Update app.py

app.py

@@ -2,18 +2,24 @@ import gradio as gr
 import pandas as pd
 import joblib
 import numpy as np
-from typing import Dict, Any, List
+from typing import Dict, Any, List, Tuple
+from reportlab.lib.pagesizes import letter
+from reportlab.pdfgen import canvas
+from reportlab.platypus import SimpleDocTemplate, Paragraph, Spacer
+from reportlab.lib.styles import getSampleStyleSheet
+from reportlab.lib.units import inch
+import io  # Import the io module
 
 # 1. Load the Model and Artifacts
 # Model and artifact paths - adjust as necessary
-model_path = 'your_model.pkl'
+model_path = 'your_model.pkl'  #  Make sure this path is correct
 scaler_path = 'scaler.pkl'  # If you used a scaler
-# encoder_path = 'encoder.pkl' # If you used an encoder
+# encoder_path = 'encoder.pkl'  # If you used an encoder
 
 try:
     model = joblib.load(model_path)
-    scaler = joblib.load(scaler_path) if scaler_path else None # Load scaler
-    # encoder = joblib.load(encoder_path) if encoder_path else None  # Load encoder
+    scaler = joblib.load(scaler_path) if scaler_path else None
+    # encoder = joblib.load(encoder_path) if encoder_path else None
 except Exception as e:
     print(f"Error loading model or artifacts: {e}")
     model = None
@@ -39,10 +45,10 @@ def preprocess_data(data: Dict[str, Any]) -> pd.DataFrame:
             'Quality_Report__c',  # Long Text Area
             'Delay_Days__c',      # Number
             'Incident_Log__c',    # Long Text Area
-            'Vendor__c',          # Lookup (Important:  Send the Salesforce ID, not the name)
+            'Vendor__c',          # Lookup (Important: Send the Salesforce ID, not the name)
             'Work_Details__c',    # Long Text Area
-            'Work_Completion_Date__c', # Date
-            'Actual_Completion_Date__c' # Date
+            'Work_Completion_Date__c',  # Date
+            'Actual_Completion_Date__c'  # Date
         ]
 
         # 1. Check for missing fields
@@ -63,7 +69,7 @@ def preprocess_data(data: Dict[str, Any]) -> pd.DataFrame:
             df['Work_Completion_Year'] = df['Work_Completion_Date__c'].dt.year
             df['Work_Completion_Month'] = df['Work_Completion_Date__c'].dt.month
             df['Work_Completion_Day'] = df['Work_Completion_Date__c'].dt.day
-            df = df.drop(columns=['Work_Completion_Date__c']) #remove original date
+            df = df.drop(columns=['Work_Completion_Date__c'])  # remove original date
 
         if 'Actual_Completion_Date__c' in df.columns:
             df['Actual_Completion_Date__c'] = pd.to_datetime(df['Actual_Completion_Date__c'])
@@ -72,10 +78,10 @@ def preprocess_data(data: Dict[str, Any]) -> pd.DataFrame:
             df['Actual_Completion_Day'] = df['Actual_Completion_Date__c'].dt.day
             df = df.drop(columns=['Actual_Completion_Date__c'])
 
-       # 4.  Text Handling (Example -  for long text areas)
-       # if 'Work_Details__c' in df.columns:
-       #    df['Work_Details__c'] = df['Work_Details__c'].fillna('')
-       #     df['Work_Details_Length'] = df['Work_Details__c'].apply(len) #example feature
+        # 4.  Text Handling (Example -  for long text areas)
+        # if 'Work_Details__c' in df.columns:
+        #    df['Work_Details__c'] = df['Work_Details__c'].fillna('')
+        #     df['Work_Details_Length'] = df['Work_Details__c'].apply(len) #example feature
 
         df = df[expected_fields]
         return df
@@ -87,26 +93,28 @@ def preprocess_data(data: Dict[str, Any]) -> pd.DataFrame:
         return pd.DataFrame()
 
 # 3. Prediction Function
-def predict_vendor_score(*args: Any) -> List[Any]: # Change return type
+def predict_vendor_score(*args: Any) -> Tuple[List[Tuple[str, float]], bytes]:
     """
     Predicts the vendor performance score based on the input data from Vendor_Log__c.
-    Returns a dictionary containing the Subcontractor_Performance_Score__c fields.
+    Returns a list of tuples, where each tuple contains the score name and its value,
+    and the PDF data as bytes.
 
     Args:
         *args: Input values from the Gradio interface, in the order matching
                 the Vendor_Log__c fields.
 
     Returns:
-        List[Any]:  A list of output values.
+        Tuple[List[Tuple[str, float]], bytes]: A tuple containing the score data and the PDF data.
     """
     if model is None:
+        error_message = "Model not loaded.  Check server logs."
         return [
-            'Model not loaded.  Check server logs.',  # Quality Score
-            'Model not loaded.  Check server logs.',  # Timeliness Score
-            'Model not loaded.  Check server logs.',  # Safety Score
-            'Model not loaded.  Check server logs.',  # Communication Score
-            'Model not loaded.  Check server logs.',  # Final Score
-        ]
+            ("Quality Score (%)", 0.0),
+            ("Timeliness Score (%)", 0.0),
+            ("Safety Score (%)", 0.0),
+            ("Communication Score (%)", 0.0),
+            ("Final Score (%)", 0.0),
+        ], generate_pdf(error_message)
 
     # 1. Prepare input data as a dictionary, mapping to Vendor_Log__c fields
     input_data = {
@@ -122,13 +130,14 @@ def predict_vendor_score(*args: Any) -> List[Any]: # Change return type
     # 2. Preprocess the data
     processed_df = preprocess_data(input_data)
     if processed_df.empty:
+        error_message = "Error in input data. Check logs."
         return [
-            'Error in input data. Check logs.', # Quality Score
-            'Error in input data. Check logs.', # Timeliness Score
-            'Error in input data. Check logs.', # Safety Score
-            'Error in input data. Check logs.', # Communication Score
-            'Error in input data. Check logs.'  # Final Score
-        ]
+            ("Quality Score (%)", 0.0),
+            ("Timeliness Score (%)", 0.0),
+            ("Safety Score (%)", 0.0),
+            ("Communication Score (%)", 0.0),
+            ("Final Score (%)", 0.0),
+        ], generate_pdf(error_message)  # Return empty PDF
 
     # 3. Make Prediction
     try:
@@ -139,23 +148,55 @@ def predict_vendor_score(*args: Any) -> List[Any]: # Change return type
         #   This is a *crucial* step where you tell Gradio how to interpret
         #   the numbers coming from your model.
         output_data = [
-            predictions[0] * 100,  # Quality Score (%)
-            predictions[1] * 100,  # Timeliness Score (%)
-            predictions[2] * 100,  # Safety Score (%)
-            predictions[3] * 100,  # Communication Score (%)
-            (predictions[0] + predictions[1] + predictions[2] + predictions[3]) / 4 * 100, # Final Score (%)
+            ("Quality Score (%)", predictions[0] * 100),
+            ("Timeliness Score (%)", predictions[1] * 100),
+            ("Safety Score (%)", predictions[2] * 100),
+            ("Communication Score (%)", predictions[3] * 100),
+            ("Final Score (%)", (predictions[0] + predictions[1] + predictions[2] + predictions[3]) / 4 * 100),
         ]
-        return output_data
+        pdf_data = generate_pdf(output_data)
+        return output_data, pdf_data
 
     except Exception as e:
+        error_message = f"Error during prediction: {e}"
         print(f"Error during prediction: {e}")
         return [
-            f'Error during prediction: {e}', # Quality Score
-            f'Error during prediction: {e}', # Timeliness Score
-            f'Error during prediction: {e}', # Safety Score
-            f'Error during prediction: {e}', # Communication Score
-            f'Error during prediction: {e}'  # Final Score
-        ]
+            ("Quality Score (%)", 0.0),
+            ("Timeliness Score (%)", 0.0),
+            ("Safety Score (%)", 0.0),
+            ("Communication Score (%)", 0.0),
+            ("Final Score (%)", 0.0),
+        ], generate_pdf(error_message)
+
+def generate_pdf(scores: List[Tuple[str, float]]) -> bytes:
+    """Generates a PDF report of the subcontractor performance scores.
+
+    Args:
+        scores (List[Tuple[str, float]]): A list of (score_name, score_value) tuples.
+    Returns:
+        bytes: The PDF data as bytes.
+    """
+    buffer = io.BytesIO()  # Use BytesIO for in-memory PDF generation
+    doc = SimpleDocTemplate(buffer, pagesize=letter)
+    styles = getSampleStyleSheet()
+    Story = []
+
+    # Add a title
+    Story.append(Paragraph("Subcontractor Performance Report", styles['Title']))
+    Story.append(Spacer(1, 0.2 * inch))
+
+    if isinstance(scores, str):  # Check if it is an error message
+        Story.append(Paragraph(scores, styles['Normal']))
+    else:
+        # Add the scores to the PDF
+        for score_name, score_value in scores:
+            Story.append(Paragraph(f"{score_name}: {score_value:.2f}", styles['Normal']))
+            Story.append(Spacer(1, 0.1 * inch))
+
+    doc.build(Story)
+    pdf_data = buffer.getvalue()
+    buffer.close()
+    return pdf_data
 
 # 4. Gradio Interface
 iface = gr.Interface(
@@ -167,21 +208,16 @@ iface = gr.Interface(
         gr.Textbox(label="Vendor ID (Text)"),  # Send ID
         gr.Textbox(label="Work Details (Long Text)", type="text"),
         gr.DateTime(label="Work Completion Date", type="datetime"),  # Corrected to "datetime"
-        gr.DateTime(label="Actual Completion Date", type="datetime"), # Corrected to "datetime"
+        gr.DateTime(label="Actual Completion Date", type="datetime"),  # Corrected to "datetime"
     ],
-    outputs=[ # Changed to a list
-        gr.HighlightedText(label="Quality Score (%)"),
-        gr.HighlightedText(label="Timeliness Score (%)"),
-        gr.HighlightedText(label="Safety Score (%)"),
-        gr.HighlightedText(label="Communication Score (%)"),
-        gr.HighlightedText(label="Final Score (%)"),
-        # "Download_PDF": gr.File(label="Download PDF"), # removed file
+    outputs=[  # Changed to a list
+        gr.HighlightedText(label="Performance Scores"),
+        gr.File(label="Download PDF Report"),
     ],
     title="Subcontractor Performance Score Calculator",
-    description="Enter Vendor Log details to calculate performance scores.",
+    description="Enter Vendor Log details to calculate performance scores and generate a PDF report.",
 )
 
 # 5. Launch the Interface
 if __name__ == "__main__":
     iface.launch()
-