|
--- |
|
license: mit |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- textvqa |
|
model-index: |
|
- name: git-base-textvqa |
|
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. --> |
|
|
|
# git-base-textvqa |
|
|
|
This model is a fine-tuned version of [microsoft/git-base-textvqa](https://huggingface.co/microsoft/git-base-textvqa) on the textvqa dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0472 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 3 |
|
- 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 | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 0.9764 | 0.2 | 500 | 0.0499 | |
|
| 0.0524 | 0.4 | 1000 | 0.0492 | |
|
| 0.0525 | 0.6 | 1500 | 0.0494 | |
|
| 0.0531 | 0.8 | 2000 | 0.0480 | |
|
| 0.0515 | 1.0 | 2500 | 0.0477 | |
|
| 0.0473 | 1.2 | 3000 | 0.0483 | |
|
| 0.0479 | 1.4 | 3500 | 0.0477 | |
|
| 0.0473 | 1.6 | 4000 | 0.0476 | |
|
| 0.0486 | 1.8 | 4500 | 0.0472 | |
|
| 0.0471 | 2.0 | 5000 | 0.0473 | |
|
| 0.0454 | 2.2 | 5500 | 0.0473 | |
|
| 0.0452 | 2.4 | 6000 | 0.0476 | |
|
| 0.0438 | 2.6 | 6500 | 0.0475 | |
|
| 0.0463 | 2.8 | 7000 | 0.0474 | |
|
| 0.0449 | 3.0 | 7500 | 0.0472 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.28.0 |
|
- Pytorch 2.0.0 |
|
- Datasets 2.12.0 |
|
- Tokenizers 0.13.3 |
|
|
