vegarsti

pandoc file.md -o file.pdf

vegarsti

• To count occurrences when aggregating (e.g. in pivot_table), use len as aggfunc!

• To drop a row: Cheatsheet

  • df[df.name != 'Tina']
  • df.drop(df.index[2])

• c['a'].shift(periods=1)

vegarsti

Github link

vegarsti

# defaultdict use case
# Finding a Pareto front, where (x1, x2) is a list of tuples satisfying
# e.g. (x1, x2) : |df[(df[var1 > x1]) & df[var2 > x2)]| < 0.1|df|,
# where |df| is the size of df, i.e., df.shape[0] (in Pandas)

pareto = defaultdict(lambda: np.Inf)
for x1, x2 in xs:
    pareto[x1] = min(pareto[x1], x2)

vegarsti

• Virtualenv & stuff (incl. autoenv): Read!
• Homebrew
• LaTeX w/ Sublime

vegarsti

Requires node.js and npm.

First, do npm install -g ijavascript. Then run ijs once.

vegarsti

One-line Tree in Python

Using Python's built-in defaultdict we can easily define a tree data structure:

def tree(): return defaultdict(tree)

That's it!

vegarsti

• plt.savefig(filename, bbox_inches='tight') --> get the entire plot when saved
• Reset to factory default matplotlib style

import matplotlib as mpl
mpl.rcParams.update(mpl.rcParamsDefault)

• plt.axes('tight') removes all space around the area of the curve
• plt.axes('equal') ensures x-y even
• plt.axhline(y=production_accuracy, linewidth=2, color='k', label='production') horizontal line

vegarsti

• save: C-x C-s (+ enter)
• open file: C-x C-f
• switch to buffer: C-x b
• meta key M is escape and left alt (option)
• C-k: kills from cursor until end of line
• C-/ (7) -- undo!
• C-a start of line
• C-e start of line
• M-f forward one word
• M-b back one word

vegarsti

• plot(x, y) + abline(a, b, col="red", lwd=2)
• plot(x, y, type="b", col="red", pch=19) "dotdash" med filled
• plot(x, y, type="b", col="red") "dotdash" uten filled
• rm(list=ls(all=TRUE)) remove stuff
• `sapply(.., simplify='array') == unlist(lapply(..))