|
--- |
|
license: mit |
|
base_model: microsoft/deberta-v3-base |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- conll2003 |
|
model-index: |
|
- name: deberta-v3-base_conll03 |
|
results: [] |
|
--- |
|
|
|
<!-- This model card has been generated automatically according to the information the Trainer had access to. You |
|
should probably proofread and complete it, then remove this comment. --> |
|
|
|
# deberta-v3-base_conll03 |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on the conll2003 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0973 |
|
- F1-type-match: 0.9316 |
|
- F1-partial: 0.9733 |
|
- F1-strict: 0.9235 |
|
- F1-exact: 0.9651 |
|
|
|
## 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 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 5 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1-type-match | F1-partial | F1-strict | F1-exact | |
|
|:-------------:|:-----:|:----:|:---------------:|:-------------:|:----------:|:---------:|:--------:| |
|
| 0.0963 | 1.0 | 439 | 0.0814 | 0.8408 | 0.8897 | 0.8323 | 0.8809 | |
|
| 0.0197 | 2.0 | 878 | 0.0803 | 0.9219 | 0.9725 | 0.9138 | 0.9648 | |
|
| 0.0108 | 3.0 | 1317 | 0.0858 | 0.9307 | 0.9728 | 0.9228 | 0.9648 | |
|
| 0.0054 | 4.0 | 1756 | 0.0922 | 0.9313 | 0.9725 | 0.9235 | 0.9643 | |
|
| 0.0033 | 5.0 | 2195 | 0.0973 | 0.9316 | 0.9733 | 0.9235 | 0.9651 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.0 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.15.0 |
|
|
