Create demo_ai_sim_pipeline.py

Author: Web4application

src/demo_ai_sim_pipeline.py

@@ -0,0 +1,74 @@
+import pandas as pd
+import numpy as np
+from sklearn.linear_model import LinearRegression
+import openpyxl
+from pathlib import Path
+
+# Path to your Aura workbook
+AURA_FILE = Path("../data/Aura.xlsl")  # treat .xlsl like .xlsx
+
+def load_data(sheet_name="SimulationInput"):
+    """Load input parameters from the Aura workbook."""
+    df = pd.read_excel(AURA_FILE, sheet_name=sheet_name)
+    return df
+
+def run_simulation(params_df, iterations=1000):
+    """
+    Simple Monte Carlo simulation:
+    Population growth with noise.
+    """
+    results = []
+    for _, row in params_df.iterrows():
+        base_growth = row["growth_rate"]
+        population = row["initial_population"]
+
+        sims = []
+        for _ in range(iterations):
+            pop = population
+            for _ in range(row["years"]):
+                # stochastic growth with random noise
+                pop = pop * (1 + base_growth + np.random.normal(0, 0.01))
+            sims.append(pop)
+        results.append({
+            "scenario": row["scenario"],
+            "expected_population": np.mean(sims),
+            "std_dev": np.std(sims)
+        })
+    return pd.DataFrame(results)
+
+def train_ai_model(results_df, params_df):
+    """
+    Train regression model to predict population outcome
+    from growth rate & years.
+    """
+    X = params_df[["growth_rate", "years"]].values
+    y = results_df["expected_population"].values
+    model = LinearRegression().fit(X, y)
+
+    # Predictions
+    preds = model.predict(X)
+    results_df["ai_prediction"] = preds
+    return results_df, model
+
+def export_results(results_df, sheet_name="SimulationResults"):
+    """Write results back to the Aura workbook."""
+    with pd.ExcelWriter(AURA_FILE, engine="openpyxl", mode="a", if_sheet_exists="replace") as writer:
+        results_df.to_excel(writer, sheet_name=sheet_name, index=False)
+
+def main():
+    print("Loading Aura data...")
+    params_df = load_data()
+
+    print("Running Monte Carlo simulation...")
+    results_df = run_simulation(params_df)
+
+    print("Training AI model...")
+    results_df, model = train_ai_model(results_df, params_df)
+
+    print("Exporting results back to Aura workbook...")
+    export_results(results_df)
+
+    print("Demo complete! Results in 'SimulationResults' sheet.")
+
+if __name__ == "__main__":
+    main()