mentalBERT-wellness-classifier

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

• Loss: 1.0471
• Accuracy: 0.722
• Auc: 0.915
• Precision Class 0: 0.765
• Precision Class 1: 0.762
• Precision Class 2: 0.745
• Precision Class 3: 0.684
• Recall Class 0: 0.736
• Recall Class 1: 0.593
• Recall Class 2: 0.651
• Recall Class 3: 0.796
• F1 Score Class 0: 0.75
• F1 Score Class 1: 0.667
• F1 Score Class 2: 0.695
• F1 Score Class 3: 0.736

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: 0.0002
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Auc	Precision Class 0	Precision Class 1	Precision Class 2	Precision Class 3	Recall Class 0	Recall Class 1	Recall Class 2	Recall Class 3	F1 Score Class 0	F1 Score Class 1	F1 Score Class 2	F1 Score Class 3
0.0887	1.0	140	1.2702	0.73	0.91	0.783	0.773	0.746	0.693	0.679	0.63	0.698	0.806	0.727	0.694	0.721	0.745
0.2058	2.0	280	1.0340	0.718	0.916	0.75	0.72	0.736	0.694	0.736	0.667	0.619	0.786	0.743	0.692	0.672	0.737
0.1921	3.0	420	1.0234	0.739	0.919	0.733	0.731	0.75	0.737	0.83	0.704	0.667	0.745	0.779	0.717	0.706	0.741
0.184	4.0	560	1.0486	0.73	0.917	0.719	0.773	0.738	0.723	0.774	0.63	0.714	0.745	0.745	0.694	0.726	0.734
0.1666	5.0	700	1.0084	0.747	0.918	0.769	0.739	0.738	0.743	0.755	0.63	0.762	0.765	0.762	0.68	0.75	0.754
0.1558	6.0	840	1.0314	0.739	0.917	0.759	0.773	0.738	0.721	0.774	0.63	0.714	0.765	0.766	0.694	0.726	0.743
0.138	7.0	980	1.0733	0.73	0.915	0.776	0.773	0.75	0.693	0.717	0.63	0.667	0.806	0.745	0.694	0.706	0.745
0.1481	8.0	1120	1.0656	0.722	0.914	0.787	0.773	0.741	0.678	0.698	0.63	0.635	0.816	0.74	0.694	0.684	0.741
0.1433	9.0	1260	1.0520	0.718	0.915	0.765	0.762	0.741	0.678	0.736	0.593	0.635	0.796	0.75	0.667	0.684	0.732
0.1577	10.0	1400	1.0471	0.722	0.915	0.765	0.762	0.745	0.684	0.736	0.593	0.651	0.796	0.75	0.667	0.695	0.736

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0