|
--- |
|
license: apache-2.0 |
|
base_model: dandelin/vilt-b32-mlm |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: vilt_finetuned_200 |
|
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. --> |
|
|
|
# vilt_finetuned_200 |
|
|
|
This model is a fine-tuned version of [dandelin/vilt-b32-mlm](https://huggingface.co/dandelin/vilt-b32-mlm) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 4.3306 |
|
|
|
## 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: 16 |
|
- 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 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 363.9675 | 0.16 | 100 | 26.1215 | |
|
| 11.4975 | 0.32 | 200 | 7.2332 | |
|
| 6.1909 | 0.48 | 300 | 5.9332 | |
|
| 5.2134 | 0.64 | 400 | 5.5186 | |
|
| 5.0189 | 0.8 | 500 | 5.3268 | |
|
| 4.7551 | 0.96 | 600 | 5.0921 | |
|
| 4.5394 | 1.12 | 700 | 4.9538 | |
|
| 4.3441 | 1.28 | 800 | 4.8967 | |
|
| 4.1436 | 1.44 | 900 | 4.7419 | |
|
| 4.1847 | 1.6 | 1000 | 4.6581 | |
|
| 4.0116 | 1.76 | 1100 | 4.5915 | |
|
| 3.918 | 1.92 | 1200 | 4.5202 | |
|
| 3.8251 | 2.08 | 1300 | 4.4634 | |
|
| 3.7981 | 2.24 | 1400 | 4.4169 | |
|
| 3.7108 | 2.4 | 1500 | 4.3954 | |
|
| 3.5706 | 2.56 | 1600 | 4.3626 | |
|
| 3.5559 | 2.72 | 1700 | 4.3374 | |
|
| 3.6951 | 2.88 | 1800 | 4.3306 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
