[Kaggle] Online Notebooks
前言
Let's go to https://www.kaggle.com/
Kaggle Notebook 有实践记录的案例。
一、线性拟合噪声数据
[Sklearn] Linear regression models to fit noisy data
二、打造 Pipeline
[Feature] Final pipeline: custom transformers
资源队列
阅读目录
- Algorithmic Trading Challenge25
- Allstate Purchase Prediction Challenge3
- Amazon.com – Employee Access Challenge6
- AMS 2013-2014 Solar Energy Prediction Contest2
- Belkin Energy Disaggregation Competition1
- Challenges in Representation Learning: Facial Expression Recognition Challenge4
- Challenges in Representation Learning: The Black Box Learning Challenge1
- Challenges in Representation Learning: Multi-modal Learning2
- Detecting Insults in Social Commentary
- EMI Music Data Science Hackathon
- Galaxy Zoo – The Galaxy Challenge
- Global Energy Forecasting Competition 2012 – Wind Forecasting
- KDD Cup 2013 – Author-Paper Identification Challenge (Track 1)2
- KDD Cup 2013 – Author Disambiguation Challenge (Track 2)1
- Large Scale Hierarchical Text Classification4
- Loan Default Prediction – Imperial College London
- Merck Molecular Activity Challenge1
- MLSP 2013 Bird Classification Challenge
- Observing the Dark World
- PAKDD 2014 – ASUS Malfunctional Components Prediction
- Personalize Expedia Hotel Searches – ICDM 2013
- Predicting a Biological Response1
- Predicting Closed Questions on Stack Overflow
- See Click Predict Fix1
- See Click Predict Fix – Hackathon1
- StumbleUpon Evergreen Classification Challenge
- [The Analytics Edge (15.071x)](The%20Analytics Edge (15.071x))
- The Marinexplore and Cornell University Whale Detection Challenge
- Walmart Recruiting – Store Sales Forecasting1
Kaggle比赛源代码和讨论的收集整理。
Algorithmic Trading Challenge25
Allstate Purchase Prediction Challenge3
- Rank 2 solution code21 by Alessandro Mariani.
- Rank 10 solution code5 by B1aine.
- Rank 36 solution cod1e by Hiroyuki.
- Rank 159 solution code by MrCanard.
- Solution thread.
Amazon.com – Employee Access Challenge6
- Rank 1 solution code24 by Paul Duan and Benjamin Solecki team.
- Rank 1 solution Q&A5 by Paul Duan.
- Rank 2 solution code1 by Owen Zhang.
- Rank 3 solution code3 by Dmitry & Leustagos.
- Rank 289 solution code by Foxtrot with original blog post here.
- Solution thread.
AMS 2013-2014 Solar Energy Prediction Contest2
- Rank 1 solution code5 and description4 by Leustagos team.
- Rank 2 solution code and description by Toulouse.
- Rank 3 solution code1 and description by Owen Zhang.
- Rank 4 solution escription by Peter Prettenhofer.
- Rank 5 solution description by Domcastro.
- Rank 58 solution code and description by Davit.
- Solution thread here.
- Ridge Regression starter code with MAE about 2.2M by Alec Radford, original thread here.
- Improved starter code by Foxtrot.
- Baseline code with MAE about 2.6M using Catmull-Rom Spline interpolation, also available in R here andhere.
Belkin Energy Disaggregation Competition1
Challenges in Representation Learning: Facial Expression Recognition Challenge4
- Rank 1 solution code6 and description2 by Charlie Tang.
- Rank 3 solution description3 by Maxim Milakov.
- Solution thread.
Challenges in Representation Learning: The Black Box Learning Challenge1
- Rank 1 solution description1 by David Thaler.
- Rank 2 solution code and description by sayit.
Challenges in Representation Learning: Multi-modal Learning2
- Rank 1 solution1 by MMDL.
- Solution thread.
Detecting Insults in Social Commentary
- Rank 1 solution description4 and code by Vivek Sharma.
- Rank 2 solution1 by tuzzeg.
- Rank 3 solution description Andrei Olariu.
- Rank 4 solution by Chris Brew.
- Rank 5 solution description by Yasser Tabandeh.
- Rank 6 solution by Andreas Mueller, code available here.
- Rank 8 solution description by Steve Poulson.
- Solution thread.
EMI Music Data Science Hackathon
- Rank 4 solution description1 by Steffen Rindle.
- Rank 18 solution code and description by Vlad Gusev.
- Rank 34 solution code and description by zenog.
- Solution thread.
Galaxy Zoo – The Galaxy Challenge
- Rank 1 solution code2 and description1 by Sander Dieleman.
- Rank 2 solution code and description by Maxim Milakov.
- Rank 3 solution code and description by tund.
- Rank 5 solution code and description by Julian de Wit.
- Rank 9 solution code and description by Soumith Chintala.
- Rank 13 solution code and description by Xiaoxiang Zhang.
- Rank 28 solution code and description by utdiscant.
- Rank 38 solution code and description by sugi.
- Rank 57 solution code and description1 by hxu.
- Rank 58 solution code and description by yr.
- Solution thread.
Global Energy Forecasting Competition 2012 – Wind Forecasting
- Rank 1 solution by Leustagos.
- Solution thread here1.
KDD Cup 2013 – Author-Paper Identification Challenge (Track 1)2
- Rank 1 solution with code and description4 by Team Algorithm, Github link to code here1.
KDD Cup 2013 – Author Disambiguation Challenge (Track 2)1
- Rank 1 solution with code and description4 by Team Algorithm, Github link to code here1.
- Rank 2 solution1 by SmallData Team.
- Rank 3 solution1 by hustmonk.
- Rank 4 solution1 by Ben S.
- Solution thread1.
Large Scale Hierarchical Text Classification4
- Rank 1 solution code and description7 by anttip.
- Rank 3 solution code2 and description2 by nagadomi.
- Solution thread one3.
- Solution thread two2.
Loan Default Prediction – Imperial College London
- Rank 2 solution and description1 by HelloWorld.
- Rank 12 solution and description by David McGarry.
- Solution thread.
Merck Molecular Activity Challenge1
MLSP 2013 Bird Classification Challenge
- Rank 1 solution code3 and description by beluga.
- Rank 2 solution code1 and description by Herbal Candy (W and thomeou).
- Rank 3 solution description by Anil Thomas.
- Rank 4 solution description by Maxim Milakov.
- Solution thread.
Observing the Dark World
- Rank 2 solution by Iain Murray, code available here.
PAKDD 2014 – ASUS Malfunctional Components Prediction
Personalize Expedia Hotel Searches – ICDM 2013
- Presentation paper/slides1 for ICDM 2013.
- Solution thread1.
Predicting a Biological Response1
- Rank 6 solution by Shea Parkes & Neil Schneider team.
- Rank 17 solution of Ensemble of RandomForests, GradientBoostingTrees and ExtraTreesRegressorby Emanuele Olivetti.
- Another solution code by Oblique Random Forest (oRF) by Shea Parkes & Neil Schneider team.
- The code of my best submission thread. Talks about Multi-core training Oblique Random Forests, and Stacking.
- Question about the process of ensemble learning thread. Talks about applying ensembles in practice, and how can problems arise and how to deal with them.
Predicting Closed Questions on Stack Overflow
- Rank 10 solution by Marco Lui.
- Rank 33 solution by Foxtrot.
See Click Predict Fix1
See Click Predict Fix – Hackathon1
StumbleUpon Evergreen Classification Challenge
- Benchmark beater 1.
- Benchmark beater 2.
- Benchmark beater 3.
- Solution thread.
- My own solution, which is a good example of what is overfitting. (Public rank: 57, Private rank: 291)
[The Analytics Edge (15.071x)](The%20Analytics Edge (15.071x))
- Rank 17 solution code and description by Foxtrot.
- Solution thread.
The Marinexplore and Cornell University Whale Detection Challenge
- Rank 1 solution by Nick Kridler.
- Rank 7 solution by Gilles Louppe and Peter Prettenhofer team.
- Rank 8 solution by Sander Dieleman.
- Rank 56 solution by Sudeep Juvekar.
- Solution discussion thread.
- Mean spectogram thread.
- Official interview from the Marinexplorer and Cornell at Kaggle.
Walmart Recruiting – Store Sales Forecasting1
- Rank 1 solution code5 and description by David Thaler.
- Rank 2 solution description1 by sriok.
- Rank 3 solution code and description1 by James King.
- Rank 5 solution description by ACS69.
- Rank 6 solution description by T. Henry.
- Rank 8 solution description by BreakfastPirate.
- Rank 9 solution description by Neil Summers.
- Rank 10 solution description by Gilberto Titericz Junior.
- Rank 11 solution description by citynight.
- Rank 16 solution code and description by yr.
- Rank 29 solution code and description by Mike Kim.
- Rank 30 solution description by dkay.
- Solution thread.
Thank you Foxtrot, James Petterson, Ben S for providing some of the links and solutions above.
[Kaggle] Online Notebooks的更多相关文章
- 使用Graphlab参加Kaggle比赛(2017-08-20 发布于知乎)
之前用学生证在graphlab上申了一年的graphlab使用权(华盛顿大学机器学习课程需要)然后今天突然想到完全可以用这个东东来参加kaggle. 下午参考了一篇教程,把notebook上面的写好了 ...
