Empezar

Clasificación de imágenes con Ultralytics YOLO

Ultralytics YOLO image classification of leaves as healthy, rust, or powdery mildew

La clasificación de imágenes es la más sencilla de las tareas compatibles y consiste en clasificar una imagen completa en una de varias clases predefinidas.

La salida de un clasificador de imágenes es una única etiqueta de clase y una puntuación de confianza. La clasificación de imágenes resulta útil cuando solo necesitas saber a qué clase pertenece una imagen y no necesitas conocer la ubicación ni la forma exacta de los objetos de esa clase.



Ver: Explora las tareas de Ultralytics YOLO: clasificación de imágenes con Ultralytics Platform
Consejo

Los modelos Classify de YOLO26 utilizan el sufijo -cls, por ejemplo, yolo26n-cls.pt, y están preentrenados con ImageNet.

Modelos#

A continuación se muestran los modelos YOLO26 Classify preentrenados con el conjunto de datos ImageNet.

Los modelos se descargan automáticamente desde la versión más reciente de Ultralytics al usarlos por primera vez.

Modelotamaño
(píxeles)
acc
top1
acc
top5
Velocidad
CPU ONNX
(ms)
Velocidad
T4 TensorRT10
(ms)
parámetros
(M)
FLOPs
(B) a 224
YOLO26n-cls22471.490.15.0 ± 0.31.1 ± 0.02.80.4
YOLO26s-cls22476.092.97.9 ± 0.21.3 ± 0.06.71.5
YOLO26m-cls22478.194.217.2 ± 0.42.0 ± 0.011.64.8
YOLO26l-cls22479.094.623.2 ± 0.32.8 ± 0.014.16.0
YOLO26x-cls22479.995.041.4 ± 0.93.8 ± 0.029.613.5
  • Los valores de acc corresponden a la precisión de los modelos en el conjunto de validación de ImageNet.
    Reprodúcelo con yolo classify val data=path/to/ImageNet device=0
  • Velocidad media en imágenes de validación de ImageNet, con ONNX en CPU y TensorRT10 en una GPU NVIDIA T4.
    Reprodúcela con yolo classify val data=path/to/ImageNet batch=1 device=0|cpu
  • Los valores de parámetros y FLOPs corresponden al modelo fusionado tras model.fuse(), que combina las capas Conv y BatchNorm. Los puntos de control preentrenados conservan la arquitectura de entrenamiento completa y pueden mostrar valores más altos.

Consulta la vista previa no publicada de YOLO27 para ver las velocidades preliminares de clasificación y los tamaños de los modelos.

Entrenar#

Entrena un modelo YOLO26n-cls con el conjunto de datos MNIST160 durante 100 épocas con un tamaño de imagen de 64.

Ejemplo
from ultralytics import YOLO

# Cargar un modelo
model = YOLO("yolo26n-cls.yaml")  # Crea un modelo nuevo a partir de YAML
model = YOLO("yolo26n-cls.pt")  # Carga un modelo preentrenado (recomendado para el entrenamiento)
model = YOLO("yolo26n-cls.yaml").load("yolo26n-cls.pt")  # Crea el modelo a partir de YAML y transfiere los pesos

# Entrenar el modelo
results = model.train(data="mnist160", epochs=100, imgsz=64)

Consulta todos los detalles del modo train en la página Entrenar. También puedes entrenar modelos de clasificación con el entrenamiento en la nube de Ultralytics Platform.

Formato del conjunto de datos#

Encontrarás todos los detalles sobre el formato de los conjuntos de datos de clasificación YOLO en la Guía de conjuntos de datos. También puedes gestionar y etiquetar conjuntos de datos de clasificación con las herramientas de anotación de Ultralytics Platform.

