Accumulate per image evaluation results and store the result in self.eval :param p: input params for evaluation :return: None

A utility function to find set of an element i (uses path compression technique)

A utility function to find set of an element i (uses path compression technique)

used for Centernet

Returns: The box centers in a Nx2 array of (x, y).

Compute iou loss of type ['iou', 'giou', 'linear_iou'] Args: inputs (tensor): pred values targets (tensor): target values

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

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

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

Forward logic of the keypoint prediction branch. Args: features (list[Tensor]): #level input features for keypoint predi

Uniform float random number between low and high.

RoiAlign with scale 1.0 and 0 sample ratio.

coords should be a N * 2 array-like, containing N couples of (x, y) points

img should be a numpy array, formatted as Height * Width * Nchannels

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

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 di

Create a parser with some common arguments used by cvpods users. Returns: argparse.ArgumentParser:

r""" Dice loss defined in the V-Net paper as: Loss_dice = 1 - D 2 * sum(p_i * g_i) D = ------------------------------

Returns: Boxes: tight bounding boxes around bit masks.

Args: main_func: a function that will be called by `main_func(*args)` num_machines (int): the total number of machines ma

Load from the given checkpoint. When path points to network file, this function has to be called on all ranks. Args:

merge all key and values of config as BaseConfig's attributes. Note that kwargs will override values in config if they have the same

Count parameters of a model and its submodules. Args: model: a torch module Returns: dict (str-> int): the key is eithe

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

Remove the field called `name`.

If `resume==True` and `cfg.OUTPUT_DIR` contains the last checkpoint (defined by a `last_checkpoint` file), resume from the file. Resu

Args: cfg (config dict): model (nn.Module): evaluators (list[DatasetEvaluator] or None): if None, will ca

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

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

Forward logic of the box prediction branch. If `self.train_on_pred_boxes is True`, the function puts predicted boxes in the `prop

With 50% prob, negate the value

RoiAlignRotated with scale 1.0 and 0 sample ratio.

Add a region inside of which background detections should be ignored. You can use these to mark a region that has deliberately been l

Register an image name/path with an image ID.

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

Apply the transform on a full-image segmentation. By default will just perform "apply_image". Args: segmentation

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

Create a Dynamic Backbone from config. Args: cfg: a config dict Returns: backbone (Backbone): backbone module, must be

Args: cfg: a cvpods `Config` instance. Returns: bifpn (Backbone): backbone module, must be a subclass of :class:

Args: cfg: a cvpods CfgNode Returns: backbone (Backbone): backbone module, must be a subclass of :class:`Backbone`.

Build an optimizer with clip and LARS wraper from config.

Create a VGG instance from config. Returns: VGG: a :class:`VGG` instance.

Create a list of :class:`TransformGen` from config. Transform list is a list of tuple which includes Transform name and parameters. Args

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

Create a seuence Conv2d->BatchNorm2d->LeakyReLu layer.

Converts dataset into COCO format and saves it to a json file. dataset_name must be registered in DatasetCatalog and in cvpods's standard for

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

Perform some basic common setups at the beginning of a job, including: 1. Set up the cvpods logger 2. Log basic information about enviro

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

Run per image evaluation on given images and store results (a list of dict) in self.evalImgs :return: None

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

Generalized IoU from https://giou.stanford.edu/ The boxes should be in [x0, y0, x1, y1] format Returns a [N, M] pairwise matrix, where N

Generate Mask Logits by Mask Branch and Dynamic Mask Head.

Args: image_size (tuple): height, width Returns: crop_size (tuple): height, width in absolute pixels

Returns: HistoryBuffer: the scalar history for name

Run model on the data_loader and evaluate the metrics with evaluator. The model will be used in eval mode. Args: model (nn.Modul

Prepare some proposals to be used to train the ROI heads. It performs box matching between `proposals` and `targets`, and assigns

Paste a set of masks that are of a fixed resolution (e.g., 28 x 28) into an image. The location, height, and width for pasting each mask is d

Returns: list[Tensor]: A list of Tensors of predicted class-specific or class-agnostic boxes for each image. Elem

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

Calculate and round number of channels based on depth multiplier. Args: channels (int): base number of channels. global_para

Scale the box with horizontal and vertical scaling factors

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

The following attributes are parsed from config: conv_dim: the output dimension of the conv layers fc_dim: the featur

Add a data object to this collection. You should use one of the below functions instead.

Args: classes (list[int] or None): scores (list[float] or None): class_names (list[str] or None): Returns: l

Dump a Markdown file that records the model evaluation metrics and corresponding scores to the current working directory. Args:

Returns: str: module name of the caller tuple: a hashable key to be used to identify different callers

Forward logic of the mask prediction branch. Args: features (list[Tensor]): #level input features for mask prediction