Edit model card

VaccinChatSentenceClassifierDutch_fromBERTje2_DAdialogQonly

This model is a fine-tuned version of outputDAQonly/checkpoint-8710 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5008
  • Accuracy: 0.9068

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: 8
  • eval_batch_size: 8
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
  • lr_scheduler_type: linear
  • num_epochs: 15.0

Training results

Training Loss Epoch Step Validation Loss Accuracy
4.0751 1.0 1320 3.1674 0.4086
2.5619 2.0 2640 2.0335 0.6426
1.8549 3.0 3960 1.3537 0.7861
1.106 4.0 5280 0.9515 0.8519
0.6698 5.0 6600 0.7152 0.8757
0.4497 6.0 7920 0.5838 0.8921
0.2626 7.0 9240 0.5300 0.8940
0.1762 8.0 10560 0.4984 0.8958
0.119 9.0 11880 0.4906 0.9059
0.0919 10.0 13200 0.4896 0.8995
0.0722 11.0 14520 0.5012 0.9022
0.0517 12.0 15840 0.4951 0.9040
0.0353 13.0 17160 0.4988 0.9040
0.0334 14.0 18480 0.5035 0.9049
0.0304 15.0 19800 0.5008 0.9068

Framework versions

  • Transformers 4.13.0.dev0
  • Pytorch 1.10.0
  • Datasets 1.16.1
  • Tokenizers 0.10.3
Downloads last month
25