judithrosell's picture
End of training
ca7c2d6 verified
metadata
license: apache-2.0
base_model: google-bert/bert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: BERT_BIOMAT_NER__ST_1000_DA
    results: []

BERT_BIOMAT_NER__ST_1000_DA

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

  • Loss: 0.4791
  • Precision: 0.4732
  • Recall: 0.6716
  • F1: 0.5552
  • Accuracy: 0.9364

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: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 397 0.2764 0.4422 0.5997 0.5090 0.9323
0.2077 2.0 794 0.3185 0.4682 0.6485 0.5438 0.9356
0.0466 3.0 1191 0.3461 0.4699 0.6592 0.5487 0.9362
0.0179 4.0 1588 0.3994 0.4567 0.6595 0.5397 0.9342
0.0179 5.0 1985 0.4091 0.4735 0.6733 0.5560 0.9369
0.0088 6.0 2382 0.4392 0.4701 0.6630 0.5501 0.9366
0.0048 7.0 2779 0.4594 0.4654 0.6644 0.5473 0.9356
0.0032 8.0 3176 0.4684 0.4740 0.6775 0.5578 0.9369
0.0024 9.0 3573 0.4763 0.4703 0.6623 0.5500 0.9359
0.0024 10.0 3970 0.4791 0.4732 0.6716 0.5552 0.9364

Framework versions

  • Transformers 4.42.4
  • Pytorch 2.4.0+cu121
  • Datasets 2.21.0
  • Tokenizers 0.19.1