机器学习非参数学习_非技术人员学习机器学习的不可思议的技巧

机器学习非参数学习

“A breakthrough in machine learning would be worth ten Microsofts.” — Bill Gates

“机器学习方面的突破将使十个微软值得。” - 比尔盖茨

In science and engineering, from computer systems to physics, machine learning methods are widely employed. Besides, marketing sites such as search engines, recommenders (i.e. Netflix, Amazon), advertisers and financial institutions use machine learning for content guideline, prediction of customer experience, compliance, or risk.

在科学和工程领域，从计算机系统到物理，机器学习方法被广泛采用。 此外，搜索引擎，推荐器(例如Netflix，Amazon)，广告商和金融机构等营销网站都使用机器学习来获取内容指南，预测客户体验，合规性或风险。

Machine learning as a discipline aims to develop and understand computer programmes that learn from experience to predict or control them.

机器学习作为一门学科，旨在开发和理解从经验中学习以预测或控制它们的计算机程序。

As a result, machine learning has become quite popular in recent years, and many machine learning tools have been developed. Some classes teach you how to learn and understand, even though you don’t have advanced knowledge of programming and mathematics at the university level.

结果，机器学习近年来变得非常流行，并且已经开发了许多机器学习工具。 即使您没有大学一级的编程和数学知识，有些课程也会教您如何学习和理解。

First, let us watch this short video and understand the fundamentals of Machine Learning. This clip allows you to understand what machine learning is, what are machine learning types, how it functions with basic examples as well as how machine learning is applied in different industries.

首先，让我们观看这段简短的视频并了解机器学习的基础知识。 该剪辑使您能够了解什么是机器学习，什么是机器学习类型，它如何通过基本示例运行以及如何在不同行业中应用机器学习。

演示地址

Simplilearn Simplilearn

Now, if you are a non-techie and are interested in machine learning, then you might be wondering and have the following set of questions in your mind:

现在，如果您不是技术人员，并且对机器学习感兴趣，那么您可能想知道并想到以下问题：

1. How to better understand machine learning if I don’t have enough basic skills/technical background?
   如果我没有足够的基本技能/技术背景，如何更好地理解机器学习？
2. Should I be part of any institute that teaches machine learning?
   我应该加入任何教授机器学习的学院吗？
3. How much would it charge me to learn machine learning?
   学习机器学习需要多少费用？
4. Would it be too challenging to learn mathematics involved in machine learning?
   学习机器学习中涉及的数学是否太有挑战性？

Machine learning uses a lot of mathematics and programming. Can a non-technical person, therefore, learn machine learning? Yes, even if you are a person not having a technical background, you can learn machine learning if you follow the right course.

机器学习使用了大量的数学和程序设计。 因此，非技术人员可以学习机器学习吗？ 是的，即使您不是一个技术背景的人，只要遵循正确的课程，就可以学习机器学习。

Here, I’ll show you the right way to address the responses to the questions earlier in the thread as well as suggestion for you to choose the courses to learn machine learning without having foundational knowledge. It would be best if you took the concrete initiatives below,

在这里，我将向您展示正确处理方法的方法，以解决该主题前面部分对问题的回答，并建议您选择无需基础知识即可学习机器学习的课程 。 最好采取以下具体措施，

• Take Andrew Ng’s machine learning course, which has no prerequisites.

  参加Andrew Ng的机器学习课程 ，该课程没有任何先决条件。

• Learn Python from MITx course on edX

  在edX上的MITx课程中学习Python

• Learn calculus, linear algebra and probability and statistics on edX

  学习微积分 ， 线性代数和概率 统计 在edX上

• Learn Python data analysis with this specific Youtube series and these lecture notes

  通过这个特定的Youtube系列和这些讲义学习Python数据分析

• Take the deep learning course by Andrew Ng

  参加Andrew Ng的深度学习课程

• Watch this YouTube series on how to apply machine learning in Python

  观看此YouTube系列文章 ，了解如何在Python中应用机器学习

• Start to apply your machine learning expertise for hands-on experience on Python, machine learning, and deep learning on Kaggle competitions

  开始将您的机器学习专业知识应用于Python的实践经验，机器学习以及Kaggle竞赛的深度学习

The reason I advise you to begin with the Andrew Ng course before you start to learn programming is that it gives you an opportunity to see if machine learning is something that you’d be interested in before spending a plenty of time on it.

