loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
*****************************************
Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. 
*****************************************
loaded library: loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1/usr/lib/x86_64-linux-gnu/libibverbs.so.1

loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1/usr/lib/x86_64-linux-gnu/libibverbs.so.1

loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
------------------------ arguments ------------------------
  batches_per_epoch ............................... 625
  channel_last .................................... False
  ddp ............................................. True
  exit_num ........................................ 300
  fuse_bn_add_relu ................................ False
  fuse_bn_relu .................................... False
  gpu_stat_file ................................... None
  grad_clipping ................................... 0.0
  graph ........................................... False
  label_smoothing ................................. 0.1
  learning_rate ................................... 2.048
  legacy_init ..................................... False
  load_path ....................................... None
  lr_decay_type ................................... cosine
  metric_local .................................... True
  metric_train_acc ................................ True
  momentum ........................................ 0.875
  nccl_fusion_max_ops ............................. 24
  nccl_fusion_threshold_mb ........................ 16
  num_classes ..................................... 1000
  num_devices_per_node ............................ 8
  num_epochs ...................................... 1
  num_nodes ....................................... 1
  ofrecord_part_num ............................... 256
  ofrecord_path ................................... /dataset/79846248
  print_interval .................................. 100
  print_timestamp ................................. False
  samples_per_epoch ............................... 1281167
  save_init ....................................... False
  save_path ....................................... None
  scale_grad ...................................... False
  skip_eval ....................................... True
  synthetic_data .................................. False
  total_batches ................................... -1
  train_batch_size ................................ 256
  train_global_batch_size ......................... 2048
  use_fp16 ........................................ False
  use_gpu_decode .................................. False
  val_batch_size .................................. 50
  val_batches_per_epoch ........................... 125
  val_global_batch_size ........................... 400
  val_samples_per_epoch ........................... 50000
  warmup_epochs ................................... 5
  weight_decay .................................... 3.0517578125e-05
  zero_init_residual .............................. True
-------------------- end of arguments ---------------------
***** Model Init *****
***** Model Init Finish, time escapled: 2.79275 s *****
[rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86755, lr: 0.000000, top1: 0.00086, throughput: 276.76 | 2022-04-29 01:57:03.397
[rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86810, lr: 0.000000, top1: 0.00074, throughput: 276.71 | 2022-04-29 01:57:03.400
[rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86739, lr: 0.000000, top1: 0.00082, throughput: 276.49 | 2022-04-29 01:57:03.436
[rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86763, lr: 0.000000, top1: 0.00090, throughput: 276.51 | 2022-04-29 01:57:03.461
[rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86794, lr: 0.000000, top1: 0.00098, throughput: 276.44 | 2022-04-29 01:57:03.493
[rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86768, lr: 0.000000, top1: 0.00086, throughput: 276.53 | 2022-04-29 01:57:03.498
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.507, Tesla V100-SXM2-32GB, 470.57.02, 71 %, 51 %, 32510 MiB, 5094 MiB, 27416 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.509, Tesla V100-SXM2-32GB, 470.57.02, 76 %, 51 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.510, Tesla V100-SXM2-32GB, 470.57.02, 71 %, 51 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.511, Tesla V100-SXM2-32GB, 470.57.02, 90 %, 59 %, 32510 MiB, 5190 MiB, 27320 MiB
[rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86784, lr: 0.000000, top1: 0.00086, throughput: 276.36 | 2022-04-29 01:57:03.512
2022/04/29 01:57:03.511, Tesla V100-SXM2-32GB, 470.57.02, 76 %, 51 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.512, Tesla V100-SXM2-32GB, 470.57.02, 73 %, 51 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.517, Tesla V100-SXM2-32GB, 470.57.02, 90 %, 59 %, 32510 MiB, 5190 MiB, 27320 MiB
[rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86833, lr: 0.000000, top1: 0.00047, throughput: 276.30 | 2022-04-29 01:57:03.521
2022/04/29 01:57:03.519, Tesla V100-SXM2-32GB, 470.57.02, 49 %, 5 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.522, Tesla V100-SXM2-32GB, 470.57.02, 73 %, 51 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.524, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 48 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.529, Tesla V100-SXM2-32GB, 470.57.02, 49 %, 5 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.530, Tesla V100-SXM2-32GB, 470.57.02, 14 %, 0 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.532, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 48 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.533, Tesla V100-SXM2-32GB, 470.57.02, 0 %, 0 %, 32510 MiB, 5130 MiB, 27380 MiB
2022/04/29 01:57:03.537, Tesla V100-SXM2-32GB, 470.57.02, 14 %, 0 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.540, Tesla V100-SXM2-32GB, 470.57.02, 0 %, 0 %, 32510 MiB, 5130 MiB, 27380 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.589, Tesla V100-SXM2-32GB, 470.57.02, 93 %, 39 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.592, Tesla V100-SXM2-32GB, 470.57.02, 47 %, 8 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.597, Tesla V100-SXM2-32GB, 470.57.02, 14 %, 0 %, 32510 MiB, 5190 MiB, 27320 MiB
