This README will be copied for each homework.

Before you start doing anything, please make sure you have a proper way of reading markdown files. You can use any tool you would like for it. Python IDEs that we recommend for this class (VSCode or PyCharm) come with Markdown readers built-in.

You only need to do setup once, then for future homeworks you can run conda activate csep546.

Before you start working with this repo, you should install Anaconda.

Before clicking in the link below read notes below:

• Linux/Mac OS:
  • If using Linux/Mac please install command line version.
  • Make sure that you choose to initialize conda at startup. This will lead to fewer headaches in the future
• Windows:
  • If using Windows, you will need to use Anaconda Terminal, which uses Bash-like syntax.
• Low storage system
  • If you are low of storage (<10GB; for example attu), then Miniconda (see link below) might be a better option.

Anaconda (default)

Miniconda (if running low on disk space)

You can find more detailed instructions for installation at this link.

First make sure you have at least ~5GB of free drive. From this directory run:

conda env create -f environment.yaml
conda activate csep546
pip install -e .

First command might take long time. Especially if your connection is slow.

Then whenever you come back to work on homeworks just run: conda activate csep546

You are welcome to use any IDE as you would like. Two popular for python are: VSCode (Microsoft) and PyCharm (JetBrains).

PyCharm is tailor made for Python, so the setup should be minimal.

Just make sure you point python interpreter to the right path for the project. To do so, you should follow these instructions.

Optional: Please note, that to run linting (optional, see below) you will need to use terminal, and run inv lint. You can try setting up linters through External Programs section. If you succeed please let us know, and we will update this section for future reference :D

VSCode is general purpose IDE, so you will need to install Python and Pylance extensions (both by Microsoft).

To setup python interpreter, follow these instructions or first reload VSCode, open a folder with homework downloaded and unzipped. Click into a python file and in bottom left you should see Python with some version. Click into that and choose one that says "csep546".

Now whenever you open terminal it should say "(csep546)" on the left hand side, and which python should also output path with csep546 in it. Sometimes, it doesn't work with the first terminal after VSCode opens, so exit it and reopen it (Ctrl+J).

Optional: For linting you will need to enable flake8, mypy and disable pylint. Rather than following the official guide, press (Ctrl/Cmd + Shift + P), select "Preferences: Open Workspace Settings" or "Preferences: Open User Settings", in search type "flake8", "mypy" and "pylint". Each should have an option called "Python > Linting: -linter- Enabled" which you should toggle on for flake8 and mypy and toggle off for pylint.

Extra Optional: Extensions for VSCode and quite helpful apart from support for specific language. I highly recommend downloading Trailing Spaces by Shardul Mahadik and Visual Studio IntelliCode by Microsoft. Also if you are advanced LaTeX user there is LaTeX Workshop by James Yu and LaTeX Utilities by tecosaur. These require you to download and install LaTeX manually though.

When you are done with coding run.

inv submit

this will generate a .zip file that should be uploaded to gradescope for automated grading.

For example running:

inv submit

will result in submission_<timestamp>.zip generated, which should be submitted under corresponding homework coding assignment on gradescope.

In this class we will use unittest framework in python to automatically grade coding problems. Some of the tests are provided to you, so that you can validate your results.

To run tests:

inv test

The output should look something like this:

> inv test
FFF..
======================================================================
FAIL: test_polyfeatures_fives (public.poly_regression.test_poly_regression.TestPolyReg)
----------------------------------------------------------------------
Traceback (most recent call last):
  File ...
AssertionError: 
Arrays are not almost equal to 6 decimals

(shapes (1,), (20, 1) mismatch)
 x: array([1.])
 y: array([[5.],
       [5.],
       [5.],...

You can see that in the top there are 3 F's and 2 .'s. F's correspond to failed tests and . correspond to correct tests.

There are few things to note:

• Not all tests are equal. Some are worth more points. This will not be displayed when you run inv test.
• We do not provide you with all tests. There are many that hidden. Even if you pass all public tests you may still fail some hidden ones.

Linting is not required for this class. However, we recommend that you keep your code linted. Especially if this is your first time learning python, or if you never had a formal introduction.

For this we provided you with a simple script to track issues & possibly fix your code. To run linter:

inv lint

which will generate output

> inv lint
flake8 homeworks
homeworks/hw1/poly_regression/polyreg.py:103:5: E303 too many blank lines (4)

You can see that issue points to file (polyreg.py), line (105) and column (3) as well as the issue "too many blank lines".

Note that we are using 4 linters (flake8, isort, black, mypy; in that order). If you have issues that are pointed out by isort or black you can run inv lint --apply to automatically apply fixes.

Again, linting is not required for this class. If you don't want to lint your code, or prefer some other linting setup that's ok.