Load semantic segmentation datasets. All files under "gt_root" with "gt_ext" extension are treated as ground truth annotations and all files

Makes a donut shaped binary mask.

Matrix NMS for multi-class masks. See: https://arxiv.org/pdf/2003.10152.pdf for more details. Args: seg_masks (Tensor): shape: [

Merge detection results from different branches of TridentNet. Return detection results by applying non-maximum suppression (NMS) on bounding

Returns: (H, W) array, a mask for all pixels that have a prediction

Restrict angles to the range of [-180, 180) degrees

Pause the timer.

Rearrange the tensor layout from the network output, i.e.: list[Tensor]: #lvl tensors of shape (N, A x K, Hi, Wi) to per-image prediction

Normalize, pad and batch the input images.

Normalize, pad and batch the input images.

Normalize, pad and batch the input images.

Normalize, pad and batch the input images.

Normalize, pad and batch the input images.

Normalize, pad and batch the input images.

Normalize, pad and batch the input images.

r""" Makes a function retry itself after encountering pytorch's CUDA OOM error. It will first retry after calling `torch.cuda.empty_cache(

We use the "DefaultRunner" which contains a number pre-defined logic for standard training workflow. They may not work for you, especially if

Set the random seed for the RNG in torch, numpy and python. Args: seed (int): if None, will use a strong random seed. Returns:

Produce additional matches for predictions that have only low-quality matches. Specifically, for each ground-truth G find the set of

Compute the smooth L1 loss for box regression. Returns: scalar Tensor

Perform the appropriate action at the given iteration. Args: iteration (int): the current iteration, ranged in [0, max_i

Return `num_samples` (or fewer, if not enough found) random samples from `labels` which is a mixture of positives & negatives. It will tr

Add a new scalar value produced at certain iteration. If the length of the buffer exceeds self._max_length, the oldest element will b

Args: results (OrderedDict[dict]): task_name -> {metric -> score} Returns: bool: whether the verification succeeds or not

Helper function to Accumulate() Considers the evaluation results applicable to a particular category, area range, and max_detections parameter setting

Helper function to Accumulate() Considers the evaluation results applicable to a particular category, area range, and max_detections parameter setting

Loads predictions from a COCO-style results file.

Helper function to Accumulate() Compute a precision recall curve given a sorted list of detected instances encoded in evaluations, evaluation_indices,

Helper function to Accumulate() Compute a precision recall curve given a sorted list of detected instances encoded in evaluations, evaluation_indices,

For each IOU threshold, greedily match each detected instance to a ground truth instance (if possible) and store the results

For each IOU threshold, greedily match each detected instance to a ground truth instance (if possible) and store the results

Sort detections from highest score to lowest, such that detection_instances[detection_sorted_indices[t]] >= detection_instances[detection_sorted_indic

Sort detections from highest score to lowest, such that detection_instances[detection_sorted_indices[t]] >= detection_instances[detection_sorted_indic

Partition the ground truth objects based on whether or not to ignore them based on area

Partition the ground truth objects based on whether or not to ignore them based on area

A function that does union of two sets of x and y (uses union by rank)

A function that does union of two sets of x and y (uses union by rank)

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

See: https://arxiv.org/pdf/1807.11164.pdf Args: num_blocks (int): the number of blocks in this stage. in_cha

Args: block_args (EasyDict): block args, see: class: `EfficientNet`. global_params (EasyDict): global args, see: clas

Build basic STEM for Dynamic Network. Args: norm (str or callable): a callable that takes the number of c

Args: box2box_transform (Box2BoxTransform/Box2BoxTransformRotated): box2box transform instance for proposal-to-de

Args: module (nn.Module): torch module to profile. enabled (bool): whether enable profile or not. use_cud

Args: name (str): the name of this registry

Args: model (nn.Module): model. save_dir (str): a directory to save and find checkpoints. save_to_disk (b

Deformable convolution. Arguments are similar to :class:`Conv2D`. Extra arguments: Args: deformable_groups (int

Create the criterion. Parameters: cfg: config matcher: module able to compute a matching between targets and proposal