|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- squad |
|
model-index: |
|
- name: questionansweringmodel |
|
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. --> |
|
|
|
# questionansweringmodel |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on the squad dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.9365 |
|
|
|
## 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: 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: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 0.7398 | 1.0 | 5520 | 0.7216 | |
|
| 0.5263 | 2.0 | 11040 | 0.7513 | |
|
| 0.3828 | 3.0 | 16560 | 0.8821 | |
|
| 0.2411 | 4.0 | 22080 | 1.0024 | |
|
| 0.1549 | 5.0 | 27600 | 1.3959 | |
|
| 0.0996 | 6.0 | 33120 | 1.7559 | |
|
| 0.0594 | 7.0 | 38640 | 2.0031 | |
|
| 0.0393 | 8.0 | 44160 | 2.3654 | |
|
| 0.0326 | 9.0 | 49680 | 2.7383 | |
|
| 0.0162 | 10.0 | 55200 | 2.9365 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.26.0 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.9.0 |
|
- Tokenizers 0.13.2 |
|
|
