|
--- |
|
license: apache-2.0 |
|
base_model: distilbert/distilbert-base-cased-distilled-squad |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: distilbert-base-cased-distilled-squad-v2 |
|
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. --> |
|
|
|
# distilbert-base-cased-distilled-squad-v2 |
|
|
|
This model is a fine-tuned version of [distilbert/distilbert-base-cased-distilled-squad](https://huggingface.co/distilbert/distilbert-base-cased-distilled-squad) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.9145 |
|
|
|
## 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: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 0.969 | 0.17 | 2500 | 0.8847 | |
|
| 0.9411 | 0.34 | 5000 | 0.8974 | |
|
| 0.9072 | 0.51 | 7500 | 0.8331 | |
|
| 0.9098 | 0.68 | 10000 | 0.8146 | |
|
| 0.866 | 0.85 | 12500 | 0.8371 | |
|
| 0.6918 | 1.02 | 15000 | 0.8752 | |
|
| 0.6142 | 1.19 | 17500 | 0.8580 | |
|
| 0.6348 | 1.36 | 20000 | 0.8042 | |
|
| 0.604 | 1.53 | 22500 | 0.8274 | |
|
| 0.5953 | 1.7 | 25000 | 0.8006 | |
|
| 0.6046 | 1.87 | 27500 | 0.8022 | |
|
| 0.4395 | 2.04 | 30000 | 0.8887 | |
|
| 0.4461 | 2.21 | 32500 | 0.9536 | |
|
| 0.4254 | 2.38 | 35000 | 0.9380 | |
|
| 0.4234 | 2.55 | 37500 | 0.9079 | |
|
| 0.396 | 2.72 | 40000 | 0.9392 | |
|
| 0.4161 | 2.89 | 42500 | 0.9145 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.37.2 |
|
- Pytorch 2.2.0+cu121 |
|
- Datasets 2.17.0 |
|
- Tokenizers 0.15.1 |
|
|
