10 minutes to Mars DataFrame

This is a short introduction to Mars DataFrame which is originated from 10 minutes to pandas.

Customarily, we import as follows:

In [1]: import mars.tensor as mt

In [2]: import mars.dataframe as md

Object creation

Creating a Series by passing a list of values, letting it create a default integer index:

In [3]: s = md.Series([1, 3, 5, mt.nan, 6, 8])

In [4]: s.execute()
Out[4]: 
0    1.0
1    3.0
2    5.0
3    NaN
4    6.0
5    8.0
dtype: float64

Creating a DataFrame by passing a Mars tensor, with a datetime index and labeled columns:

In [5]: dates = md.date_range('20130101', periods=6)

In [6]: dates.execute()
Out[6]: 
DatetimeIndex(['2013-01-01', '2013-01-02', '2013-01-03', '2013-01-04',
               '2013-01-05', '2013-01-06'],
              dtype='datetime64[ns]', freq='D')

In [7]: df = md.DataFrame(mt.random.randn(6, 4), index=dates, columns=list('ABCD'))

In [8]: df.execute()
Out[8]: 
                   A         B         C         D
2013-01-01  1.380039 -0.303651  0.344097  1.020950
2013-01-02 -0.683558  0.947133 -1.996983 -1.215373
2013-01-03  1.208587  0.539496 -0.206538 -0.089361
2013-01-04 -1.856555  0.776693 -0.152847 -0.712563
2013-01-05 -0.828458 -0.096248  0.764279  0.217016
2013-01-06  0.080176 -0.973207  1.474233 -0.528768

Creating a DataFrame by passing a dict of objects that can be converted to series-like.

In [9]: df2 = md.DataFrame({'A': 1.,
   ...:                     'B': md.Timestamp('20130102'),
   ...:                     'C': md.Series(1, index=list(range(4)), dtype='float32'),
   ...:                     'D': mt.array([3] * 4, dtype='int32'),
   ...:                     'E': 'foo'})
   ...: 

In [10]: df2.execute()
Out[10]: 
     A          B    C  D    E
0  1.0 2013-01-02  1.0  3  foo
1  1.0 2013-01-02  1.0  3  foo
2  1.0 2013-01-02  1.0  3  foo
3  1.0 2013-01-02  1.0  3  foo

The columns of the resulting DataFrame have different dtypes.

In [11]: df2.dtypes
Out[11]: 
A           float64
B    datetime64[ns]
C           float32
D             int32
E            object
dtype: object

Viewing data

Here is how to view the top and bottom rows of the frame:

In [12]: df.head().execute()
Out[12]: 
                   A         B         C         D
2013-01-01  1.380039 -0.303651  0.344097  1.020950
2013-01-02 -0.683558  0.947133 -1.996983 -1.215373
2013-01-03  1.208587  0.539496 -0.206538 -0.089361
2013-01-04 -1.856555  0.776693 -0.152847 -0.712563
2013-01-05 -0.828458 -0.096248  0.764279  0.217016

In [13]: df.tail(3).execute()
Out[13]: 
                   A         B         C         D
2013-01-04 -1.856555  0.776693 -0.152847 -0.712563
2013-01-05 -0.828458 -0.096248  0.764279  0.217016
2013-01-06  0.080176 -0.973207  1.474233 -0.528768

Display the index, columns:

In [14]: df.index.execute()
Out[14]: 
DatetimeIndex(['2013-01-01', '2013-01-02', '2013-01-03', '2013-01-04',
               '2013-01-05', '2013-01-06'],
              dtype='datetime64[ns]', freq='D')

In [15]: df.columns.execute()
Out[15]: Index(['A', 'B', 'C', 'D'], dtype='object')

DataFrame.to_tensor() gives a Mars tensor representation of the underlying data. Note that this can be an expensive operation when your DataFrame has columns with different data types, which comes down to a fundamental difference between DataFrame and tensor: tensors have one dtype for the entire tensor, while DataFrames have one dtype per column. When you call DataFrame.to_tensor(), Mars DataFrame will find the tensor dtype that can hold all of the dtypes in the DataFrame. This may end up being object, which requires casting every value to a Python object.

