01/26/2022 10:56:56 - WARNING - __main__ - Process rank: -1, device: cuda:0, n_gpu: 1distributed training: False, 16-bits training: True
01/26/2022 10:56:56 - INFO - __main__ - Training/evaluation parameters TrainingArguments(
_n_gpu=1,
adafactor=False,
adam_beta1=0.9,
adam_beta2=0.999,
adam_epsilon=1e-08,
bf16=False,
bf16_full_eval=False,
dataloader_drop_last=False,
dataloader_num_workers=0,
dataloader_pin_memory=True,
ddp_bucket_cap_mb=None,
ddp_find_unused_parameters=None,
debug=[],
deepspeed=None,
disable_tqdm=False,
do_eval=True,
do_predict=False,
do_train=True,
eval_accumulation_steps=None,
eval_steps=500,
evaluation_strategy=IntervalStrategy.STEPS,
fp16=True,
fp16_backend=auto,
fp16_full_eval=False,
fp16_opt_level=O1,
gradient_accumulation_steps=2,
gradient_checkpointing=True,
greater_is_better=None,
group_by_length=True,
half_precision_backend=auto,
hub_model_id=None,
hub_strategy=HubStrategy.EVERY_SAVE,
hub_token=<HUB_TOKEN>,
ignore_data_skip=False,
label_names=None,
label_smoothing_factor=0.0,
learning_rate=0.0003,
length_column_name=input_length,
load_best_model_at_end=False,
local_rank=-1,
log_level=-1,
log_level_replica=-1,
log_on_each_node=True,
logging_dir=./runs/Jan26_10-56-56_ganymede.eafit.edu.co,
logging_first_step=False,
logging_nan_inf_filter=True,
logging_steps=100,
logging_strategy=IntervalStrategy.STEPS,
lr_scheduler_type=SchedulerType.COSINE,
max_grad_norm=1.0,
max_steps=-1,
metric_for_best_model=None,
mp_parameters=,
no_cuda=False,
num_train_epochs=500.0,
optim=OptimizerNames.ADAMW_HF,
output_dir=./,
overwrite_output_dir=True,
past_index=-1,
per_device_eval_batch_size=80,
per_device_train_batch_size=80,
prediction_loss_only=False,
push_to_hub=True,
push_to_hub_model_id=None,
push_to_hub_organization=None,
push_to_hub_token=<PUSH_TO_HUB_TOKEN>,
remove_unused_columns=True,
report_to=['wandb'],
resume_from_checkpoint=None,
run_name=cosine+drop_proj+low_specaugment-300M,
save_on_each_node=False,
save_steps=500,
save_strategy=IntervalStrategy.STEPS,
save_total_limit=1,
seed=42,
sharded_ddp=[],
skip_memory_metrics=True,
tf32=None,
tpu_metrics_debug=False,
tpu_num_cores=None,
use_legacy_prediction_loop=False,
warmup_ratio=0.0,
warmup_steps=500,
weight_decay=0.0,
xpu_backend=None,
)
01/26/2022 10:57:00 - WARNING - datasets.builder - Reusing dataset common_voice (/home/sagrilaft/.cache/huggingface/datasets/mozilla-foundation___common_voice/et/7.0.0/33e08856cfa0d0665e837bcad73ffd920a0bc713ce8c5fffb55dbdf1c084d5ba)
01/26/2022 10:57:02 - WARNING - datasets.builder - Reusing dataset common_voice (/home/sagrilaft/.cache/huggingface/datasets/mozilla-foundation___common_voice/et/7.0.0/33e08856cfa0d0665e837bcad73ffd920a0bc713ce8c5fffb55dbdf1c084d5ba)
01/26/2022 10:57:02 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/sagrilaft/.cache/huggingface/datasets/mozilla-foundation___common_voice/et/7.0.0/33e08856cfa0d0665e837bcad73ffd920a0bc713ce8c5fffb55dbdf1c084d5ba/cache-f4479ce33e07f5c7.arrow
01/26/2022 10:57:02 - WARNING - datasets.arrow_dataset - Loading cached processed dataset at /home/sagrilaft/.cache/huggingface/datasets/mozilla-foundation___common_voice/et/7.0.0/33e08856cfa0d0665e837bcad73ffd920a0bc713ce8c5fffb55dbdf1c084d5ba/cache-105cefde64f772a9.arrow
01/26/2022 10:58:44 - WARNING - huggingface_hub.repository - /home/sagrilaft/Project/audio/xls-r-et/./ is already a clone of https://huggingface.co/shpotes/xls-r-et. Make sure you pull the latest changes with `repo.git_pull()`.

