metadata

license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
datasets:
  - szeged_ner
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: hun_wnut_modell
    results:
      - task:
          name: Token Classification
          type: token-classification
        dataset:
          name: szeged_ner
          type: szeged_ner
          config: business
          split: test
          args: business
        metrics:
          - name: Precision
            type: precision
            value: 0.8590342679127726
          - name: Recall
            type: recall
            value: 0.9004081632653061
          - name: F1
            type: f1
            value: 0.8792347548824233
          - name: Accuracy
            type: accuracy
            value: 0.9881996563884619

hun_wnut_modell

This model is a fine-tuned version of distilbert-base-uncased on the szeged_ner dataset. It achieves the following results on the evaluation set:

Loss: 0.0419
Precision: 0.8590
Recall: 0.9004
F1: 0.8792
Accuracy: 0.9882

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: 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: 5

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2035	1.0	511	0.0665	0.8124	0.8343	0.8232	0.9813
0.075	2.0	1022	0.0501	0.8280	0.8841	0.8551	0.9847
0.0498	3.0	1533	0.0444	0.8452	0.8914	0.8677	0.9866
0.0354	4.0	2044	0.0417	0.8661	0.8980	0.8818	0.9885
0.0275	5.0	2555	0.0419	0.8590	0.9004	0.8792	0.9882

Framework versions

Transformers 4.32.0
Pytorch 2.0.1+cu118
Datasets 2.14.4
Tokenizers 0.13.3