Update app.py

app.py

@@ -13,7 +13,7 @@ import tempfile
 # Configure logging to match the log format
 logging.basicConfig(level=logging.INFO, format='%(asctime)s,%(msecs)03d - %(levelname)s - %(message)s')
 
-# CSS styling for the Gradio interface with a dark theme and high-contrast text
+# CSS styling for the Gradio interface with a dark theme and blue button
 css = """
 @import url('https://fonts.googleapis.com/css2?family=Roboto:wght@400;500;700&display=swap');
 @import url('https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css');
@@ -34,7 +34,7 @@ h1 {
 }
 
 .gr-button {
-    background-color: #2DD4BF;
+    background-color: #3B82F6;
     color: #1F2937;
     border: none;
     border-radius: 8px;
@@ -44,7 +44,7 @@ h1 {
 }
 
 .gr-button:hover {
-    background-color: #26A69A;
+    background-color: #2563EB;
 }
 
 .dashboard-container {
@@ -226,21 +226,24 @@ def validate_csv(df):
     Validate that the CSV has the required columns.
     Returns True if valid, False otherwise with an error message.
     """
-    required_columns = ['equipment', 'usage_count', 'status', 'amc_expiry']
+    required_columns = ['device_id', 'usage_hours', 'amc_date', 'status']
     missing_columns = [col for col in required_columns if col not in df.columns]
     if missing_columns:
         return False, f"Missing required columns: {', '.join(missing_columns)}"
     # Validate data types
     try:
-        df['usage_count'] = pd.to_numeric(df['usage_count'], errors='raise')
-        df['amc_expiry'] = pd.to_datetime(df['amc_expiry'], errors='raise')
+        df['usage_hours'] = pd.to_numeric(df['usage_hours'], errors='raise')
+        df['amc_date'] = pd.to_datetime(df['amc_date'], errors='raise')
+        # Handle 'downtime' if present
+        if 'downtime' in df.columns:
+            df['downtime'] = pd.to_numeric(df['downtime'], errors='raise')
     except Exception as e:
         return False, f"Invalid data types: {str(e)}"
     return True, ""
 
 def generate_device_cards(df, anomaly_df):
     """
-    Generate HTML for device cards showing health, usage count, and status.
+    Generate HTML for device cards showing health, usage hours, downtime, status, log type, and timestamp.
     Returns an HTML string.
     """
     if anomaly_df is not None:
@@ -249,15 +252,21 @@ def generate_device_cards(df, anomaly_df):
         df['anomaly'] = "Unknown"
 
     html = []
-    for equipment in df['equipment'].unique():
-        device_data = df[df['equipment'] == equipment].iloc[-1]  # Latest record
+    for device_id in df['equipment'].unique():
+        device_data = df[df['equipment'] == device_id].iloc[-1]  # Latest record
         anomaly_class = "anomaly-unusual" if device_data['anomaly'] == "Unusual" else "anomaly-normal"
+        downtime = device_data.get('downtime', 'N/A')
+        log_type = device_data.get('log_type', 'N/A')
+        timestamp = device_data.get('timestamp', 'N/A')
         html.append(f"""
         <div class="card device-card">
-            <h2><i class="fas fa-microchip"></i> {equipment}</h2>
+            <h2><i class="fas fa-microchip"></i> {device_id}</h2>
             <p><strong>Status:</strong> {device_data['status']}</p>
-            <p><strong>Usage Count:</strong> {device_data['usage_count']}</p>
+            <p><strong>Usage Hours:</strong> {device_data['usage_count']}</p>
+            <p><strong>Downtime (hrs):</strong> {downtime}</p>
             <p><strong>Activity:</strong> <span class="anomaly-badge {anomaly_class}">{device_data['anomaly']}</span></p>
+            <p><strong>Log Type:</strong> {log_type}</p>
+            <p><strong>Last Log:</strong> {timestamp}</p>
             <p><strong>AMC Expiry:</strong> {device_data['amc_expiry'].strftime('%Y-%m-%d')}</p>
         </div>
         """)
@@ -265,7 +274,7 @@ def generate_device_cards(df, anomaly_df):
 
 def generate_summary(combined_df, anomaly_df, amc_df, plot_path, pdf_path):
     """
-    Generate a detailed and easy-to-understand summary of the processing results.
+    Generate a detailed and easy-to-understand summary of the processing results, including downtime.
     Returns a markdown string for display in the Gradio interface.
     """
     summary = []
@@ -274,7 +283,9 @@ def generate_summary(combined_df, anomaly_df, amc_df, plot_path, pdf_path):
     summary.append("## Overview")
     total_records = len(combined_df)
     unique_devices = combined_df['equipment'].unique()
+    total_downtime = combined_df['downtime'].sum() if 'downtime' in combined_df.columns else 0
     summary.append(f"We processed **{total_records} log entries** for **{len(unique_devices)} devices** ({', '.join(unique_devices)}).")
+    summary.append(f"Total downtime recorded: **{total_downtime} hours**.")
     summary.append("This dashboard provides real-time insights into device health, usage patterns, and maintenance needs.\n")
 