Transformaciones personalizadas#

La clasificación de Ultralytics YOLO utiliza torchvision.transforms.RandomResizedCrop para el entrenamiento y torchvision.transforms.CenterCrop para la validación y la inferencia. Estas transformaciones basadas en recortes presuponen que las imágenes de entrada son cuadradas y pueden eliminar regiones importantes de imágenes con relaciones de aspecto extremas. Para conservar la imagen completa sin estirarla, sustituye los recortes por la transformación existente ClassifyLetterBox, que cambia el tamaño de cada imagen para que quepa dentro de imgsz × imgsz y rellena el área restante, preservando la relación de aspecto. Úsala en la canalización de torchvision basada en PIL mediante un ClassificationDataset, ClassificationTrainer y ClassificationValidator personalizados:

import numpy as np
import torch
import torchvision.transforms as T

from ultralytics import YOLO
from ultralytics.data.augment import ClassifyLetterBox
from ultralytics.data.dataset import ClassificationDataset
from ultralytics.models.yolo.classify import ClassificationTrainer, ClassificationValidator

class CustomizedDataset(ClassificationDataset):
    """A customized dataset class for image classification with enhanced data augmentation transforms."""

    def __init__(self, root: str, args, augment: bool = False, prefix: str = "", names=None):
        """Initialize a customized classification dataset with enhanced data augmentation transforms."""
        super().__init__(root, args, augment, prefix, names)
        letterbox = T.Compose([np.asarray, ClassifyLetterBox(args.imgsz), T.ToPILImage()])

        # Añade aquí tus transformaciones de entrenamiento personalizadas
        train_transforms = T.Compose(
            [
                letterbox,
                T.RandomHorizontalFlip(p=args.fliplr),
                T.RandomVerticalFlip(p=args.flipud),
                T.RandAugment(interpolation=T.InterpolationMode.BILINEAR),
                T.ColorJitter(brightness=args.hsv_v, contrast=args.hsv_v, saturation=args.hsv_s, hue=args.hsv_h),
                T.ToTensor(),
                T.Normalize(mean=torch.tensor(0), std=torch.tensor(1)),
                T.RandomErasing(p=args.erasing, inplace=True),
            ]
        )

        # Añade aquí tus transformaciones de validación personalizadas
        val_transforms = T.Compose(
            [
                letterbox,
                T.ToTensor(),
                T.Normalize(mean=torch.tensor(0), std=torch.tensor(1)),
            ]
        )
        self.torch_transforms = train_transforms if augment else val_transforms

class CustomizedTrainer(ClassificationTrainer):
    """A customized trainer class for YOLO classification models with enhanced dataset handling."""

    def build_dataset(self, img_path: str, mode: str = "train", batch=None):
        """Build a customized dataset for classification training and the validation during training."""
        return CustomizedDataset(
            root=img_path,
            args=self.args,
            augment=mode == "train",
            prefix="train" if mode == "train" else self.args.split,
            names=self.data["names"],
        )

class CustomizedValidator(ClassificationValidator):
    """A customized validator class for YOLO classification models with enhanced dataset handling."""

    def build_dataset(self, img_path: str):
        """Build a customized dataset for classification standalone validation (no augmentation)."""
        return CustomizedDataset(root=img_path, args=self.args, augment=False, prefix=self.args.split, names=self.names)

model = YOLO("yolo26n-cls.pt")
model.train(data="imagenet", trainer=CustomizedTrainer, epochs=10, imgsz=224, batch=64)
model.val(data="imagenet", validator=CustomizedValidator, imgsz=224, batch=64)

Validar#

Valida la precisión de un modelo YOLO26n-cls entrenado. No se necesitan argumentos, ya que model conserva los atributos de entrenamiento data y sus argumentos: path/to/best.pt del ejemplo de entrenamiento valida con MNIST160. Los pesos oficiales registran una ruta al conjunto de datos de entrenamiento que no existe en tu equipo, así que recurren al valor predeterminado de la tarea imagenet10 y muestran una advertencia. Pasa data para validar con otro conjunto de datos.

Ejemplo
from ultralytics import YOLO

# Cargar un modelo
model = YOLO("yolo26n-cls.pt")  # carga un modelo oficial
model = YOLO("path/to/best.pt")  # cargar un modelo personalizado

# Valida el modelo
metrics = model.val()  # no hacen falta argumentos; se recuerdan el conjunto de datos y los ajustes
metrics.top1  # top1 accuracy
metrics.top5  # top5 accuracy
Consejo

Un modelo entrenado con transformaciones personalizadas necesita que se pase el ClassificationValidator correspondiente a val(); de lo contrario, la validación recortará las imágenes que el entrenamiento había redimensionado.

