Working in the fast paced world of data science? Here are some habits that are bound to increase your productivity.

在快速发展的数据科学世界中工作？ 这里有一些习惯必将提高您的生产力。

Data science is quite the trend right now. The pay is high and great opportunities abound. Its applications are numerous and many industries, tech or not, are beginning to see the importance of making sense of their data.

数据科学是当前的趋势。 薪水很高，机会很多。 它的应用广泛，无论有无技术，许多行业都开始意识到理解其数据的重要性。

While it is a hot topic right now, here are 8 habits that will make you a better data scientist.

虽然这是当前的热门话题，但这里有8个习惯可以使您成为更好的数据科学家。

了解工作角色 (Know the Job Roles)

Before you apply for any job, make sure you know what you are getting yourself into. There are many job roles/titles that exist and you need to be familiar with them and know just what you will be doing. It is also good to note that even with clearly defined roles, you can find yourself doing much more depending on the company you work for. So be ready.

在申请任何工作之前，请确保您知道自己正在从事什么。 存在许多工作角色/职务，您需要熟悉它们并知道您将要做什么。 还要注意的是，即使角色定义明确，您也会发现自己在做更多的工作，这取决于您所工作的公司。 所以准备好

Some of the job roles that exist include:

存在的一些工作角色包括：

• Data scientist
  数据科学家
• Data analyst
  数据分析师
• Business analyst
  业务分析师
• Machine learning engineer
  机器学习工程师
• Database administrator
  数据库管理员

… To name a few. While they all require the same basic skill set (Python, R, SQL) and knowledge with working with data, their requirements can differ. So before you apply for that job, check the job description and see if they align with your skill and ability.

……仅举几例。 尽管它们都需要相同的基本技能(Python，R，SQL)和处理数据的知识，但它们的要求可能有所不同。 因此，在申请该职位之前，请检查职位描述，并查看它们是否与您的技能和能力相符。

Stress on skill because you’re going to need a lot of it.

强调技能，因为您将需要很多技能。

时间意识 (Time Conscious)

You need to be time conscious for various reasons. If you are working on a project and you can’t seem to get it right, you should know when to drop it. You can’t do the same thing and expect different results.

出于各种原因，您需要时间意识。 如果您正在开发一个项目，但似乎无法正确完成，则应该知道何时删除它。 您不能做相同的事情并且期望得到不同的结果。

In my experience, when a project seems overwhelming I do one of the following:

以我的经验，当一个项目似乎不堪重负时，我将执行以下操作之一：

• Rest: Sometimes its just fatigue and you need to rest your head and come at the work with renewed vigor. You can use this time to engage in any activity really that makes you happy and relieves stress.

  休息 ：有时它只是疲劳，您需要休息一下，以新的活力来工作。 您可以利用这段时间从事任何真正能让您开心并减轻压力的活动。

• Search for similar problems and see how they were solved. A good place for this is Kaggle. You can go through notebooks on kaggle on different topics. You can also take a short course or go through examples in a data science textbook.
  搜索类似的问题，看看如何解决。 Kaggle是个不错的选择。 您可以在kaggle上浏览有关不同主题的笔记本。 您还可以参加一门简短的课程，或者查阅数据科学教科书中的示例。

• Abandon ship: Yes, when work is pulling you under, you need to know when to let it go. It’s not easy especially when you feel like one more try will get you there. But, if you work in an office environment, they expect results and telling your boss you’ve been working on one project for a month with no solution in view can come off the wrong way. You don’t want that, trust me.

  弃船 ：是的，当工作使您陷入困境时，您需要知道何时放手。 这并非易事，尤其是当您觉得再尝试一次就能到达那里时。 但是，如果您在办公室环境中工作，他们会期望结果，并告诉您的老板您已经在一个项目上工作了一个月而没有解决方案，这可能会出错。 你不要那个，相信我。

请求帮忙 (Ask for Help)

One thing I love about data science is the community. Its honestly amazing — online and offline. When I’m stuck on a problem, I’m very quick to call a friend or search online for solutions. The words no one is an island of knowledge really rings through here.

我喜欢数据科学的一件事是社区。 老实说，它令人惊讶-在线和离线。 当我遇到问题时，我会很快打电话给朋友或在线搜索解决方案。 “ 没有人是知识之岛 ”这两个词真的在这里响彻了头。

Platforms such as Stack Overflow, Kaggle and the rest are extremely useful. Another good source of help is Discussion Forums. I admit, I never saw the use of discussion forums at first and I always felt it was a distraction especially when taking a course. But I was wrong. The right discussion forum is one with people of like interest and goals such as discussion forums for competitions or courses or it could be you and your friends coming together to help each other out. It is also a good way to network, so keep that in mind.

