UConn NumPy tutorial development

Prof. Ryan C. Cooper

This is our own repository for testing git and GitHub

This semester we are developing Jupyter notebooks for the NumPy tutorials repository. We are going to use a few tools to build tutorials, share code, and review code:

Building tutorials

Python, Jupyter, and the SciPy stack: You can install Python and NumPy with the miniconda installer. Once its installed, you can run use the conda terminal to install NumPy, Jupyter, SciPy, and Matplotlib

$ conda install numpy jupyterlab scipy matplotlib

If you're new to NumPy, get started with the Absolute Beginner's Tutorial to NumPy.

In your conda terminal, run

$ jupyter-lab

to get Jupyter running in a web browser (I prefer Firefox or Chrome). Now, you can run some Python commands and make some progress.

Sharing code

GitHub and Git: NumPy is a free open source software. The code is updated using a version control software called git. NumPy is shared and reviewed by a worldwide community of developers on the website <www.github.com>. You can download, share, and suggest changes to software using git and GitHub. Start by installing git, then use a terminal to clone this repository

$ git clone https://github.com/cooperrc/uconn_numpy_tutorials

Once you have run this command, you now have a copy of this repository on your computer. Try opening the practice_notebook.ipynb in the notebooks folder and running the commands. You can play around with the inputs and even write your own code.

Reviewing code

GitHub Pull Requests: We will get code reviewed by creating a pull request (PR) on an existing GitHub repository. Practice creating a PR with these steps:

1. Create a fork of the repository uconn_numpy_tutorials

2. Clone your fork to your computer with the git clone ...

3. Create a new "remote" for the upstream repo with the command

git remote add upstream https://github.com/cooperrc/uconn_numpy_tutorials

4. Use the git command to create a "branch" called my-notebook

git branch my-notebook

5. Use the git command to "checkout the branch"

git checkout my-notebook

6. Create your own Jupyter notebook in the notebooks folder, use your username to make it unique e.g. I would name the notebook, cooperrc_notebook.ipynb

a. In your notebook include a NumPy example that you think is interesting

b. Save your work and close the notebook

7. Use the git command to (i) add your file, (ii) commit your work, and (iii) push your changes to your fork

i. git add notebooks/cooperrc_notebook.ipynb ii. git commit -m "my first branch commit for my notebook" iii. git push -u origin my-notebook

8. Now, you can create a PR with this repository and I will merge your work.

Bonus: contributing to your first open source project