predictive_analytics.py

import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler
from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier
from sklearn.metrics import mean_squared_error, r2_score, accuracy_score, classification_report

class PredictiveAnalytics:
    def __init__(self):
        self.model = None
        self.scaler = StandardScaler()
        self.target_column = None

    def predict(self, data):
        # Identify the target column (assuming it's the last column)
        self.target_column = data.columns[-1]
        
        # Prepare the data
        X = data.drop(columns=[self.target_column])
        y = data[self.target_column]

        # Split the data
        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

        # Scale the features
        X_train_scaled = self.scaler.fit_transform(X_train)
        X_test_scaled = self.scaler.transform(X_test)

        # Determine if it's a regression or classification problem
        if y.dtype == 'object' or len(np.unique(y)) < 10:
            self.model = RandomForestClassifier(n_estimators=100, random_state=42)
            is_classification = True
        else:
            self.model = RandomForestRegressor(n_estimators=100, random_state=42)
            is_classification = False

        # Train the model
        self.model.fit(X_train_scaled, y_train)

        # Make predictions
        y_pred = self.model.predict(X_test_scaled)

        # Evaluate the model
        if is_classification:
            accuracy = accuracy_score(y_test, y_pred)
            report = classification_report(y_test, y_pred)
            return f"Classification Results:\nAccuracy: {accuracy:.2f}\n\nClassification Report:\n{report}"
        else:
            mse = mean_squared_error(y_test, y_pred)
            r2 = r2_score(y_test, y_pred)
            return f"Regression Results:\nMean Squared Error: {mse:.2f}\nR-squared Score: {r2:.2f}"

    def get_feature_importance(self):
        if self.model is None:
            return "Model has not been trained yet."
        
        feature_importance = pd.DataFrame({
            'feature': self.model.feature_names_in_,
            'importance': self.model.feature_importances_
        }).sort_values('importance', ascending=False)
        
        return feature_importance