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: /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: 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 ------------------------ arguments ------------------------ batches_per_epoch ............................... 625 channel_last .................................... False ddp ............................................. True exit_num ........................................ 300 fuse_bn_add_relu ................................ False fuse_bn_relu .................................... False gpu_stat_file ................................... None grad_clipping ................................... 0.0 graph ........................................... False 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 ...................................... False 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: 2.90758 s ***** [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86766, lr: 0.000000, top1: 0.00078, throughput: 277.33 | 2022-05-06 22:01:22.265 [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86752, lr: 0.000000, top1: 0.00055, throughput: 277.21 | 2022-05-06 22:01:22.267 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86761, lr: 0.000000, top1: 0.00094, throughput: 277.22 | 2022-05-06 22:01:22.277 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86730, lr: 0.000000, top1: 0.00113, throughput: 277.05 | 2022-05-06 22:01:22.324 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86749, lr: 0.000000, top1: 0.00098, throughput: 277.03 | 2022-05-06 22:01:22.343 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86766, lr: 0.000000, top1: 0.00113, throughput: 277.06 | 2022-05-06 22:01:22.347 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86755, lr: 0.000000, top1: 0.00090, throughput: 276.98 | 2022-05-06 22:01:22.359 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86749, lr: 0.000000, top1: 0.00129, throughput: 277.29 | 2022-05-06 22:01:22.262 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/06 22:01:22.401, Tesla V100-SXM2-32GB, 470.57.02, 30 %, 20 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.405, Tesla V100-SXM2-32GB, 470.57.02, 64 %, 12 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.410, Tesla V100-SXM2-32GB, 470.57.02, 18 %, 13 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.415, Tesla V100-SXM2-32GB, 470.57.02, 65 %, 29 %, 32510 MiB, 5166 MiB, 27344 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] 2022/05/06 22:01:22.418, Tesla V100-SXM2-32GB, 470.57.02, 30 %, 17 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.419, Tesla V100-SXM2-32GB, 470.57.02, 30 %, 20 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.419, Tesla V100-SXM2-32GB, 470.57.02, 30 %, 20 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.423, Tesla V100-SXM2-32GB, 470.57.02, 57 %, 13 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.424, Tesla V100-SXM2-32GB, 470.57.02, 64 %, 12 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.425, Tesla V100-SXM2-32GB, 470.57.02, 64 %, 12 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.428, Tesla V100-SXM2-32GB, 470.57.02, 5 %, 0 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.430, Tesla V100-SXM2-32GB, 470.57.02, 18 %, 13 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.431, Tesla V100-SXM2-32GB, 470.57.02, 18 %, 13 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.434, Tesla V100-SXM2-32GB, 470.57.02, 19 %, 0 %, 32510 MiB, 5130 MiB, 27380 MiB 2022/05/06 22:01:22.436, Tesla V100-SXM2-32GB, 470.57.02, 65 %, 29 %, 32510 MiB, 5166 MiB, 27344 MiB 2022/05/06 22:01:22.436, Tesla V100-SXM2-32GB, 470.57.02, 65 %, 29 %, 32510 MiB, 5166 MiB, 27344 MiB 2022/05/06 22:01:22.442, Tesla V100-SXM2-32GB, 470.57.02, 30 %, 17 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.443, Tesla V100-SXM2-32GB, 470.57.02, 30 %, 17 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.455, Tesla V100-SXM2-32GB, 470.57.02, 57 %, 13 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.456, Tesla V100-SXM2-32GB, 470.57.02, 57 %, 13 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.460, Tesla V100-SXM2-32GB, 470.57.02, 5 %, 0 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.460, Tesla V100-SXM2-32GB, 470.57.02, 5 %, 0 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.464, Tesla V100-SXM2-32GB, 470.57.02, 19 %, 0 %, 32510 MiB, 5130 MiB, 27380 MiB 2022/05/06 22:01:22.464, Tesla V100-SXM2-32GB, 470.57.02, 19 %, 0 %, 32510 MiB, 5130 MiB, 27380 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/06 22:01:22.562, Tesla V100-SXM2-32GB, 470.57.02, 68 %, 47 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.583, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 87 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.609, Tesla V100-SXM2-32GB, 470.57.02, 68 %, 52 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.643, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5166 MiB, 27344 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] 2022/05/06 22:01:22.660, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.679, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.679, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.703, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 56 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.706, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 51 