jenslemmens's picture
End of training
eb59411 verified
metadata
license: mit
base_model: DTAI-KULeuven/robbert-2023-dutch-base
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
model-index:
  - name: robbert-2023-dutch-base-gender
    results: []

robbert-2023-dutch-base-gender

This model is a fine-tuned version of DTAI-KULeuven/robbert-2023-dutch-base on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6591
  • Precision: 0.6282
  • Recall: 0.6290
  • Fscore: 0.6278
  • Accuracy: 0.6285

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: 1e-05
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 42
  • gradient_accumulation_steps: 2
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 1.0
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Precision Recall Fscore Accuracy
0.616 0.29 2000 0.6498 0.6295 0.6299 0.6266 0.6267
0.6033 0.59 4000 0.6584 0.6278 0.6274 0.6228 0.6228
0.5896 0.88 6000 0.6600 0.6285 0.6293 0.6282 0.6290

Framework versions

  • Transformers 4.36.2
  • Pytorch 2.1.2+cu121
  • Datasets 2.14.5
  • Tokenizers 0.15.0