|
--- |
|
license: apache-2.0 |
|
base_model: facebook/deit-tiny-patch16-224 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- imagefolder |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: hushem_40x_deit_tiny_adamax_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: 0.8536585365853658 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# hushem_40x_deit_tiny_adamax_0001_fold5 |
|
|
|
This model is a fine-tuned version of [facebook/deit-tiny-patch16-224](https://huggingface.co/facebook/deit-tiny-patch16-224) on the imagefolder dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.9183 |
|
- Accuracy: 0.8537 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:| |
|
| 0.0769 | 1.0 | 220 | 0.5269 | 0.8293 | |
|
| 0.0227 | 2.0 | 440 | 0.7159 | 0.8293 | |
|
| 0.0008 | 3.0 | 660 | 1.1921 | 0.7805 | |
|
| 0.0031 | 4.0 | 880 | 0.8277 | 0.8537 | |
|
| 0.0006 | 5.0 | 1100 | 0.9658 | 0.8049 | |
|
| 0.0 | 6.0 | 1320 | 0.5880 | 0.8780 | |
|
| 0.0007 | 7.0 | 1540 | 0.9103 | 0.8293 | |
|
| 0.0006 | 8.0 | 1760 | 0.4548 | 0.8780 | |
|
| 0.0029 | 9.0 | 1980 | 1.0525 | 0.8293 | |
|
| 0.0 | 10.0 | 2200 | 0.8563 | 0.8293 | |
|
| 0.0 | 11.0 | 2420 | 0.7625 | 0.8537 | |
|
| 0.0 | 12.0 | 2640 | 0.7582 | 0.8537 | |
|
| 0.0 | 13.0 | 2860 | 0.7509 | 0.8537 | |
|
| 0.0 | 14.0 | 3080 | 0.7628 | 0.8537 | |
|
| 0.0 | 15.0 | 3300 | 0.7694 | 0.8537 | |
|
| 0.0 | 16.0 | 3520 | 0.7739 | 0.8537 | |
|
| 0.0 | 17.0 | 3740 | 0.7753 | 0.8537 | |
|
| 0.0 | 18.0 | 3960 | 0.7841 | 0.8780 | |
|
| 0.0 | 19.0 | 4180 | 0.7863 | 0.8780 | |
|
| 0.0 | 20.0 | 4400 | 0.8002 | 0.8537 | |
|
| 0.0 | 21.0 | 4620 | 0.7935 | 0.8537 | |
|
| 0.0 | 22.0 | 4840 | 0.8153 | 0.8537 | |
|
| 0.0 | 23.0 | 5060 | 0.8135 | 0.8537 | |
|
| 0.0 | 24.0 | 5280 | 0.8169 | 0.8537 | |
|
| 0.0 | 25.0 | 5500 | 0.8219 | 0.8537 | |
|
| 0.0 | 26.0 | 5720 | 0.8262 | 0.8537 | |
|
| 0.0 | 27.0 | 5940 | 0.8371 | 0.8537 | |
|
| 0.0 | 28.0 | 6160 | 0.8311 | 0.8537 | |
|
| 0.0 | 29.0 | 6380 | 0.8379 | 0.8537 | |
|
| 0.0 | 30.0 | 6600 | 0.8615 | 0.8537 | |
|
| 0.0 | 31.0 | 6820 | 0.8776 | 0.8537 | |
|
| 0.0 | 32.0 | 7040 | 0.8507 | 0.8537 | |
|
| 0.0 | 33.0 | 7260 | 0.8611 | 0.8537 | |
|
| 0.0 | 34.0 | 7480 | 0.8621 | 0.8537 | |
|
| 0.0 | 35.0 | 7700 | 0.8793 | 0.8537 | |
|
| 0.0 | 36.0 | 7920 | 0.8811 | 0.8537 | |
|
| 0.0 | 37.0 | 8140 | 0.8839 | 0.8537 | |
|
| 0.0 | 38.0 | 8360 | 0.8830 | 0.8537 | |
|
| 0.0 | 39.0 | 8580 | 0.8720 | 0.8537 | |
|
| 0.0 | 40.0 | 8800 | 0.8855 | 0.8537 | |
|
| 0.0 | 41.0 | 9020 | 0.8862 | 0.8537 | |
|
| 0.0 | 42.0 | 9240 | 0.9183 | 0.8537 | |
|
| 0.0 | 43.0 | 9460 | 0.8971 | 0.8537 | |
|
| 0.0 | 44.0 | 9680 | 0.9190 | 0.8537 | |
|
| 0.0 | 45.0 | 9900 | 0.9193 | 0.8537 | |
|
| 0.0 | 46.0 | 10120 | 0.9157 | 0.8537 | |
|
| 0.0 | 47.0 | 10340 | 0.9115 | 0.8537 | |
|
| 0.0 | 48.0 | 10560 | 0.9160 | 0.8537 | |
|
| 0.0 | 49.0 | 10780 | 0.9155 | 0.8537 | |
|
| 0.0 | 50.0 | 11000 | 0.9183 | 0.8537 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.32.1 |
|
- Pytorch 2.1.1+cu121 |
|
- Datasets 2.12.0 |
|
- Tokenizers 0.13.2 |
|
|