教小白精通编程

面向小白的编程教学博客

策略评估

policy evaluation

策略评估的伪代码如下: Policy_Evaluation(输入:环境$p(s’,r s,a )$,策略$\pi$) 初始化所有状态的价值为0:$V(s), s\in S$ for $k = 0,1,\cdots $: 初始化逼近误差 $\delta \leftarrow 0$ for...

移民

How to migrate

各种指数 人均年收入(美元) 美国63,080 中国9,460 香港50,300 民主指数或经济学人智库2018年民主指数: 美国25(得分7.96) 中国130(得分3.32) 香港73(得分) 挪威第一(得分9.87),朝鲜倒数第一(1.08) (2019年) **清廉指数 cpi2018 **:美国23 中国80 香港16 环境可持续指数: 美国 27(7...

谷歌云VPS+V2ray+WebSocket+TLS+Web_CDN

google cloud_VPS+V2ray+WebSocket+TLS+Web_CDN

1. 申请谷歌云(或购买其他VPS)+安装v2ray 2. WebSocket+TLS+Web https://guide.v2fly.org/en_US/advanced/wss_and_web.html 1. 申请域名+配置解析+验证 2. 安装Caddy Install Caddy with PHP & HTTPS using Let’s Encrypt on Ub...

Resources for policy gradient

layout: post title: Resources for policy gradient subtitle: Resources for policy gradient date: 2020-01-29 13:51:00 author: “xuepro” header-img: “img/home_bg.jpg” ...

A PyTorch implementation of deep Q-learning Network (DQN) for Atari games

A PyTorch implementation of deep Q-learning Network (DQN) for Atari games

Deep Q-learning Network (DQN) can be used to train an agent to play Atari games: We often use continuous frames to represent an state of the enviroment. DQN use replay mempry to store ecxperienc...

install swig and box2d-py for windows

install swig and box2d-py for windows

I encountered an error when I want to use the lunar_lander environment in gym. The error is : error: “module ‘gym.envs.box2d’ has no attribute ‘LunarLander’” I searched on internet and found the ...

强化学习资源

Resources for reinforcement learning

Awesome Reinforcement Learning https://github.com/ZhiqiangHo/awesome-reinforcement-learning courses: http://rail.eecs.berkeley.edu/deeprlcourse/ A brief introduction to reinforcement learning ...

比较Pytorch和我的卷积实现

Comparison of my convolution implementation to pytorch

the Test code for pytorch conv2d # pytorch import torch import torch.nn as nn x = torch.tensor(range(50),dtype=torch.float) x = torch.reshape(x, (1,2,5,5)) #N,C,H,W x.requires_grad = True print(x.s...

如何用google colab运行你的深度学习代码

how to use google colab for your deep learning

google Colaboratory(简称 google colab)是谷歌提供的免费的云端的Jupyter notebook环境,google colab已经自带了各种常用的python程序包包括著名的深度学习库tensorflow和pytorch,特别是google colab提供了免费的GPU,使得不需要做任何安装和配置就能在浏览器编写和执行python代码(包括机器学习、深度学习代...

test the deep learning library. in my book

test the deep learning library. in my book

test dense layer import numpy as np import matplotlib.pyplot as plt %matplotlib inline from NeuralNetwork import * import util np.random.seed(1) pts = 100 k = 2 b =1 X = np.random.randn(pts,1)*...