Data Science and Machine Learning in Python: Linear models

掌握 Python 中最流行的数据科学和机器学习算法（线性回归、逻辑…）。

您将学到什么
从头开始​​逐步实施我们的所有模型。您将了解他们的理论和实践的每一个细节。
从根本上了解最流行的机器学习算法。
掌握 Python 中的主要机器学习库：scikit-learn、numpy、pandas、matplotlib 等。
了解数据科学工作流程以及如何从头到尾解决预测问题。
诊断和解决我们模型中的问题。当他们的模型失败时，您将成为您的同事所寻求的人。

要求
python 的基本知识（变量、循环、类等）
使用 pandas 和可视化工具的经验会有所帮助，但不是必需的。

描述
为什么要学习数据科学？

公司有一个问题：他们每天收集和存储大量数据。问题是他们没有从这些数据中提取知识和做出决策的工具和能力。但这种情况正在改变。几年来，对数据科学家的需求呈指数级增长。如此之多，以至于拥有这些技能的人数不足以填补所有职位空缺。在 Glassdoor 或 Indeed 上进行基本搜索，您会发现为什么数据科学家的薪水近年来增长如此之快。

为什么选这门课？

几乎所有的课程要么太理论，要么太实用。大学课程通常不会培养从头开始解决数据科学问题所需的技能，也不会教你如何流利地使用必要的软件。另一方面，许多在线课程和训练营会教你如何使用这些技术，而无需深入了解它们，只是从表面上了解理论。

我们的课程结合了每种方法的精华。一方面，我们将研究这些方法的来源以及使用它们的原因，了解它们为何以这种方式工作。另一方面，我们将使用 Python 中最流行的数据科学和机器学习库从头开始编写这些方法。只有当您准确了解每种算法的工作原理后，我们才会学习如何将它们与高级 Python 库一起使用。

课程内容

机器学习和数据科学简介。

简单的线性回归。我们将学习如何研究不同现象之间的关系。

多元线性回归。我们将创建具有多个变量的模型来研究感兴趣变量的行为。

套索回归。多元线性回归的高级版本，能够过滤最有用的变量。

岭回归。更稳定的多元线性回归版本。

逻辑回归。最流行的分类和检测算法。它将允许我们研究不同变量和某些对象类之间的关系。

泊松回归。算法将允许我们查看几个变量如何影响事件发生的次数。

数据科学的核心概念（过拟合与欠拟合、交叉验证、变量准备等）。

任何问题？请记住，我们提供 30 天全额退款保证。对你没有风险。这就是我们确信您会喜欢这门课程的方式。

Published 08/2022
MP4 | Video: h264, 1280×720 | Audio: AAC, 44.1 KHz, 2 Ch
Genre: eLearning | Language: English | Duration: 46 lectures (3h 57m) | Size: 887.1 MB

Master the most popular data science and machine learning algorithms in Python (linear regression, logistic…).

What you’ll learn
Implement all our models from scratch, step by step. You will learn every detail of their theory and practice.
Fundamentally understand the most popular machine learning algorithms.
Master the main machine learning libraries in Python: scikit-learn, numpy, pandas, matplotlib, etc.
Understand the data science workflow and how to solve a prediction problem from beginning to end.
Diagnose and solve problems in our models. You’ll be the person your colleagues will seek when their models fail.

Requirements
Basic knowledge of python (variables, loops, classes, etc)
Experience with pandas and visualization tools helps, but is not required.

Description
Why study data science?

Companies have a problem: they collect and store huge amounts of data on a daily basis. The problem is that they don’t have the tools and capabilities to extract knowledge and make decisions from that data. But that is changing. For some years now, the demand for data scientists has grown exponentially. So much so, that the number of people with these skills is not enough to fill all the job openings. A basic search on Glassdoor or Indeed will reveal to you why data scientist salaries have grown so much in recent years.

Why this course?

Almost every course out there is either too theoretical or too practical. University courses don’t usually develop the skills needed to tackle data science problems from scratch, nor do they teach you how to use the necessary software fluently. On the other hand, many online courses and bootcamps teach you how to use these techniques without getting a deep understanding of them, going through the theory superficially.

Our course combines the best of each method. On the one hand, we will look at where these methods come from and why they are used, understanding why they work the way they do. On the other, we will program these methods from scratch, using the most popular data science and machine learning libraries in Python. Only when you have understood exactly how each algorithm works, we will learn how to use them with advanced Python libraries.

Course content

Introduction to machine learning and data science.

Simple linear regression. We will learn how to study the relationship between different phenomena.

Multiple linear regression. We will create models with more than one variable to study the behavior of a variable of interest.

Lasso regression. Advanced version of multiple linear regression with the ability to filter the most useful variables.

Ridge regression. More stable version of multiple linear regression.

Logistic regression. Most popular classification and detection algorithm. It will allow us to study the relationship between different variables and certain object classes.

Poisson regression. Algorithm that will allow us to see how several variables affect the number of times an event occurs.

Central concepts in data science (overfitting vs underfitting, cross validation, variable preparation, etc).

Any questions? Remember that we have a 30-day full money back guarantee. No risk for you. That’s how convinced we are that you will love the course.

Who this course is for
Students interested in landing a job in the field of Data Science / Machine Learning.
Professionals who want to apply predictive modelling to solve their hardest business problems.
Machine learning practitioners who want to develop a deep understanding of how their algorithms work.