udpos28-sm-all-POS
This model is a fine-tuned version of bert-base-cased on the udpos28 dataset. It achieves the following results on the evaluation set:
- Loss: 0.1479
- Precision: 0.9587
- Recall: 0.9589
- F1: 0.9588
- Accuracy: 0.9648
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.1261 | 1.0 | 4978 | 0.1358 | 0.9513 | 0.9510 | 0.9512 | 0.9581 |
0.0788 | 2.0 | 9956 | 0.1326 | 0.9578 | 0.9578 | 0.9578 | 0.9642 |
0.0424 | 3.0 | 14934 | 0.1479 | 0.9587 | 0.9589 | 0.9588 | 0.9648 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.10.2+cu102
- Datasets 2.2.2
- Tokenizers 0.12.1
- Downloads last month
- 239
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
Evaluation results
- Precision on udpos28self-reported0.959
- Recall on udpos28self-reported0.959
- F1 on udpos28self-reported0.959
- Accuracy on udpos28self-reported0.965