add LIME

.gitignore

@@ -0,0 +1,2 @@
+.DS_Store
+venv/

app.py

@@ -0,0 +1,11 @@
+import streamlit as st
+import src.prompt_config as prompt_params
+
+
+def main():
+    st.title("Explainable AI")
+    st.markdown(prompt_params.APP_INTRODUCTION)
+
+
+if __name__ == '__main__':
+    main()

pages/LIME.py

@@ -0,0 +1,136 @@
+# Import Libraries
+import numpy as np
+import pandas as pd
+import matplotlib.pyplot as plt
+import streamlit as st
+import src.prompt_config as prompt_params
+import streamlit.components.v1 as components
+# Models
+import xgboost
+from sklearn.model_selection import train_test_split
+
+# XAI (Explainability)
+import shap
+import lime
+# from anchor import anchor_tabular
+from sklearn.inspection import PartialDependenceDisplay
+
+# 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 define_features():
+    """ Define feature names and categorical mappings. """
+    global feature_names, categorical_features, class_names, categorical_names
+    
+    feature_names = ["Age", "Workclass",
+                     "Education-Num", "Marital Status", "Occupation",
+                     "Relationship", "Race", "Sex", "Capital Gain",
+                     "Capital Loss", "Hours per week", "Country"]
+    
+    categorical_features = ["Workclass", "Marital Status", "Occupation", "Relationship", "Race", "Sex", "Country"]
+    
+    class_names = ['<=50K', '>50K']
+    
+    categorical_names = {
+        1: ['Private', 'Self-emp-not-inc', 'Self-emp-inc', 'Federal-gov', 'Local-gov', 'State-gov', 'Without-pay',
+            'Never-worked'],
+        3: ['Married-civ-spouse', 'Divorced', 'Never-married', 'Separated', 'Widowed', 'Married-spouse-absent',
+            'Married-AF-spouse'],
+        4: ['Tech-support', 'Craft-repair', 'Other-service', 'Sales', 'Exec-managerial', 'Prof-specialty',
+            'Handlers-cleaners',
+            'Machine-op-inspct', 'Adm-clerical', 'Farming-fishing', 'Transport-moving', 'Priv-house-serv',
+            'Protective-serv', 'Armed-Forces'],
+        5: ['Wife', 'Own-child', 'Husband', 'Not-in-family', 'Other-relative', 'Unmarried'],
+        6: ['White', 'Asian-Pac-Islander', 'Amer-Indian-Eskimo', 'Other', 'Black'],
+        7: ['Female', 'Male'],
+        11: ['United-States', 'Cambodia', 'England', 'Puerto-Rico', 'Canada', 'Germany', 'Outlying-US(Guam-USVI-etc)',
+             'India',
+             'Japan', 'Greece', 'South', 'China', 'Cuba', 'Iran', 'Honduras', 'Philippines', 'Italy', 'Poland',
+             'Jamaica', 'Vietnam',
+             'Mexico', 'Portugal', 'Ireland', 'France', 'Dominican-Republic', 'Laos', 'Ecuador', 'Taiwan', 'Haiti',
+             'Columbia', 'Hungary',
+             'Guatemala', 'Nicaragua', 'Scotland', 'Thailand', 'Yugoslavia', 'El-Salvador', 'Trinadad&Tobago', 'Peru',
+             'Hong', 'Holand-Netherlands']
+    }
+
+
+def explain_example(kernel_width, example_idx):
+    """ 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()
+    
+    # Initialize LIME Explainer
+    explainer = lime.lime_tabular.LimeTabularExplainer(
+        X_train.values,
+        class_names=class_names,
+        feature_names=feature_names,
+        categorical_features=categorical_features,
+        categorical_names=categorical_names,
+        kernel_width=kernel_width
+    )
+    
+    # Explain the selected sample
+    exp = explainer.explain_instance(X_test.values[example_idx], global_model.predict_proba, num_features=12)
+
+    # Generate HTML explanation
+    explanation_html = exp.as_html()
+
+    # Display explanation in Streamlit
+    components.html(explanation_html, height=600, scrolling=True)
+
+
+def main():
+    global global_model
+    
+    # Ensure the model is trained only once
+    if global_model is None:
+        train_model()
+    
+    # Streamlit UI Controls
+    lime_kernel_width = st.sidebar.slider(
+        label="Set the `kernel` value:",
+        min_value=0.0,
+        max_value=100.0,
+        value=3.0,  # Default value
+        step=0.1,  # Step size
+        help=prompt_params.LIME_KERNEL_WIDTH_HELP,
+    )
+    
+    example_idx = st.sidebar.slider(
+        label="Select the sample index to explain:",
+        min_value=0,
+        max_value=len(X_test) - 1,  # Ensures the index is within range
+        value=1,  # Default value
+        step=1,  # Step size
+        help=prompt_params.EXAMPLE_BE_EXPLAINED_IDX,
+    )
+    
+    # Explain the selected sample
+    if st.button("Explain Sample"):
+        explain_example(lime_kernel_width, example_idx)
+
+
+if __name__ == '__main__':
+    main()

src/prompt_config.py

@@ -0,0 +1,27 @@
+APP_INTRODUCTION = """
+This application provides explainability for machine learning models using LIME and SHAP.
+It allows users to explore how different features influence model predictions by selecting
+specific samples and visualizing their explanations interactively.
+"""
+
+
+LIME_KERNEL_WIDTH_HELP = """
+The `kernel_width` parameter in LIME controls the size of the neighborhood used to generate perturbations around a sample
+for explanation. It determines how far the generated synthetic data points will be from the original instance.
+
+### How It Works:
+- **Smaller Values (e.g., 1.0 - 3.0):** Focus on very local explanations, meaning LIME will give more weight to points closer to the original sample.
+- **Larger Values (e.g., 10.0 - 100.0):** Expand the neighborhood, leading to more global explanations that consider a broader range of feature values.
+
+### Recommended Settings:
+- **For simple models or small datasets:** Start with `kernel_width = 3.0`.
+- **For complex models or high-dimensional data:** A larger value (e.g., `10.0 - 25.0`) may provide better stability.
+- **For debugging or fine-tuning:** Experiment with different values to see how it impacts feature importance rankings.
+
+⚠️ **Note:** A very large `kernel_width` can make explanations less interpretable, as it may include too many outliers.
+"""
+
+
+EXAMPLE_BE_EXPLAINED_IDX="""
+Select the index of the example you want to explain.
+"""