IT博客汇
  • 首页
  • 精华
  • 技术
  • 设计
  • 资讯
  • 扯淡
  • 权利声明
  • 登录 注册

    love 0
    Christian Lorentzen
    发表于
    2025-05-01 03:00:00

    Model Diagnostics: Statistics vs Machine Learning

    [social4i size="small" align="align-left"][This article was first published onR – Michael's and Christian's Blog, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.In this post, we show how differentuse casesrequire differentmodel diagnostics. In short, we compare (statistical)inferenceandprediction.As an example, we use a simple linear model for the Munich rent index dataset, which was kindly provided by the authors ofRegression – Models, Methods an ...继续阅读 (83)


    love 0
    R on Nicola Rennie
    发表于
    2025-05-01 00:00:00

    30 Day Chart Challenge 2025

    [social4i size="small" align="align-left"][This article was first published onR on Nicola Rennie, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.The30 Day Chart Challengeis a data visualisation challenge organised by Cédric Scherer and Dominic Royé. Participants make one chart each day of the challenge, inspired by the daily prompt. The prompts are also split across 5 different categories, which for 2025 were comparisons, distributions, relationships, time ...继续阅读 (88)


    love 0
    Stat's What It's All About
    发表于
    2025-04-30 21:00:00

    Working with Ordinal Ranks in {marginaleffects}

    [social4i size="small" align="align-left"][This article was first published onStat's What It's All About, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Given an ordinal regression model, it is relatively easy to get class-wise predictions – the conditional predicted probability of each level of the outcome. However,often, one might be interested in summarizing effects not on the class-wise probability scale (nor on the latent scale), but in ...继续阅读 (86)


    love 0
    Richard Tyler
    发表于
    2025-04-30 19:40:39

    How to Improve Customer Experience with a Customer Service Virtual Assistant

    In today’s hyperconnected world, customer expectations are higher than ever. People want fast, personalized, and round-the-clock support, regardless of business size or industry. For growing companies, meeting these expectations can be a challenge. That’s where a customer service virtual assistant can make a big impact. With the support of reliable virtual assistant companies in the […]The postHow to Improve Customer Experience with a Customer Service Virtual Assistantappeared first onJ L I Blog. ...继续阅读 (92)


    love 0
    Michael Mayer
    发表于
    2025-04-30 10:51:39

    Fast Grouped Counts and Means in R

    [social4i size="small" align="align-left"][This article was first published onR – Michael's and Christian's Blog, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.From time to time, the following questions pop up:How to calculate grouped counts and (weighted) means?What are fast ways to do it in R?This blog post presents a couple of approaches and then compares their speed with a naive benchmark.Base RThere are many ways to calculate grouped counts ...继续阅读 (70)


    love 0
    Liyun
    发表于
    2025-04-30 07:28:05

    从心所欲

    吾十有五而志于学,三十而立,四十而不惑,五十而知天命,六十而耳顺,七十而从心所欲,不逾矩年轻的时候并不是很能读懂这段话,年纪大了才恍然时间快进。现在的日子基本上是 知天命、耳顺、从心所欲的阶段了。每天看着聒噪的人群,我已经心生羡慕。和长者的间隙已经远远小于和二十多岁年轻人的代沟。是从什么时候开始,经历了什么,让我的人生越来越勇敢,也越来越无所谓了呢。初心还在吗,还在的。梦想还在吗,还在的。那一束束不曾泯灭的光亮,闪耀着年少与年长。放下的越多,只能说明人越来越老,力不从心。跨越千山万里,只为回眸一笑,是一种人生。和猫一起睡到天昏地暗,也是一种人生。习惯了疯狂之后,人生好像也没什么界限了。从心所欲,不逾矩,却也不设限。 ...继续阅读 (102)


    love 0
    Richard Tyler
    发表于
    2025-04-30 06:57:07

    Energy-Efficient Water Heating Solutions for Filipino Households

    As energy prices continue to rise in the Philippines, more homeowners are looking for ways to save on electricity without sacrificing comfort. One often overlooked area is the water heater. A well-chosen unit can significantly reduce power consumption while still providing the hot water needed for daily routines. For households seeking smarter energy use, exploring […]The postEnergy-Efficient Water Heating Solutions for Filipino Householdsappeared first onJ L I Blog. ...继续阅读 (104)


