bert-base-uncased-finetuned-iemocap8

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

• Loss: 1.8968
• Accuracy: 0.6654
• F1: 0.6723

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: 4.319412088241492e-05
• train_batch_size: 64
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
No log	1.0	51	1.0531	0.5597	0.5655
1.0284	2.0	102	0.9370	0.6227	0.6304
1.0284	3.0	153	0.8796	0.6722	0.6765
0.4432	4.0	204	0.9785	0.6654	0.6727
0.4432	5.0	255	1.0664	0.6586	0.6634
0.2492	6.0	306	1.1291	0.6499	0.6606
0.2492	7.0	357	1.1847	0.6702	0.6777
0.1707	8.0	408	1.4084	0.6508	0.6534
0.1707	9.0	459	1.3468	0.6702	0.6762
0.1461	10.0	510	1.4245	0.6634	0.6710
0.1461	11.0	561	1.4865	0.6499	0.6600
0.1262	12.0	612	1.4616	0.6576	0.6656
0.1262	13.0	663	1.5335	0.6663	0.6711
0.1203	14.0	714	1.4855	0.6731	0.6806
0.1203	15.0	765	1.5825	0.6712	0.6792
0.1023	16.0	816	1.7145	0.6731	0.6794
0.1023	17.0	867	1.6676	0.6751	0.6823
0.0976	18.0	918	1.8013	0.6693	0.6719
0.0976	19.0	969	1.7192	0.6673	0.6755
0.0937	20.0	1020	1.7837	0.6654	0.6731
0.0937	21.0	1071	1.7779	0.6760	0.6831
0.0901	22.0	1122	1.8352	0.6615	0.6687
0.0901	23.0	1173	1.8601	0.6596	0.6656
0.0844	24.0	1224	1.9129	0.6625	0.6719
0.0844	25.0	1275	1.8507	0.6731	0.6784
0.0829	26.0	1326	1.8582	0.6673	0.6735
0.0829	27.0	1377	1.8670	0.6770	0.6825
0.0839	28.0	1428	1.8763	0.6741	0.6800
0.0839	29.0	1479	1.8925	0.6702	0.6769
0.0802	30.0	1530	1.8968	0.6654	0.6723

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.0
• Tokenizers 0.13.2