learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.1
num_epochs: 10
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Constrast Loss	Div Loss
845.359	0.6289	100	955.3924	941.3030	140.8944
774.0615	1.2579	200	921.1964	908.8786	123.1788
839.9893	1.8868	300	932.0772	920.1297	119.4754
994.4874	2.5157	400	885.8154	874.5105	113.0493
772.1645	3.1447	500	933.2662	921.5520	117.1420
1298.7873	3.7736	600	920.5379	909.0538	114.8405
775.2367	4.4025	700	904.8465	893.6350	112.1154
1198.9295	5.0314	800	870.4927	859.6562	108.3643
732.0786	5.6604	900	911.9991	900.8566	111.4251
1380.1144	6.2893	1000	917.1681	905.9143	112.5378
776.9523	6.9182	1100	867.9667	857.2078	107.5892
1467.2303	7.5472	1200	897.4872	886.5748	109.1243
715.7438	8.1761	1300	885.5976	874.7515	108.4605
1226.158	8.8050	1400	927.9101	916.7936	111.1654
624.3603	9.4340	1500	902.7379	891.6586	110.7930

nguyencaong/wav2vec2-pretraining-demo

wav2vec2-pretraining-demo