Update app.py

app.py

@@ -4,7 +4,9 @@ from queue import Queue
 import time
 from prometheus_client import start_http_server, Counter, Histogram
 import threading
+import psutil
 
+# Load the model
 gr.load("models/Sevixdd/roberta-base-finetuned-ner").launch()
 
 # --- Prometheus Metrics Setup ---
@@ -37,27 +39,42 @@ def chat_function(message, history):
             logging.error(f"Error in chat processing: {e}")
             return "An error occurred. Please try again."
 
-# --- Gradio Interface with Metrics Display ---
-with gr.Blocks() as demo:
-    gr.Markdown("## PLOD Filtered")
-    chatbot = gr.ChatInterface(fn=chat_function)
-
-    with gr.Row():
-        request_count_display = gr.Textbox(label="Request Count")
-        avg_latency_display = gr.Textbox(label="Avg. Response Time (s)")
-
-    # --- Update Metrics Display ---
-    def update_metrics():
-        while True:
-            request_count_display.update(REQUEST_COUNT.collect()[0].samples[0].value)
-            avg_latency_display.update(round(REQUEST_LATENCY.collect()[0].samples[0].value, 2))
-            time.sleep(5)  # Update every 5 seconds
-
-    # Start the metrics server in a separate thread
+# --- Gradio Interface with Three Windows ---
+with gr.Blocks(title="PLOD Filtered with Monitoring") as demo:
+    with gr.Tab("Chat"):
+        gr.Markdown("## Chat with the Bot")
+        chatbot = gr.ChatInterface(fn=chat_function)
+
+    with gr.Tab("Performance Metrics"):
+        with gr.Row():
+            request_count_display = gr.Textbox(label="Request Count")
+            avg_latency_display = gr.Textbox(label="Avg. Response Time (s)")
+
+        def update_metrics():
+            while True:
+                request_count_display.update(REQUEST_COUNT.collect()[0].samples[0].value)
+                avg_latency_display.update(round(REQUEST_LATENCY.collect()[0].samples[0].value, 2))
+                time.sleep(5)  # Update every 5 seconds
+
+    with gr.Tab("Resource Usage"):
+        with gr.Row():
+            cpu_usage_display = gr.Textbox(label="CPU Usage (%)")
+            mem_usage_display = gr.Textbox(label="Memory Usage (%)")
+
+        def update_usage():
+            while True:
+                cpu_usage_display.update(psutil.cpu_percent())
+                mem_usage_display.update(psutil.virtual_memory().percent)
+                time.sleep(5) 
+
+    # --- Start Threads ---
     threading.Thread(target=start_http_server, args=(8000,), daemon=True).start()
-    demo.load(update_metrics, None, None, every=1)  # Call update_metrics every second after initial page load
+    threading.Thread(target=update_metrics, daemon=True).start()
+    threading.Thread(target=update_usage, daemon=True).start()
+
+# Launch the app
+demo.launch(share=True)
 
-#demo.launch(share=True)  # For Hugging Face Spaces, set share=True to make the UI public