From e6c9817a988ad97298a9bea290090e57d0aea824 Mon Sep 17 00:00:00 2001
From: Masonmason <masonhuang0320@gmail.com>
Date: Thu, 17 Jul 2025 10:22:48 +0800
Subject: [PATCH] feat: Add real-time inference results display to deployment
 UI
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

- Add result callback mechanism to WorkflowOrchestrator
- Implement result_updated signal in DeploymentWorker
- Create detailed inference results display with timestamps and formatted output
- Support both tuple and dict result formats
- Add auto-scrolling results panel with history management
- Connect pipeline results to Live View tab for real-time monitoring

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
---
 .../core/functions/workflow_orchestrator.py   | 11 ++++
 cluster4npu_ui/ui/dialogs/deployment.py       | 51 +++++++++++++++++++
 2 files changed, 62 insertions(+)

diff --git a/cluster4npu_ui/core/functions/workflow_orchestrator.py b/cluster4npu_ui/core/functions/workflow_orchestrator.py
index cef8042..2ba6ba7 100644
--- a/cluster4npu_ui/core/functions/workflow_orchestrator.py
+++ b/cluster4npu_ui/core/functions/workflow_orchestrator.py
@@ -31,6 +31,7 @@ class WorkflowOrchestrator:
         self.running = False
         self._stop_event = threading.Event()
         self.frame_callback = None
+        self.result_callback = None
 
     def start(self):
         """
@@ -94,6 +95,12 @@ class WorkflowOrchestrator:
         Sets the callback function for frame updates.
         """
         self.frame_callback = callback
+    
+    def set_result_callback(self, callback):
+        """
+        Sets the callback function for inference results.
+        """
+        self.result_callback = callback
 
     def _create_data_source(self) -> Optional[Any]:
         """
@@ -157,6 +164,10 @@ class WorkflowOrchestrator:
                     self.pipeline.pipeline_name,
                     format=self.output_config.get('format', 'json').lower()
                 )
+                
+                # Also call the result callback if set
+                if self.result_callback:
+                    self.result_callback(result_dict)
             except Exception as e:
                 print(f"Error handling result: {e}")
 
diff --git a/cluster4npu_ui/ui/dialogs/deployment.py b/cluster4npu_ui/ui/dialogs/deployment.py
index c79e2f8..4582c5a 100644
--- a/cluster4npu_ui/ui/dialogs/deployment.py
+++ b/cluster4npu_ui/ui/dialogs/deployment.py
@@ -65,6 +65,7 @@ class DeploymentWorker(QThread):
     deployment_completed = pyqtSignal(bool, str)  # success, message
     error_occurred = pyqtSignal(str)
     frame_updated = pyqtSignal('PyQt_PyObject')  # For live view
+    result_updated = pyqtSignal(dict)  # For inference results
     
     def __init__(self, pipeline_data: Dict[str, Any]):
         super().__init__()
@@ -131,6 +132,7 @@ class DeploymentWorker(QThread):
                 # Create and start the orchestrator
                 self.orchestrator = WorkflowOrchestrator(pipeline, config.input_config, config.output_config)
                 self.orchestrator.set_frame_callback(self.frame_updated.emit)
+                self.orchestrator.set_result_callback(self.result_updated.emit)
                 self.orchestrator.start()
                 
                 self.progress_updated.emit(100, "Pipeline deployed successfully!")
@@ -525,6 +527,7 @@ Stage Configurations:
         self.deployment_worker.deployment_completed.connect(self.on_deployment_completed)
         self.deployment_worker.error_occurred.connect(self.on_deployment_error)
         self.deployment_worker.frame_updated.connect(self.update_live_view)
+        self.deployment_worker.result_updated.connect(self.update_inference_results)
         
         self.deployment_worker.start()
     
@@ -634,6 +637,54 @@ Stage Configurations:
         except Exception as e:
             print(f"Error updating live view: {e}")
     
+    def update_inference_results(self, result_dict):
+        """Update the inference results display."""
+        try:
+            import json
+            from datetime import datetime
+            
+            # Format the results for display
+            timestamp = datetime.fromtimestamp(result_dict.get('timestamp', 0)).strftime("%H:%M:%S.%f")[:-3]
+            stage_results = result_dict.get('stage_results', {})
+            
+            result_text = f"[{timestamp}] Pipeline ID: {result_dict.get('pipeline_id', 'Unknown')}\n"
+            
+            # Display results from each stage
+            for stage_id, result in stage_results.items():
+                result_text += f"  {stage_id}:\n"
+                if isinstance(result, tuple) and len(result) == 2:
+                    # Handle tuple results (probability, result_string)
+                    probability, result_string = result
+                    result_text += f"    Result: {result_string}\n"
+                    result_text += f"    Probability: {probability:.3f}\n"
+                elif isinstance(result, dict):
+                    # Handle dict results
+                    for key, value in result.items():
+                        if key == 'probability':
+                            result_text += f"    Probability: {value:.3f}\n"
+                        else:
+                            result_text += f"    {key}: {value}\n"
+                else:
+                    result_text += f"    {result}\n"
+            
+            result_text += "-" * 50 + "\n"
+            
+            # Append to results display (keep last 100 lines)
+            current_text = self.results_text.toPlainText()
+            lines = current_text.split('\n')
+            if len(lines) > 100:
+                lines = lines[-50:]  # Keep last 50 lines
+                current_text = '\n'.join(lines)
+            
+            self.results_text.setPlainText(current_text + result_text)
+            
+            # Auto-scroll to bottom
+            scrollbar = self.results_text.verticalScrollBar()
+            scrollbar.setValue(scrollbar.maximum())
+            
+        except Exception as e:
+            print(f"Error updating inference results: {e}")
+    
     def apply_theme(self):
         """Apply consistent theme to the dialog."""
         self.setStyleSheet("""