Edit model card
YAML Metadata Error: "widget[0]" must be of type object

BERT_swedish-ner

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

  • Loss: 0.1316
  • Precision: 0.9340
  • Recall: 0.9419
  • F1: 0.9379
  • Accuracy: 0.9800

Model description

Finetuned the model from KB/bert-base-swedish-cased for Swedish NER task. The model can classify three categories:

  • PER (person names)
  • LOC (Location)
  • ORG (Organization)

Intended uses & limitations

NER, token classification

Training and evaluation data

wikiann-SV dataset

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 5e-05
  • train_batch_size: 12
  • eval_batch_size: 12
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5

Training results

Framework versions

  • Transformers 4.22.1
  • Pytorch 1.12.1+cu113
  • Datasets 2.4.0
  • Tokenizers 0.12.1
Downloads last month
22
Safetensors
Model size
124M params
Tensor type
I64
·
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train hkaraoguz/BERT_swedish-ner

Evaluation results