 0.000 (0.030) Loss 3.7425 (2.7470) -Epoch: [1][330/706] Time 0.209 (0.189) Data 0.051 (0.029) Loss 2.1833 (2.7570) -Epoch: [1][340/706] Time 0.146 (0.188) Data 0.014 (0.029) Loss 1.9368 (2.7528) -Epoch: [1][350/706] Time 0.114 (0.188) Data 0.002 (0.028) Loss 4.2374 (2.7586) -Epoch: [1][360/706] Time 0.147 (0.187) Data 0.000 (0.028) Loss 1.7814 (2.7556) -Epoch: [1][370/706] Time 0.139 (0.186) Data 0.010 (0.027) Loss 2.4780 (2.7540) -Epoch: [1][380/706] Time 0.181 (0.186) Data 0.000 (0.026) Loss 4.4340 (2.7657) -Epoch: [1][390/706] Time 0.147 (0.185) Data 0.006 (0.026) Loss 4.0507 (2.7714) -Epoch: [1][400/706] Time 0.348 (0.185) Data 0.003 (0.025) Loss 2.8494 (2.7727) -Epoch: [1][410/706] Time 0.170 (0.184) Data 0.000 (0.025) Loss 2.5537 (2.7687) -Epoch: [1][420/706] Time 0.164 (0.184) Data 0.006 (0.025) Loss 2.6789 (2.7726) -Epoch: [1][430/706] Time 0.174 (0.184) Data 0.000 (0.024) Loss 4.1589 (2.7655) -Epoch: [1][440/706] Time 0.147 (0.184) Data 0.000 (0.024) Loss 2.2839 (2.7617) -Epoch: [1][450/706] Time 0.173 (0.183) Data 0.017 (0.023) Loss 1.9115 (2.7582) -Epoch: [1][460/706] Time 0.259 (0.183) Data 0.000 (0.023) Loss 4.9441 (2.7561) -Epoch: [1][470/706] Time 0.191 (0.182) Data 0.000 (0.022) Loss 6.0829 (2.7611) -Epoch: [1][480/706] Time 0.161 (0.182) Data 0.059 (0.022) Loss 3.1898 (2.7529) -Epoch: [1][490/706] Time 0.146 (0.181) Data 0.000 (0.022) Loss 2.4645 (2.7539) -Epoch: [1][500/706] Time 0.194 (0.181) Data 0.000 (0.021) Loss 3.6184 (2.7673) -Epoch: [1][510/706] Time 0.154 (0.180) Data 0.000 (0.021) Loss 2.3011 (2.7689) -Epoch: [1][520/706] Time 0.168 (0.180) Data 0.000 (0.021) Loss 3.6966 (2.7675) -Epoch: [1][530/706] Time 0.142 (0.180) Data 0.019 (0.020) Loss 2.2218 (2.7628) -Epoch: [1][540/706] Time 0.219 (0.180) Data 0.013 (0.020) Loss 2.7425 (2.7665) -Epoch: [1][550/706] Time 0.146 (0.179) Data 0.000 (0.020) Loss 2.6945 (2.7692) -Epoch: [1][560/706] Time 0.137 (0.179) Data 0.000 (0.020) Loss 2.4023 (2.7653) -Epoch: [1][570/706] Time 0.163 (0.179) Data 0.005 (0.019) Loss 3.5916 (2.7625) -Epoch: [1][580/706] Time 0.163 (0.179) Data 0.000 (0.019) Loss 3.9644 (2.7611) -Epoch: [1][590/706] Time 0.176 (0.178) Data 0.017 (0.019) Loss 2.3585 (2.7577) -Epoch: [1][600/706] Time 0.080 (0.178) Data 0.000 (0.018) Loss 2.2000 (2.7543) -Epoch: [1][610/706] Time 0.309 (0.178) Data 0.014 (0.018) Loss 1.3961 (2.7591) -Epoch: [1][620/706] Time 0.255 (0.178) Data 0.000 (0.018) Loss 3.3538 (2.7555) -Epoch: [1][630/706] Time 0.181 (0.178) Data 0.004 (0.018) Loss 3.4288 (2.7545) -Epoch: [1][640/706] Time 0.196 (0.177) Data 0.000 (0.018) Loss 2.2273 (2.7506) -Epoch: [1][650/706] Time 0.139 (0.177) Data 0.000 (0.017) Loss 1.7347 (2.7467) -Epoch: [1][660/706] Time 0.137 (0.177) Data 0.000 (0.017) Loss 2.5334 (2.7465) -Epoch: [1][670/706] Time 0.115 (0.177) Data 0.006 (0.017) Loss 3.3579 (2.7456) -Epoch: [1][680/706] Time 0.183 (0.177) Data 0.010 (0.017) Loss 2.3344 (2.7503) -Epoch: [1][690/706] Time 0.215 (0.177) Data 0.000 (0.017) Loss 2.9407 (2.7509) -Epoch: [1][700/706] Time 0.026 (0.175) Data 0.000 (0.017) Loss 3.2773 (2.7521) -inside validate -Epoch: [1][0/706] Time 7.386 (7.386) Data 7.333 (7.333) Loss 2.9276 (2.9276) -Epoch: [1][10/706] Time 0.125 (1.170) Data 0.000 (0.670) Loss 4.4593 (3.0785) -Epoch: [1][20/706] Time 0.180 (0.701) Data 0.013 (0.358) Loss 1.5825 (2.9004) -Epoch: [1][30/706] Time 0.160 (0.534) Data 0.006 (0.243) Loss 3.0443 (2.9953) -Epoch: [1][40/706] Time 0.116 (0.446) Data 0.000 (0.184) Loss 3.6802 (2.9498) -Epoch: [1][50/706] Time 0.097 (0.392) Data 0.000 (0.149) Loss 2.5292 (2.9550) -Epoch: [1][60/706] Time 0.138 (0.354) Data 0.000 (0.126) Loss 2.1673 (2.9876) -Epoch: [1][70/706] Time 0.176 (0.328) Data 0.000 (0.109) Loss 2.7099 (2.9382) -Epoch: [1][80/706] Time 0.176 (0.309) Data 0.000 (0.096) Loss 3.3677 (2.9394) -Epoch: [1][90/706] Time 0.274 (0.294) Data 0.124 (0.088) Loss 2.6444 (2.9494) -Epoch: [1][100/706] Time 0.105 (0.280) Data 0.000 (0.080) Loss 2.5985 (2.9478) -Epoch: [1][110/706] Time 0.301 (0.270) Data 0.000 (0.073) Loss 5.8280 (2.9894) -Epoch: [1][120/706] Time 0.154 (0.261) Data 0.021 (0.067) Loss 2.5715 (2.9530) -Epoch: [1][130/706] Time 0.156 (0.254) Data 0.010 (0.063) Loss 3.0661 (2.9519) -Epoch: [1][140/706] Time 0.180 (0.248) Data 0.000 (0.059) Loss 2.2430 (2.9186) -Epoch: [1][150/706] Time 0.146 (0.242) Data 0.000 (0.055) Loss 3.0410 (2.9358) -Epoch: [1][160/706] Time 0.134 (0.239) Data 0.000 (0.052) Loss 4.0817 (2.9561) -Epoch: [1][170/706] Time 0.138 (0.233) Data 0.000 (0.049) Loss 2.9662 (2.9476) -Epoch: [1][180/706] Time 0.254 (0.232) Data 0.000 (0.047) Loss 2.6805 (2.9551) -Epoch: [1][190/706] Time 0.152 (0.227) Data 0.000 (0.045) Loss 3.7318 (2.9587) -Epoch: [1][200/706] Time 0.133 (0.224) Data 0.000 (0.043) Loss 2.3003 (2.9362) -Epoch: [1][210/706] Time 0.135 (0.221) Data 0.013 (0.041) Loss 1.8793 (2.9401) -Epoch: [1][220/706] Time 0.071 (0.218) Data 0.000 (0.039) Loss 1.7475 (2.9251) -Epoch: [1][230/706] Time 0.123 (0.215) Data 0.002 (0.038) Loss 1.2947 (2.9334) -Epoch: [1][240/706] Time 0.256 (0.213) Data 0.027 (0.036) Loss 2.2602 (2.9327) -Epoch: [1][250/706] Time 0.191 (0.211) Data 0.002 (0.035) Loss 1.9645 (2.9425) -Epoch: [1][260/706] Time 0.065 (0.210) Data 0.007 (0.034) Loss 2.2612 (2.9598) -Epoch: [1][270/706] Time 0.098 (0.207) Data 0.004 (0.034) Loss 2.3464 (2.9432) -Epoch: [1][280/706] Time 0.128 (0.206) Data 0.002 (0.033) Loss 4.0180 (2.9562) -Epoch: [1][290/706] Time 0.181 (0.205) Data 0.000 (0.032) Loss 2.8119 (2.9463) -Epoch: [1][300/706] Time 0.219 (0.204) Data 0.048 (0.031) Loss 1.7371 (2.9444) -Epoch: [1][310/706] Time 0.067 (0.202) Data 0.008 (0.030) Loss 2.4412 (2.9237) -Epoch: [1][320/706] Time 0.150 (0.201) Data 0.000 (0.029) Loss 3.5263 (2.9339) -Epoch: [1][330/706] Time 0.221 (0.200) Data 0.005 (0.029) Loss 3.5551 (2.9282) -Epoch: [1][340/706] Time 0.242 (0.199) Data 0.000 (0.028) Loss 2.2587 (2.9239) -Epoch: [1][350/706] Time 0.133 (0.198) Data 0.000 (0.027) Loss 2.1590 (2.9076) -Epoch: [1][360/706] Time 0.099 (0.197) Data 0.000 (0.027) Loss 2.6811 (2.8884) -Epoch: [1][370/706] Time 0.356 (0.196) Data 0.010 (0.026) Loss 2.8790 (2.8936) -Epoch: [1][380/706] Time 0.241 (0.195) Data 0.000 (0.026) Loss 2.1578 (2.8864) -Epoch: [1][390/706] Time 0.083 (0.194) Data 0.007 (0.025) Loss 2.8117 (2.8803) -Epoch: [1][400/706] Time 0.312 (0.194) Data 0.000 (0.025) Loss 3.3439 (2.8883) -Epoch: [1][410/706] Time 0.087 (0.193) Data 0.011 (0.024) Loss 2.4047 (2.8964) -Epoch: [1][420/706] Time 0.146 (0.192) Data 0.000 (0.024) Loss 2.0619 (2.8851) -Epoch: [1][430/706] Time 0.132 (0.192) Data 0.015 (0.023) Loss 4.4919 (2.8700) -Epoch: [1][440/706] Time 0.133 (0.192) Data 0.000 (0.023) Loss 2.3322 (2.8589) -Epoch: [1][450/706] Time 0.055 (0.191) Data 0.000 (0.022) Loss 1.6384 (2.8549) -Epoch: [1][460/706] Time 0.157 (0.191) Data 0.000 (0.022) Loss 3.6163 (2.8638) -Epoch: [1][470/706] Time 0.108 (0.190) Data 0.000 (0.021) Loss 4.7202 (2.8758) -Epoch: [1][480/706] Time 0.156 (0.189) Data 0.000 (0.021) Loss 2.6321 (2.8698) -Epoch: [1][490/706] Time 0.155 (0.189) Data 0.000 (0.021) Loss 1.8572 (2.8535) -Epoch: [1][500/706] Time 0.089 (0.188) Data 0.000 (0.020) Loss 3.8271 (2.8456) -Epoch: [1][510/706] Time 0.203 (0.187) Data 0.010 (0.020) Loss 3.0813 (2.8421) -Epoch: [1][520/706] Time 0.140 (0.187) Data 0.000 (0.020) Loss 2.8524 (2.8485) -Epoch: [1][530/706] Time 0.115 (0.187) Data 0.000 (0.019) Loss 2.6206 (2.8459) -Epoch: [1][540/706] Time 0.256 (0.187) Data 0.000 (0.019) Loss 3.0770 (2.8454) -Epoch: [1][550/706] Time 0.088 (0.186) Data 0.000 (0.019) Loss 1.2149 (2.8456) -Epoch: [1][560/706] Time 0.127 (0.186) Data 0.000 (0.019) Loss 2.7004 (2.8418) -Epoch: [1][570/706] Time 0.143 (0.185) Data 0.000 (0.018) Loss 3.6877 (2.8354) -Epoch: [1][580/706] Time 0.128 (0.185) Data 0.000 (0.018) Loss 2.0427 (2.8339) -Epoch: [1][590/706] Time 0.072 (0.185) Data 0.000 (0.018) Loss 4.1425 (2.8394) -Epoch: [1][600/706] Time 0.233 (0.185) Data 0.000 (0.018) Loss 2.7209 (2.8374) -Epoch: [1][610/706] Time 0.309 (0.184) Data 0.000 (0.017) Loss 3.4429 (2.8366) -Epoch: [1][620/706] Time 0.187 (0.184) Data 0.000 (0.017) Loss 3.9014 (2.8403) -Epoch: [1][630/706] Time 0.156 (0.184) Data 0.022 (0.017) Loss 3.1802 (2.8402) -Epoch: [1][640/706] Time 0.095 (0.183) Data 0.000 (0.017) Loss 3.9291 (2.8430) -Epoch: [1][650/706] Time 0.143 (0.183) Data 0.000 (0.017) Loss 3.2793 (2.8357) -Epoch: [1][660/706] Time 0.123 (0.182) Data 0.000 (0.016) Loss 2.1582 (2.8357) -Epoch: [1][670/706] Time 0.141 (0.182) Data 0.000 (0.016) Loss 2.5436 (2.8356) -Epoch: [1][680/706] Time 0.135 (0.182) Data 0.013 (0.016) Loss 2.7325 (2.8263) -Epoch: [1][690/706] Time 0.192 (0.182) Data 0.000 (0.016) Loss 2.8372 (2.8240) -Epoch: [1][700/706] Time 0.025 (0.180) Data 0.000 (0.016) Loss 1.5988 (2.8198) -inside validate -Epoch: [1][0/706] Time 8.172 (8.172) Data 8.135 (8.135) Loss 1.8691 (1.8691) -Epoch: [1][10/706] Time 0.137 (1.127) Data 0.010 (0.746) Loss 2.5244 (2.2988) -Epoch: [1][20/706] Time 0.176 (0.682) Data 0.008 (0.392) Loss 3.9815 (2.4213) -Epoch: [1][30/706] Time 0.247 (0.523) Data 0.000 (0.268) Loss 1.4299 (2.5581) -Epoch: [1][40/706] Time 0.277 (0.440) Data 0.000 (0.203) Loss 3.2377 (2.6313) -Epoch: [1][50/706] Time 0.134 (0.386) Data 0.000 (0.163) Loss 3.4636 (2.7408) -Epoch: [1][60/706] Time 0.218 (0.349) Data 0.050 (0.138) Loss 4.0224 (2.7315) -Epoch: [1][70/706] Time 0.168 (0.324) Data 0.000 (0.120) Loss 2.5272 (2.6682) -Epoch: [1][80/706] Time 0.147 (0.303) Data 0.015 (0.106) Loss 2.3523 (2.6971) -Epoch: [1][90/706] Time 0.149 (0.288) Data 0.000 (0.095) Loss 1.8953 (2.6984) -Epoch: [1][100/706] Time 0.071 (0.277) Data 0.000 (0.087) Loss 3.7283 (2.7417) -Epoch: [1][110/706] Time 0.061 (0.266) Data 0.007 (0.080) Loss 3.7512 (2.7377) -Epoch: [1][120/706] Time 0.141 (0.258) Data 0.000 (0.073) Loss 3.5630 (2.7146) -Epoch: [1][130/706] Time 0.098 (0.251) Data 0.015 (0.068) Loss 3.2608 (2.7197) -Epoch: [1][140/706] Time 0.162 (0.245) Data 0.008 (0.064) Loss 2.1765 (2.7038) -Epoch: [1][150/706] Time 0.152 (0.241) Data 0.004 (0.061) Loss 3.4354 (2.6952) -Epoch: [1][160/706] Time 0.128 (0.236) Data 0.000 (0.057) Loss 2.3135 (2.6959) -Epoch: [1][170/706] Time 0.146 (0.232) Data 0.000 (0.054) Loss 4.3729 (2.7070) -Epoch: [1][180/706] Time 0.120 (0.228) Data 0.010 (0.051) Loss 1.6344 (2.6955) -Epoch: [1][190/706] Time 0.130 (0.225) Data 0.017 (0.049) Loss 2.7072 (2.7189) -Epoch: [1][200/706] Time 0.200 (0.222) Data 0.000 (0.047) Loss 3.7376 (2.7420) -Epoch: [1][210/706] Time 0.061 (0.219) Data 0.000 (0.045) Loss 2.2764 (2.7343) -Epoch: [1][220/706] Time 0.200 (0.216) Data 0.001 (0.043) Loss 2.4411 (2.7243) -Epoch: [1][230/706] Time 0.087 (0.214) Data 0.005 (0.041) Loss 2.2979 (2.7212) -Epoch: [1][240/706] Time 0.188 (0.212) Data 0.020 (0.040) Loss 1.5351 (2.7460) -Epoch: [1][250/706] Time 0.203 (0.210) Data 0.000 (0.038) Loss 3.2823 (2.7463) -Epoch: [1][260/706] Time 0.146 (0.208) Data 0.041 (0.037) Loss 2.3084 (2.7554) -Epoch: [1][270/706] Time 0.158 (0.206) Data 0.000 (0.036) Loss 4.0953 (2.7827) -Epoch: [1][280/706] Time 0.102 (0.205) Data 0.003 (0.035) Loss 4.6031 (2.7845) -Epoch: [1][290/706] Time 0.197 (0.204) Data 0.000 (0.034) Loss 2.9450 (2.7897) -Epoch: [1][300/706] Time 0.254 (0.202) Data 0.005 (0.033) Loss 3.9542 (2.8002) -Epoch: [1][310/706] Time 0.093 (0.201) Data 0.010 (0.032) Loss 2.1256 (2.7953) -Epoch: [1][320/706] Time 0.143 (0.200) Data 0.000 (0.031) Loss 3.7112 (2.7896) -Epoch: [1][330/706] Time 0.177 (0.199) Data 0.000 (0.030) Loss 1.5947 (2.7920) -Epoch: [1][340/706] Time 0.275 (0.199) Data 0.000 (0.029) Loss 0.9572 (2.7778) -Epoch: [1][350/706] Time 0.078 (0.197) Data 0.000 (0.028) Loss 2.0956 (2.7757) -Epoch: [1][360/706] Time 0.134 (0.196) Data 0.019 (0.028) Loss 1.3265 (2.7782) -Epoch: [1][370/706] Time 0.081 (0.195) Data 0.000 (0.027) Loss 1.9289 (2.7799) -Epoch: [1][380/706] Time 0.205 (0.194) Data 0.009 (0.027) Loss 2.1615 (2.7823) -Epoch: [1][390/706] Time 0.158 (0.193) Data 0.000 (0.026) Loss 3.3651 (2.7833) -Epoch: [1][400/706] Time 0.303 (0.193) Data 0.000 (0.026) Loss 1.6468 (2.7886) -Epoch: [1][410/706] Time 0.161 (0.192) Data 0.000 (0.025) Loss 3.7928 (2.7940) -Epoch: [1][420/706] Time 0.138 (0.191) Data 0.015 (0.025) Loss 5.2301 (2.7964) -Epoch: [1][430/706] Time 0.233 (0.191) Data 0.000 (0.025) Loss 2.5495 (2.7875) -Epoch: [1][440/706] Time 0.098 (0.191) Data 0.000 (0.024) Loss 2.6987 (2.7904) -Epoch: [1][450/706] Time 0.090 (0.190) Data 0.000 (0.024) Loss 2.9283 (2.7936) -Epoch: [1][460/706] Time 0.214 (0.190) Data 0.000 (0.023) Loss 2.7819 (2.7923) -Epoch: [1][470/706] Time 0.149 (0.189) Data 0.032 (0.023) Loss 2.5404 (2.7875) -Epoch: [1][480/706] Time 0.240 (0.189) Data 0.007 (0.023) Loss 2.2970 (2.7948) -Epoch: [1][490/706] Time 0.153 (0.188) Data 0.000 (0.022) Loss 2.1309 (2.7962) -Epoch: [1][500/706] Time 0.150 (0.187) Data 0.004 (0.022) Loss 1.1430 (2.8027) -Epoch: [1][510/706] Time 0.150 (0.187) Data 0.000 (0.021) Loss 1.8856 (2.7975) -Epoch: [1][520/706] Time 0.285 (0.186) Data 0.006 (0.021) Loss 3.5290 (2.8036) -Epoch: [1][530/706] Time 0.068 (0.186) Data 0.000 (0.021) Loss 3.0544 (2.8019) -Epoch: [1][540/706] Time 0.215 (0.186) Data 0.000 (0.021) Loss 3.6192 (2.8036) -Epoch: [1][550/706] Time 0.164 (0.185) Data 0.000 (0.020) Loss 1.2971 (2.8028) -Epoch: [1][560/706] Time 0.214 (0.185) Data 0.000 (0.020) Loss 0.8718 (2.7976) -Epoch: [1][570/706] Time 0.116 (0.184) Data 0.000 (0.020) Loss 3.3693 (2.8029) -Epoch: [1][580/706] Time 0.190 (0.184) Data 0.020 (0.020) Loss 2.5732 (2.7965) -Epoch: [1][590/706] Time 0.148 (0.184) Data 0.026 (0.020) Loss 0.9345 (2.7951) -Epoch: [1][600/706] Time 0.165 (0.184) Data 0.000 (0.019) Loss 2.8028 (2.7953) -Epoch: [1][610/706] Time 0.120 (0.184) Data 0.009 (0.019) Loss 2.7986 (2.8078) -Epoch: [1][620/706] Time 0.159 (0.183) Data 0.000 (0.019) Loss 2.0212 (2.8040) -Epoch: [1][630/706] Time 0.182 (0.183) Data 0.000 (0.019) Loss 2.8470 (2.8090) -Epoch: [1][640/706] Time 0.169 (0.182) Data 0.010 (0.018) Loss 1.9418 (2.8146) -Epoch: [1][650/706] Time 0.188 (0.182) Data 0.000 (0.018) Loss 1.9739 (2.8071) -Epoch: [1][660/706] Time 0.175 (0.182) Data 0.000 (0.018) Loss 2.8421 (2.8064) -Epoch: [1][670/706] Time 0.195 (0.182) Data 0.012 (0.018) Loss 2.8870 (2.7988) -Epoch: [1][680/706] Time 0.053 (0.181) Data 0.000 (0.018) Loss 2.2218 (2.7992) -Epoch: [1][690/706] Time 0.109 (0.181) Data 0.000 (0.018) Loss 3.6717 (2.7936) -Epoch: [1][700/706] Time 0.026 (0.180) Data 0.000 (0.017) Loss 3.0761 (2.7900) -inside validate -Epoch: [1][0/706] Time 7.965 (7.965) Data 7.938 (7.938) Loss 2.9929 (2.9929) -Epoch: [1][10/706] Time 0.098 (1.158) Data 0.010 (0.729) Loss 4.2573 (3.1021) -Epoch: [1][20/706] Time 0.103 (0.704) Data 0.006 (0.384) Loss 2.2838 (2.8906) -Epoch: [1][30/706] Time 0.131 (0.532) Data 0.019 (0.271) Loss 4.1170 (2.8947) -Epoch: [1][40/706] Time 0.197 (0.448) Data 0.017 (0.206) Loss 2.4738 (2.9165) -Epoch: [1][50/706] Time 0.092 (0.392) Data 0.000 (0.166) Loss 3.2507 (2.8633) -Epoch: [1][60/706] Time 0.141 (0.354) Data 0.000 (0.140) Loss 3.7715 (2.8323) -Epoch: [1][70/706] Time 0.298 (0.329) Data 0.000 (0.120) Loss 3.9568 (2.8663) -Epoch: [1][80/706] Time 0.099 (0.308) Data 0.000 (0.107) Loss 2.1466 (2.9147) -Epoch: [1][90/706] Time 0.279 (0.294) Data 0.000 (0.096) Loss 2.5663 (2.9102) -Epoch: [1][100/706] Time 0.069 (0.281) Data 0.037 (0.087) Loss 2.3916 (2.8741) -Epoch: [1][110/706] Time 0.152 (0.270) Data 0.004 (0.081) Loss 1.5952 (2.8893) -Epoch: [1][120/706] Time 0.200 (0.262) Data 0.000 (0.075) Loss 2.9951 (2.8829) -Epoch: [1][130/706] Time 0.137 (0.254) Data 0.000 (0.069) Loss 2.1510 (2.8574) -Epoch: [1][140/706] Time 0.251 (0.249) Data 0.000 (0.065) Loss 3.4168 (2.8693) -Epoch: [1][150/706] Time 0.237 (0.242) Data 0.000 (0.061) Loss 1.9660 (2.8247) -Epoch: [1][160/706] Time 0.185 (0.238) Data 0.013 (0.057) Loss 2.6635 (2.8210) -Epoch: [1][170/706] Time 0.139 (0.233) Data 0.000 (0.054) Loss 2.3100 (2.7978) -Epoch: [1][180/706] Time 0.094 (0.230) Data 0.000 (0.051) Loss 3.2187 (2.7963) -Epoch: [1][190/706] Time 0.161 (0.227) Data 0.000 (0.049) Loss 3.6556 (2.7962) -Epoch: [1][200/706] Time 0.161 (0.224) Data 0.000 (0.047) Loss 2.8892 (2.8173) -Epoch: [1][210/706] Time 0.083 (0.221) Data 0.000 (0.045) Loss 1.5370 (2.8037) -Epoch: [1][220/706] Time 0.108 (0.218) Data 0.000 (0.043) Loss 2.4067 (2.8305) -Epoch: [1][230/706] Time 0.203 (0.215) Data 0.000 (0.041) Loss 2.5249 (2.8128) -Epoch: [1][240/706] Time 