{'loss': 4.5903, 'learning_rate': 0.00019800000000000002, 'epoch': 2.78}
{'loss': 2.9083, 'learning_rate': 0.000398, 'epoch': 5.56}
{'loss': 1.2823, 'learning_rate': 0.000598, 'epoch': 8.33}
{'loss': 0.3962, 'learning_rate': 0.0007980000000000001, 'epoch': 11.11}
{'loss': 0.3034, 'learning_rate': 0.000998, 'epoch': 13.89}
{'eval_loss': 0.38567885756492615, 'eval_wer': 0.4354438434616608, 'eval_runtime': 133.1769, 'eval_samples_per_second': 19.59, 'eval_steps_per_second': 0.248, 'epoch': 13.89}
{'loss': 0.2639, 'learning_rate': 0.0009999210371864342, 'epoch': 16.67}
{'loss': 0.2354, 'learning_rate': 0.0009996809758056298, 'epoch': 19.44}
{'loss': 0.2158, 'learning_rate': 0.000999279885163942, 'epoch': 22.22}
{'loss': 0.2063, 'learning_rate': 0.000998717894518502, 'epoch': 25.0}
{'loss': 0.1926, 'learning_rate': 0.000997995184978743, 'epoch': 27.78}
{'eval_loss': 0.4142022132873535, 'eval_wer': 0.4174355711104041, 'eval_runtime': 135.7258, 'eval_samples_per_second': 19.223, 'eval_steps_per_second': 0.243, 'epoch': 27.78}
{'loss': 0.1771, 'learning_rate': 0.0009971119894480345, 'epoch': 30.56}
{'loss': 0.1709, 'learning_rate': 0.0009960685925486266, 'epoch': 33.33}
{'loss': 0.1724, 'learning_rate': 0.0009948653305299256, 'epoch': 36.11}
{'loss': 0.1632, 'learning_rate': 0.0009935025911601335, 'epoch': 38.89}
{'loss': 0.1609, 'learning_rate': 0.0009919808136012837, 'epoch': 41.67}
{'eval_loss': 0.4701250195503235, 'eval_wer': 0.42958956411072224, 'eval_runtime': 133.1213, 'eval_samples_per_second': 19.599, 'eval_steps_per_second': 0.248, 'epoch': 41.67}
01/26/2022 15:02:33 - WARNING - huggingface_hub.repository - Adding files tracked by Git LFS: ['wandb/run-20220126_105847-1xdiy2kf/run-1xdiy2kf.wandb']. This may take a bit of time if the files are large.