    love 0
    Joseph Rickert
    发表于
    2025-04-30 00:00:00

    March 2025 Top 40 New CRAN Packages

    [social4i size="small" align="align-left"][This article was first published onR Works, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.In March, one hundred eighty-two new packages made it to CRAN. Here are my Top 40 picks in sixteen categories: Agriculture, Archaeology, Biology, Climate Modeling, Computational Methods, Data, Ecology, Epidemiology, Genomics, Machine Learning, Medicine, Risk Forecasting, Statistics, Time Series, Utilities, and Visualization.A ...继续阅读 (65)


    love 0
    Jakub Nowosad
    发表于
    2025-04-30 00:00:00

    Spatial machine learning with R: caret, tidymodels, and mlr3

    [social4i size="small" align="align-left"][This article was first published ongeocompx, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.This is the first part of a blog post series on spatial machine learning with R.The R language has a variety of packages for machine learning, and many of them can be used for machine learning tasks in a spatial context (spatial machine learning). Spatial machine learning is generally different from traditional machine learn ...继续阅读 (79)


    love 0
    R/exams
    发表于
    2025-04-29 22:00:00

    vowels: Phonetics cloze quiz about Cardinal vowels

    [social4i size="small" align="align-left"][This article was first published onR/exams, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Cloze exercise in which two randomly-selected Cardinal vowels have to be described according to the three parameters vowel height, frontness-backness, and lip rounding.Name:vowelsType:clozePreview:Describe the following Cardinal vowels according to the three parameters vowel height, frontness-backness, and lip rounding:[e] is ...继续阅读 (82)


    love 0
    R/exams
    发表于
    2025-04-29 22:00:00

    flags: Fun with Flags

    [social4i size="small" align="align-left"][This article was first published onR/exams, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Single-choice knowledge quiz question in which the correct flag for a country has to be selected from a list of flags.Name:flagsType:schoiceRelated:capitals,countrycodesPreview:Which is the flag of Bolivia?False.is the flag of Chile.True.is the flag of Bolivia.False.is the flag of Peru.False.is the flag of Brazil.False.is the ...继续阅读 (78)


    love 0
    M. Fatih Tüzen
    发表于
    2025-04-29 21:00:00

    Explained vs. Predictive Power: R², Adjusted R², and Beyond

    [social4i size="small" align="align-left"][This article was first published onA Statistician's R Notebook, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.1IntroductionYou trust R². Should you?You proudly present a model with R² = 0.95. Everyone applauds.But what if your model fails miserably on the next new data?When building a statistical model, one of the first numbers analysts and data scientists often cite is theR², or coefficient of determination. ...继续阅读 (76)


    love 0
    Dariia Mykhailyshyna
    发表于
    2025-04-29 13:34:48

    surveydown: An Open-Source, Markdown-Based Platform for Interactive and Reproducible Surveys workshop

    [social4i size="small" align="align-left"][This article was first published onR-posts.com, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Join our workshop onsurveydown: An Open-Source, Markdown-Based Platform for Interactive and Reproducible Surveys, which is a part of our workshops for Ukraine series!Here’s some more info:Title:surveydown: An Open-Source, Markdown-Based Platform for Interactive and Reproducible SurveysDate:Thursday, May 29th, 18:00 – 20:0 ...继续阅读 (49)


    love 0
    未曾
    发表于
    2025-04-29 03:33:21

    耕织图卷

    《耕织图》是南宋时期楼璹以诗配画的方式绍耕织技术的著作。此《耕织图》卷传为元代程棨摹绘宋代楼璹本。分作:耕作图(自浸种至入仓共二十一节)、蚕织图(自浴蚕至剪帛共二十四节)两卷。各节附标题及五言诗一首,诗文用篆书(另配楷书小字)。此两卷现藏于弗瑞尔·赛克勒美术馆。 ...继续阅读 (92)


    love 0
    R/exams
    发表于
    2025-04-28 22:00:00

    sumdiff: Sum Minus Difference

    [social4i size="small" align="align-left"][This article was first published onR/exams, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Exercise template for a very basic arithmetic task, namely computing the sum of two random numbers minus their difference.Name:sumdiffType:numRelated:derivPreview:By how much is the sum of the numbers 52 and 30 greater than the difference between these numbers?The result of the calculation is (52 + 30) – (52 – 30) ...继续阅读 (46)


