bert-base-cased-finetuned-ner-BC2GM-IOB

This model is a fine-tuned version of bert-base-cased. It achieves the following results on the evaluation set:

• Loss: 0.0813
• Gene
  • Precision: 0.752111423914654
  • Recall: 0.8025296442687747
  • F1: 0.7765029830197338
  • Number: 6325
• Overall
  • Precision: 0.7521
  • Recall: 0.8025
  • F1: 0.7765
  • Accuracy: 0.9736

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/Monolingual/EMBO-BLURB/NER%20Project%20Using%20EMBO-BLURB%20Dataset.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology.

Training and evaluation data

Dataset Source: https://huggingface.co/datasets/EMBO/BLURB

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: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Gene Precision	Gene Recall	Gene F1	Gene Number	Overall Precision	Overall Recall	Overall F1	Overall Accuracy
0.0882	1.0	786	0.0771	0.7383	0.7538	0.7460	6325	0.7383	0.7538	0.7460	0.9697
0.0547	2.0	1572	0.0823	0.7617	0.7758	0.7687	6325	0.7617	0.7758	0.7687	0.9732
0.0356	3.0	2358	0.0813	0.7521	0.8025	0.7765	6325	0.7521	0.8025	0.7765	0.9736

*All values in the above chart are rounded to the nearest ten-thousandth.

Framework versions

• Transformers 4.28.1
• Pytorch 2.0.0
• Datasets 2.11.0
• Tokenizers 0.13.3