{'loss': 0.1521, 'learning_rate': 0.0009903004882677155, 'epoch': 44.44}
{'loss': 0.1531, 'learning_rate': 0.0009884621566680308, 'epoch': 47.22}
{'loss': 0.1514, 'learning_rate': 0.0009864664112305846, 'epoch': 50.0}
{'loss': 0.1468, 'learning_rate': 0.0009843138951125682, 'epoch': 52.78}
{'loss': 0.1501, 'learning_rate': 0.000982005301992741, 'epoch': 55.56}
{'eval_loss': 0.45478546619415283, 'eval_wer': 0.43805281578110083, 'eval_runtime': 133.376, 'eval_samples_per_second': 19.561, 'eval_steps_per_second': 0.247, 'epoch': 55.56}
{'loss': 0.1427, 'learning_rate': 0.0009795413758478834, 'epoch': 58.33}
{'loss': 0.1441, 'learning_rate': 0.0009769229107130409, 'epoch': 61.11}
{'loss': 0.1399, 'learning_rate': 0.0009741507504256327, 'epoch': 63.89}
{'loss': 0.1379, 'learning_rate': 0.0009712257883535152, 'epoch': 66.67}
{'loss': 0.1382, 'learning_rate': 0.0009681489671070803, 'epoch': 69.44}
{'eval_loss': 0.523300290107727, 'eval_wer': 0.4544066178810054, 'eval_runtime': 131.9566, 'eval_samples_per_second': 19.772, 'eval_steps_per_second': 0.25, 'epoch': 69.44}
{'loss': 0.1372, 'learning_rate': 0.0009649212782354872, 'epoch': 72.22}
{'loss': 0.1345, 'learning_rate': 0.0009615437619071199, 'epoch': 75.0}
{'loss': 0.1346, 'learning_rate': 0.0009580175065743799, 'epoch': 77.78}
{'loss': 0.1367, 'learning_rate': 0.0009543436486229162, 'epoch': 80.56}
{'loss': 0.1287, 'learning_rate': 0.0009505233720054086, 'epoch': 83.33}
{'eval_loss': 0.48360341787338257, 'eval_wer': 0.43658924594336623, 'eval_runtime': 132.2498, 'eval_samples_per_second': 19.728, 'eval_steps_per_second': 0.25, 'epoch': 83.33}
{'loss': 0.1226, 'learning_rate': 0.0009465579078600233, 'epoch': 86.11}
{'loss': 0.1258, 'learning_rate': 0.0009424485341136599, 'epoch': 88.89}
{'loss': 0.1222, 'learning_rate': 0.0009381965750701229, 'epoch': 91.67}
{'loss': 0.1182, 'learning_rate': 0.0009338034009833446, 'epoch': 94.44}
{'loss': 0.1164, 'learning_rate': 0.0009292704276158022, 'epoch': 97.22}
{'eval_loss': 0.5202445983886719, 'eval_wer': 0.43948456888323256, 'eval_runtime': 132.7793, 'eval_samples_per_second': 19.649, 'eval_steps_per_second': 0.249, 'epoch': 97.22}
{'loss': 0.1212, 'learning_rate': 0.0009245991157822682, 'epoch': 100.0}
{'loss': 0.1139, 'learning_rate': 0.000919790970879042, 'epoch': 102.78}
{'loss': 0.115, 'learning_rate': 0.000914847542398815, 'epoch': 105.56}
{'loss': 0.1123, 'learning_rate': 0.0009097704234313245, 'epoch': 108.33}
{'loss': 0.1101, 'learning_rate': 0.0009045612501499573, 'epoch': 111.11}
{'eval_loss': 0.5120447874069214, 'eval_wer': 0.43827553293032134, 'eval_runtime': 131.3001, 'eval_samples_per_second': 19.871, 'eval_steps_per_second': 0.251, 'epoch': 111.11}
{'loss': 0.1092, 'learning_rate': 0.0008992217012844713, 'epoch': 113.89}
{'loss': 0.1104, 'learning_rate': 0.0008937534975799996, 'epoch': 116.67}
{'loss': 0.1021, 'learning_rate': 0.0008881584012425165, 'epoch': 119.44}
{'loss': 0.1063, 'learning_rate': 0.0008824382153709421, 'epoch': 122.22}
{'loss': 0.1107, 'learning_rate': 0.0008765947833760678, 'epoch': 125.0}
