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: /usr/lib/x86_64-linux-gnu/libibverbs.so.1 loaded library: 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/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: 3.11486 s ***** [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86693, top1: 0.00133, throughput: 263.86 | 2022-05-10 02:05:40.165 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86718, top1: 0.00113, throughput: 263.86 | 2022-05-10 02:05:40.168 [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86717, top1: 0.00086, throughput: 263.86 | 2022-05-10 02:05:40.165 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86731, top1: 0.00098, throughput: 263.85 | 2022-05-10 02:05:40.167 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86739, top1: 0.00133, throughput: 263.87 | 2022-05-10 02:05:40.165 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86747, top1: 0.00113, throughput: 263.86 | 2022-05-10 02:05:40.166 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86760, top1: 0.00098, throughput: 263.88 | 2022-05-10 02:05:40.166 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86740, top1: 0.00090, throughput: 263.85 | 2022-05-10 02:05:40.167 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/10 02:05:40.456, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 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] 2022/05/10 02:05:40.463, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB 2022/05/10 02:05:40.465, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.467, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.468, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.468, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.475, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/10 02:05:40.476, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/10 02:05:40.476, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/10 02:05:40.477, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/10 02:05:40.479, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.481, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.480, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.481, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.481, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.482, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.482, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.482, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 7751 MiB, 24759 MiB 2022/05/10 02:05:40.487, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.489, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.490, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB 2022/05/10 02:05:40.492, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.492, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB 2022/05/10 02:05:40.493, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.493, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB 2022/05/10 02:05:40.494, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 60 %, 32510 MiB, 7668 MiB, 24842 MiB 2022/05/10 02:05:40.497, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.499, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.500, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.500, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.501, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.501, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.502, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.503, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7884 MiB, 24626 MiB 2022/05/10 02:05:40.506, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.508, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.508, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.508, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.509, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.509, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.510, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.513, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 7860 MiB, 24650 MiB 2022/05/10 02:05:40.516, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.518, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.519, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.520, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.520, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.521, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.522, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 7876 MiB, 24634 MiB 2022/05/10 02:05:40.525, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.527, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.527, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.528, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.529, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.529, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.532, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7760 MiB, 24750 MiB 2022/05/10 02:05:40.538, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.539, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.540, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.541, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 7630 MiB, 24880 MiB 2022/05/10 02:05:40.546, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.546, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.553, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB 2022/05/10 02:05:40.554, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 7852 MiB, 24658 MiB [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86718, top1: 0.00141, throughput: 372.46 | 2022-05-10 02:06:48.897 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86724, top1: 0.00113, throughput: 372.43 | 2022-05-10 02:06:48.904 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86727, top1: 0.00098, throughput: 372.41 | 2022-05-10 02:06:48.905 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86743, top1: 0.00125, throughput: 372.42 | 2022-05-10 02:06:48.905 [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86751, top1: 0.00133, throughput: 372.42 | 2022-05-10 02:06:48.904 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86755, top1: 0.00078, throughput: 372.42 | 2022-05-10 02:06:48.906 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86751, top1: 0.00109, throughput: 372.42 | 2022-05-10 02:06:48.906 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86719, top1: 0.00090, throughput: 372.42 | 2022-05-10 02:06:48.908 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86741, top1: 0.00133, throughput: 377.72 | 2022-05-10 02:07:56.679 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86720, top1: 0.00137, throughput: 377.72 | 2022-05-10 02:07:56.679 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86688, top1: 0.00102, throughput: 377.68 | 2022-05-10 02:07:56.679 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86750, top1: 0.00113, throughput: 377.73 | 2022-05-10 02:07:56.680 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86736, top1: 0.00105, throughput: 377.73 | 2022-05-10 02:07:56.679 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86738, top1: 0.00105, throughput: 377.73 | 2022-05-10 02:07:56.679 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86744, top1: 0.00102, throughput: 377.73 | 2022-05-10 02:07:56.680 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86738, top1: 0.00090, throughput: 377.73 | 2022-05-10 02:07:56.681