Data Wrangling
F
with pandas M A Cheat Sheet
http://pandas.pydata.org
Syntax – Creating DataFrames
Tidy Data – A foundation for wrangling in pandas
In a tidy data set:
FMA
&
Each variable is saved in its own column
Tidy data complements pandas’s vectorized
operations. pandas will
automatically preserve observations as you manipulate variables. No other
format works as intuitively with pandas.
Each observation is saved in its own row Reshaping
Data – Change the layout of a data set
abc
1 4 7 10
2 5 8 11
3 6 9 12 df
= pd.DataFrame(
{"a" : [4 ,5, 6], "b" : [7, 8, 9], "c" : [10, 11, 12]}, index = [1, 2, 3]) Specify values for
each column.
df = pd.DataFrame(
[[4, 7, 10], [5, 8, 11], [6, 9, 12]], index=[1, 2, 3], columns=['a', 'b', 'c']) Specify values
for each row.
abc
nv
1 4 7 10
d 2 5 8 11
e 2 6 9 12 df
= pd.DataFrame(
{"a" : [4 ,5, 6], "b" : [7, 8, 9], "c" : [10, 11, 12]}, index = pd.MultiIndex.from_tuples(
[('d',1),('d',2),('e',2)],
names=['n','v']))) Create DataFrame with a MultiIndex
Method Chaining
Most pandas methods return a DataFrame so that another pandas method can be
applied to the result. This improves readability of code. df = (pd.melt(df)
.rename(columns={
'variable' : 'var', 'value' : 'val'}) .query('val >= 200') )
df[['width','length','species']] df[df.Length > 7]
Extract rows that meet logical criteria. df.drop_duplicates()
Remove duplicate rows (only considers columns).
df.sample(frac=0.5)
Randomly select fraction of rows. df.sample(n=10)
Randomly select n rows. df.iloc[10:20]
Select rows by position.
Select multiple columns with specific names. df['width'] or df.width
Select single column with specific name. df.filter(regex='regex')
Select columns whose name matches regular expression regex.
df.head(n)
df.nlargest(n, 'value') Select first n rows.
Select and order top n entries. df.tail(n)
df.nsmallest(n, 'value') Select last n rows.
Select and order bottom n entries.
Logic in Python (and pandas)
< Less than != Not equal to
df.loc[:,'x2':'x4'] > Greater than df.column.isin(values) Group membership
Select all columns between x2 and x4 (inclusive).
== Equals pd.isnull(obj) Is NaN
df.iloc[:,[1,2,5]]
<= Less than or equals pd.notnull(obj) Is not NaN
>= Greater than or equals &,|,~,^,df.any(),df.all() Logical and, or, not, xor, any, all
regex (Regular Expressions) Examples
'\.' Matches strings containing a period '.'
'Length$' Matches strings ending with word 'Length'
'^Sepal' Matches strings beginning with the word 'Sepal'
'^x[1-5]$' Matches strings beginning with 'x' and ending with 1,2,3,4,5
''^(?!Species$).*' Matches strings except the string 'Species'
Select columns in positions 1, 2 and 5 (first column is 0). df.loc[df['a'] > 10, ['a','c']]
Select rows meeting logical condition, and only the specific columns . http://pandas.pydata.org/
This cheat sheet inspired by Rstudio Data Wrangling Cheatsheet
(https://www.rstudio.com/wp-content/uploads/2015/02/data-wrangling-cheatsheet.pdf) Written by Irv Lustig, Princeton Consultants
M A pd.melt(df) Gather columns into rows.
M
* AF
*
df.pivot(columns='var', values='val')
Spread rows into columns.
pd.concat([df1,df2])
Append rows of DataFrames
df.sort_values('mpg')
Order rows by values of a column (low to high).
df.sort_values('mpg',ascending=False) Order rows by values of a column (high to
low).
df.rename(columns = {'y':'year'})
Rename the columns of a DataFrame
df.sort_index()
Sort the index of a DataFrame
df.reset_index()
Reset index of DataFrame to row numbers, moving index to columns.
