本文整理汇总了Python中mmdet.core.anchor_target方法的典型用法代码示例。如果您正苦于以下问题：Python core.anchor_target方法的具体用法？Python core.anchor_target怎么用？Python core.anchor_target使用的例子？那么, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类mmdet.core的用法示例。
在下文中一共展示了core.anchor_target方法的2个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于系统推荐出更棒的Python代码示例。
示例1: loss
# 需要导入模块: from mmdet import core [as 别名]
# 或者: from mmdet.core import anchor_target [as 别名]
def loss(self,
cls_scores,
bbox_preds,
gt_bboxes,
gt_labels,
img_metas,
cfg,
gt_bboxes_ignore=None):
featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores]
assert len(featmap_sizes) == len(self.anchor_generators)
anchor_list, valid_flag_list = self.get_anchors(
featmap_sizes, img_metas)
cls_reg_targets = anchor_target(
anchor_list,
valid_flag_list,
gt_bboxes,
img_metas,
self.target_means,
self.target_stds,
cfg,
gt_bboxes_ignore_list=gt_bboxes_ignore,
gt_labels_list=gt_labels,
label_channels=1,
sampling=False,
unmap_outputs=False)
if cls_reg_targets is None:
return None
(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list,
num_total_pos, num_total_neg) = cls_reg_targets
num_images = len(img_metas)
all_cls_scores = torch.cat([
s.permute(0, 2, 3, 1).reshape(
num_images, -1, self.cls_out_channels) for s in cls_scores
], 1)
all_labels = torch.cat(labels_list, -1).view(num_images, -1)
all_label_weights = torch.cat(label_weights_list,
-1).view(num_images, -1)
all_bbox_preds = torch.cat([
b.permute(0, 2, 3, 1).reshape(num_images, -1, 4)
for b in bbox_preds
], -2)
all_bbox_targets = torch.cat(bbox_targets_list,
-2).view(num_images, -1, 4)
all_bbox_weights = torch.cat(bbox_weights_list,
-2).view(num_images, -1, 4)
losses_cls, losses_bbox = multi_apply(
self.loss_single,
all_cls_scores,
all_bbox_preds,
all_labels,
all_label_weights,
all_bbox_targets,
all_bbox_weights,
num_total_samples=num_total_pos,
cfg=cfg)
return dict(loss_cls=losses_cls, loss_bbox=losses_bbox) 示例2: loss
# 需要导入模块: from mmdet import core [as 别名]
# 或者: from mmdet.core import anchor_target [as 别名]
def loss(self,
cls_scores,
bbox_preds,
gt_bboxes,
gt_labels,
img_metas,
cfg,
gt_bboxes_ignore=None):
featmap_sizes = [featmap.size()[-2:] for featmap in cls_scores]
assert len(featmap_sizes) == len(self.anchor_generators)
device = cls_scores[0].device
anchor_list, valid_flag_list = self.get_anchors(
featmap_sizes, img_metas, device=device)
cls_reg_targets = anchor_target(
anchor_list,
valid_flag_list,
gt_bboxes,
img_metas,
self.target_means,
self.target_stds,
cfg,
gt_bboxes_ignore_list=gt_bboxes_ignore,
gt_labels_list=gt_labels,
label_channels=1,
sampling=False,
unmap_outputs=False)
if cls_reg_targets is None:
return None
(labels_list, label_weights_list, bbox_targets_list, bbox_weights_list,
num_total_pos, num_total_neg) = cls_reg_targets
num_images = len(img_metas)
all_cls_scores = torch.cat([
s.permute(0, 2, 3, 1).reshape(
num_images, -1, self.cls_out_channels) for s in cls_scores
], 1)
all_labels = torch.cat(labels_list, -1).view(num_images, -1)
all_label_weights = torch.cat(label_weights_list,
-1).view(num_images, -1)
all_bbox_preds = torch.cat([
b.permute(0, 2, 3, 1).reshape(num_images, -1, 4)
for b in bbox_preds
], -2)
all_bbox_targets = torch.cat(bbox_targets_list,
-2).view(num_images, -1, 4)
all_bbox_weights = torch.cat(bbox_weights_list,
-2).view(num_images, -1, 4)
losses_cls, losses_bbox = multi_apply(
self.loss_single,
all_cls_scores,
all_bbox_preds,
all_labels,
all_label_weights,
all_bbox_targets,
all_bbox_weights,
num_total_samples=num_total_pos,
cfg=cfg)
return dict(loss_cls=losses_cls, loss_bbox=losses_bbox)