🐧 Palmer Penguins Species Classifier

A lightweight, high-accuracy Machine Learning pipeline built with Scikit-Learn and deployed to predict the species of Antarctic penguins (Adélie, Chinstrap, or Gentoo) based on anatomical measurements and island habitat.

📌 Model Summary

• Model Type: Random Forest Classifier (n_estimators=100)
• Preprocessing: Median Imputation + StandardScaler for numeric features; Mode Imputation + OneHotEncoder for categorical features wrapped inside a Scikit-Learn ColumnTransformer & Pipeline.
• Dataset: Palmer Archipelago (Antarctica) Penguin Dataset (344 samples).
• Test Accuracy: 100.0%

---

🎯 Input Features

Feature	Type	Description
island	Categorical	Island where penguin was observed (Torgersen, Biscoe, Dream)
bill_length_mm	Float	Culmen length (mm)
bill_depth_mm	Float	Culmen depth (mm)
flipper_length_mm	Float	Flipper length (mm)
body_mass_g	Float	Body mass in grams
sex	Categorical	Biological sex (Male, Female)

---

🚀 How to Use

You can load and run inference with this model directly using joblib and huggingface_hub:

import joblib
import pandas as pd
from huggingface_hub import hf_hub_download

# Download artifacts from Hugging Face Hub
repo_id = "<YOUR_HF_USERNAME>/Penguins"
model_path = hf_hub_download(repo_id=repo_id, filename="artifacts/model_pipeline.joblib")
labels_path = hf_hub_download(repo_id=repo_id, filename="artifacts/labels.joblib")

# Load pipeline & labels
pipeline = joblib.load(model_path)
classes = joblib.load(labels_path)

# Sample penguin
sample_data = pd.DataFrame([{
    "island": "Torgersen",
    "bill_length_mm": 39.1,
    "bill_depth_mm": 18.7,
    "flipper_length_mm": 181.0,
    "body_mass_g": 3750.0,
    "sex": "Male"
}])

prediction = pipeline.predict(sample_data)[0]
print(f"Predicted Species: {prediction}")

---

📜 License

MIT License