pipert.contrib.detection_demo.utils

Module Contents

Classes

FocalLoss

Functions

floatn(x, n=3)
load_classes(path)
model_info(model, report='summary')
labels_to_class_weights(labels, nc=80)
labels_to_image_weights(labels, nc=80, class_weights=np.ones(80))
coco_class_weights()
coco80_to_coco91_class()
weights_init_normal(m)
xyxy2xywh(x)
xywh2xyxy(x)
scale_coords(img1_shape, coords, img0_shape)
clip_coords(boxes, img_shape)
ap_per_class(tp, conf, pred_cls, target_cls)	Compute the average precision, given the recall and precision curves.
compute_ap(recall, precision)	Compute the average precision, given the recall and precision curves.
bbox_iou(box1, box2, x1y1x2y2=True, GIoU=False)
wh_iou(box1, box2)
compute_loss(p, targets, model)
build_targets(model, targets)
non_max_suppression(prediction, conf_thres=0.5, nms_thres=0.5)	Removes detections with lower object confidence score than ‘conf_thres’
get_yolo_layers(model)
print_model_biases(model)
strip_optimizer(f='weights/last.pt')
create_backbone(f='weights/last.pt')
coco_class_count(path='../coco/labels/train2014/')
coco_only_people(path='../coco/labels/val2014/')
select_best_evolve(path='evolve*.txt')
coco_single_class_labels(path='../coco/labels/train2014/', label_class=43)
print_mutation(hyp, results, bucket='')
fitness(x)
plot_one_box(x, img, color=None, label=None, line_thickness=None)
plot_wh_methods()
plot_images(imgs, targets, paths=None, fname='images.jpg')
plot_test_txt()
plot_targets_txt()
plot_evolution_results(hyp)
plot_results(start=0, stop=0)
plot_results_overlay(start=0, stop=0)
version_to_tuple(version)

pipert.contrib.detection_demo.utils.floatn(x, n=3)

pipert.contrib.detection_demo.utils.load_classes(path)

pipert.contrib.detection_demo.utils.model_info(model, report='summary')

pipert.contrib.detection_demo.utils.labels_to_class_weights(labels, nc=80)

pipert.contrib.detection_demo.utils.labels_to_image_weights(labels, nc=80, class_weights=np.ones(80))

pipert.contrib.detection_demo.utils.coco_class_weights()

pipert.contrib.detection_demo.utils.coco80_to_coco91_class()

pipert.contrib.detection_demo.utils.weights_init_normal(m)

pipert.contrib.detection_demo.utils.xyxy2xywh(x)

pipert.contrib.detection_demo.utils.xywh2xyxy(x)

pipert.contrib.detection_demo.utils.scale_coords(img1_shape, coords, img0_shape)

pipert.contrib.detection_demo.utils.clip_coords(boxes, img_shape)

pipert.contrib.detection_demo.utils.ap_per_class(tp, conf, pred_cls, target_cls)

Compute the average precision, given the recall and precision curves. Source: https://github.com/rafaelpadilla/Object-Detection-Metrics. # Arguments

tp: True positives (list). conf: Objectness value from 0-1 (list). pred_cls: Predicted object classes (list). target_cls: True object classes (list).

# Returns

The average precision as computed in py-faster-rcnn.

pipert.contrib.detection_demo.utils.compute_ap(recall, precision)

Compute the average precision, given the recall and precision curves. Source: https://github.com/rbgirshick/py-faster-rcnn. # Arguments

recall: The recall curve (list). precision: The precision curve (list).

# Returns

The average precision as computed in py-faster-rcnn.

pipert.contrib.detection_demo.utils.bbox_iou(box1, box2, x1y1x2y2=True, GIoU=False)

pipert.contrib.detection_demo.utils.wh_iou(box1, box2)

class pipert.contrib.detection_demo.utils.FocalLoss(loss_fcn, gamma=0.5, alpha=1, reduction='mean')

Bases: torch.nn.Module

forward(self, input, target)

pipert.contrib.detection_demo.utils.compute_loss(p, targets, model)

pipert.contrib.detection_demo.utils.build_targets(model, targets)

pipert.contrib.detection_demo.utils.non_max_suppression(prediction, conf_thres=0.5, nms_thres=0.5)

Removes detections with lower object confidence score than ‘conf_thres’ Non-Maximum Suppression to further filter detections. Returns detections with shape:

(x1, y1, x2, y2, object_conf, class_conf, class)

pipert.contrib.detection_demo.utils.get_yolo_layers(model)

pipert.contrib.detection_demo.utils.print_model_biases(model)

pipert.contrib.detection_demo.utils.strip_optimizer(f='weights/last.pt')

pipert.contrib.detection_demo.utils.create_backbone(f='weights/last.pt')

pipert.contrib.detection_demo.utils.coco_class_count(path='../coco/labels/train2014/')

pipert.contrib.detection_demo.utils.coco_only_people(path='../coco/labels/val2014/')

pipert.contrib.detection_demo.utils.select_best_evolve(path='evolve*.txt')

pipert.contrib.detection_demo.utils.coco_single_class_labels(path='../coco/labels/train2014/', label_class=43)

pipert.contrib.detection_demo.utils.print_mutation(hyp, results, bucket='')

pipert.contrib.detection_demo.utils.fitness(x)

pipert.contrib.detection_demo.utils.plot_one_box(x, img, color=None, label=None, line_thickness=None)

pipert.contrib.detection_demo.utils.plot_wh_methods()

pipert.contrib.detection_demo.utils.plot_images(imgs, targets, paths=None, fname='images.jpg')

pipert.contrib.detection_demo.utils.plot_test_txt()

pipert.contrib.detection_demo.utils.plot_targets_txt()

pipert.contrib.detection_demo.utils.plot_evolution_results(hyp)

pipert.contrib.detection_demo.utils.plot_results(start=0, stop=0)

pipert.contrib.detection_demo.utils.plot_results_overlay(start=0, stop=0)

pipert.contrib.detection_demo.utils.version_to_tuple(version)