mva_ner_2

This model is a fine-tuned version of prajjwal1/bert-tiny on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.0026
• Overall Precision: 0.9873
• Overall Recall: 0.9873
• Overall F1: 0.9873
• Overall Accuracy: 0.9987
• Year F1: 1.0
• Years Ago F1: 0.9844

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: 0.001
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Overall Precision	Overall Recall	Overall F1	Overall Accuracy	Year F1	Years Ago F1
0.0099	35.71	1000	0.0225	0.9625	0.9747	0.9686	0.9960	1.0	0.9612
0.0078	71.43	2000	0.0157	0.9625	0.9747	0.9686	0.9960	1.0	0.9612
0.0078	107.14	3000	0.0075	0.9873	0.9873	0.9873	0.9987	1.0	0.9844
0.0061	142.86	4000	0.0062	0.9873	0.9873	0.9873	0.9987	1.0	0.9844
0.0053	178.57	5000	0.0032	0.9873	0.9873	0.9873	0.9987	1.0	0.9844
0.0049	214.29	6000	0.0179	0.9747	0.9747	0.9747	0.9973	1.0	0.9688
0.0049	250.0	7000	0.0011	1.0	1.0	1.0	1.0	1.0	1.0
0.0034	285.71	8000	0.0064	0.9747	0.9747	0.9747	0.9973	1.0	0.9688
0.0037	321.43	9000	0.0148	0.9875	1.0	0.9937	0.9987	1.0	0.9922
0.0035	357.14	10000	0.0006	1.0	1.0	1.0	1.0	1.0	1.0
0.003	392.86	11000	0.0007	1.0	1.0	1.0	1.0	1.0	1.0
0.0028	428.57	12000	0.0032	0.9873	0.9873	0.9873	0.9987	1.0	0.9844
0.0025	464.29	13000	0.0006	1.0	1.0	1.0	1.0	1.0	1.0
0.0024	500.0	14000	0.0026	0.9873	0.9873	0.9873	0.9987	1.0	0.9844

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1