You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
for k in range(30):
x = self.vgg[k](x)###############
if 22 == k:
s = self.conv4_3_L2Norm(x)
sources.append(s)
elif 29 == k:
s = self.conv5_3_L2Norm(x)
sources.append(s)
# apply vgg up to fc7
aaa = len(self.vgg)
for k in range(30, len(self.vgg)):
x = self.vgg[k](x)
sources.append(x)
测试模式,当第一次运行到refinedet.py的forward的以下代码的时候:
在 x = self.vggk这行,用vgg跑前向的时候,
当i=5的时候,执行x = self.vggk这句话,显存会快速达到4851M然后快速降到800多M
当i=7的时候,执行x = self.vggk这句话,显存会快速达到1609M然后快速降到800多M
。。。
看了下都是卷积层
但是为什么是第一次的时候,后面不会有问题了!!!!so,why第一次的时候会出现后面不会??
用的320大小的图片
The text was updated successfully, but these errors were encountered: