learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 8
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 22
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Accuracy	Acc Dem 5	Acc No 1	Acc No 2	Acc No 3	Binary Dem Accuracy	Binary Dem F1
No log	2.0	34	1.7914	0.2660	0.0	0.0	0.0	0.9615	1.0	1.0
No log	4.0	68	1.7699	0.2553	0.0	0.0	0.1053	0.8462	1.0	1.0
No log	6.0	102	1.7839	0.2660	0.0	0.0	0.1053	0.8846	1.0	1.0
1.7786	8.0	136	1.7959	0.2021	0.0	0.0	1.0	0.0	1.0	1.0
1.7786	10.0	170	1.7681	0.2234	0.0	0.3333	0.7368	0.0	1.0	1.0
1.7786	12.0	204	1.7108	0.3404	0.0	0.3333	0.1053	0.8846	1.0	1.0
1.7376	14.0	238	1.7089	0.3830	0.0	0.3810	0.1579	0.9615	1.0	1.0
1.7376	16.0	272	1.6353	0.4362	0.0	0.6190	0.2105	0.9231	1.0	1.0
1.7376	18.0	306	1.6161	0.4362	0.0833	0.5714	0.2105	0.9231	1.0	1.0
1.5323	20.0	340	1.6172	0.3936	0.0	0.6190	0.2632	0.7308	1.0	1.0
1.5323	22.0	374	1.6181	0.4149	0.0833	0.6190	0.3158	0.7308	1.0	1.0

dpraitality/experts_3split_wav2vec2-large-xls-r-300m-dm32

experts_3split_wav2vec2-large-xls-r-300m-dm32