Weather Image Classification

---

📚 Table of Contents

• Overview
• Dataset
• Approach
• Classes
• Results
• Technology Stack
• Licence

---

🔍 Overview

This program uses Python and the Tensorflow / Keras modules to classify Weather Images from a Kaggle dataset.

---

📊 Dataset

Download from OFFICIAL

---

🧠 Approach

There was a testing of parameters in the model(lr, different optimizers , dropout rates , pre-trained models and data-augmentation). The model was trained in 20 epochs, with a MobileNetV2/ResNet150V2 with early stopping for overfitting prevention. The results and the plots are being shown in the .ipynb file as well as the commentary.

---

🏷️ Classes

The pictures are divided into 11 classes:

1. dew

2. fog/smog

3. frost

4. glaze

5. hail

6. lightning

7. rain

8. rainbow

9. rime

10. sandstorm

11. snow

---

📊 Results

---

🛠️ Technology Stack

• Language: Python (v.??)
• Modules: Tensorflow/Keras, Numpy, Matplotlib (& other)
• Platform: Google Colab

---

⚖️ Licence

See the Licence