Deploy Auto-Quantization MVP

app.py

@@ -173,21 +173,27 @@ Upload a model to HuggingFace Hub to trigger automatic quantization!
             "failed": "❌"
         }.get(job["status"], "❓")
 
-        jobs_text += f"""
-### {status_emoji} Job #{job['id']} - {job['status'].upper()}
+        # Truncate model ID if too long
+        model_display = job['model_id']
+        if len(model_display) > 50:
+            model_display = model_display[:47] + "..."
 
-**Model:** `{job['model_id']}`
-**Method:** {job['method']}
-**Time:** {job['timestamp']}
-"""
+        jobs_text += f"\n### {status_emoji} Job #{job['id']} - {job['status'].upper()}\n\n"
+        jobs_text += f"**Model:** `{model_display}`  \n"
+        jobs_text += f"**Method:** {job['method']}  \n"
+        jobs_text += f"**Time:** {job['timestamp']}  \n"
 
         if job["status"] == "completed" and "output_repo" in job:
             jobs_text += f"**✨ Output:** [{job['output_repo']}](https://huggingface.co/{job['output_repo']})  \n"
 
         if job["status"] == "failed" and "error" in job:
-            jobs_text += f"**Error:** {job['error'][:200]}...  \n"
+            # Truncate long errors and make them more readable
+            error_msg = job['error']
+            if len(error_msg) > 150:
+                error_msg = error_msg[:150] + "..."
+            jobs_text += f"**Error:** {error_msg}  \n"
 
-        jobs_text += "---\n\n"
+        jobs_text += "\n---\n"
 
     return jobs_text
 
@@ -206,9 +212,21 @@ def get_metrics():
 
     total = len(job_queue)
     completed = len([j for j in job_queue if j["status"] == "completed"])
-    failed = len([j for j in job_queue if j["status"] == "failed"])
 
-    success_rate = f"{(completed/(completed+failed)*100):.1f}%" if (completed + failed) > 0 else "N/A"
+    # Only count legitimate failures (not "already quantized" or validation errors)
+    legitimate_failures = []
+    for j in job_queue:
+        if j["status"] == "failed":
+            error = j.get("error", "")
+            # Skip validation failures like "already quantized"
+            if "already quantized" not in error.lower() and "skipping" not in error.lower():
+                legitimate_failures.append(j)
+
+    failed = len(legitimate_failures)
+
+    # Calculate success rate based only on legitimate attempts
+    legitimate_attempts = completed + failed
+    success_rate = f"{(completed/legitimate_attempts*100):.1f}%" if legitimate_attempts > 0 else "N/A"
 
     # Estimated time saved (30 min per model)
     time_saved = completed * 0.5