Update app.py

app.py

@@ -1,30 +1,83 @@
+import pickle 
+import pandas as pd
+import shap
+from shap.plots._force_matplotlib import draw_additive_plot
 import gradio as gr
-import pickle
 import numpy as np
+import matplotlib.pyplot as plt
 
-# Load the trained model
-with open("clf.pkl", "rb") as f:
-    clf = pickle.load(f)
+# Load the XGBoost model
+with open("h22_xgb.pkl", "rb") as f:
+    loaded_model = pickle.load(f)
+
+# Setup SHAP Explainer
+explainer = shap.Explainer(loaded_model)  # PLEASE DO NOT CHANGE THIS.
 
 # Define the prediction function
-def predict(SupportiveGM, Merit, LearningDevelopment, WorkEnvironmente, Engagement, WellBeing, ChainScale):
-    # Convert inputs into a NumPy array
-    input_data = np.array([[SupportiveGM, Merit, LearningDevelopment, WorkEnvironmente, Engagement, WellBeing, ChainScale]])
+def main_func(SupportiveGM, Merit, LearningDevelopment, WorkEnvironment, Engagement, WellBeing):
+    new_row = pd.DataFrame.from_dict({'SupportiveGM': SupportiveGM, 'Merit': Merit,
+              'LearningDevelopment': LearningDevelopment, 'WorkEnvironment': WorkEnvironment, 
+              'Engagement': Engagement, 'WellBeing': WellBeing}, orient='index').transpose()
+    
+    # Predict probabilities
+    prob = loaded_model.predict_proba(new_row)
+    
+    # Compute SHAP values
+    shap_values = explainer(new_row)
+    plot = shap.plots.bar(shap_values[0], max_display=6, order=shap.Explanation.abs, show_data='auto', show=False)
+
+    plt.tight_layout()
+    local_plot = plt.gcf()
+    plt.rcParams['figure.figsize'] = (6, 4)
+    plt.close()
+    
+    return {"Leave": float(prob[0][0]), "Stay": 1 - float(prob[0][0])}, local_plot
+
+# Create the UI
+title = "**Mod 3 Team 5: Employee Turnover Predictor**"
+description1 = """
+This app takes six inputs about employees' satisfaction with different aspects of their work (such as work-life balance, ...) 
+and predicts whether the employee intends to stay with the employer or leave. The outputs include:
+1. The predicted probability of staying or leaving.
+2. A SHAP plot that visualizes how different factors impact the prediction.
+"""
+
+description2 = """
+To use the app, adjust the values of the six employee satisfaction factors and click **Analyze**. ✨ 
+""" 
+
+with gr.Blocks(title=title) as demo:
+    gr.Markdown(f"## {title}")
+    gr.Markdown(description1)
+    gr.Markdown("""---""")
+    gr.Markdown(description2)
+    gr.Markdown("""---""")
+
+    with gr.Row():        
+        with gr.Column():
+            SupportiveGM = gr.Slider(label="Supportive GM Score", minimum=1, maximum=5, value=4, step=0.1)
+            Merit = gr.Slider(label="Merit Score", minimum=1, maximum=5, value=4, step=0.1)
+            LearningDevelopment = gr.Slider(label="Learning & Development Score", minimum=1, maximum=5, value=4, step=0.1)
+            WorkEnvironment = gr.Slider(label="Work Environment Score", minimum=1, maximum=5, value=4, step=0.1)
+            Engagement = gr.Slider(label="Engagement Score", minimum=1, maximum=5, value=4, step=0.1)
+            WellBeing = gr.Slider(label="Well-Being Score", minimum=1, maximum=5, value=4, step=0.1)
+            submit_btn = gr.Button("Analyze")
+        
+        with gr.Column(visible=True, scale=1, min_width=600) as output_col:
+            label = gr.Label(label="Predicted Turnover Probability")
+            local_plot = gr.Plot(label="SHAP Plot:")
+
+            submit_btn.click(
+                main_func,
+                [SupportiveGM, Merit, LearningDevelopment, WorkEnvironment, Engagement, WellBeing],
+                [label, local_plot], api_name="Employee_Turnover"
+            )
     
-    # Make prediction using the model
-    prediction = clf.predict(input_data)
-
-    return f"Predicted Turnover Probability: {prediction[0]}"
-
-# Create the Gradio interface
-iface = gr.Interface(
-    fn=predict,
-    inputs=["number"] * 7,
-    outputs="text",
-    title="Employee Turnover Prediction",
-    api_name="/Employee_Turnover"
-)
-
-if __name__ == "__main__":
-    iface.launch()
+    gr.Markdown("### Click on an example below to see how it works:")
+    gr.Examples([[4, 4, 4, 4, 5, 5], [5, 4, 5, 4, 4, 4]], 
+                [SupportiveGM, Merit, LearningDevelopment, WorkEnvironment, Engagement, WellBeing], 
+                [label, local_plot], main_func, cache_examples=True)
+
+demo.launch()
+