Deadbody42/procit_ner_street_city_names

procit_ner_street_city_name

This model is a fine-tuned version of roberta-base on an procit006/NER_Street_City_Name_Dataset_Aug27. It achieves the following results on the evaluation set:

• Loss: 0.0075
• Loc: {'precision': 0.9770341725199274, 'recall': 1.0, 'f1-score': 0.9883836972575946, 'support': 12380}
• Per: {'precision': 1.0, 'recall': 1.0, 'f1-score': 1.0, 'support': 5590}
• Micro avg: {'precision': 0.9840643995400032, 'recall': 1.0, 'f1-score': 0.9919682040241781, 'support': 17970}
• Macro avg: {'precision': 0.9885170862599637, 'recall': 1.0, 'f1-score': 0.9941918486287973, 'support': 17970}
• Weighted avg: {'precision': 0.9841782446186255, 'recall': 1.0, 'f1-score': 0.9919972271590997, 'support': 17970}

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: 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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Loc	Per	Micro avg	Macro avg	Weighted avg
0.0918	0.2667	500	0.0146	{'precision': 0.9762637016008201, 'recall': 1.0, 'f1-score': 0.9879893060931328, 'support': 12380}	{'precision': 0.9969566774078052, 'recall': 0.9962432915921288, 'f1-score': 0.9965998568360772, 'support': 5590}	{'precision': 0.9825915585482017, 'recall': 0.9988313856427379, 'f1-score': 0.9906449209371637, 'support': 17970}	{'precision': 0.9866101895043127, 'recall': 0.9981216457960644, 'f1-score': 0.992294581464605, 'support': 17970}	{'precision': 0.9827007486103386, 'recall': 0.9988313856427379, 'f1-score': 0.9906678246603592, 'support': 17970}
0.0231	0.5333	1000	0.0087	{'precision': 0.99983429991715, 'recall': 0.9747980613893377, 'f1-score': 0.987157464212679, 'support': 12380}	{'precision': 0.998569896317483, 'recall': 0.9992844364937388, 'f1-score': 0.9989270386266094, 'support': 5590}	{'precision': 0.9994338768115942, 'recall': 0.9824151363383417, 'f1-score': 0.9908514340236854, 'support': 17970}	{'precision': 0.9992020981173165, 'recall': 0.9870412489415382, 'f1-score': 0.9930422514196442, 'support': 17970}	{'precision': 0.9994409768163075, 'recall': 0.9824151363383417, 'f1-score': 0.9908186729480085, 'support': 17970}
0.012	0.8	1500	0.0154	{'precision': 0.9770341725199274, 'recall': 1.0, 'f1-score': 0.9883836972575946, 'support': 12380}	{'precision': 0.9941186954197113, 'recall': 0.9978533094812164, 'f1-score': 0.9959825015623606, 'support': 5590}	{'precision': 0.9822776501476862, 'recall': 0.9993322203672788, 'f1-score': 0.9907315458457464, 'support': 17970}	{'precision': 0.9855764339698194, 'recall': 0.9989266547406082, 'f1-score': 0.9921830994099776, 'support': 17970}	{'precision': 0.9823487236056142, 'recall': 0.9993322203672788, 'f1-score': 0.9907474878009247, 'support': 17970}
0.0155	1.0667	2000	0.0082	{'precision': 1.0, 'recall': 0.9747980613893377, 'f1-score': 0.9872382198952879, 'support': 12380}	{'precision': 0.9996420261320924, 'recall': 0.9991055456171736, 'f1-score': 0.9993737138767111, 'support': 5590}	{'precision': 0.9998867176437269, 'recall': 0.9823594880356149, 'f1-score': 0.9910456140350876, 'support': 17970}	{'precision': 0.9998210130660462, 'recall': 0.9869518035032556, 'f1-score': 0.9933059668859996, 'support': 17970}	{'precision': 0.9998886436326321, 'recall': 0.9823594880356149, 'f1-score': 0.9910132566986355, 'support': 17970}
