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: /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 ------------------------ 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: 2.88586 s ***** [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86750, lr: 0.000000, top1: 0.00125, throughput: 276.68 | 2022-05-22 18:59:36.228 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86725, lr: 0.000000, top1: 0.00145, throughput: 276.59 | 2022-05-22 18:59:36.247 [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86739, lr: 0.000000, top1: 0.00105, throughput: 276.53 | 2022-05-22 18:59:36.272 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86731, lr: 0.000000, top1: 0.00133, throughput: 276.47 | 2022-05-22 18:59:36.280 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86758, lr: 0.000000, top1: 0.00117, throughput: 276.43 | 2022-05-22 18:59:36.290 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86725, lr: 0.000000, top1: 0.00125, throughput: 276.45 | 2022-05-22 18:59:36.306 [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86746, lr: 0.000000, top1: 0.00109, throughput: 276.48 | 2022-05-22 18:59:36.338 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86737, lr: 0.000000, top1: 0.00141, throughput: 276.72 | 2022-05-22 18:59:36.222 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/22 18:59:36.402, Tesla V100-SXM2-32GB, 470.57.02, 44 %, 5 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.408, Tesla V100-SXM2-32GB, 470.57.02, 9 %, 4 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.413, Tesla V100-SXM2-32GB, 470.57.02, 7 %, 0 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.417, Tesla V100-SXM2-32GB, 470.57.02, 42 %, 30 %, 32510 MiB, 5160 MiB, 27350 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/22 18:59:36.421, Tesla V100-SXM2-32GB, 470.57.02, 50 %, 1 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.423, Tesla V100-SXM2-32GB, 470.57.02, 44 %, 5 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.426, Tesla V100-SXM2-32GB, 470.57.02, 28 %, 0 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.430, Tesla V100-SXM2-32GB, 470.57.02, 9 %, 4 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.432, Tesla V100-SXM2-32GB, 470.57.02, 58 %, 16 %, 32510 MiB, 5044 MiB, 27466 MiB 2022/05/22 18:59:36.437, Tesla V100-SXM2-32GB, 470.57.02, 7 %, 0 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.441, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 53 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.444, Tesla V100-SXM2-32GB, 470.57.02, 87 %, 44 %, 32510 MiB, 5160 MiB, 27350 MiB 2022/05/22 18:59:36.450, Tesla V100-SXM2-32GB, 470.57.02, 50 %, 1 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.457, Tesla V100-SXM2-32GB, 470.57.02, 28 %, 0 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.464, Tesla V100-SXM2-32GB, 470.57.02, 58 %, 16 %, 32510 MiB, 5044 MiB, 27466 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/22 18:59:36.469, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 53 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.475, Tesla V100-SXM2-32GB, 470.57.02, 44 %, 5 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.482, Tesla V100-SXM2-32GB, 470.57.02, 10 %, 0 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.492, Tesla V100-SXM2-32GB, 470.57.02, 7 %, 0 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.497, Tesla V100-SXM2-32GB, 470.57.02, 87 %, 44 %, 32510 MiB, 5160 MiB, 27350 MiB 2022/05/22 18:59:36.515, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 80 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.534, Tesla V100-SXM2-32GB, 470.57.02, 83 %, 72 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.554, Tesla V100-SXM2-32GB, 470.57.02, 58 %, 16 %, 32510 MiB, 5044 MiB, 27466 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] 2022/05/22 18:59:36.558, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 51 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.575, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 86 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.575, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 86 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.576, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 86 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.609, Tesla V100-SXM2-32GB, 470.57.02, 10 %, 0 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.610, Tesla V100-SXM2-32GB, 470.57.02, 10 %, 0 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.620, Tesla V100-SXM2-32GB, 470.57.02, 10 %, 0 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.638, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 83 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.640, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 83 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.642, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 83 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.655, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5160 MiB, 27350 MiB 2022/05/22 18:59:36.656, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5160 MiB, 27350 MiB 2022/05/22 18:59:36.657, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5160 MiB, 27350 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/22 18:59:36.679, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.679, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.681, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.699, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 52 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.709, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 66 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.711, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 66 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.712, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 66 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.715, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 71 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.717, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 87 %, 32510 MiB, 5044 MiB, 27466 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/22 18:59:36.717, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 47 %, 32510 MiB, 5044 MiB, 27466 MiB 2022/05/22 18:59:36.718, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 47 %, 32510 MiB, 5044 MiB, 27466 MiB 2022/05/22 18:59:36.720, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 55 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.722, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.736, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.737, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.736, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 52 %, 32510 MiB, 5088 MiB, 27422 MiB 2022/05/22 18:59:36.739, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5160 MiB, 27350 MiB 2022/05/22 18:59:36.743, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 71 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/05/22 18:59:36.746, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.751, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 55 %, 32510 MiB, 5184 MiB, 27326 MiB 2022/05/22 18:59:36.755, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 66 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.757, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 44 %, 32510 MiB, 5160 MiB, 27350 MiB 2022/05/22 18:59:36.764, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 47 %, 32510 MiB, 5044 MiB, 27466 MiB 2022/05/22 18:59:36.766, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 57 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/05/22 18:59:36.770, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 45 %, 32510 MiB, 5124 MiB, 27386 MiB 2022/05/22 18:59:36.791, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 66 %, 32510 MiB, 5100 MiB, 27410 MiB 2022/05/22 18:59:36.794, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 47 %, 32510 MiB, 5044 MiB, 27466 MiB 2022/05/22 18:59:36.796, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 45 %, 32510 MiB, 5124 MiB, 27386 MiB [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86727, lr: 0.000000, top1: 0.00109, throughput: 293.57 | 2022-05-22 19:01:03.431 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86742, lr: 0.000000, top1: 0.00129, throughput: 293.73 | 2022-05-22 19:01:03.436 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86765, lr: 0.000000, top1: 0.00102, throughput: 293.50[rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86736, lr: 0.000000, top1: 0.00113, throughput: 293.90 | 2022-05-22 19:01:03.444 | 2022-05-22 19:01:03.444 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86752, lr: 0.000000, top1: 0.00078, throughput: 293.59 | 2022-05-22 19:01:03.444 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86729, lr: 0.000000, top1: 0.00148, throughput: 293.52 | 2022-05-22 19:01:03.507 [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86713, lr: 0.000000, top1: 0.00102, throughput: 293.45 | 2022-05-22 19:01:03.509 [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86737, lr: 0.000000, top1: 0.00133, throughput: 293.80 | 2022-05-22 19:01:03.440 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86757, lr: 0.000000, top1: 0.00129, throughput: 294.57 | 2022-05-22 19:02:30.341 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86752, lr: 0.000000, top1: 0.00137, throughput: 294.81 | 2022-05-22 19:02:30.342 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86779, lr: 0.000000, top1: 0.00113, throughput: 294.79 | 2022-05-22 19:02:30.352 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86744, lr: 0.000000, top1: 0.00121, throughput: 294.53 | 2022-05-22 19:02:30.362 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86729, lr: 0.000000, top1: 0.00102, throughput: 294.47 | 2022-05-22 19:02:30.380 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86750, lr: 0.000000, top1: 0.00059, throughput: 294.42 | 2022-05-22 19:02:30.381 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86751, lr: 0.000000, top1: 0.00129, throughput: 294.42 | 2022-05-22 19:02:30.391 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86722, lr: 0.000000, top1: 0.00148, throughput: 294.77 | 2022-05-22 19:02:30.292