Options and Settings

Overview

pandas has an options system that lets you customize some aspects of it’s behaviour, display-related options being those the user is most likely to adjust.

Options have a full “dotted-style”, case-insensitive name (e.g. display.max_rows), You can get/set options directly as attributes of the top-level options attribute:

In [1]: import pandas as pd

In [2]: pd.options.display.max_rows
Out[2]: 15

In [3]: pd.options.display.max_rows = 999

In [4]: pd.options.display.max_rows
Out[4]: 999

There is also an API composed of 5 relevant functions, available directly from the pandas namespace, and they are:

Note: developers can check out pandas/core/config.py for more info.

All of the functions above accept a regexp pattern (re.search style) as an argument, and so passing in a substring will work - as long as it is unambiguous :

In [5]: pd.get_option("display.max_rows")
Out[5]: 999

In [6]: pd.set_option("display.max_rows",101)

In [7]: pd.get_option("display.max_rows")
Out[7]: 101

In [8]: pd.set_option("max_r",102)

In [9]: pd.get_option("display.max_rows")
Out[9]: 102

The following will not work because it matches multiple option names, e.g. display.max_colwidth, display.max_rows, display.max_columns:

In [10]: try:
   ....:     pd.get_option("column")
   ....: except KeyError as e:
   ....:     print(e)
   ....: 
'Pattern matched multiple keys'

Note: Using this form of shorthand may cause your code to break if new options with similar names are added in future versions.

You can get a list of available options and their descriptions with describe_option. When called with no argument describe_option will print out the descriptions for all available options.

Getting and Setting Options

As described above, get_option() and set_option() are available from the pandas namespace. To change an option, call set_option('option regex', new_value)

In [11]: pd.get_option('mode.sim_interactive')
Out[11]: False

In [12]: pd.set_option('mode.sim_interactive', True)

In [13]: pd.get_option('mode.sim_interactive')
Out[13]: True

All options also have a default value, and you can use reset_option to do just that:

In [14]: pd.get_option("display.max_rows")
Out[14]: 60

In [15]: pd.set_option("display.max_rows",999)

In [16]: pd.get_option("display.max_rows")
Out[16]: 999

In [17]: pd.reset_option("display.max_rows")

In [18]: pd.get_option("display.max_rows")
Out[18]: 60

It’s also possible to reset multiple options at once (using a regex):

In [19]: pd.reset_option("^display")
height has been deprecated.

line_width has been deprecated, use display.width instead (currently both are
identical)

option_context context manager has been exposed through the top-level API, allowing you to execute code with given option values. Option values are restored automatically when you exit the with block:

In [20]: with pd.option_context("display.max_rows",10,"display.max_columns", 5):
   ....:      print(pd.get_option("display.max_rows"))
   ....:      print(pd.get_option("display.max_columns"))
   ....: 
10
5

In [21]: print(pd.get_option("display.max_rows"))
60

In [22]: print(pd.get_option("display.max_columns"))
20

Frequently Used Options

The following is a walkthrough of the more frequently used display options.

display.max_rows and display.max_columns sets the maximum number of rows and columns displayed when a frame is pretty-printed. Truncated lines are replaced by an ellipsis.

In [23]: df=pd.DataFrame(np.random.randn(7,2))

In [24]: pd.set_option('max_rows', 7)

In [25]: df
Out[25]: 
          0         1
0  0.469112 -0.282863
1 -1.509059 -1.135632
2  1.212112 -0.173215
3  0.119209 -1.044236
4 -0.861849 -2.104569
5 -0.494929  1.071804
6  0.721555 -0.706771

In [26]: pd.set_option('max_rows', 5)

In [27]: df
Out[27]: 
           0         1
0   0.469112 -0.282863
1  -1.509059 -1.135632
..       ...       ...
5  -0.494929  1.071804
6   0.721555 -0.706771

[7 rows x 2 columns]

In [28]: pd.reset_option('max_rows')

display.expand_frame_repr allows for the the representation of dataframes to stretch across pages, wrapped over the full column vs row-wise.

