|
--- |
|
license: apache-2.0 |
|
base_model: google/vit-base-patch16-224-in21k |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- imagefolder |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: emotion_face_image_classification |
|
results: |
|
- task: |
|
name: Image Classification |
|
type: image-classification |
|
dataset: |
|
name: imagefolder |
|
type: imagefolder |
|
config: default |
|
split: train |
|
args: default |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.55 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# emotion_face_image_classification |
|
|
|
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 1.2556 |
|
- Accuracy: 0.55 |
|
|
|
## 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: 64 |
|
- eval_batch_size: 64 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 50 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| No log | 1.0 | 10 | 2.0561 | 0.1437 | |
|
| No log | 2.0 | 20 | 1.9946 | 0.325 | |
|
| No log | 3.0 | 30 | 1.9059 | 0.3563 | |
|
| No log | 4.0 | 40 | 1.7640 | 0.3937 | |
|
| No log | 5.0 | 50 | 1.6329 | 0.3812 | |
|
| No log | 6.0 | 60 | 1.5935 | 0.375 | |
|
| No log | 7.0 | 70 | 1.5206 | 0.4313 | |
|
| No log | 8.0 | 80 | 1.4580 | 0.4813 | |
|
| No log | 9.0 | 90 | 1.4422 | 0.4938 | |
|
| No log | 10.0 | 100 | 1.4297 | 0.4437 | |
|
| No log | 11.0 | 110 | 1.4020 | 0.5062 | |
|
| No log | 12.0 | 120 | 1.3344 | 0.5125 | |
|
| No log | 13.0 | 130 | 1.3541 | 0.5125 | |
|
| No log | 14.0 | 140 | 1.2990 | 0.5125 | |
|
| No log | 15.0 | 150 | 1.2482 | 0.5687 | |
|
| No log | 16.0 | 160 | 1.2726 | 0.4875 | |
|
| No log | 17.0 | 170 | 1.2523 | 0.5375 | |
|
| No log | 18.0 | 180 | 1.2857 | 0.4813 | |
|
| No log | 19.0 | 190 | 1.2476 | 0.5375 | |
|
| No log | 20.0 | 200 | 1.2991 | 0.4625 | |
|
| No log | 21.0 | 210 | 1.3108 | 0.5062 | |
|
| No log | 22.0 | 220 | 1.3052 | 0.5 | |
|
| No log | 23.0 | 230 | 1.2606 | 0.4813 | |
|
| No log | 24.0 | 240 | 1.1568 | 0.5687 | |
|
| No log | 25.0 | 250 | 1.2242 | 0.5437 | |
|
| No log | 26.0 | 260 | 1.2031 | 0.5875 | |
|
| No log | 27.0 | 270 | 1.1649 | 0.5625 | |
|
| No log | 28.0 | 280 | 1.3134 | 0.4938 | |
|
| No log | 29.0 | 290 | 1.2542 | 0.5062 | |
|
| No log | 30.0 | 300 | 1.1542 | 0.5687 | |
|
| No log | 31.0 | 310 | 1.1619 | 0.5625 | |
|
| No log | 32.0 | 320 | 1.2700 | 0.5 | |
|
| No log | 33.0 | 330 | 1.1943 | 0.55 | |
|
| No log | 34.0 | 340 | 1.1441 | 0.575 | |
|
| No log | 35.0 | 350 | 1.2405 | 0.5687 | |
|
| No log | 36.0 | 360 | 1.2228 | 0.5375 | |
|
| No log | 37.0 | 370 | 1.1598 | 0.6 | |
|
| No log | 38.0 | 380 | 1.1894 | 0.5687 | |
|
| No log | 39.0 | 390 | 1.2493 | 0.5188 | |
|
| No log | 40.0 | 400 | 1.1716 | 0.5687 | |
|
| No log | 41.0 | 410 | 1.1558 | 0.5813 | |
|
| No log | 42.0 | 420 | 1.2529 | 0.5312 | |
|
| No log | 43.0 | 430 | 1.2515 | 0.5625 | |
|
| No log | 44.0 | 440 | 1.2576 | 0.5188 | |
|
| No log | 45.0 | 450 | 1.1793 | 0.5687 | |
|
| No log | 46.0 | 460 | 1.2252 | 0.5625 | |
|
| No log | 47.0 | 470 | 1.1876 | 0.575 | |
|
| No log | 48.0 | 480 | 1.2288 | 0.5375 | |
|
| No log | 49.0 | 490 | 1.2623 | 0.5125 | |
|
| 0.989 | 50.0 | 500 | 1.1603 | 0.5875 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.33.2 |
|
- Pytorch 2.0.1+cu118 |
|
- Datasets 2.14.5 |
|
- Tokenizers 0.13.3 |
|
|