0.233 (0.213) Data 0.002 (0.040) Loss 3.9131 (2.7957) -Epoch: [1][250/706] Time 0.164 (0.211) Data 0.003 (0.038) Loss 1.3588 (2.7832) -Epoch: [1][260/706] Time 0.199 (0.210) Data 0.008 (0.037) Loss 3.8534 (2.7942) -Epoch: [1][270/706] Time 0.179 (0.208) Data 0.000 (0.036) Loss 3.8419 (2.8059) -Epoch: [1][280/706] Time 0.066 (0.206) Data 0.000 (0.034) Loss 1.4718 (2.7819) -Epoch: [1][290/706] Time 0.080 (0.205) Data 0.000 (0.033) Loss 1.9299 (2.7822) -Epoch: [1][300/706] Time 0.179 (0.203) Data 0.000 (0.033) Loss 2.3038 (2.7723) -Epoch: [1][310/706] Time 0.066 (0.202) Data 0.000 (0.032) Loss 1.9034 (2.7741) -Epoch: [1][320/706] Time 0.117 (0.201) Data 0.000 (0.031) Loss 3.6061 (2.7720) -Epoch: [1][330/706] Time 0.214 (0.200) Data 0.000 (0.030) Loss 2.4616 (2.7756) -Epoch: [1][340/706] Time 0.213 (0.199) Data 0.000 (0.029) Loss 5.7976 (2.7776) -Epoch: [1][350/706] Time 0.131 (0.198) Data 0.013 (0.029) Loss 2.0360 (2.7725) -Epoch: [1][360/706] Time 0.203 (0.197) Data 0.000 (0.028) Loss 2.7978 (2.7633) -Epoch: [1][370/706] Time 0.144 (0.196) Data 0.013 (0.027) Loss 2.4522 (2.7797) -Epoch: [1][380/706] Time 0.150 (0.195) Data 0.011 (0.027) Loss 1.6006 (2.7626) -Epoch: [1][390/706] Time 0.185 (0.194) Data 0.000 (0.026) Loss 4.8104 (2.7613) -Epoch: [1][400/706] Time 0.059 (0.194) Data 0.001 (0.026) Loss 4.7956 (2.7684) -Epoch: [1][410/706] Time 0.173 (0.193) Data 0.000 (0.026) Loss 2.4201 (2.7784) -Epoch: [1][420/706] Time 0.154 (0.193) Data 0.000 (0.025) Loss 1.5339 (2.7726) -Epoch: [1][430/706] Time 0.275 (0.192) Data 0.005 (0.025) Loss 3.3720 (2.7627) -Epoch: [1][440/706] Time 0.066 (0.192) Data 0.000 (0.024) Loss 3.2141 (2.7634) -Epoch: [1][450/706] Time 0.157 (0.191) Data 0.000 (0.024) Loss 1.6384 (2.7628) -Epoch: [1][460/706] Time 0.326 (0.191) Data 0.057 (0.024) Loss 1.9809 (2.7605) -Epoch: [1][470/706] Time 0.226 (0.190) Data 0.004 (0.023) Loss 2.6847 (2.7689) -Epoch: [1][480/706] Time 0.123 (0.189) Data 0.000 (0.023) Loss 3.4692 (2.7736) -Epoch: [1][490/706] Time 0.111 (0.189) Data 0.007 (0.022) Loss 3.0328 (2.7838) -Epoch: [1][500/706] Time 0.146 (0.188) Data 0.000 (0.022) Loss 1.8988 (2.7814) -Epoch: [1][510/706] Time 0.181 (0.188) Data 0.000 (0.021) Loss 2.6046 (2.7865) -Epoch: [1][520/706] Time 0.165 (0.187) Data 0.000 (0.021) Loss 1.6155 (2.7828) -Epoch: [1][530/706] Time 0.075 (0.187) Data 0.000 (0.021) Loss 2.7840 (2.7862) -Epoch: [1][540/706] Time 0.216 (0.187) Data 0.033 (0.021) Loss 0.5647 (2.7826) -Epoch: [1][550/706] Time 0.202 (0.186) Data 0.007 (0.020) Loss 2.2202 (2.7772) -Epoch: [1][560/706] Time 0.081 (0.186) Data 0.007 (0.020) Loss 2.3639 (2.7740) -Epoch: [1][570/706] Time 0.152 (0.185) Data 0.030 (0.020) Loss 3.3143 (2.7894) -Epoch: [1][580/706] Time 0.156 (0.185) Data 0.000 (0.020) Loss 4.0019 (2.7874) -Epoch: [1][590/706] Time 0.086 (0.185) Data 0.000 (0.020) Loss 2.4914 (2.7915) -Epoch: [1][600/706] Time 0.181 (0.185) Data 0.000 (0.019) Loss 1.7180 (2.7870) -Epoch: [1][610/706] Time 0.270 (0.184) Data 0.000 (0.019) Loss 2.3274 (2.7833) -Epoch: [1][620/706] Time 0.139 (0.184) Data 0.000 (0.019) Loss 1.9608 (2.7836) -Epoch: [1][630/706] Time 0.240 (0.184) Data 0.000 (0.019) Loss 2.5158 (2.7754) -Epoch: [1][640/706] Time 0.154 (0.183) Data 0.000 (0.018) Loss 3.1979 (2.7727) -Epoch: [1][650/706] Time 0.128 (0.183) Data 0.000 (0.018) Loss 3.1471 (2.7767) -Epoch: [1][660/706] Time 0.158 (0.182) Data 0.000 (0.018) Loss 1.3908 (2.7866) -Epoch: [1][670/706] Time 0.162 (0.182) Data 0.015 (0.018) Loss 3.7232 (2.7884) -Epoch: [1][680/706] Time 0.188 (0.182) Data 0.000 (0.018) Loss 4.2668 (2.7950) -Epoch: [1][690/706] Time 0.151 (0.182) Data 0.000 (0.017) Loss 2.9129 (2.7944) -Epoch: [1][700/706] Time 0.025 (0.180) Data 0.000 (0.017) Loss 3.7150 (2.7897) -inside validate -Epoch: [1][0/706] Time 7.600 (7.600) Data 7.530 (7.530) Loss 2.0233 (2.0233) -Epoch: [1][10/706] Time 0.083 (1.163) Data 0.004 (0.689) Loss 2.3627 (2.5885) -Epoch: [1][20/706] Time 0.093 (0.703) Data 0.000 (0.363) Loss 2.4825 (2.5702) -Epoch: [1][30/706] Time 0.200 (0.534) Data 0.017 (0.247) Loss 3.5206 (2.7114) -Epoch: [1][40/706] Time 0.148 (0.448) Data 0.000 (0.187) Loss 3.2703 (2.5875) -Epoch: [1][50/706] Time 0.144 (0.394) Data 0.000 (0.153) Loss 2.8004 (2.6854) -Epoch: [1][60/706] Time 0.127 (0.356) Data 0.000 (0.128) Loss 2.9912 (2.7619) -Epoch: [1][70/706] Time 0.243 (0.329) Data 0.000 (0.111) Loss 3.6869 (2.7813) -Epoch: [1][80/706] Time 0.053 (0.309) Data 0.000 (0.098) Loss 3.2545 (2.8323) -Epoch: [1][90/706] Time 0.078 (0.294) Data 0.000 (0.088) Loss 1.4959 (2.8188) -Epoch: [1][100/706] Time 0.036 (0.281) Data 0.000 (0.080) Loss 2.9103 (2.8362) -Epoch: [1][110/706] Time 0.302 (0.271) Data 0.007 (0.073) Loss 1.8333 (2.7771) -Epoch: [1][120/706] Time 0.173 (0.262) Data 0.024 (0.067) Loss 2.8809 (2.7846) -Epoch: [1][130/706] Time 0.204 (0.255) Data 0.014 (0.063) Loss 2.1370 (2.7686) -Epoch: [1][140/706] Time 0.281 (0.250) Data 0.000 (0.059) Loss 3.3004 (2.8070) -Epoch: [1][150/706] Time 0.169 (0.242) Data 0.011 (0.056) Loss 2.2475 (2.8227) -Epoch: [1][160/706] Time 0.129 (0.238) Data 0.000 (0.053) Loss 2.8708 (2.8351) -Epoch: [1][170/706] Time 0.190 (0.233) Data 0.012 (0.050) Loss 3.5116 (2.8401) -Epoch: [1][180/706] Time 0.225 (0.231) Data 0.013 (0.047) Loss 2.5201 (2.8410) -Epoch: [1][190/706] Time 0.104 (0.228) Data 0.000 (0.045) Loss 1.6411 (2.8133) -Epoch: [1][200/706] Time 0.213 (0.224) Data 0.000 (0.043) Loss 2.9047 (2.8065) -Epoch: [1][210/706] Time 0.092 (0.221) Data 0.000 (0.041) Loss 2.1656 (2.8055) -Epoch: [1][220/706] Time 0.076 (0.218) Data 0.008 (0.040) Loss 1.7587 (2.8078) -Epoch: [1][230/706] Time 0.136 (0.216) Data 0.000 (0.039) Loss 4.2237 (2.7941) -Epoch: [1][240/706] Time 0.210 (0.213) Data 0.004 (0.037) Loss 2.4514 (2.7963) -Epoch: [1][250/706] Time 0.167 (0.212) Data 0.000 (0.036) Loss 3.6589 (2.7949) -Epoch: [1][260/706] Time 0.138 (0.210) Data 0.000 (0.035) Loss 4.5676 (2.7841) -Epoch: [1][270/706] Time 0.142 (0.207) Data 0.005 (0.033) Loss 3.0379 (2.7778) -Epoch: [1][280/706] Time 0.180 (0.207) Data 0.000 (0.033) Loss 2.0022 (2.7708) -Epoch: [1][290/706] Time 0.185 (0.205) Data 0.000 (0.032) Loss 2.0787 (2.7586) -Epoch: [1][300/706] Time 0.083 (0.204) Data 0.023 (0.031) Loss 4.3004 (2.7756) -Epoch: [1][310/706] Time 0.209 (0.203) Data 0.000 (0.030) Loss 1.5033 (2.7720) -Epoch: [1][320/706] Time 0.186 (0.201) Data 0.078 (0.030) Loss 1.9281 (2.7520) -Epoch: [1][330/706] Time 0.148 (0.200) Data 0.000 (0.029) Loss 3.9858 (2.7553) -Epoch: [1][340/706] Time 0.086 (0.199) Data 0.000 (0.028) Loss 2.7909 (2.7527) -Epoch: [1][350/706] Time 0.157 (0.198) Data 0.023 (0.028) Loss 2.1509 (2.7668) -Epoch: [1][360/706] Time 0.190 (0.197) Data 0.011 (0.027) Loss 2.6735 (2.7716) -Epoch: [1][370/706] Time 0.148 (0.197) Data 0.000 (0.027) Loss 3.5404 (2.7769) -Epoch: [1][380/706] Time 0.104 (0.195) Data 0.000 (0.026) Loss 1.2944 (2.7858) -Epoch: [1][390/706] Time 0.080 (0.194) Data 0.000 (0.026) Loss 3.2209 (2.7796) -Epoch: [1][400/706] Time 0.332 (0.194) Data 0.000 (0.025) Loss 3.4945 (2.7733) -Epoch: [1][410/706] Time 0.289 (0.193) Data 0.000 (0.025) Loss 3.1314 (2.7866) -Epoch: [1][420/706] Time 0.126 (0.192) Data 0.000 (0.024) Loss 3.1050 (2.7823) -Epoch: [1][430/706] Time 0.204 (0.192) Data 0.000 (0.024) Loss 1.6548 (2.7718) -Epoch: [1][440/706] Time 0.080 (0.192) Data 0.000 (0.023) Loss 1.9408 (2.7719) -Epoch: [1][450/706] Time 0.157 (0.192) Data 0.000 (0.023) Loss 3.5182 (2.7631) -Epoch: [1][460/706] Time 0.144 (0.191) Data 0.013 (0.023) Loss 2.4596 (2.7705) -Epoch: [1][470/706] Time 0.131 (0.190) Data 0.000 (0.023) Loss 4.0671 (2.7808) -Epoch: [1][480/706] Time 0.219 (0.190) Data 0.003 (0.022) Loss 2.2991 (2.7809) -Epoch: [1][490/706] Time 0.124 (0.189) Data 0.004 (0.022) Loss 1.3342 (2.7686) -Epoch: [1][500/706] Time 0.179 (0.188) Data 0.000 (0.022) Loss 1.8959 (2.7620) -Epoch: [1][510/706] Time 0.111 (0.187) Data 0.000 (0.022) Loss 3.3576 (2.7675) -Epoch: [1][520/706] Time 0.132 (0.187) Data 0.000 (0.021) Loss 3.0397 (2.7626) -Epoch: [1][530/706] Time 0.196 (0.187) Data 0.000 (0.021) Loss 2.3533 (2.7682) -Epoch: [1][540/706] Time 0.264 (0.187) Data 0.017 (0.021) Loss 1.6353 (2.7676) -Epoch: [1][550/706] Time 0.238 (0.187) Data 0.000 (0.020) Loss 