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 ...................................... 65536
  batch_size_per_proc ............................. 65536
  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: 0.5048305392265320, latency(ms): 164.5923647657036781 | 2022-04-15 01:48:18.522
timestamp, name, driver_version, utilization.gpu [%], utilization.memory [%], memory.total [MiB], memory.free [MiB], memory.used [MiB]
2022/04/15 01:48:20.310, Tesla V100-SXM2-32GB, 470.57.02, 0 %, 0 %, 32510 MiB, 28966 MiB, 3544 MiB
[rank:0] iter: 200/1100, loss: 0.4715388119220734, latency(ms): 132.8966336324810982 | 2022-04-15 01:48:31.812
[rank:0] iter: 300/1100, loss: 0.4639216959476471, latency(ms): 115.6409234553575516 | 2022-04-15 01:48:43.376
[rank:0] iter: 400/1100, loss: 0.4608530700206757, latency(ms): 113.9072391763329506 | 2022-04-15 01:48:54.766
[rank:0] iter: 500/1100, loss: 0.4578500986099243, latency(ms): 115.7326907292008400 | 2022-04-15 01:49:06.340
[rank:0] iter: 600/1100, loss: 0.4527823328971863, latency(ms): 115.6345367059111595 | 2022-04-15 01:49:17.903
[rank:0] iter: 700/1100, loss: 0.4472089409828186, latency(ms): 121.3133496791124344 | 2022-04-15 01:49:30.035
[rank:0] iter: 800/1100, loss: 0.4479774236679077, latency(ms): 114.5928063616156578 | 2022-04-15 01:49:41.494
[rank:0] iter: 900/1100, loss: 0.4472298920154572, latency(ms): 122.0076050236821175 | 2022-04-15 01:49:53.695
[rank:0] iter: 1000/1100, loss: 0.4466403722763062, latency(ms): 120.7192019000649452 | 2022-04-15 01:50:05.766
[rank:0] iter: 1100/1100, loss: 0.4467601478099823, latency(ms): 124.7422039881348610 | 2022-04-15 01:50:18.241