soba2204/wav2vec2-asv19training

working

This model is a fine-tuned version of lighteternal/wav2vec2-large-xlsr-53-greek on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0076
• Accuracy: 0.9984

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 550

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3693	0.0039	10	0.4063	0.8983
0.4466	0.0079	20	0.4230	0.8983
0.3205	0.0118	30	0.2852	0.8983
0.3037	0.0158	40	0.2173	0.9009
0.3259	0.0197	50	0.2704	0.8554
0.2012	0.0236	60	0.1175	0.9446
0.1654	0.0276	70	0.0382	0.9894
0.0635	0.0315	80	0.0470	0.9819
0.0182	0.0355	90	0.2888	0.9159
0.2049	0.0394	100	0.1801	0.9486
0.1128	0.0433	110	0.0171	0.9953
0.0246	0.0473	120	0.0321	0.9896
0.1879	0.0512	130	0.0337	0.9903
0.0582	0.0552	140	0.0077	0.9978
0.0052	0.0591	150	0.1915	0.9494
0.0023	0.0630	160	0.0554	0.9866
0.0634	0.0670	170	0.0636	0.9850
0.0058	0.0709	180	0.2431	0.9446
0.0365	0.0749	190	0.0071	0.9982
0.0046	0.0788	200	0.0180	0.9953
0.0227	0.0827	210	0.0186	0.9955
0.0003	0.0867	220	0.0268	0.9943
0.1036	0.0906	230	0.0303	0.9929
0.0764	0.0946	240	0.0070	0.9986
0.0013	0.0985	250	0.0051	0.9990
0.1667	0.1024	260	0.0078	0.9986
0.0399	0.1064	270	0.0085	0.9982
0.0058	0.1103	280	0.0387	0.9894
0.0767	0.1143	290	0.0238	0.9931
0.0011	0.1182	300	0.0148	0.9965
0.0019	0.1221	310	0.0123	0.9972
0.0518	0.1261	320	0.0087	0.9976
0.0019	0.1300	330	0.0213	0.9943
0.0006	0.1340	340	0.0309	0.9927
0.0007	0.1379	350	0.0360	0.9911
0.0007	0.1418	360	0.0356	0.9913
0.0022	0.1458	370	0.0431	0.9901
0.0009	0.1497	380	0.0349	0.9919
0.0005	0.1537	390	0.0247	0.9941
0.0242	0.1576	400	0.0211	0.9953
0.0003	0.1615	410	0.0097	0.9976
0.0003	0.1655	420	0.0076	0.9986
0.0004	0.1694	430	0.0070	0.9988
0.0003	0.1734	440	0.0070	0.9990
0.0198	0.1773	450	0.0036	0.9990
0.0006	0.1812	460	0.0027	0.9994
0.0002	0.1852	470	0.0025	0.9994
0.0011	0.1891	480	0.0034	0.9992
0.0006	0.1931	490	0.0049	0.9990
0.0002	0.1970	500	0.0054	0.9990
0.1014	0.2009	510	0.0066	0.9990
0.0004	0.2049	520	0.0069	0.9990
0.0004	0.2088	530	0.0070	0.9990
0.001	0.2128	540	0.0075	0.9986
0.0377	0.2167	550	0.0076	0.9984

Framework versions

• Transformers 4.41.0.dev0
• Pytorch 2.1.2
• Datasets 2.19.2.dev0
• Tokenizers 0.19.1