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: 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 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: 3.16430 s ***** [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86724, top1: 0.00109, throughput: 275.74 | 2022-05-02 02:10:21.533 [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86754, top1: 0.00109, throughput: 275.73 | 2022-05-02 02:10:21.534 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86756, top1: 0.00078, throughput: 275.74 | 2022-05-02 02:10:21.537 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86778, top1: 0.00074, throughput: 275.75 | 2022-05-02 02:10:21.534 [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86762, top1: 0.00105, throughput: 275.74 | 2022-05-02 02:10:21.535 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86755, top1: 0.00105, throughput: 275.75 | 2022-05-02 02:10:21.536 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86763, top1: 0.00082, throughput: 275.76 | 2022-05-02 02:10:21.536 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86750, top1: 0.00102, throughput: 275.73 | 2022-05-02 02:10:21.537 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] 2022/05/02 02:10:21.855, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/02 02:10:21.864, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/02 02:10:21.868, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/02 02:10:21.872, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.872, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/02 02:10:21.875, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB 2022/05/02 02:10:21.875, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 8664 MiB, 23846 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] 2022/05/02 02:10:21.879, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB 2022/05/02 02:10:21.881, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.881, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.884, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.884, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.883, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB 2022/05/02 02:10:21.885, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB 2022/05/02 02:10:21.886, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 63 %, 32510 MiB, 8516 MiB, 23994 MiB 2022/05/02 02:10:21.890, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.890, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.891, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.894, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 59 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.894, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.894, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.895, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.897, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 61 %, 32510 MiB, 8468 MiB, 24042 MiB 2022/05/02 02:10:21.899, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.900, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.900, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.903, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.903, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.904, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.911, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.913, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 63 %, 32510 MiB, 8664 MiB, 23846 MiB 2022/05/02 02:10:21.915, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.916, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.916, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.919, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.920, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 60 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.920, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.921, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.923, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 62 %, 32510 MiB, 8554 MiB, 23956 MiB 2022/05/02 02:10:21.925, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.925, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.926, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.929, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.929, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.929, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.930, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.937, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8600 MiB, 23910 MiB 2022/05/02 02:10:21.939, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.939, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.940, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.943, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.943, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.945, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.946, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8494 MiB, 24016 MiB 2022/05/02 02:10:21.948, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.949, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.952, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.952, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.953, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.955, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 65 %, 32510 MiB, 8382 MiB, 24128 MiB 2022/05/02 02:10:21.956, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.961, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.962, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB 2022/05/02 02:10:21.963, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 66 %, 32510 MiB, 8544 MiB, 23966 MiB [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86754, top1: 0.00105, throughput: 380.85 | 2022-05-02 02:11:28.753 [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86766, top1: 0.00102, throughput: 380.84 | 2022-05-02 02:11:28.753 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86725, top1: 0.00090, throughput: 380.85 | 2022-05-02 02:11:28.754 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86723, top1: 0.00129, throughput: 380.86 | 2022-05-02 02:11:28.753 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86771, top1: 0.00094, throughput: 380.84 | 2022-05-02 02:11:28.755 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86768, top1: 0.00090, throughput: 380.84 | 2022-05-02 02:11:28.754 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86760, top1: 0.00070, throughput: 380.85 | 2022-05-02 02:11:28.755 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86764, top1: 0.00094, throughput: 380.84 | 2022-05-02 02:11:28.756 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86709, top1: 0.00086, throughput: 378.93 | 2022-05-02 02:12:36.311 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86748, top1: 0.00066, throughput: 378.93 | 2022-05-02 02:12:36.311 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86758, top1: 0.00066, throughput: 378.94 | 2022-05-02 02:12:36.312 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86749, top1: 0.00086, throughput: 378.94 | 2022-05-02 02:12:36.311 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86750, top1: 0.00105, throughput: 378.92 | 2022-05-02 02:12:36.312 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86744, top1: 0.00090, throughput: 378.93 | 2022-05-02 02:12:36.314 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86761, top1: 0.00109, throughput: 378.94 | 2022-05-02 02:12:36.313 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86720, top1: 0.00137, throughput: 378.93 | 2022-05-02 02:12:36.314