Predecir#

Usa un modelo YOLO26n-cls entrenado para ejecutar predicciones en imágenes.

Ejemplo
from ultralytics import YOLO

# Cargar un modelo
model = YOLO("yolo26n-cls.pt")  # carga un modelo oficial
model = YOLO("path/to/best.pt")  # cargar un modelo personalizado

# Haz predicciones con el modelo
results = model("https://ultralytics.com/images/bus.jpg")  # haz una predicción sobre una imagen

# Accede a los resultados
for result in results:
    top1 = result.probs.top1  # ID de la clase predicha con mayor puntuación
    top1_conf = result.probs.top1conf  # confianza de la predicción con mayor puntuación
    top1_name = result.names[top1]  # nombre de la clase predicha con mayor puntuación

Consulta todos los detalles del modo predict en la página de predicción.

Resultados de salida#

La clasificación de imágenes devuelve un objeto Results por imagen. El campo principal de predicción es result.probs, que contiene el vector de probabilidades de clase y métodos auxiliares para las predicciones principales.

AtributoTipoFormaDescripción
result.probsProbs(C,)Probabilidades de clase.
result.probs.datatorch.float32(C,)Probabilidad por clase.
result.probs.top1int()ID de la clase principal.
result.probs.top1conftorch.float32()Nivel de confianza principal.
result.probs.top5list[int](<=5)ID de las 5 clases principales.

Para consultar los campos Results específicos de cada tarea, visita la sección resultados de predicción por tarea.

Exportar#

Exporta un modelo YOLO26n-cls a otro formato, como ONNX, CoreML, etc.

Ejemplo
from ultralytics import YOLO

# Cargar un modelo
model = YOLO("yolo26n-cls.pt")  # carga un modelo oficial
model = YOLO("path/to/best.pt")  # cargar un modelo personalizado

# Exporta el modelo
model.export(format="onnx")

Los formatos de exportación disponibles para YOLO26-cls se muestran en la tabla siguiente. Puedes exportar a cualquier formato con el argumento format, por ejemplo, format='onnx' o format='engine'. Puedes hacer predicciones o validar directamente con los modelos exportados, por ejemplo, yolo predict model=yolo26n-cls.onnx. Cuando termine la exportación, se mostrarán ejemplos de uso para tu modelo.

FormatoArgumento formatModeloMetadatosArgumentos
PyTorch-yolo26n-cls.pt✅-
TorchScripttorchscriptyolo26n-cls.torchscript✅imgsz, quantize, dynamic, nms, batch, device
ONNXonnxyolo26n-cls.onnx✅imgsz, quantize, dynamic, simplify, opset, nms, batch, data, fraction, device
OpenVINOopenvinoyolo26n-cls_openvino_model/✅imgsz, quantize, dynamic, nms, batch, data, fraction, device
TensorRTengineyolo26n-cls.engine✅imgsz, quantize, dynamic, simplify, opset, workspace, nms, batch, data, fraction, device
CoreMLcoremlyolo26n-cls.mlpackage✅imgsz, dynamic, quantize, nms, batch, device
Apple Core AIcoreaiyolo26n-cls.aimodel✅imgsz, batch, quantize
TF SavedModelsaved_modelyolo26n-cls_saved_model/✅imgsz, quantize, opset, nms, batch, data, fraction, device
TF GraphDefpbyolo26n-cls.pb❌imgsz, opset, batch, device
TF Edge TPUedgetpuyolo26n-cls_edgetpu.tflite✅imgsz, quantize, opset, data, fraction, device
LiteRTlitertyolo26n-cls.tflite✅imgsz, quantize, batch, data, fraction, device
PaddlePaddlepaddleyolo26n-cls_paddle_model/✅imgsz, batch, device
MNNmnnyolo26n-cls.mnn✅imgsz, batch, dynamic, quantize, simplify, opset, nms, device
NCNNncnnyolo26n-cls_ncnn_model/✅imgsz, quantize, device
IMX500imxyolo11n_imx_model/✅imgsz, quantize, data, fraction, nms, device
RKNNrknnyolo26n-cls_rknn_model/✅imgsz, batch, name, quantize, simplify, opset, data, fraction, device
ExecuTorchexecutorchyolo26n-cls_executorch_model/✅imgsz, quantize, batch, data, fraction, device
Axeleraaxelerayolo26n-cls_axelera_model/✅imgsz, batch, quantize, data, fraction, device
DEEPXdeepxyolo26n-cls_deepx_model/✅imgsz, quantize, simplify, opset, data, optimize, device
Qualcomm QNNqnnyolo26n-cls_qnn.onnx✅imgsz, batch, name, quantize, simplify, opset, data, fraction, device
Hailohailoyolo26n-cls_hailo_model/✅imgsz, name, quantize, data, fraction, simplify, conf, iou, device
Huawei Ascendascendyolo26n-cls_ascend_model/✅imgsz, batch, name, quantize, opset, simplify, nms, device
AMD Xilinxxilinxyolo26n-cls_xilinx_model/✅imgsz, name, quantize, data, fraction, opset, simplify, device
NXPnxpyolo26n-cls_nxp.tflite✅imgsz, quantize, data, fraction, device
ST Edge AIstedgeaiyolo26n-cls_stedgeai.onnx✅imgsz, quantize, data, fraction, opset, simplify, device

