Upload 3 files

README.md

@@ -4,7 +4,7 @@ emoji: 🕳️
 colorFrom: yellow
 colorTo: red
 sdk: gradio
-sdk_version: 4.42.0
+sdk_version: 5.12.0
 app_file: app.py
 pinned: false
 license: mit
@@ -59,7 +59,7 @@ The model was trained on a synthetically generated dataset of `10,000` samples d
 
 All features are min-max scaled `[0,1]`.
 
-### Training Procedure 
+### Training Procedure
 
 - **Algorithm:** XGBoost
 - **Objective:** `reg:squarederror`
@@ -69,7 +69,7 @@ All features are min-max scaled `[0,1]`.
 
 ## 📊 Performance & Interpretability
 
-### Model Metrics 
+### Model Metrics
 The model demonstrates high precision in predicting the severity score $S$, which controls civic resource allocation.
 
 | Metric | Value | Interpretation |
@@ -95,8 +95,6 @@ The bar chart below shows the mean absolute SHAP value, identifying which featur
 ![SHAP Bar Plot](shap_bar_plot.png)
 
 #### Detailed Impact (Beeswarm)
-The summary plot shows how high vs. low values of a feature affect out outcome. For example, high values of **C (Centrality)** push the score significantly higher.
+The summary plot shows how high vs. low values of a feature affect the outcome. For example, high values of **C (Centrality)** push the score significantly higher.
 
 ![SHAP Dot Plot](shap_dot_plot.png)
-
-## Training Details

app.py

@@ -1,22 +1,17 @@
 import sys
 
-# SHIM FOR PYTHON 3.13 TO FIX GRADIO/PYDUB "audioop" ERROR
-# This must happen before anything else is imported.
+# SHIM FOR PYTHON 3.13: fake audioop module before any imports
 try:
     import audioop
 except ImportError:
     import types
-    # Fake the audioop module to satisfy pydub's import check
-    module = types.ModuleType("audioop")
-    sys.modules["audioop"] = module
-    print("applied audioop shim for Python 3.13 compatibility")
+    sys.modules["audioop"] = types.ModuleType("audioop")
 
 import gradio as gr
 import xgboost as xgb
 import joblib
 import json
 import numpy as np
-import os
 
 # --- Load Assets ---
 MODEL_PATH = "severity_model.json"
@@ -39,22 +34,18 @@ def get_label(score):
     return "High 🔴"
 
 def predict(*args):
-    # Map arguments to feature list
     input_dict = dict(zip(feature_names, args))
     row = np.array([[input_dict[f] for f in feature_names]], dtype=np.float32)
-    
-    # Scale and predict
     scaled_row = scaler.transform(row)
     prediction = float(model.predict(scaled_row)[0])
     score = max(0, min(1, prediction))
-    
     return round(score, 4), get_label(score)
 
-# --- UI Setup ---
+# --- UI ---
 with gr.Blocks(theme=gr.themes.Soft()) as demo:
     gr.Markdown("# 🕳️ Pothole Severity Predictor (Civic AI)")
     gr.Markdown("Adjust the sliders below to simulate pothole features and predict repair priority.")
-    
+
     with gr.Row():
         with gr.Column():
             a = gr.Slider(0, 1, value=0.1, label="Area Ratio (A)", info="Size of pothole")
@@ -68,9 +59,9 @@ with gr.Blocks(theme=gr.themes.Soft()) as demo:
             p = gr.Slider(0, 1, value=0.1, label="Critical Infra (P)", info="Proximity to hospitals/schools")
             f = gr.Slider(0, 1, value=0.1, label="Recurrence (F)", info="Historical failure")
             x = gr.Slider(0, 1, value=0.0, label="Reopen Count (X)", info="Failed repairs")
-    
+
     btn = gr.Button("Calculate Severity Score", variant="primary")
-    
+
     with gr.Row():
         out_score = gr.Number(label="Severity Score (0-1)")
         out_label = gr.Textbox(label="Priority Level")

requirements.txt

@@ -3,5 +3,3 @@ pandas
 scikit-learn
 xgboost
 joblib
-gradio
-huggingface_hub