For df, our DataFrame of all floating-point values, DataFrame.to_tensor() is fast and doesn’t require copying data.

In [16]: df.to_tensor().execute()
Out[16]: 
array([[ 1.38003874, -0.30365081,  0.3440968 ,  1.02094996],
       [-0.68355761,  0.94713257, -1.99698258, -1.21537257],
       [ 1.20858735,  0.5394963 , -0.20653791, -0.0893615 ],
       [-1.85655497,  0.77669312, -0.15284695, -0.71256303],
       [-0.82845757, -0.09624842,  0.76427906,  0.21701598],
       [ 0.08017614, -0.97320724,  1.47423281, -0.52876787]])

For df2, the DataFrame with multiple dtypes, DataFrame.to_tensor() is relatively expensive.

In [17]: df2.to_tensor().execute()
Out[17]: 
array([[1.0, Timestamp('2013-01-02 00:00:00'), 1.0, 3, 'foo'],
       [1.0, Timestamp('2013-01-02 00:00:00'), 1.0, 3, 'foo'],
       [1.0, Timestamp('2013-01-02 00:00:00'), 1.0, 3, 'foo'],
       [1.0, Timestamp('2013-01-02 00:00:00'), 1.0, 3, 'foo']],
      dtype=object)

Note

DataFrame.to_tensor() does not include the index or column labels in the output.

describe() shows a quick statistic summary of your data:

In [18]: df.describe().execute()
Out[18]: 
              A         B         C         D
count  6.000000  6.000000  6.000000  6.000000
mean  -0.116628  0.148369  0.037707 -0.218017
std    1.256216  0.735920  1.176547  0.784437
min   -1.856555 -0.973207 -1.996983 -1.215373
25%   -0.792233 -0.251800 -0.193115 -0.666614
50%   -0.301691  0.221624  0.095625 -0.309065
75%    0.926485  0.717394  0.659233  0.140422
max    1.380039  0.947133  1.474233  1.020950

Sorting by an axis:

In [19]: df.sort_index(axis=1, ascending=False).execute()
Out[19]: 
                   D         C         B         A
2013-01-01  1.020950  0.344097 -0.303651  1.380039
2013-01-02 -1.215373 -1.996983  0.947133 -0.683558
2013-01-03 -0.089361 -0.206538  0.539496  1.208587
2013-01-04 -0.712563 -0.152847  0.776693 -1.856555
2013-01-05  0.217016  0.764279 -0.096248 -0.828458
2013-01-06 -0.528768  1.474233 -0.973207  0.080176

Sorting by values:

In [20]: df.sort_values(by='B').execute()
Out[20]: 
                   A         B         C         D
2013-01-06  0.080176 -0.973207  1.474233 -0.528768
2013-01-01  1.380039 -0.303651  0.344097  1.020950
2013-01-05 -0.828458 -0.096248  0.764279  0.217016
2013-01-03  1.208587  0.539496 -0.206538 -0.089361
2013-01-04 -1.856555  0.776693 -0.152847 -0.712563
2013-01-02 -0.683558  0.947133 -1.996983 -1.215373

Selection

Note

While standard Python / Numpy expressions for selecting and setting are intuitive and come in handy for interactive work, for production code, we recommend the optimized DataFrame data access methods, .at, .iat, .loc and .iloc.

Getting

Selecting a single column, which yields a Series, equivalent to df.A:

In [21]: df['A'].execute()
Out[21]: 
2013-01-01    1.380039
2013-01-02   -0.683558
2013-01-03    1.208587
2013-01-04   -1.856555
2013-01-05   -0.828458
2013-01-06    0.080176
Freq: D, Name: A, dtype: float64

Selecting via [], which slices the rows.

