🧠 Image Classifier Trainer UI

A desktop GUI tool to train and test image classifiers using TensorFlow and Tkinter. Created by CS42.org.

---

📦 Features

• Select folders of categorized images for training.
• Customize image size, number of epochs, and test split.
• Track training progress with animated accuracy plot.
• See final validation accuracy after training.
• Display 5 random prediction results from validation set.
• Save/load model in .keras format.
• Toggle between dark and light themes.
• Reset the program anytime.

---

🖼 Folder Structure Example

├── airplane/
│   ├── 1.jpg
│   ├── 2.jpg
├── automobile/
│   ├── 1.jpg
│   ├── 2.jpg

---

🚀 Getting Started

1. Install dependencies

pip install -r requirements.txt

2. Run the app

python Image_Classifier_Code.py

---

🔧 Model Info

• 3 Conv2D + MaxPooling layers
• Dense layers with ReLU and Softmax
• Optimizer: Adam
• Loss: Sparse Categorical Crossentropy

---

📈 Screenshots

✅ GUI in action

📊 Live Accuracy Plot

---

💾 Outputs

• Trained model: model_ui.keras
• Class list: model_ui_classes.txt

---

📜 License: MIT

✔️ You can:

• Use commercially or personally
• Modify and redistribute

❗ You must:

• Include this license
• Attribute the original author

No warranty is provided with this software.

---

🧑‍💻 Built by CS42.org