Miniprojects

Hello and welcome to The Data Incubator!

First Assignment

This first assignment functions as a sanity check, to ensure you understand the basic flow of completing these miniprojects. Once you've completed this, you'll be ready to try your hand at the more substantive miniprojects. (Though you'll probably want to go through some of the lessons, first!)

1. Run git checkout master. Our provisioning code auto-checks-out a different branch (you can run git status) to see, but you'll want to be on master.

2. Open up assignment1/README.md. Read the instructions, and then edit assignment1/__init__.py

   Each module lives in its own directory with a README.md (with instructions) and an __init__.py file (where your solutions go). All code used to generate solutions should live in the respective module's directory (possibly in a src sub-directory).

3. In assignment1/__init__.py, you'll see a function add, along with some decorators up top. Make that function add the two numbers up. Do not modify the decorators.

4. Each question has some a decorator specifying the return type, along with some test cases. Make sure that your function works correctly on the test cases and always returns the right type. You can run validate.py to do this for you.

5. Go ahead and implement the add function, and then run validate.py to make sure you return the right output format.

6. Finally, git add and git commit your solution, then run git push grader master to submit your answer. You should get a bunch of output from the grader, along with your grade. You should also login to your dashboard - check the gradebook linked there to confirm that your grade is being recorded correctly.

That's it! Good luck with the course!

Useful Notes

• If you're having problems with Python imports, you can run python fellow.py to test that you get no ImportErrors

• Make sure to run validate.py

• There's an environment.yml, which specifies the packages that the grader installs before running your code. If you need to add packages, make sure to edit the file. You can update it automatically via:

  $ conda env export --name root > environment.yml

  Just be careful --- if you install lots of packages, grading might take longer!

Workflow

Because of your limited time, it's advisable to think about your workflow before launching into a project. Often it's much more efficient to code a test or proof-of-concept early in the week to avoid running into significant roadblocks later in the week.

For the graph project this might look like:

1. Can I successfully scrape a page and access the text strings?
2. Can I parse names and get the pairs I expect?
3. Can I construct a graph from pairs of names and do its attributes make sense?

This work will iron out (or at least raise) a lot of important questions early on such as what extra libraries you may or may not need.

Deployment

In general, you'll need to edit __init__.py with your answers, commit to the git repo and then push the grader. It makes sense to keep your source code, binaries, databases, etc. out of the repository since they won't be needed by the grader and they'll just take up memory and time.