ultralytics 8.3.127__tar.gz → 8.3.128__tar.gz

{ultralytics-8.3.127/ultralytics.egg-info → ultralytics-8.3.128}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.127
+Version: 8.3.128
 Summary: Ultralytics YOLO 🚀 for SOTA object detection, multi-object tracking, instance segmentation, pose estimation and image classification.
 Author-email: Glenn Jocher <glenn.jocher@ultralytics.com>, Jing Qiu <jing.qiu@ultralytics.com>
 Maintainer-email: Ultralytics <hello@ultralytics.com>

{ultralytics-8.3.127 → ultralytics-8.3.128}/pyproject.toml

@@ -134,7 +134,7 @@ ultralytics = "ultralytics.cfg:entrypoint"
 [tool.setuptools]  # configuration specific to the `setuptools` build backend.
 packages = { find = { where = ["."], include = ["ultralytics", "ultralytics.*"] } }
 # Tests included below for checking Conda builds in https://github.com/conda-forge/ultralytics-feedstock
-package-data = { "ultralytics" = ["**/*.yaml", "**/*.sh", "../tests/*.py"], "ultralytics.assets" = ["*.jpg"] }
+package-data = { "ultralytics" = ["**/*.yaml", "**/*.sh", "../tests/*.py"], "ultralytics.assets" = ["*.jpg"], "ultralytics.solutions.templates" = ["*.html"]}
 
 [tool.setuptools.dynamic]
 version = { attr = "ultralytics.__version__" }

{ultralytics-8.3.127 → ultralytics-8.3.128}/tests/test_cuda.py

@@ -71,8 +71,13 @@ def test_export_engine_matrix(task, dynamic, int8, half, batch):
 @pytest.mark.skipif(not DEVICES, reason="No CUDA devices available")
 def test_train():
     """Test model training on a minimal dataset using available CUDA devices."""
-    device = DEVICES if len(DEVICES) > 1 else DEVICES[0]
-    YOLO(MODEL).train(data="coco8.yaml", imgsz=64, epochs=1, device=device)  # requires imgsz>=64
+    import os
+
+    device = tuple(DEVICES) if len(DEVICES) > 1 else DEVICES[0]
+    results = YOLO(MODEL).train(data="coco8.yaml", imgsz=64, epochs=1, device=device)  # requires imgsz>=64
+    visible = eval(os.environ["CUDA_VISIBLE_DEVICES"])
+    assert visible == device, f"Passed GPUs '{device}', but used GPUs '{visible}'"
+    assert results is (None if len(DEVICES) > 1 else not None)  # DDP returns None, single-GPU returns metrics
 
 
 @pytest.mark.slow

{ultralytics-8.3.127 → ultralytics-8.3.128}/tests/test_python.py

@@ -188,11 +188,11 @@ def test_track_stream():
     model.track(video_url, imgsz=160, tracker="bytetrack.yaml")
     model.track(video_url, imgsz=160, tracker="botsort.yaml", save_frames=True)  # test frame saving also
 
-    # Test Global Motion Compensation (GMC) methods
-    for gmc in "orb", "sift", "ecc":
+    # Test Global Motion Compensation (GMC) methods and ReID
+    for gmc, reidm in zip(["orb", "sift", "ecc"], ["auto", "auto", "yolo11n-cls.pt"]):
         default_args = YAML.load(ROOT / "cfg/trackers/botsort.yaml")
         custom_yaml = TMP / f"botsort-{gmc}.yaml"
-        YAML.save(custom_yaml, {**default_args, "gmc_method": gmc})
+        YAML.save(custom_yaml, {**default_args, "gmc_method": gmc, "with_reid": True, "model": reidm})
         model.track(video_url, imgsz=160, tracker=custom_yaml)
 
 

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/__init__.py

@@ -1,6 +1,6 @@
 # Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
 
-__version__ = "8.3.127"
+__version__ = "8.3.128"
 
 import os
 

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/engine/exporter.py

@@ -82,6 +82,7 @@ from ultralytics.utils import (
     ARM64,
     DEFAULT_CFG,
     IS_COLAB,
+    IS_JETSON,
     LINUX,
     LOGGER,
     MACOS,
@@ -682,6 +683,7 @@ class Exporter:
     @try_export
     def export_paddle(self, prefix=colorstr("PaddlePaddle:")):
         """YOLO Paddle export."""
+        assert not IS_JETSON, "Jetson Paddle exports not supported yet"
         check_requirements(("paddlepaddle-gpu" if torch.cuda.is_available() else "paddlepaddle>=3.0.0", "x2paddle"))
         import x2paddle  # noqa
         from x2paddle.convert import pytorch2paddle  # noqa

