Adding pandas dataframe calculated columns like in Excel

Question

I'm currently using panda for data science purposes, which is very unoriginal, so I'm sure there's a better way than mine of adding columns to a dataframe calculated from the others, in an excel fashion.

For example, I've got a dataframe in which I've got a mass and a period, and I'd like to add a column with (M/P)^(1/3), M and P being converted in the appropriate units. I then write

day_sec=60.0*60.0*24.0
Msun=1.989 * 10**30
clean=reduced.dropna()
v=pd.DataFrame(clean['orbital_period'].apply(lambda x: x**(1/3.0))/clean['star_mass'].apply(lambda x: x**(1/3.0)), columns=list('v'))*day_sec/(Msun**(1/3.0))
clean.append(v)

Which doesn't even act like I want, and is very complicated both to write and read (here, the equation was very very simple). Any suggestion?

If you want to create a new column from existing columns, the best way to do it is something like df['c'] = df['a'] + df['b']. — Demetri Pananos
– Demetri Pananos, Commented Dec 20, 2016 at 17:16

Ted Petrou · Accepted Answer · 2016-12-20 17:21:31Z

5

You should not use apply at all here. You should write the new column as one vectorized operation that looks something like this.

clean['v'] = clean['orbital_period'] ** (1/3.0) / clean['star_mass'] ** (1/3.0) *day_sec/(Msun**(1/3.0))

Which looks like it can be reduced to the following:

power = 1/3.0
(clean.orbital_period / (clean.star_mass * Msun)) ** power * day_sec

edited Dec 20, 2016 at 17:21

answered Dec 20, 2016 at 17:13

Ted Petrou

62.4k19 gold badges139 silver badges139 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

lmo Over a year ago

Or with pow, something like pow(clean.orbital_period / (clean.star_mass * Msun), 1/3.0) * day_sec should work.

Matt Over a year ago

When I run your code, I get : /Users/Home/anaconda/lib/python2.7/site-packages/ipykernel/__main__.py:5: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead

Collectives™ on Stack Overflow

Adding pandas dataframe calculated columns like in Excel

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related