- kaggle——NFL Big Data Bowl 2020 Official Starter Notebook
Introduction In this competition you will predict how many yards a team will gain on a rushing play ...
- kaggle新手如何在平台学习大神的代码
原创:数据臭皮匠 [导读]Kaggle ,作为听说它很牛X但从未接触过的同学,可能仅仅了解这是一个参加数据挖掘比赛的网站,殊不知Kaggle也会有赛题相关的数据集, 比如我们熟知的房价预测.泰坦尼克 ...
- TorchVision Faster R-CNN 微调，实战 Kaggle 小麦检测
本文将利用 TorchVision Faster R-CNN 预训练模型,于 Kaggle: 全球小麦检测 上实践迁移学习中的一种常用技术:微调(fine tuning). 本文相关的 Kaggle ...
- 基于Colab Pro & Google Drive的Kaggle实战
原文:https://hippocampus-garden.com/kaggle_colab/ 原文标题:How to Kaggle with Colab Pro & Google Drive ...
- kaggle入门2——改进特征
1:改进我们的特征 在上一个任务中,我们完成了我们在Kaggle上一个机器学习比赛的第一个比赛提交泰坦尼克号:灾难中的机器学习. 可是我们提交的分数并不是非常高.有三种主要的方法可以让我们能够提高他: ...
- Kaggle入门教程
此为中文翻译版 1:竞赛 我们将学习如何为Kaggle竞赛生成一个提交答案(submisson).Kaggle是一个你通过完成算法和全世界机器学习从业者进行竞赛的网站.如果你的算法精度是给出数据集中最 ...
- 如何使用Python在Kaggle竞赛中成为Top15
如何使用Python在Kaggle竞赛中成为Top15 Kaggle比赛是一个学习数据科学和投资时间的非常的方式,我自己通过Kaggle学习到了很多数据科学的概念和思想,在我学习编程之后的几个月就开始 ...
- kaggle实战记录 =>Digit Recognizer
date:2016-09-13 今天开始注册了kaggle,从digit recognizer开始学习, 由于是第一个案例对于整个流程目前我还不够了解,首先了解大神是怎么运行怎么构思,然后模仿.这样的 ...
随机推荐
- clipse调试web项目配置调试配置——没有Server
文章:eclipse环境下如何配置tomcat(包含没有Server解决办法) 地址:https://blog.csdn.net/TimliangL/article/details/78882566
- everything in javascript can act like an object,
- 解读 v8 排序源码
前言 v8 是 Chrome 的 JavaScript 引擎,其中关于数组的排序完全采用了 JavaScript 实现. 排序采用的算法跟数组的长度有关,当数组长度小于等于 10 时,采用插入排序,大 ...
- Go语言在国产CPU平台上应用前景的探索与思考
http://www.chinaaet.com/article/3000087559 0 引言 CPU是电子产品的核心,代表着信息产业的发展水平.CPU发展至今已经有四十多年的历史了,实际就是Inte ...
- Dubbo源码分析：Serialization
背景 顺序化逻缉处理! 类图 获取Serialization对象时序图 序列化
- 应用安全测试技术DAST、SAST、IAST对比分析【转】
转自:https://blog.csdn.net/qq_29277155/article/details/92411079 一.全球面临软件安全危机 2010年,大型社交网站rockyou.com被曝 ...
- 铺砖头问题（完美）——爆搜&&插头DP
题意 给定一个 $n \times m$ 的格子,每个格子被染成了黑色或白色.现在要用 $1 \times 2$ 的砖块覆盖这些格子,要求块与块之间互不重叠,且覆盖了所有白色的格子,但不覆盖任意黑色格 ...
- Bias, Variance and the Trade-off
偏差,方差以及两者权衡 偏差是由模型简化的假设,使目标函数更容易学习. 一般来说,参数化算法有很高的偏差,使它们学习起来更快,更容易理解,但通常不那么灵活.反过来,它们在复杂问题上的预测性能更低,无法 ...
- Java web开发——文件的上传和下载
一. 功能性需求与非功能性需求 要求操作便利,一次选择多个文件和文件夹进行上传:支持PC端全平台操作系统,Windows,Linux,Mac 支持文件和文件夹的批量下载,断点续传.刷新页面后继续传输. ...
- CLR内部异常(中)
不捕捉某一个异常 常常有这种情况,代码不需要捕捉异常,但需要执行一些清理或者修正操作.虽然不总是,支持物(holders)经常用在这种场景里.在支持物(holders)不适用的情况里,CLR提供了两个 ...