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: 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 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 .................................. 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: 2.77439 s ***** [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86731, top1: 0.00094, throughput: 276.03 | 2022-04-15 12:02:19.881 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86748, top1: 0.00074, throughput: 275.94 | 2022-04-15 12:02:19.880 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86723, top1: 0.00055, throughput: 275.90 | 2022-04-15 12:02:19.881 [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86745, top1: 0.00066, throughput: 276.17 | 2022-04-15 12:02:19.887 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86730, top1: 0.00051, throughput: 275.90 | 2022-04-15 12:02:19.886 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86733, top1: 0.00082, throughput: 276.07 | 2022-04-15 12:02:19.895 [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86724, top1: 0.00066, throughput: 276.11 | 2022-04-15 12:02:19.886 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86734, top1: 0.00070, throughput: 275.85 | 2022-04-15 12:02:19.885 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/15 12:02:20.216, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 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] 2022/04/15 12:02:20.224, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.225, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/15 12:02:20.235, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.234, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/15 12:02:20.237, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 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] timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/15 12:02:20.244, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.245, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.265, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.265, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/15 12:02:20.266, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/15 12:02:20.276, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.275, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/15 12:02:20.275, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/15 12:02:20.276, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.275, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/15 12:02:20.278, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.279, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.280, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.285, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 62 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.285, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.285, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.285, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.286, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8500 MiB, 24010 MiB 2022/04/15 12:02:20.288, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.289, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.289, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.294, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.295, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.295, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.295, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.295, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/04/15 12:02:20.297, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 62 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.298, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.299, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.304, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.304, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.304, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.304, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 62 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.305, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8580 MiB, 23930 MiB 2022/04/15 12:02:20.307, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.307, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.308, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.313, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.314, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.314, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.314, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8616 MiB, 23894 MiB 2022/04/15 12:02:20.316, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.317, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.322, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.328, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.328, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.329, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.329, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 64 %, 32510 MiB, 8432 MiB, 24078 MiB 2022/04/15 12:02:20.332, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.333, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.339, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.339, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.340, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8390 MiB, 24120 MiB 2022/04/15 12:02:20.345, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.346, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.350, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.350, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB 2022/04/15 12:02:20.351, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8548 MiB, 23962 MiB [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86737, top1: 0.00062, throughput: 381.16 | 2022-04-15 12:03:27.045 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86744, top1: 0.00051, throughput: 381.16 | 2022-04-15 12:03:27.044 [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86726, top1: 0.00125, throughput: 381.16 | 2022-04-15 12:03:27.044 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86727, top1: 0.00059, throughput: 381.15 | 2022-04-15 12:03:27.051 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86718, top1: 0.00082, throughput: 381.21 | 2022-04-15 12:03:27.049 [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86715, top1: 0.00094, throughput: 381.16 | 2022-04-15 12:03:27.050 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86724, top1: 0.00055, throughput: 381.16 | 2022-04-15 12:03:27.048 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86735, top1: 0.00098, throughput: 381.15 | 2022-04-15 12:03:27.052 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86722, top1: 0.00043, throughput: 379.59 | 2022-04-15 12:04:34.486 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86733, top1: 0.00062, throughput: 379.60 | 2022-04-15 12:04:34.487 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86749, top1: 0.00059, throughput: 379.59 | 2022-04-15 12:04:34.485 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86704, top1: 0.00094, throughput: 379.58 | 2022-04-15 12:04:34.486 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86729, top1: 0.00102, throughput: 379.59 | 2022-04-15 12:04:34.492 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86704, top1: 0.00082, throughput: 379.59 | 2022-04-15 12:04:34.490 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86783, top1: 0.00074, throughput: 379.59 | 2022-04-15 12:04:34.489 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86728, top1: 0.00047, throughput: 379.60 | 2022-04-15 12:04:34.490