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: loaded library: loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1/usr/lib/x86_64-linux-gnu/libibverbs.so.1/usr/lib/x86_64-linux-gnu/libibverbs.so.1/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 ------------------------ arguments ------------------------ batches_per_epoch ............................... 625 channel_last .................................... False ddp ............................................. False exit_num ........................................ 300 fuse_bn_add_relu ................................ True fuse_bn_relu .................................... True gpu_stat_file ................................... None grad_clipping ................................... 0.0 graph ........................................... True 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 ...................................... True skip_eval ....................................... True synthetic_data .................................. False total_batches ................................... -1 train_batch_size ................................ 256 train_global_batch_size ......................... 2048 use_fp16 ........................................ False use_gpu_decode .................................. True 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: 3.07742 s ***** [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86746, top1: 0.00098, throughput: 263.56 | 2022-05-26 18:01:11.524 [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86767, top1: 0.00145, throughput: 263.63 | 2022-05-26 18:01:11.526 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86749, top1: 0.00062, throughput: 263.56 | 2022-05-26 18:01:11.525 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86731, top1: 0.00094, throughput: 263.55 | 2022-05-26 18:01:11.527 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86758, top1: 0.00062, throughput: 263.64 | 2022-05-26 18:01:11.526 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86733, top1: 0.00078, throughput: 263.57 | 2022-05-26 18:01:11.524 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86760, top1: 0.00082, throughput: 263.55 | 2022-05-26 18:01:11.527 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86755, top1: 0.00062, throughput: 263.65 | 2022-05-26 18:01:11.525 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/26 18:01:11.812, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/26 18:01:11.820, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.820, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 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] timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/26 18:01:11.830, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.831, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/26 18:01:11.831, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 MiB 2022/05/26 18:01:11.831, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 MiB 2022/05/26 18:01:11.831, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 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] 2022/05/26 18:01:11.836, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.837, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.838, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.838, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.838, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.837, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 MiB 2022/05/26 18:01:11.839, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 MiB 2022/05/26 18:01:11.840, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7719 MiB, 24791 MiB 2022/05/26 18:01:11.843, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.845, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.845, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.846, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.846, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.846, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.848, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.849, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.852, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.853, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.854, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.854, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.854, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.855, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.857, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.858, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7882 MiB, 24628 MiB 2022/05/26 18:01:11.862, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.863, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.864, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.864, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.865, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.865, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.867, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.868, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 7866 MiB, 24644 MiB 2022/05/26 18:01:11.871, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.872, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.873, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.873, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.873, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.873, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.875, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.876, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7870 MiB, 24640 MiB 2022/05/26 18:01:11.880, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.881, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.882, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.882, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.882, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.884, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.885, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 7722 MiB, 24788 MiB 2022/05/26 18:01:11.890, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.890, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.890, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.891, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.892, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.893, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7600 MiB, 24910 MiB 2022/05/26 18:01:11.898, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.900, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/05/26 18:01:11.901, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7790 MiB, 24720 MiB [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86727, top1: 0.00125, throughput: 377.12 | 2022-05-26 18:02:19.407 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86737, top1: 0.00070, throughput: 377.13 | 2022-05-26 18:02:19.407 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86761, top1: 0.00074, throughput: 377.12 | 2022-05-26 18:02:19.408 [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86749, top1: 0.00109, throughput: 377.11 | 2022-05-26 18:02:19.408 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86771, top1: 0.00094, throughput: 377.12 | 2022-05-26 18:02:19.408 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86736, top1: 0.00094, throughput: 377.12 | 2022-05-26 18:02:19.409 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86737, top1: 0.00082, throughput: 377.12 | 2022-05-26 18:02:19.410 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86754, top1: 0.00098, throughput: 377.12 | 2022-05-26 18:02:19.410 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86733, top1: 0.00062, throughput: 374.30 | 2022-05-26 18:03:27.802 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86722, top1: 0.00062, throughput: 374.31 | 2022-05-26 18:03:27.801 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86757, top1: 0.00059, throughput: 374.30 | 2022-05-26 18:03:27.801 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86728, top1: 0.00113, throughput: 374.30 | 2022-05-26 18:03:27.802 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86772, top1: 0.00090, throughput: 374.31 | 2022-05-26 18:03:27.803 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86767, top1: 0.00074, throughput: 374.30 | 2022-05-26 18:03:27.802 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86743, top1: 0.00102, throughput: 374.30 | 2022-05-26 18:03:27.805 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86776, top1: 0.00121, throughput: 374.29 | 2022-05-26 18:03:27.804