test_model_dir

This model is a fine-tuned version of facebook/wav2vec2-base on the minds14 dataset. It achieves the following results on the evaluation set:

• Loss: 2.7181
• Accuracy: 0.0885

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: 3e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	2	2.6433	0.0354
No log	2.0	4	2.6427	0.0265
No log	3.0	6	2.6412	0.0619
No log	4.0	8	2.6391	0.0885
2.6378	5.0	10	2.6384	0.1239
2.6378	6.0	12	2.6380	0.0973
2.6378	7.0	14	2.6375	0.0708
2.6378	8.0	16	2.6415	0.0796
2.6378	9.0	18	2.6399	0.0531
2.6288	10.0	20	2.6450	0.0796
2.6288	11.0	22	2.6450	0.0619
2.6288	12.0	24	2.6452	0.0708
2.6288	13.0	26	2.6479	0.0708
2.6288	14.0	28	2.6496	0.0619
2.6185	15.0	30	2.6522	0.0796
2.6185	16.0	32	2.6558	0.0796
2.6185	17.0	34	2.6567	0.0708
2.6185	18.0	36	2.6572	0.0619
2.6185	19.0	38	2.6611	0.0619
2.6069	20.0	40	2.6629	0.0619
2.6069	21.0	42	2.6621	0.0531
2.6069	22.0	44	2.6663	0.0531
2.6069	23.0	46	2.6672	0.0442
2.6069	24.0	48	2.6645	0.0531
2.599	25.0	50	2.6670	0.0708
2.599	26.0	52	2.6692	0.0531
2.599	27.0	54	2.6653	0.0708
2.599	28.0	56	2.6669	0.0885
2.599	29.0	58	2.6797	0.0619
2.5767	30.0	60	2.6781	0.0354
2.5767	31.0	62	2.6861	0.0265
2.5767	32.0	64	2.6852	0.0442
2.5767	33.0	66	2.6733	0.0442
2.5767	34.0	68	2.6881	0.0708
2.5771	35.0	70	2.6800	0.0708
2.5771	36.0	72	2.6777	0.0619
2.5771	37.0	74	2.6761	0.0708
2.5771	38.0	76	2.6657	0.0619
2.5771	39.0	78	2.6667	0.0708
2.5636	40.0	80	2.6681	0.0708
2.5636	41.0	82	2.6649	0.0796
2.5636	42.0	84	2.6598	0.0796
2.5636	43.0	86	2.6627	0.0619
2.5636	44.0	88	2.6596	0.0796
2.5608	45.0	90	2.6511	0.0796
2.5608	46.0	92	2.6522	0.0708
2.5608	47.0	94	2.6610	0.0708
2.5608	48.0	96	2.6638	0.0531
2.5608	49.0	98	2.6642	0.0619
2.5432	50.0	100	2.6596	0.0796
2.5432	51.0	102	2.6675	0.0885
2.5432	52.0	104	2.6964	0.0885
2.5432	53.0	106	2.7030	0.0531
2.5432	54.0	108	2.7016	0.0531
2.5295	55.0	110	2.6918	0.0619
2.5295	56.0	112	2.6893	0.0619
2.5295	57.0	114	2.6936	0.0708
2.5295	58.0	116	2.6905	0.0885
2.5295	59.0	118	2.6838	0.0796
2.5207	60.0	120	2.6845	0.0708
2.5207	61.0	122	2.6896	0.0708
2.5207	62.0	124	2.6965	0.0796
2.5207	63.0	126	2.6971	0.1062
2.5207	64.0	128	2.6982	0.0973
2.5015	65.0	130	2.7037	0.0885
2.5015	66.0	132	2.7065	0.0973
2.5015	67.0	134	2.7078	0.0973
2.5015	68.0	136	2.7055	0.0973
2.5015	69.0	138	2.7023	0.0973
2.4869	70.0	140	2.6923	0.1062
2.4869	71.0	142	2.6906	0.1062
2.4869	72.0	144	2.6989	0.1062
2.4869	73.0	146	2.7078	0.0885
2.4869	74.0	148	2.7106	0.0973
2.4638	75.0	150	2.7117	0.0796
2.4638	76.0	152	2.7119	0.0796
2.4638	77.0	154	2.7153	0.0708
2.4638	78.0	156	2.7111	0.0708
2.4638	79.0	158	2.7086	0.0885
2.4408	80.0	160	2.7000	0.1150
2.4408	81.0	162	2.6915	0.1062
2.4408	82.0	164	2.6907	0.1062
2.4408	83.0	166	2.6908	0.0973
2.4408	84.0	168	2.6926	0.0796
2.4688	85.0	170	2.6984	0.1062
2.4688	86.0	172	2.7039	0.1062
2.4688	87.0	174	2.7053	0.0973
2.4688	88.0	176	2.7098	0.0796
2.4688	89.0	178	2.7100	0.0885
2.4379	90.0	180	2.7113	0.1062
2.4379	91.0	182	2.7121	0.0973
2.4379	92.0	184	2.7127	0.0973
2.4379	93.0	186	2.7162	0.0973
2.4379	94.0	188	2.7189	0.0973
2.4385	95.0	190	2.7199	0.0885
2.4385	96.0	192	2.7186	0.0796
2.4385	97.0	194	2.7182	0.0885
2.4385	98.0	196	2.7183	0.0885
2.4385	99.0	198	2.7182	0.0885
2.4402	100.0	200	2.7181	0.0885

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1