python personal data analysis to learn to read notes - Table of Contents Index
Tang Yu Di, Ph.D. computer science, artificial intelligence, Netease cloud classroom certification experts, 51CTO Lecturer, CSDN blog expert, lecturer. Has many years of experience training in the field of artificial intelligence, to lead the research and development team total development curriculum AI course of more than 60, covering current hot areas of Artificial Intelligence
book combines machine learning, data analysis and Python languages, user-friendly way through the case to explain how the algorithm is applied to the actual task. The 20 chapters, divided into four portions. The first part describes the Python toolkit, including scientific computing library Numpy, data analysis library Pandas, visualization library Matplotlib; Part 2 explains the classical algorithm in machine learning, such as regression algorithm, decision trees, ensemble algorithm, support vector machine, clustering algorithm; part 3 describes the depth of the learning algorithm used, including neural networks, convolutional neural network, recurrent neural network; part 4 project combat, based on real data sets, the algorithm is applied to the actual business model .
The book is suitable interested in artificial intelligence, machine learning, data analysis and other beginners and enthusiasts.
The book recommended Learning Roadmap:
"Python data analysis and machine learning practical - Tang Yu Di" study notes Chapter 1 - Machine Learning concept, entry environment
1.1 machine learning applications:
2, machine learning process:
In general, the machine learning process is roughly divided into the following steps:
a first step ①: data collection and pre-processing . For example, the news will be a lot of factors unrelated doping special characters and advertising, we first have to weed out these. In addition, the article may also be used for word, extract keywords and other operations, which will be analyzed in detail in subsequent cases.
The first step ②: Feature Project, also known as feature extraction . For example, some news, describing "Kobe's career drawing to a successful conclusion, officially retired today." Obviously, this is a sports-related news, but the computer may not recognize Kobe Bryant, so it needs to convert people can read characters into a computer can recognize the value. This step looks easy, but doing it is very difficult, and very important part of machine learning how to construct appropriate input feature.
Step ③: Model Building . This step is just to train a classifier can, of course, the modeling process also involves a lot of work parameter adjustment, just to establish a similar model is easy, but you want to be perfect model also requires a lot of experiments.
The first step ④: Assessment and Forecast . Finally, to complete the construction of the model can predict the judge, and then pretreated article is passed in the model, the machine will tell us which it is derived in accordance with the data to learn what results.
1.3 environment configuration:
1) Recommended download Anaconda integrated environment.
Download the corresponding version can be installed. Integrated jupyter notebook, skyder and so on. Note Tsinghua modified mirror, reference: https://www.cnblogs.com/downmoon/p/12447551.html
However, the invitation personally love Eclipse, after all, a multi-language development environment.
If there is an individual package can not use pip install, where you can try to download whl. https://www.lfd.uci.edu/~gohlke/pythonlibs/
2）jupyter notebook
A great tool for teaching.
If the installation or configuration have any questions, can be found here: https://www.cnblogs.com/downmoon/p/12598135.html
1.4 Other resources:
Github:https://github.com/
GitHub offers a wealth of open source projects and code.
kaggle community: https://www.kaggle.com/
Its contents are relevant and scientific data, we can use it as a race site, including not only all walks of life data sets, but also the brightest Great God solutions.
Summary: This chapter describes the Python route learning and machine learning as a whole, the book used by Anaconda environment only one to get.
Chapter 1 finish.
python personal data analysis to learn to read notes - Table of Contents Index
The book resources download, go to asynchronous community: https://www.epubit.com