🐧 Penguin-Species-Classifier[Machine Learning App]

This is an interactive web application built using Streamlit that predicts the species of penguins based on user-provided features. The app uses a Random Forest Classifier machine learning model for prediction and provides insights into the dataset with visualizations.

---

📊 Features

1. Penguin Species Prediction:

   • Predicts the species of penguins (“Adelie”, “Chinstrap”, or “Gentoo”).
   • Uses features like bill length, bill depth, flipper length, body mass, island, and gender.

2. Data Visualization:

   • Provides a scatter plot to visualize the relationship between penguin features such as bill length and body mass.

3. Interactive Input:

   • Allows users to input penguin features through sliders and dropdowns for real-time predictions.

4. Probability Display:

   • Shows the probability of each penguin species based on the input features.

---

🔎 Dataset

The app uses the Palmer Archipelago (Antarctica) penguin dataset:

• Source: Palmer Penguins Dataset
• Features:
  • island: Island where the penguin was found.
  • bill_length_mm: Length of the penguin’s bill (in mm).
  • bill_depth_mm: Depth of the penguin’s bill (in mm).
  • flipper_length_mm: Length of the penguin’s flipper (in mm).
  • body_mass_g: Weight of the penguin (in grams).
  • sex: Gender of the penguin.
• Target: species (Adelie, Chinstrap, Gentoo).

---

⚖️ Machine Learning Model

The app uses the Random Forest Classifier:

• A robust, ensemble-based machine learning algorithm.
• Combines predictions from multiple decision trees for high accuracy.
• Encodes categorical variables like island and sex using one-hot encoding.
• Maps target labels (“Adelie”, “Chinstrap”, “Gentoo”) to numerical values for training.

---

🔧 Installation

Follow these steps to run the app locally:

Prerequisites

• Python 3.8 or higher
• Install required Python libraries:

  pip install streamlit pandas numpy scikit-learn

Clone the Repository

git clone <repository-url>
cd <repository-folder>

Run the App

streamlit run app.py

---

📊 How It Works

1. Load the Data: Loads the penguin dataset from a public URL.
2. Explore the Data: Displays raw data and visualizations.
3. Get User Inputs: Accepts user inputs via sliders and dropdown menus for features like bill length, bill depth, flipper length, etc.
4. Prepare Data: Encodes categorical data and scales numerical features as needed.
5. Train the Model: Trains a Random Forest Classifier on the dataset.
6. Make Predictions:
   • Predicts the penguin species based on user inputs.
   • Displays probabilities for each species.

---

🔄 Workflow Diagram

graph TD
A[Load Dataset] --> B[Explore Dataset]
B --> C[Visualize Data]
C --> D[Input User Features]
D --> E[Prepare Data for Model]
E --> F[Train Random Forest Classifier]
F --> G[Predict Species]
G --> H[Display Results]

Loading

---

🎮 App Interface

1. Sidebar:

   • Input penguin features (island, bill length, bill depth, etc.)
   • Real-time predictions based on the inputs.

2. Main Page:

   • Raw data table.
   • Scatter plot visualization of bill length vs. body mass.
   • Probabilities of each species and the predicted species.

---

🚀 Example Usage

1. Open the app by running the command streamlit run app.py.
2. Input the following example values:
   • Island: Dream
   • Bill length: 45 mm
   • Bill depth: 18 mm
   • Flipper length: 200 mm
   • Body mass: 4000 g
   • Gender: Male
3. The app predicts the penguin species (e.g., Gentoo) and displays probabilities for all species.

---

🔧 Technologies Used

• Frontend: Streamlit for interactive UI.
• Backend: Python for processing data and machine learning.
• Machine Learning: Scikit-learn’s Random Forest Classifier.
• Data Visualization: Streamlit’s charting tools.

---

⚙️ Future Improvements

• Add additional visualizations (e.g., histograms for each feature).
• Allow users to upload their datasets for predictions.
• Include more advanced machine learning algorithms for better accuracy.

---

🛠️ Troubleshooting

1. Missing Libraries:

   • Ensure all required libraries are installed.
   • Run: pip install -r requirements.txt.

2. Dataset Issues:

   • Ensure you have internet access as the dataset is fetched from a URL.

---

📢 Credits

• Dataset Source: Palmer Penguins Dataset
• Developed by: [Sandip Kushwaha]

---

📚 License

This project is licensed under the MIT License.