2.4834 (2.7704) -Epoch: [1][560/706] Time 0.149 (0.186) Data 0.015 (0.020) Loss 2.3309 (2.7704) -Epoch: [1][570/706] Time 0.168 (0.185) Data 0.000 (0.020) Loss 2.1342 (2.7647) -Epoch: [1][580/706] Time 0.154 (0.185) Data 0.000 (0.020) Loss 4.9859 (2.7661) -Epoch: [1][590/706] Time 0.151 (0.185) Data 0.000 (0.019) Loss 3.1977 (2.7729) -Epoch: [1][600/706] Time 0.170 (0.185) Data 0.001 (0.019) Loss 2.7716 (2.7777) -Epoch: [1][610/706] Time 0.375 (0.184) Data 0.000 (0.019) Loss 4.1654 (2.7849) -Epoch: [1][620/706] Time 0.124 (0.184) Data 0.000 (0.019) Loss 3.5024 (2.7888) -Epoch: [1][630/706] Time 0.150 (0.184) Data 0.000 (0.018) Loss 4.3528 (2.8006) -Epoch: [1][640/706] Time 0.198 (0.184) Data 0.000 (0.018) Loss 2.6179 (2.7963) -Epoch: [1][650/706] Time 0.141 (0.183) Data 0.000 (0.018) Loss 2.6837 (2.7951) -Epoch: [1][660/706] Time 0.151 (0.182) Data 0.000 (0.018) Loss 4.8596 (2.7903) -Epoch: [1][670/706] Time 0.080 (0.182) Data 0.000 (0.017) Loss 1.2609 (2.7891) -Epoch: [1][680/706] Time 0.026 (0.182) Data 0.000 (0.017) Loss 2.1239 (2.7810) -Epoch: [1][690/706] Time 0.191 (0.182) Data 0.000 (0.017) Loss 2.7035 (2.7728) -Epoch: [1][700/706] Time 0.023 (0.180) Data 0.000 (0.017) Loss 3.7018 (2.7789) -inside validate -Epoch: [1][0/706] Time 7.875 (7.875) Data 7.845 (7.845) Loss 1.5818 (1.5818) -Epoch: [1][10/706] Time 0.081 (1.152) Data 0.000 (0.721) Loss 2.2306 (3.0219) -Epoch: [1][20/706] Time 0.124 (0.697) Data 0.006 (0.381) Loss 2.3317 (2.9669) -Epoch: [1][30/706] Time 0.042 (0.530) Data 0.000 (0.259) Loss 3.5473 (3.0988) -Epoch: [1][40/706] Time 0.202 (0.446) Data 0.020 (0.196) Loss 2.4480 (3.1019) -Epoch: [1][50/706] Time 0.107 (0.390) Data 0.000 (0.160) Loss 3.7922 (3.1044) -Epoch: [1][60/706] Time 0.138 (0.355) Data 0.000 (0.135) Loss 3.4316 (3.0832) -Epoch: [1][70/706] Time 0.102 (0.327) Data 0.006 (0.116) Loss 3.7976 (3.0319) -Epoch: [1][80/706] Time 0.082 (0.307) Data 0.000 (0.102) Loss 3.1491 (3.0122) -Epoch: [1][90/706] Time 0.291 (0.293) Data 0.014 (0.092) Loss 3.2850 (2.9968) -Epoch: [1][100/706] Time 0.049 (0.280) Data 0.000 (0.083) Loss 2.6995 (2.9534) -Epoch: [1][110/706] Time 0.251 (0.269) Data 0.005 (0.076) Loss 4.6647 (2.9540) -Epoch: [1][120/706] Time 0.148 (0.262) Data 0.000 (0.071) Loss 3.0663 (2.9598) -Epoch: [1][130/706] Time 0.064 (0.254) Data 0.000 (0.066) Loss 2.2907 (2.9445) -Epoch: [1][140/706] Time 0.022 (0.247) Data 0.000 (0.061) Loss 1.6919 (2.9165) -Epoch: [1][150/706] Time 0.170 (0.242) Data 0.000 (0.058) Loss 3.5653 (2.9135) -Epoch: [1][160/706] Time 0.222 (0.238) Data 0.000 (0.054) Loss 3.5569 (2.9292) -Epoch: [1][170/706] Time 0.234 (0.233) Data 0.003 (0.051) Loss 2.5909 (2.9027) -Epoch: [1][180/706] Time 0.213 (0.231) Data 0.000 (0.049) Loss 2.5077 (2.8910) -Epoch: [1][190/706] Time 0.112 (0.227) Data 0.000 (0.047) Loss 3.1480 (2.8918) -Epoch: [1][200/706] Time 0.156 (0.223) Data 0.002 (0.045) Loss 3.4281 (2.8805) -Epoch: [1][210/706] Time 0.205 (0.221) Data 0.036 (0.043) Loss 2.9188 (2.8784) -Epoch: [1][220/706] Time 0.104 (0.217) Data 0.004 (0.042) Loss 2.2177 (2.8750) -Epoch: [1][230/706] Time 0.043 (0.214) Data 0.000 (0.040) Loss 2.1131 (2.8469) -Epoch: [1][240/706] Time 0.197 (0.213) Data 0.000 (0.039) Loss 4.5511 (2.8462) -Epoch: [1][250/706] Time 0.159 (0.211) Data 0.000 (0.037) Loss 1.3967 (2.8470) -Epoch: [1][260/706] Time 0.082 (0.210) Data 0.008 (0.036) Loss 2.9580 (2.8494) -Epoch: [1][270/706] Time 0.230 (0.207) Data 0.000 (0.035) Loss 1.8898 (2.8268) -Epoch: [1][280/706] Time 0.179 (0.207) Data 0.000 (0.034) Loss 1.9493 (2.8286) -Epoch: [1][290/706] Time 0.172 (0.205) Data 0.000 (0.033) Loss 3.3490 (2.8409) -Epoch: [1][300/706] Time 0.025 (0.203) Data 0.000 (0.032) Loss 4.0876 (2.8434) -Epoch: [1][310/706] Time 0.098 (0.202) Data 0.002 (0.031) Loss 4.7028 (2.8466) -Epoch: [1][320/706] Time 0.192 (0.200) Data 0.001 (0.030) Loss 3.7637 (2.8479) -Epoch: [1][330/706] Time 0.300 (0.200) Data 0.000 (0.029) Loss 6.1994 (2.8726) -Epoch: [1][340/706] Time 0.199 (0.199) Data 0.000 (0.029) Loss 3.2447 (2.8811) -Epoch: [1][350/706] Time 0.166 (0.198) Data 0.000 (0.028) Loss 2.4149 (2.8929) -Epoch: [1][360/706] Time 0.192 (0.197) Data 0.011 (0.027) Loss 3.2247 (2.8931) -Epoch: [1][370/706] Time 0.175 (0.196) Data 0.000 (0.027) Loss 2.3533 (2.8799) -Epoch: [1][380/706] Time 0.226 (0.195) Data 0.017 (0.026) Loss 2.5315 (2.8854) -Epoch: [1][390/706] Time 0.148 (0.194) Data 0.000 (0.025) Loss 3.4746 (2.8984) -Epoch: [1][400/706] Time 0.285 (0.193) Data 0.000 (0.025) Loss 1.9322 (2.8956) -Epoch: [1][410/706] Time 0.226 (0.193) Data 0.014 (0.024) Loss 4.3691 (2.8957) -Epoch: [1][420/706] Time 0.123 (0.192) Data 0.000 (0.024) Loss 2.9859 (2.8852) -Epoch: [1][430/706] Time 0.228 (0.192) Data 0.000 (0.024) Loss 1.1231 (2.8782) -Epoch: [1][440/706] Time 0.075 (0.192) Data 0.000 (0.023) Loss 2.7306 (2.8768) -Epoch: [1][450/706] Time 0.121 (0.191) Data 0.000 (0.023) Loss 2.2293 (2.8655) -Epoch: [1][460/706] Time 0.217 (0.191) Data 0.000 (0.022) Loss 1.5868 (2.8685) -Epoch: [1][470/706] Time 0.179 (0.190) Data 0.000 (0.022) Loss 1.8233 (2.8715) -Epoch: [1][480/706] Time 0.130 (0.189) Data 0.000 (0.021) Loss 3.9584 (2.8757) -Epoch: [1][490/706] Time 0.148 (0.189) Data 0.000 (0.021) Loss 2.1675 (2.8733) -Epoch: [1][500/706] Time 0.097 (0.188) Data 0.000 (0.021) Loss 3.5215 (2.8845) -Epoch: [1][510/706] Time 0.187 (0.187) Data 0.018 (0.020) Loss 2.0367 (2.8862) -Epoch: [1][520/706] Time 0.159 (0.187) Data 0.000 (0.020) Loss 2.6973 (2.8838) -Epoch: [1][530/706] Time 0.128 (0.187) Data 0.000 (0.020) Loss 5.1316 (2.8878) -Epoch: [1][540/706] Time 0.222 (0.186) Data 0.000 (0.019) Loss 4.7437 (2.8956) -Epoch: [1][550/706] Time 0.050 (0.186) Data 0.001 (0.019) Loss 2.2789 (2.8966) -Epoch: [1][560/706] Time 0.165 (0.186) Data 0.005 (0.019) Loss 2.4493 (2.8899) -Epoch: [1][570/706] Time 0.135 (0.185) Data 0.000 (0.019) Loss 4.6442 (2.8888) -Epoch: [1][580/706] Time 0.149 (0.185) Data 0.000 (0.018) Loss 2.6693 (2.8864) -Epoch: [1][590/706] Time 0.147 (0.185) Data 0.000 (0.018) Loss 3.0104 (2.8901) -Epoch: [1][600/706] Time 0.183 (0.184) Data 0.026 (0.018) Loss 0.9689 (2.8861) -Epoch: [1][610/706] Time 0.208 (0.184) Data 0.015 (0.018) Loss 1.9505 (2.8876) -Epoch: [1][620/706] Time 0.142 (0.184) Data 0.000 (0.018) Loss 1.8614 (2.8800) -Epoch: [1][630/706] Time 0.151 (0.184) Data 0.000 (0.017) Loss 1.9069 (2.8672) -Epoch: [1][640/706] Time 0.090 (0.183) Data 0.001 (0.017) Loss 4.4883 (2.8699) -Epoch: [1][650/706] Time 0.127 (0.183) Data 0.005 (0.017) Loss 4.4437 (2.8630) -Epoch: [1][660/706] Time 0.132 (0.182) Data 0.000 (0.017) Loss 1.8941 (2.8562) -Epoch: [1][670/706] Time 0.204 (0.182) Data 0.003 (0.017) Loss 2.2388 (2.8553) -Epoch: [1][680/706] Time 0.121 (0.182) Data 0.000 (0.017) Loss 3.3050 (2.8517) -Epoch: [1][690/706] Time 0.273 (0.182) Data 0.000 (0.017) Loss 2.9801 (2.8488) -Epoch: [1][700/706] Time 0.026 (0.180) Data 0.000 (0.016) Loss 2.7222 (2.8395) -inside validate -Epoch: [1][0/706] Time 9.140 (9.140) Data 9.095 (9.095) Loss 1.8743 (1.8743) -Epoch: [1][10/706] Time 0.126 (1.157) Data 0.000 (0.829) Loss 2.3971 (2.5783) -Epoch: [1][20/706] Time 0.092 (0.697) Data 0.000 (0.435) Loss 3.2651 (2.7625) -Epoch: [1][30/706] Time 0.138 (0.531) Data 0.007 (0.297) Loss 0.9824 (2.7733) -Epoch: [1][40/706] Time 0.147 (0.446) Data 0.000 (0.225) Loss 4.1958 (2.7878) -Epoch: [1][50/706] Time 0.180 (0.393) Data 0.000 (0.182) Loss 2.4730 (2.7922) -Epoch: [1][60/706] Time 0.157 (0.355) Data 0.000 (0.152) Loss 3.9957 (2.7630) -Epoch: [1][70/706] Time 0.218 (0.329) Data 0.003 (0.132) Loss 1.9863 (2.7301) -Epoch: [1][80/706] Time 0.205 (0.308) Data 0.010 (0.116) Loss 4.0899 (2.7491) -Epoch: [1][90/706] Time 0.148 (0.293) Data 0.001 (0.104) Loss 2.1512 (2.7322) -Epoch: [1][100/706] Time 0.063 (0.280) Data 0.000 (0.094) Loss 2.0056 (2.7513) -Epoch: [1][110/706] Time 0.158 (0.270) Data 0.000 (0.086) Loss 4.0647 (2.7338) -Epoch: [1][120/706] Time 0.154 (0.262) Data 0.000 (0.079) Loss 3.7097 (2.7137) -Epoch: [1][130/706] Time 0.125 (0.253) Data 0.000 (0.073) Loss 2.2076 (2.7071) -Epoch: [1][140/706] Time 0.174 (0.248) Data 0.000 (0.068) Loss 1.8095 (2.7169) -Epoch: [1][150/706] Time 0.276 (0.243) Data 0.000 (0.064) Loss 4.0658 (2.7514) -Epoch: [1][160/706] Time 0.152 (0.238) Data 0.000 (0.060) Loss 5.7526 (2.7868) -Epoch: [1][170/706] Time 0.272 (0.233) Data 0.002 (0.058) Loss 1.7649 (2.7855) -Epoch: [1][180/706] Time 0.226 (0.230) Data 0.008 (0.055) Loss 1.9856 (2.7728) -Epoch: [1][190/706] Time 0.219 (0.228) Data 0.000 (0.052) Loss 4.8110 (2.7949) -Epoch: [1][200/706] Time 0.133 (0.223) Data 0.000 (0.049) Loss 2.6968 (2.7917) -Epoch: [1][210/706] Time 0.114 (0.221) Data 0.000 (0.048) Loss 2.2982 (2.7975) -Epoch: [1][220/706] Time 0.176 (0.218) Data 0.000 (0.046) Loss 2.7444 (2.7810) -Epoch: [1][230/706] Time 0.118 (0.215) Data 0.001 (0.044) Loss 2.3349 (2.7752) -Epoch: [1][240/706] Time 0.261 (0.213) Data 0.000 (0.043) Loss 4.5976 (2.7879) -Epoch: [1][250/706] Time 0.197 (0.211) Data 0.000 (0.041) Loss 4.2362 (2.7817) -Epoch: [1][260/706] Time 0.128 (0.210) Data 0.035 (0.040) Loss 4.3060 (2.7879) -Epoch: [1][270/706] Time 0.145 (0.208) Data 0.000 (0.039) Loss 2.6125 (2.7644) -Epoch: [1][280/706] Time 0.209 (0.206) Data 0.000 (0.037) Loss 1.6084 (2.7509) -Epoch: [1][290/706] Time 0.175 (0.205) Data 0.007 (0.036) Loss 2.4064 (2.7410) -Epoch: [1][300/706] Time 0.218 (0.203) Data 0.000 (0.035) Loss 4.1552 (2.7566) -Epoch: [1][310/706] Time 0.096 (0.202) Data 0.015 (0.034) Loss 4.2839 (2.7589) -Epoch: [1][320/706] Time 0.130 (0.201) Data 0.000 (0.033) Loss 3.4151 (2.7523) -Epoch: [1][330/706] Time 0.121 (0.200) Data 0.000 (0.032) Loss 1.7756 (2.7457) -Epoch: [1][340/706] Time 0.158 (0.199) Data 0.000 (0.032) Loss 4.6262 (2.7521) -Epoch: [1][350/706] Time 0.112 (0.198) Data 0.010 (0.031) Loss 4.0016 (2.7732) -Epoch: [1][360/706] Time 0.158 (0.197) Data 0.008 (0.030) Loss 2.7264 (2.7737) -Epoch: [1][370/706] Time 0.241 (0.196) Data 0.008 (0.030) Loss 4.9696 (2.7747) -Epoch: [1][380/706] Time 0.078 (0.195) Data 0.013 (0.029) Loss 1.5298 (2.7739) -Epoch: [1][390/706] Time 0.141 (0.194) Data 0.000 (0.028) Loss 5.0953 (2.7707) -Epoch: [1][400/706] Time 0.298 (0.193) Data 0.004 (0.028) Loss 2.6838 (2.7751) -Epoch: [1][410/706] Time 0.172 (0.193) Data 0.000 (0.027) Loss 3.7518 (2.7790) -Epoch: [1][420/706] Time 0.262 (0.192) Data 0.000 (0.027) Loss 3.5482 (2.7752) -Epoch: [1][430/706] Time 0.201 (0.192) Data 0.000 (0.026) Loss 2.6669 (2.7763) -Epoch: [1][440/706] Time 0.176 (0.192) Data 0.000 (0.026) Loss 2.8985 (2.7870) -Epoch: [1][450/706] Time 0.170 (0.191) Data 0.000 (0.025) Loss 2.4100 (2.7824) -Epoch: [1][460/706] Time 0.215 (0.191) Data 0.000 (0.025) Loss 2.2517 (2.7768) -Epoch: [1][470/706] Time 0.143 (0.190) Data 0.000 (0.024) Loss 2.7022 (2.7771) -Epoch: [1][480/706] Time 0.187 (0.189) Data 0.024 (0.024) Loss 1.6251 (2.7826) -Epoch: [1][490/706] Time 0.137 (0.188) Data 0.000 (0.024) Loss 2.2819 (2.7837) -Epoch: [1][500/706] Time 0.137 (0.188) Data 0.000 (0.023) Loss 3.9802 (2.7780) -Epoch: [1][510/706] Time 0.152 (0.187) Data 0.000 (0.023) Loss 2.1689 (2.7796) -Epoch: [1][520/706] Time 0.172 (0.187) Data 0.000 (0.022) Loss 2.1648 (2.7805) -Epoch: [1][530/706] Time 0.147 (0.187) Data 0.000 (0.022) Loss 2.1394 (2.7728) -Epoch: [1][540/706] Time 0.166 (0.186) Data 0.065 (0.022) Loss 1.7170 (2.7713) -Epoch: [1][550/706] Time 0.092 (0.186) Data 0.000 (0.021) Loss 2.4311 (2.7667) -Epoch: [1][560/706] Time 0.144 (0.186) Data 0.000 (0.021) Loss 1.7246 (2.7622) -Epoch: [1][570/706] Time 0.162 (0.185) Data 0.000 (0.021) Loss 2.7124 (2.7629) -Epoch: [1][580/706] Time 0.145 (0.185) Data 0.000 (0.021) Loss 3.9032 (2.7625) -Epoch: [1][590/706] Time 0.179 (0.185) Data 0.007 (0.020) Loss 3.0152 (2.7607) -Epoch: [1][600/706] Time 0.194 (0.184) Data 0.000 (0.020) Loss 3.7052 (2.7624) -Epoch: [1][610/706] Time 0.167 (0.184) Data 0.006 (0.020) Loss 2.7110 (2.7615) -Epoch: [1][620/706] Time 0.139 (0.184) Data 0.007 (0.019) Loss 3.1074 (2.7585) -Epoch: [1][630/706] Time 0.262 (0.184) Data 0.000 (0.019) Loss 2.3018 (2.7520) -Epoch: [1][640/706] Time 0.171 (0.183) Data 0.006 (0.019) Loss 2.1350 (2.7480) -Epoch: [1][650/706] Time 0.149 (0.183) Data 0.000 (0.019) Loss 1.7519 (2.7498) -Epoch: [1][660/706] Time 0.182 (0.182) Data 0.013 (0.018) Loss 3.9452 (2.7507) -Epoch: [1][670/706] Time 0.171 (0.182) Data 0.000 (0.018) Loss 2.2518 (2.7542) -Epoch: [1][680/706] Time 0.202 (0.182) Data 0.084 (0.018) Loss 2.8007 (2.7492) -Epoch: [1][690/706] Time 0.159 (0.182) Data 0.000 (0.018) Loss 3.0325 (2.7480) -Epoch: [1][700/706] Time 0.024 (0.180) Data 0.000 (0.018) Loss 3.4793 (2.7432) -inside validate -Epoch: [1][0/706] Time 7.668 (7.668) Data 7.629 (7.629) Loss 4.1692 (4.1692) -Epoch: [1][10/706] Time 0.079 (1.167) Data 0.001 (0.699) Loss 2.1916 (3.0725) -Epoch: [1][20/706] Time 0.094 (0.703) Data 0.000 (0.367) Loss 2.6397 (3.0716) -Epoch: [1][30/706] Time 0.130 (0.534) Data 0.000 (0.249) Loss 1.6454 (2.8984) -Epoch: [1][40/706] Time 0.185 (0.453) Data 0.000 (0.189) Loss 3.3467 (2.9965) -Epoch: [1][50/706] Time 0.143 (0.392) Data 0.013 (0.153) Loss 3.0622 (2.9972) -Epoch: [1][60/706] Time 0.139 (0.355) Data 0.000 (0.129) Loss 1.8846 (2.9747) -Epoch: [1][70/706] Time 0.133 (0.329) Data 0.000 (0.111) Loss 2.0375 (2.9092) -Epoch: [1][80/706] Time 0.118 (0.309) Data 0.000 (0.097) Loss 2.1848 (2.8628) -Epoch: [1][90/706] Time 0.162 (0.293) Data 0.009 (0.088) Loss 3.7628 (2.7921) -Epoch: [1][100/706] Time 0.079 (0.281) Data 0.000 (0.080) Loss 2.7192 (2.8360) -Epoch: [1][110/706] Time 0.282 (0.270) Data 0.010 (0.073) Loss 1.2978 (2.8142) -Epoch: [1][120/706] Time 0.070 (0.262) Data 0.000 (0.067) Loss 3.7067 (2.7758) -Epoch: [1][130/706] Time 0.119 (0.254) Data 0.000 (0.062) Loss 1.1144 (2.7559) -Epoch: [1][140/706] Time 0.107 (0.248) Data 0.000 (0.058) Loss 2.9481 (2.7339) -Epoch: [1][150/706] Time 0.274 (0.242) Data 0.005 (0.055) Loss 2.7959 (2.7542) -Epoch: [1][160/706] Time 0.123 (0.238) Data 0.002 (0.052) Loss 3.6224 (2.7626) -Epoch: [1][170/706] Time 0.277 (0.234) Data 0.011 (0.049) Loss 3.5945 (2.7557) -Epoch: [1][180/706] Time 0.228 (0.231) Data 0.008 (0.047) Loss 3.4475 (2.7882) -Epoch: [1][190/706] Time 0.186 (0.228) Data 0.003 (0.045) Loss 2.3373 (2.7923) -Epoch: [1][200/706] Time 0.169 (0.224) Data 0.005 (0.043) Loss 3.6264 (2.7751) -Epoch: [1][210/706] Time 0.112 (0.221) Data 0.000 (0.041) Loss 2.5371 (2.7565) -Epoch: [1][220/706] Time 0.159 (0.219) Data 0.000 (0.039) Loss 3.6955 (2.7646) -Epoch: [1][230/706] Time 0.115 (0.215) Data 0.008 (0.038) Loss 3.1402 (2.7496) -Epoch: [1][240/706] Time 0.148 (0.213) Data 0.000 (0.036) Loss 3.9417 (2.7413) -Epoch: [1][250/706] Time 0.239 (0.212) Data 0.010 (0.035) Loss 2.4285 (2.7338) -Epoch: [1][260/706] Time 0.167 (0.210) Data 0.000 (0.034) Loss 3.0744 (2.7333) -Epoch: [1][270/706] Time 0.284 (0.208) Data 0.000 (0.033) Loss 3.1270 (2.7524) -Epoch: [1][280/706] Time 0.219 (0.207) Data 0.000 (0.032) Loss 3.5401 (2.7658) -Epoch: [1][290/706] Time 0.115 (0.205) Data 0.000 (0.031) Loss 1.6758 (2.7775) -Epoch: [1][300/706] Time 0.207 (0.204) Data 0.006 (0.030) Loss 3.3397 (2.7883) -Epoch: [1][310/706] Time 0.200 (0.203) Data 0.000 (0.030) Loss 3.8018 (2.7750) -Epoch: [1][320/706] Time 0.135 (0.201) Data 0.000 (0.029) Loss 3.3675 (2.7676) -Epoch: [1][330/706] Time 0.192 (0.201) Data 0.005 (0.028) Loss 2.9357 (2.7619) -Epoch: [1][340/706] Time 0.109 (0.199) Data 0.000 (0.027) Loss 1.7684 (2.7636) -Epoch: [1][350/706] Time 0.109 (0.198) Data 0.000 (0.027) Loss 3.1795 (2.7588) -Epoch: [1][360/706] Time 0.234 (0.197) Data 0.005 (0.026) Loss 4.8257 (2.7714) -Epoch: [1][370/706] Time 0.103 (0.196) Data 0.013 (0.026) Loss 3.8567 (2.7713) -Epoch: [1][380/706] Time 0.236 (0.195) Data 0.006 (0.025) Loss 2.2547 (2.7662) -Epoch: [1][390/706] Time 0.141 (0.194) Data 0.000 (0.025) Loss 1.8532 (2.7625) -Epoch: [1][400/706] Time 0.306 (0.194) Data 0.001 (0.024) Loss 2.6702 (2.7502) -Epoch: [1][410/706] Time 0.180 (0.193) Data 0.002 (0.023) Loss 3.8972 (2.7502) -Epoch: [1][420/706] Time 0.185 (0.192) Data 0.000 (0.023) Loss 3.8072 (2.7579) -Epoch: [1][430/706] Time 0.224 (0.192) Data 0.000 (0.023) Loss 1.6312 (2.7446) -Epoch: [1][440/706] Time 0.177 (0.192) Data 0.003 (0.022) Loss 4.4454 (2.7412) -Epoch: [1][450/706] Time 0.134 (0.192) Data 0.000 (0.022) Loss 1.8144 (2.7341) -Epoch: [1][460/706] Time 0.269 (0.191) Data 0.000 (0.021) Loss 3.5565 (2.7381) -Epoch: [1][470/706] Time 0.182 (0.190) Data 