<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[bear507的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/bear507</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; bear507]]></copyright><item><title><![CDATA[Hadoop 3.3 Spark 3.2 M1 MacOS 伪分布式配置]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/123876262</link><guid>https://blog.youkuaiyun.com/bear507/article/details/123876262</guid><author>bear507</author><pubDate>Thu, 31 Mar 2022 16:20:40 +0800</pubDate><description><![CDATA[Hadoop 3.3 + Spark 3.2 + M1 MacOS伪分布式配置
安装Java 8
一开始装了Java 17发现Hadoop没法开启yarn，只好降低版本，适用于M1的Java 8下载：Java Download | Java 8, Java 11, Java 13 - Linux, Windows &amp; macOS (azul.com)
Java多版本管理
打开~/.zshrc输入
export JAVA_17_HOME="/Library/Java/JavaVirtualMachin]]></description><category></category></item><item><title><![CDATA[Elliptic Curve Cryptography (ECC) and Pairings 椭圆曲线密码学与配对]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/116055405</link><guid>https://blog.youkuaiyun.com/bear507/article/details/116055405</guid><author>bear507</author><pubDate>Fri, 23 Apr 2021 14:38:20 +0800</pubDate><description><![CDATA[本文是Dan Boneh 和Victor Shoup所写书籍A Graduate Course in Applied Cryptography的笔记。
The group of points of an elliptic curve
最早由Diophantus提出了一个问题：已知二元多项式方程f(x,y)=0f(x,y)=0f(x,y)=0，如何找到满足方程的有理数点(x∈Q,y∈Q)(x\in Q,y\in Q)(x∈Q,y∈Q)？该问题可以等价于找到满足y2=x3−x+9y^2=x^3-x+9y2=x3]]></description><category></category></item><item><title><![CDATA[Flask和Docker部署Pytorch深度学习模型]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/108330373</link><guid>https://blog.youkuaiyun.com/bear507/article/details/108330373</guid><author>bear507</author><pubDate>Mon, 31 Aug 2020 21:54:07 +0800</pubDate><description><![CDATA[模型训练
预先训练深度学习模型，保存训练后的模型或参数，此处采用图像分类模型。
Flask搭建模型
Flask是一个使用 Python 编写的轻量级Web应用框架，可通过$pip install flask进行安装。
首先编写app.py文件
import os
from flask import Flask, render_template, request, redirect
import io
from PIL import Image
import uuid
import vision 

app ]]></description><category></category></item><item><title><![CDATA[【一周算法实践进阶】任务3 模型融合(Stacking)]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/86726962</link><guid>https://blog.youkuaiyun.com/bear507/article/details/86726962</guid><author>bear507</author><pubDate>Thu, 31 Jan 2019 22:03:59 +0800</pubDate><description><![CDATA[导入本次任务所用到的包：
import pandas as pd
from sklearn.model_selection import train_test_split, KFold, GridSearchCV
import numpy as np
import warnings
import matplotlib.pyplot as plt
from sklearn.metrics impor...]]></description><category></category></item><item><title><![CDATA[【一周算法实践进阶】任务2 特征工程]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/86696246</link><guid>https://blog.youkuaiyun.com/bear507/article/details/86696246</guid><author>bear507</author><pubDate>Tue, 29 Jan 2019 21:48:11 +0800</pubDate><description><![CDATA[导入本次任务所用到的包：
import pandas as pd
import numpy as np
from scipy import stats
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.linear_model...]]></description><category></category></item><item><title><![CDATA[【一周算法实践进阶】任务1 数据预处理]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/86649069</link><guid>https://blog.youkuaiyun.com/bear507/article/details/86649069</guid><author>bear507</author><pubDate>Fri, 25 Jan 2019 17:05:58 +0800</pubDate><description><![CDATA[导入本次实践过程中所需的包：
import pandas as pd
from sklearn.preprocessing import LabelBinarizer, Imputer

准备数据
数据集下载
实践数据的下载地址 https://pan.baidu.com/s/1wO9qJRjnrm8uhaSP67K0lw
说明：这份数据集是金融数据（非原始数据，已经处理过了），我们要做的是预测贷...]]></description><category></category></item><item><title><![CDATA[【数据分析实践】Task 1.3 模型调优]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/86183272</link><guid>https://blog.youkuaiyun.com/bear507/article/details/86183272</guid><author>bear507</author><pubDate>Wed, 09 Jan 2019 22:03:58 +0800</pubDate><description><![CDATA[导入本次实践过程中所需的包：
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split, GridSearchCV
from sklearn.linear_model import LogisticRegres...]]></description><category></category></item><item><title><![CDATA[【数据分析实践】Task 1.3 模型评估]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/85922463</link><guid>https://blog.youkuaiyun.com/bear507/article/details/85922463</guid><author>bear507</author><pubDate>Mon, 07 Jan 2019 21:49:27 +0800</pubDate><description><![CDATA[构建随机森林、GBDT、XGBoost和LightGBM这四个模型，并对每一个模型进行评分，评分方式任意，例如准确度和auc值。

