Handy IPython Tips

the magics

I/O

reload modules you just modified on the fly, autoreload:

%load_ext autoreload
%autoreload 2

save object for later use, storemagic:

# for example you got a counter
c = Counter("a b c".split())
# you can save it in IPython's database and quit
%store c
exit()
# when you want your counter back, simply type:
%store -r

save the ad hoc scripts you typed in IPython session, logstart:

%logstart
from awesome import me
exit()
# cat ipython_log.py

profile

the basic one, timeit:

%timeit reduce(lambda x, _: x + [x[-2] + x[-1]], [0] * (n-2), [0, 1])
#19.4 µs ± 47.1 ns per loop (mean ± std. dev. of 7 runs, 10000 loops each)

memory usage, memory_profiler:

%load_ext memory_profiler
# usage
%mprun -f func_a -f func_b func_c()

Note that, only functions after -f will do the memory profile. The command above implies that func_c will call func_a and func_b.

Wrap it up

magic description
%autoreload debug mode
%storemagic stash object
%logstart stash script
%timeit simple benchmark
%mprun memory profile

references: - Built-in magic commands — IPython 7.22.0 documentation - autoreload — IPython 7.22.0 documentation - storemagic — IPython 7.22.0 documentation - Fibonacci in One Line Python | Finxter


Python

2020-04-10 05:43 +0000