wavlm-emotion

This model is a fine-tuned version of microsoft/wavlm-base on the None dataset. It achieves the following results on the evaluation set:

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 32
eval_batch_size: 32
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
num_epochs: 50
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss
No log	1.0	96	1.9509
1.9473	2.0	192	1.9486
1.9469	3.0	288	1.9512
1.9469	4.0	384	1.9495
1.9471	5.0	480	1.9488
1.9468	6.0	576	1.9474
1.9478	7.0	672	1.9510
1.9487	8.0	768	1.9510
1.9487	9.0	864	1.9466
1.9471	10.0	960	1.9465
1.9471	11.0	1056	1.9505
1.9477	12.0	1152	1.9467
1.9479	13.0	1248	1.9465
1.948	14.0	1344	1.9439
1.9473	15.0	1440	1.9488
1.9488	16.0	1536	1.9432
1.9472	17.0	1632	1.9491
1.9465	18.0	1728	1.9461
1.9468	19.0	1824	1.9507
1.9481	20.0	1920	1.9480
1.9471	21.0	2016	1.9442
1.9477	22.0	2112	1.9466
1.9474	23.0	2208	1.9504
1.9468	24.0	2304	1.9496
1.9476	25.0	2400	1.9464
1.9476	26.0	2496	1.9466
1.9471	27.0	2592	1.9473
1.9467	28.0	2688	1.9485
1.9468	29.0	2784	1.9477
1.9471	30.0	2880	1.9484
1.9458	31.0	2976	1.9472
1.9476	32.0	3072	1.9475
1.9468	33.0	3168	1.9482
1.9468	34.0	3264	1.9495
1.9463	35.0	3360	1.9497
1.9474	36.0	3456	1.9490
1.9462	37.0	3552	1.9481
1.9458	38.0	3648	1.9490
1.9461	39.0	3744	1.9486
1.9446	40.0	3840	1.9488
1.946	41.0	3936	1.9490
1.9467	42.0	4032	1.9487
1.9466	43.0	4128	1.9485
1.9463	44.0	4224	1.9486
1.9459	45.0	4320	1.9486
1.9458	46.0	4416	1.9487
1.9464	47.0	4512	1.9485
1.946	48.0	4608	1.9485
1.9459	49.0	4704	1.9485
1.9459	50.0	4800	1.9485