nms=None genera salidas sin procesar de forma predeterminada para aplicar NMS externamente (YOLO26 mantiene su cabezal sin NMS en Axelera). Establece nms=False para seleccionar un cabezal sin NMS disponible; los formatos no compatibles recurren a su ruta de salida nativa. Las entradas nms anteriores identifican los formatos que pueden integrar NMS con nms=True.

Consulta todos los detalles de export en la página Exportar.

Preguntas frecuentes#

  • Los modelos YOLO26, como yolo26n-cls.pt, están diseñados para la clasificación de imágenes eficiente. Asignan una única etiqueta de clase a toda una imagen, junto con una puntuación de confianza. Esto resulta útil cuando basta con conocer la clase de una imagen y no necesitas saber la ubicación o la forma de los objetos que contiene.

  • Puedes entrenar un modelo YOLO26 con comandos de Python o de la CLI. Por ejemplo, para entrenar un modelo yolo26n-cls con el conjunto de datos MNIST160 durante 100 épocas y con un tamaño de imagen de 64:

    Ejemplo
    from ultralytics import YOLO
    
    # Cargar un modelo
    model = YOLO("yolo26n-cls.pt")  # Carga un modelo preentrenado (recomendado para el entrenamiento)
    
    # Entrenar el modelo
    results = model.train(data="mnist160", epochs=100, imgsz=64)

    Visita la página de Configuración para ver más opciones de configuración.

  • Puedes explorar, probar y descargar modelos de clasificación YOLO26 preentrenados en Ultralytics Platform. Modelos como yolo26n-cls.pt, yolo26s-cls.pt y yolo26m-cls.pt están preentrenados con el conjunto de datos ImageNet y se descargan automáticamente la primera vez que se utilizan.

  • Para validar la precisión de un modelo entrenado en un conjunto de datos como MNIST160, puedes usar los siguientes comandos de Python o CLI:

    Ejemplo
    from ultralytics import YOLO
    
    # Cargar un modelo
    model = YOLO("path/to/best.pt")  # carga el modelo entrenado
    
    # Valida el modelo
    metrics = model.val()  # no se necesitan argumentos; usa el conjunto de datos y la configuración del entrenamiento
    metrics.top1  # top1 accuracy
    metrics.top5  # top5 accuracy

    Para obtener más información, visita la sección Validar.

  • Puedes exportar un modelo YOLO26 entrenado a varios formatos mediante comandos de Python o CLI. Por ejemplo, para exportar un modelo al formato ONNX:

    Ejemplo
    from ultralytics import YOLO
    
    # Cargar un modelo
    model = YOLO("yolo26n-cls.pt")  # carga el modelo entrenado
    
    # Exporta el modelo a ONNX
    model.export(format="onnx")

    Para consultar las opciones de exportación detalladas, visita la página Exportar.

