finetuned_roberta-base-on-IEMOCAP_4

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

• Loss: 1.5143
• Accuracy: 0.6867

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: 2e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.8876	1.0	113	0.9530	0.6341
0.7153	2.0	226	0.8473	0.6807
0.5127	3.0	339	0.8841	0.6907
0.3844	4.0	452	0.9751	0.7062
0.3104	5.0	565	0.9363	0.7029
0.2374	6.0	678	1.1047	0.6896
0.2128	7.0	791	1.1239	0.7040
0.2104	8.0	904	1.1717	0.7129
0.1896	9.0	1017	1.1580	0.6973
0.1211	10.0	1130	1.2320	0.7029
0.1737	11.0	1243	1.3389	0.6973
0.1022	12.0	1356	1.3537	0.7062
0.0934	13.0	1469	1.3742	0.7140
0.0987	14.0	1582	1.4039	0.7151
0.1065	15.0	1695	1.4150	0.7173

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1