    love 0
    Will Hipson
    发表于
    2025-04-28 03:00:00

    Announcing… schematic

    [social4i size="small" align="align-left"][This article was first published ondata-in-flight, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.I’m thrilled to announce the release ofschematic, an R package that helpsyou(the developer) communicate data validation problems to non-technical users. With schematic, you can leveragetidyselectselectors and other conveniences to compare incoming data against a schema, avoiding punishing issues caused by invalid or po ...继续阅读 (55)


    love 0
    datascienceconfidential - r
    发表于
    2025-04-28 00:00:00

    How the DSGE sausage is made

    [social4i size="small" align="align-left"][This article was first published ondatascienceconfidential - r, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.IntroductionDynamic Stochastic General Equilibrium (DSGE) models are a class of models which attempt to model the entire economy of a nation. The purpose of this post is to hand-code a DSGE model in R, fit it to data, and check that the results agree with the software packages Dynare and gEcon.I got intere ...继续阅读 (46)


    love 0
    rOpenSci - open tools for open science
    发表于
    2025-04-28 00:00:00

    rOpenSci News Digest, April 2025

    [social4i size="small" align="align-left"][This article was first published onrOpenSci - open tools for open science, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Before sending DELETE THE INDEX_CACHE and re-knit!Dear rOpenSci friends, it’s time for our monthly news roundup!blablaYou can read this poston our blog. Now let’s dive into the activity at and around rOpenSci!rOpenSci HQEvent: Career Paths for U.S. Federal Data ScientistsrOpenSci and Openscapes ...继续阅读 (58)


    love 0
    xi\'an
    发表于
    2025-04-27 22:25:53

    proximal sampler

    [social4i size="small" align="align-left"][This article was first published onR – Xi'an's Og, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.At theColumbia workshoplast week, Andre Wibisono presented work related with arecent arXivalon the exponentially fast convergence of both unadjusted Langevin and  proximal sampler algorithms under strong [definitely strong] log-concavity assumptions. The idea behind the proximal sampler is to target the demar ...继续阅读 (58)


    love 0
    Zhenguo Zhang
    发表于
    2025-04-26 00:00:00

    [R] How to create errorbars with overlaid points using ggplot

    [social4i size="small" align="align-left"][This article was first published onR on Zhenguo Zhang's Blog, and kindly contributed toR-bloggers]. (You can report issue about the content on this pagehere)Want to share your content on R-bloggers?click hereif you have a blog, orhereif you don't.Zhenguo Zhang’s Blog /2025/04/26/r-how-to-create-an-error-barplot-with-overlaid-points-using-ggplot/ –library(ggplot2) library(dplyr)Sometimes you may want to create a plot with the following features:a point to indicate the mean of a grouperror bars to indicate the standard deviation of the ...继续阅读 (55)


    love 0
    未曾
    发表于
    2025-04-23 08:01:39

    童蒙必读书

    此《童蒙必读书》由清代涂宗瀛辑编。全书包含:弟子规、小儿语、弟子职、程子四箴、范氏心箴、朱子敬斋箴、朱子小学题辞、陈氏忍字箴、性理字训、训蒙千字文等十四种家塾理学训蒙课本。此为清光绪九年武昌书局校刊本,手书上板,字大墨莹。 ...继续阅读 (82)


    love 0
    MacTalk
    发表于
    2025-04-22 10:40:38

    墨问便签正成为历史

    墨问便签这款产品,最初是用来做 demo 的,试试小程序的能力。另外,有一点纪念之前做的锤子便签的意味,就取了个名字,叫墨问便签。没想到这个小产品慢慢有了生命,形成了社区,小程序的能力可以承载图文、音频、视频和订阅等很多功能,于是 demo 转正,我们在 2023 年 10 月 28 日正式发布了墨问便签 1.0。时至今日,墨问便签早就跳出了便签的范畴,有了 AI 和社区属性,未来我们肯定会做 Web 版本。所以我想给它一个独立的名字,就叫墨问好了。事实上墨友们有很长一段时间都是这么叫的。二爷说,批准。这等同于 TheFacebook 改名为 Facebook。毕竟,Google 原名是 BackRubApple Computer 改成了 AppleTheFacebook 改名为 Facebook墨问便签改为墨问,一切就算是对了。这么地吧,撒个花[庆祝][庆祝][庆祝]https://note.mowen.cn/note/detail?noteUuid=jbJv5KSkoUPQfmFWVhY1o2025年4月22日 ...继续阅读 (92)


