Interface for Python

Apply the resizing transform on rotated boxes. For details of how these (approximation) formulas are derived, please refer to :meth:`RotatedB

Args: dict: a detection dataset dict Returns: list[dict]: a list of dataset dicts, which con

Same input/output format as :meth:`GeneralizedRCNN.forward`

Args: match_quality_matrix (Tensor[float]): an MxN tensor, containing the pairwise quality between M ground-truth

Args: dataset_dict (dict): Metadata of one image, in Detectron2 Dataset format. Returns: dict: a format that

Args: original_image (np.ndarray): an image of shape (H, W, C) (in BGR order). Returns: predictions (dict):

An interface that wraps around a caffe2 model and mimics detectron2's models' input & output format. This is used to compare the caff

equivalent to FastRCNNOutputs.inference

equivalent to mask_head.mask_rcnn_inference

Returns: Boxes: Create a new :class:`Boxes` by indexing. The following usage are allowed: 1. `new_boxes = boxes

Access the individual image in its original size. Returns: Tensor: an image of shape (H, W) or (C_1, ..., C_K, H, W) whe

Create a new `Keypoints` by indexing on this `Keypoints`. The following usage are allowed: 1. `new_kpts = kpts[3]`: return

Returns: RotatedBoxes: Create a new :class:`RotatedBoxes` by indexing. The following usage are allowed: 1. `new

Args: item: an index-like object and will be used to index all the fields. Returns: If `item` is a string, r

Returns: BitMasks: Create a new :class:`BitMasks` by indexing. The following usage are allowed: 1. `new_masks =

Support indexing over the instances and return a `PolygonMasks` object. `item` can be: 1. An integer. It will return an obje

Args: cfg (CfgNode): model (GeneralizedRCNN): a GeneralizedRCNN to apply TTA on. tta_mapper (callable): t

Args: weights (4-element tuple): Scaling factors that are applied to the (dx, dy, dw, dh) deltas. In Fast R-CNN,

Args: weights (5-element tuple): Scaling factors that are applied to the (dx, dy, dw, dh, da) deltas. These are t

Args: output_size (int, tuple[int] or list[int]): output size of the pooled region, e.g., 14 x 14. If tuple or li

Args: thresholds (list): a list of thresholds used to stratify predictions into levels. labels (list)

The `__init__` method of any subclass can specify its own set of arguments.

The `__init__` method of any subclass should also contain these arguments. Args: in_channels (int): out_chan

Args: norm (str or callable): a callable that takes the number of channels and return a `nn.Module`, or a pre-def

Similar to :class:`BottleneckBlock`, but with deformable conv in the 3x3 convolution.

Args: norm (str or callable): a callable that takes the number of channels and return a `nn.Module`, or a pre-def

The following attributes are parsed from config: conv_dims: an iterable of output channel counts for each conv in the head

The following attributes are parsed from config: num_conv, num_fc: the number of conv/fc layers conv_dim/fc_dim: the

Args: input_size (int): channels, or (channels, height, width) num_classes (int): number of foreground classes

The following attributes are parsed from config: num_conv: the number of conv layers conv_dim: the dimension of the c

The following attributes are parsed from config: num_conv: the number of conv layers conv_dim: the dimension of the c

The following attributes are parsed from config: num_conv: the number of conv layers conv_dim: the dimension of the c

Args: box2box_transform (Box2BoxTransformRotated): :class:`Box2BoxTransformRotated` instance for anchor-proposal

Args: box2box_transform (Box2BoxTransform): :class:`Box2BoxTransform` instance for anchor-proposal transformation

Args: metadata (MetadataCatalog): image metadata.

Args: json_file (str): path to the json file. New data will be appended if the file exists. window_size (int): the wi

Args: log_dir (str): the directory to save the output events window_size (int): the scalars will be median-smoothed b

Args: max_iter (int): the maximum number of iterations to train. Used to compute ETA.

Args: start_iter (int): the iteration number to start with

Args: img (ndarray): an RGB image of shape (H, W, 3). scale (float): scale the input image

Args: img_rgb: a numpy array of shape (H, W, C), where H and W correspond to the height and width of the image re

Args: lst (list): a list which contains elements to produce. copy (bool): whether to deepcopy the element when produc

Args: dataset: an iterable. Each element must be a dict with keys "width" and "height", which will be used to bat

Args: size (int): the total number of data of the underlying dataset to sample from shuffle (bool): whether to shuffl

Args: dataset_dicts (list[dict]): annotations in Detectron2 dataset format. repeat_thresh (float): frequency threshol

Args: size (int): the total number of data of the underlying dataset to sample from

Args: sampler (Sampler): Base sampler. group_ids (list[int]): If the sampler produces indices in range [0, N),

Args: prob (float): probability of flip. horizontal (boolean): whether to apply horizontal flipping verti

Args: short_edge_length (list[int]): If ``sample_style=="range"``, a [min, max] interval from which to sample the

Args: crop_type (str): one of "relative_range", "relative", "absolute". See `config/defaults.py` for explanation.

Args: output_size (h, w): Dimensions of output image scale_range (l, h): Range of input-to-output size scaling factor

Args: intensity_min (float): Minimum augmentation intensity_max (float): Maximum augmentation

Args: intensity_min (float): Minimum augmentation intensity_max (float): Maximum augmentation

Args: intensity_min (float): Minimum augmentation (1 preserves input). intensity_max (float): Maximum augmentation (1

Args: scale (float): Standard deviation of principal component weighting.

Args: h, w (int): original image size new_h, new_w (int): new image size interp: PIL interpolation method

Each argument is a function that takes one argument: the trainer.

Args: warmup_iter (int): the number of iterations at the beginning to exclude from timing.

Args: writers (list[EventWriter]): a list of EventWriter objects period (int):

Args: optimizer (torch.optim.Optimizer): scheduler (torch.optim._LRScheduler)

Args: enable_predicate (callable[trainer -> bool]): a function which takes a trainer, and returns whether to enab

Args: eval_period (int): the period to run `eval_function`. eval_function (callable): a function which takes no argum

Args: period (int): the period this hook is run, or 0 to not run during training. The hook will always run in the