CornerNet 测试：

源码地址：

https://github.com/feifeiwei/Pytorch-CornerNet

经过测试，p100上，测试直接内存溢出了。

# -*- coding: utf-8 -*-
"""
Created on Tue Dec  4 10:49:24 2018

@author: 60236
"""
import time
import torch
import torch.nn as nn

from select_gpu import usegpu
from backbone import ResNet50
from layers import  conv_bn, conv_bn_relu, conv_relu
from corner_pooling import top_pool, left_pool,bottom_pool, right_pool

        
class corner_net(nn.Module):
    def __init__(self, num_classes, inplanes=256, backbone=ResNet50):
        super(corner_net,self).__init__()
        self.features = backbone()
        self.relu = nn.ReLU(inplace=True)
        self.num_classes = num_classes    

        self.conv_bn_relu1 = conv_bn_relu(in

文章来源: blog.csdn.net，作者：网奇，版权归原作者所有，如需转载，请联系作者。

原文链接：blog.csdn.net/jacke121/article/details/86764521