bert-finetuned-ner_swedish_test_large_set

This model is a fine-tuned version of KBLab/bert-base-swedish-cased-ner on the suc3 dataset. It achieves the following results on the evaluation set:

• eval_loss: 0.0265
• eval_precision: 0.8542
• eval_recall: 0.8468
• eval_f1: 0.8505
• eval_accuracy: 0.9919
• eval_runtime: 1076.8307
• eval_samples_per_second: 10.685
• eval_steps_per_second: 1.336
• epoch: 1.0
• step: 5754

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: 3

Framework versions

• Transformers 4.19.3
• Pytorch 1.7.1
• Datasets 2.2.2
• Tokenizers 0.12.1