Bert-v1

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

• Loss: 0.5738
• F1: 0.9637
• Accuracy: 0.9300
• Precision: 0.9302
• Recall: 0.9997

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: 0.002
• train_batch_size: 1
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 10

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Accuracy	Precision	Recall
0.6498	0.0000	1	0.2875	0.9635	0.9295	0.9303	0.9991
0.2169	0.0000	2	0.2707	0.9635	0.9295	0.9303	0.9991
0.0354	0.0000	3	0.3281	0.9634	0.9294	0.9302	0.9991
0.0225	0.0000	4	0.3941	0.9635	0.9295	0.9302	0.9992
0.0021	0.0000	5	0.4481	0.9635	0.9295	0.9302	0.9992
0.0207	0.0000	6	0.4928	0.9635	0.9296	0.9302	0.9993
0.0131	0.0000	7	0.5282	0.9636	0.9298	0.9302	0.9995
0.0017	0.0001	8	0.5525	0.9637	0.9300	0.9302	0.9997
0.0002	0.0001	9	0.5672	0.9637	0.9300	0.9302	0.9997
0.0003	0.0001	10	0.5738	0.9637	0.9300	0.9302	0.9997

Framework versions

• PEFT 0.13.0
• Transformers 4.45.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.1
• Tokenizers 0.20.0