vit-base-patch16-224-in21k_dog_vs_cat_image_classification
This model is a fine-tuned version of google/vit-base-patch16-224-in21k. It achieves the following results on the evaluation set:
- Loss: 0.0404
- Accuracy: 0.99
- F1: 0.9897
- Recall: 0.9909
- Precision: 0.9885
Model description
This is a binary classification model to distinguish between cats and dogs.
For more information on how it was created, check out the following link: https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Image%20Classification/Binary%20Classification/Dogs%20or%20Cats%20Image%20Classification/Dog_v_Cat_ViT.ipynb
Intended uses & limitations
This model is intended to demonstrate my ability to solve a complex problem using technology.
Training and evaluation data
Dataset Source: https://www.kaggle.com/datasets/shaunthesheep/microsoft-catsvsdogs-dataset
Sample Images From Dataset:
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
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision |
|---|---|---|---|---|---|---|---|
| 0.0896 | 1.0 | 1250 | 0.0590 | 0.979 | 0.9783 | 0.9728 | 0.9838 |
| 0.0253 | 2.0 | 2500 | 0.0543 | 0.9842 | 0.9837 | 0.9802 | 0.9871 |
| 0.0066 | 3.0 | 3750 | 0.0404 | 0.99 | 0.9897 | 0.9909 | 0.9885 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.12.1
- Datasets 2.8.0
- Tokenizers 0.12.1