In [22]: df[0:3].execute()
Out[22]: 
                   A         B         C         D
2013-01-01  1.380039 -0.303651  0.344097  1.020950
2013-01-02 -0.683558  0.947133 -1.996983 -1.215373
2013-01-03  1.208587  0.539496 -0.206538 -0.089361

In [23]: df['20130102':'20130104'].execute()
Out[23]: 
                   A         B         C         D
2013-01-02 -0.683558  0.947133 -1.996983 -1.215373
2013-01-03  1.208587  0.539496 -0.206538 -0.089361
2013-01-04 -1.856555  0.776693 -0.152847 -0.712563

Selection by label

For getting a cross section using a label:

In [24]: df.loc['20130101'].execute()
Out[24]: 
A    1.380039
B   -0.303651
C    0.344097
D    1.020950
Name: 2013-01-01 00:00:00, dtype: float64

Selecting on a multi-axis by label:

In [25]: df.loc[:, ['A', 'B']].execute()
Out[25]: 
                   A         B
2013-01-01  1.380039 -0.303651
2013-01-02 -0.683558  0.947133
2013-01-03  1.208587  0.539496
2013-01-04 -1.856555  0.776693
2013-01-05 -0.828458 -0.096248
2013-01-06  0.080176 -0.973207

Showing label slicing, both endpoints are included:

In [26]: df.loc['20130102':'20130104', ['A', 'B']].execute()
Out[26]: 
                   A         B
2013-01-02 -0.683558  0.947133
2013-01-03  1.208587  0.539496
2013-01-04 -1.856555  0.776693

Reduction in the dimensions of the returned object:

In [27]: df.loc['20130102', ['A', 'B']].execute()
Out[27]: 
A   -0.683558
B    0.947133
Name: 2013-01-02 00:00:00, dtype: float64

For getting a scalar value:

In [28]: df.loc['20130101', 'A'].execute()
Out[28]: 1.3800387404592631

For getting fast access to a scalar (equivalent to the prior method):

In [29]: df.at['20130101', 'A'].execute()
Out[29]: 1.3800387404592631

Selection by position

Select via the position of the passed integers:

In [30]: df.iloc[3].execute()
Out[30]: 
A   -1.856555
B    0.776693
C   -0.152847
D   -0.712563
Name: 2013-01-04 00:00:00, dtype: float64

By integer slices, acting similar to numpy/python:

In [31]: df.iloc[3:5, 0:2].execute()
Out[31]: 
                   A         B
2013-01-04 -1.856555  0.776693
2013-01-05 -0.828458 -0.096248

By lists of integer position locations, similar to the numpy/python style:

In [32]: df.iloc[[1, 2, 4], [0, 2]].execute()
Out[32]: 
                   A         C
2013-01-02 -0.683558 -1.996983
2013-01-03  1.208587 -0.206538
2013-01-05 -0.828458  0.764279

For slicing rows explicitly:

In [33]: df.iloc[1:3, :].execute()
Out[33]: 
                   A         B         C         D
2013-01-02 -0.683558  0.947133 -1.996983 -1.215373
2013-01-03  1.208587  0.539496 -0.206538 -0.089361

For slicing columns explicitly:

In [34]: df.iloc[:, 1:3].execute()
Out[34]: 
                   B         C
2013-01-01 -0.303651  0.344097
2013-01-02  0.947133 -1.996983
2013-01-03  0.539496 -0.206538
2013-01-04  0.776693 -0.152847
2013-01-05 -0.096248  0.764279
2013-01-06 -0.973207  1.474233

For getting a value explicitly:

In [35]: df.iloc[1, 1].execute()
Out[35]: 0.947132573716667

For getting fast access to a scalar (equivalent to the prior method):

In [36]: df.iat[1, 1].execute()
Out[36]: 0.947132573716667

Boolean indexing

Using a single column’s values to select data.

In [37]: df[df['A'] > 0].execute()
Out[37]: 
                   A         B         C         D
2013-01-01  1.380039 -0.303651  0.344097  1.020950
2013-01-03  1.208587  0.539496 -0.206538 -0.089361
2013-01-06  0.080176 -0.973207  1.474233 -0.528768

Selecting values from a DataFrame where a boolean condition is met.

In [38]: df[df > 0].execute()
Out[38]: 
                   A         B         C         D
2013-01-01  1.380039       NaN  0.344097  1.020950
2013-01-02       NaN  0.947133       NaN       NaN
2013-01-03  1.208587  0.539496       NaN       NaN
2013-01-04       NaN  0.776693       NaN       NaN
2013-01-05       NaN       NaN  0.764279  0.217016
2013-01-06  0.080176       NaN  1.474233       NaN

Operations

Stats

Operations in general exclude missing data.