In [29]: df=pd.DataFrame(np.random.randn(5,10))

In [30]: pd.set_option('expand_frame_repr', True)

In [31]: df
Out[31]: 
          0         1         2         3         4         5         6  \
0 -1.039575  0.271860 -0.424972  0.567020  0.276232 -1.087401 -0.673690   
1  0.404705  0.577046 -1.715002 -1.039268 -0.370647 -1.157892 -1.344312   
2  1.643563 -1.469388  0.357021 -0.674600 -1.776904 -0.968914 -1.294524   
3 -0.013960 -0.362543 -0.006154 -0.923061  0.895717  0.805244 -1.206412   
4 -1.170299 -0.226169  0.410835  0.813850  0.132003 -0.827317 -0.076467   

          7         8         9  
0  0.113648 -1.478427  0.524988  
1  0.844885  1.075770 -0.109050  
2  0.413738  0.276662 -0.472035  
3  2.565646  1.431256  1.340309  
4 -1.187678  1.130127 -1.436737  

In [32]: pd.set_option('expand_frame_repr', False)

In [33]: df
Out[33]: 
          0         1         2         3         4         5         6         7         8         9
0 -1.039575  0.271860 -0.424972  0.567020  0.276232 -1.087401 -0.673690  0.113648 -1.478427  0.524988
1  0.404705  0.577046 -1.715002 -1.039268 -0.370647 -1.157892 -1.344312  0.844885  1.075770 -0.109050
2  1.643563 -1.469388  0.357021 -0.674600 -1.776904 -0.968914 -1.294524  0.413738  0.276662 -0.472035
3 -0.013960 -0.362543 -0.006154 -0.923061  0.895717  0.805244 -1.206412  2.565646  1.431256  1.340309
4 -1.170299 -0.226169  0.410835  0.813850  0.132003 -0.827317 -0.076467 -1.187678  1.130127 -1.436737

In [34]: pd.reset_option('expand_frame_repr')

display.large_repr lets you select whether to display dataframes that exceed max_columns or max_rows as a truncated frame, or as a summary.

In [35]: df=pd.DataFrame(np.random.randn(10,10))

In [36]: pd.set_option('max_rows', 5)

In [37]: pd.set_option('large_repr', 'truncate')

In [38]: df
Out[38]: 
           0         1         2         3         4         5         6  \
0  -1.413681  1.607920  1.024180  0.569605  0.875906 -2.211372  0.974466   
1   0.545952 -1.219217 -1.226825  0.769804 -1.281247 -0.727707 -0.121306   
..       ...       ...       ...       ...       ...       ...       ...   
8  -2.484478 -0.281461  0.030711  0.109121  1.126203 -0.977349  1.474071   
9  -1.071357  0.441153  2.353925  0.583787  0.221471 -0.744471  0.758527   

           7         8         9  
0  -2.006747 -0.410001 -0.078638  
1  -0.097883  0.695775  0.341734  
..       ...       ...       ...  
8  -0.064034 -1.282782  0.781836  
9   1.729689 -0.964980 -0.845696  

[10 rows x 10 columns]

In [39]: pd.set_option('large_repr', 'info')

In [40]: df
Out[40]: 
<class 'pandas.core.frame.DataFrame'>
Int64Index: 10 entries, 0 to 9
Data columns (total 10 columns):
0    10 non-null float64
1    10 non-null float64
2    10 non-null float64
3    10 non-null float64
4    10 non-null float64
5    10 non-null float64
6    10 non-null float64
7    10 non-null float64
8    10 non-null float64
9    10 non-null float64
dtypes: float64(10)

In [41]: pd.reset_option('large_repr')

In [42]: pd.reset_option('max_rows')

display.max_columnwidth sets the maximum width of columns. Cells of this length or longer will be truncated with an elipsis.