{'eval_loss': 0.5130612850189209, 'eval_wer': 0.4446706967865097, 'eval_runtime': 133.0961, 'eval_samples_per_second': 19.602, 'eval_steps_per_second': 0.248, 'epoch': 125.0}
{'loss': 0.1054, 'learning_rate': 0.000870629988386491, 'epoch': 127.78}
{'loss': 0.1037, 'learning_rate': 0.0008645457526417502, 'epoch': 130.56}
{'loss': 0.1, 'learning_rate': 0.0008583440368728572, 'epoch': 133.33}
{'loss': 0.0967, 'learning_rate': 0.0008520268396704223, 'epoch': 136.11}
{'loss': 0.0942, 'learning_rate': 0.0008455961968405809, 'epoch': 138.89}
{'eval_loss': 0.5256752967834473, 'eval_wer': 0.43748011454024816, 'eval_runtime': 133.1352, 'eval_samples_per_second': 19.597, 'eval_steps_per_second': 0.248, 'epoch': 138.89}
{'loss': 0.0987, 'learning_rate': 0.0008390541807489265, 'epoch': 141.67}
{'loss': 0.094, 'learning_rate': 0.0008324028996526611, 'epoch': 144.44}
{'loss': 0.0935, 'learning_rate': 0.0008256444970211793, 'epoch': 147.22}
{'loss': 0.0906, 'learning_rate': 0.000818781150845306, 'epoch': 150.0}
{'loss': 0.0909, 'learning_rate': 0.0008118150729354068, 'epoch': 152.78}
{'eval_loss': 0.5312919020652771, 'eval_wer': 0.47133312122176263, 'eval_runtime': 133.5185, 'eval_samples_per_second': 19.54, 'eval_steps_per_second': 0.247, 'epoch': 152.78}
{'loss': 0.0881, 'learning_rate': 0.0008047485082086017, 'epoch': 155.56}
{'loss': 0.0921, 'learning_rate': 0.0007975837339653084, 'epoch': 158.33}
{'loss': 0.0838, 'learning_rate': 0.0007903230591553503, 'epoch': 161.11}
{'loss': 0.0775, 'learning_rate': 0.0007829688236338643, 'epoch': 163.89}
{'loss': 0.0807, 'learning_rate': 0.0007755233974072504, 'epoch': 166.67}
{'eval_loss': 0.5847254991531372, 'eval_wer': 0.44708876869233216, 'eval_runtime': 133.1812, 'eval_samples_per_second': 19.59, 'eval_steps_per_second': 0.248, 'epoch': 166.67}
{'loss': 0.0801, 'learning_rate': 0.0007679891798694021, 'epoch': 169.44}
{'loss': 0.0952, 'learning_rate': 0.000760368599028469, 'epoch': 172.22}
{'loss': 0.0813, 'learning_rate': 0.0007526641107243966, 'epoch': 175.0}
{'loss': 0.0795, 'learning_rate': 0.0007448781978374969, 'epoch': 177.78}
{'loss': 0.0767, 'learning_rate': 0.0007370133694883051, 'epoch': 180.56}
{'eval_loss': 0.5318175554275513, 'eval_wer': 0.4534521158129176, 'eval_runtime': 135.804, 'eval_samples_per_second': 19.212, 'eval_steps_per_second': 0.243, 'epoch': 180.56}
{'loss': 0.0762, 'learning_rate': 0.0007290721602289801, 'epoch': 183.33}
{'loss': 0.0729, 'learning_rate': 0.0007210571292265096, 'epoch': 186.11}
{'loss': 0.0751, 'learning_rate': 0.0007129708594379816, 'epoch': 188.89}
{'loss': 0.0743, 'learning_rate': 0.0007048159567781905, 'epoch': 191.67}
{'loss': 0.0729, 'learning_rate': 0.0006965950492798443, 'epoch': 194.44}
{'eval_loss': 0.5796982645988464, 'eval_wer': 0.4584155265669742, 'eval_runtime': 132.2018, 'eval_samples_per_second': 19.735, 'eval_steps_per_second': 0.25, 'epoch': 194.44}
{'loss': 0.0791, 'learning_rate': 0.0006883107862466433, 'epoch': 197.22}
{'loss': 0.0737, 'learning_rate': 0.0006799658373995054, 'epoch': 200.0}
{'loss': 0.0694, 'learning_rate': 0.0006715628920162093, 'epoch': 202.78}
