metadata
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: VaccinChatSentenceClassifierDutch_fromBERTje2_DAdialogQonly09
results: []
VaccinChatSentenceClassifierDutch_fromBERTje2_DAdialogQonly09
This model is a fine-tuned version of outputDAQonly09/ on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4964
- 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 |
|---|---|---|---|---|
| 4.0196 | 1.0 | 1320 | 3.1691 | 0.4232 |
| 2.5409 | 2.0 | 2640 | 2.0495 | 0.6435 |
| 1.8396 | 3.0 | 3960 | 1.3566 | 0.7861 |
| 1.0986 | 4.0 | 5280 | 0.9565 | 0.8556 |
| 0.6711 | 5.0 | 6600 | 0.7119 | 0.8803 |
| 0.4456 | 6.0 | 7920 | 0.5805 | 0.8912 |
| 0.2563 | 7.0 | 9240 | 0.5124 | 0.8995 |
| 0.1686 | 8.0 | 10560 | 0.4953 | 0.9022 |
| 0.1135 | 9.0 | 11880 | 0.4863 | 0.9022 |
| 0.0846 | 10.0 | 13200 | 0.4893 | 0.9049 |
| 0.07 | 11.0 | 14520 | 0.4860 | 0.9077 |
| 0.0487 | 12.0 | 15840 | 0.4889 | 0.9059 |
| 0.0342 | 13.0 | 17160 | 0.4926 | 0.9068 |
| 0.0332 | 14.0 | 18480 | 0.4951 | 0.9068 |
| 0.029 | 15.0 | 19800 | 0.4964 | 0.9068 |
Framework versions
- Transformers 4.13.0.dev0
- Pytorch 1.10.0
- Datasets 1.16.1
- Tokenizers 0.10.3