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/engine/trainer.py

@@ -105,7 +105,8 @@ class BaseTrainer:
         self.args = get_cfg(cfg, overrides)
         self.check_resume(overrides)
         self.device = select_device(self.args.device, self.args.batch)
-        self.args.device = str(self.device)  # ensure -1 is updated to selected CUDA device
+        # update "-1" devices so post-training val does not repeat search
+        self.args.device = os.getenv("CUDA_VISIBLE_DEVICES") if "cuda" in str(self.device) else str(self.device)
         self.validator = None
         self.metrics = None
         self.plots = {}

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/models/yolo/detect/predict.py

@@ -82,11 +82,9 @@ class DetectionPredictor(BasePredictor):
 
     def get_obj_feats(self, feat_maps, idxs):
         """Extract object features from the feature maps."""
-        from math import gcd
-
         import torch
 
-        s = gcd(*[x.shape[1] for x in feat_maps])  # find smallest vector length
+        s = min([x.shape[1] for x in feat_maps])  # find smallest vector length
         obj_feats = torch.cat(
             [x.permute(0, 2, 3, 1).reshape(x.shape[0], -1, s, x.shape[1] // s).mean(dim=-1) for x in feat_maps], dim=1
         )  # mean reduce all vectors to same length

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/nn/autobackend.py

@@ -290,8 +290,8 @@ class AutoBackend(nn.Module):
         elif engine:
             LOGGER.info(f"Loading {w} for TensorRT inference...")
 
-            if IS_JETSON and check_version(PYTHON_VERSION, "<=3.8.0"):
-                # fix error: `np.bool` was a deprecated alias for the builtin `bool` for JetPack 4 with Python <= 3.8.0
+            if IS_JETSON and check_version(PYTHON_VERSION, "<=3.8.10"):
+                # fix error: `np.bool` was a deprecated alias for the builtin `bool` for JetPack 4 and JetPack 5 with Python <= 3.8.10
                 check_requirements("numpy==1.23.5")
 
             try:  # https://developer.nvidia.com/nvidia-tensorrt-download

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/solutions/similarity_search.py

@@ -29,16 +29,16 @@ class VisualAISearch(BaseSolution):
     def __init__(self, **kwargs):
         """Initializes the VisualAISearch class with the FAISS index file and CLIP model."""
         super().__init__(**kwargs)
-        check_requirements(["open-clip-torch", "faiss-cpu"])
+        check_requirements(["git+https://github.com/ultralytics/CLIP.git", "faiss-cpu"])
+        import clip
         import faiss
-        import open_clip
 
         self.faiss = faiss
-        self.open_clip = open_clip
+        self.clip = clip
 
         self.faiss_index = "faiss.index"
         self.data_path_npy = "paths.npy"
-        self.model_name = "ViT-B-32-quickgelu"
+        self.model_name = "ViT-B/32"
         self.data_dir = Path(self.CFG["data"])
         self.device = select_device(self.CFG["device"])
 
@@ -51,11 +51,7 @@ class VisualAISearch(BaseSolution):
             safe_download(url=f"{ASSETS_URL}/images.zip", unzip=True, retry=3)
             self.data_dir = Path("images")
 
-        self.clip_model, _, self.preprocess = self.open_clip.create_model_and_transforms(
-            self.model_name, pretrained="openai"
-        )
-        self.clip_model = self.clip_model.to(self.device).eval()
-        self.tokenizer = self.open_clip.get_tokenizer(self.model_name)
+        self.model, self.preprocess = clip.load(self.model_name, device=self.device)
 
         self.index = None
         self.image_paths = []
@@ -67,13 +63,13 @@ class VisualAISearch(BaseSolution):
         image = Image.open(path)
         tensor = self.preprocess(image).unsqueeze(0).to(self.device)
         with torch.no_grad():
-            return self.clip_model.encode_image(tensor).cpu().numpy()
+            return self.model.encode_image(tensor).cpu().numpy()
 
     def extract_text_feature(self, text):
         """Extract CLIP text embedding."""
-        tokens = self.tokenizer([text]).to(self.device)
+        tokens = self.clip.tokenize([text]).to(self.device)
         with torch.no_grad():
-            return self.clip_model.encode_text(tokens).cpu().numpy()
+            return self.model.encode_text(tokens).cpu().numpy()
 
     def load_or_build_index(self):
         """Loads FAISS index or builds a new one from image features."""

