机器学习入门�_机器学习入门

机器学习入门�

什么是机器学习？ (What is Machine Learning?)

Machine Learning is the science (and art) of programming computers so they can learn from data.

机器学习是编程计算机的科学(和艺术)，因此它们可以从数据中学习。

General Definition:

一般定义：

Machine Learning is the field of study that gives computers the ability to learn without being explicitly programmed. [Arthur Samuel, 1959 ]

机器学习是使计算机无需明确编程即可学习的能力的研究领域。 [亚瑟·塞缪尔，1959年]

Applied Machine Learning: Here applied means how to interpret the machine learning algorithms at a high level.

应用的机器学习：这里的应用是指如何从较高层次解释机器学习算法。

为什么要机器学习？ (Why Machine Learning?)

Consider how you would write a spam filter using traditional programming techniques.

考虑一下如何使用传统的编程技术编写垃圾邮件过滤器。

Traditional Programming Approach

传统编程方法

Machine Learning Approach

机器学习方法

Machine Learning Approach

机器学习方法

Machine Learning Approach

机器学习方法

Machine Learning is Great For:

机器学习非常适合：

• Problems for which existing solutions require a lot of hand-tuning.
  现有解决方案的问题需要大量的手动调整。
• For complex problems which can not solved by using traditional approach.
  对于传统方法无法解决的复杂问题。
• For fluctuating environments where machine learning can adopt a new data easily.
  在不断变化的环境中，机器学习可以轻松采用新数据。

机器学习中的关键概念 (Key Concept In Machine Learning)

Machine Learning systems can be classified according to the amount and type of supervision they get during training.

可以根据培训期间获得的监督的数量和类型对机器学习系统进行分类。

Supervised learning (Train Me!) : Once the model gets trained it can start making a prediction or decision when new data is given to it.

监督学习(训练我！)：一旦训练了模型，就可以开始进行预测或做出决定，何时提供新数据。

Supervised Algorithms:

监督算法：

k-Nearest Neighbors, Linear Regression ,Logistic Regression ,Support Vector Machines (SVMs) ,Decision Trees and Random Forests.

k最近邻，线性回归，逻辑回归，支持向量机(SVM)，决策树和随机森林。

Unsupervised Learning: I am self sufficient in learning.

无监督学习：我在学习方面是自已的。

Unsupervised Algorithms: Clustering k-Means ,Hierarchical Cluster Analysis (HCA) ,Principal Component Analysis (PCA).

无监督算法：聚类k均值，层次聚类分析(HCA)，主成分分析(PCA)。

Semi supervised Learning:

半监督学习：

• Semi supervised learning algorithms are combinations of unsupervised and supervised algorithms.
  半监督学习算法是非监督算法和监督算法的组合。
• These Algorithms deals with partially labeled training data,usually a lot of unlabeled data and a little bit of labeled data
  这些算法处理部分标记的训练数据，通常是很多未标记的数据和少量标记的数据

• For example, deep belief networks (DBNs)

  例如，深度信任网络( DBN)

Reinforcement learning — my life my rules! (Hit & trial)

强化学习-我的生活就是我的规则！ (命中和试用)

• This Machine learning system use ‘Agent’ who can observe the environment, select and perform actions, and get rewards in return.
  该机器学习系统使用“ Agent”，他可以观察环境，选择和执行动作，并获得回报。

• For example, many robots implement Reinforcement Learning algorithms to learn how to walk. DeepMind’s AlphaGo program is also a good example of Reinforcement Learning.

  例如，许多机器人都采用了强化学习算法来学习如何走路。 DeepMind的AlphaGo程序也是强化学习的一个很好的例子。

您将如何应用机器学习来解决您的问题？ (How would you apply machine learning to solve your problem?)

There are three basic steps:

有三个基本步骤：

• Representation:How to represent learning problems in terms of something that a computer can understand. In this step, you have to also decide what kind of algorithm to apply to selected data.

  表示法 ：如何用计算机可以理解的东西来表示学习问题。 在此步骤中，您还必须决定对所选数据应用哪种算法。

• Evaluation: In this step, we can calculate the quality and accuracy score for the prediction coming out from the machine learning algorithm typically called classifier.

  评估 ：在这一步中，我们可以计算出机器学习算法(通常称为分类器)得出的预测的质量和准确性得分。

• Optimisation: In this step, we need to search for an optimal classifier that gives the best outcome for the selected problem.

  优化 ：在这一步中，我们需要搜索一个最佳分类器，该分类器可以为所选问题提供最佳结果。

机器学习的Python工具 (Python Tools For Machine Learning)

• scikit-learn: scikit learn is the most widely used Python library for machine learning. It is an open-source project. Read more about scikit-learn using below link

  scikit-learn： scikit学习是机器学习中使用最广泛的Python库。 这是一个开源项目。 使用以下链接阅读有关scikit-learn的更多信息

• SciPy: sciPy is a python library that supports data manipulation and commonly used in scientific computing which includes statistical distribution, optimization of functions, linear algebra, and variety of specialized mathematical functions. Read more about Scipy using below link

  SciPy： sciPy是一个python库，支持数据操作，通常用于科学计算中，包括统计分布，函数优化，线性代数和各种专门的数学函数。 使用以下链接阅读有关Scipy的更多信息

• Numpy : Numpy is a scientific computing python library that contains the support for fundamentals data structure used by scikit-learn.such as multidimensional-array.Read more about Numpy using below link

  Numpy ：Numpy是一个科学的计算python库，包含对scikit-learn使用的基本数据结构的支持，例如多维数组。

• Pandas: Pandas is a python library for data manipulation and analysis. It supports data structures like Data Frame. Read more about Pandas using below link.

  Pandas ：Pandas是用于数据处理和分析的python库。 它支持数据结构，如数据框架。 使用下面的链接阅读有关熊猫的更多信息。

• Matplotlib: matplotlib is widely used python 2D plotting library that produces high-quality figures in a variety of formats.Read more about Matplotlib using below link.

  Matplotlib ：matplotlib是广泛使用的python 2D绘图库，可生成各种格式的高质量图形。使用下面的链接阅读有关Matplotlib的更多信息。

Conclusion:

结论：

This Article will give you quick overview of Machine Learning.Thank you for reading. Please give it a try, have fun and let me know your feedback!

本文将为您提供机器学习的快速概述。感谢您阅读。 请尝试一下，玩得开心，让我知道您的反馈！

翻译自: https://medium.com/analytics-vidhya/an-introduction-to-machine-learning-1175f63f87fb

机器学习入门