我建议您在开始学习编程之前就先开始吴德华(Andrew Ng)课程，原因是它使您有机会在花大量时间学习机器学习之前是否对它感兴趣。

技术机器学习的方式 (How technical machine learning is)

Machine learning is, overall, quite a technical area as it uses programming, statistics and mathematics to apply mathematical algorithms to data.

总体而言，机器学习是一个技术领域，因为它使用编程，统计和数学将数学算法应用于数据。

To use the machine learning models, you may need to know how to code in a programming language such as Python.

要使用机器学习模型，您可能需要知道如何使用Python之类的编程语言进行编码。

The computer models use a lot of mathematics to function. Linear algebra, calculus and probability theory are part of the mathematics used in machine learning models. Nevertheless, some courses teach you how models of machine learning work if you do not already know about mathematics. I recommend that you start there and then learn more about machine learning mathematics.

计算机模型使用许多数学函数。 线性代数，微积分和概率论是机器学习模型中使用的数学的一部分。 尽管如此，如果您还不了解数学，一些课程会教您如何学习机器学习模型。 我建议您从那里开始，然后进一步了解机器学习数学。

If you want to work with large data sets or want a career in machine learning, then you will also have to learn SQL.

如果您想使用大型数据集或想要从事机器学习，那么您还必须学习SQL。

如果我没有足够的基本技能/技术背景，如何更好地理解机器学习？ (How to better understand machine learning if I don’t have enough basic skills/technical background?)

I would suggest you continue with the course Andrew Ng from Stanford University if you are interested to learn machine learning.

如果您想学习机器学习，我建议您继续学习斯坦福大学的Andrew Ng课程。

In this course, you will learn about the most efficient methods of machine learning and how to practice them and allow them to work for yourself. Above all, you will learn not just the theoretical concepts of learning but also the practical know-how to apply these concepts efficiently and effectively to new problems. It is perhaps the most popular machine learning course as Andrew Ng does an outstanding job of intuitively describing machine learning and models.

在本课程中，您将学习最有效的机器学习方法，以及如何实践它们并使它们自己工作。 最重要的是，您不仅会学习学习的理论概念，还将学到将这些概念有效地应用于新问题的实践知识。 这可能是最受欢迎的机器学习课程，因为Andrew Ng在直观描述机器学习和模型方面做得非常出色。

I advise you to start there so you can see if machine learning is something you’re interested in before you spend a lot of time studying programming and mathematics.

我建议您从那里开始，这样您就可以在花大量时间学习编程和数学之前查看一下机器学习是否对您感兴趣。

如何学习和练习机器学习编码？ (How to learn and practice machine learning coding?)

Before being able to implement the learning models of the machine, you would need to know how to program.

在实现机器的学习模型之前，您需要了解如何编程。

What is the smartest programming language to practice for machine learning?

机器学习中最聪明的编程语言是什么？

Now, watch the short video here to explore more about this.

现在，请在此处观看短片以进一步了解。

演示地址

Siraj Raval Siraj Raval

The most preferred programming language will be Python. Because it is the most common machine language, it is relatively easy to study, and it has several integrated libraries which allow you to implement algorithms for machine learning quickly. To order to learn Python, I would encourage you to take this MIT course on edX.

最优选的编程语言是Python。 因为它是最常见的机器语言，所以它相对容易学习，并且它具有几个集成的库，可让您快速实现用于机器学习的算法。 为了学习Python，我鼓励您参加edX上的MIT课程 。

When you have learned how to program in Python, I will encourage you to do some simple Python data analysis. Data analysis is used a great deal in machine learning, particularly in preparing the data to function well with the machine learning models. For learn data analysis in Python, I would recommend you use this YouTube series to learn pandas and Numpy with these lecture notes.

当您学习了如何使用Python进行编程时，我将鼓励您进行一些简单的Python数据分析。 数据分析在机器学习中被大量使用，特别是在准备数据以使其与机器学习模型良好配合的过程中。 要使用Python学习数据分析，我建议您使用本YouTube系列丛书 ，以这些讲义学习熊猫和Numpy。

For machine learning, there are several other programming languages, and one of them is R. It is a programming language built specifically for statistics and is therefore also very popular in statistical fields. This has a variety of built-in libraries and a reasonably large community of support. However, I would recommend that since most machine tutorials are in Python, you learn Python over R.

对于机器学习，还有其他几种编程语言，其中一种是R。这是一种专门为统计而构建的编程语言，因此在统计领域也非常流行。 它具有各种内置库和相当大的支持社区。 但是，由于大多数机器教程都使用Python，因此我建议您使用R来学习Python。

获得编程技能后该怎么办？ (What to do after acquiring programming skills?)

