metadata
license: mit
base_model: microsoft/deberta-v3-base
tags:
- generated_from_trainer
datasets:
- ontonotes5
model-index:
- name: deberta-v3-base_on5
results: []
deberta-v3-base_on5
This model is a fine-tuned version of microsoft/deberta-v3-base on the ontonotes5 dataset. It achieves the following results on the evaluation set:
- Loss: 0.0598
- F1-type-match: 0.6780
- F1-partial: 0.6872
- F1-strict: 0.6565
- F1-exact: 0.6729
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | F1-type-match | F1-partial | F1-strict | F1-exact |
|---|---|---|---|---|---|---|---|
| 0.0738 | 1.0 | 936 | 0.0624 | 0.5568 | 0.5632 | 0.5322 | 0.5479 |
| 0.0432 | 2.0 | 1873 | 0.0591 | 0.5773 | 0.5848 | 0.5559 | 0.5709 |
| 0.0289 | 3.0 | 2808 | 0.0598 | 0.6780 | 0.6872 | 0.6565 | 0.6729 |
Framework versions
- Transformers 4.36.0
- Pytorch 2.0.0
- Datasets 2.1.0
- Tokenizers 0.15.0