|
--- |
|
license: apache-2.0 |
|
base_model: facebook/deit-small-patch16-224 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- imagefolder |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: smids_5x_deit_small_rms_0001_fold5 |
|
results: |
|
- task: |
|
name: Image Classification |
|
type: image-classification |
|
dataset: |
|
name: imagefolder |
|
type: imagefolder |
|
config: default |
|
split: test |
|
args: default |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.915 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# smids_5x_deit_small_rms_0001_fold5 |
|
|
|
This model is a fine-tuned version of [facebook/deit-small-patch16-224](https://huggingface.co/facebook/deit-small-patch16-224) on the imagefolder dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8725 |
|
- Accuracy: 0.915 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_ratio: 0.1 |
|
- num_epochs: 50 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:| |
|
| 0.2465 | 1.0 | 375 | 0.3246 | 0.8817 | |
|
| 0.1757 | 2.0 | 750 | 0.3108 | 0.8983 | |
|
| 0.1416 | 3.0 | 1125 | 0.4018 | 0.8917 | |
|
| 0.0805 | 4.0 | 1500 | 0.5614 | 0.8733 | |
|
| 0.0831 | 5.0 | 1875 | 0.4489 | 0.8783 | |
|
| 0.0313 | 6.0 | 2250 | 0.6222 | 0.8833 | |
|
| 0.0541 | 7.0 | 2625 | 0.5461 | 0.9017 | |
|
| 0.0151 | 8.0 | 3000 | 0.5573 | 0.9017 | |
|
| 0.0125 | 9.0 | 3375 | 0.6886 | 0.8817 | |
|
| 0.0415 | 10.0 | 3750 | 0.5097 | 0.8967 | |
|
| 0.0264 | 11.0 | 4125 | 0.6554 | 0.89 | |
|
| 0.0122 | 12.0 | 4500 | 0.5509 | 0.8983 | |
|
| 0.004 | 13.0 | 4875 | 0.6519 | 0.895 | |
|
| 0.0092 | 14.0 | 5250 | 0.8038 | 0.88 | |
|
| 0.0491 | 15.0 | 5625 | 0.5627 | 0.9117 | |
|
| 0.0096 | 16.0 | 6000 | 0.7380 | 0.885 | |
|
| 0.003 | 17.0 | 6375 | 0.6721 | 0.8967 | |
|
| 0.0153 | 18.0 | 6750 | 0.6251 | 0.9017 | |
|
| 0.0001 | 19.0 | 7125 | 0.7496 | 0.8933 | |
|
| 0.0067 | 20.0 | 7500 | 0.7015 | 0.895 | |
|
| 0.0015 | 21.0 | 7875 | 0.7054 | 0.9017 | |
|
| 0.0117 | 22.0 | 8250 | 0.7731 | 0.8933 | |
|
| 0.0015 | 23.0 | 8625 | 0.6632 | 0.9033 | |
|
| 0.0003 | 24.0 | 9000 | 0.7347 | 0.9 | |
|
| 0.0 | 25.0 | 9375 | 0.8344 | 0.8917 | |
|
| 0.0002 | 26.0 | 9750 | 0.7080 | 0.905 | |
|
| 0.0026 | 27.0 | 10125 | 0.8121 | 0.8933 | |
|
| 0.0001 | 28.0 | 10500 | 0.7303 | 0.915 | |
|
| 0.0285 | 29.0 | 10875 | 0.7237 | 0.905 | |
|
| 0.0021 | 30.0 | 11250 | 0.7546 | 0.9 | |
|
| 0.0096 | 31.0 | 11625 | 0.8217 | 0.8933 | |
|
| 0.0 | 32.0 | 12000 | 0.8254 | 0.9017 | |
|
| 0.0 | 33.0 | 12375 | 0.7430 | 0.905 | |
|
| 0.0 | 34.0 | 12750 | 0.7668 | 0.9083 | |
|
| 0.0002 | 35.0 | 13125 | 0.7980 | 0.9117 | |
|
| 0.0018 | 36.0 | 13500 | 0.8507 | 0.9017 | |
|
| 0.0 | 37.0 | 13875 | 0.7364 | 0.9133 | |
|
| 0.0038 | 38.0 | 14250 | 0.7573 | 0.9133 | |
|
| 0.0 | 39.0 | 14625 | 0.7786 | 0.9083 | |
|
| 0.0 | 40.0 | 15000 | 0.7610 | 0.915 | |
|
| 0.0 | 41.0 | 15375 | 0.7902 | 0.9117 | |
|
| 0.0 | 42.0 | 15750 | 0.7539 | 0.9133 | |
|
| 0.0 | 43.0 | 16125 | 0.8516 | 0.9133 | |
|
| 0.0 | 44.0 | 16500 | 0.8335 | 0.9183 | |
|
| 0.0 | 45.0 | 16875 | 0.8847 | 0.905 | |
|
| 0.0 | 46.0 | 17250 | 0.9773 | 0.8967 | |
|
| 0.0028 | 47.0 | 17625 | 0.8669 | 0.915 | |
|
| 0.0 | 48.0 | 18000 | 0.8681 | 0.9133 | |
|
| 0.0 | 49.0 | 18375 | 0.8719 | 0.9133 | |
|
| 0.0019 | 50.0 | 18750 | 0.8725 | 0.915 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.32.1 |
|
- Pytorch 2.1.0+cu121 |
|
- Datasets 2.12.0 |
|
- Tokenizers 0.13.2 |
|
|
