Update app.py

app.py

@@ -1,59 +1,79 @@
+#New code post Reza meeting - takes out the All option
+
 import pickle
 import pandas as pd
 import shap
 import gradio as gr
+import numpy as np
 import matplotlib.pyplot as plt
 
 # Load the model from disk
-loaded_model = pickle.load(open("h37_xgb.pkl", 'rb'))
+loaded_model = pickle.load(open("h42_xgb.pkl", 'rb'))
 
 # Setup SHAP
 explainer = shap.Explainer(loaded_model)  # PLEASE DO NOT CHANGE THIS.
 
-# Map the Generation selection to numeric values (5 for Millennials, 6 for Gen Z)
-generation_mapping = {
-    "Millennials": 5,
-    "Gen Z": 6
-}
+# Mapping categorical values to numerical
+generation_mapping = {"Millennials": 5, "Gen Z": 6}
+chain_scale_mapping = {"Upper Midscale": 3, "Upscale": 4, "Upper Upscale": 5}
+region_mapping = {"Northeast": 1, "Midwest": 2, "South": 3, "West": 4}
+recommend_mapping = {"1": 1, "2": 2, "3": 3, "4": 4, "5": 5}
 
-def process_generation(selected_value):
-    return f"You selected: {selected_value}"
+generation_choices = list(generation_mapping.keys())
+chain_scale_choices = list(chain_scale_mapping.keys())
+region_choices = list(region_mapping.keys())
+recommend_choices = list(recommend_mapping.keys())
 
-# Create the main function for server
-def main_func(Generation, Tenure, Engage2, WorkEnv3, GM3, WellBeing, SupportiveGM, 
-              ManagementLevel=2, RecommendToOthers=4, ChainScale=3, HotelRegion=2):
-    
-    # Convert selected Generation string to the corresponding numeric value
-    Generation_numeric = generation_mapping[Generation]
-    
-    # Create the dataframe with the mapped generation value
-    new_row = pd.DataFrame.from_dict({'Generation': Generation_numeric, 'Tenure': Tenure, 'ManagementLevel': ManagementLevel, 
-                                      'Engage2': Engage2, 'RecommendToOthers': RecommendToOthers, 'WorkEnv3': WorkEnv3,
-                                      'GM3': GM3, 'ChainScale': ChainScale, 'WellBeing': WellBeing, 'SupportiveGM': SupportiveGM,
-                                      'HotelRegion': HotelRegion}, orient='index').transpose()
-    
+# Define main function for prediction
+def main_func(Generation, Engage2, RecommendToOthers, Voice, Merit, Workload, WellBeing, SupportiveGM, ChainScale, HotelRegion, ManagementLevel=2):
+    # Map inputs to numerical values
+    gen = generation_mapping[Generation]
+    cs = chain_scale_mapping[ChainScale]
+    reg = region_mapping[HotelRegion]
+    recommend = recommend_mapping[RecommendToOthers]
+
+    # Create DataFrame for model prediction
+    new_row = pd.DataFrame.from_dict({
+        'Generation': gen, 
+        'ManagementLevel': ManagementLevel, 
+        'Engage2': Engage2,
+        'RecommendToOthers': recommend, 
+        'Voice': Voice,
+        'Merit': Merit, 
+        'Workload': Workload, 
+        'ChainScale': cs, 
+        'WellBeing': WellBeing, 
+        'SupportiveGM': SupportiveGM, 
+        'HotelRegion': reg
+    }, orient = 'index').transpose()
+
+    # Predict probabilities
     prob = loaded_model.predict_proba(new_row)
-    shap_values = explainer(new_row)
     
+    # Compute SHAP values
+    shap_values = explainer(new_row)
+    selected_features = ["Engage2", "RecommendToOthers", "Voice", "Merit", "Workload", "WellBeing", "SupportiveGM"]
+    shap_values_filtered = shap_values[:, selected_features]
+
     # Generate SHAP plot
-    plot = shap.plots.bar(shap_values[0], max_display=6, order=shap.Explanation.abs, show_data='auto', show=False)
-    
+    shap.plots.bar(shap_values_filtered[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.rcParams['figure.figsize'] = (6, 4)
     plt.close()
-    
-    return {"Leave": float(prob[0][0]), "Stay": 1 - float(prob[0][0])}, local_plot
+
+    return {"Leave": float(prob[0][0]), "Stay": float(prob[0][0])}, local_plot
 
