|
--- |
|
license: apache-2.0 |
|
base_model: google/vit-base-patch16-224-in21k |
|
tags: |
|
- image-classification |
|
- bird species identification |
|
- generated_from_trainer |
|
datasets: |
|
- imagefolder |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: image_classification_obipix_birdID |
|
results: |
|
- task: |
|
name: Image Classification |
|
type: image-classification |
|
dataset: |
|
name: private crawled images |
|
type: imagefolder |
|
config: default |
|
split: train |
|
args: default |
|
metrics: |
|
- name: Accuracy |
|
type: accuracy |
|
value: 0.9719696025912545 |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# image_classification_obipix_birdID |
|
|
|
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 private crawled images dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1150 |
|
- Accuracy: 0.9720 |
|
|
|
## 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: 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: 5 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:| |
|
| 6.9257 | 0.18 | 1000 | 5.3830 | 0.1638 | |
|
| 3.9727 | 0.35 | 2000 | 2.7695 | 0.4797 | |
|
| 2.057 | 0.53 | 3000 | 1.5070 | 0.6936 | |
|
| 1.2103 | 0.7 | 4000 | 0.9727 | 0.7842 | |
|
| 0.8513 | 0.88 | 5000 | 0.7101 | 0.8318 | |
|
| 0.5836 | 1.06 | 6000 | 0.5797 | 0.8561 | |
|
| 0.3545 | 1.23 | 7000 | 0.5066 | 0.8730 | |
|
| 0.314 | 1.41 | 8000 | 0.4521 | 0.8818 | |
|
| 0.2858 | 1.58 | 9000 | 0.3915 | 0.8960 | |
|
| 0.2482 | 1.76 | 10000 | 0.3564 | 0.9056 | |
|
| 0.2192 | 1.93 | 11000 | 0.3131 | 0.9148 | |
|
| 0.1271 | 2.11 | 12000 | 0.2916 | 0.9207 | |
|
| 0.0779 | 2.29 | 13000 | 0.2727 | 0.9260 | |
|
| 0.0749 | 2.46 | 14000 | 0.2597 | 0.9309 | |
|
| 0.0682 | 2.64 | 15000 | 0.2415 | 0.9355 | |
|
| 0.0615 | 2.81 | 16000 | 0.2268 | 0.9385 | |
|
| 0.0566 | 2.99 | 17000 | 0.2084 | 0.9440 | |
|
| 0.0197 | 3.17 | 18000 | 0.1951 | 0.9475 | |
|
| 0.0158 | 3.34 | 19000 | 0.1843 | 0.9513 | |
|
| 0.0145 | 3.52 | 20000 | 0.1746 | 0.9541 | |
|
| 0.0118 | 3.69 | 21000 | 0.1649 | 0.9573 | |
|
| 0.0103 | 3.87 | 22000 | 0.1531 | 0.9599 | |
|
| 0.006 | 4.05 | 23000 | 0.1379 | 0.9644 | |
|
| 0.0016 | 4.22 | 24000 | 0.1316 | 0.9668 | |
|
| 0.0013 | 4.4 | 25000 | 0.1265 | 0.9686 | |
|
| 0.0014 | 4.57 | 26000 | 0.1232 | 0.9697 | |
|
| 0.0009 | 4.75 | 27000 | 0.1189 | 0.9712 | |
|
| 0.001 | 4.92 | 28000 | 0.1150 | 0.9720 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.0 |
|
|
