# Dataset Description  
The dataset consists of Pokémon images from 150 categories, with around 55 images per category. These images have been curated to represent different Pokémon from the iconic franchise.

## Files  

### 1. Train Data  
- The train dataset is organized into 150 subfolders, each named after a Pokémon.  
- Each subfolder contains images corresponding to that Pokémon. These images serve as labeled data to train your classification model.  


### 2. Test Data  
- The test dataset contains **2,195 images**, labeled sequentially from `1.jpg` to `2195.jpg`.  
- These images are unlabeled, and participants need to predict the corresponding Pokémon category for each image.  

### 3. Submission Template  
- The submission template provides the required format for participants to submit their predictions.  
- Submissions must contain two columns:  
- **Img_name :** The name of the test image (e.g., `1.jpg`).  
- **Class Prediction:** The predicted Pokémon name corresponding to the test image.  
- **Example structure:**  

## Key Information  

### What files do I need?  
You need the **train data**, **test data**, and the **submission template**.  

### What is the data format?  
- **Training data:** Images organized into subfolders named after Pokémon categories.  
- **Test data:** Unlabeled images named sequentially (e.g., `1.jpg`, `2.jpg`).  
- **Submission:** CSV file with `Img_name` and `Class Prediction` columns.  

### What are you predicting?  
You are predicting the **Pokémon name** (category) for each test image.  

### What acronyms or terms might I encounter?  
- **CNN:** Convolutional Neural Network, a type of deep learning model.  
- **Img_name :** The file name of the test image.  
- **Class Prediction:** The Pokémon category predicted by the model.  

---

Take a moment to explore the dataset files before starting your modeling process. Good luck, and may your CNN catch ‘em all! 🚀