add ale

pages/ALE.py

@@ -0,0 +1,75 @@
+# Import Libraries
+import matplotlib.pyplot as plt
+import streamlit as st
+import src.prompt_config as prompt_params
+# Models
+import xgboost
+from sklearn.model_selection import train_test_split
+from alepython import ale_plot
+# XAI (Explainability)
+import shap
+
+# Global Variables to Store Model & Data
+global_model = None
+X_train, X_test, y_train, y_test = None, None, None, None
+
+
+def train_model():
+    """ Train the XGBoost model only once and store it globally. """
+    global global_model, X_train, X_test, y_train, y_test
+    
+    if global_model is None:
+        # Load Data from SHAP library
+        X, y = shap.datasets.adult()
+        
+        # Split data
+        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.5, random_state=42)
+        
+        # Train XGBoost model
+        global_model = xgboost.XGBClassifier()
+        global_model.fit(X_train, y_train)
+        
+        print("XGBoost Model training completed!")
+
+def explain_example():
+    """ Explain a given sample without retraining the model. """
+    global global_model, X_train, X_test, y_train, y_test
+    
+    if global_model is None:
+        train_model()
+    
+    fig, ax = plt.subplots(figsize=(10, 5))
+    st.write("1D Main Effect ALE Plot")
+    ale_plot(
+        global_model,
+        X_test,
+        "Age",
+        bins=5,
+        monte_carlo=True,
+        monte_carlo_rep=30,
+        monte_carlo_ratio=0.5,
+    )
+    
+    st.pyplot(fig)
+
+    fig, ax = plt.subplots(figsize=(10, 5))
+    st.write("2D Second-Order ALE Plot")
+    ale_plot(global_model, X_test, X_train.columns[:2], bins=10)
+    st.pyplot(fig)
+
+def main():
+    global global_model
+    
+    # Ensure the model is trained only once
+    if global_model is None:
+        train_model()
+    
+    st.title("ALE (Accumulated Local Effects)")
+    st.write(prompt_params.ALE_INTRODUCTION)
+    # Explain the selected sample
+    if st.button("Explain Sample"):
+        explain_example()
+
+
+if __name__ == '__main__':
+    main()

pages/ICE_and_PDP.py

@@ -52,6 +52,7 @@ def main():
     # Define feature names
     feature_names = ["Age", "Workclass", "Education-Num", "Marital Status", "Occupation",
                      "Relationship", "Race", "Sex", "Capital Gain", "Capital Loss", "Hours per week", "Country"]
+    print(X_test.columns)  # Check the actual feature names
 
     selected_feature = st.sidebar.selectbox("Select a feature for PDP/ICE analysis:", feature_names)
     

src/prompt_config.py

@@ -105,4 +105,16 @@ When `kind` is selected:
 - **both**: Displays both ICE and PDP.
 - **individual**: Displays only ICE.
 - **average**: Displays only PDP.
-"""
+"""
+
+ALE_INTRODUCTION = """
+ALE (Accumulated Local Effects) is an interpretable machine learning technique that quantifies the impact of a feature on model predictions while accounting for feature dependencies.
+
+The process of ALE includes the following steps:
+1. **Bin the Feature**: Divide the feature into intervals (bins) to segment the data.
+2. **Compute Local Effects**: Measure the change in predictions when the feature moves from the lower to the upper edge of each bin.
+3. **Accumulate Effects**: Sum the local effects sequentially across bins to observe the overall influence of the feature.
+4. **Centering**: Normalize the accumulated effects by subtracting the mean to focus on relative deviations from the average prediction.
+
+By using ALE, **interpretability** improves by capturing localized effects while mitigating bias from correlated features, making model explanations more reliable.
+"""