4. How to use this book (… if you ask us)#
This textbook has been crafted with attention to detail to facilitate your journey into data science. However, the essence of its usefulness lies in how you engage with its content. Here’s some advice to optimize your learning experience:
Progress at Your Own Speed: Data science is a vast field, and everyone has their own pace. If you find certain concepts daunting, pause and seek additional resources. The internet is full of tutorials, videos, courses, and forums on practically all aspects that we cover in this book. The key is patience and persistence.
Learning by Doing: Reading this book alone won’t transform you into a data scientist. To truly learn data science, you need to actively engage in coding and working with real data sets. Observing code and nodding along saying, “yeah, I got it” is often one of the biggest mistakes that students can make. The quick impression you understood something is usually far away from knowing how to do it yourself!
We can’t stress this enough - the most prevalent pitfall many students succumb to is passive learning! The true test of your understanding comes when you can independently apply the skills, starting from scratch, not merely copying, pasting, and executing someone else’s code.Lean on Tech, But Don’t Rely on It: While chatbots can serve as helpful data science assistants, they are not a substitute for your active learning. We firmly believe that the path to becoming a competent data scientist involves learning how to solve challenges on your own. Relying on a chatbot for answers is akin to copying solutions from another student or a forum. While this approach may help you complete your exercises in the short run, it won’t contribute much to your long-term learning and skill development.
4.1. Code is the easy part#
The next paragraphs will be on code and programming. And for many students, in particular students without a strong computer science background, the coding parts often seems to be the main challenge in the beginning. So, just as a small warning before we talk Python: A very good data scientist for sure knows how to work with code. But that would not nearly be enough. You will later see, that many data science techniques -including very fancy ones- can often we executed with just a few lines of code. The real skill therefore will not be to be able to run all available techniques on any dataset you’ll get. It is the ability to know how to use those methods, but also to know when to use them! When you analyze data it will often come down to asking the right questions, and knowing which tools can help you answer them.
But we will come to this many times again. For now, let’s get started with the very technical parts.
4.2. Skills you should have to work with this book#
Python Programming Basics: Understanding of variables, fundamental data types, loops, functions, and module imports are essential. If you’re new to Python or need a refresher, consider resources like Python.org’s Beginner’s Guide, RealPython, or GeeksforGeeks Python Programming Language.
At least High School Math Skills: Basic knowledge of statistics and algebra will serve you well, for instance to better understand the methods and their caveats.
4.3. Why Python?#
Python, over the years, has emerged as the lingua franca of data science. But why?
Versatile Libraries: Python boasts an array of libraries, such as NumPy, pandas, and scikit-learn, specifically tailored for data science tasks.
Community Support: The vast Python community ensures regular updates, myriad tutorials, and instant help on forums.
Intuitive Syntax: Python’s clear and readable syntax makes it perfect for beginners.
Flexibility: Python seamlessly integrates with other languages, making it ideal for diverse tasks.
Industry Adoption: Major tech companies use Python, ensuring ample job opportunities.
While languages like R and tools like KNIME have their strengths, Python’s holistic ecosystem and adaptability make it the preferred choice for many. All code examples in this book will be written in Python.
4.4. Open source libraries#
Python’s prowess in data science can be attributed to its incredible libraries. These libraries, backed by a robust open-source community, are the backbone of various data operations:
NumPy: Fundamental package for numerical computations in Python. NumPy Documentation
pandas: Offers data structures and operations for manipulating numerical tables and time series. pandas Documentation
SciPy: Builds on NumPy and provides a large number of functions that operate on NumPy arrays and are useful for different types of scientific and engineering applications. SciPy Documentation
scikit-learn: Simple and efficient tools for data mining and data analysis. scikit-learn Documentation
4.5. This Book is full of Code, and you Should Run It Yourself!#
This book is created using Jupyter Book [Community, 2020] and the entire code to produce this book is freely available on GitHub.
Every section of this book that contains code can be found as a Jupyter Notebook, so that you can go ahead and re-run the code, play with it, adapt it, expand it. To do so, we suggest two possible routes. You can (1) run the individual notebooks directly in your browser by clicking on the small rocket button on top which will open the provided notebook either in Google Colab, or using Binder.
Or, if you want to have full control and either know how to set up an environment or want to learn how to do this, you can choose open (2) to run all code locally on you own machine. For this we recommend setting up a new environment by following the steps below:
4.5.1. Prerequisites#
Before we begin, please ensure you have Anaconda or Miniconda installed on your machine. If you haven’t already done so, you can download and install Anaconda from here. Anaconda is a free and open-source distribution of Python for scientific computing. It comes with conda, a package, dependency, and environment manager. Miniconda is a smaller, “minimal” version of Anaconda that includes only conda and Python.
4.5.2. Step 1: Clone the Repository#
The first step is to clone the repository for this course. This can be done by using the git clone command:
git clone https://github.com/florian-huber/data_science_course.git
This command will create a new directory named data_science_course in your current directory, and download the entire repository into that directory.
4.5.4. Step 3: Create a New Environment#
Once you’re in the data_science_course directory, you can create a new conda environment for this course. This can be done using the conda env create command, along with the -f flag to specify the file defining the environment:
conda env create -f environment.yml
This command will create a new conda environment named data_science_course, based on the specifications in the environment.yml file. The environment will include Python 3.9, along with a number of other Python packages that are used in this course.
4.5.5. Step 4: Activate the New Environment#
After the new environment has been created, you can activate it using the conda activate command:
conda activate data_science_course
Once the environment is activated, you can start working on the course materials. Remember, every time you want to work on the course, you should first activate the environment.
4.5.6. Step 5: Updating the Environment#
In case of updates to the environment.yml file, you can update your conda environment using the conda env update command:
conda env update -f environment.yml --prune
The --prune option causes conda to remove any dependencies that are no longer required from the environment.
With these steps, you should have everything set up and ready to start diving into the material for this data science course.
Happy data science coding!