app-penguins.py

import streamlit as st
import pandas as pd
from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.preprocessing import StandardScaler, OneHotEncoder
from sklearn.compose import ColumnTransformer
from sklearn.neighbors import KNeighborsClassifier
from sklearn.metrics import accuracy_score

# Load your data (replace with your actual data loading)
penguins = pd.read_csv('penguins_lter.csv')

# Data Cleaning (same as your existing code)
penguins_cleaned = penguins.dropna()
penguins_cleaned = penguins_cleaned.drop_duplicates()

# Fill missing values (same as your existing code)
numerical_cols = penguins.select_dtypes(include=['number']).columns
penguins[numerical_cols] = penguins[numerical_cols].fillna(penguins[numerical_cols].mean())
categorical_cols = penguins.select_dtypes(include=['object']).columns
penguins[categorical_cols] = penguins[categorical_cols].fillna(penguins[categorical_cols].mode().iloc[0])


# Feature Engineering and Model Training (same as your existing code)
X = penguins.drop('Species', axis=1)
y = penguins['Species']
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

numerical_features = ['Culmen Length (mm)', 'Culmen Depth (mm)', 'Flipper Length (mm)', 'Body Mass (g)']
categorical_features = ['Island', 'Sex']

numerical_transformer = Pipeline(steps=[('scaler', StandardScaler())])
categorical_transformer = Pipeline(steps=[('onehot', OneHotEncoder(handle_unknown='ignore'))])

preprocessor = ColumnTransformer(
    transformers=[
        ('num', numerical_transformer, numerical_features),
        ('cat', categorical_transformer, categorical_features)
    ])

pipeline = Pipeline(steps=[
    ('preprocessor', preprocessor),
    ('classifier', KNeighborsClassifier())
])

pipeline.fit(X_train, y_train)
y_pred = pipeline.predict(X_test)
accuracy = accuracy_score(y_test, y_pred)

# Streamlit App
st.title("Penguin Species Classification")

st.write("This app predicts the species of a penguin based on its features.")

# Display the accuracy
st.write(f"Model Accuracy: {accuracy}")

# Input features for prediction
culmen_length = st.number_input("Culmen Length (mm)", min_value=0.0)
culmen_depth = st.number_input("Culmen Depth (mm)", min_value=0.0)
flipper_length = st.number_input("Flipper Length (mm)", min_value=0.0)
body_mass = st.number_input("Body Mass (g)", min_value=0.0)
island = st.selectbox("Island", penguins['Island'].unique())
sex = st.selectbox("Sex", penguins['Sex'].unique())


# Create a DataFrame for prediction
new_penguin = pd.DataFrame({
    'Culmen Length (mm)': [culmen_length],
    'Culmen Depth (mm)': [culmen_depth],
    'Flipper Length (mm)': [flipper_length],
    'Body Mass (g)': [body_mass],
    'Island': [island],
    'Sex': [sex]
    })


# Make prediction
if st.button("Predict Species"):
  prediction = pipeline.predict(new_penguin)
  st.write(f"Predicted Species: {prediction[0]}")