Comentarios207

Haz una pregunta, comparte tus resultados o ayuda a otra persona.

  1. hellotenetGitHub

    According to my observation, during the training of YOLO for classification tasks, it randomly selects a portion of data from the test set to evaluate the accuracy of this round. If I want to use all the data in the test set for testing in each epoch and provide top_acc1 and top_acc5, which parameter should I add in the yaml file?

    1
  2. eladoni1GitHub(editado)

    Hi there! Quick briefing - the images are of size Width = 1216, Height = 292.

    TL;DR - After debug into train.py & predict.py, realized the images are resized + cropped strangely from what I gathered in the docs/forums, even with specified imgsz and rect=True.

    Full explanation -

    Training was : model.train(data=..., imgsz=1216, rect=True)

    Expectations : Was expecting the dataset images to be resized BY THE LARGER side, so 1216 will remain 1216, and so is the smaller side.

    (Even when I tried to do imgsz as a tuple - (h,w) aka (292,1216), a warning showed up, telling me the imgsz will be automatically selected to 1216, as it should have been an integer.. not a tuple.)

    What actually happened : What happened is the smaller side (height, 292) resized to 1216, then the original, bigger side (width, 1216), was resized to 5063 (same scalar the smaller side was multiplied).

    Meaning, what previously was image of size 1216x292, turned into image of size 5063x1216 (This is due to Resize transformation of pytorch)

    And its not over - the data taken was only the INNER 1216x1216 - less than 1/4 of the actual data.. (This is due to CenterCrop transformation of pytorch)

    So, basically, I am asking - Is there a quote from the docs I have missed that said classify must be square sized? Or is there another way (except from padding to 1216x1216 - adding 924 black pixels to the height), to make classify work with rectangular images?

    Thank you so much, and sorry if I wasnt clear.. hopefully it was.

    1
    1. Glenn JocherUltralytics

      Hey there! 🙌 Thanks for the detailed message.

      It seems like you've encountered an interesting scenario with image sizes and the rect=True parameter. In YOLOv8, when using rectangular training (rect=True), the model does expect square input sizes for simplicity and efficiency. The behavior you've noticed, where the smaller dimension gets resized to match the specified imgsz, and then the image is center-cropped, is by design for handling non-square inputs.

      For your specific case with rectangular images (1216x292), instead of resizing or padding the images manually to make them square, a recommended approach is to work with the default behavior but ensure your imgsz is set to the dimension that avoids data loss. If further customization is necessary, it might involve custom modifications to the data preprocessing pipeline.

      Unfortunately, there isn't a built-in switch for direct handling of non-square images without cropping or resizing as you described. This is an area for potential improvement in future updates!

      For now, perhaps consider either adjusting your pipeline to accommodate this behavior or implementing a custom preprocessing step that ensures your images maintain their aspect ratio without significant data loss. Padding might be an unavoidable step for cases like these.

      If there's anything else we can help with or if you have suggestions, feel free to reach out. Thanks for diving deep into this, and we're here to support your projects! 🚀

  3. lixjohnGitHub

    If I can train Yolo-cls mode to identify the level of object moisture content? Such as coal? Thanks.

    1. Glenn JocherUltralytics

      Yes, Ultralytics YOLO26 classification can work for coal moisture levels if you label them as discrete classes like dry, medium, and wet and train with the standard classification dataset format; if you need an exact moisture percentage rather than bins, that’s a regression problem, and if the visual differences are subtle you may need NIR/thermal imaging instead of RGB alone.

    2. lixjohnGitHub

      Thanks, I think classification is enough because we just need high level of coal moisture not exact percentage.

    3. Glenn JocherUltralytics

      Yes, that should be a good fit for classify; use a few clear bins like low, medium, and high, keep lighting/camera distance as consistent as possible, and start from a pretrained yolo26n-cls.pt model—if you’d like, share a sample class layout or training command here and we can sanity-check it.

    4. lixjohnGitHub

      Thank you very much, that makes sense. I just start project and not go to training phase yet, but I am happy to share sample and training command once I have to get more help on this.