{'loss': 0.0687, 'learning_rate': 0.0006631046580647384, 'epoch': 205.56}
{'loss': 0.072, 'learning_rate': 0.0006545938613305993, 'epoch': 208.33}
{'eval_loss': 0.5266388654708862, 'eval_wer': 0.44339802736239264, 'eval_runtime': 130.3709, 'eval_samples_per_second': 20.012, 'eval_steps_per_second': 0.253, 'epoch': 208.33}
{'loss': 0.0647, 'learning_rate': 0.000646033244538398, 'epoch': 211.11}
{'loss': 0.0631, 'learning_rate': 0.000637425566467959, 'epoch': 213.89}
{'loss': 0.0606, 'learning_rate': 0.0006287736010652693, 'epoch': 216.67}
{'loss': 0.0645, 'learning_rate': 0.000620080136548535, 'epoch': 219.44}
{'loss': 0.0593, 'learning_rate': 0.0006113479745096386, 'epoch': 222.22}
{'eval_loss': 0.579177737236023, 'eval_wer': 0.43888005090677695, 'eval_runtime': 133.0716, 'eval_samples_per_second': 19.606, 'eval_steps_per_second': 0.248, 'epoch': 222.22}
{'loss': 0.0596, 'learning_rate': 0.0006025799290112865, 'epoch': 225.0}
{'loss': 0.0606, 'learning_rate': 0.0005937788256801371, 'epoch': 227.78}
{'loss': 0.0591, 'learning_rate': 0.0005849475007962031, 'epoch': 230.56}
{'loss': 0.0544, 'learning_rate': 0.0005760888003788196, 'epoch': 233.33}
{'loss': 0.0524, 'learning_rate': 0.0005672055792694753, 'epoch': 236.11}
{'eval_loss': 0.5740395784378052, 'eval_wer': 0.43732103086223356, 'eval_runtime': 132.5642, 'eval_samples_per_second': 19.681, 'eval_steps_per_second': 0.249, 'epoch': 236.11}
{'loss': 0.0504, 'learning_rate': 0.0005583007002117978, 'epoch': 238.89}
{'loss': 0.0494, 'learning_rate': 0.0005493770329289953, 'epoch': 241.67}
{'loss': 0.0484, 'learning_rate': 0.0005404374531990474, 'epoch': 244.44}
{'loss': 0.0489, 'learning_rate': 0.0005314848419279442, 'epoch': 247.22}
{'loss': 0.0494, 'learning_rate': 0.0005225220842212748, 'epoch': 250.0}
{'eval_loss': 0.5520852208137512, 'eval_wer': 0.44358892777601017, 'eval_runtime': 133.6258, 'eval_samples_per_second': 19.525, 'eval_steps_per_second': 0.247, 'epoch': 250.0}
{'loss': 0.0499, 'learning_rate': 0.0005135520684544602, 'epoch': 252.78}
{'loss': 0.0487, 'learning_rate': 0.0005045776853419344, 'epoch': 255.56}
{'loss': 0.0523, 'learning_rate': 0.0004956018270055703, 'epoch': 258.33}
{'loss': 0.0479, 'learning_rate': 0.00048662738604265206, 'epoch': 261.11}
{'loss': 0.0469, 'learning_rate': 0.0004776572545936954, 'epoch': 263.89}
{'eval_loss': 0.5630596876144409, 'eval_wer': 0.4345847916003818, 'eval_runtime': 130.7272, 'eval_samples_per_second': 19.958, 'eval_steps_per_second': 0.252, 'epoch': 263.89}
{'loss': 0.0467, 'learning_rate': 0.0004686943234104146, 'epoch': 266.67}
{'loss': 0.0438, 'learning_rate': 0.00045974148092413597, 'epoch': 269.44}
{'loss': 0.0422, 'learning_rate': 0.00045080161231496064, 'epoch': 272.22}
{'loss': 0.042, 'learning_rate': 0.00044187759858197486, 'epoch': 275.0}
{'loss': 0.0417, 'learning_rate': 0.0004329723156148063, 'epoch': 277.78}
{'eval_loss': 0.5777372121810913, 'eval_wer': 0.43057588291441296, 'eval_runtime': 130.7332, 'eval_samples_per_second': 19.957, 'eval_steps_per_second': 0.252, 'epoch': 277.78}
