depression_tweet

This model is a fine-tuned version of rafalposwiata/deproberta-large-v1 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0646
• Accuracy: 0.9836
• Precision: 0.9656
• Recall: 0.9977
• F1: 0.9814

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: 128
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
No log	0.2	50	0.1556	0.9684	0.9508	0.9777	0.9641
No log	0.4	100	0.1399	0.9646	0.9354	0.9865	0.9603
No log	0.61	150	0.1118	0.9631	0.9279	0.9920	0.9589
No log	0.81	200	0.1090	0.9659	0.9333	0.9922	0.9619
No log	1.01	250	0.0819	0.9759	0.9556	0.9905	0.9727
No log	1.21	300	0.0548	0.9831	0.9831	0.9777	0.9804
No log	1.42	350	0.1162	0.9587	0.9435	0.9624	0.9529
No log	1.62	400	0.1167	0.9657	0.9303	0.9955	0.9618
No log	1.82	450	0.0859	0.9776	0.9549	0.9955	0.9747
0.0575	2.02	500	0.0564	0.9848	0.9707	0.9950	0.9827
0.0575	2.23	550	0.0591	0.9839	0.9693	0.9945	0.9817
0.0575	2.43	600	0.0913	0.9814	0.9623	0.9962	0.9790
0.0575	2.63	650	0.0633	0.9847	0.9686	0.9970	0.9826
0.0575	2.83	700	0.1171	0.9762	0.9493	0.9985	0.9733
0.0575	3.04	750	0.0646	0.9836	0.9656	0.9977	0.9814

Framework versions

• Transformers 4.37.1
• Pytorch 2.0.1
• Datasets 2.16.1
• Tokenizers 0.15.1