Animal Classification App

This is an animal classifier application based on the Mindspore framework and using a fine-tuned version of the ResNet50 Model. The application is built using Gradio.

Setup

1️⃣ Fork the repository

Fork this repository and enter the root folder

2️⃣ Install Dependencies

pip install -r requirements.txt

User Guide (note: all commands should be entered via the terminal)

---

1️⃣ Train Fine-Tuned ResNet50 Model

python train_resnet50.py

📁 Output:

• The trained model will be stored at models/resnet50_animals_best.ckpt

---

2️⃣ Launch the Animal Classification App

python app.py

The terminal should output something like: Running on local URL: http://127.0.0.1:7860. Copy the URL into the browser to open the app.

You can select one of the available photos of animals or upload one of your own to see what animal the model predicts it to be!