distilbert-base-uncased-finetuned-depression

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

• Loss: 0.6721
• Precision: 0.9018
• Recall: 0.8881
• F1: 0.8946
• Accuracy: 0.9168

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: 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	469	0.3914	0.9136	0.7542	0.8087	0.8678
0.5449	2.0	938	0.3944	0.8652	0.8677	0.8644	0.8977
0.2679	3.0	1407	0.4355	0.8717	0.8713	0.8703	0.9009
0.1516	4.0	1876	0.4509	0.8757	0.8809	0.8779	0.9083
0.0989	5.0	2345	0.4762	0.8861	0.8846	0.8854	0.9094
0.0666	6.0	2814	0.4829	0.8878	0.8890	0.8883	0.9126
0.0563	7.0	3283	0.5768	0.8918	0.8866	0.8885	0.9115
0.0349	8.0	3752	0.6874	0.8898	0.8644	0.8758	0.8987
0.0444	9.0	4221	0.6256	0.8804	0.8822	0.8790	0.9019
0.0301	10.0	4690	0.6354	0.8897	0.8750	0.8814	0.9030
0.0318	11.0	5159	0.7172	0.8894	0.8682	0.8770	0.9009
0.0222	12.0	5628	0.6906	0.9001	0.8700	0.8834	0.9019
0.0243	13.0	6097	0.7263	0.8898	0.8732	0.8800	0.9019
0.0172	14.0	6566	0.6936	0.8945	0.8766	0.8846	0.9072
0.0204	15.0	7035	0.7428	0.9081	0.8730	0.8889	0.9051
0.0162	16.0	7504	0.7202	0.8966	0.8748	0.8846	0.9062
0.0162	17.0	7973	0.6721	0.9018	0.8881	0.8946	0.9168
0.0172	18.0	8442	0.7664	0.9037	0.8706	0.8854	0.9030
0.0156	19.0	8911	0.7166	0.8985	0.8784	0.8876	0.9094
0.0158	20.0	9380	0.7327	0.8966	0.8748	0.8846	0.9062

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1