 # Create the UI
 title = "**Employee Turnover Predictor & Interpreter** 🪐"
 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. There are two outputs from the app: 1- the predicted probability of stay or leave, 2- Shapley's force-plot which visualizes the extent to which each factor impacts the stay/leave prediction.   
+This app takes six inputs about employees' satisfaction with different aspects of their work and predicts whether the employee intends to stay with the employer or leave.  
+It provides:
+1. The predicted probability of staying or leaving.  
+2. A SHAP interpretation plot showing feature importance.
 """
 
-description2 = """
-To use the app, click on one of the examples, or adjust the values of the six employee satisfaction factors, and click on Analyze. ✨ 
-""" 
+description2 = """Adjust the values of different employee satisfaction factors and click "Analyze" to get a prediction."""
 
 with gr.Blocks(title=title) as demo:
     gr.Markdown(f"## {title}")
@@ -61,31 +81,31 @@ with gr.Blocks(title=title) as demo:
     gr.Markdown("""---""")
     gr.Markdown(description2)
     gr.Markdown("""---""")
+    
     with gr.Row():        
         with gr.Column():
-            # Dropdown to select Generation (Millennials or Gen Z)
-            Generation = gr.Dropdown(label="Select Generation", choices=["Millennials", "Gen Z"], value="Millennials")
-            Engage2 = gr.Slider(label="Engage2", minimum=1, maximum=5, value=4, step=.1)
-            Tenure = gr.Slider(label="Tenure", minimum=1, maximum=5, value=4, step=.1)
-            WorkEnv3 = gr.Slider(label="WorkEnv3", minimum=1, maximum=5, value=4, step=.1)
-            GM3 = gr.Slider(label="GM3", minimum=1, maximum=5, value=4, step=.1)
-            WellBeing = gr.Slider(label="Well Being", minimum=1, maximum=5, value=4, step=.1)
-            SupportiveGM = gr.Slider(label="SupportiveGM", minimum=1, maximum=5, value=4, step=.1)
+            Generation = gr.Dropdown(label="Select Generation 🧍🏽", choices=generation_choices, value="Millennials")
+            ChainScale = gr.Dropdown(label="Select Chain Scale 🧳", choices=chain_scale_choices, value="Upscale")
+            HotelRegion = gr.Dropdown(label="Select Hotel Region 🏨", choices=region_choices, value="Midwest")
+            RecommendToOthers = gr.Dropdown(label="Recommend to Others", choices=recommend_choices, value="4")
+            
+            Engage2 = gr.Slider(label="Engagement (Engage2)", minimum=1, maximum=5, value=4, step=0.1)
+            Voice = gr.Slider(label="Voice", minimum=1, maximum=5, value=4, step=0.1)
+            Merit = gr.Slider(label="Merit", minimum=1, maximum=5, value=4, step=0.1)
+            Workload = gr.Slider(label="Workload", minimum=1, maximum=5, value=4, step=0.1)
+            WellBeing = gr.Slider(label="Well-being", minimum=1, maximum=5, value=4, step=0.1)
+            SupportiveGM = gr.Slider(label="Supportive GM", 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 Label")
-            local_plot = gr.Plot(label='Shap:')
+            local_plot = gr.Plot(label="SHAP Impact")
         
-            # When the button is clicked, process the inputs
             submit_btn.click(
                 main_func,
-                [Generation, Engage2, Tenure, WorkEnv3, GM3, WellBeing, SupportiveGM],
+                [Generation, Engage2, RecommendToOthers, Voice, Merit, Workload, WellBeing, SupportiveGM, ChainScale, HotelRegion],
                 [label, local_plot], api_name="Employee_Turnover"
             )
-    
-    gr.Markdown("### Click on any of the examples below to see how it works:")
-    gr.Examples([["Gen Z", 4, 4, 4, 5, 5], ["Millennials", 4, 5, 4, 4, 4]], 
-                [Generation, Engage2, Tenure, WorkEnv3, GM3, WellBeing, SupportiveGM], 
-                [label, local_plot], main_func, cache_examples=True)
 
-demo.launch( )
+demo.launch( )