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: /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: 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: 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: 2.76744 s ***** [rank:6] [train], epoch: 0/1, iter: 100/312, loss: 0.86740, top1: 0.00115, throughput: 421.06 | 2022-04-09 01:49:38.453 [rank:4] [train], epoch: 0/1, iter: 100/312, loss: 0.86739, top1: 0.00098, throughput: 421.07 | 2022-04-09 01:49:38.453 [rank:5] [train], epoch: 0/1, iter: 100/312, loss: 0.86742, top1: 0.00092, throughput: 421.06 | 2022-04-09 01:49:38.452 [rank:7] [train], epoch: 0/1, iter: 100/312, loss: 0.86745, top1: 0.00102, throughput: 421.05 | 2022-04-09 01:49:38.455 [rank:3] [train], epoch: 0/1, iter: 100/312, loss: 0.86741, top1: 0.00115, throughput: 421.06 | 2022-04-09 01:49:38.455 [rank:0] [train], epoch: 0/1, iter: 100/312, loss: 0.86743, top1: 0.00098, throughput: 421.03 | 2022-04-09 01:49:38.459 [rank:1] [train], epoch: 0/1, iter: 100/312, loss: 0.86748, top1: 0.00115, throughput: 421.07 | 2022-04-09 01:49:38.457 [rank:2] [train], epoch: 0/1, iter: 100/312, loss: 0.86717, top1: 0.00143, throughput: 421.05 | 2022-04-09 01:49:38.455 timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/09 01:49:38.731, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 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] 2022/04/09 01:49:38.739, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 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] 2022/04/09 01:49:38.742, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/09 01:49:38.744, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/09 01:49:38.745, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 61 %, 32510 MiB, 7932 MiB, 24578 MiB timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB] 2022/04/09 01:49:38.748, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB 2022/04/09 01:49:38.749, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.749, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB 2022/04/09 01:49:38.751, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.750, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB 2022/04/09 01:49:38.751, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB 2022/04/09 01:49:38.752, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.752, Tesla V100-SXM2-32GB, 470.57.02, 95 %, 85 %, 32510 MiB, 7773 MiB, 24737 MiB 2022/04/09 01:49:38.754, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.755, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.756, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.758, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.758, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.759, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.759, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 73 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.760, Tesla V100-SXM2-32GB, 470.57.02, 92 %, 82 %, 32510 MiB, 7700 MiB, 24810 MiB 2022/04/09 01:49:38.763, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.764, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.765, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.766, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.767, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.767, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.768, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.768, Tesla V100-SXM2-32GB, 470.57.02, 69 %, 60 %, 32510 MiB, 7932 MiB, 24578 MiB 2022/04/09 01:49:38.770, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.771, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 73 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.772, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.773, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 73 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.774, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.774, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.775, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.776, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 68 %, 32510 MiB, 7892 MiB, 24618 MiB 2022/04/09 01:49:38.777, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 73 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.778, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.779, Tesla V100-SXM2-32GB, 470.57.02, 67 %, 56 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.781, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.781, Tesla V100-SXM2-32GB, 470.57.02, 67 %, 56 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.782, Tesla V100-SXM2-32GB, 470.57.02, 67 %, 56 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.782, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.783, Tesla V100-SXM2-32GB, 470.57.02, 67 %, 56 %, 32510 MiB, 7890 MiB, 24620 MiB 2022/04/09 01:49:38.785, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.786, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.787, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.788, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.788, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.789, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.790, Tesla V100-SXM2-32GB, 470.57.02, 84 %, 73 %, 32510 MiB, 7790 MiB, 24720 MiB 2022/04/09 01:49:38.792, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.793, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.794, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.796, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.796, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.797, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.798, Tesla V100-SXM2-32GB, 470.57.02, 100 %, 81 %, 32510 MiB, 7620 MiB, 24890 MiB 2022/04/09 01:49:38.800, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.802, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.803, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.804, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB 2022/04/09 01:49:38.805, Tesla V100-SXM2-32GB, 470.57.02, 78 %, 69 %, 32510 MiB, 7776 MiB, 24734 MiB [rank:5] [train], epoch: 0/1, iter: 200/312, loss: 0.86752, top1: 0.00082, throughput: 1365.63 | 2022-04-09 01:50:15.944 [rank:2] [train], epoch: 0/1, iter: 200/312, loss: 0.86729, top1: 0.00092, throughput: 1365.62 | 2022-04-09 01:50:15.947 [rank:3] [train], epoch: 0/1, iter: 200/312, loss: 0.86758, top1: 0.00115, throughput: 1365.63 | 2022-04-09 01:50:15.947 [rank:1] [train], epoch: 0/1, iter: 200/312, loss: 0.86757, top1: 0.00090, throughput: 1365.58 | 2022-04-09 01:50:15.950 [rank:6] [train], epoch: 0/1, iter: 200/312, loss: 0.86754, top1: 0.00082, throughput: 1365.54 | 2022-04-09 01:50:15.947 [rank:7] [train], epoch: 0/1, iter: 200/312, loss: 0.86765, top1: 0.00094, throughput: 1365.58 | 2022-04-09 01:50:15.949 [rank:0] [train], epoch: 0/1, iter: 200/312, loss: 0.86737, top1: 0.00111, throughput: 1365.70 | 2022-04-09 01:50:15.949 [rank:4] [train], epoch: 0/1, iter: 200/312, loss: 0.86746, top1: 0.00102, throughput: 1365.56 | 2022-04-09 01:50:15.946 [rank:5] [train], epoch: 0/1, iter: 300/312, loss: 0.86737, top1: 0.00082, throughput: 1338.53 | 2022-04-09 01:50:54.195 [rank:6] [train], epoch: 0/1, iter: 300/312, loss: 0.86737, top1: 0.00105, throughput: 1338.55 | 2022-04-09 01:50:54.198 [rank:3] [train], epoch: 0/1, iter: 300/312, loss: 0.86742, top1: 0.00104, throughput: 1338.50 | 2022-04-09 01:50:54.199 [rank:7] [train], epoch: 0/1, iter: 300/312, loss: 0.86765, top1: 0.00100, throughput: 1338.54 | 2022-04-09 01:50:54.200 [rank:0] [train], epoch: 0/1, iter: 300/312, loss: 0.86748, top1: 0.00119, throughput: 1338.50 | 2022-04-09 01:50:54.201 [rank:1] [train], epoch: 0/1, iter: 300/312, loss: 0.86757, top1: 0.00084, throughput: 1338.56 | 2022-04-09 01:50:54.200 [rank:2] [train], epoch: 0/1, iter: 300/312, loss: 0.86749, top1: 0.00092, throughput: 1338.47 | 2022-04-09 01:50:54.199 [rank:4] [train], epoch: 0/1, iter: 300/312, loss: 0.86734, top1: 0.00105, throughput: 1338.55 | 2022-04-09 01:50:54.197