finetuned_bert-base-on-IEMOCAP_4

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

• Loss: 1.3568
• Accuracy: 0.6712

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
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2828	1.0	113	1.2221	0.4557
0.9978	2.0	226	0.9164	0.6341
0.6524	3.0	339	0.8095	0.6851
0.5501	4.0	452	0.8434	0.6874
0.4188	5.0	565	0.8785	0.6951
0.3462	6.0	678	0.9251	0.7084
0.2205	7.0	791	1.0525	0.6951
0.2596	8.0	904	1.0655	0.7062
0.242	9.0	1017	1.1137	0.7062
0.1651	10.0	1130	1.1869	0.7118
0.1943	11.0	1243	1.2023	0.7029
0.1442	12.0	1356	1.2443	0.6962
0.1569	13.0	1469	1.2965	0.7106
0.1396	14.0	1582	1.3718	0.7007
0.1002	15.0	1695	1.3720	0.7062
0.1051	16.0	1808	1.3800	0.7018
0.1604	17.0	1921	1.4143	0.6962
0.1211	18.0	2034	1.4186	0.7018
0.1083	19.0	2147	1.4322	0.7029
0.0939	20.0	2260	1.4358	0.7007

Framework versions

• Transformers 4.30.1
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3