week_04

Week 4: Setup

Welcome to week 4! Before running the exercise, we need to do a bit of set up. This week we'll be learning about the Python modules numpy, pandas, and matplotlib.

If you think back to week 1, when we set up our conda environment, you'll remember that we installed certain packages. More specifically, in the file week_01/README.md, we ran the following commands:

conda create -n ppchem python=3.10
conda activate ppchem
pip install pandas  # installs numpy as dependency
pip install rdkit==2022.09.5
pip install jupyterlab

You can see that by running these, in our environment ppchem, we should already have numpy and pandas installed. Open a terminal / powershell / command line prompt, conda activate ppchem (or whatever you called your environment) and run the command conda list. The output should look something like this (output shortened):

# packages in environment at /opt/miniforge3/envs/ppchem:
#
# Name                    Version                   Build  Channel
anyio                     4.2.0                    pypi_0    pypi
appnope                   0.1.4                    pypi_0    pypi
...
jupyterlab                4.1.0                    pypi_0    pypi
...
numpy                     1.26.4                   pypi_0    pypi
...
pandas                    2.2.0                    pypi_0    pypi
pandocfilters             1.5.1                    pypi_0    pypi
parso                     0.8.3                    pypi_0    pypi
pexpect                   4.9.0                    pypi_0    pypi
pillow                    10.2.0                   pypi_0    pypi
pip                       24.0               pyhd8ed1ab_0    conda-forge
...
websocket-client          1.7.0                    pypi_0    pypi
wheel                     0.42.0             pyhd8ed1ab_0    conda-forge
xz                        5.2.6                h57fd34a_0    conda-forge

and you should be able to see numpy and pandas there.

Milestone: installing new packages and updating the environment file

Note: this section assumes that you are up to date with your milestones, and have pushed your exported environment file to your personal ppchem repository. If this is not the case, make sure you have followed all of the instructions in the Week 1 exercises.

Then, follow these instructions:

1. In your terminal / powershell application, navigate to your personal <username>/ppchem repository.
2. Make sure you have activated your environment: conda activate ppchem (or whatever you named your env)
3. Install the new package: pip install matplotlib
4. Create a new branch: git checkout -b update-env
5. Export the environment file: conda env export > env.yml
6. Inspect the changes to the environment file compared to the last commit. This can be done by running git diff env.yml. Use your arrow keys to scroll. There may be a few changes, but most importantly you should see a line like: + - matplotlib==3.8.3. This tells us that, relative to the last commit, matplotlib has been installed in the environment, at version number 3.8.3. Press q to quit the git diff viewer.
7. Add the changes: git add env.yml
8. Commit them with a meaningful message: git commit -m "Updated environment to include matplotlib"
9. Push to your fork. As the remote doesn't yet know that we have created the branch update-env locally, we need to push with: git push --set-upstream origin update-env

Navigate to your repository on Github, at URL: https://github.com/<username>/ppchem. You should see a page like this:

Click the branch drop down menu where it says "main" to select a branch, and select the branch "update-env":

You should see that your branch update-env is 1 commit ahead of main. We want to create a pull request for this branch, so will click on the "contribute" button, and select "Open pull request":

this will open a new page for opening a pull request:

Make sure you add a title and a short description of your pull request - i.e the changes you have made and want to merge. Then, select "Create pull request". This will take you to the pull request page.

On this page, this is typically where code reviews will be posted. Usually, if you are contributing to an open source package, and want to merge some of your changes into the main branch of the code, someone will review your work, request changes and leave comments. This all happens on this page.

As this is just your personal repository and the changes to the code weren't significant, for now we will not do any review and just merge into main. Select "Merge pull request":

and "Confirm merge":

then your pull request is merged! You can safely delete the branch associated with the PR, as all the changes are now in main:

The pull request is accessible in the "Pull Requests" tab of the main repository page, but will be in the 'closed' section.

Navigate back to your main repository landing page, i.e. https://github.com/<username>/ppchem and check that the changes are there:

Finally, update the "Open a pull request" row of your Personal Milestones table with the URL of the pull request.

For example, the URL of my (Joe's) PR was: jwa7/ppchem#3

Good job! Now onto the week 4 exercises...