coding_cheat_sheet.md

Ocean's Six Coding Cheat Sheet

Tips

• Brackets (<>) indicate that you should replace the brackets and words inside with your own input
• Spaces are very important in coding!
  • Terminal/bash scripts use a lot of spacing between commands. i.e. git pull upstream master
  • In Python, there are very few spaces needed. i.e. np.array([[1,2,3], [2,3,4]])
• Frequently, my own coding errors come from adding a space when there shouldn't be one (or vice versa) or mispelling something.

Numpy Arrays

• Remember, all arrays begin at the 0 index! So a[0] picks the first element, a[1] picks the second, and so on
• In these tips, a, b and c refer to different numpy arrays.

1D arrays

• This command selects everything: a[:]
• This command selects the 0th, 1st and 2nd rows (does not select the 3rd row): a[0:3]
  • The numbers inside the brackets can be changed to select whatever values you want.

2D arrays

• This command selects all rows and all columns: b[:,:]
• This command selects all rows, and the zeroth column: b[:,0]
• This command selects the zeroth row, and all columns in that row: b[0,:]
• This column selects the zeroth and first row, but NOT the second. It includes all the columns in that row: b[0:2,:]

3D arrays

Technically, the first dimension is known as the row, the second as column and the third as depth, but I find that confusing. We'll call the first dimension the "group," the second "row," and third, "column" (as usual).

• This command selects everything. c[:,:,:] #This command selects everything.
• This command selects the 0th group, and all rows/columns in that group: c[0,:,:]
• This command selects the 0th row from every group: c[:,0,:]
• This command selects the 0th row from every group: c[:,:,0]
• This command tells you the shape of your array: c.shape
  • The first value is the size of the first dimension, and so on

NetCDF files

This tutorial was written in Python 2.7, so the print command is slightly different, but it's a helpful read to understand how these files work.

Follow these first steps in order:

1. Import the tools to open a dataset: from netCDF4 import Dataset
2. Open a dataset: dataset = Dataset('path/to/filename.nc')

Use these tools in any order:

• View the dataset's attributes: dataset.ncattrs()
• Access a specific attribute: dataset.attribute_name
• View the dataset's dimensions: dataset.dimensions
• View a specific dimension: dataset.dimensions['name of dimension']
• View the dataset's variables: dataset.variables
• View a specific variable: dataset.variables['name of variable']
• See a variable's values: dataset.variables['name of variable'][ : ]

Terminal

• To show all the files in the folder you're in: ls
• To navigate to a new folder: cd <path name> (enter the name of the folder or path to the folder)
• To navigate to the home director: cd
• To navigate to the previous folder: cd ..
• To open a file: open
• To find the path of a file: Find the file in Finder, then drag it into the Terminal window. The path will automatically appear!

Git

• To clone a repository: git clone <url> (i.e., copy all files from an online repository to a folder on your computer)
• To add a new remote: git add remote <name> <url> (i.e., connecting your local repository (on your computer) to a remote repository (online)
• To see your remotes: git remote -v
• To add a change: git add <file-name> (i.e., you've edited a file and want to make Git aware of those changes)
• To commit that change: git commit -m "message" (this makes your changes official. It's basically like "saving" your work to Git.)
• To push those changes to your remote repository: git push origin master (i.e., sending any updated files to your online repository)
• To sync your forked repository with any changes made to the original: git pull upstream master (i.e., you're adding any new files from the main online repository to your computer)