pd.concat([df1,df2], axis=1)
df.drop(columns=['Length','Height']) Append columns of DataFrames
Drop columns from DataFrame
Subset Observations (Rows)
Subset Variables (Columns)
http://pandas.pydata.org/ This cheat sheet inspired by Rstudio Data Wrangling Cheatsheet
(https://www.rstudio.com/wp-content/uploads/2015/02/data-wrangling-cheatsheet.pdf) Written by Irv Lustig, Princeton Consultants
Summarize Data
Make New Columns
Combine Data Sets
df['w'].value_counts()
Count number of rows with each unique value of variable len(df)
# of rows in DataFrame. df['w'].nunique()
# of distinct values in a column. df.describe()
Basic descriptive statistics for each column (or GroupBy)
pandas provides a large set of summary functions that operate on different kinds of
pandas objects (DataFrame columns, Series, GroupBy, Expanding and Rolling (see
below)) and produce single values for each of the groups. When applied to a
DataFrame, the result is returned as a pandas Series for each column. Examples:
sum()
Sum values of each object. count()
Count non-NA/null values of each object. median()
Median value of each object. quantile([0.25,0.75]) Quantiles of each object.
apply(function)
Apply function to each object.
Handling Missing Data
df.dropna()
Drop rows with any column having NA/null data. df.fillna(value)
Replace all NA/null data with value.
Plotting
df.plot.hist()
Histogram for each column
adf bdf x1 x2 A 1 B 2 C 3 Standard Joins
x1 x2 x3 A 1 T B 2 F C 3 NaN
x1 x2 x3 A 1.0 T B 2.0 F D NaN T
x1 x2 x3 A 1 T B 2 F
x1 x2 x3 A 1 T B 2 F C 3 NaN D NaN T
x1 x3 A T B F D T
pd.merge(adf, bdf,
how='left', on='x1') Join matching rows from bdf to adf.
df.assign(Area=lambda df: df.Length*df.Height)
Compute and append one or more new columns.
pd.merge(adf, bdf, df['Volume'] = df.Length*df.Height*df.Depth
how='right', on='x1') Add single column.
Join matching rows from adf to bdf. pd.qcut(df.col, n, labels=False)
Bin column into n buckets. min()
Minimum value in each object. max()
pd.merge(adf, bdf,
how='inner', on='x1') Vector function
Join data. Retain only rows in both sets. Maximum value in each object. mean()
Mean value of each object. var()
Vector function
pd.merge(adf, bdf, pandas provides a large set of vector functions that operate on all
how='outer', on='x1') columns of a DataFrame or a single selected column (a pandas
Join data. Retain all values, all rows. Variance of each object. std()
Series). These functions produce vectors of values for each of the columns, or a single
Series for the individual Series. Examples: Standard deviation of each
Filtering Joins object.
x1 x2 A 1 B 2
x1 x2 C 3
shift(1)
Copy with values shifted by 1. rank(method='dense')
Ranks with no gaps. rank(method='min')
Ranks. Ties get min rank. rank(pct=True)
Ranks rescaled to interval [0, 1]. rank(method='first')
Ranks. Ties go to first value.
min(axis=1)
Element-wise min. abs()
Absolute value.
The examples below can also be applied to groups. In this case, the function is applied
on a per-group basis, and the returned vectors are of the length of the original
DataFrame.
Windows
df.expanding()
Return an Expanding object allowing summary functions to be applied cumulatively.
df.rolling(n)
Return a Rolling object allowing summary functions to be applied to windows of length
n.
max(axis=1)
Element-wise max. clip(lower=-10,upper=10) Trim values at input thresholds
adf[adf.x1.isin(bdf.x1)] Group Data
All rows in adf that have a match in bdf.
df.groupby(by="col")
adf[~adf.x1.isin(bdf.x1)] Return a GroupBy object,
All rows in adf that do not have a match in bdf. grouped by values in column named
"col".
df.groupby(level="ind")
Return a GroupBy object, grouped by values in index level named "ind".
x1 x2 A 1 B 2 C 3
All of the summary functions listed above can be applied to a group. Additional GroupBy
functions:
shift(-1)
ydf zdf Copy with values lagged by 1. cumsum()
Cumulative sum. cummax()
Cumulative max. cummin()
Cumulative min. cumprod()
Cumulative product.
Set-like Operations x1 x2 B 2 C 3
x1 x2 A 1 B 2 C 3 D 4
x1 x2 A 1
x1 x2 B 2 C 3 D 4
pd.merge(ydf, zdf) size()
agg(function)
Rows that appear in both ydf and zdf Size of each group.
Aggregate group using function.
(Intersection).
pd.merge(ydf, zdf, how='outer')
Rows that appear in either or both ydf and zdf
df.plot.scatter(x='w',y='h')
(Union).
Scatter chart using pairs of points
pd.merge(ydf, zdf, how='outer',
indicator=True) .query('_merge == "left_only"') .drop(columns=['_merge'])
Rows that appear in ydf but not zdf (Setdiff).
