wav2vec2-large-xlsr-53-intent-classification-ori

This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7682
• Accuracy: 0.4167

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: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 45

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.2017	1.0	14	2.2113	0.1042
2.2013	2.0	28	2.2078	0.0625
2.197	3.0	42	2.1981	0.0625
2.1917	4.0	56	2.1760	0.3125
2.1759	5.0	70	2.1534	0.3333
2.1713	6.0	84	2.1305	0.3333
2.1431	7.0	98	2.1044	0.3333
2.1366	8.0	112	2.0838	0.3333
2.1482	9.0	126	2.0649	0.3333
2.1074	10.0	140	2.0532	0.3333
2.1066	11.0	154	2.0506	0.3333
2.089	12.0	168	2.0624	0.3333
2.0625	13.0	182	2.0580	0.3333
2.1106	14.0	196	2.0419	0.3333
2.0714	15.0	210	2.0350	0.3333
2.0256	16.0	224	2.0333	0.3333
2.1226	17.0	238	2.0286	0.3333
2.0451	18.0	252	2.0195	0.3333
2.0822	19.0	266	1.9968	0.3333
2.0991	20.0	280	1.9883	0.3333
2.0537	21.0	294	1.9767	0.3333
1.973	22.0	308	1.9524	0.3333
2.0429	23.0	322	1.9432	0.3333
2.0091	24.0	336	1.9402	0.3333
2.0309	25.0	350	1.9295	0.3333
2.0261	26.0	364	1.9167	0.3333
2.0081	27.0	378	1.9083	0.3333
2.023	28.0	392	1.9013	0.3333
2.0	29.0	406	1.8623	0.375
1.936	30.0	420	1.8483	0.3958
1.9809	31.0	434	1.8344	0.3958
1.9645	32.0	448	1.8428	0.4167
1.9788	33.0	462	1.8372	0.3958
1.9484	34.0	476	1.8246	0.3958
1.9553	35.0	490	1.7941	0.4167
1.9321	36.0	504	1.7824	0.4167
1.9759	37.0	518	1.7884	0.3958
1.9424	38.0	532	1.7875	0.3958
1.9592	39.0	546	1.7901	0.3958
1.9425	40.0	560	1.7812	0.3958
1.8899	41.0	574	1.7736	0.3958
1.9361	42.0	588	1.7711	0.4167
1.9023	43.0	602	1.7711	0.4167
1.9203	44.0	616	1.7688	0.4167
1.8921	45.0	630	1.7682	0.4167

Framework versions

• Transformers 4.20.1
• Pytorch 1.11.0
• Datasets 2.1.0
• Tokenizers 0.12.1