机器学习实践项目（一）- Rossman商店销售预测 - 认识这个项目

Rossmann operates over 3,000 drug stores in 7 European countries. Currently, Rossmann store managers are tasked with predicting their daily sales for up to six weeks in advance. Store sales are influenced by many factors, including promotions, competition, school and state holidays, seasonality, and locality. With thousands of individual managers predicting sales based on their unique circumstances, the accuracy of results can be quite varied.
In their first Kaggle competition, Rossmann is challenging you to predict 6 weeks of daily sales for 1,115 stores located across Germany. Reliable sales forecasts enable store managers to create effective staff schedules that increase productivity and motivation. By helping Rossmann create a robust prediction model, you will help store managers stay focused on what’s most important to them: their customers and their teams!
Prizes
–1st place - $15,000
–2nd place - $10,000
–3rd place - $5,000
In addition, a single $5,000 reward will go to the team whose methodology is implemented by Rossmann. This award may be given to a team at any position on the leaderboard.
Rossmann is interested in hiring top Kagglers from this competition. If you’re interested in a position with Rossmann, you may append “JOB” next to your team name for consideration.

罗仕曼（Rossmann）在欧洲7个国家运营着超过3,000家药店。目前，罗仕曼的门店经理需要提前预测最多六周的每日销售额。门店销售受多种因素影响，包括促销活动、竞争对手、学校与国家假期、季节性和地理位置。由于数千名经理根据各自独特的情况预测销售，结果的准确性可能差异较大。
在罗仕曼的首届Kaggle竞赛中，参赛者需为德国境内的1,115家门店预测未来6周的每日销售额。可靠的销售预测能让门店经理制定高效的员工排班计划，从而提升工作效率和团队积极性。通过帮助罗仕曼构建稳健的预测模型，参赛者将助力门店经理专注于最重要的任务：服务顾客与带领团队！
奖项设置
第一名 - 15,000美元
第二名 - 10,000美元
第三名 - 5,000美元
额外设立一项5,000美元奖金，颁发给其方法论被Rossmann采用的团队。该奖项可能授予排行榜上任意名次的团队。
Rossmann有意从本次竞赛中招募顶尖Kaggle选手。若对Rossmann的工作职位感兴趣，可在团队名称旁标注“JOB”以纳入考虑范围。

以上是这个项目在Kaggle上的介绍（项目连接），特地翻译成了中文。可见Kaggle上的竞赛还有奖金，如果表现出色可能还有工作机会，所以搞好机器学习还是为我们CS人能提供多一份的机会。

言归正传，我们继续认识这个项目。按照项目的要求，通过算法计算出来的销售金额是否正确，可以通过 均方根百分比误差(Root Mean Square Percentage Error (RMSPE)) 来评估，公式如下：

$$\text{RMSPE}=\sqrt{\frac{1}{n}\sum _{i=1}^n{\left(\frac{y_i-{\hat{y}}_i}{y_i}\right)}^2}\times 100\%$$

其中，$y_i$表示商店的实际销售金额，$\widehat{y_i}$表示经过算法预测的预测销售金额。

项目提供了完整的数据集供下载，但是前提是要正式参与这个项目，不过我已经把数据集上传到我的机器学习项目中了：代码仓库，数据集放在了 这里，有兴趣的同学可以去下载一下。