Access an attribute and return its value if exists. Otherwise return default.

Efficient version of torch.cat that avoids a copy if there is only a single element in a list

Add a scalar `value` to the `HistoryBuffer` associated with `name`. Args: smoothing_hint (bool): a 'hint' on whether thi

Concatenates a list of Boxes into a single Boxes Arguments: boxes_list (list[Boxes]) Returns: Boxes

Get a copy of the default config. Returns: a detectron2 CfgNode instance.

Return intersection-over-union (Jaccard index) of boxes. Both sets of boxes are expected to be in (x_center, y_center, width, height, an

Args: box: can be a k-tuple, k-list or an Nxk array/tensor, where k = 4 or 5 from_mode, to_mode (BoxMode) Re

This function is a replacement of .to(another_device) such that it allows the casting to be traced properly by explicitly calling the underly

Clone the Boxes. Returns: Boxes

Initialize the detectron2 logger and set its verbosity level to "INFO". Args: output (str): a file name or a directory to save log.

Returns: The :class:`EventStorage` object that's currently being used. Throws an error if no :class`EventStorage` is currently en

Args: norm (str or callable): Returns: nn.Module or None: the normalization layer

Clip (in place) the boxes by limiting x coordinates to the range [0, width] and y coordinates to the range [0, height]. Args

Args: boxes (Boxes, RotatedBoxes or ndarray): either a :class:`Boxes`, or an Nx4 numpy array of XYXY_ABS format f

Args: filepath (str): a string that contains the absolute path, including the file name, where the visualized ima

Returns: ndarray: the visualized image of shape (H, W, 3) (RGB) in uint8 type. The shape is scaled w.r.t the input

Set multiple metadata with kwargs.

Build the whole model architecture, defined by ``cfg.MODEL.META_ARCHITECTURE``. Note that it does not load any weights from ``cfg``.

Given two lists of rotated boxes of size N and M, compute the IoU (intersection over union) between __all__ N x M pairs of boxes. The

Set the field named `name` to `value`. The length of `value` must be the number of instances, and must agree with other exist

Computes the area of all the boxes. Returns: torch.Tensor: a vector with areas of each box.

Build a backbone from `cfg.MODEL.BACKBONE.NAME`. Returns: an instance of :class:`Backbone`

Args: name (str): the name that identifies a dataset, e.g. "coco_2014_train". func (callable): a callable which takes

Args: start_iter, max_iter (int): See docs above

Args: shape: (h, w) tuple or a int interp: PIL interpolation method

Args: tensors: a tuple or list of `torch.Tensors`, each of shape (Hi, Wi) or (C_1, ..., C_K, Hi, Wi) where K >= 1

Remove the field called `name`.

User should call this function at the beginning of each iteration, to notify the storage of the start of a new iteration. The

Similar to `build_detection_train_loader`. But this function uses the given `dataset_name` argument (instead of the names in cfg), and us

Build a proposal generator from `cfg.MODEL.PROPOSAL_GENERATOR.NAME`. The name can be "PrecomputedProposals" to use no proposal generator.

Crop each bitmask by the given box, and resize results to (mask_size, mask_size). This can be used to prepare training targets for Ma

Resize the output instances. The input images are often resized when entering an object detector. As a result, we often need the outputs

Log only for the first n times. Args: lvl (int): the logging level msg (str): n (int): name (str): name of t

Find boxes that are non-empty. A box is considered empty, if either of its side is no larger than threshold. Returns:

Given two lists of boxes of size N and M, compute the IoU (intersection over union) between __all__ N x M pairs of boxes. The box ord

If `resume==True`, and last checkpoint exists, resume from it. Otherwise, load a model specified by the config. Args:

A data loader is created by the following steps: 1. Use the dataset names in config to query :class:`DatasetCatalog`, and obtain a list of d

Create a small table using the keys of small_dict as headers. This is only suitable for small dictionaries. Args: small_dict (di

Draw annotations/segmentaions in Detectron2 Dataset format. Args: dic (dict): annotation/segmentation data of one image,

Compute the default losses for box head in Fast(er) R-CNN, with softmax cross entropy loss and smooth L1 loss. Returns:

Performs non-maximum suppression (NMS) on the rotated boxes according to their intersection-over-union (IoU). Rotated NMS iteratively re

Create a grayscale version of the original image. The colors in masked area, if given, will be kept.

RoiAlign with scale 1.0 and 0 sample ratio.

Same as torchvision.ops.boxes.batched_nms, but safer.

Returns: str: a yaml string representation of the config

Get box regression transformation deltas (dx, dy, dw, dh) that can be used to transform the `src_boxes` into the `target_boxes`. That

Returns: bool: whether the field called `name` exists.

Arguments: box_cls, box_delta: Same as the output of :meth:`RetinaNetHead.forward` anchors (list[list[Boxes]]): a lis

Args: polygons (list[ndarray]): each array has shape (Nx2,) height, width (int) Returns: ndarray: a bool mask of shap

Add an `img_tensor` to the `_vis_data` associated with `img_name`. Args: img_name (str): The name of the image to put in

Put multiple scalars from keyword arguments. Examples: storage.put_scalars(loss=my_loss, accuracy=my_accuracy, smoothin

Args: rgb (bool): whether to return RGB colors or BGR colors. maximum (int): either 255 or 1 Returns: ndarray: a vec

Given a list of N Instances (for N images), each containing a `gt_classes` field, return a list of Instances that contain only instances with

Args: stem (nn.Module): a stem module stages (list[list[ResNetBlock]]): several (typically 4) stages,

Depending on the brightness_factor, gives a lighter or darker color i.e. a color with less or more saturation than the original color

RoiAlignRotated with scale 1.0 and 0 sample ratio.

Apply transformation `deltas` (dx, dy, dw, dh) to `boxes`. Args: deltas (Tensor): transformation deltas of shape (N, k*4

Computes the area of all the boxes. Returns: torch.Tensor: a vector with areas of each box.

Performs non-maximum suppression in a batched fashion. Each index value correspond to a category, and NMS will not be applied between el

Built an anchor generator from `cfg.MODEL.ANCHOR_GENERATOR.NAME`.

Build a box head defined by `cfg.MODEL.ROI_BOX_HEAD.NAME`.

Build an optimizer from config.

Build ROIHeads defined by `cfg.MODEL.ROI_HEADS.NAME`.

Clip (in place) the boxes by limiting x coordinates to the range [0, width] and y coordinates to the range [0, height]. For

Draw instance-level prediction results on an image. Args: predictions (Instances): the output of an instance detection/s

Args: x_data (list[int]): a list containing x values of all the points being drawn. Length of list should match t

Args: input: NCHW images rois: Bx5 boxes. First column is the index into N. The other 4 columns are xyxy.