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: /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 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 ............................................. 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: 2.70797 s ***** [rank:3] [train], epoch: 0/1, iter: 100/625, loss: 0.86765, lr: 0.000000, top1: 0.00082, throughput: 276.53 | 2022-04-22 10:25:05.147 [rank:5] [train], epoch: 0/1, iter: 100/625, loss: 0.86788, lr: 0.000000, top1: 0.00070, throughput: 276.52 | 2022-04-22 10:25:05.162 [rank:4] [train], epoch: 0/1, iter: 100/625, loss: 0.86767, lr: 0.000000, top1: 0.00070, throughput: 276.42 | 2022-04-22 10:25:05.189 [rank:0] [train], epoch: 0/1, iter: 100/625, loss: 0.86767, lr: 0.000000, top1: 0.00074, throughput: 276.45 | 2022-04-22 10:25:05.190 [rank:7] [train], epoch: 0/1, iter: 100/625, loss: 0.86761, lr: 0.000000, top1: 0.00070, throughput: 276.35 | 2022-04-22 10:25:05.208 [rank:2] [train], epoch: 0/1, iter: 100/625, loss: 0.86741, lr: 0.000000, top1: 0.00082, throughput: 276.32 | 2022-04-22 10:25:05.217 [rank:6] [train], epoch: 0/1, iter: 100/625, loss: 0.86790, lr: 0.000000, top1: 0.00078, throughput: 276.39 | 2022-04-22 10:25:05.228 [rank:1] [train], epoch: 0/1, iter: 100/625, loss: 0.86766, lr: 0.000000, top1: 0.00082, throughput: 276.31 | 2022-04-22 10:25:05.260 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/22 10:25:05.293, Tesla V100-SXM2-32GB, 470.57.02, 54 %, 1 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.298, Tesla V100-SXM2-32GB, 470.57.02, 4 %, 0 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.302, Tesla V100-SXM2-32GB, 470.57.02, 26 %, 18 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.306, Tesla V100-SXM2-32GB, 470.57.02, 64 %, 39 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.310, Tesla V100-SXM2-32GB, 470.57.02, 50 %, 35 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.314, Tesla V100-SXM2-32GB, 470.57.02, 72 %, 10 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/04/22 10:25:05.319, Tesla V100-SXM2-32GB, 470.57.02, 22 %, 12 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.323, Tesla V100-SXM2-32GB, 470.57.02, 25 %, 13 %, 32510 MiB, 5132 MiB, 27378 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] 2022/04/22 10:25:05.343, Tesla V100-SXM2-32GB, 470.57.02, 54 %, 1 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.348, Tesla V100-SXM2-32GB, 470.57.02, 54 %, 1 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.351, Tesla V100-SXM2-32GB, 470.57.02, 4 %, 0 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.352, Tesla V100-SXM2-32GB, 470.57.02, 4 %, 0 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.355, Tesla V100-SXM2-32GB, 470.57.02, 26 %, 18 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.357, Tesla V100-SXM2-32GB, 470.57.02, 26 %, 18 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.360, Tesla V100-SXM2-32GB, 470.57.02, 98 %, 31 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.363, Tesla V100-SXM2-32GB, 470.57.02, 98 %, 31 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.366, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 38 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.368, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 38 %, 32510 MiB, 5176 MiB, 27334 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/22 10:25:05.371, Tesla V100-SXM2-32GB, 470.57.02, 72 %, 10 %, 32510 MiB, 5108 MiB, 27402 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/22 10:25:05.373, Tesla V100-SXM2-32GB, 470.57.02, 72 %, 10 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/04/22 10:25:05.374, Tesla V100-SXM2-32GB, 470.57.02, 54 %, 1 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.376, Tesla V100-SXM2-32GB, 470.57.02, 22 %, 12 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.382, Tesla V100-SXM2-32GB, 470.57.02, 54 %, 1 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.383, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 48 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.385, Tesla V100-SXM2-32GB, 470.57.02, 4 %, 0 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.387, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 34 %, 32510 MiB, 5132 MiB, 27378 MiB 2022/04/22 10:25:05.389, Tesla V100-SXM2-32GB, 470.57.02, 4 %, 0 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.390, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 34 %, 32510 MiB, 5132 MiB, 27378 MiB 2022/04/22 10:25:05.391, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 52 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.403, Tesla V100-SXM2-32GB, 470.57.02, 91 %, 52 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.405, Tesla V100-SXM2-32GB, 470.57.02, 98 %, 31 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.409, Tesla V100-SXM2-32GB, 470.57.02, 98 %, 31 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.411, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 37 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.423, Tesla V100-SXM2-32GB, 470.57.02, 88 %, 37 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.425, Tesla V100-SXM2-32GB, 470.57.02, 72 %, 10 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/04/22 10:25:05.430, Tesla V100-SXM2-32GB, 470.57.02, 72 %, 10 