0.000 (0.021) Loss 0.7889 (2.7269) -Epoch: [1][480/706] Time 0.162 (0.189) Data 0.036 (0.021) Loss 2.2826 (2.7314) -Epoch: [1][490/706] Time 0.134 (0.189) Data 0.000 (0.020) Loss 3.7174 (2.7381) -Epoch: [1][500/706] Time 0.186 (0.188) Data 0.000 (0.020) Loss 3.1611 (2.7419) -Epoch: [1][510/706] Time 0.154 (0.187) Data 0.074 (0.020) Loss 1.8019 (2.7396) -Epoch: [1][520/706] Time 0.173 (0.187) Data 0.000 (0.020) Loss 1.9572 (2.7405) -Epoch: [1][530/706] Time 0.162 (0.187) Data 0.002 (0.019) Loss 2.6166 (2.7443) -Epoch: [1][540/706] Time 0.143 (0.187) Data 0.001 (0.019) Loss 2.7315 (2.7549) -Epoch: [1][550/706] Time 0.138 (0.186) Data 0.000 (0.019) Loss 1.0470 (2.7529) -Epoch: [1][560/706] Time 0.125 (0.186) Data 0.000 (0.019) Loss 2.0568 (2.7512) -Epoch: [1][570/706] Time 0.301 (0.186) Data 0.006 (0.018) Loss 3.0265 (2.7460) -Epoch: [1][580/706] Time 0.164 (0.185) Data 0.000 (0.018) Loss 2.6263 (2.7434) -Epoch: [1][590/706] Time 0.281 (0.185) Data 0.000 (0.018) Loss 1.7456 (2.7489) -Epoch: [1][600/706] Time 0.101 (0.185) Data 0.000 (0.018) Loss 3.7756 (2.7606) -Epoch: [1][610/706] Time 0.188 (0.184) Data 0.000 (0.017) Loss 5.3003 (2.7681) -Epoch: [1][620/706] Time 0.232 (0.184) Data 0.022 (0.017) Loss 3.1824 (2.7818) -Epoch: [1][630/706] Time 0.166 (0.184) Data 0.000 (0.017) Loss 3.2932 (2.7796) -Epoch: [1][640/706] Time 0.172 (0.183) Data 0.000 (0.017) Loss 2.4397 (2.7775) -Epoch: [1][650/706] Time 0.132 (0.183) Data 0.011 (0.017) Loss 1.8975 (2.7808) -Epoch: [1][660/706] Time 0.144 (0.183) Data 0.000 (0.017) Loss 1.8997 (2.7792) -Epoch: [1][670/706] Time 0.313 (0.183) Data 0.019 (0.016) Loss 1.4331 (2.7763) -Epoch: [1][680/706] Time 0.095 (0.182) Data 0.027 (0.016) Loss 2.0639 (2.7745) -Epoch: [1][690/706] Time 0.096 (0.182) Data 0.000 (0.016) Loss 1.1622 (2.7728) -Epoch: [1][700/706] Time 0.026 (0.180) Data 0.000 (0.016) Loss 2.8812 (2.7707) -inside validate -Test: [0/435] Time 5.440 (5.440) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.026 (0.527) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.019 (0.294) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.018 (0.208) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.028 (0.166) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.134 (0.146) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.013 (0.127) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.041 (0.116) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.074 (0.108) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.040 (0.101) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.066 (0.096) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.008 (0.092) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.039 (0.091) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.042 (0.088) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.032 (0.086) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.066 (0.085) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.051 (0.083) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.044 (0.082) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.024 (0.083) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.088 (0.083) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.038 (0.083) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.017 (0.082) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.026 (0.083) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.062 (0.082) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.030 (0.083) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.084 (0.082) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.050 (0.082) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.055 (0.082) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.019 (0.081) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.010 (0.081) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.057 (0.081) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.067 (0.081) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.030 (0.082) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.046 (0.081) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.008 (0.082) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.043 (0.082) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.031 (0.082) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.047 (0.082) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.008 (0.082) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.013 (0.082) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.048 (0.081) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.091 (0.081) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.046 (0.081) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.029 (0.080) Loss 15.8373 (2.7909) -Test: [0/435] Time 5.979 (5.979) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.016 (0.580) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.042 (0.318) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.015 (0.228) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.029 (0.181) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.021 (0.153) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.051 (0.137) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.024 (0.123) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.035 (0.112) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.008 (0.105) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.039 (0.099) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.011 (0.095) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.132 (0.093) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.009 (0.090) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.073 (0.087) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.041 (0.086) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.051 (0.086) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.052 (0.084) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.008 (0.083) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.106 (0.083) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.057 (0.082) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.052 (0.082) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.033 (0.083) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.059 (0.082) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.072 (0.082) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.034 (0.082) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.040 (0.082) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.036 (0.082) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.121 (0.082) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.037 (0.081) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.033 (0.081) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.063 (0.081) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.279 (0.082) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.115 (0.082) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.023 (0.083) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.039 (0.082) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.457 (0.084) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.029 (0.083) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.050 (0.084) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.051 (0.083) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.194 (0.084) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.067 (0.084) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.099 (0.083) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.027 (0.082) Loss 15.8373 (2.7909) -Test: [0/435] Time 6.012 (6.012) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.023 (0.584) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.016 (0.324) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.018 (0.230) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.120 (0.182) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.034 (0.153) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.016 (0.134) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.043 (0.122) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.169 (0.115) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.029 (0.107) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.039 (0.100) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.048 (0.095) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.156 (0.093) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.034 (0.090) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.056 (0.089) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.049 (0.087) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.451 (0.088) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.029 (0.087) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.062 (0.086) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.028 (0.085) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.369 (0.086) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.063 (0.085) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.038 (0.085) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.086 (0.086) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.185 (0.087) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.019 (0.086) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.023 (0.086) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.060 (0.086) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.223 (0.087) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.055 (0.086) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.014 (0.086) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.079 (0.086) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.109 (0.086) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.048 (0.086) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.023 (0.086) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.048 (0.086) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.100 (0.086) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.064 (0.085) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.014 (0.086) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.062 (0.086) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.271 (0.085) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.121 (0.085) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.028 (0.084) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.028 (0.084) Loss 15.8373 (2.7909) -Test: [0/435] Time 5.529 (5.529) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.014 (0.529) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.012 (0.290) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.076 (0.208) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.014 (0.165) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.033 (0.141) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.038 (0.125) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.151 (0.114) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.034 (0.105) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.018 (0.097) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.045 (0.091) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.044 (0.088) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.241 (0.086) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.042 (0.085) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.065 (0.084) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.043 (0.085) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.192 (0.083) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.056 (0.083) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.046 (0.083) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.049 (0.082) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.336 (0.084) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.052 (0.085) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.045 (0.084) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.093 (0.084) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.415 (0.085) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.059 (0.086) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.071 (0.086) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.079 (0.085) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.260 (0.087) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.058 (0.087) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.033 (0.086) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.062 (0.087) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.450 (0.087) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.058 (0.087) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.091 (0.087) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.057 (0.086) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.120 (0.087) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.013 (0.086) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.051 (0.086) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.057 (0.086) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.250 (0.086) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.030 (0.086) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.051 (0.085) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.043 (0.084) Loss 15.8373 (2.7909) -Test: [0/435] Time 7.929 (7.929) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.099 (0.776) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.048 (0.422) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.040 (0.300) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.033 (0.238) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.164 (0.202) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.017 (0.179) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.041 (0.164) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.033 (0.150) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.293 (0.143) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.054 (0.135) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.070 (0.129) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.056 (0.125) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.050 (0.122) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.007 (0.117) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.081 (0.118) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.066 (0.115) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.027 (0.113) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.035 (0.111) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.209 (0.109) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.140 (0.107) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.046 (0.106) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.041 (0.104) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.223 (0.104) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.153 (0.103) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.062 (0.101) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.026 (0.100) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.008 (0.099) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.271 (0.099) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.068 (0.098) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.055 (0.097) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.050 (0.097) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.069 (0.097) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.064 (0.096) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.013 (0.096) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.032 (0.096) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.032 (0.096) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.016 (0.095) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.041 (0.094) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.009 (0.093) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.036 (0.092) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.029 (0.091) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.010 (0.090) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.008 (0.088) Loss 15.8373 (2.7909) -Test: [0/435] Time 12.652 (12.652) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.037 (1.199) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.040 (0.655) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.030 (0.465) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.091 (0.370) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.034 (0.310) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.040 (0.271) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.022 (0.247) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.067 (0.226) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.073 (0.211) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.060 (0.200) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.050 (0.192) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.068 (0.181) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.040 (0.173) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.048 (0.168) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.009 (0.162) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.020 (0.156) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.034 (0.151) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.038 (0.146) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.017 (0.143) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.046 (0.139) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.042 (0.135) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.063 (0.132) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.044 (0.130) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.028 (0.128) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.065 (0.126) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.071 (0.124) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.044 (0.122) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.082 (0.121) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.032 (0.119) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.063 (0.117) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.030 (0.115) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.031 (0.113) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.020 (0.112) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.020 (0.110) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.040 (0.108) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.084 (0.106) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.011 (0.104) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.021 (0.102) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.026 (0.100) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.008 (0.098) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.010 (0.096) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.010 (0.094) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.007 (0.093) Loss 15.8373 (2.7909) -Test: [0/435] Time 10.482 (10.482) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.036 (1.017) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.032 (0.564) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.064 (0.401) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.465 (0.329) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.051 (0.279) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.018 (0.245) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.058 (0.223) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.281 (0.206) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.064 (0.192) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.052 (0.181) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.053 (0.171) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.260 (0.165) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.061 (0.157) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.068 (0.152) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.027 (0.145) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.322 (0.143) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.032 (0.141) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.112 (0.137) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.055 (0.133) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.352 (0.132) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.054 (0.130) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.058 (0.128) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.081 (0.127) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.043 (0.125) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.186 (0.123) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.056 (0.121) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.055 (0.119) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.256 (0.118) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.046 (0.117) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.059 (0.115) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.057 (0.114) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.058 (0.113) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.028 (0.111) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.064 (0.109) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.019 (0.107) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.032 (0.105) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.011 (0.103) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.053 (0.101) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.021 (0.100) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.008 (0.098) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.021 (0.097) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.028 (0.095) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.015 (0.093) Loss 15.8373 (2.7909) -Test: [0/435] Time 16.932 (16.932) Loss 13.4768 (13.4768) -Test: [10/435] Time 0.102 (1.642) Loss 3.8265 (7.4022) -Test: [20/435] Time 0.008 (0.901) Loss 10.8776 (7.2060) -Test: [30/435] Time 0.052 (0.641) Loss 10.0520 (6.3861) -Test: [40/435] Time 0.047 (0.498) Loss 4.0857 (6.4776) -Test: [50/435] Time 0.009 (0.421) Loss 2.3633 (5.9823) -Test: [60/435] Time 0.045 (0.363) Loss 2.5386 (5.5355) -Test: [70/435] Time 0.061 (0.328) Loss 1.6233 (5.0419) -Test: [80/435] Time 0.029 (0.293) Loss 4.9065 (4.8347) -Test: [90/435] Time 0.010 (0.272) Loss 0.3864 (4.4399) -Test: [100/435] Time 0.102 (0.252) Loss 2.6647 (4.2093) -Test: [110/435] Time 0.015 (0.235) Loss 0.8263 (3.8973) -Test: [120/435] Time 0.072 (0.223) Loss 1.3630 (3.7581) -Test: [130/435] Time 0.021 (0.212) Loss 0.3643 (3.6380) -Test: [140/435] Time 0.070 (0.204) Loss 0.8837 (3.4330) -Test: [150/435] Time 0.057 (0.194) Loss 0.7218 (3.3113) -Test: [160/435] Time 0.068 (0.186) Loss 0.4155 (3.1299) -Test: [170/435] Time 0.057 (0.181) Loss 1.9887 (2.9747) -Test: [180/435] Time 0.035 (0.175) Loss 0.1545 (2.8421) -Test: [190/435] Time 0.025 (0.171) Loss 0.1481 (2.7088) -Test: [200/435] Time 0.059 (0.166) Loss 0.0549 (2.6782) -Test: [210/435] Time 0.072 (0.163) Loss 0.0519 (2.5848) -Test: [220/435] Time 0.045 (0.158) Loss 0.4416 (2.5026) -Test: [230/435] Time 0.053 (0.154) Loss 0.4162 (2.4140) -Test: [240/435] Time 0.067 (0.151) Loss 0.3122 (2.3264) -Test: [250/435] Time 0.027 (0.147) Loss 0.0276 (2.2567) -Test: [260/435] Time 0.034 (0.142) Loss 0.1919 (2.2134) -Test: [270/435] Time 0.030 (0.139) Loss 0.0665 (2.1634) -Test: [280/435] Time 0.027 (0.135) Loss 0.5928 (2.1213) -Test: [290/435] Time 0.022 (0.131) Loss 0.4286 (2.0869) -Test: [300/435] Time 0.023 (0.128) Loss 1.5573 (2.0447) -Test: [310/435] Time 0.020 (0.124) Loss 2.2205 (2.0447) -Test: [320/435] Time 0.024 (0.121) Loss 3.0426 (2.0438) -Test: [330/435] Time 0.033 (0.118) Loss 2.3030 (2.0327) -Test: [340/435] Time 0.013 (0.115) Loss 1.6866 (2.0461) -Test: [350/435] Time 0.015 (0.113) Loss 0.5024 (2.0714) -Test: [360/435] Time 0.008 (0.110) Loss 4.9589 (2.0963) -Test: [370/435] Time 0.028 (0.108) Loss 1.0733 (2.1245) -Test: [380/435] Time 0.027 (0.106) Loss 4.3960 (2.1709) -Test: [390/435] Time 0.009 (0.104) Loss 5.0647 (2.2146) -Test: [400/435] Time 0.015 (0.101) Loss 6.0071 (2.2936) -Test: [410/435] Time 0.007 (0.099) Loss 9.5637 (2.3691) -Test: [420/435] Time 0.019 (0.097) Loss 15.2775 (2.5429) -Test: [430/435] Time 0.007 (0.095) Loss 15.8373 (2.7909) diff --git a/torch/train.sh b/torch/train.sh deleted file mode 100755 index 81d338a7..00000000 --- a/torch/train.sh +++ /dev/null @@ -1 +0,0 @@ -python main.py ../dataset --multiprocessing-distributed -j 8 --batch-size 128 --epochs 50 > log.txt