gnurt2041/bert-base-cased-tuned

results

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

• Loss: 0.0837
• Accuracy: 0.975
• Precision: 0.9751
• Recall: 0.975
• F1: 0.9750

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: 3e-05
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 8
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 5
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.334	0.9895	59	0.1713	0.9583	0.9596	0.9583	0.9584
0.0852	1.9958	119	0.2023	0.95	0.9522	0.95	0.9500
0.0369	2.9853	178	0.2496	0.9417	0.9450	0.9417	0.9417
0.0022	3.9916	238	0.1342	0.9583	0.9596	0.9583	0.9584
0.0839	4.9979	298	0.1378	0.975	0.9763	0.975	0.9750
0.0024	5.9874	357	0.1526	0.9583	0.9617	0.9583	0.9583
0.0577	6.9937	417	0.0837	0.975	0.9751	0.975	0.9750
0.0014	8.0	477	0.1215	0.975	0.9751	0.975	0.9750
0.0008	8.9895	536	0.1326	0.975	0.9751	0.975	0.9750
0.0008	9.8952	590	0.1340	0.975	0.9751	0.975	0.9750

Framework versions

• Transformers 4.44.2
• Pytorch 2.5.0+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1