bert-finetuned-ner / README.md
andrewoh's picture
bert-finetuned-ner-chemreact
aae4d4c verified
|
raw
history blame
No virus
1.88 kB
metadata
license: apache-2.0
base_model: bert-base-cased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-finetuned-ner
    results: []

bert-finetuned-ner

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

  • Loss: 0.8089
  • Precision: 0.3730
  • Recall: 0.5764
  • F1: 0.4529
  • Accuracy: 0.7512

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: 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: 5

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
No log 1.0 69 0.8052 0.3835 0.3229 0.3506 0.7630
No log 2.0 138 0.7310 0.3635 0.4809 0.4141 0.7549
No log 3.0 207 0.7309 0.3881 0.5208 0.4448 0.7621
No log 4.0 276 0.7683 0.3926 0.5330 0.4521 0.7642
No log 5.0 345 0.8089 0.3730 0.5764 0.4529 0.7512

Framework versions

  • Transformers 4.41.2
  • Pytorch 2.3.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1