suicide-distilbert-2

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: 1.4305
• Accuracy: {'accuracy': 0.4}
• Precision: 0.2743
• Recall: 0.4
• Fscore: 0.2992

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: 7e-05
• train_batch_size: 32
• eval_batch_size: 32
• 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	Precision	Recall	Fscore
No log	1.0	10	1.6207	{'accuracy': 0.31}	0.0981	0.31	0.1490
No log	2.0	20	1.5899	{'accuracy': 0.31}	0.0961	0.31	0.1467
No log	3.0	30	1.5121	{'accuracy': 0.35}	0.2121	0.35	0.2258
No log	4.0	40	1.4855	{'accuracy': 0.37}	0.2008	0.37	0.2468
No log	5.0	50	1.4565	{'accuracy': 0.36}	0.1844	0.36	0.2381
No log	6.0	60	1.4514	{'accuracy': 0.38}	0.1923	0.38	0.2495
No log	7.0	70	1.4609	{'accuracy': 0.39}	0.2949	0.39	0.2934
No log	8.0	80	1.4545	{'accuracy': 0.4}	0.2776	0.4	0.3212
No log	9.0	90	1.4673	{'accuracy': 0.37}	0.2889	0.37	0.2762
No log	10.0	100	1.4305	{'accuracy': 0.4}	0.2743	0.4	0.2992

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0