0.0098	1.3333	2500	0.0085	{'precision': 0.9770305470044992, 'recall': 0.9998384491114701, 'f1-score': 0.9883029262645217, 'support': 12380}	{'precision': 1.0, 'recall': 0.9992844364937388, 'f1-score': 0.9996420901932713, 'support': 5590}	{'precision': 0.9840591618734593, 'recall': 0.9996661101836394, 'f1-score': 0.9918012422360247, 'support': 17970}	{'precision': 0.9885152735022495, 'recall': 0.9995614428026045, 'f1-score': 0.9939725082288965, 'support': 17970}	{'precision': 0.9841757469068281, 'recall': 0.9996661101836394, 'f1-score': 0.9918302454833147, 'support': 17970}
0.0093	1.6	3000	0.0085	{'precision': 0.9769479750532881, 'recall': 0.9995961227786753, 'f1-score': 0.9881422924901185, 'support': 12380}	{'precision': 0.9991061851984269, 'recall': 0.9998211091234347, 'f1-score': 0.9994635193133047, 'support': 5590}	{'precision': 0.9837358304583539, 'recall': 0.9996661101836394, 'f1-score': 0.9916369959427009, 'support': 17970}	{'precision': 0.9880270801258575, 'recall': 0.999708615951055, 'f1-score': 0.9938029059017116, 'support': 17970}	{'precision': 0.9838408183872519, 'recall': 0.9996661101836394, 'f1-score': 0.991664031941516, 'support': 17970}
0.0109	1.8667	3500	0.0077	{'precision': 1.0, 'recall': 0.9747980613893377, 'f1-score': 0.9872382198952879, 'support': 12380}	{'precision': 1.0, 'recall': 1.0, 'f1-score': 1.0, 'support': 5590}	{'precision': 1.0, 'recall': 0.9826377295492488, 'f1-score': 0.9912428427079825, 'support': 17970}	{'precision': 1.0, 'recall': 0.9873990306946688, 'f1-score': 0.9936191099476439, 'support': 17970}	{'precision': 1.0, 'recall': 0.9826377295492488, 'f1-score': 0.9912080780358188, 'support': 17970}
0.0067	2.1333	4000	0.0083	{'precision': 0.9770341725199274, 'recall': 1.0, 'f1-score': 0.9883836972575946, 'support': 12380}	{'precision': 1.0, 'recall': 1.0, 'f1-score': 1.0, 'support': 5590}	{'precision': 0.9840643995400032, 'recall': 1.0, 'f1-score': 0.9919682040241781, 'support': 17970}	{'precision': 0.9885170862599637, 'recall': 1.0, 'f1-score': 0.9941918486287973, 'support': 17970}	{'precision': 0.9841782446186255, 'recall': 1.0, 'f1-score': 0.9919972271590997, 'support': 17970}
0.0075	2.4	4500	0.0076	{'precision': 1.0, 'recall': 0.9747980613893377, 'f1-score': 0.9872382198952879, 'support': 12380}	{'precision': 1.0, 'recall': 1.0, 'f1-score': 1.0, 'support': 5590}	{'precision': 1.0, 'recall': 0.9826377295492488, 'f1-score': 0.9912428427079825, 'support': 17970}	{'precision': 1.0, 'recall': 0.9873990306946688, 'f1-score': 0.9936191099476439, 'support': 17970}	{'precision': 1.0, 'recall': 0.9826377295492488, 'f1-score': 0.9912080780358188, 'support': 17970}
0.0077	2.6667	5000	0.0076	{'precision': 0.9770341725199274, 'recall': 1.0, 'f1-score': 0.9883836972575946, 'support': 12380}	{'precision': 1.0, 'recall': 1.0, 'f1-score': 1.0, 'support': 5590}	{'precision': 0.9840643995400032, 'recall': 1.0, 'f1-score': 0.9919682040241781, 'support': 17970}	{'precision': 0.9885170862599637, 'recall': 1.0, 'f1-score': 0.9941918486287973, 'support': 17970}	{'precision': 0.9841782446186255, 'recall': 1.0, 'f1-score': 0.9919972271590997, 'support': 17970}
0.0071	2.9333	5500	0.0075	{'precision': 0.9770341725199274, 'recall': 1.0, 'f1-score': 0.9883836972575946, 'support': 12380}	{'precision': 1.0, 'recall': 1.0, 'f1-score': 1.0, 'support': 5590}	{'precision': 0.9840643995400032, 'recall': 1.0, 'f1-score': 0.9919682040241781, 'support': 17970}	{'precision': 0.9885170862599637, 'recall': 1.0, 'f1-score': 0.9941918486287973, 'support': 17970}	{'precision': 0.9841782446186255, 'recall': 1.0, 'f1-score': 0.9919972271590997, 'support': 17970}

Framework versions

• Transformers 4.53.0
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2