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.708, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 51 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.720, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 59 %, 32510 MiB, 5050 MiB, 27460 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] 2022/05/06 22:01:22.725, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 74 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.726, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 74 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.727, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 78 %, 32510 MiB, 5130 MiB, 27380 MiB 2022/05/06 22:01:22.732, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.733, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5166 MiB, 27344 MiB 2022/05/06 22:01:22.733, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5166 MiB, 27344 MiB 2022/05/06 22:01:22.733, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5094 MiB, 27416 MiB 2022/05/06 22:01:22.747, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 51 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.749, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.751, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.754, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 51 %, 32510 MiB, 5114 MiB, 27396 MiB 2022/05/06 22:01:22.761, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 74 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.762, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 56 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.762, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 56 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.762, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 74 %, 32510 MiB, 5190 MiB, 27320 MiB 2022/05/06 22:01:22.775, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5166 MiB, 27344 MiB 2022/05/06 22:01:22.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 73 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 73 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.776, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 76 %, 32510 MiB, 5166 MiB, 27344 MiB 2022/05/06 22:01:22.783, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.783, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 5130 MiB, 27380 MiB 2022/05/06 22:01:22.783, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 5130 MiB, 27380 MiB 2022/05/06 22:01:22.784, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5174 MiB, 27336 MiB 2022/05/06 22:01:22.796, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 56 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.797, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 56 %, 32510 MiB, 5106 MiB, 27404 MiB 2022/05/06 22:01:22.801, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 73 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.802, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 73 %, 32510 MiB, 5050 MiB, 27460 MiB 2022/05/06 22:01:22.806, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 5130 MiB, 27380 MiB 2022/05/06 22:01:22.807, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 58 %, 32510 MiB, 5130 MiB, 27380 MiB [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86754, lr: 0.000000, top1: 0.00102, throughput: 295.44 | 2022-05-06 22:02:48.918 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86732, lr: 0.000000, top1: 0.00074, throughput: 295.69 | 2022-05-06 22:02:48.921 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86759, lr: 0.000000, top1: 0.00074, throughput: 295.54 | 2022-05-06 22:02:48.945 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86743, lr: 0.000000, top1: 0.00105, throughput: 295.57 | 2022-05-06 22:02:48.958 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86706, lr: 0.000000, top1: 0.00117, throughput: 295.25 | 2022-05-06 22:02:48.967 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86719, lr: 0.000000, top1: 0.00113, throughput: 295.24 | 2022-05-06 22:02:48.975 [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86727, lr: 0.000000, top1: 0.00102, throughput: 295.25 | 2022-05-06 22:02:48.984 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86724, lr: 0.000000, top1: 0.00078, throughput: 295.41 | 2022-05-06 22:02:49.018 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86784, lr: 0.000000, top1: 0.00094, throughput: 295.08 | 2022-05-06 22:04:15.722 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86688, lr: 0.000000, top1: 0.00102, throughput: 295.00 | 2022-05-06 22:04:15.725 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86765, lr: 0.000000, top1: 0.00086, throughput: 295.06 | 2022-05-06 22:04:15.737 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86745, lr: 0.000000, top1: 0.00102, throughput: 294.80 | 2022-05-06 22:04:15.756 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86748, lr: 0.000000, top1: 0.00102, throughput: 294.89 | 2022-05-06 22:04:15.770 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86714, lr: 0.000000, top1: 0.00098, throughput: 294.76 | 2022-05-06 22:04:15.772 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86728, lr: 0.000000, top1: 0.00094, throughput: 295.34 | 2022-05-06 22:04:15.697 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86748, lr: 0.000000, top1: 0.00098, throughput: 295.21 | 2022-05-06 22:04:15.701