learning_rate: 0.0001
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Use 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_steps: 50
num_epochs: 3
mixed_precision_training: Native AMP

Training Loss	Epoch	Step	Validation Loss	Cer
28.9281	0.1289	200	5.0567	100.0
4.8764	0.2579	400	4.6646	100.0
4.734	0.3868	600	4.6652	99.7709
4.6908	0.5158	800	4.6114	97.9147
4.6114	0.6447	1000	4.5419	97.6328
4.5786	0.7737	1200	4.5616	97.4507
4.5379	0.9026	1400	4.5366	96.3052
4.4209	1.0316	1600	4.4376	96.9396
4.1586	1.1605	1800	3.9708	77.6903
3.3937	1.2895	2000	3.1347	59.2751
2.792	1.4184	2200	2.7514	50.6403
2.4482	1.5474	2400	2.3231	42.6046
2.21	1.6763	2600	2.1510	41.2065
2.0125	1.8053	2800	1.9830	36.6953
1.813	1.9342	3000	1.7184	32.3896
1.6918	2.0632	3200	1.6239	31.3146
1.5507	2.1921	3400	1.4690	29.0061
1.4465	2.3211	3600	1.4636	29.1295
1.3874	2.4500	3800	1.3202	26.0456
1.3271	2.5790	4000	1.3076	26.3980
1.2861	2.7079	4200	1.2258	24.7415
1.2229	2.8369	4400	1.2062	24.4831
1.1965	2.9658	4600	1.2018	24.5242

Gummybear05/wav2vec2-E50_freq_pause_

wav2vec2-E50_freq_pause_