This project explores the application of convolutional neural networks (CNNs) for the task of animal image classification. I have curated a dataset of diverse animal images and employed transfer learning to fine-tune pre-trained models on our specific task. The goal is to achieve high classification accuracy and contribute to the field of computer vision.
https://huggingface.co/spaces/dielz/animals-classifier-demo
- Using a dataset with a total amount of 10000+ data
- Training and test accuracy of at least 95%
- The model was able to classify at least 3 different classes
To support the define objectives, the dataset was obtained from kaggle with 12 different classes and 17000+ data. However, due to limited resources, only 10 classes were used in this project.
| Model | Structure | Data Splitting | Training Accuracy | Validation Accuracy |
|---|---|---|---|---|
| 1 | Sequential + MobileNetV2 + Conv2D + MaxPool2D | 80/20 | 98% | 95% |
Seen from the summary and prediction results of the model that has fulfilled all the points that are the objective of this project.