Explore Sklearn

本课程旨在让学生了解 Python 的机器学习库 sklearn。该课程分为七个部分，分别是：-

一、简介

2. 监督学习

3. 半监督学习

4. 无监督学习

5. 降维

6. 特征选择

7. 其他话题

学生将获得有关如何使用 sklearn 的广泛指导。Sklearn 的搜索引擎将用于研究 sklearn 的许多功能，其中包括：-

1.预处理功能

2.分类模型

3.回归模型

4. 半监督模型

5. 聚类模型

6.降维函数

7.特征选择功能

8. 度量函数

除了了解 sklearn 中众多不同类型的函数外，学生还将复习十二个 Jupyter Notebook 的代码。这些笔记本的主题是：-

1.有监督的分类问题

2. 监督回归问题

3. 半监督分类问题

4. 半监督回归问题

5. 无监督分类问题

6.降维

7. 通过选择最佳特征进行特征选择

8. 通过选择一定百分比的最佳特征来选择特征

9. 逻辑回归与决策树

10.机器学习生命周期

学生将使用 sklearn 和其他编码，从头到尾涵盖整个机器学习生命周期。这将包括：-

1. 在 Google Colab 中创建 Jupyter Notebook

2. 将 Python 库导入 Jupyter Notebook

3. 从 sklearn、openml 或 Github 加载数据集

4. 通过处理任何空值来清理数据

5. 对数据进行编码以将对象特征转换为数字特征

6. 使用可视化技术分析数据

7. 必要时从回归模型中删除任何异常值

8. 在必要时删除任何具有高相关性的特征

9. 必要时减少数据集的维数

10. 必要时减少数据集的特征

11. 分配因变量和自变量

12. 必要时将数据集拆分为训练集和验证集

13.选择最合适的模型

14. 对模型进行预测

15. 使用度量函数分析模型的准确性

Published 10/2022
MP4 | Video: h264, 1280×720 | Audio: AAC, 44.1 KHz
Language: English | Size: 2.90 GB | Duration: 6h 16m

to develop machine learning skills

What you’ll learn

• Students will learn about sklearn, Python’s machine learning library
• Students will learn about and go over the code of supervised learning classification and regression problems
• Students will learn about and go over the code of semi-supervised classification and regression problems
• Students will learn about and go over the code of unsupervised regression problems
• Students will learn about and go over the code of principal component analysis
• Students will learn about and go over the code of feature selection techniques

Requirements

Basic Python programming is a prerequisite to this course

Description

This course is intended to give the student an overview of Python’s machine learning library, sklearn. The course is broken down into seven sections, being:-

1. Introduction

2. Supervised learning

3. Semi-supervised learning

4. Unsupervised learning

5. Dimensionality reduction

6. Feature selection

7. Other topics

The student will receive extensive guidance on how to use sklearn. Sklearn’s search engine will be used to research sklearn’s many functions, which include:-

1. Preprocessing functions

2. Classification models

3. Regression models

4. Semi-supervision models

5. Clustering models

6. Dimensionality reduction functions

7. Feature selection functions

8. Metrics functions

In addition to learning about the numerous and varied types of functions in sklearn, The student will go over the code of twelve Jupyter Notebooks. The subject matter of these notebooks are:-

1. Supervised classification problems

2. Supervised regression problems

3. Semi-supervised classification problems

4. Semi-supervised regression problems

5. Unsupervised classification problems

6. Dimensionality reduction

7. Feature selection by selecting the best features

8. Feature selection by selecting a percentage of the best features

9. Logistic regression versus decision tree

10. The machine learning life cycle

The student will, using sklearn and other coding, cover the entire machine learning life cycle from the beginning to the end. This will cover:-

1. Creating a Jupyter Notebook in Google Colab

2. Importing Python libraries into the Jupyter Notebook

3. Loading the dataset from either sklearn, openml, or Github

4. Cleaning the data by taking care of any null values

5. Encoding the data to covert object features to numeric features

6. Using visualisation techniques to analyse the data

7. Removing any outliers from regression models where necessary

8. Removing any feastures that have a high correlation where necessary

9. Reducing the dimensionality of a dataset where necessary

10. Reducing the features of a dataset where necessary

11. Assigning dependent and independent variables

12. Splitting the dataset into training and validation sets where necessary

13. Selecting the most appropriate model

14. making predictions on the model

15. Analysing the accuracy of the model by using metric functions