     # Downtime Insights (Anomalies)
@@ -283,9 +294,10 @@ def generate_summary(combined_df, anomaly_df, amc_df, plot_path, pdf_path):
         num_anomalies = sum(anomaly_df['anomaly'] == -1)
         if num_anomalies > 0:
             summary.append(f"**{num_anomalies} potential downtime risks** detected:")
-            anomaly_records = anomaly_df[anomaly_df['anomaly'] == -1][['equipment', 'usage_count', 'status']]
+            anomaly_records = anomaly_df[anomaly_df['anomaly'] == -1][['equipment', 'usage_count', 'status', 'downtime']]
             for _, row in anomaly_records.iterrows():
-                summary.append(f"- **{row['equipment']}** (Usage: {row['usage_count']}, Status: {row['status']}) - Indicates possible overuse or underuse.")
+                downtime = row['downtime'] if 'downtime' in row else 'N/A'
+                summary.append(f"- **{row['equipment']}** (Usage: {row['usage_count']}, Status: {row['status']}, Downtime: {downtime} hrs) - Indicates possible overuse or underuse.")
         else:
             summary.append("No potential downtime risks detected. All devices are operating within expected patterns.")
     else:
@@ -321,8 +333,8 @@ def generate_flowchart_html():
     """
     steps = [
         ("Upload CSV File(s)", "User uploads log files in CSV format."),
-        ("Validate Data", "Checks for required columns (equipment, usage_count, status, amc_expiry) and correct data types."),
-        ("Generate Usage Chart", "Creates a bar chart showing usage counts by device and status (e.g., Active, Inactive)."),
+        ("Validate Data", "Checks for required columns (device_id, usage_hours, amc_date, status) and correct data types."),
+        ("Generate Usage Chart", "Creates a bar chart showing usage hours by device and status (e.g., ok, warning)."),
         ("Detect Downtime Risks", "Uses Local Outlier Factor to identify devices with unusual usage patterns (e.g., too high or too low)."),
         ("Check Maintenance Dates", "Identifies devices with AMC expiries within 7 days from 2025-06-05."),
         ("Create PDF Report", "Generates a detailed PDF with data tables, insights, and this flowchart.")
@@ -360,6 +372,12 @@ def process_files(uploaded_files):
         try:
             df = pd.read_csv(file.name)
             logging.info(f"Loaded {len(df)} records from {file.name}")
+            # Rename columns to match expected names
+            df = df.rename(columns={
+                'device_id': 'equipment',
+                'usage_hours': 'usage_count',
+                'amc_date': 'amc_expiry'
+            })
             # Validate CSV structure
             is_valid, error_msg = validate_csv(df)
             if not is_valid:
@@ -437,7 +455,7 @@ def generate_usage_plot(df):
     try:
         plt.figure(figsize=(12, 6))
         # Define colors for statuses (adjusted for dark theme visibility)
-        status_colors = {'Active': '#2DD4BF', 'Inactive': '#F87171', 'Down': '#FBBF24', 'Online': '#10B981'}
+        status_colors = {'ok': '#2DD4BF', 'warning': '#F87171', 'normal': '#10B981', 'down': '#FBBF24'}
         for status in df['status'].unique():
             subset = df[df['status'] == status]
             plt.bar(
@@ -447,7 +465,7 @@ def generate_usage_plot(df):
                 color=status_colors.get(status, '#6B7280')
             )
         plt.xlabel("Equipment (Status)", fontsize=12, color='#D1D5DB')
-        plt.ylabel("Usage Count", fontsize=12, color='#D1D5DB')
+        plt.ylabel("Usage Hours", fontsize=12, color='#D1D5DB')
         plt.title("Device Usage Overview", fontsize=14, color='#FFFFFF')
         plt.legend(title="Status")
         plt.xticks(rotation=45, ha='right', color='#D1D5DB')
@@ -541,13 +559,16 @@ def generate_pdf_report(original_df, anomaly_df, amc_df):
             c.drawString(50, y, f"Total Records: {len(original_df)}")
             y -= 20
             c.drawString(50, y, f"Unique Devices: {', '.join(original_df['equipment'].unique())}")
+            y -= 20
+            total_downtime = original_df['downtime'].sum() if 'downtime' in original_df.columns else 0
+            c.drawString(50, y, f"Total Downtime: {total_downtime} hours")
             y -= 40
 