2022/04/29 01:57:03.599, Tesla V100-SXM2-32GB, 470.57.02, 90 %, 28 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.603, Tesla V100-SXM2-32GB, 470.57.02, 49 %, 5 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.607, Tesla V100-SXM2-32GB, 470.57.02, 10 %, 0 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.611, Tesla V100-SXM2-32GB, 470.57.02, 14 %, 0 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.614, Tesla V100-SXM2-32GB, 470.57.02, 0 %, 0 %, 32510 MiB, 5130 MiB, 27380 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.647, Tesla V100-SXM2-32GB, 470.57.02, 93 %, 39 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.655, Tesla V100-SXM2-32GB, 470.57.02, 47 %, 8 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.657, Tesla V100-SXM2-32GB, 470.57.02, 14 %, 0 %, 32510 MiB, 5190 MiB, 27320 MiB
2022/04/29 01:57:03.659, Tesla V100-SXM2-32GB, 470.57.02, 90 %, 28 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.672, Tesla V100-SXM2-32GB, 470.57.02, 49 %, 5 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.677, Tesla V100-SXM2-32GB, 470.57.02, 10 %, 0 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.690, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 73 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.694, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 67 %, 32510 MiB, 5130 MiB, 27380 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.734, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.740, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 88 %, 32510 MiB, 5114 MiB, 27396 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.757, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 68 %, 32510 MiB, 5190 MiB, 27320 MiB
2022/04/29 01:57:03.771, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.772, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.772, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 84 %, 32510 MiB, 5166 MiB, 27344 MiB
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/29 01:57:03.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 88 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 88 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.777, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 85 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.777, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5094 MiB, 27416 MiB
2022/04/29 01:57:03.794, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 66 %, 32510 MiB, 5190 MiB, 27320 MiB
2022/04/29 01:57:03.794, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 66 %, 32510 MiB, 5190 MiB, 27320 MiB
2022/04/29 01:57:03.794, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.796, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 88 %, 32510 MiB, 5114 MiB, 27396 MiB
2022/04/29 01:57:03.799, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 84 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.799, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 84 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.800, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 71 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.810, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 66 %, 32510 MiB, 5190 MiB, 27320 MiB
2022/04/29 01:57:03.813, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 85 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.814, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 85 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.814, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 65 %, 32510 MiB, 5130 MiB, 27380 MiB
2022/04/29 01:57:03.815, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 84 %, 32510 MiB, 5166 MiB, 27344 MiB
2022/04/29 01:57:03.819, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.819, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.828, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 85 %, 32510 MiB, 5174 MiB, 27336 MiB
2022/04/29 01:57:03.833, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 71 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.833, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 71 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.835, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 5106 MiB, 27404 MiB
2022/04/29 01:57:03.838, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 65 %, 32510 MiB, 5130 MiB, 27380 MiB
2022/04/29 01:57:03.839, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 65 %, 32510 MiB, 5130 MiB, 27380 MiB
2022/04/29 01:57:03.847, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 67 %, 32510 MiB, 5050 MiB, 27460 MiB
2022/04/29 01:57:03.858, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 71 %, 32510 MiB, 5130 MiB, 27380 MiB
[rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86744, lr: 0.000000, top1: 0.00078, throughput: 294.19 | 2022-04-29 01:58:30.454
[rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86792, lr: 0.000000, top1: 0.00086, throughput: 294.36 | 2022-04-29 01:58:30.465
[rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86747, lr: 0.000000, top1: 0.00090, throughput: 293.95 | 2022-04-29 01:58:30.490
[rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86757, lr: 0.000000, top1: 0.00082, throughput: 294.23 | 2022-04-29 01:58:30.518
[rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86741, lr: 0.000000, top1: 0.00078, throughput: 294.15[rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86784, lr: 0.000000, top1: 0.00098, throughput: 294.24 | 2022-04-29 01:58:30.524 
| 2022-04-29 01:58:30.524
[rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86769, lr: 0.000000, top1: 0.00090, throughput: 293.95 | 2022-04-29 01:58:30.551
[rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86761, lr: 0.000000, top1: 0.00168, throughput: 293.73 | 2022-04-29 01:58:30.553
[rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86759, lr: 0.000000, top1: 0.00102, throughput: 294.99 | 2022-04-29 01:59:57.335
[rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86774, lr: 0.000000, top1: 0.00090, throughput: 294.85 | 2022-04-29 01:59:57.347
[rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86757, lr: 0.000000, top1: 0.00086, throughput: 294.79 | 2022-04-29 01:59:57.358
[rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86765, lr: 0.000000, top1: 0.00094, throughput: 294.76 | 2022-04-29 01:59:57.373
[rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86793, lr: 0.000000, top1: 0.00078, throughput: 294.86 | 2022-04-29 01:59:57.373
[rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86734, lr: 0.000000, top1: 0.00035, throughput: 294.46 | 2022-04-29 01:59:57.393
[rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86756, lr: 0.000000, top1: 0.00145, throughput: 294.49 | 2022-04-29 01:59:57.395
[rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86737, lr: 0.000000, top1: 0.00105, throughput: 294.56 | 2022-04-29 01:59:57.399