Partition the largest cluster into two sub-clusters.

Paste a list of masks that are of various resolutions (e.g., 28 x 28) into an image. The location, height, and width for pasting each mask is

Post-process the output boxes for TensorMask. The input images are often resized when entering an object detector. As a result, we often

Re-direct empty clusters.

Register a module. Args: module (:obj:`nn.Module`): Module to be registered.

Args: features (list[Tensor]): #lvl input features to ROIHeads proposals (list[Instances]): #image Instances, with th

Args: Same as in :meth:`__init__()`. Returns: fig (matplotlib.pyplot.figure): top level container for all th

Returns: tuple[str]: tasks that can be evaluated under the given configuration.

Returns: tuple[str]: tasks that can be evaluated under the given configuration.

Args: metrics_dict (dict): dict of scalar metrics

Computes the accuracy over the k top predictions for the specified values of k

Implement operator-level activations counting using jit. This is a wrapper of fvcore.nn.activation_count, that supports standard detection mo

Add config for additional category-related dataset options - category whitelisting - category mapping

Add config for densepose head.

Augment `proposals` with ground-truth boxes from `gt_boxes`. Args: Same as `add_ground_truth_to_proposals`, but with gt_boxes and pr

Add config for Panoptic-DeepLab.

Add config for PointRend.

Add config for TensorMask.

Add config for tridentnet.

Called after each iteration.

Called after the last iteration.

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

Map each box in `box_lists` to a feature map level index and return the assignment vector. Args: box_lists (list[Boxes] | list[R

When the config is defined for certain number of workers (according to ``cfg.SOLVER.REFERENCE_WORLD_SIZE``) that's different from the

Called before each iteration.

Called before the first iteration.

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

Build backbone.

Build DensePose data filter which selects data for training Args: cfg (CfgNode): configuration options Return: Callable

Build DensePose head based on configurations options Args: cfg (CfgNode): configuration options input_channels (int): input

Build DensePose loss based on configurations options Args: cfg (CfgNode): configuration options Return: An instance of D

Create an instance of DensePose predictor based on configuration options. Args: cfg (CfgNode): configuration options input_c

Create evaluator(s) for a given dataset. This uses the special metadata "evaluator_type" associated with each builtin dataset.

Returns: DatasetEvaluator or None It is not implemented by default.

Build a list of default hooks, including timing, evaluation, checkpointing, lr scheduling, precise BN, writing events. Retur

Build a instance embedding branch from `cfg.MODEL.INS_EMBED_HEAD.NAME`.

Build a keypoint head from `cfg.MODEL.ROI_KEYPOINT_HEAD.NAME`.

It now calls :func:`detectron2.solver.build_lr_scheduler`. Overwrite it if you'd like a different scheduler.

Returns: torch.optim.Optimizer: It now calls :func:`detectron2.solver.build_optimizer`. Overwrite it if you'd li

Build an RPN head defined by `cfg.MODEL.RPN.HEAD_NAME`.

Returns: iterable It now calls :func:`detectron2.data.build_detection_test_loader`. Overwrite it if you'd like a

Build a list of writers to be used. By default it contains writers that write metrics to the screen, a json file, and a tenso

Computes the AP under the precision recall curve.

For each location of the prediction `sem_seg_logits` we estimate uncerainty as the difference between top first and top second predicted

Concatenates a list of BitMasks into a single BitMasks Arguments: bitmasks_list (list[BitMasks]) Returns:

Delete all the stored histograms for visualization. This should be called after histograms are written to tensorboard.

Delete all the stored images for visualization. This should be called after images are written to tensorboard.

Clone the RotatedBoxes. Returns: RotatedBoxes

Filter LVIS instance segmentation annotations to remove all categories that are not included in COCO. The new json files can be used to evalu

Compresses quantized DensePose chart-based result Args: result (DensePoseChartResultQuantized): quantized DensePose chart-based resu

Construct the init_net from params dictionary

Convert DensePose predictor outputs to DensePoseResult using some registered converter. Does recursive lookup for base classes, so th

Apply some basic name conversion to names in C2 weights. It only deals with typical backbone models. Args: original_keys (list[s

Convert all boxes in `box_lists` to the low-level format used by ROI pooling ops (see description under Returns). Args: box_list

Map Caffe2 Detectron weight names to Detectron2 names. Args: weights (dict): name -> tensor Returns: dict: detectron2 n