{'loss': 0.0406, 'learning_rate': 0.0004240886332668285, 'epoch': 280.56}
{'loss': 0.0385, 'learning_rate': 0.00041522941443031025, 'epoch': 283.33}
{'loss': 0.0384, 'learning_rate': 0.0004063975141138074, 'epoch': 286.11}
{'loss': 0.0369, 'learning_rate': 0.00039759577852209704, 'epoch': 288.89}
{'loss': 0.0379, 'learning_rate': 0.0003888270441389465, 'epoch': 291.67}
{'eval_loss': 0.5934527516365051, 'eval_wer': 0.42828507795100224, 'eval_runtime': 131.6738, 'eval_samples_per_second': 19.814, 'eval_steps_per_second': 0.251, 'epoch': 291.67}
{'loss': 0.0399, 'learning_rate': 0.0003800941368130171, 'epoch': 294.44}
{'loss': 0.0359, 'learning_rate': 0.0003713998708471946, 'epoch': 297.22}
{'loss': 0.0345, 'learning_rate': 0.0003627470480916394, 'epoch': 300.0}
{'loss': 0.0346, 'learning_rate': 0.00035413845704085215, 'epoch': 302.78}
{'loss': 0.0319, 'learning_rate': 0.00034557687193504045, 'epoch': 305.56}
{'eval_loss': 0.5892442464828491, 'eval_wer': 0.41880369074132995, 'eval_runtime': 130.692, 'eval_samples_per_second': 19.963, 'eval_steps_per_second': 0.253, 'epoch': 305.56}
{'loss': 0.0318, 'learning_rate': 0.00033706505186608136, 'epoch': 308.33}
{'loss': 0.0301, 'learning_rate': 0.00032860573988836573, 'epoch': 311.11}
{'loss': 0.0322, 'learning_rate': 0.0003202016621348088, 'epoch': 313.89}
{'loss': 0.0331, 'learning_rate': 0.00031185552693831594, 'epoch': 316.67}
{'loss': 0.0311, 'learning_rate': 0.0003035700239589846, 'epoch': 319.44}
{'eval_loss': 0.6010639071464539, 'eval_wer': 0.4210626789691378, 'eval_runtime': 132.2371, 'eval_samples_per_second': 19.73, 'eval_steps_per_second': 0.25, 'epoch': 319.44}
{'loss': 0.0291, 'learning_rate': 0.00029534782331732204, 'epoch': 322.22}
{'loss': 0.0287, 'learning_rate': 0.00028719157473376303, 'epoch': 325.0}
{'loss': 0.0279, 'learning_rate': 0.00027918443530500935, 'epoch': 327.78}
{'loss': 0.0294, 'learning_rate': 0.0002711672294401907, 'epoch': 330.56}
{'loss': 0.0269, 'learning_rate': 0.00026322376817181157, 'epoch': 333.33}
{'eval_loss': 0.5972930192947388, 'eval_wer': 0.4232262169901368, 'eval_runtime': 133.0245, 'eval_samples_per_second': 19.613, 'eval_steps_per_second': 0.248, 'epoch': 333.33}
{'loss': 0.0251, 'learning_rate': 0.0002553566113926006, 'epoch': 336.11}
{'loss': 0.0261, 'learning_rate': 0.00024756829440508705, 'epoch': 338.89}
{'loss': 0.025, 'learning_rate': 0.0002398613271045611, 'epoch': 341.67}
{'loss': 0.0264, 'learning_rate': 0.00023223819317022654, 'epoch': 344.44}
{'loss': 0.0271, 'learning_rate': 0.00022470134926479935, 'epoch': 347.22}
{'eval_loss': 0.6094068288803101, 'eval_wer': 0.4428253261215399, 'eval_runtime': 132.9769, 'eval_samples_per_second': 19.62, 'eval_steps_per_second': 0.248, 'epoch': 347.22}
{'loss': 0.0246, 'learning_rate': 0.00021725322424281286, 'epoch': 350.0}
{'loss': 0.0245, 'learning_rate': 0.00020989621836788614, 'epoch': 352.78}
{'loss': 0.0229, 'learning_rate': 0.00020263270253920428, 'epoch': 355.56}
{'loss': 0.0229, 'learning_rate': 0.00019546501752746288, 'epoch': 358.33}
{'loss': 0.0236, 'learning_rate': 0.0001883954732205217, 'epoch': 361.11}
