bert-base-multilingual-cased-fine_tuned-ner-WikiNeural_Multilingual

This model is a fine-tuned version of bert-base-multilingual-cased.

It achieves the following results on the evaluation set:

• Loss: 0.0168
• Overall
  • Precision: 0.9957
  • Recall: 0.9961
  • F1: 0.9959
  • Accuracy: 0.9947
• Loc
  • Precision: 0.9983410191680872
  • Recall: 0.99820099576644
  • F1: 0.9982710025571356
  • Number: 1,932,180
• Misc
  • Precision: 0.9809396911027518
  • Recall: 0.9833044214778437
  • F1: 0.9821206328547606
  • Number: 122,787
• Org
  • Precision: 0.9868919798954698
  • Recall: 0.9881129520338388
  • F1: 0.9875020885547201
  • Number: 59,813
• Per
  • Precision: 0.9386096837531854
  • Recall: 0.9516901050491359
  • F1: 0.9451046377116415
  • Number: 47,216

Model description

For more information on how it was created, check out the following link: https://github.com/DunnBC22/NLP_Projects/blob/main/Token%20Classification/Multilingual/Babelscape-WikiNeural-Joined%20Dataset/Babelscape%20WikiNeural%20Joined%20Dataset%20With%20Multilingual%20BERT.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology. You are welcome to test and experiment with this model, but it is at your own risk/peril.

Training and evaluation data

Dataset Soruce: https://huggingface.co/datasets/dmargutierrez/Babelscape-wikineural-joined

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Train Loss	Epoch	Step	Valid Loss	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.015	1.0	102700	0.0168	0.9957	0.9961	0.9959	0.9947

Train Loss	Epoch	Valid Loss	LOC Precision	LOC Recall	LOC F1	LOC Number	MISC Precision	MISC Recall	MISC F1	MISC Number	ORG Precision	ORG Recall	ORG F1	ORG Number	PER Precision	PER Recall	PER F1	PER Number
0.015	1.0	0.0168	0.9983	0.9982	0.9983	1,932,180	0.9809	0.9833	0.9821	122,787	0.9869	0.9881	0.9875	59,813	0.9386	0.9517	0.9451	47,216

Framework versions

• Transformers 4.31.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3