Past one month I have been reading a book titled "Approaching Almost Any Machine Learning Problems" authored by Abhishek Thakur.
This book is divided into 13 parts with each titled as below:
Setting up your working environment
Supervised vs unsupervised learning
Cross-validation
Evaluation metrics
Arranging machine learning projects
Approaching categorical variables
Feature engineering
Feature selection
Hyperparameter optimization
Approaching image classification & segmentation
Approaching text classification/regression
Approaching ensembling and stacking
Approaching reproducible code & model serving
Few key points I liked about this book:
Thoroughly Explained:
As per the title given, the author explains each part of the book in detail not only with theory but also with complete code with execution results.
Significant but Rare Topics Discussed:
The 1st and 5th parts are not really discussed in any end-to-end data science projects, and I am glad that the author has taken the time to explain them in specific detail.
Codes for Hands-on:
Codes provided will help the starters to dive into the topics hands-on and ensures successful execution.
Thoughtful of Readers:
In the last section of the book. the author has provided blank pages for notes taking. This is a very thoughtful gesture from the author.
A topic that is needed to be discussed on a "Large Scale":
The last part of the book is all about “reproducible code and model serving”. This is an important part to be discussed because in reality the company/departments that build a machine learning or any analytics project want their projects to be reused. This is clearly explained technically.
Overall this book is a wonderful addition to your data book collection. People who are starting up their DS journey can start building projects with the codes provided starting from a simple Regression Model to Neural Network.
I totally recommend this book to people who are willing to test their skills hands-on. Finally, I would this book 4.5 why? Because as a reader, I felt that the chapters could be well divided between each page in order to better keep track of each chapter.
If you like this post, please like, share, and subscribe to this blog for more data science-related posts.
Comments