loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
loaded library: /usr/lib/x86_64-linux-gnu/libibverbs.so.1
------------------------ arguments ------------------------
  batch_size ...................................... 512
  batch_size_per_proc ............................. 512
  data_dir ........................................ /dataset/f9f659c5/wdl_ofrecord
  data_part_name_suffix_length .................... 5
  data_part_num ................................... 256
  dataset_format .................................. ofrecord
  ddp ............................................. True
  deep_dropout_rate ............................... 0.5
  deep_embedding_vec_size ......................... 16
  deep_vocab_size ................................. 2322444
  eval_after_training ............................. False
  eval_batchs ..................................... 20
  eval_interval ................................... 0
  execution_mode .................................. eager
  hidden_size ..................................... 1024
  hidden_units_num ................................ 2
  learning_rate ................................... 0.001
  loss_print_every_n_iter ......................... 100
  max_iter ........................................ 1100
  model_load_dir .................................. 
  model_save_dir .................................. ./checkpoint
  num_deep_sparse_fields .......................... 26
  num_dense_fields ................................ 13
  num_wide_sparse_fields .......................... 2
  save_initial_model .............................. False
  save_model_after_each_eval ...................... False
  test_name ....................................... noname_test
  wide_vocab_size ................................. 2322444
-------------------- end of arguments ---------------------
[rank:0] iter: 100/1100, loss: 0.5278729796409607, latency(ms): 41.1160516366362572 | 2022-04-18 02:02:54.169
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/18 02:02:54.200, Tesla V100-SXM2-32GB, 470.57.02, 99 %, 97 %, 32510 MiB, 30682 MiB, 1828 MiB
[rank:0] iter: 200/1100, loss: 0.5001709461212158, latency(ms): 4.2639669030904770 | 2022-04-18 02:02:54.595
[rank:0] iter: 300/1100, loss: 0.4970759451389313, latency(ms): 4.0406133607029915 | 2022-04-18 02:02:54.999
[rank:0] iter: 400/1100, loss: 0.4919593930244446, latency(ms): 4.0576609224081039 | 2022-04-18 02:02:55.405
[rank:0] iter: 500/1100, loss: 0.4937639236450195, latency(ms): 4.0554801747202873 | 2022-04-18 02:02:55.810
[rank:0] iter: 600/1100, loss: 0.4814880490303040, latency(ms): 4.0541280806064606 | 2022-04-18 02:02:56.216
[rank:0] iter: 700/1100, loss: 0.4863997399806976, latency(ms): 4.0749396756291389 | 2022-04-18 02:02:56.623
[rank:0] iter: 800/1100, loss: 0.4852460026741028, latency(ms): 4.0606321394443512 | 2022-04-18 02:02:57.029
[rank:0] iter: 900/1100, loss: 0.4858795404434204, latency(ms): 4.0901892632246017 | 2022-04-18 02:02:57.438
[rank:0] iter: 1000/1100, loss: 0.4822021722793579, latency(ms): 4.0713366121053696 | 2022-04-18 02:02:57.846
[rank:0] iter: 1100/1100, loss: 0.4822394549846649, latency(ms): 4.0750683471560478 | 2022-04-18 02:02:58.253