    love 0
    未曾
    发表于
    2025-04-20 03:27:26

    历代诗话

    此《历代诗话》由清代何文焕编订。全书共十六册,辑录南朝梁至明代的诗话,如:诗品、二十四诗品、全唐诗话、石林诗话、韵语阳秋、山房随笔、夷白斋诗话等共二十七种。内容涉及作家作品评论、诗歌理论的发挥、诗人遗闻轶事。此为清乾隆三十五年刊本,末附《历代诗话考索》。 ...继续阅读 (116)


    love 0
    未曾
    发表于
    2025-04-17 10:03:56

    草虫图卷

    此《草虫图》卷由清代朱汝琳绘。画卷描绘了各式广腰或细腰蜂类、飞蛾、蜻蜓、豆娘、螳螂、蟑螂、天牛、金蛛、椿象、蝗虫、步行虫等七十只草虫。作者观察实体后描绘这些草虫,各个栩栩如生,而且笔力细健,色彩雅洁。此卷绘制于清康熙五十年,现藏于台北故宫博物院。 ...继续阅读 (87)


    love 0
    未曾
    发表于
    2025-04-15 01:02:53

    植物名实图考

    此《植物名实图考》由清代植物学家吴其濬著。全书三十八卷(另附长编二十二卷),分为谷类、蔬类、山草类、隰草类、石草类、水草类、蔓草类、芳草类、毒草类、群芳类、果类、木类等十二大类。其中收录植物一千七百多条(绘图一千八百多幅)。此为清道光时期陆应谷校刊本。 ...继续阅读 (115)


    love 0
    Prologue 序章
    发表于
    2025-04-12 00:00:00

    2025 Q1经济杂谈

    已知的已知也可以是黑天鹅2025 Q1经济杂谈《娱乐至死》中的政治事务娱乐化是真实的。有一些人已经把聊国家大事当成一种谈资,闲暇时总和别人聊一些耸人听闻的宏大叙事:大棋论、地缘政治战略、金融战等等,左手拿着《货币战争》,右手指点江山,就像是各种媒体的复读机,有反乌托邦小说的元素。最近听得最多的是人们关于特朗普关税和台海议题的讨论,这种事物一般是不会有对错是非之分,只有如何把握机遇一说。特朗普关税很显然中美是双输的,特朗普政府把关税当作增长国家安全和制造业回流的手段,而中国和欧元区作为乙方试图讨价还价。不理解的是很多人把中国关税反击当作一种胜利,他们知道相互加征关税的后果是什么吗?原产地在美国的一些精密仪器和芯片零件的采购成本暴增,如果人民币不竞争性贬值,对美转口贸易和直接出口可能因此熄火。我们很容易地拿着商务部的统计数据看,中国2023年对美贸易敞口也就不到2成,但问题是转口贸易可能占比多少?美国也对墨西哥加征关税以及取消小额包裹免税,转口贸易和离岸跨境电商的道路已经被堵死了。保守估计,至少也有3成出口是直接或间受特朗普关税影响,所以当下是美国在出口“内需”,拿本国人民对商品的大量需求作为筹码去威胁他国,中国和欧元区如果不去谈妥,降低双方相互增加的关税,就会加剧通缩,被迫去产能,结局是不得不走上通往长期通缩的道路。通缩不一定意味着严重的经济衰退(如GDP负增长),但大部分人的经济生 ...继续阅读 (79)