In [43]: df=pd.DataFrame(np.array([['foo', 'bar', 'bim', 'uncomfortably long string'],
   ....:                      ['horse', 'cow', 'banana', 'apple']]))
   ....: 

In [44]: pd.set_option('max_colwidth',40)

In [45]: df
Out[45]: 
       0    1       2                          3
0    foo  bar     bim  uncomfortably long string
1  horse  cow  banana                      apple

In [46]: pd.set_option('max_colwidth', 6)

In [47]: df
Out[47]: 
       0    1      2      3
0    foo  bar    bim  un...
1  horse  cow  ba...  apple

In [48]: pd.reset_option('max_colwidth')

display.max_info_columns sets a threshold for when by-column info will be given.

In [49]: df=pd.DataFrame(np.random.randn(10,10))

In [50]: pd.set_option('max_info_columns', 11)

In [51]: df.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 10 entries, 0 to 9
Data columns (total 10 columns):
0    10 non-null float64
1    10 non-null float64
2    10 non-null float64
3    10 non-null float64
4    10 non-null float64
5    10 non-null float64
6    10 non-null float64
7    10 non-null float64
8    10 non-null float64
9    10 non-null float64
dtypes: float64(10)
In [52]: pd.set_option('max_info_columns', 5)

In [53]: df.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 10 entries, 0 to 9
Columns: 10 entries, 0 to 9
dtypes: float64(10)
In [54]: pd.reset_option('max_info_columns')

display.max_info_rows: df.info() will usually show null-counts for each column. For large frames this can be quite slow. max_info_rows and max_info_cols limit this null check only to frames with smaller dimensions then specified.

In [55]: df=pd.DataFrame(np.random.choice([0,1,np.nan],size=(10,10)))

In [56]: df
Out[56]: 
    0   1   2   3   4   5   6   7   8   9
0   0   1   1   0   1   1   0 NaN   1 NaN
1   1 NaN   0   0   1   1 NaN   1   0   1
2 NaN NaN NaN   1   1   0 NaN   0   1 NaN
3   0   1   1 NaN   0 NaN   1 NaN NaN   0
4   0   1   0   0   1   0   0 NaN   0   0
5   0 NaN   1 NaN NaN NaN NaN   0   1 NaN
6   0   1   0   0 NaN   1 NaN NaN   0 NaN
7   0 NaN   1   1 NaN   1   1   1   1 NaN
8   0   0 NaN   0 NaN   1   0   0 NaN NaN
9 NaN NaN   0 NaN NaN NaN   0   1   1 NaN

In [57]: pd.set_option('max_info_rows', 11)

In [58]: df.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 10 entries, 0 to 9
Data columns (total 10 columns):
0    8 non-null float64
1    5 non-null float64
2    8 non-null float64
3    7 non-null float64
4    5 non-null float64
5    7 non-null float64
6    6 non-null float64
7    6 non-null float64
8    8 non-null float64
9    3 non-null float64
dtypes: float64(10)
In [59]: pd.set_option('max_info_rows', 5)

In [60]: df.info()
<class 'pandas.core.frame.DataFrame'>
Int64Index: 10 entries, 0 to 9
Data columns (total 10 columns):
0    float64
1    float64
2    float64
3    float64
4    float64
5    float64
6    float64
7    float64
8    float64
9    float64
dtypes: float64(10)
In [61]: pd.reset_option('max_info_rows')

display.precision sets the output display precision. This is only a suggestion.

In [62]: df=pd.DataFrame(np.random.randn(5,5))

In [63]: pd.set_option('precision',7)

In [64]: df
Out[64]: 
          0         1         2         3         4
0 -2.049028  2.846612 -1.208049 -0.450392  2.423905
1  0.121108  0.266916  0.843826 -0.222540  2.021981
2 -0.716789 -2.224485 -1.061137 -0.232825  0.430793
3 -0.665478  1.829807 -1.406509  1.078248  0.322774
4  0.200324  0.890024  0.194813  0.351633  0.448881

In [65]: pd.set_option('precision',4)

In [66]: df
Out[66]: 
       0      1      2      3      4
0 -2.049  2.847 -1.208 -0.450  2.424
1  0.121  0.267  0.844 -0.223  2.022
2 -0.717 -2.224 -1.061 -0.233  0.431
3 -0.665  1.830 -1.407  1.078  0.323
4  0.200  0.890  0.195  0.352  0.449

display.chop_threshold sets at what level pandas rounds to zero when it displays a Series of DataFrame. Note, this does not effect the precision at which the number is stored.