诸如Stack Overflow ， Kaggle等平台非常有用。 另一个有用的帮助资源是“ 论坛” 。 我承认，起初我从未见过使用讨论论坛，而且我总是觉得这很分散注意力，尤其是在上课时。 但是我错了。 合适的讨论论坛是一个有兴趣和目标的人，例如比赛或课程的讨论论坛，也可以是您和您的朋友聚在一起互相帮助。 这也是建立联系的一种好方法，因此请记住这一点。

Part of the mistakes I made when I started off was thinking I could do it all. I had no mentor or guidance and I just went off binge watching one tutorial after another. I got the skills alright but I also had many abandoned projects that I could not finish because I didn’t have any help when I was lost.

一开始我犯的部分错误是我认为自己可以做到。 我没有导师或指导，只是暴饮暴食，看着一个又一个的教程。 我的技能还不错，但是我也有很多无法完成的废弃项目，因为迷路时我没有任何帮助。

Even if you are a shy person, you can enjoy the anonymity of asking questions online or simply searching the web.

即使您是一个害羞的人，您也可以享受在网上匿名提问或只是在网上搜索的匿名性。

编码可重用功能 (Coding Re-usable functions)

Some of the process when dealing with data can be repetitive. Spare yourself the stress of having to write the same code over and over again.

处理数据时的某些过程可能是重复的。 避免自己不得不一次又一次地编写相同的代码。

An example of a process which you can automate through functions is filling missing data. You can write a function to loop through the columns and fill in the missing values. You should try that now if you haven’t.

您可以通过函数自动执行的过程示例是填充丢失的数据。 您可以编写一个函数以遍历各列并填写缺少的值。 如果还没有，应该立即尝试。

If you started as a programmer before diving into data science, you will be familiar with the joy of functions. I’ve gone through a couple of data science courses and I haven’t heard them mention functions or how useful functions are even when I see them using functions.

如果您是在开始从事数据科学之前是一名程序员，那么您将熟悉功能的乐趣。 我已经完成了几门数据科学课程，但我还没有听说过它们提到函数，甚至即使我看到使用函数的函数也没有多有用。

Functions can save you a lot of time and increase your productivity.

功能可以节省大量时间并提高生产率。

注释和文档 (Comments and Documentation)

A simple ‘This code does this’ or ‘This function does that’ is enough. When I started learning how to program, I did not see the need for comments. I always felt well, I wrote the code, how can I not know what it does.

一个简单的“此代码可以执行此操作”或“此函数可以执行此操作”就足够了。 当我开始学习如何编程时，我没有看到注释的必要。 我一直感觉很好，我编写了代码，我怎么不知道它的作用。

However, if you do happen to drop a project for sometime and return to it, things can get muddled up. For one, you may not remember the thought process that led you to do a particular activity. Maybe you multiplied a particular feature by a constant for some reason and now you’re looking at your work wondering why.

但是，如果您确实碰巧放弃某个项目并返回该项目，则事情可能会变得混乱。 首先，您可能不记得导致您进行特定活动的思考过程。 也许出于某种原因，您将某个特定功能乘以常数，现在您正在寻找原因。

Comments and documentation are especially useful when working in teams. You don’t want your teammates getting lost neither do you want to keep answering the same question numerous times.

团队合作时，注释和文档特别有用。 您不希望您的队友迷路，也不想多次回答相同的问题。

You also want to document any failed processes so you or your teammates don’t repeat the same mistakes.

您还希望记录任何失败的流程，以免您或您的团队成员重复同样的错误。

Commenting and documenting increase productivity and is a good habit to develop. You can start by writing one line comment describing each part of your personal work and before you know it, you become an expert documenter.

评论和记录可提高生产率，是养成良好的习惯。 您可以先写一行注释来描述您个人工作的每个部分，然后在您知道之前，您就成为了专业的文档编制者 。

自我发展 (Self Development)

Being a data scientist is so savvy these days that people often dive head first into it and don’t come up for air. As a data scientist, you need a blend of soft and hard skills. Yes, you can deploy models and what not but so can a lot of people. You need something to set you aside from the pack. Soft skills.

