fosh-detector-v2-bge

This model is a fine-tuned version of David-ger/fosh-detector-v2-bge on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0281
• Accuracy: 0.9926
• Precision: 1.0
• Recall: 0.9858
• F1: 0.9929

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.1095	0.0625	50	0.0719	0.9926	1.0	0.9858	0.9929
0.2441	0.125	100	0.0526	0.9778	0.9787	0.9787	0.9787
0.0676	0.1875	150	0.0095	0.9963	0.9930	1.0	0.9965
0.072	0.25	200	0.1507	0.9815	1.0	0.9645	0.9819
0.0986	0.3125	250	0.0672	0.9778	1.0	0.9574	0.9783
0.1446	0.375	300	0.0986	0.9778	0.9592	1.0	0.9792
0.075	0.4375	350	0.0341	0.9926	0.9929	0.9929	0.9929
0.0539	0.5	400	0.0490	0.9852	1.0	0.9716	0.9856
0.0656	0.5625	450	0.0407	0.9889	1.0	0.9787	0.9892
0.0825	0.625	500	0.0484	0.9889	0.9859	0.9929	0.9894
0.0547	0.6875	550	0.0286	0.9963	1.0	0.9929	0.9964
0.1036	0.75	600	0.0228	0.9963	1.0	0.9929	0.9964
0.0426	0.8125	650	0.0234	0.9963	1.0	0.9929	0.9964
0.061	0.875	700	0.0285	0.9926	1.0	0.9858	0.9929
0.0137	0.9375	750	0.0282	0.9926	1.0	0.9858	0.9929
0.0322	1.0	800	0.0281	0.9926	1.0	0.9858	0.9929

Framework versions

• Transformers 4.50.3
• Pytorch 2.6.0+cu124
• Datasets 3.3.2
• Tokenizers 0.21.1