metadata

license: apache-2.0
base_model: facebook/wav2vec2-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: SER_model_xapiens
    results: []

SER_model_xapiens

This model is a fine-tuned version of facebook/wav2vec2-base on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.7902
Accuracy: 0.7333

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	1	1.6064	0.1833
No log	2.0	2	1.6067	0.1667
No log	3.0	3	1.6075	0.1
No log	4.0	4	1.6086	0.2
No log	5.0	5	1.6099	0.1667
No log	6.0	6	1.6110	0.1833
No log	7.0	7	1.6132	0.1833
No log	8.0	8	1.6156	0.1667
No log	9.0	9	1.6173	0.25
1.5968	10.0	10	1.6201	0.2833
1.5968	11.0	11	1.6210	0.3167
1.5968	12.0	12	1.6208	0.3
1.5968	13.0	13	1.6162	0.35
1.5968	14.0	14	1.6076	0.3667
1.5968	15.0	15	1.5950	0.4
1.5968	16.0	16	1.5843	0.3833
1.5968	17.0	17	1.5706	0.4
1.5968	18.0	18	1.5506	0.4
1.5968	19.0	19	1.5278	0.4167
1.5049	20.0	20	1.5029	0.4167
1.5049	21.0	21	1.4745	0.4333
1.5049	22.0	22	1.4454	0.4667
1.5049	23.0	23	1.4123	0.4833
1.5049	24.0	24	1.3828	0.4833
1.5049	25.0	25	1.3580	0.4833
1.5049	26.0	26	1.3328	0.4833
1.5049	27.0	27	1.3053	0.4667
1.5049	28.0	28	1.2808	0.4833
1.5049	29.0	29	1.2581	0.4667
1.2782	30.0	30	1.2392	0.45
1.2782	31.0	31	1.2202	0.45
1.2782	32.0	32	1.1858	0.45
1.2782	33.0	33	1.1552	0.4667
1.2782	34.0	34	1.1385	0.4667
1.2782	35.0	35	1.1221	0.4667
1.2782	36.0	36	1.1088	0.4667
1.2782	37.0	37	1.0941	0.4667
1.2782	38.0	38	1.0764	0.4667
1.2782	39.0	39	1.0624	0.4667
1.0507	40.0	40	1.0446	0.45
1.0507	41.0	41	1.0348	0.5
1.0507	42.0	42	1.0177	0.4833
1.0507	43.0	43	1.0010	0.4667
1.0507	44.0	44	0.9913	0.4833
1.0507	45.0	45	0.9695	0.4833
1.0507	46.0	46	0.9485	0.5167
1.0507	47.0	47	0.9873	0.4667
1.0507	48.0	48	0.9856	0.4667
1.0507	49.0	49	0.9353	0.4667
0.906	50.0	50	0.9049	0.5333
0.906	51.0	51	0.8960	0.5167
0.906	52.0	52	0.8972	0.5167
0.906	53.0	53	0.8801	0.5167
0.906	54.0	54	0.8654	0.55
0.906	55.0	55	0.8813	0.55
0.906	56.0	56	0.8894	0.5333
0.906	57.0	57	0.8758	0.5333
0.906	58.0	58	0.8794	0.5333
0.906	59.0	59	0.8952	0.55
0.8081	60.0	60	0.8949	0.55
0.8081	61.0	61	0.8789	0.5833
0.8081	62.0	62	0.8545	0.6167
0.8081	63.0	63	0.8440	0.6333
0.8081	64.0	64	0.8344	0.6667
0.8081	65.0	65	0.8368	0.6333
0.8081	66.0	66	0.8504	0.6333
0.8081	67.0	67	0.8604	0.6167
0.8081	68.0	68	0.8637	0.6167
0.8081	69.0	69	0.8641	0.65
0.7364	70.0	70	0.8660	0.65
0.7364	71.0	71	0.8604	0.65
0.7364	72.0	72	0.8546	0.65
0.7364	73.0	73	0.8461	0.65
0.7364	74.0	74	0.8426	0.6833
0.7364	75.0	75	0.8443	0.6833
0.7364	76.0	76	0.8413	0.6833
0.7364	77.0	77	0.8385	0.6833
0.7364	78.0	78	0.8399	0.65
0.7364	79.0	79	0.8432	0.65
0.6787	80.0	80	0.8443	0.6167
0.6787	81.0	81	0.8433	0.6167
0.6787	82.0	82	0.8370	0.6167
0.6787	83.0	83	0.8315	0.6
0.6787	84.0	84	0.8275	0.6167
0.6787	85.0	85	0.8190	0.6333
0.6787	86.0	86	0.8134	0.6667
0.6787	87.0	87	0.8104	0.6833
0.6787	88.0	88	0.8108	0.6833
0.6787	89.0	89	0.8140	0.6833
0.6359	90.0	90	0.8145	0.6833
0.6359	91.0	91	0.8153	0.6667
0.6359	92.0	92	0.8135	0.6833
0.6359	93.0	93	0.8108	0.6833
0.6359	94.0	94	0.8072	0.6833
0.6359	95.0	95	0.8017	0.6833
0.6359	96.0	96	0.7965	0.7
0.6359	97.0	97	0.7928	0.7
0.6359	98.0	98	0.7908	0.7333
0.6359	99.0	99	0.7904	0.7333
0.6092	100.0	100	0.7902	0.7333

Framework versions

Transformers 4.42.0.dev0
Pytorch 2.3.0
Datasets 2.19.2.dev0
Tokenizers 0.19.1