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/04/22 10:25:05.435, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 48 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.447, Tesla V100-SXM2-32GB, 470.57.02, 79 %, 47 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.448, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 34 %, 32510 MiB, 5132 MiB, 27378 MiB 2022/04/22 10:25:05.468, Tesla V100-SXM2-32GB, 470.57.02, 70 %, 34 %, 32510 MiB, 5132 MiB, 27378 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/22 10:25:05.505, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 88 %, 32510 MiB, 5096 MiB, 27414 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/22 10:25:05.511, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 69 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.527, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 86 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.543, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.548, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 69 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.567, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 80 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.568, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.578, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.584, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 80 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.587, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 87 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/04/22 10:25:05.589, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.600, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 77 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.602, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 87 %, 32510 MiB, 5108 MiB, 27402 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/22 10:25:05.605, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 82 %, 32510 MiB, 5132 MiB, 27378 MiB 2022/04/22 10:25:05.606, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 77 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.620, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 86 %, 32510 MiB, 5096 MiB, 27414 MiB 2022/04/22 10:25:05.623, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 82 %, 32510 MiB, 5132 MiB, 27378 MiB 2022/04/22 10:25:05.625, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 66 %, 32510 MiB, 5116 MiB, 27394 MiB 2022/04/22 10:25:05.630, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 75 %, 32510 MiB, 5192 MiB, 27318 MiB 2022/04/22 10:25:05.647, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 80 %, 32510 MiB, 5168 MiB, 27342 MiB 2022/04/22 10:25:05.660, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 78 %, 32510 MiB, 5176 MiB, 27334 MiB 2022/04/22 10:25:05.663, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 50 %, 32510 MiB, 5108 MiB, 27402 MiB 2022/04/22 10:25:05.682, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 77 %, 32510 MiB, 5052 MiB, 27458 MiB 2022/04/22 10:25:05.685, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 82 %, 32510 MiB, 5132 MiB, 27378 MiB [rank:3] [train], epoch: 0/1, iter: 200/625, loss: 0.86767, lr: 0.000000, top1: 0.00086, throughput: 294.76 | 2022-04-22 10:26:31.998 [rank:6] [train], epoch: 0/1, iter: 200/625, loss: 0.86748, lr: 0.000000, top1: 0.00090, throughput: 295.03 | 2022-04-22 10:26:31.999 [rank:7] [train], epoch: 0/1, iter: 200/625, loss: 0.86781, lr: 0.000000, top1: 0.00078, throughput: 294.95 | 2022-04-22 10:26:32.003 [rank:2] [train], epoch: 0/1, iter: 200/625, loss: 0.86748, lr: 0.000000, top1: 0.00102, throughput: 294.94 | 2022-04-22 10:26:32.013 [rank:4] [train], epoch: 0/1, iter: 200/625, loss: 0.86777, lr: 0.000000, top1: 0.00055, throughput: 294.69 | 2022-04-22 10:26:32.060 [rank:0] [train], epoch: 0/1, iter: 200/625, loss: 0.86765, lr: 0.000000, top1: 0.00074, throughput: 294.67 | 2022-04-22 10:26:32.067 [rank:5] [train], epoch: 0/1, iter: 200/625, loss: 0.86737, lr: 0.000000, top1: 0.00086, throughput: 294.51 | 2022-04-22 10:26:32.088 [rank:1] [train], epoch: 0/1, iter: 200/625, loss: 0.86768, lr: 0.000000, top1: 0.00059, throughput: 294.72 | 2022-04-22 10:26:32.122 [rank:6] [train], epoch: 0/1, iter: 300/625, loss: 0.86763, lr: 0.000000, top1: 0.00090, throughput: 296.38 | 2022-04-22 10:27:58.374 [rank:4] [train], epoch: 0/1, iter: 300/625, loss: 0.86766, lr: 0.000000, top1: 0.00055, throughput: 296.59 | 2022-04-22 10:27:58.374 [rank:1] [train], epoch: 0/1, iter: 300/625, loss: 0.86744, lr: 0.000000, top1: 0.00082, throughput: 296.68 | 2022-04-22 10:27:58.411 [rank:2] [train], epoch: 0/1, iter: 300/625, loss: 0.86756, lr: 0.000000, top1: 0.00059, throughput: 296.25 | 2022-04-22 10:27:58.425 [rank:0] [train], epoch: 0/1, iter: 300/625, loss: 0.86748, lr: 0.000000, top1: 0.00113, throughput: 296.40 | 2022-04-22 10:27:58.437 [rank:7] [train], epoch: 0/1, iter: 300/625, loss: 0.86755, lr: 0.000000, top1: 0.00062, throughput: 296.18 | 2022-04-22 10:27:58.437 [rank:3] [train], epoch: 0/1, iter: 300/625, loss: 0.86793, lr: 0.000000, top1: 0.00059, throughput: 296.08 | 2022-04-22 10:27:58.461 [rank:5] [train], epoch: 0/1, iter: 300/625, loss: 0.86763, lr: 0.000000, top1: 0.00066, throughput: 296.73 | 2022-04-22 10:27:58.361