relation-bert-because

This model is a fine-tuned version of bert-base-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1546
• Precision: 0.2346
• Recall: 0.7273
• F1: 0.3548
• Accuracy: 0.9605
• Relation P: 0.2346
• Relation R: 0.7273
• Relation F1: 0.3548

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: 4e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy	Relation P	Relation R	Relation F1
0.7173	0.1242	20	0.3552	0.0577	0.4636	0.1026	0.8295	0.0577	0.4636	0.1026
0.7173	0.2484	40	0.1944	0.1468	0.7045	0.2429	0.9291	0.1468	0.7045	0.2429
0.7173	0.3727	60	0.1984	0.2552	0.7318	0.3784	0.9629	0.2552	0.7318	0.3784
0.7173	0.4969	80	0.1609	0.2241	0.7273	0.3426	0.9566	0.2241	0.7273	0.3426
0.7173	0.6211	100	0.1379	0.1896	0.7136	0.2996	0.9475	0.1896	0.7136	0.2996
0.7173	0.7453	120	0.1787	0.2601	0.7318	0.3838	0.9654	0.2601	0.7318	0.3838
0.7173	0.8696	140	0.1667	0.2453	0.7182	0.3657	0.9631	0.2453	0.7182	0.3657
0.7173	0.9938	160	0.1547	0.2346	0.7273	0.3548	0.9605	0.2346	0.7273	0.3548

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1