suicide-distilbert-2-5

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: 1.6324
• Accuracy: {'accuracy': 0.51}
• Precision: 0.4799
• Recall: 0.51
• Fscore: 0.4853

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: 6
• eval_batch_size: 6
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	Fscore
No log	1.0	50	1.4888	{'accuracy': 0.36}	0.3280	0.36	0.2117
No log	2.0	100	1.3987	{'accuracy': 0.44}	0.2745	0.44	0.3224
No log	3.0	150	1.3294	{'accuracy': 0.47}	0.2877	0.47	0.3461
No log	4.0	200	1.3637	{'accuracy': 0.43}	0.3077	0.43	0.3231
No log	5.0	250	1.3532	{'accuracy': 0.44}	0.3478	0.44	0.3428
No log	6.0	300	1.3604	{'accuracy': 0.44}	0.3743	0.44	0.3765
No log	7.0	350	1.3632	{'accuracy': 0.46}	0.3910	0.46	0.4065
No log	8.0	400	1.3834	{'accuracy': 0.45}	0.3968	0.45	0.4022
No log	9.0	450	1.4246	{'accuracy': 0.47}	0.4205	0.47	0.4303
1.0098	10.0	500	1.4542	{'accuracy': 0.43}	0.4257	0.43	0.4228
1.0098	11.0	550	1.5015	{'accuracy': 0.5}	0.4576	0.5	0.4718
1.0098	12.0	600	1.5163	{'accuracy': 0.45}	0.4564	0.45	0.4416
1.0098	13.0	650	1.5486	{'accuracy': 0.5}	0.4645	0.5	0.4751
1.0098	14.0	700	1.6468	{'accuracy': 0.46}	0.4270	0.46	0.4329
1.0098	15.0	750	1.6324	{'accuracy': 0.51}	0.4799	0.51	0.4853

Framework versions

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