ch01

Python Machine Learning - Code Examples

Chapter 1: Giving Computers the Ability to Learn from Data

---

For Existing Conda Users: Creating a New Conda Environment (Optional)

If you are already a conda user, you can skip the next section and get all the recommended package versions in a fresh conda environment (called "pyml-book") via

make -f Makefile

from the main folder in this repository.

Setting Up Your Python Environment (Manually)

This chapter does not contain any code examples, but we recommend you to set up and check your Python before you proceed with the next chapters.

For more detailed setup instructions, please refer to the section Installing Python and packages from the Python Package Index in Chapter 1.

Conda

If you are using conda (we recommend installing conda via Miniforge), you can create a new environment as follows:

conda create -n "pyml-book" python=3.9 numpy=1.21.2 scipy=1.7.0 scikit-learn=1.0 matplotlib=3.4.3 pandas=1.3.2

After creating this environment, you can activate it via

conda activate "pyml-book"

Pip and virtualenv

If you prefer using pip, you can go ahead and install the required packages via

pip install numpy==1.21.2 scipy==1.7.0 scikit-learn==1.0 matplotlib==3.4.3 pandas==1.3.2

However, we additionally recommend creating a new virtual environment for this book. You can create a new virtual environment with a specific Python version using virtualenv as follows:

pip install virtualenv
cd /path/to/where/you/want/your/environment
virtualenv pyml-book
source pyml-book/bin/activate 

After activating your environment, you can install the required packages via

pip install numpy==1.21.2 scipy==1.7.0 scikit-learn==1.0 matplotlib==3.4.3 pandas==1.3.2

Checking Your Python Environment

To verify that your Python environment is set up for the following chapters, we recommend running the ../python_environment_check.py script provided in the main folder of this repository.

You can run the python_environment_check.py script via

python python_environment_check.py

Shown below is an example output:

(base) sebastian@MacBook-Air ~/Desktop/Python-Machine-Learning-PyTorch-Edition/ch01 % python ../python_environment_check.py
[OK] Your Python version is 3.9.6 | packaged by conda-forge | (default, Jul 11 2021, 03:35:11)
[Clang 11.1.0 ]
[OK] numpy 1.21.2
[OK] scipy 1.7.0
[OK] matplotlib 3.4.3
[OK] sklearn 1.0
[OK] pandas 1.3.2

Jupyter Notebooks

Some readers were wondering about the .ipynb of the code files contained in this repository -- these files are Jupyter notebooks (formerly known as IPython notebooks).

Compared to regular .py scripts, Jupyter notebooks allow us to have everything in one place:

• Our code.
• The results from executing the code.
• Plots of our data.
• Documentation supporting the handy Markdown and LaTeX syntax for typing and rendering mathematical notation.

Please see the https://jupyter.org/install website for the latest installation instructions.

Two official applications can open Jupyter notebooks: the original Jupyter Notebook app and the newer Jupyter Lab app (and VS Code has Jupyter notebook support, too). The notebooks provided in this repository are compatible with both.

We recommend installing Jupyter Lab via Jupyter Lab can be installed via

conda install -c conda-forge jupyterlab

or

pip install jupyterlab

Finally, please note that the Jupyter notebooks provided in this repository are optional, although we highly recommend them. All code examples found in this book are also available via .py script files (which were converted from the Jupyter notebooks to ensure that they contain the identical code.)