    5. Glenn JocherUltralytics

      Sounds good—when you reach training, share your dataset folder structure and exact yolo classify train ... command here, and we’ll help sanity-check it against the classification guide.

    6. lixjohnGitHub

      Thank you! I got another question related to segment, hopefully I can get your suggestion as well, it is a little bit tricky :-). The coal is very small ( around 10mm one piece), but we have identify them in a truck (10m*3m) in order to ensure the truck carry right size of coal, one picture for one truck, it means there are huge member of small coal in one picture, I don't have image resolution yet, but I suppose it is clear enough to show details for each coal, my questions are

      1. how to label data before training, I can use SAM to label each coal but I have to say it is impossible to label all coal in picture because the coal member is huge, if I can just label some of them in pictures? I think we can use around 300 pictures to start training, but not all coal labeled in one picture. Does it will impact accuracy of model?
      2. Is it possible for Yolo26-segment to identify each coal in this situation from algorithm and performance point of view? In one picture, the truck is big but carry large member of similar small objects (10mm vs. 10m). Any solution we can take for this situation?

      Apricate for any suggestions...

    7. Glenn JocherUltralytics

      Yes, but you should not leave visible coals unlabeled in training images because the model will treat them as background and recall will drop; for this kind of dense small-object problem, the practical approach is to split each truck image into smaller overlapping tiles, fully annotate every visible coal inside the selected tiles, and train on those crops rather than the full frame. If each coal is still only a few pixels after tiling, segment will struggle and detect or even patch-level classify may be a better fit, and our data collection and annotation guide is a good reference—feel free to post one sample image here when you have it and we can suggest the best labeling strategy.

  4. jiangshouyeGitHub

    如何指定分类标签的次序呢?训练完成后,如何获得模型里的标签顺序呢?

    1. Glenn JocherUltralytics

      分类任务里类别索引顺序是从你的数据集目录结构推断出来的(通常按 train/ 下各类别子文件夹名称排序生成),如果你想“指定顺序”,最稳妥的方法是把类别文件夹命名成可排序的形式(例如 00_cat/、01_dog/);训练完成后直接读取模型里的 model.names 就能看到标签到索引的对应关系,参考 图像分类文档。

      from ultralytics import YOLO
      
      model = YOLO("runs/classify/train/weights/best.pt")
      print(model.names)  # {0: '00_cat', 1: '01_dog', ...}
  5. serenamansurGitHub

    hello, hope all is well. @glenn-jocher i would like to ask you few questions

    1- in classification task , when finetuning YOLOv11m-cls, is it enough to set "autoaugment" for data augmentation or i can use it with other augmentation args such as scale, hsv_v, hsv_s ??


    2- i will give you an example on this one, lets say i am training with epoch = 50 with patience = 8 and close_mosaic i set to = 5 and early stopping occured at epoch 20/50.

    Is close_mosaic applied here or is it set from the start that it should be disabled at epoch 45? and if it was disabled before epoch 20, how does this affect the result and validation?

    Thank you in advance!!

    1. Glenn JocherUltralytics

      For classify, auto_augment can be used together with other augmentation args like scale and hsv_h/hsv_s/hsv_v (they target different parts of the classification transform pipeline), but since policies like randaugment are already quite strong you’ll usually want to keep extra color jitter mild and tune one knob at a time; see the Image Classification docs and the classification transforms/augmentations reference.

      Also note close_mosaic is a mosaic-specific setting for detection-style tasks and isn’t used for *-cls training, so it won’t affect YOLO classification; if you meant detection, close_mosaic=5 disables mosaic only for the last 5 scheduled epochs (i.e., starting at epochs - close_mosaic), so if early stopping ends at epoch 20/50 that “close mosaic” switch never occurs.

    2. serenamansurGitHub

      Hello @glenn-jocher, Thank you for your previous reply, it was very helpful. I have another question related to class imbalance in computer vision tasks. In many ML setups, when classes are imbalanced, we apply per-class loss weights so that rare classes contribute more to the overall training loss and are not ignored during optimization.

