release: freeze Nexus One AI 1.0.0-rc1

ansible/roles/cezen-backend/files/train_qlora.py

@@ -73,7 +73,7 @@ def set_status(status: str):
 # ── Dataset loading ───────────────────────────────────────────────────────────
 
 def load_dataset_from_file(path: str):
-    """Load JSONL or CSV dataset into a list of dicts with 'text' or 'prompt'/'completion' keys."""
+    """Load JSON, JSONL, or CSV training records as a list of objects."""
     p = Path(path)
     rows = []
     if p.suffix.lower() == ".csv":
@@ -82,6 +82,16 @@ def load_dataset_from_file(path: str):
             reader = csv.DictReader(f)
             for row in reader:
                 rows.append(dict(row))
+    elif p.suffix.lower() == ".json":
+        with open(path, encoding="utf-8", errors="replace") as f:
+            payload = json.load(f)
+        if isinstance(payload, list):
+            rows = payload
+        elif isinstance(payload, dict):
+            for key in ("samples", "data", "records"):
+                if isinstance(payload.get(key), list):
+                    rows = payload[key]
+                    break
     else:
         with open(path, encoding="utf-8", errors="replace") as f:
             for line in f:
@@ -90,7 +100,11 @@ def load_dataset_from_file(path: str):
                     try:
                         rows.append(json.loads(line))
                     except Exception:
-                        pass
+                        raise ValueError("Dataset contains an invalid JSONL row")
+    if not rows:
+        raise ValueError("Dataset has no training records")
+    if not all(isinstance(row, dict) for row in rows):
+        raise ValueError("Every training record must be a JSON object or CSV row")
     return rows
 
 def format_row(row: dict) -> str: