To-do list for 2022

[jenniferagraham]

A few more comments to add (from what I noted):

Website:

• Don't type bash before activating environment.
• Fix link to GitHub site (main rather than master)
• Do we want a bit more info for miniconda (Cefas) users?

Notebooks

02

• Add markdown cheatsheet link to jupyter notebook intro?

Day 1 exercises

• add note about range on day 1 exercises?

• A bit of earlier info on indexing or slices?

• Link to cheatsheet for string formatting (when introducing f strings): https://kapeli.com/cheat_sheets/Python_Format_Strings.docset/Contents/Resources/Documents/index

• Q on how to get co2e to 2dp?
  co2e_2dp = [float(f"{x:.2f}") for x in co2e]
  (can also use built in round() function)

14 - Numpy:

• Flag up numpy for matlab link earlier? https://numpy.org/doc/stable/user/numpy-for-matlab-users.html

• Update section on common mistakes:

try:
    a = np.array(1,2,3,4)   # WRONG, only 2 non-keyword arguments accepted
except TypeError as err:
    print('TypeError: ',err)
# help(np.array)

• Fix link to lists: https://github.com/ueapy/pythoncourse2020-materials/blob/master/notebooks/07-Built-in-Data-Structures.ipynb

• a[2:9:3] # [start:end:step]

  • note that 9 could be removed to include all values to end.

• Indexing in practice:
  x.sort()

• Ex2 - typo in solutions

• Note boolean arrays have to have same shape as those they're passed to.

• Copies - mention earlier link to pointers.

• Operations over axes - improve description of who this works, e.g. chosen axis will "collapse"?

• NB. reshape used before discussed later.

• Ex5 - typo (on-of)

• Solutions: need to reshape 5x5

• Broadcasting:

  • Note this would also work: a1d.reshape(5,1) + a2d

• Q - How to extract mask from masked array?

• Q - How to find number of nan values in an array? - np.sum(np.isnan(a))?

15:

• fix link to colormaps
• remove clim reference in mini exercise (not the right option here - but perhaps introduce elsewhere?).
• fix x1 -> x1d (typo)
• Final exercise: Note colorbar is the tricky part - leave to the end!

xarray

• Q on interpolation: flag link for further info http://xarray.pydata.org/en/stable/user-guide/interpolation.html
  • Note other interp methods exist e.g. kdtree, scipy ...?
• slice appears inclusive for time, but not others - should be inclusive for all, so is this a rounding/precision issue?
• Ex1: follow -> following
• Emphasis somewhere that data isn't downloaded/read until it is used?
  • (Easy to quickly check attributes/size or extract specific values?)

18

• how to handle missing values on print: e.g. na_rep='NULL'
• df = pd.DataFrame({'OLR': olr_df.Value, 'SOI': soi_df.Value})
  • Value here is name of variable - suggest renaming, as easily confused with .values properties