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 .................................. 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: 2.95242 s ***** [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86730, top1: 0.00094, throughput: 270.83 | 2022-04-08 10:06:50.403 [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86742, top1: 0.00113, throughput: 270.82 | 2022-04-08 10:06:50.405 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86697, top1: 0.00109, throughput: 270.81 | 2022-04-08 10:06:50.404 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86700, top1: 0.00129, throughput: 270.81 | 2022-04-08 10:06:50.407 [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86707, top1: 0.00082, throughput: 270.82 | 2022-04-08 10:06:50.404 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86732, top1: 0.00090, throughput: 270.83 | 2022-04-08 10:06:50.407 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86716, top1: 0.00055, throughput: 270.81 | 2022-04-08 10:06:50.407 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86750, top1: 0.00090, throughput: 270.80 | 2022-04-08 10:06:50.406 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] 2022/04/08 10:06:50.736, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB 2022/04/08 10:06:50.736, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/08 10:06:50.737, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/08 10:06:50.744, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.744, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.746, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.745, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/08 10:06:50.749, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/08 10:06:50.753, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.754, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/08 10:06:50.755, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.755, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.756, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB 2022/04/08 10:06:50.757, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.761, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.761, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.761, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB 2022/04/08 10:06:50.762, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.763, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.762, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8424 MiB, 24086 MiB 2022/04/08 10:06:50.771, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.771, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.777, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.777, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.778, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.778, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8442 MiB, 24068 MiB 2022/04/08 10:06:50.781, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.781, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.785, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.785, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.786, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.787, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.788, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.788, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8594 MiB, 23916 MiB 2022/04/08 10:06:50.790, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.791, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.795, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.795, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.795, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.796, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.797, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.797, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8490 MiB, 24020 MiB 2022/04/08 10:06:50.800, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.800, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.804, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.804, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.804, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.805, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.806, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.806, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8558 MiB, 23952 MiB 2022/04/08 10:06:50.808, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.809, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.813, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.815, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.815, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8444 MiB, 24066 MiB 2022/04/08 10:06:50.817, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.817, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.824, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.826, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 59 %, 32510 MiB, 8332 MiB, 24178 MiB 2022/04/08 10:06:50.829, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.833, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB 2022/04/08 10:06:50.834, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8512 MiB, 23998 MiB [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86720, top1: 0.00082, throughput: 389.10 | 2022-04-08 10:07:56.198 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86714, top1: 0.00098, throughput: 389.11 | 2022-04-08 10:07:56.199 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86696, top1: 0.00078, throughput: 389.10 | 2022-04-08 10:07:56.198 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86722, top1: 0.00121, throughput: 389.09 | 2022-04-08 10:07:56.198 [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86758, top1: 0.00129, throughput: 389.09 | 2022-04-08 10:07:56.198 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86733, top1: 0.00062, throughput: 389.11 | 2022-04-08 10:07:56.199 [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86720, top1: 0.00102, throughput: 389.09 | 2022-04-08 10:07:56.198 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86721, top1: 0.00086, throughput: 389.09 | 2022-04-08 10:07:56.201 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86731, top1: 0.00078, throughput: 388.93 | 2022-04-08 10:09:02.020 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86712, top1: 0.00082, throughput: 388.92 | 2022-04-08 10:09:02.021 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86721, top1: 0.00105, throughput: 388.92 | 2022-04-08 10:09:02.021 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86706, top1: 0.00066, throughput: 388.92 | 2022-04-08 10:09:02.022 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86720, top1: 0.00086, throughput: 388.92 | 2022-04-08 10:09:02.021 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86703, top1: 0.00090, throughput: 388.92 | 2022-04-08 10:09:02.022 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86737, top1: 0.00113, throughput: 388.92 | 2022-04-08 10:09:02.022 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86712, top1: 0.00082, throughput: 388.93 | 2022-04-08 10:09:02.023