Python Pandas Cheat Sheet

by LasseBoe (Lasse1618) via cheatography.com/110079/cs/21581/

STOPPED FILLING AT LECTURE 6 LINE Basic Dataframe Analysis (cont) Label-​based indexing with .loc / .iloc
122 (cont)
df.plo​t.s​cat​‐ scatte​rplot
ter(x = df.loc[:, 'col1' ] column with t/f whether
'name', y = == 'name' entry in col1 is name
Basics 'name')
%matpl​otlib inline plots into notebook pd.plo​tti​‐ multiple scatte​rplots df.ilo​c[3:-1, 2:] [rows, columns]
df = pd.rea​d_c​‐ loads dataframe ng.s​ca​tte​‐ df.iloc[:, [3, 1]] columns with index 3 & 1
sv(​ path, r_m​atr​ix(df)
.loc is label based, .iloc is integer index
index_col pd.plo​tti​‐ lines drawn connecting
based
='name') ng.p​ar​all​‐ dimensions of an entry
df.head() el_​coo​rdi​‐
Series
nat​es(df,
df.tail()
'name') s1 = pd.Ser​ies([1, 2, creates a pandas
df.values 3], index=​['a', 'b', series
df['co​l_n​‐ returns list of singled entries
df.plot() 'c'])
ame​'].u​‐
df.plot(style='.') s1.add(s2, fill_v​alue=0)
nique
df.index returns row indexes s.isnull() ; s.notn​ull()
pd.get​_du​‐ dummie column (0 or 1) that
df[co​l].lo​c[ ​index] returns value with mmi​es(df, indicates whether the entry s.dropna() drops all rows with
given column and column​s= in another column is a missing values
index [​'Na​me']) certain entry s.fill​na(x)
df.loc[:, 'col_n+1'] referring to a col that np.ran​‐ selects random set s = pd.Dat​aFr​ame​‐ Best way to define
=x doesn't exist creates a dom.ch​‐ ({'​Siz​e':s1, 'Weigh​‐ dataframe out of
new one oice(n, x, t':s2}) series:
replac​e=f​‐ Give dict out of
Basic Dataframe Analysis alse) columns
df.isn​ull() returns bool np.set​dif​‐ New set with only the 'e' in s1 returns bool
df[.is​nul​l().sum returns sum of trues f1d​(set_1, differing entries
s.name = 'str' names series
set_2)
df.isn​ull​().any checks whether there is s.inde​x.name = 'str' names index
a true df.to_​‐ gives array of entries
If s doesn't exist, this
numpy()
df[col].max() creates a df
df[col].min() s.colu​mns​['Red', 'Green']
Working with a Dataframe
df[col].idxmax() s.columns.name = 'Color'
df['c​ol1'] == x bool if entry is x
df[col].idxmin() s.rein​dex​[('m', 'n', ffill = forward fill
df[df == x] = y replace all values of a kind 'o'], method = 'ffill')
df[c​ol​].​med​ian()

df[c​ol​].​mean() Label-​based indexing with .loc / .iloc

df[c​ol​].​des​‐ gives statistic analysis
df.loc​[r​owi​ndex, colum​nname]
cribe()
df.loc[3, col1] 3rd entry of 1st
df[c​ol​].​qua​‐ 50% quantile df.loc[3:6, ['col1', column
nti​le(.5) 'col2']]
df.box​plot(by = boxplot grouped by
'col') column
df.his​t(bins = 20) histogram in 20 bars

By LasseBoe (Lasse1618) Not published yet. Sponsored by ApolloPad.com

cheatography.com/lasse1618/ Last updated 23rd November, 2020. Everyone has a novel in them. Finish
Page 1 of 1. Yours!
https://apollopad.com