In [67]: df=pd.DataFrame(np.random.randn(6,6))

In [68]: pd.set_option('chop_threshold', 0)

In [69]: df
Out[69]: 
       0      1      2      3      4      5
0 -0.198  0.966 -1.523 -0.117  0.296 -1.048
1  1.641  1.906  2.772  0.089 -1.144 -0.633
2  0.925 -0.006 -0.820 -0.601 -1.039  0.825
3 -0.824 -0.338 -0.928 -0.840  0.249 -0.109
4  0.432 -0.461  0.337 -3.208 -1.536  0.410
5 -0.673 -0.741 -0.111 -2.673  0.864  0.061

In [70]: pd.set_option('chop_threshold', .5)

In [71]: df
Out[71]: 
       0      1      2      3      4      5
0  0.000  0.966 -1.523  0.000  0.000 -1.048
1  1.641  1.906  2.772  0.000 -1.144 -0.633
2  0.925  0.000 -0.820 -0.601 -1.039  0.825
3 -0.824  0.000 -0.928 -0.840  0.000  0.000
4  0.000  0.000  0.000 -3.208 -1.536  0.000
5 -0.673 -0.741  0.000 -2.673  0.864  0.000

In [72]: pd.reset_option('chop_threshold')

display.colheader_justify controls the justification of the headers. Options are ‘right’, and ‘left’.

In [73]: df=pd.DataFrame(np.array([np.random.randn(6), np.random.randint(1,9,6)*.1, np.zeros(6)]).T, columns=['A', 'B', 'C'], dtype='float')

In [74]: pd.set_option('colheader_justify', 'right')

In [75]: df
Out[75]: 
       A    B  C
0  0.933  0.3  0
1  0.289  0.2  0
2  1.325  0.2  0
3  0.589  0.7  0
4  0.531  0.1  0
5 -1.199  0.7  0

In [76]: pd.set_option('colheader_justify', 'left')

In [77]: df
Out[77]: 
   A      B    C
0  0.933  0.3  0
1  0.289  0.2  0
2  1.325  0.2  0
3  0.589  0.7  0
4  0.531  0.1  0
5 -1.199  0.7  0

In [78]: pd.reset_option('colheader_justify')