    love 0
    MacTalk
    发表于
    2025-04-11 11:24:09

    使用海外产品(包括 AI)的时候,发现个秘密

    这个发现叫做慢操作终于把 2025 墨问 AI 内参专栏的第一季交付完了。为了写这个第一季,我们下了挺多功夫。包括把自己的 App Store 切换美区,下载了很多海外的 AI 产品。包括购买了推特的会员,也就是 X 的蓝V,并且重启了推特创作,在上面关注了不少有趣的 AI 行业的人,自己也写了不少内容。在使用这些软件的过程中,我发现了一个挺有意思的现象。国外的软件好像没有那么着急,没有像中国的快递那样必须要求立等可取,我觉得这可能反映了不同区域人们的生活态度。比如,我注册推特的蓝 V,花了钱之后大概用了 3 天的时间才把这个蓝 V 的标给我挂上,系统提示就是在审核,我也不知道在审核什么。如果是在国内做产品的话,我花了几百块钱买了你一个产品,你不能让我立刻用,不能让我立刻有那个标识。这怎么可以呢?我这外卖 20 分钟都到了,你这儿买一个产品居然要三天才能给我挂上蓝 V,简直难以想象,这产品基本上就不能要了。包括前一阵儿。我们使用这个新兴的 AI 浏览器 Dia,它采用了一种邀请码的方式。这个邀请码并不是官方来发放的,它是一种缓慢传播的方式。比如说,我从小盖那儿拿到了一个邀请码,我就可以使用这个产品,我也有了 10 个邀请码。我可以用这 10 个邀请码邀请别人,让别人去试用。小盖的邀请码是哪儿来的?因为他特别想用这个产品,就去推特上找一个老外要的。大概就是这样一个传播路径。路径不复杂 ...继续阅读 (73)


    love 0
    MacTalk
    发表于
    2025-04-10 04:00:52

    如何理解段永平这周的股市操作

    前几天我看到段在雪球上说:卖了点苹果以及其他的 put。懂王威武,感觉起码要折腾一阵子,卖点 put 来回一起折腾也许是个好玩的游戏。还是老话,千万别用 ma­r­g­in。如果 put 进来的价格从 10 年后看回来是便宜的价格,那这个投资就值得做,长短皆宜。//段永平这段话是什么意思?我不是很清楚,但咱可以问 Gimini 2.5 Pro,回复如下:Gimini 2.5 Pro:这段话是段永平在分享他对于市场波动时期的一种投资策略和理念,主要围绕着卖出看跌期权(Selling Put Options)展开。我们可以分点来理解:1、“卖了点苹果以及其他的 put”:核心操作: 他卖出了(Sell)一些苹果公司(AAPL)和其他他认为不错的公司的看跌期权(Put Options)。2、卖出看跌期权意味着什么: 卖出看跌期权的投资者(在这里是段永平)获得一笔权利金(Premium)。作为交换,他承担了一项义务:如果该期权的买方在到期日或之前选择行权(Exercise),那么卖方(段永平)必须按照约定的行权价(Strike Price)从买方手中买入对应数量的股票。3、卖方的期望: 通常,卖出看跌期权的人期望股价在期权到期时高于或等于行权价,这样期权就不会被行权(或者价值很低),他就能赚取全部的权利金。4、“懂王威武,感觉起码要折腾一阵子”:背景/动机: “懂王”通常指美国前总统特朗普 ...继续阅读 (96)


    love 0
    未曾
    发表于
    2025-04-04 23:00:15

    克拉普的中国摄影集

    此套《克拉普的中国摄影集》由美国地质学家弗雷德里克·克拉普(Frederick G. Clapp)拍摄。此次整理共432幅,记录了民国初期克拉普团队在中国西北石油勘探途中拍摄的山川地质以及人文风情,拍摄地主要有:北京、辽宁、河北、山西、陕西、天津等地。此套现藏于威斯康星大学密尔沃基分校图书馆。 ...继续阅读 (118)


    love 0
    未曾
    发表于
    2025-03-31 04:03:34

    丽人行图卷

    此《丽人行图》卷传为宋代李公麟绘。本幅根据杜甫《丽人行》一诗,描绘杨贵妃的姊姊虢国夫人等一行共九人,在初春时节骑马出游的情景。女子体态丰腴,额鼻皆留白。马匹肥壮,纵辔徐行,优游闲适。全作之人马造型、发髻衣饰、设色画法,皆有唐风。此卷现藏于台北故宫博物院。 ...继续阅读 (107)

上一页 18/840 下一页
沪ICP备19023445号-2号
友情链接