randomroate by some change

def mapper(dataset_dict):
    dataset_dict = copy.deepcopy(dataset_dict)  # it will be modified by code below
    image = utils.read_image(dataset_dict["file_name"], format="BGR")    
    transform_list = [
                  
                     T.ResizeShortestEdge(short_edge_length=(640, 672, 704, 736, 768, 800), max_size=1333, sample_style='choice')
                     ,T.RandomRotation([10,15])
                
                      ]
    image, transforms = T.apply_transform_gens(transform_list, image)
    dataset_dict["image"] = torch.as_tensor(image.transpose(2, 0, 1).astype("float32"))

    
    #print('image_shape->',image.shape,image.shape[:2])

    annos = [
        utils.transform_instance_annotations(obj, transforms, image.shape[:2])
        for obj in dataset_dict.pop("annotations")
        if obj.get("iscrowd", 0) == 0
    ]

    instances = utils.annotations_to_instances(annos, image.shape[:2])
    dataset_dict["instances"] = instances
    #dataset_dict["instances"] = utils.filter_empty_instances(instances)
    return dataset_dict

this is my mapper for augmentation.
is T.RandomRotation([10,15]) happen every image? or by some change.
if it apply to every images. how should I apply it by only some percentage?

cc @vfdev-5

Start with the mapper shown in the issue and the documented behavior of T.RandomRotation. Check whether the transform is applied deterministically or probabilistically, and identify the documented way to configure a target application percentage; the issue names no repository file or test.