Print information about the annotation file. :return:

generate loss and postprocess

This function will load the weights from a file. If the file extension is ``.pt``, it will be considered as a `pytorch pickle file <http://py

Create an annotation from a string or other box object. Args: obj (Box or string, optional): Bounding box object to copy attribu

Get ann ids that satisfy given filter conditions. default skips that filter :param imgIds (int array) : get anns for given imgs

Load anns with the specified ids. :param ids (int array) : integer ids specifying anns :return: anns (object array) : l

Compute IOU between all boxes from ``boxes1`` with all boxes from ``boxes2``. Args: boxes1 (torch.Tensor): List of bounding boxes

Load anns with the specified ids. :param ids (int array) : integer ids specifying img :return: imgs (object array) : lo

This function will save the weights to a file. If the file extension is ``.pt``, it will be considered as a `pytorch pickle file <http://pyto

Creates a roidb from pre-computed proposals of a particular methods.

Add a rate to the engine. Rates are object attributes that automatically change according to the current batch number. Args:

all_boxes is a list of length number-of-classes. Each list element is a list of length number-of-images. Each of those list e

filtering parameters. default skips that filter. :param catNms (str array) : get cats for given cat names :param supNms (str

Get img ids that satisfy given filter conditions. :param imgIds (int array) : get imgs for given ids :param catIds (int array

Return the directory where experimental artifacts are placed. If the directory does not exist, it is created. A canonical path is built using

Compute the best match (largest overlap area) between a given detection and a list of annotations. Args: detection (brambox.boxes.detect

Visual debugging of detections.

This function randomly changes the the input dimension of the dataset.

Load pre-computed proposals in the format provided by Jan Hosang: http://www.mpi-inf.mpg.de/departments/computer-vision-and-multimoda

Change coordinates of an annotation, according to the previous letterboxing

Update rates according to batch size. |br| This function gets automatically called every batch, and should generally not be called by the use

This function will compute a new size and update it on the next mini_batch. Args: multiple (int or tuple, optional): value (or v

Create a bounding box from a string or other detection object. Args: obj (Box or string, optional): Bounding box object to copy

Create a detection from a string or other box object. Args: obj (Box or string, optional): Bounding box object to copy attribute

Deserialize a detection file into a dictionary of detections

Compute Yolo loss.

Return the database of ground-truth regions of interest. This function loads/saves from/to a cache file to speed up future calls.

Return the database of ground-truth regions of interest. This function loads/saves from/to a cache file to speed up future calls.

Calculate the intersection area between two boxes. Args: a (brambox.boxes.box.Box): First bounding box b (brambox.boxes.box.Box)

Match detection results with gound truth and return true and false positive rates. This function will return a list of values as the true and fal

This function will recursively loop over all module children. Args: mod (torch.nn.Module, optional): Module to loop over; Defaul

This function gets called after every training epoch and decides if the training cycle continues. Return: Boolean: Whether are n

Serialize input dictionary of detections into one string

Yield input elements with given stride and offset.

Compute and display summary metrics for evaluation results. Note this functin can *only* be applied on the default parameter setting

Compare prediction boxes and ground truths, convert ground truths to network output tensors

Compare prediction boxes and ground truths, convert ground truths to network output tensors

Finds proposals that are inside crowd regions and marks them with overlap = -1 (for all gt rois), which means they will be excluded from

Loads COCO bounding-box instance annotations. Crowd instances are handled by marking their overlaps (with all categories) to -1. This

Load the indexes listed in this dataset's image set file.

Load image ids.

Load the indexes listed in this dataset's image set file.

Load image and bounding boxes info from XML file in the PASCAL VOC format.

Merge config dictionary a into config dictionary b, clobbering the options in b whenever they are also specified in a.

Non maximum suppression. Args: boxes (tensor): Bounding boxes of one image Return: (tensor): Pruned boxes

Prepare ._gts and ._dts for evaluation based on params :return: None