{'eval_loss': 0.611167311668396, 'eval_wer': 0.42373528475978367, 'eval_runtime': 132.3797, 'eval_samples_per_second': 19.708, 'eval_steps_per_second': 0.249, 'epoch': 361.11}
{'loss': 0.0229, 'learning_rate': 0.00018142634787900996, 'epoch': 363.89}
{'loss': 0.0221, 'learning_rate': 0.00017455988740212576, 'epoch': 366.67}
{'loss': 0.0208, 'learning_rate': 0.0001677983046038624, 'epoch': 369.44}
{'loss': 0.0208, 'learning_rate': 0.00016114377849989813, 'epoch': 372.22}
{'loss': 0.0202, 'learning_rate': 0.00015459845360537756, 'epoch': 375.0}
{'eval_loss': 0.605171799659729, 'eval_wer': 0.4233534839325485, 'eval_runtime': 132.1498, 'eval_samples_per_second': 19.743, 'eval_steps_per_second': 0.25, 'epoch': 375.0}
{'loss': 0.0197, 'learning_rate': 0.0001481644392438113, 'epoch': 377.78}
{'loss': 0.0195, 'learning_rate': 0.00014184380886731607, 'epoch': 380.56}
{'loss': 0.0186, 'learning_rate': 0.0001356385993884165, 'epoch': 383.33}
{'loss': 0.0186, 'learning_rate': 0.00012955081052362072, 'epoch': 386.11}
{'loss': 0.0175, 'learning_rate': 0.00012358240414898276, 'epoch': 388.89}
{'eval_loss': 0.631072998046875, 'eval_wer': 0.4188673242125358, 'eval_runtime': 131.5592, 'eval_samples_per_second': 19.831, 'eval_steps_per_second': 0.251, 'epoch': 388.89}
{'loss': 0.0164, 'learning_rate': 0.00011773530366786184, 'epoch': 391.67}
{'loss': 0.0166, 'learning_rate': 0.0001120113933910767, 'epoch': 394.44}
{'loss': 0.0163, 'learning_rate': 0.000106412517929661, 'epoch': 397.22}
{'loss': 0.0159, 'learning_rate': 0.00010094048160041091, 'epoch': 400.0}
{'loss': 0.0158, 'learning_rate': 9.559704784441814e-05, 'epoch': 402.78}
{'eval_loss': 0.618556797504425, 'eval_wer': 0.4137130130448616, 'eval_runtime': 132.4333, 'eval_samples_per_second': 19.7, 'eval_steps_per_second': 0.249, 'epoch': 402.78}
{'loss': 0.015, 'learning_rate': 9.038393865877726e-05, 'epoch': 405.56}
{'loss': 0.0156, 'learning_rate': 8.530283404164612e-05, 'epoch': 408.33}
{'loss': 0.0146, 'learning_rate': 8.03553714508441e-05, 'epoch': 411.11}
{'loss': 0.0146, 'learning_rate': 7.554314527615763e-05, 'epoch': 413.89}
{'loss': 0.0141, 'learning_rate': 7.086770632552514e-05, 'epoch': 416.67}
{'eval_loss': 0.6352781057357788, 'eval_wer': 0.412249443207127, 'eval_runtime': 132.6132, 'eval_samples_per_second': 19.674, 'eval_steps_per_second': 0.249, 'epoch': 416.67}
{'loss': 0.0138, 'learning_rate': 6.633056132526849e-05, 'epoch': 419.44}
{'loss': 0.0138, 'learning_rate': 6.193317243452734e-05, 'epoch': 422.22}
{'loss': 0.0131, 'learning_rate': 5.7676956774059205e-05, 'epoch': 425.0}
{'loss': 0.0131, 'learning_rate': 5.356328596955179e-05, 'epoch': 427.78}
{'loss': 0.013, 'learning_rate': 4.963246731369253e-05, 'epoch': 430.56}
{'eval_loss': 0.6343812942504883, 'eval_wer': 0.40687241489023224, 'eval_runtime': 131.5808, 'eval_samples_per_second': 19.828, 'eval_steps_per_second': 0.251, 'epoch': 430.56}
{'loss': 0.0127, 'learning_rate': 4.580635924447718e-05, 'epoch': 433.33}
{'loss': 0.0126, 'learning_rate': 4.212662149914887e-05, 'epoch': 436.11}
