|
--- |
|
license: apache-2.0 |
|
base_model: google-t5/t5-small |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: t5-small-mrqa |
|
results: [] |
|
datasets: |
|
- enriquesaou/mrqa-squadded-sample |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/favcowboy/huggingface/runs/k381y37g) |
|
# t5-small-mrqa |
|
|
|
This model is a fine-tuned version of [google-t5/t5-small](https://huggingface.co/google-t5/t5-small) on an MRQA sample. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8647 |
|
|
|
## 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: 14 |
|
- eval_batch_size: 14 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 3 |
|
- total_train_batch_size: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 8 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:------:|:----:|:---------------:| |
|
| No log | 0.9991 | 357 | 0.9669 | |
|
| 1.0947 | 1.9981 | 714 | 0.9170 | |
|
| 0.9558 | 3.0 | 1072 | 0.8990 | |
|
| 0.9558 | 3.9991 | 1429 | 0.8855 | |
|
| 0.9023 | 4.9981 | 1786 | 0.8680 | |
|
| 0.8684 | 6.0 | 2144 | 0.8680 | |
|
| 0.8542 | 6.9991 | 2501 | 0.8668 | |
|
| 0.8542 | 7.9925 | 2856 | 0.8647 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.42.0.dev0 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.2 |
|
- Tokenizers 0.19.1 |
