|
|
--- |
|
|
license: cc-by-4.0 |
|
|
base_model: deepset/bert-base-cased-squad2 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
datasets: |
|
|
- covid_qa_deepset |
|
|
model-index: |
|
|
- name: bert-covidqa-1 |
|
|
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. --> |
|
|
|
|
|
# bert-covidqa-1 |
|
|
|
|
|
This model is a fine-tuned version of [deepset/bert-base-cased-squad2](https://huggingface.co/deepset/bert-base-cased-squad2) on the covid_qa_deepset dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.4528 |
|
|
|
|
|
## 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: 3e-05 |
|
|
- train_batch_size: 16 |
|
|
- eval_batch_size: 16 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 1 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|
|:-------------:|:-----:|:----:|:---------------:| |
|
|
| 1.0353 | 0.04 | 5 | 0.7327 | |
|
|
| 1.1116 | 0.09 | 10 | 0.5674 | |
|
|
| 0.8086 | 0.13 | 15 | 0.5025 | |
|
|
| 0.814 | 0.18 | 20 | 0.5620 | |
|
|
| 0.4168 | 0.22 | 25 | 0.6628 | |
|
|
| 0.7069 | 0.26 | 30 | 0.5637 | |
|
|
| 0.4168 | 0.31 | 35 | 0.4855 | |
|
|
| 0.5636 | 0.35 | 40 | 0.4708 | |
|
|
| 0.398 | 0.39 | 45 | 0.4712 | |
|
|
| 0.4681 | 0.44 | 50 | 0.5235 | |
|
|
| 0.34 | 0.48 | 55 | 0.5863 | |
|
|
| 0.2484 | 0.53 | 60 | 0.6422 | |
|
|
| 0.4526 | 0.57 | 65 | 0.6614 | |
|
|
| 0.2941 | 0.61 | 70 | 0.6210 | |
|
|
| 0.7383 | 0.66 | 75 | 0.5334 | |
|
|
| 0.7337 | 0.7 | 80 | 0.4612 | |
|
|
| 0.4082 | 0.75 | 85 | 0.4447 | |
|
|
| 0.3517 | 0.79 | 90 | 0.4429 | |
|
|
| 0.341 | 0.83 | 95 | 0.4446 | |
|
|
| 0.2751 | 0.88 | 100 | 0.4536 | |
|
|
| 0.4916 | 0.92 | 105 | 0.4566 | |
|
|
| 0.4895 | 0.96 | 110 | 0.4528 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.34.1 |
|
|
- Pytorch 2.1.0+cu118 |
|
|
- Datasets 2.14.5 |
|
|
- Tokenizers 0.14.1 |
|
|
|
