NER-TotalAmount

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

• Loss: 0.0973
• Precision: 0.8889
• Recall: 0.9308
• F1: 0.9094
• Accuracy: 0.9794

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: 2e-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: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	18	0.5186	0.0524	0.0440	0.0479	0.8261
No log	2.0	36	0.2669	0.3287	0.3679	0.3472	0.8936
No log	3.0	54	0.1462	0.725	0.8208	0.7699	0.9516
No log	4.0	72	0.0991	0.8006	0.8962	0.8457	0.9668
No log	5.0	90	0.0937	0.8421	0.9057	0.8727	0.9718
No log	6.0	108	0.0774	0.8813	0.9340	0.9069	0.9775
No log	7.0	126	0.0764	0.8710	0.9340	0.9014	0.9794
No log	8.0	144	0.0753	0.8824	0.9434	0.9119	0.9794
No log	9.0	162	0.0831	0.8689	0.9591	0.9118	0.9775
No log	10.0	180	0.0871	0.8696	0.9434	0.9050	0.9783
No log	11.0	198	0.0906	0.8794	0.9403	0.9088	0.9786
No log	12.0	216	0.0843	0.8832	0.9277	0.9049	0.9779
No log	13.0	234	0.0882	0.8892	0.9591	0.9228	0.9802
No log	14.0	252	0.0977	0.8779	0.9497	0.9124	0.9786
No log	15.0	270	0.0831	0.8919	0.9340	0.9124	0.9794
No log	16.0	288	0.0881	0.8876	0.9434	0.9146	0.9802
No log	17.0	306	0.0898	0.8728	0.9497	0.9096	0.9794
No log	18.0	324	0.0890	0.8856	0.9497	0.9165	0.9809
No log	19.0	342	0.0900	0.8830	0.9497	0.9152	0.9805
No log	20.0	360	0.0933	0.8886	0.9528	0.9196	0.9809
No log	21.0	378	0.0941	0.8912	0.9528	0.9210	0.9805
No log	22.0	396	0.0979	0.8909	0.9497	0.9193	0.9798
No log	23.0	414	0.0998	0.8935	0.9497	0.9207	0.9802
No log	24.0	432	0.0975	0.8889	0.9308	0.9094	0.9794
No log	25.0	450	0.0973	0.8889	0.9308	0.9094	0.9794

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.0
• Datasets 2.12.0
• Tokenizers 0.15.1