      For example, if I have 3 detection classes with instance counts like:

      class0 = 30 instances class1 = 500 instances class2 = 700 instances

      By default, all classes contribute equally (weight = 1), but ideally I’d like to assign a higher weight to the rare class (class0).


      My question: Is there a supported way in Ultralytics YOLO to apply per-class weights for the detection classification loss or bounding boxes loss (cls loss component)? If yes, where can I access and modify those weights?? Thank you in advance!

    3. Glenn JocherUltralytics

      At the moment Ultralytics YOLO doesn’t have a supported “per-class loss weight” hook for detection via data.yaml/CLI (for cls_loss or box losses); the only built-in knobs are the overall loss term weights like cls, box, and dfl in the configuration settings. If you need true per-class weighting you’ll have to implement it in a custom trainer/loss (the core detection loss lives in the loss implementation and can be swapped via the advanced customization guide); otherwise the practical workaround is to rebalance the dataset by oversampling/adding/augmenting the rare-class examples.

    4. maxexplorerGitHub

      Hi Ultralytics team, YOLO-classify always outputs class probabilities (probs) that sum to 1, even for completely unknown or out-of-distribution (OOD) inputs. As a result, unknown objects can be classified with high confidence (e.g., 0.8), and using a threshold does not always filter them. Are there any recommended strategies, best practices, or built-in options to reduce false high-confidence predictions on unknown data?
      Would combining classify with detect help in such cases? Thanks!

    5. Glenn JocherUltralytics

      You’re right: *-cls uses a softmax-style head, so the returned probs are normalized to sum to 1 by design, which means “unknown/OOD” will still get forced into a known class with some confidence; there isn’t a built-in OOD/reject option in Ultralytics YOLO classification today (beyond simple thresholding). A practical approach is to (a) add an explicit unknown (or “background”) class with diverse negative examples and train for rejection, and/or (b) run a detection model first and only classify detected crops—if nothing is detected you can safely treat the frame as unknown; see the Ultralytics YOLO image classification docs for the standard classify pipeline.

      If you want a lightweight post-process, you can gate predictions using max-softmax and an uncertainty score like entropy (often works better than max-prob alone):

      from ultralytics import YOLO
      import torch
      
      model = YOLO("yolo26n-cls.pt")
      results = model("path/to/img.jpg")
      
      r = results[0]
      p = r.probs.data  # class probabilities (tensor)
      max_p = float(p.max())
      entropy = float(-(p * (p + 1e-12).log()).sum())
      
      is_unknown = (max_p < 0.6) or (entropy > 2.5)  # tune on a held-out OOD set
  6. sagargwalGitHub

    is thier a way to use grad cam or any interpretability? method on yolov11-cls model

  7. theoderickdayGitHub

    Hi I have a simple bit of code based on your example above and using a model I created myself. It works fine at identifying birds and I get output of the form:-

    0: 640x640 nut_hatch 0.30, house_sparrow 0.18, bluetit 0.17, cat 0.13, robin 0.12, 273.5ms Speed: 29.8ms preprocess, 273.5ms inference, 0.1ms postprocess per image at shape (1, 3, 640, 640) 2025-10-27.jpg

    I use the following lines to pass data from the Raspberry PI camera to the model:-

    frame = picam2.capture_array() results = model(frame)

    what I want to do is extract the probabilities from the results structure so I can save specific images when the prediction is over say 80%

    Any ideas?

  8. wdcs-fenilpatelGitHub

    @glenn-jocher

    I'm encountering a strange and seemingly unpredictable issue during inference with a YOLO classification model (using ultralytics package). The confidence score for every prediction is always exactly 0.5, regardless of the input image.

    All inference results return confidence = 0.5.

    The predicted class index might vary, but the confidence is always 0.5.

    This issue starts unpredictably—everything works fine for a while, then suddenly this behavior appears.

    Sometimes, simply rebooting the system resolves the issue temporarily, but it may reappear.

  9. Ranjith3436GitHub

    I am trying to train classification model. why you prefer you architecture. why don't you move transformer architecture? is it capable than ViT.

  10. wowosoGitHub

    What if we can't obtain metrics such as F1 score, mAP, recall, and precision in image classification tasks?