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: 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/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 ............................... 312 channel_last .................................... True 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 ................................... 4.096 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 ................................ 512 train_global_batch_size ......................... 4096 use_fp16 ........................................ True 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: 2.92974 s ***** [rank:7] [train], epoch: 0/1, iter: 100/312, loss: 0.86738, top1: 0.00098, throughput: 429.61 | 2022-05-08 01:59:07.270 [rank:1] [train], epoch: 0/1, iter: 100/312, loss: 0.86756, top1: 0.00105, throughput: 429.60 | 2022-05-08 01:59:07.271 [rank:0] [train], epoch: 0/1, iter: 100/312, loss: 0.86778, top1: 0.00109, throughput: 429.61 | 2022-05-08 01:59:07.271 [rank:4] [train], epoch: 0/1, iter: 100/312, loss: 0.86781, top1: 0.00072, throughput: 429.62 | 2022-05-08 01:59:07.272 [rank:2] [train], epoch: 0/1, iter: 100/312, loss: 0.86796, top1: 0.00109, throughput: 429.61 | 2022-05-08 01:59:07.272 [rank:6] [train], epoch: 0/1, iter: 100/312, loss: 0.86739, top1: 0.00119, throughput: 429.65 | 2022-05-08 01:59:07.271 [rank:3] [train], epoch: 0/1, iter: 100/312, loss: 0.86751, top1: 0.00111, throughput: 429.64 | 2022-05-08 01:59:07.272 [rank:5] [train], epoch: 0/1, iter: 100/312, loss: 0.86753, top1: 0.00111, throughput: 429.62 | 2022-05-08 01:59:07.272 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] 2022/05/08 01:59:07.515, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB 2022/05/08 01:59:07.515, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/08 01:59:07.521, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.521, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 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] 2022/05/08 01:59:07.526, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.525, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB 2022/05/08 01:59:07.526, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/08 01:59:07.527, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/05/08 01:59:07.527, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB 2022/05/08 01:59:07.528, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB 2022/05/08 01:59:07.531, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.531, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.531, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.533, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.532, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB 2022/05/08 01:59:07.534, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 86 %, 32510 MiB, 7769 MiB, 24741 MiB 2022/05/08 01:59:07.534, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.535, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.538, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.539, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.539, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.545, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.546, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.547, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 81 %, 32510 MiB, 7730 MiB, 24780 MiB 2022/05/08 01:59:07.547, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.548, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.551, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.551, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.552, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.553, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.554, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.555, Tesla V100-SXM2-32GB, 470.57.02, 97 %, 86 %, 32510 MiB, 7952 MiB, 24558 MiB 2022/05/08 01:59:07.555, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.555, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.558, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.559, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.559, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.561, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.561, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.562, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 72 %, 32510 MiB, 7920 MiB, 24590 MiB 2022/05/08 01:59:07.562, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.563, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.566, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.566, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.566, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.568, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.569, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.569, Tesla V100-SXM2-32GB, 470.57.02, 77 %, 68 %, 32510 MiB, 7926 MiB, 24584 MiB 2022/05/08 01:59:07.569, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.570, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.574, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.575, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.576, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.577, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 75 %, 32510 MiB, 7792 MiB, 24718 MiB 2022/05/08 01:59:07.577, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.578, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.580, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.582, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.583, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.583, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 61 %, 32510 MiB, 7696 MiB, 24814 MiB 2022/05/08 01:59:07.583, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.584, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.589, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB 2022/05/08 01:59:07.590, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 88 %, 32510 MiB, 7820 MiB, 24690 MiB [rank:1] [train], epoch: 0/1, iter: 200/312, loss: 0.86743, top1: 0.00137, throughput: 1281.46 | 2022-05-08 01:59:47.225 [rank:7] [train], epoch: 0/1, iter: 200/312, loss: 0.86770, top1: 0.00080, throughput: 1281.45 | 2022-05-08 01:59:47.224 [rank:0] [train], epoch: 0/1, iter: 200/312, loss: 0.86736, top1: 0.00115, throughput: 1281.49 | 2022-05-08 01:59:47.225 [rank:4] [train], epoch: 0/1, iter: 200/312, loss: 0.86733, top1: 0.00119, throughput: 1281.49 | 2022-05-08 01:59:47.225 [rank:3] [train], epoch: 0/1, iter: 200/312, loss: 0.86763, top1: 0.00094, throughput: 1281.49 | 2022-05-08 01:59:47.225 [rank:5] [train], epoch: 0/1, iter: 200/312, loss: 0.86752, top1: 0.00109, throughput: 1281.49 | 2022-05-08 01:59:47.225 [rank:2] [train], epoch: 0/1, iter: 200/312, loss: 0.86750, top1: 0.00148, throughput: 1281.41 | 2022-05-08 01:59:47.227 [rank:6] [train], epoch: 0/1, iter: 200/312, loss: 0.86722, top1: 0.00111, throughput: 1281.44 | 2022-05-08 01:59:47.226 [rank:5] [train], epoch: 0/1, iter: 300/312, loss: 0.86757, top1: 0.00104, throughput: 1339.22 | 2022-05-08 02:00:25.457 [rank:0] [train], epoch: 0/1, iter: 300/312, loss: 0.86744, top1: 0.00121, throughput: 1339.16 | 2022-05-08 02:00:25.458 [rank:1] [train], epoch: 0/1, iter: 300/312, loss: 0.86749, top1: 0.00115, throughput: 1339.20 | 2022-05-08 02:00:25.457 [rank:2] [train], epoch: 0/1, iter: 300/312, loss: 0.86733, top1: 0.00121, throughput: 1339.29 | 2022-05-08 02:00:25.457 [rank:4] [train], epoch: 0/1, iter: 300/312, loss: 0.86736, top1: 0.00141, throughput: 1339.21 | 2022-05-08 02:00:25.456 [rank:3] [train], epoch: 0/1, iter: 300/312, loss: 0.86759, top1: 0.00100, throughput: 1339.09 | 2022-05-08 02:00:25.460 [rank:6] [train], epoch: 0/1, iter: 300/312, loss: 0.86733, top1: 0.00082, throughput: 1339.23 | 2022-05-08 02:00:25.457 [rank:7] [train], epoch: 0/1, iter: 300/312, loss: 0.86745, top1: 0.00111, throughput: 1339.16 | 2022-05-08 02:00:25.457