Python for Data Analysis 3e: Data Wrangling with Pandas, Numpy, and Jupyter

This book is very thorough and written very clearly. I bought it to assist my learning about DS/ML/AI. As Wes says in the book, the vast majority of the work in this area is data "wrangling", or getting the data into a form suitable for analysis. This book is very comprehensive and covers every aspect of the process.It starts with some introductions to Python, IPython and Jupyter. It doesn't really assume any Python knowledge, but I knew and worked with Python already and would describe it as a "whistle-stop tour". IPython is a more interactive (and dare I say, better) Python shell/interpreter which is great for exploring / testing stuff out. Jupyter is simply a web-based tool which can use IPython and has a slightly different workflow based on "cells" which can be executed.After that, there is a brief section on NumPy, and it highlights the key ideas: vectorised operations with multi-dimensional arrays. That is, to create the result of summing the elements of two arrays, a and b, you can simply write a + b. This makes for quick and convenient batch operations on data with no for loops in sight.These arrays offer a compact and highly efficient way to work with data, and they the "lingua franca" of the Python data world.The bulk of what follows is on pandas, which makes working with numpy even easier and is a heavyweight tool for loading data from a variety of formats (and storing it), cleaning it, processing it, aggregating, joining, visualising (with help from matplotlib) - everything you could want really. pandas is amazing software, and like most great software projects, it has the right abstractions (like the DataFrame), integrates seamlessly with a lot of other data libraries, and starts to feels intuitive after a while.There is also a lot of information on working with time data, and a brief chapter on building models.My favourite part was probably the end-to-end data analysis examples where Wes explores datasets from movies to baby names. Seeing the whole process there was great.The appendices are well worth reading, if not skimming, since there are more NumPy and IPython tricks.Could this book be improved in any way? Yes, I think a lot of reads a bit like a reference. Some more "real-world" examples would have been better in a lot of cases - but you do see those in the data analysis chapter. Also, the reason for using simpler examples was probably to demonstrate the point more clearly and independent of other chapters.Overall though, this gets a strong recommendation from me if you want to get more into data and machine learning.Thanks to Wes McKinney, we not only have pandas, but this book to navigate it.

Good gift useful

I have been learning Python and it's modules for few months and I bought three books.This is the best one by far.It takes you through Python itself, Jupyter, Numpy, Pandas and some other very important modules. Nicely explained, lots of examples and you can download the code, too.Highly recommended!

Damaged product

In the ever-evolving world of data analysis, "Python for Data Analysis: Data Wrangling with pandas, NumPy, and Jupyter" shines as a beacon for beginners and experienced data enthusiasts alike. It is a testament to the book's caliber that it manages to comprehensively cover Python's powerful data manipulation tools in such an approachable manner.The book's primary strength lies in its thorough exploration of data wrangling. For the uninitiated, data wrangling is the art of maneuvering raw data into a more digestible form for analysis, and this book nails it. By delving deep into the intricacies of libraries like pandas and NumPyWhat's particularly commendable is the balance between theory and practice. While it's brimming with technical details and explanations, it never feels overwhelming. Instead, readers are constantly engaged with practical examples, ensuring that learning is both meaningful and applicable.Whether you're a beginner stepping into the world of data or a seasoned pro, this book offers invaluable insights and skills that will enhance your data-wrangling journey.

Recomiendo ir por la versión electrónica ya que los imágenes (gráficos, notas, figuras, etc.) son a color. el libro en físico no tiene ninguna calidad, esto resulta difícil por ejemplo en la sección de Matplotlib.

Perfect conditions

Ottimo testo non adatto ai principianti x Python

Muy bueno