Modelpublic

cuadron11/suicide-distilbert-5-5

sourceHugging Faceapache-2.0updated 3y agoView on Hugging Face

0likes6downloads

Model Card

suicide-distilbert-5-5

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: 1.8866
—Accuracy: {'accuracy': 0.41}
—Precision: 0.3515
—Recall: 0.41
—Fscore: 0.3748

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: 5.5e-05
—trainbatchsize: 32
—evalbatchsize: 32
—seed: 42
—optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
—lrschedulertype: linear
—num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	Fscore
No log	1.0	10	1.6137	{'accuracy': 0.31}	0.0961	0.31	0.1467
No log	2.0	20	1.5891	{'accuracy': 0.31}	0.0961	0.31	0.1467
No log	3.0	30	1.5298	{'accuracy': 0.29}	0.0927	0.29	0.1405
No log	4.0	40	1.4865	{'accuracy': 0.34}	0.1649	0.34	0.2211
No log	5.0	50	1.4554	{'accuracy': 0.35}	0.2536	0.35	0.2760
No log	6.0	60	1.4830	{'accuracy': 0.39}	0.3336	0.39	0.3132
No log	7.0	70	1.5124	{'accuracy': 0.37}	0.3383	0.37	0.3351
No log	8.0	80	1.5414	{'accuracy': 0.4}	0.3535	0.4	0.3620
No log	9.0	90	1.6319	{'accuracy': 0.36}	0.3216	0.36	0.3306
No log	10.0	100	1.7549	{'accuracy': 0.36}	0.3183	0.36	0.3164
No log	11.0	110	1.7023	{'accuracy': 0.37}	0.3293	0.37	0.3459
No log	12.0	120	1.8304	{'accuracy': 0.37}	0.3190	0.37	0.3365
No log	13.0	130	1.8078	{'accuracy': 0.38}	0.3541	0.38	0.3487
No log	14.0	140	1.8710	{'accuracy': 0.41}	0.3515	0.41	0.3748
No log	15.0	150	1.8866	{'accuracy': 0.41}	0.3515	0.41	0.3748

Framework versions

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