Exercises

Exercises#

These exercises are meant to prepare the DTU Python supporters for a variety of things that they may encounter.

It also covers a bit more than what the courses expects, we need to be prepared for a variety a things.

Note

These exercises are intended with low guidance, we strive to let our supporters be explorative, and that they ensure they know where to find information.

As such, supporters should be able to find the information on-line if they encounter problems they have not experienced before. Or, use their peers to get a solution. So please use search engines for solving the problems.

Best practices#

After all steps of an installation/upgrade of packages it is important to understand and ensure that the environment is behaving as expected.

There can be numerous paths to confirm this, here are some check-points that can be useful:

  • check the pip/conda list of packages and ensure that the package version and package is installed as requested.

  • open up a Python shell and ensure it can be imported

  • ensure the module’s version after import is as requested. Most likely the module has a variable: <modulename>.__version__ which can be checked for current imported version.

  • do the above points also in their editor to ensure that their editor does not launch in a virtual environment.

For instance, after having installed numpy==1.22 one can check:

pip list
# or "grep"'ing for a specific package
pip list | grep numpy

conda list
# or "grep"'ing for a specific package
conda list | grep numpy

Once the package is listed in the package list, check version and import:

python3 -c "import numpy ; print(numpy.__version__)

this will fail with errors if it can not import it, and if it succeeds, it will print the version, if the variable (__version__) is part of the package.

As will be clear in Exercise 4, the above test is also important in the IDE that the questioner is working with.

Warning

Please always use a virtual environment (pip/conda) when performing these exercises. It ensures that you do not mess up your own installation.

Exercise 1#

  1. Install the official Python distribution.

    Windows only: If your terminal will not recognize the python executable, most likely your forgot to check the Add ... to PATH. It might be a good idea if you also try and install Python like this (forcefully) to know how the error message looks.

  2. Install conda in its minimal installation (MiniConda).

    In the semester 2023; the recommended usage will be pip. Likely there will be some students with a prior conda installation. It is thus important that you know how to have a conda installation and navigate a regular Python installation.

    Create a new environment in a local folder.

    Understand that:

    1. Environments created in local folders are not automatically listed when listing available environments

    2. How to activate a local environment.

    3. How to install packages into a local environment

      1. Packages can be installed from an external command, such as conda install -n <path to environment> <packages>,

      2. Packages will be installed directly if the environment is already the activated one: conda activate <path to environment>

    4. How can you see if you are in a conda environment?

    5. Ensure you can navigate between both the conda and the official Python installation.

Hint

The command python -c "import sys ; print(sys.exec_prefix)" can give a hint at which Python interpreter is being used.

Exercise 2#

Create two virtual environments, using venv.

# first environment
numpy=1.24
matplotlib=3.6

# second environment
numpy=1.23
matplotlib=3.6

Check that you can easily swap between these two environments.

Note

One cannot copy paste the above in pip, ensure you change the package specification to match the package installers terminology.

Note

If you have gone through all exercises, then do this again with conda!

Exercise 3#

Install any package, and figure out its location in the file-system. It is important you check against an importable package.

Hint

  • __file__

  • pip

Exercise 4#

For the currently known recommend IDE’s:

In order of priority, if time is limited, only do the first.

  • VS Code (please also read the page linked for some additional information)

  • Spyder

  • PyCharm

Figure out the following:

  1. How to swap environment (interpreter) in the IDE instead of the default Python executable

  2. How to check the packages that are installed (some IDE’s allows calling Pip directly in the IPython console)

Exercise 5#

Run through the 02002 installation instructions. Then run through these extra steps: https://lab.compute.dtu.dk/cp/02002students/-/wikis/testing

It would also be great to test some of these things in a Jupyter Notebook, to see how well they fare in a more constrained environment.

Exercise 6#

Go to pip dependencies and provoke the output shown, in a virtual environment. Understand all output of the commands, especially the warnings and errors. Use pip check as well.

Also resolve the pip check errors.

Exercise 7 (not necessary)#

Complete exercise 2 using virtualenv which works slightly different from venv.

Exercise 8 (not necessary)#

Install the package pyparsing at a specific Git commit c8b7664 using pip.

Contents