|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: google/vit-base-patch16-224 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- medmnist-v2 |
|
metrics: |
|
- accuracy |
|
- f1 |
|
model-index: |
|
- name: ViT_breastmnist_std_30 |
|
results: |
|
- task: |
|
name: Image Classification |
|
type: image-classification |
|
dataset: |
|
name: medmnist-v2 |
|
type: medmnist-v2 |
|
config: breastmnist |
|
split: validation |
|
args: breastmnist |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.8269230769230769 |
|
- name: F1 |
|
type: f1 |
|
value: 0.7314974182444062 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# ViT_breastmnist_std_30 |
|
|
|
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the medmnist-v2 dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3936 |
|
- Accuracy: 0.8269 |
|
- F1: 0.7315 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 64 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 10 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | |
|
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:| |
|
| 0.5034 | 0.2597 | 20 | 0.4719 | 0.7436 | 0.4708 | |
|
| 0.4414 | 0.5195 | 40 | 0.4457 | 0.7821 | 0.6400 | |
|
| 0.3762 | 0.7792 | 60 | 0.4212 | 0.8205 | 0.7248 | |
|
| 0.4059 | 1.0390 | 80 | 0.3988 | 0.8462 | 0.7641 | |
|
| 0.3249 | 1.2987 | 100 | 0.3829 | 0.8333 | 0.7606 | |
|
| 0.2991 | 1.5584 | 120 | 0.4080 | 0.8462 | 0.7743 | |
|
| 0.2948 | 1.8182 | 140 | 0.3932 | 0.8462 | 0.7833 | |
|
| 0.2667 | 2.0779 | 160 | 0.4388 | 0.8333 | 0.7502 | |
|
| 0.2049 | 2.3377 | 180 | 0.4047 | 0.8333 | 0.7606 | |
|
| 0.1639 | 2.5974 | 200 | 0.4301 | 0.8333 | 0.7502 | |
|
| 0.1732 | 2.8571 | 220 | 0.4028 | 0.8333 | 0.7606 | |
|
| 0.1138 | 3.1169 | 240 | 0.3755 | 0.8718 | 0.8194 | |
|
| 0.1099 | 3.3766 | 260 | 0.4019 | 0.8590 | 0.7886 | |
|
| 0.1285 | 3.6364 | 280 | 0.3739 | 0.8590 | 0.7974 | |
|
| 0.1265 | 3.8961 | 300 | 0.3714 | 0.8590 | 0.8051 | |
|
| 0.0735 | 4.1558 | 320 | 0.3820 | 0.8718 | 0.8194 | |
|
| 0.0515 | 4.4156 | 340 | 0.3910 | 0.8462 | 0.7833 | |
|
| 0.0577 | 4.6753 | 360 | 0.3984 | 0.8462 | 0.7833 | |
|
| 0.0584 | 4.9351 | 380 | 0.4314 | 0.8590 | 0.7974 | |
|
| 0.0241 | 5.1948 | 400 | 0.4040 | 0.8718 | 0.8194 | |
|
| 0.015 | 5.4545 | 420 | 0.4201 | 0.8718 | 0.8194 | |
|
| 0.023 | 5.7143 | 440 | 0.4276 | 0.8718 | 0.8194 | |
|
| 0.0254 | 5.9740 | 460 | 0.4271 | 0.8846 | 0.8342 | |
|
| 0.0086 | 6.2338 | 480 | 0.4149 | 0.8718 | 0.8194 | |
|
| 0.012 | 6.4935 | 500 | 0.4738 | 0.8718 | 0.8120 | |
|
| 0.0052 | 6.7532 | 520 | 0.4314 | 0.8846 | 0.8342 | |
|
| 0.0123 | 7.0130 | 540 | 0.4363 | 0.8718 | 0.8194 | |
|
| 0.0026 | 7.2727 | 560 | 0.4477 | 0.8846 | 0.8342 | |
|
| 0.0018 | 7.5325 | 580 | 0.4447 | 0.8718 | 0.8194 | |
|
| 0.0024 | 7.7922 | 600 | 0.4588 | 0.8718 | 0.8194 | |
|
| 0.0076 | 8.0519 | 620 | 0.4517 | 0.8718 | 0.8194 | |
|
| 0.0013 | 8.3117 | 640 | 0.4535 | 0.8718 | 0.8194 | |
|
| 0.0012 | 8.5714 | 660 | 0.4479 | 0.8846 | 0.8342 | |
|
| 0.001 | 8.8312 | 680 | 0.4477 | 0.8846 | 0.8342 | |
|
| 0.0015 | 9.0909 | 700 | 0.4509 | 0.8846 | 0.8342 | |
|
| 0.001 | 9.3506 | 720 | 0.4529 | 0.8846 | 0.8342 | |
|
| 0.0009 | 9.6104 | 740 | 0.4569 | 0.8846 | 0.8342 | |
|
| 0.001 | 9.8701 | 760 | 0.4563 | 0.8846 | 0.8342 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.1 |
|
- Pytorch 2.4.0 |
|
- Datasets 3.0.1 |
|
- Tokenizers 0.20.0 |
|
|
