easycv.core.evaluation.custom_cocotools package

Submodules

easycv.core.evaluation.custom_cocotools.cocoeval module

class easycv.core.evaluation.custom_cocotools.cocoeval.COCOeval(cocoGt=None, cocoDt=None, iouType='segm', sigmas=None)

Bases: object

__init__(cocoGt=None, cocoDt=None, iouType='segm', sigmas=None)

Initialize CocoEval using coco APIs for gt and dt :param cocoGt: coco object with ground truth annotations :param cocoDt: coco object with detection results :param iouType: type of iou to be computed, bbox for detection task,

segm for segmentation task

Parameters

sigmas – keypoint labelling sigmas.

Returns

None

evaluate()

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

computeIoU(imgId, catId)

computeOks(imgId, catId)

evaluateImg(imgId, catId, aRng, maxDet)

perform evaluation for single category and image :param imgId: image id, string :param catId: category id, string :param aRng: area range, tuple :param maxDet: maximum detection number

Returns

dict (single image results)

accumulate(p=None)

Accumulate per image evaluation results and store the result in self.eval :param param p: input params for evaluation

Returns

None

summarize()

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

summarize_per_category()

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

filter_annotations(annotations, catIds)

makeplot(recThrs, precisions, name, save_dir=None)

analyze()

Analyze errors

class easycv.core.evaluation.custom_cocotools.cocoeval.Params(iouType='segm')

Bases: object

Params for coco evaluation api

setDetParams()

setKpParams()

__init__(iouType='segm')

Initialize self. See help(type(self)) for accurate signature.