🌟 Pokémon – Professor Oak's Call 🔎

⚡ The Young Trainer’s Mission

On a clear morning in the Kanto region, you are invited by Professor Oak, the most respected Pokémon researcher, to take on a new challenge where you can demonstrate your talent and skills.

A power outage in his laboratory has disorganized part of his precious collection of Pokémon images and data, and now only a young trainer with an analytical mind and a passion for science can help him.

Professor Oak managed to recover two essential CSV files (train.csv and test.csv) containing the following columns:

• SampleID – a unique numeric identifier (e.g., 000001, 000002, …)
• ImagePath – path to the image (e.g., images/000001.png)
• Type – Pokémon type (only in train.csv; missing in test.csv)

All images are stored in the images/ directory, named according to the rule:
images/<SampleID>.png

These are the only recovered pieces of information and the only data you are allowed to use to reconstruct the dataset.

🎯 Your Task

Professor Oak needs your help to recover his lost data and continue his study on Pokémon species recognition from images.

Your mission is to select and train the best model capable of intelligently identifying the various Pokémon types using the images provided.

The dataset contains already labeled examples for training, and your final challenge is to use your trained model to classify, accurately and elegantly, the Pokémon images in test.csv.

Only the best trainers can combine science with intuition, and now it’s your turn to prove you deserve the title of Pokémon Master.

📘 Example of Final Output

SampleID,Type
000001,Grass
000002,Fire
000003,Water

📊 Evaluation

Predictions will be compared to the true types and accuracy will be calculated:

accuracy = (number_of_correct_predictions / total_number_of_predictions)

The final score is determined based on the accuracy using the following rules:

• accuracy ≥ 90% → 100 points
• accuracy < 20% → 0 points
• For intermediate values, the score is scaled proportionally between 0 and 100.

⚠️ Important: Only classification of the images in test.csv counts for evaluation. The submission file must include the columns SampleID,Type and be in csv format.

🏆 Good luck, young trainer! The fate of the Pokémon encyclopedia depends on you.