wav2vec2-base-finetuned-ravdess

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

• Loss: 0.8783
• Accuracy: 0.7535

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: 32
• eval_batch_size: 32
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	9	2.0739	0.1562
2.0781	2.0	18	2.0611	0.1181
2.0668	3.0	27	2.0308	0.2535
2.0429	4.0	36	1.9606	0.2604
1.974	5.0	45	1.8449	0.2847
1.8594	6.0	54	1.7678	0.2917
1.7675	7.0	63	1.7700	0.2708
1.6932	8.0	72	1.6049	0.3889
1.5656	9.0	81	1.5510	0.4444
1.4658	10.0	90	1.4535	0.4583
1.4658	11.0	99	1.4101	0.4514
1.3843	12.0	108	1.3687	0.5
1.3085	13.0	117	1.3333	0.5035
1.2264	14.0	126	1.3208	0.5208
1.1349	15.0	135	1.3048	0.5312
1.0861	16.0	144	1.2428	0.5799
0.9836	17.0	153	1.1886	0.5799
0.9273	18.0	162	1.1574	0.6146
0.8686	19.0	171	1.1356	0.6111
0.814	20.0	180	1.1261	0.6285
0.814	21.0	189	1.0796	0.6007
0.7279	22.0	198	1.0277	0.6493
0.6845	23.0	207	1.0408	0.6840
0.6283	24.0	216	0.9708	0.7153
0.5835	25.0	225	0.9926	0.6875
0.5445	26.0	234	1.0126	0.6840
0.497	27.0	243	0.9502	0.6979
0.4508	28.0	252	0.9432	0.7118
0.4331	29.0	261	0.9246	0.7014
0.4023	30.0	270	0.9649	0.6875
0.4023	31.0	279	0.9114	0.7049
0.3924	32.0	288	0.9460	0.7118
0.3797	33.0	297	0.9605	0.7118
0.3494	34.0	306	0.8505	0.7396
0.3195	35.0	315	0.8830	0.7188
0.3148	36.0	324	0.9352	0.7014
0.2856	37.0	333	0.8551	0.7292
0.2831	38.0	342	0.8505	0.7326
0.2718	39.0	351	0.8800	0.7396
0.2624	40.0	360	0.8991	0.7153
0.2624	41.0	369	0.8724	0.7465
0.2612	42.0	378	0.9138	0.7049
0.2511	43.0	387	0.8914	0.7257
0.2324	44.0	396	0.8783	0.7535
0.2228	45.0	405	0.9215	0.7188
0.2244	46.0	414	0.8904	0.7431
0.2192	47.0	423	0.9142	0.7326
0.217	48.0	432	0.8891	0.7361
0.2146	49.0	441	0.9009	0.7326
0.215	50.0	450	0.8994	0.7361

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3