I advise you to take the course on deep learning taught by Andrew Ng after knowing Python.

我建议您在了解Python之后参加由Andrew Ng教授的深度学习课程 。

Deep learning is a subset of machine learning with algorithms that function very well when a lot of data is available. We have become very common in recent years because of the sharp increase in data available to companies.

深度学习是机器学习的一个子集，其算法在大量数据可用时运行良好。 近年来，由于公司可用数据的急剧增加，我们变得非常普遍。

First, it is worth taking a more practical training course on the computer that will teach you how to apply the various Python machine learning algorithms. I‘d advise this book. Nonetheless, it will be a great option to watch this YouTube series.

首先，值得在计算机上参加更实际的培训课程，该课程将教您如何应用各种Python机器学习算法。 我建议这本书 。 尽管如此，观看此YouTube系列视频还是一个不错的选择。

You will now be capable of applying your own machine learning algorithms on your datasets. You will do so by using your personal data sets to use machine learning or compete in Kaggle machine learning competitions.

您现在可以将自己的机器学习算法应用于数据集。 您将通过使用您的个人数据集来使用机器学习或参加Kaggle机器学习竞赛来做到这一点。

学习机器学习中涉及的数学是否太有挑战性？ (Would it be too challenging to learn mathematics involved in machine learning?)

If you’d like to understand how mathematically the machine learning models behave, how the machine learning methods can be optimized wholly, and if you intend to have a career or research role in machine learning, so you need to study maths related to it.

如果您想了解机器学习模型在数学上的行为方式，如何全面优化机器学习方法，以及您打算在机器学习中担任职业或研究角色，那么您需要学习与之相关的数学。

In mathematics, we need to learn and practice three essential topics, which are given below:

在数学中，我们需要学习和实践以下三个基本主题：

• Calculus(Integration, Derivatives and Partial Derivatives) by MIT

  麻省理工学院的微积分 ( 积分，导数和偏导数 )

  Calculus(Integration, Derivatives and Partial Derivatives) by MITNote: This youtube playlist will give you a thorough understanding of Calculus.

  麻省理工学院的微积分 ( 积分，微分和偏导数 ) 注：此youtube播放列表将使您对微积分有透彻的了解。

• Probability and Statistics by MIT

  麻省理工学院的概率 统计

  Probability and Statistics by MITNote: The link here brings you to a great website with engaging graphs/animations/charts that allow you to play and understand the concepts.

  麻省理工学院的概率与统计 注： 这里的链接带您进入一个具有吸引力的图形/动画/图表的出色网站，使您可以玩耍和理解概念。

• Youtube series of Linear Algebra (The University of Texas at Austin)

  YouTube系列线性代数(德克萨斯大学奥斯汀分校)

  Youtube series of Linear Algebra (The University of Texas at Austin)Note: I would also suggest this playlist for a conceptual understanding of Linear algebra.

  Youtube系列线性代数(德克萨斯大学奥斯汀分校) 注：我也建议使用此播放列表 ，以从概念上理解线性代数。

机器学习的高级课程 (A premium course of machine learning)

If you need there are premium machine learning courses as well, from which you can learn from no prior knowledge in programming to be able to develop machine learning models on your own.

如果需要，还提供高级机器学习课程，您无需学习任何编程知识就可以从中学习，从而能够自行开发机器学习模型。

The paid course that I would suggest the one offered by DataCamp, which would teach you Python skills needed by a data scientist to succeed, Mathematics, and how machine learning models can be applied.

我会推荐DataCamp提供的付费课程，该课程将教您数据科学家成功所需要的Python技能，数学以及如何应用机器学习模型。

You can get the free unlimited access of DataCamp through September, 9.

您可以在9月9日之前免费无限次访问DataCamp。

关键要点 (The key takeaway)

After the above steps, learn to read and implement research articles for deeper understanding, try as well to take part in hackathons and Kaggle competitions to prove your skills.

完成上述步骤后，学习阅读和实施研究文章以加深理解，并尝试参加黑客马拉松和Kaggle竞赛以证明您的技能。

So now start to learn machine learning. Best wishes!

因此，现在开始学习机器学习。 最好的祝愿！

If you have any question in your mind or have suggestions, please comment below.

如果您有任何疑问或建议，请在下面评论。

Thanks for the reading.

感谢您的阅读。

翻译自: https://medium.com/datadriveninvestor/incredible-tips-for-a-non-techie-to-learn-machine-learning-d86e36a88e94

机器学习非参数学习