本文整理汇总了Python中model.config.cfg.POOLING_MODE属性的典型用法代码示例。如果您正苦于以下问题：Python cfg.POOLING_MODE属性的具体用法？Python cfg.POOLING_MODE怎么用？Python cfg.POOLING_MODE使用的例子？那么恭喜您, 这里精选的属性代码示例或许可以为您提供帮助。您也可以进一步了解该属性所在类model.config.cfg的用法示例。
在下文中一共展示了cfg.POOLING_MODE属性的3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: _predict
# 需要导入模块: from model.config import cfg [as 别名]
# 或者: from model.config.cfg import POOLING_MODE [as 别名]
def _predict(self, net_conv):
# This is just _build_network in tf-faster-rcnn
# torch.backends.cudnn.benchmark = False
# net_conv = self._image_to_head()
'''
ROI pooling on SELECTIVE SEARCH boxes
'''
if cfg.POOLING_MODE == 'crop':
pool5 = self._crop_pool_layer(net_conv, self._boxes)
else:
pool5 = self._roi_pool_layer(net_conv, self._boxes)
if self._mode == 'TRAIN':
torch.backends.cudnn.benchmark = True # benchmark because now the input size are fixed
fc7 = self._head_to_tail(pool5)
cls_prob, bbox_prob, fuse_prob, image_prob = self._region_classification(fc7)
# for k in self._predictions.keys():
# self._score_summaries[k] = self._predictions[k]
self._score_summaries['image_prob'] = self._predictions['image_prob']
#print(id(net_conv))
return fuse_prob
# return net_conv, cls_prob, bbox_prob, fuse_prob, image_prob 开发者ID:Sunarker，项目名称:Collaborative-Learning-for-Weakly-Supervised-Object-Detection，代码行数:28，代码来源:network.py
示例2: _predict
# 需要导入模块: from model.config import cfg [as 别名]
# 或者: from model.config.cfg import POOLING_MODE [as 别名]
def _predict(self):
# This is just _build_network in tf-faster-rcnn
torch.backends.cudnn.benchmark = False
net_conv = self._image_to_head()
# build the anchors for the image
if cfg.FPN:
self._anchor_component_fpn(net_conv)
rois = self._region_proposal_fpn(net_conv)
pool5 = self._crop_pool_layer_fpn(net_conv, rois)
pool5 = pool5.view(pool5.size(0),-1)
else:
self._anchor_component(net_conv.size(2), net_conv.size(3))
rois = self._region_proposal(net_conv)
if cfg.POOLING_MODE == 'crop':
pool5 = self._crop_pool_layer(net_conv, rois)
else:
pool5 = self._roi_pool_layer(net_conv, rois)
if self._mode == 'TRAIN':
torch.backends.cudnn.benchmark = True # benchmark because now the input size are fixed
fc7 = self._head_to_tail(pool5)
cls_prob, bbox_pred = self._region_classification(fc7)
for k in self._predictions.keys():
self._score_summaries[k] = self._predictions[k]
return rois, cls_prob, bbox_pred 示例3: _build_network
# 需要导入模块: from model.config import cfg [as 别名]
# 或者: from model.config.cfg import POOLING_MODE [as 别名]
def _build_network(self, is_training=True):
# select initializers
if cfg.TRAIN.TRUNCATED:
initializer = tf.truncated_normal_initializer(mean=0.0, stddev=0.01)
initializer_bbox = tf.truncated_normal_initializer(mean=0.0, stddev=0.001)
else:
initializer = tf.random_normal_initializer(mean=0.0, stddev=0.01)
initializer_bbox = tf.random_normal_initializer(mean=0.0, stddev=0.001)
net_conv = self._image_to_head(is_training)
with tf.variable_scope(self._scope, self._scope):
# build the anchors for the image
self._anchor_component()
# region proposal network
rois = self._region_proposal(net_conv, is_training, initializer)
# region of interest pooling
if cfg.POOLING_MODE == 'crop':
pool5 = self._crop_pool_layer(net_conv, rois, "pool5")
else:
raise NotImplementedError
fc7 = self._head_to_tail(pool5, is_training)
with tf.variable_scope(self._scope, self._scope):
# region classification
cls_prob, bbox_pred = self._region_classification(fc7, is_training,
initializer, initializer_bbox)
self._score_summaries.update(self._predictions)
return rois, cls_prob, bbox_pred