sandeshrajx/test_trainer_20240918_1411

bert-fraud-classification-test

This model is a fine-tuned version of google-bert/bert-base-multilingual-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.6431
• F1: 0.7435
• Precision: 0.6605
• Val Accuracy: {'accuracy': 0.706}

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: 5e-05
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1	Precision	Val Accuracy
0.601	0.32	40	0.6152	0.7039	0.6597	{'accuracy': 0.682}
0.6332	0.64	80	0.6143	0.7068	0.6515	{'accuracy': 0.679}
0.4862	0.96	120	0.5791	0.7151	0.7137	{'accuracy': 0.714}
0.6297	1.28	160	0.6323	0.7281	0.6495	{'accuracy': 0.69}
0.6164	1.6	200	0.5010	0.7522	0.8345	{'accuracy': 0.774}
0.6333	1.92	240	0.5824	0.7310	0.6828	{'accuracy': 0.71}
0.4465	2.24	280	0.5335	0.7579	0.7695	{'accuracy': 0.761}
0.5342	2.56	320	0.5065	0.7644	0.8040	{'accuracy': 0.775}
0.5462	2.88	360	0.6431	0.7435	0.6605	{'accuracy': 0.706}

Framework versions

• Transformers 4.45.0.dev0
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1