gpt2-finetuned-depression

This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6647
• Precision: 0.8917
• Recall: 0.8648
• F1: 0.8772
• Accuracy: 0.9104

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	0.4102	0.8382	0.7154	0.7590	0.8614
0.652	2.0	938	0.3335	0.8669	0.8254	0.8439	0.8838
0.3283	3.0	1407	0.4627	0.8978	0.8464	0.8689	0.9041
0.1988	4.0	1876	0.5853	0.9021	0.8324	0.8628	0.8987
0.1613	5.0	2345	0.6426	0.9034	0.8385	0.8672	0.8987
0.1013	6.0	2814	0.6247	0.8682	0.8611	0.8643	0.9041
0.0863	7.0	3283	0.7673	0.8930	0.8375	0.8617	0.8987
0.0757	8.0	3752	0.6647	0.8917	0.8648	0.8772	0.9104
0.0511	9.0	4221	0.6658	0.8768	0.8674	0.8720	0.9030
0.0581	10.0	4690	0.7686	0.9104	0.8595	0.8824	0.9094
0.0311	11.0	5159	0.6830	0.8918	0.8488	0.8685	0.8977
0.0537	12.0	5628	0.7438	0.9078	0.8563	0.8795	0.9062
0.0436	13.0	6097	0.7950	0.8933	0.8438	0.8663	0.8987
0.042	14.0	6566	0.7248	0.8986	0.8507	0.8726	0.9030
0.0374	15.0	7035	0.6973	0.8884	0.8504	0.8681	0.9009
0.0371	16.0	7504	0.7294	0.8874	0.8554	0.8703	0.9030
0.0371	17.0	7973	0.7649	0.8937	0.8486	0.8692	0.9030
0.0318	18.0	8442	0.7576	0.8879	0.8467	0.8657	0.9009
0.0307	19.0	8911	0.7556	0.8937	0.8486	0.8692	0.9030
0.0264	20.0	9380	0.7647	0.8930	0.8486	0.8689	0.9030

Framework versions

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