Update app.py

app.py

@@ -1,20 +1,26 @@
 import pickle
 import pandas as pd
 import shap
-from shap.plots._force_matplotlib import draw_additive_plot
 import gradio as gr
 import numpy as np
 import matplotlib.pyplot as plt
-
-# load the model from disk
-loaded_model = pickle.load(open("salar_xgb_team.pkl", 'rb'))
-
+ 
+# Load the model
+loaded_model = pickle.load(open("salar_xgb_team.pkl", "rb"))
+ 
 # Setup SHAP
-explainer = shap.Explainer(loaded_model) # PLEASE DO NOT CHANGE THIS.
-
-# Main prediction function
+explainer = shap.Explainer(loaded_model)  # DO NOT CHANGE THIS
+ 
+# Main function
 def main_func(age, education_num, sex, capital_gain, capital_loss, hours_per_week):
+    # Input validation
+    if age < 18 or age > 100 or education_num < 1 or hours_per_week < 1 or hours_per_week > 100:
+        return {"≤50K": 0.0, ">50K": 0.0}, None, "❌ Invalid inputs. Please check your entries."
+ 
+    # Process categorical
     sex_binary = 0 if sex == "Male" else 1
+ 
+    # Create input row
     new_row = pd.DataFrame({
         'age': [age],
         'education-num': [education_num],
@@ -24,21 +30,32 @@ def main_func(age, education_num, sex, capital_gain, capital_loss, hours_per_wee
         'hours-per-week': [hours_per_week]
     })
  
+    # Predict
     prob = loaded_model.predict_proba(new_row)
     shap_values = explainer(new_row)
  
-    # SHAP bar plot
-    plot = shap.plots.bar(shap_values[0], max_display=6, show=False)
+    # SHAP plot
+    plt.figure(figsize=(8, 4))
+    shap.plots.bar(shap_values[0], max_display=6, show=False)
     plt.tight_layout()
     local_plot = plt.gcf()
     plt.close()
  
-    return {"≤50K": float(prob[0][0]), ">50K": float(prob[0][1])}, local_plot
-
-# Gradio UI
-title = "**Salary Predictor & SHAP Explainer** 💼"
-description1 = "This app uses demographic and financial info to predict if someone earns over 50K annually."
-description2 = "Adjust the inputs below and click Analyze to see the prediction and SHAP feature importance."
+    # Predicted class and confidence
+    pred_class = ">50K" if prob[0][1] > 0.5 else "≤50K"
+    confidence = round(prob[0][1] if pred_class == ">50K" else prob[0][0], 2)
+ 
+    interpretation = f"💼 Prediction: **{pred_class}**\nConfidence: {confidence * 100:.2f}%"
+ 
+    return {
+        "≤50K": round(prob[0][0], 2),
+        ">50K": round(prob[0][1], 2)
+    }, local_plot, interpretation
+ 
+# ----------- Gradio UI -----------
+title = "**Salary Predictor & SHAP Explainer** 💰"
+description1 = "This app uses demographic and financial info to predict whether someone earns more than $50K annually."
+description2 = "Adjust the sliders and inputs below, then click **Analyze** to see the prediction and SHAP explanation."
  
 with gr.Blocks(title=title) as demo:
     gr.Markdown(f"## {title}")
@@ -47,23 +64,26 @@ with gr.Blocks(title=title) as demo:
     gr.Markdown(description2)
     gr.Markdown("---")
  
-    age = gr.Slider(label="Age", minimum=18, maximum=70, value=35)
-    education_num = gr.Slider(label="Education Number", minimum=1, maximum=16, value=10)
-    sex = gr.Radio(["Male", "Female"], label="Sex")
-    capital_gain = gr.Number(label="Capital Gain", value=0)
-    capital_loss = gr.Number(label="Capital Loss", value=0)
-    hours_per_week = gr.Slider(label="Hours per Week", minimum=1, maximum=100, value=40)
+    with gr.Row():
+        with gr.Column(scale=1):
+            age = gr.Slider(label="Age (Years)", minimum=18, maximum=100, value=35, info="Enter age between 18 and 100")
+            education_num = gr.Slider(label="Education Level (Numerical)", minimum=1, maximum=16, value=10, info="E.g., 1 = Preschool, 16 = Doctorate")
+            sex = gr.Radio(["Male", "Female"], label="Sex")
+            capital_gain = gr.Number(label="Capital Gain", value=0)
+            capital_loss = gr.Number(label="Capital Loss", value=0)
+            hours_per_week = gr.Slider(label="Hours Worked per Week", minimum=1, maximum=100, value=40)
  
-    submit_btn = gr.Button("Analyze")
+            submit_btn = gr.Button("🔍 Analyze")
  
-    with gr.Column(visible=True) as output_col:
-        label = gr.Label(label="Predicted Probability")
-        local_plot = gr.Plot(label="SHAP Feature Importance")
+        with gr.Column(scale=1):
+            label = gr.Label(label="Predicted Probabilities")
+            local_plot = gr.Plot(label="SHAP Feature Importance")
+            result_text = gr.Textbox(label="Prediction Summary", lines=2)
  
     submit_btn.click(
         main_func,
         [age, education_num, sex, capital_gain, capital_loss, hours_per_week],
-        [label, local_plot],
+        [label, local_plot, result_text],
         api_name="Salary_Predictor"
     )
  
@@ -75,9 +95,12 @@ with gr.Blocks(title=title) as demo:
             [35, 9, "Male", 0, 1902, 40]
         ],
         inputs=[age, education_num, sex, capital_gain, capital_loss, hours_per_week],
-        outputs=[label, local_plot],
+        outputs=[label, local_plot, result_text],
         fn=main_func,
         cache_examples=True
     )
  
+    gr.Markdown("---")
+    gr.Markdown("Built with ❤️ by Tania Ramesh for the 2025 AI Applications Project.")
+ 
 demo.launch()