{'loss': 0.0125, 'learning_rate': 3.859443992523443e-05, 'epoch': 438.89}
{'loss': 0.0127, 'learning_rate': 3.521095281819786e-05, 'epoch': 441.67}
{'loss': 0.0117, 'learning_rate': 3.1977250554608205e-05, 'epoch': 444.44}
{'eval_loss': 0.6442536115646362, 'eval_wer': 0.40801781737193765, 'eval_runtime': 132.782, 'eval_samples_per_second': 19.649, 'eval_steps_per_second': 0.249, 'epoch': 444.44}
{'loss': 0.0122, 'learning_rate': 2.8894375240750613e-05, 'epoch': 447.22}
{'loss': 0.0127, 'learning_rate': 2.5963320376792666e-05, 'epoch': 450.0}
{'loss': 0.0125, 'learning_rate': 2.318503053661475e-05, 'epoch': 452.78}
{'loss': 0.0121, 'learning_rate': 2.056040106340823e-05, 'epoch': 455.56}
{'loss': 0.0122, 'learning_rate': 1.809027778113803e-05, 'epoch': 458.33}
{'eval_loss': 0.6512282490730286, 'eval_wer': 0.4069360483614381, 'eval_runtime': 133.1307, 'eval_samples_per_second': 19.597, 'eval_steps_per_second': 0.248, 'epoch': 458.33}
{'loss': 0.0113, 'learning_rate': 1.5775456721964065e-05, 'epoch': 461.11}
{'loss': 0.0113, 'learning_rate': 1.3616683869708658e-05, 'epoch': 463.89}
{'loss': 0.0117, 'learning_rate': 1.1614654919452406e-05, 'epoch': 466.67}
{'loss': 0.0112, 'learning_rate': 9.770015053337078e-06, 'epoch': 469.44}
{'loss': 0.0115, 'learning_rate': 8.083358732646206e-06, 'epoch': 472.22}
{'eval_loss': 0.6556624174118042, 'eval_wer': 0.4069360483614381, 'eval_runtime': 132.4623, 'eval_samples_per_second': 19.696, 'eval_steps_per_second': 0.249, 'epoch': 472.22}
{'loss': 0.0111, 'learning_rate': 6.555229506231608e-06, 'epoch': 475.0}
{'loss': 0.0118, 'learning_rate': 5.186119835347169e-06, 'epoch': 477.78}
{'loss': 0.0112, 'learning_rate': 3.976470934946075e-06, 'epoch': 480.56}
{'loss': 0.0117, 'learning_rate': 2.92667263149321e-06, 'epoch': 483.33}
{'loss': 0.0111, 'learning_rate': 2.037063237338044e-06, 'epoch': 486.11}
{'eval_loss': 0.6513079404830933, 'eval_wer': 0.4059497295577474, 'eval_runtime': 131.4726, 'eval_samples_per_second': 19.844, 'eval_steps_per_second': 0.251, 'epoch': 486.11}
{'loss': 0.0109, 'learning_rate': 1.3144256109794418e-06, 'epoch': 488.89}
{'loss': 0.0112, 'learning_rate': 7.44394288673067e-07, 'epoch': 491.67}
{'loss': 0.0113, 'learning_rate': 3.352551457218578e-07, 'epoch': 494.44}
{'loss': 0.011, 'learning_rate': 8.714003300058781e-08, 'epoch': 497.22}
{'loss': 0.0113, 'learning_rate': 1.2890911316176812e-10, 'epoch': 500.0}
{'eval_loss': 0.6521313190460205, 'eval_wer': 0.40601336302895324, 'eval_runtime': 132.4384, 'eval_samples_per_second': 19.7, 'eval_steps_per_second': 0.249, 'epoch': 500.0}
{'train_runtime': 176050.2174, 'train_samples_per_second': 16.203, 'train_steps_per_second': 0.102, 'train_loss': 0.11533763631184896, 'epoch': 500.0}
***** train metrics *****
  epoch                    =              500.0
  train_loss               =             0.1153
  train_runtime            = 2 days, 0:54:10.21
  train_samples            =               5705
  train_samples_per_second =             16.203
  train_steps_per_second   =              0.102
01/28/2022 11:53:03 - INFO - __main__ - *** Evaluate ***