Сохранять каждый запуск датасета в отдельную папку и показывать прогресс генерации.

Файлы пишутся в sonar_dataset/дата-время-модель без удаления прошлых запусков; кнопка «Генерация…» заполняется по мере записи сцен через NDJSON-стрим.

Co-authored-by: Cursor <cursoragent@cursor.com>

backend/dataset_generator.py

@@ -474,6 +474,35 @@ def resolve_output_dir(output_dir: str | Path = "sonar_dataset") -> Path:
     return out
 
 
+def _safe_object_stem(object_name: str | None) -> str:
+    stem = Path(object_name or "object").stem
+    safe = "".join(ch if ch.isalnum() or ch in "_-" else "_" for ch in stem).strip("._-")
+    return (safe or "object")[:80]
+
+
+def make_generation_run_dir(base_dir: Path, object_name: str | None = None) -> Path:
+    """Create a new run subdirectory: YYYY-MM-DD_HH-MM-SS-<object_stem>.
+
+    Previous runs under base_dir are left untouched.
+    """
+    from datetime import datetime
+
+    base_dir.mkdir(parents=True, exist_ok=True)
+    stamp = datetime.now().strftime("%Y-%m-%d_%H-%M-%S")
+    folder = f"{stamp}-{_safe_object_stem(object_name)}"
+    path = base_dir / folder
+    if path.exists():
+        n = 2
+        while True:
+            candidate = base_dir / f"{folder}_{n}"
+            if not candidate.exists():
+                path = candidate
+                break
+            n += 1
+    path.mkdir(parents=True, exist_ok=False)
+    return path
+
+
 def _downsample_points(points: list[list[float]], max_points: int) -> list[list[float]]:
     max_points = max(100, int(max_points))
     if len(points) <= max_points:
@@ -588,7 +617,7 @@ def write_scene_files(
     return {"npy": str(npy_path), "obj": str(obj_path), "stem": stem}
 
 
-def generate_dataset(
+def iter_generate_dataset(
     *,
     count: int = 5,
     seed: int = 42,
@@ -599,16 +628,13 @@ def generate_dataset(
     object_scale_is_max: bool = False,
     beam_count: int = 45,
     length_count: int | None = None,
-) -> dict[str, Any]:
-    """Generate `count` unique scenes into output_dir with Area_X naming.
+):
+    """Yield NDJSON-friendly progress events, then a final ``done`` payload.
 
-    object_points: normalized template vertices from user .obj (class 1 = object).
-    object_scale: relative size multiplier vs unit-normalized mesh (1.0 = default).
-    object_scale_is_max: if True, treat object_scale as upper bound and sample
-        per-scene scale uniformly from [1, object_scale] (AUV altitude variation).
-    beam_count: across-track beams (width resolution, X).
-    length_count: along-track pings (length resolution, Y). Defaults to beam_count.
-    Each scene has exactly beam_count × length_count sonar returns (first-hit casting).
+    Events:
+      {"type":"start","total":N,"outputDir":"...","runName":"..."}
+      {"type":"progress","current":k,"total":N,"entry":{...}}
+      {"type":"done","result":{...}}
     """
     count = int(count)
     if count < 1:
@@ -636,7 +662,8 @@ def generate_dataset(
     if length_count > 1024:
         raise ValueError("length_count (Длина) must be <= 1024")
 
-    out = resolve_output_dir(output_dir)
+    base = resolve_output_dir(output_dir)
+    run_dir = make_generation_run_dir(base, object_name)
     template = normalize_object_points(object_points)
 
     labels = plan_scene_labels(count, seed)
@@ -655,6 +682,14 @@ def generate_dataset(
     preview_stem: str | None = None
     preview_has_object = False
 
+    yield {
+        "type": "start",
+        "total": count,
+        "outputDir": str(run_dir),
+        "baseDir": str(base),
+        "runName": run_dir.name,
+    }
+
     for i in range(count):
         visibility = labels[i]
         scene_seed = int(seed) + i * 10007 + 17
@@ -675,7 +710,7 @@ def generate_dataset(
             length_count=length_count,
         )
         area, scene_no, stem = scene_index_to_area_name(i)
-        paths = write_scene_files(scene, out, stem)
+        paths = write_scene_files(scene, run_dir, stem)
 
         entry = {
             "index": i,
@@ -702,6 +737,14 @@ def generate_dataset(
             preview_stem = stem
             preview_has_object = bool(scene["hasObject"])
 
+        yield {
+            "type": "progress",
+            "current": i + 1,
+            "total": count,
+            "entry": entry,
+            "outputDir": str(run_dir),
+        }
+
     preview: dict[str, Any] | None = None
     if preview_points is not None:
         pts = _downsample_points(preview_points, 25000)
@@ -713,8 +756,10 @@ def generate_dataset(
             "classLabels": {"0": "background", "1": "object"},
         }
 
-    return {
-        "outputDir": str(out),
+    result = {
+        "outputDir": str(run_dir),
+        "baseDir": str(base),
+        "runName": run_dir.name,
         "count": count,
         "seed": int(seed),
         "beamCount": beam_count,
@@ -728,3 +773,36 @@ def generate_dataset(
         "written": written,
         "preview": preview,
     }
+    yield {"type": "done", "result": result}
+
+
+def generate_dataset(
+    *,
+    count: int = 5,
+    seed: int = 42,
+    output_dir: str | Path = "sonar_dataset",
+    object_points: list[list[float]],
+    object_name: str | None = None,
+    object_scale: float = 1.0,
+    object_scale_is_max: bool = False,
+    beam_count: int = 45,
+    length_count: int | None = None,
+) -> dict[str, Any]:
+    """Generate `count` unique scenes into a new timestamped run folder under output_dir."""
+    result: dict[str, Any] | None = None
+    for event in iter_generate_dataset(
+        count=count,
+        seed=seed,
+        output_dir=output_dir,
+        object_points=object_points,
+        object_name=object_name,
+        object_scale=object_scale,
+        object_scale_is_max=object_scale_is_max,
+        beam_count=beam_count,
+        length_count=length_count,
+    ):
+        if event.get("type") == "done":
+            result = event["result"]
+    if result is None:
+        raise RuntimeError("Dataset generation produced no result.")
+    return result