ultralytics-8.3.128/ultralytics/solutions/templates/similarity-search.html

@@ -0,0 +1,160 @@
+<!-- Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license -->
+
+<!--Similarity search webpage-->
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>Semantic Image Search</title>
+    <link
+      href="https://fonts.googleapis.com/css2?family=Inter:wght@400;600&display=swap"
+      rel="stylesheet"
+    />
+    <style>
+      body {
+        background: linear-gradient(135deg, #f0f4ff, #f9fbff);
+        font-family: "Inter", sans-serif;
+        color: #111e68;
+        padding: 2rem;
+        margin: 0;
+        min-height: 100vh;
+      }
+
+      h1 {
+        text-align: center;
+        margin-bottom: 2rem;
+        font-size: 2.5rem;
+        font-weight: 600;
+      }
+
+      form {
+        display: flex;
+        flex-wrap: wrap;
+        justify-content: center;
+        align-items: center;
+        gap: 1rem;
+        margin-bottom: 3rem;
+        animation: fadeIn 1s ease-in-out;
+      }
+
+      input[type="text"] {
+        width: 300px;
+        padding: 0.75rem 1rem;
+        font-size: 1rem;
+        border-radius: 10px;
+        border: 1px solid #ccc;
+        box-shadow: 0 2px 6px rgba(0, 0, 0, 0.05);
+        transition: box-shadow 0.3s ease;
+      }
+
+      input[type="text"]:focus {
+        outline: none;
+        box-shadow: 0 0 0 3px rgba(17, 30, 104, 0.2);
+      }
+
+      button {
+        background-color: #111e68;
+        color: white;
+        font-weight: 600;
+        font-size: 1rem;
+        padding: 0.75rem 1.5rem;
+        border-radius: 10px;
+        border: none;
+        cursor: pointer;
+        transition:
+          background-color 0.3s ease,
+          transform 0.2s ease;
+      }
+
+      button:hover {
+        background-color: #1f2e9f;
+        transform: translateY(-2px);
+      }
+
+      .grid {
+        display: grid;
+        grid-template-columns: repeat(auto-fill, minmax(260px, 1fr));
+        gap: 1.5rem;
+        max-width: 1600px;
+        margin: auto;
+        animation: fadeInUp 1s ease-in-out;
+      }
+
+      .card {
+        background: white;
+        border-radius: 16px;
+        overflow: hidden;
+        box-shadow: 0 6px 14px rgba(0, 0, 0, 0.08);
+        transition:
+          transform 0.3s ease,
+          box-shadow 0.3s ease;
+      }
+
+      .card:hover {
+        transform: translateY(-6px);
+        box-shadow: 0 10px 20px rgba(0, 0, 0, 0.1);
+      }
+
+      .card img {
+        width: 100%;
+        height: 100%;
+        object-fit: cover;
+        display: block;
+      }
+
+      @keyframes fadeIn {
+        0% {
+          opacity: 0;
+          transform: scale(0.95);
+        }
+        100% {
+          opacity: 1;
+          transform: scale(1);
+        }
+      }
+
+      @keyframes fadeInUp {
+        0% {
+          opacity: 0;
+          transform: translateY(20px);
+        }
+        100% {
+          opacity: 1;
+          transform: translateY(0);
+        }
+      }
+    </style>
+  </head>
+  <body>
+    <div style="text-align: center; margin-bottom: 1rem">
+      <img
+        src="https://raw.githubusercontent.com/ultralytics/assets/main/logo/favicon.png"
+        alt="Ultralytics Logo"
+        style="height: 40px"
+      />
+    </div>
+    <h1>Semantic Image Search with AI</h1>
+
+    <!-- Search box -->
+    <form method="POST">
+      <input
+        type="text"
+        name="query"
+        placeholder="Describe the scene (e.g., man walking)"
+        value="{{ request.form['query'] }}"
+        required
+      />
+      <button type="submit">Search</button>
+    </form>
+
+    <!-- Search results grid -->
+    <div class="grid">
+      {% for img in results %}
+      <div class="card">
+        <img src="{{ url_for('static', filename=img) }}" alt="Result Image" />
+      </div>
+      {% endfor %}
+    </div>
+  </body>
+</html>

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/trackers/bot_sort.py

@@ -255,6 +255,6 @@ class ReID:
     def __call__(self, img, dets):
         """Extract embeddings for detected objects."""
         feats = self.model([save_one_box(det, img, save=False) for det in xywh2xyxy(torch.from_numpy(dets[:, :4]))])
-        if feats.shape[0] != dets.shape[0] and feats[0].shape[0] == dets.shape[0]:
+        if len(feats) != dets.shape[0] and feats[0].shape[0] == dets.shape[0]:
             feats = feats[0]  # batched prediction with non-PyTorch backend
         return [f.cpu().numpy() for f in feats]

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/trackers/track.py

@@ -44,6 +44,8 @@ def on_predict_start(predictor: object, persist: bool = False) -> None:
     if cfg.tracker_type not in {"bytetrack", "botsort"}:
         raise AssertionError(f"Only 'bytetrack' and 'botsort' are supported for now, but got '{cfg.tracker_type}'")
 
+    predictor._feats = None  # reset in case used earlier
+    predictor.save_feats = False
     if cfg.tracker_type == "botsort" and cfg.with_reid and cfg.model == "auto":
         from ultralytics.nn.modules.head import Detect
 
@@ -55,7 +57,6 @@ def on_predict_start(predictor: object, persist: bool = False) -> None:
             cfg.model = "yolo11n-cls.pt"
         else:
             predictor.save_feats = True
-            predictor._feats = None
 
             # Register hook to extract input of Detect layer
             def pre_hook(module, input):

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics/utils/benchmarks.py

@@ -40,7 +40,7 @@ import torch.cuda
 from ultralytics import YOLO, YOLOWorld
 from ultralytics.cfg import TASK2DATA, TASK2METRIC
 from ultralytics.engine.exporter import export_formats
-from ultralytics.utils import ARM64, ASSETS, LINUX, LOGGER, MACOS, TQDM, WEIGHTS_DIR, YAML
+from ultralytics.utils import ARM64, ASSETS, IS_JETSON, LINUX, LOGGER, MACOS, TQDM, WEIGHTS_DIR, YAML
 from ultralytics.utils.checks import IS_PYTHON_3_13, check_imgsz, check_requirements, check_yolo, is_rockchip
 from ultralytics.utils.downloads import safe_download
 from ultralytics.utils.files import file_size
@@ -126,7 +126,7 @@ def benchmark(
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 Paddle exports not supported yet"
                 assert model.task != "obb", "Paddle OBB bug https://github.com/PaddlePaddle/Paddle/issues/72024"
                 assert not is_end2end, "End-to-end models not supported by PaddlePaddle yet"
-                assert LINUX or MACOS, "Windows Paddle exports not supported yet"
+                assert (LINUX and not IS_JETSON) or MACOS, "Windows and Jetson Paddle exports not supported yet"
             if i == 12:  # MNN
                 assert not isinstance(model, YOLOWorld), "YOLOWorldv2 MNN exports not supported yet"
             if i == 13:  # NCNN

{ultralytics-8.3.127 → ultralytics-8.3.128/ultralytics.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultralytics
-Version: 8.3.127
+Version: 8.3.128
 Summary: Ultralytics YOLO 🚀 for SOTA object detection, multi-object tracking, instance segmentation, pose estimation and image classification.
 Author-email: Glenn Jocher <glenn.jocher@ultralytics.com>, Jing Qiu <jing.qiu@ultralytics.com>
 Maintainer-email: Ultralytics <hello@ultralytics.com>

{ultralytics-8.3.127 → ultralytics-8.3.128}/ultralytics.egg-info/SOURCES.txt

@@ -238,6 +238,7 @@ ultralytics/solutions/speed_estimation.py
 ultralytics/solutions/streamlit_inference.py
 ultralytics/solutions/trackzone.py
 ultralytics/solutions/vision_eye.py
+ultralytics/solutions/templates/similarity-search.html
 ultralytics/trackers/__init__.py
 ultralytics/trackers/basetrack.py
 ultralytics/trackers/bot_sort.py