List of Options

Option Default Function
display.chop_threshold None If set to a float value, all float values smaller then the given threshold will be displayed as exactly 0 by repr and friends.
display.colheader_justify right Controls the justification of column headers. used by DataFrameFormatter.
display.column_space 12 No description available.
display.date_dayfirst False When True, prints and parses dates with the day first, eg 20/01/2005
display.date_yearfirst False When True, prints and parses dates with the year first, eg 2005/01/20
display.encoding UTF-8 Defaults to the detected encoding of the console. Specifies the encoding to be used for strings returned by to_string, these are generally strings meant to be displayed on the console.
display.expand_frame_repr True Whether to print out the full DataFrame repr for wide DataFrames across multiple lines, max_columns is still respected, but the output will wrap-around across multiple “pages” if it’s width exceeds display.width.
display.float_format None The callable should accept a floating point number and return a string with the desired format of the number. This is used in some places like SeriesFormatter. See core.format.EngFormatter for an example.
display.height 60 Deprecated. Use display.max_rows instead.
display.large_repr truncate For DataFrames exceeding max_rows/max_cols, the repr (and HTML repr) can show a truncated table (the default from 0.13), or switch to the view from df.info() (the behaviour in earlier versions of pandas). allowable settings, [‘truncate’, ‘info’]
display.line_width 80 Deprecated. Use display.width instead.
display.max_columns 20 max_rows and max_columns are used in __repr__() methods to decide if to_string() or info() is used to render an object to a string. In case python/IPython is running in a terminal this can be set to 0 and pandas will correctly auto-detect the width the terminal and swap to a smaller format in case all columns would not fit vertically. The IPython notebook, IPython qtconsole, or IDLE do not run in a terminal and hence it is not possible to do correct auto-detection. ‘None’ value means unlimited.
display.max_colwidth 50 The maximum width in characters of a column in the repr of a pandas data structure. When the column overflows, a ”...” placeholder is embedded in the output.
display.max_info_columns 100 max_info_columns is used in DataFrame.info method to decide if per column information will be printed.
display.max_info_rows 1690785 df.info() will usually show null-counts for each column. For large frames this can be quite slow. max_info_rows and max_info_cols limit this null check only to frames with smaller dimensions then specified.
display.max_rows 60 This sets the maximum number of rows pandas should output when printing out various output. For example, this value determines whether the repr() for a dataframe prints out fully or just a summary repr. ‘None’ value means unlimited.
display.max_seq_items 100 when pretty-printing a long sequence, no more then max_seq_items will be printed. If items are omitted, they will be denoted by the addition of ”...” to the resulting string. If set to None, the number of items to be printed is unlimited.
display.mpl_style None Setting this to ‘default’ will modify the rcParams used by matplotlib to give plots a more pleasing visual style by default. Setting this to None/False restores the values to their initial value.
display.multi_sparse True “Sparsify” MultiIndex display (don’t display repeated elements in outer levels within groups)
display.notebook_repr_html True When True, IPython notebook will use html representation for pandas objects (if it is available).
display.pprint_nest_depth 3 Controls the number of nested levels to process when pretty-printing
display.precision 7 Floating point output precision (number of significant digits). This is only a suggestion
display.show_dimensions truncate Whether to print out dimensions at the end of DataFrame repr. If ‘truncate’ is specified, only print out the dimensions if the frame is truncated (e.g. not display all rows and/or columns)
display.width 80 Width of the display in characters. In case python/IPython is running in a terminal this can be set to None and pandas will correctly auto-detect the width. Note that the IPython notebook, IPython qtconsole, or IDLE do not run in a terminal and hence it is not possible to correctly detect the width.
io.excel.xls.writer xlwt The default Excel writer engine for ‘xls’ files.
io.excel.xlsm.writer openpyxl The default Excel writer engine for ‘xlsm’ files. Available options: ‘openpyxl’ (the default).
io.excel.xlsx.writer openpyxl The default Excel writer engine for ‘xlsx’ files.
io.hdf.default_format None default format writing format, if None, then put will default to ‘fixed’ and append will default to ‘table’
io.hdf.dropna_table True drop ALL nan rows when appending to a table
mode.chained_assignment warn Raise an exception, warn, or no action if trying to use chained assignment, The default is warn
mode.sim_interactive False Whether to simulate interactive mode for purposes of testing
mode.use_inf_as_null False True means treat None, NaN, -INF, INF as null (old way), False means None and NaN are null, but INF, -INF are not null (new way).

Number Formatting

pandas also allow you to set how numbers are displayed in the console. This option is not set through the set_options API.

Use the set_eng_float_format function to alter the floating-point formatting of pandas objects to produce a particular format.

For instance:

In [79]: import numpy as np

In [80]: pd.set_eng_float_format(accuracy=3, use_eng_prefix=True)

In [81]: s = pd.Series(np.random.randn(5), index=['a', 'b', 'c', 'd', 'e'])

In [82]: s/1.e3
Out[82]: 
a   -236.866u
b    846.974u
c   -685.597u
d    609.099u
e   -303.961u
dtype: float64

In [83]: s/1.e6
Out[83]: 
a   -236.866n
b    846.974n
c   -685.597n
d    609.099n
e   -303.961n
dtype: float64