如今，成为一名数据科学家是如此的机灵，以至于人们经常首先涉足其中，而不是一无所获。 作为数据科学家，您需要软和硬技能的融合。 是的，您可以部署模型，但不能部署模型，但是很多人也可以。 您需要一些东西来将您放在一边。 软技能。

Data scientist find themselves in different industries performing various tasks, under different titles such as Data engineer, ML engineer, software engineer, business analyst and so on. You need to have enough domain knowledge of the field you are working it. This will guide the techniques you employ and help you look at a problem critically. Having domain knowledge will also prevent you from going down rabbit holes. You have a clearer purpose and you can better allocate your resources to relevant problems

数据科学家发现自己身处不同行业中，以不同的职务执行不同的任务，例如数据工程师，机器学习工程师，软件工程师，业务分析师等等。 您需要对正在工作的领域有足够的领域知识 。 这将引导你使用你的问题批评的眼光看待技术和帮助。 拥有领域知识还可以防止您陷入困境。 您的目标更加明确，可以更好地将资源分配给相关问题

You also need to be a good communicator. After you have analyzed data and come up with your solution or next steps, your next job is to communicate your findings. The best way you can do this is through visualizations and storytelling. Your intricate models will not matter if you cannot effectively communicate your results and sell your idea to your company.

您还需要成为一个良好的沟通者。 分析数据并提出解决方案或后续步骤之后，下一个工作就是传达您的发现。 做到这一点的最佳方法是可视化和讲故事 。 如果您无法有效地传达您的结果并将您的想法出售给公司，那么复杂的模型将无关紧要。

However if your company doesn’t take your recommendation, don’t feel bad. It might not be the right time just yet but your efforts would be appreciated.

但是，如果您的公司不接受您的建议，那就不要难过。 可能还不是时候，但是您的努力将不胜感激。

So as you acquire technical skills, make effort to develop those immeasurable skills that will make you a true asset.

因此，当您掌握技术技能时，请努力发展那些不可估量的技能，这些技能将使您成为真正的资产。

不要成为完美主义者 (Don’t be a perfectionist)

Yes, you read that right. The best way to become an expert data scientist is by doing projects. You’ve taken python courses, R, data science and machine learning but you’re still looking for another course to try because you don’t feel you’re quite there yet.

是的，你看的没错。 成为专家数据科学家的最佳方法是进行项目。 您已经学过python课程，R，数据科学和机器学习，但是您仍在寻找另一门课程，因为您还不觉得自己去那儿了。

It could happen to anybody. I took a couple of python courses but still referred to myself as a beginner because although I took all those courses, I did not feel confident in my ability. I had not tried them on my own. What I essentially did was follow courses from beginning to end and then nothing. No personal projects, No practice.

任何人都可能发生。 我参加了几门python课程，但仍然称自己为初学者，因为尽管我参加了所有这些课程，但我对自己的能力并不自信。 我没有自己尝试过。 我本质上所做的就是从头到尾地学习课程，然后什么也不做。 没有个人项目，没有实践。

If you do find yourself doing the same thing, here I am telling you to stop. Get some data and practice the basics you know. See what results you can get and then work to get better results.

如果您确实发现自己在做同样的事情，请告诉我停下来。 获取一些数据并练习您知道的基础知识。 查看您可以获得什么结果，然后努力获得更好的结果。

The best way to learn data science is to do data science. The icing on the cake is that you get to build your portfolio at the same time. How awesome is that?!

学习数据科学的最好方法是做数据科学。 锦上添花的是，您可以同时建立自己的投资组合。 那太棒了!!

回顾你的工作 (Review your work)

Make sure all parts of your project work. Remove any repetitive (use functions) or irrelevant sections.

确保项目的所有部分都在工作。 删除所有重复的(使用功能)或无关的部分。

Make sure you made the right assumptions, run unit tests on your model. Train, validate and test.

确保您做出正确的假设，对模型运行单元测试。 培训，验证和测试 。

Tie the bow up nicely in your projects, no matter how small they are.

不管项目多么小，都可以在项目中很好地打结。

If you form this habit when working with small projects, you will easily do the same in major projects.

如果您在处理小型项目时养成这种习惯，那么在大型项目中您将很容易这样做。

结论 (CONCLUSION)

The path to becoming a data scientist is a fast one. It can take you some months, few weeks or even days. With the amazing tutorials and resources available online, you can learn anything you want to learn. You just need a little determination. These habits will spur you onwards and help you become your very best.

成为数据科学家的道路是一条捷径。 这可能需要几个月，几周甚至几天的时间。 借助在线提供的出色教程和资源，您可以学到任何想要学习的东西。 您只需要一点决心。 这些习惯会刺激您，并帮助您成为最好的。

What other habits do you feel make for an effective data scientist?

对于有效的数据科学家，您觉得还有哪些其他习惯？