本篇代码均建立在上一篇(【数据分析实践】 Task1.1 模型构建)代码已运行的基础上

模型构建
导入包
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection...]]></description><category></category></item><item><title><![CDATA[【数据分析实践】Task 1.2 模型构建之集成模型]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/85802417</link><guid>https://blog.youkuaiyun.com/bear507/article/details/85802417</guid><author>bear507</author><pubDate>Fri, 04 Jan 2019 21:45:43 +0800</pubDate><description><![CDATA[模型构建之集成模型
构建随机森林、GBDT、XGBoost和LightGBM这四个模型，并对每一个模型进行评分，评分方式任意，例如准确度和auc值。

本篇代码运行均建立在上一篇(【数据分析实践】 Task1.1 模型构建)代码的运行基础上

集成模型构建
模型训练：
rf_clf = RandomForestClassifier(random_state=2018).fit(X_train, y...]]></description><category></category></item><item><title><![CDATA[【数据分析实践】 Task1.1 模型构建]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/85715485</link><guid>https://blog.youkuaiyun.com/bear507/article/details/85715485</guid><author>bear507</author><pubDate>Thu, 03 Jan 2019 19:46:45 +0800</pubDate><description><![CDATA[导入本次实践过程中所需的包：
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LogisticRegression
from skle...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 19 Introduction to Deep Learning]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82354418</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82354418</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 20:34:08 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 19 Introduction to Deep Learning



Introduction

目前为止，本课程已经包含了一系列经典的计算机视觉技术，包括边缘检测 (edge detection)、聚类 (clustering) 方法、分类器 (classi...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 1 Course Introduction]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316169</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316169</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:07:30 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 1 Course Introduction



什么是计算机视觉？



定义


从数字图像中提取信息。这些信息可以随着定义不同而改变，可以是空间测量、现实增强等等
构造一个可以理解图片意思并能应用的算法。




一个艰难的问题

50 年来，计算视觉都没...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 2 Color and Linear Algebra]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316159</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316159</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:07:15 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 2 Color and Linear Algebra



颜色物理学



什么是颜色？

颜色是环境中的物理光和我们视觉系统交互的结果。它是一种我们在看物品和光时的视觉体验的心理性质，而不是这些物品和光的物理性质。



颜色和光

白光由几乎平均能量的可视...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 3 Linear Algebra Primer Part 2]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316153</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316153</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:06:59 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正




Lecture 3 Linear Algebra Primer Part 2



Transformation Matrices

矩阵可以用多种方式将向量转换，最简单的例子是缩放：

[sx00sy][sx00sy]\left [ \begin{matrix} s_{x...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 4 Pixels and Filters]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316149</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316149</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:06:44 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 4 Pixels and Filters



Image Sampling and Quantization



Image Type


Binary Type 每个像素不是黑 (0) 就是白 (1)
Grayscale Images 比起 Binary ...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 5 Edge detection]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316142</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316142</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:06:28 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 5 Edge Detection



Edge Detection in Mammals



Hubel &amp;amp;amp; Wiesel

两个人的实验表明，猫的视觉神经在碰到特定方向的边缘时，会产生反应。



Biederman

Biederman 的实验表...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 6 Edge Detection]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316138</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316138</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:06:16 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 6 Edge Detection



Edge Detection



Edge Basic

图片上的边缘主要有四个可能的来源：不连续的表面（表面角度突然改变）、深度不连续（一个表面重叠在另一个上）、表面颜色不连续、光照不连续（明 / 暗）。从数学上来看，...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 6-7 Features and FittingFeature Descriptor]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316122</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316122</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:05:54 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 6-7 Features and Fitting/Feature Descriptor



Local Invariant(局部不变) Features



Motivation

使用局部不变的特征的目的出于它在广泛环境下的有用性，之前讨论的方法在这些环境...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 8 Feature Descriptors and Resizing]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316113</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316113</guid><author>bear507</author><pubDate>Mon, 03 Sep 2018 08:05:18 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 8 Feature Descriptors and Resizing



Scale invariant keypoint detection



Motivation

给定有较大尺度差别的两张相同情景下的图片，目标是独立的检测每张图中相同关键点。解决方案...]]></description><category></category></item><item><title><![CDATA[[CS131] Lecture 9 Image Resizing and Segmentation]]></title><link>https://blog.youkuaiyun.com/bear507/article/details/82316096</link><guid>https://blog.youkuaiyun.com/bear507/article/details/82316096</guid><author>bear507</author><pubDate>Sun, 02 Sep 2018 15:47:06 +0800</pubDate><description><![CDATA[
  根据 Stanford CS131 课程写的笔记（大部分为 note 翻译），英语水平一般，如有错误请评论指正


Lecture 9 Image Resizing and Segmentation



Introduction

为了适应不同尺寸和形状的播放器，图片或影片需要调整大小，本章介绍如何在调整大小的同时保存重要内容并减少瑕疵。



Problem Statement

输入大...]]></description><category></category></item></channel></rss>