两派 1. 新的卷机计算方法 这种是直接提出新的卷机计算方式,从而减少参数,达到压缩模型的效果,例如SqueezedNet,mobileNet SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size 修改网络结构,类似于mobileNet MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Appli
学习笔记: 1.基于TCP协议的Socket网络编程: (1)Socket类构造方法:在客户端和服务器端建立连接 Socket s = new Socket(hostName,port);以主机名和端口号作为参数来创建一个Socket对象. Socket s = new Socket(address,port);以InetAddress对象和端口号作为参数来创建一个Socket对象. 创建Socket对象时可能抛出UnknownHostException或IOException异常,必须捕获它们
原来泛型可以这样用: 网络返回基类,返回一个code,msg,body,其中body不确定,所以,我们把它写成泛型 import org.json.JSONObject; /** * 网络请求的基类 * Created by on 16/7/14. */ public class NetData<T> { public static final int STATUS_OK = 0; public int code = -1; public String msg; public T body;
!pip install gym import random import numpy as np import matplotlib.pyplot as plt from keras.layers import Dense, Dropout, Activation from keras.models import Sequential from keras.optimizers import Adam from keras import backend as K from collection