             # Device Log Details
             y = draw_section_title("Device Log Details", y)
             c.setFont("Helvetica-Bold", 10)
-            headers = ["Equipment", "Usage Count", "Status", "AMC Expiry", "Activity"]
-            x_positions = [50, 150, 250, 350, 450]
+            headers = ["Equipment", "Timestamp", "Usage Hours", "Downtime (hrs)", "Status", "Log Type", "AMC Expiry", "Activity"]
+            x_positions = [50, 110, 190, 260, 320, 370, 430, 490]
             for i, header in enumerate(headers):
                 c.drawString(x_positions[i], y, header)
             c.line(50, y - 5, width - 50, y - 5)
@@ -559,10 +580,13 @@ def generate_pdf_report(original_df, anomaly_df, amc_df):
                 output_df['anomaly'] = anomaly_df['anomaly'].map({1: "Normal", -1: "Unusual"})
             for _, row in output_df.iterrows():
                 c.drawString(50, y, str(row['equipment']))
-                c.drawString(150, y, str(row['usage_count']))
-                c.drawString(250, y, str(row['status']))
-                c.drawString(350, y, str(row['amc_expiry'].strftime('%Y-%m-%d')))
-                c.drawString(450, y, str(row['anomaly']))
+                c.drawString(110, y, str(row.get('timestamp', 'N/A')))
+                c.drawString(190, y, str(row['usage_count']))
+                c.drawString(260, y, str(row.get('downtime', 'N/A')))
+                c.drawString(320, y, str(row['status']))
+                c.drawString(370, y, str(row.get('log_type', 'N/A')))
+                c.drawString(430, y, str(row['amc_expiry'].strftime('%Y-%m-%d')))
+                c.drawString(490, y, str(row['anomaly']))
                 y -= 20
                 if y < 50:
                     c.showPage()
@@ -578,12 +602,13 @@ def generate_pdf_report(original_df, anomaly_df, amc_df):
                 c.drawString(50, y, f"Potential Downtime Risks Detected: {num_anomalies}")
                 y -= 20
                 if num_anomalies > 0:
-                    anomaly_records = anomaly_df[anomaly_df['anomaly'] == -1][['equipment', 'usage_count', 'status']]
+                    anomaly_records = anomaly_df[anomaly_df['anomaly'] == -1][['equipment', 'usage_count', 'status', 'downtime']]
                     c.drawString(50, y, "Details:")
                     y -= 20
                     c.setFont("Helvetica-Oblique", 10)
                     for _, row in anomaly_records.iterrows():
-                        c.drawString(50, y, f"{row['equipment']}: Usage Count = {row['usage_count']}, Status = {row['status']}")
+                        downtime = row['downtime'] if 'downtime' in row else 'N/A'
+                        c.drawString(50, y, f"{row['equipment']}: Usage Hours = {row['usage_count']}, Status = {row['status']}, Downtime = {downtime} hrs")
                         y -= 20
                         c.drawString(70, y, "Note: This device’s usage is significantly higher or lower than others, which may indicate overuse or underuse.")
                         y -= 20
@@ -642,8 +667,8 @@ def generate_pdf_report(original_df, anomaly_df, amc_df):
             c.setFont("Helvetica", 10)
             flowchart = [
                 ("1. Upload CSV File(s)", "User uploads log files in CSV format containing device usage data."),
-                ("2. Validate Data", "Ensures all required columns (equipment, usage_count, status, amc_expiry) are present and data types are correct (e.g., usage_count as numeric, amc_expiry as date)."),
-                ("3. Generate Usage Chart", "Creates a bar chart showing usage counts by device and status (e.g., Active, Inactive) to visualize usage patterns."),
+                ("2. Validate Data", "Ensures all required columns (device_id, usage_hours, amc_date, status) are present and data types are correct (e.g., usage_hours as numeric, amc_date as date)."),
+                ("3. Generate Usage Chart", "Creates a bar chart showing usage hours by device and status (e.g., ok, warning) to visualize usage patterns."),
                 ("4. Detect Downtime Risks", "Uses Local Outlier Factor (LOF) algorithm to identify devices with unusual usage patterns by comparing local density of usage counts (contamination=0.1, n_neighbors=5)."),
                 ("5. Check Maintenance Dates", "Identifies devices with AMC expiries within 7 days from 2025-06-05, calculating days left and urgency (urgent if ≤3 days)."),
                 ("6. Create PDF Report", "Generates this PDF with a data table, downtime insights, maintenance alerts, and this detailed flowchart.")