|
--- |
|
license: apache-2.0 |
|
base_model: google/vit-base-patch32-224-in21k |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- imagefolder |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: adam_ViTB-32-224-in21k-2e-4-batch_16_epoch_4_classes_24 |
|
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: 0.9568965517241379 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# adam_ViTB-32-224-in21k-2e-4-batch_16_epoch_4_classes_24 |
|
|
|
This model is a fine-tuned version of [google/vit-base-patch32-224-in21k](https://huggingface.co/google/vit-base-patch32-224-in21k) on the imagefolder dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2485 |
|
- Accuracy: 0.9569 |
|
|
|
## 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: 0.0002 |
|
- 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 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:| |
|
| 1.1079 | 0.07 | 100 | 1.0858 | 0.8506 | |
|
| 0.5485 | 0.14 | 200 | 0.6077 | 0.8649 | |
|
| 0.4432 | 0.21 | 300 | 0.4838 | 0.8822 | |
|
| 0.3463 | 0.28 | 400 | 0.4479 | 0.8764 | |
|
| 0.3089 | 0.35 | 500 | 0.3687 | 0.8908 | |
|
| 0.2341 | 0.42 | 600 | 0.4974 | 0.8635 | |
|
| 0.2635 | 0.49 | 700 | 0.3657 | 0.8894 | |
|
| 0.2038 | 0.56 | 800 | 0.2892 | 0.9080 | |
|
| 0.1374 | 0.63 | 900 | 0.3617 | 0.8865 | |
|
| 0.2198 | 0.7 | 1000 | 0.3332 | 0.9037 | |
|
| 0.2532 | 0.77 | 1100 | 0.3292 | 0.9037 | |
|
| 0.1897 | 0.84 | 1200 | 0.2957 | 0.9167 | |
|
| 0.1718 | 0.91 | 1300 | 0.2398 | 0.9339 | |
|
| 0.1637 | 0.97 | 1400 | 0.3514 | 0.9009 | |
|
| 0.0794 | 1.04 | 1500 | 0.2616 | 0.9224 | |
|
| 0.0541 | 1.11 | 1600 | 0.3213 | 0.9124 | |
|
| 0.0475 | 1.18 | 1700 | 0.3717 | 0.9124 | |
|
| 0.1251 | 1.25 | 1800 | 0.2938 | 0.9195 | |
|
| 0.0712 | 1.32 | 1900 | 0.2988 | 0.9181 | |
|
| 0.1021 | 1.39 | 2000 | 0.3862 | 0.9009 | |
|
| 0.0073 | 1.46 | 2100 | 0.2492 | 0.9310 | |
|
| 0.0114 | 1.53 | 2200 | 0.2902 | 0.9267 | |
|
| 0.0487 | 1.6 | 2300 | 0.2301 | 0.9411 | |
|
| 0.0856 | 1.67 | 2400 | 0.2682 | 0.9411 | |
|
| 0.0028 | 1.74 | 2500 | 0.2948 | 0.9325 | |
|
| 0.0028 | 1.81 | 2600 | 0.3002 | 0.9282 | |
|
| 0.0279 | 1.88 | 2700 | 0.2797 | 0.9353 | |
|
| 0.0768 | 1.95 | 2800 | 0.2721 | 0.9368 | |
|
| 0.0251 | 2.02 | 2900 | 0.2896 | 0.9325 | |
|
| 0.0645 | 2.09 | 3000 | 0.2802 | 0.9397 | |
|
| 0.0022 | 2.16 | 3100 | 0.2387 | 0.9468 | |
|
| 0.0073 | 2.23 | 3200 | 0.2074 | 0.9540 | |
|
| 0.0016 | 2.3 | 3300 | 0.2271 | 0.9440 | |
|
| 0.0016 | 2.37 | 3400 | 0.2513 | 0.9526 | |
|
| 0.0656 | 2.44 | 3500 | 0.2889 | 0.9411 | |
|
| 0.0013 | 2.51 | 3600 | 0.2750 | 0.9397 | |
|
| 0.0014 | 2.58 | 3700 | 0.2463 | 0.9526 | |
|
| 0.0011 | 2.65 | 3800 | 0.2723 | 0.9483 | |
|
| 0.0012 | 2.72 | 3900 | 0.2631 | 0.9511 | |
|
| 0.0012 | 2.79 | 4000 | 0.2584 | 0.9540 | |
|
| 0.0012 | 2.86 | 4100 | 0.2572 | 0.9540 | |
|
| 0.001 | 2.92 | 4200 | 0.2481 | 0.9569 | |
|
| 0.0011 | 2.99 | 4300 | 0.2485 | 0.9569 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.3 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
