ultralytics 8.3.69 New Results to_sql() method for SQL format (#18921)

Signed-off-by: Glenn Jocher <glenn.jocher@ultralytics.com>
Co-authored-by: UltralyticsAssistant <web@ultralytics.com>
Co-authored-by: Glenn Jocher <glenn.jocher@ultralytics.com>

.github/workflows/ci.yml

@@ -98,7 +98,8 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        os: [ubuntu-latest, windows-latest, macos-15, ubuntu-24.04-arm]
+        # Temporarily disable windows-latest due to https://github.com/ultralytics/ultralytics/actions/runs/13020330819/job/36319338854?pr=18921
+        os: [ubuntu-latest, macos-15, ubuntu-24.04-arm]
         python-version: ["3.11"]
         model: [yolo11n]
     steps:
@@ -151,7 +152,9 @@ jobs:
       - name: Benchmark Summary
         run: |
           cat benchmarks.log
-          echo "$(cat benchmarks.log)" >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
+          cat benchmarks.log >> $GITHUB_STEP_SUMMARY
+          echo '```' >> $GITHUB_STEP_SUMMARY
 
   Tests:
     if: github.event_name != 'workflow_dispatch' || github.event.inputs.tests == 'true'

docs/en/modes/predict.md

@@ -508,7 +508,12 @@ All Ultralytics `predict()` calls will return a list of `Results` objects:
 | `verbose()`   | `str`           | Return log string for each task.                                                    |
 | `save_txt()`  | `None`          | Save predictions into a txt file.                                                   |
 | `save_crop()` | `None`          | Save cropped predictions to `save_dir/cls/file_name.jpg`.                           |
-| `tojson()`    | `str`           | Convert the object to JSON format.                                                  |
+| `summary()`   | `List[Dict]`    | A list of dictionaries, each containing summarized information for results          |
+| `to_df()`     | `DataFrame`     | Convert the results to Pandas Dataframe.                                            |
+| `to_csv()`    | `str`           | Convert the result to CSV (comma separated values) format.                          |
+| `to_xml()`    | `str`           | Convert the results to XML (Extensible Markup Language) format.                     |
+| `to_json()`   | `str`           | Convert the results to JSON format.                                                 |
+| `to_sql()`    | `None`          | Dump the results into the SQL database.                                             |
 
 For more details see the [`Results` class documentation](../reference/engine/results.md).
 

ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.68"
+__version__ = "8.3.69"
 
 import os
 

ultralytics/engine/results.py

@@ -937,6 +937,75 @@ class Results(SimpleClass):
 
         return json.dumps(self.summary(normalize=normalize, decimals=decimals), indent=2)
 
+    def to_sql(self, table_name="results", normalize=False, decimals=5, db_path="results.db"):
+        """
+        Converts detection results to an SQL-compatible format.
+
+        This method serializes the detection results into a format compatible with SQL databases.
+        It includes information about detected objects such as bounding boxes, class names, confidence scores,
+        and optionally segmentation masks, keypoints or oriented bounding boxes.
+
+        Args:
+            table_name (str): Name of the SQL table where the data will be inserted. Defaults to "detection_results".
+            normalize (bool): Whether to normalize the bounding box coordinates by the image dimensions.
+                If True, coordinates will be returned as float values between 0 and 1. Defaults to False.
+            decimals (int): Number of decimal places to round the bounding boxes values to. Defaults to 5.
+            db_path (str): Path to the SQLite database file. Defaults to "results.db".
+
+        Examples:
+            >>> results = model("path/to/image.jpg")
+            >>> results[0].to_sql()
+            >>> print("SQL data written successfully.")
+        """
+        import json
+        import sqlite3
+
+        # Convert results to a list of dictionaries
+        data = self.summary(normalize=normalize, decimals=decimals)
+        if not data:
+            LOGGER.warning("⚠️ No results to save to SQL. Results dict is empty")
+            return
+
+        # Connect to the SQLite database
+        conn = sqlite3.connect(db_path)
+        cursor = conn.cursor()
+
+        # Create table if it doesn't exist
+        columns = (
+            "id INTEGER PRIMARY KEY AUTOINCREMENT, class_name TEXT, confidence REAL, "
+            "box TEXT, masks TEXT, kpts TEXT, obb TEXT"
+        )
+        cursor.execute(f"CREATE TABLE IF NOT EXISTS {table_name} ({columns})")
+
+        # Insert data into the table
+        for i, item in enumerate(data):
+            detect, obb = None, None  # necessary to reinit these variables inside for loop to avoid duplication
+            class_name = item.get("name")
+            box = item.get("box", {})
+            # Serialize the box as JSON for 'detect' and 'obb' based on key presence
+            if all(key in box for key in ["x1", "y1", "x2", "y2"]) and not any(key in box for key in ["x3", "x4"]):
+                detect = json.dumps(box)
+            if all(key in box for key in ["x1", "y1", "x2", "y2", "x3", "x4"]):
+                obb = json.dumps(box)
+
+            cursor.execute(
+                f"INSERT INTO {table_name} (class_name, confidence, box, masks, kpts, obb) VALUES (?, ?, ?, ?, ?, ?)",
+                (
+                    class_name,
+                    item.get("confidence"),
+                    detect,
+                    json.dumps(item.get("segments", {}).get("x", [])),
+                    json.dumps(item.get("keypoints", {}).get("x", [])),
+                    obb,
+                ),
+            )
+
+        # Commit and close the connection
+        conn.commit()
+        conn.close()
+
+        LOGGER.info(f"✅ Detection results successfully written to SQL table '{table_name}' in database '{db_path}'.")
+
 
 class Boxes(BaseTensor):
     """

ultralytics/utils/benchmarks.py

@@ -173,10 +173,10 @@ def benchmark(
     df = pd.DataFrame(y, columns=["Format", "Status❔", "Size (MB)", key, "Inference time (ms/im)", "FPS"])
 
     name = model.model_name
-    s = f"\nBenchmarks complete for {name} on {data} at imgsz={imgsz} ({time.time() - t0:.2f}s)\n{df.fillna('-')}\n"
+    dt = time.time() - t0
+    legend = "Benchmarks legend:  - ✅ Success  - ❎ Export passed but validation failed  - ❌️ Export failed"
+    s = f"\nBenchmarks complete for {name} on {data} at imgsz={imgsz} ({dt:.2f}s)\n{legend}\n{df.fillna('-')}\n"
     LOGGER.info(s)
-    LOGGER.info("Status legends:")
-    LOGGER.info("✅ - Benchmark passed | ❎ - Export passed but validation failed | ❌️ - Export failed")
     with open("benchmarks.log", "a", errors="ignore", encoding="utf-8") as f:
         f.write(s)