Performing a descriptive statistic:

In [39]: df.mean().execute()
Out[39]: 
A   -0.116628
B    0.148369
C    0.037707
D   -0.218017
dtype: float64

Same operation on the other axis:

In [40]: df.mean(1).execute()
Out[40]: 
2013-01-01    0.610359
2013-01-02   -0.737195
2013-01-03    0.363046
2013-01-04   -0.486318
2013-01-05    0.014147
2013-01-06    0.013108
Freq: D, dtype: float64

Operating with objects that have different dimensionality and need alignment. In addition, Mars DataFrame automatically broadcasts along the specified dimension.

In [41]: s = md.Series([1, 3, 5, mt.nan, 6, 8], index=dates).shift(2)

In [42]: s.execute()
Out[42]: 
2013-01-01    NaN
2013-01-02    NaN
2013-01-03    1.0
2013-01-04    3.0
2013-01-05    5.0
2013-01-06    NaN
Freq: D, dtype: float64

In [43]: df.sub(s, axis='index').execute()
Out[43]: 
                   A         B         C         D
2013-01-01       NaN       NaN       NaN       NaN
2013-01-02       NaN       NaN       NaN       NaN
2013-01-03  0.208587 -0.460504 -1.206538 -1.089361
2013-01-04 -4.856555 -2.223307 -3.152847 -3.712563
2013-01-05 -5.828458 -5.096248 -4.235721 -4.782984
2013-01-06       NaN       NaN       NaN       NaN

Apply

Applying functions to the data:

In [44]: df.apply(lambda x: x.max() - x.min()).execute()
Out[44]: 
A    3.236594
B    1.920340
C    3.471215
D    2.236323
dtype: float64

String Methods

Series is equipped with a set of string processing methods in the str attribute that make it easy to operate on each element of the array, as in the code snippet below. Note that pattern-matching in str generally uses regular expressions by default (and in some cases always uses them). See more at Vectorized String Methods.

In [45]: s = md.Series(['A', 'B', 'C', 'Aaba', 'Baca', mt.nan, 'CABA', 'dog', 'cat'])

In [46]: s.str.lower().execute()
Out[46]: 
0       a
1       b
2       c
3    aaba
4    baca
5     NaN
6    caba
7     dog
8     cat
dtype: object

Merge

Concat

Mars DataFrame provides various facilities for easily combining together Series and DataFrame objects with various kinds of set logic for the indexes and relational algebra functionality in the case of join / merge-type operations.

Concatenating DataFrame objects together with concat():

In [47]: df = md.DataFrame(mt.random.randn(10, 4))

In [48]: df.execute()
Out[48]: 
          0         1         2         3
0  0.009272 -0.659995 -0.378903 -1.410914
1 -1.824559  1.399596 -0.953777 -1.109440
2  0.611343  0.584350  1.876943 -1.694674
3 -0.427432 -1.107903 -0.436457  0.290908
4  0.609468  0.096552  0.178083  1.152755
5  1.876762  0.334645  0.290916  0.525072
6  1.280286  1.981358 -2.052386  0.717608
7  0.768622  0.810061  0.151142 -0.855658
8 -0.816228  0.791419 -1.255221  0.401390
9 -1.781242 -1.562540  0.345979  0.392550

# break it into pieces
In [49]: pieces = [df[:3], df[3:7], df[7:]]

In [50]: md.concat(pieces).execute()
Out[50]: 
          0         1         2         3
0  0.009272 -0.659995 -0.378903 -1.410914
1 -1.824559  1.399596 -0.953777 -1.109440
2  0.611343  0.584350  1.876943 -1.694674
3 -0.427432 -1.107903 -0.436457  0.290908
4  0.609468  0.096552  0.178083  1.152755
5  1.876762  0.334645  0.290916  0.525072
6  1.280286  1.981358 -2.052386  0.717608
7  0.768622  0.810061  0.151142 -0.855658
8 -0.816228  0.791419 -1.255221  0.401390
9 -1.781242 -1.562540  0.345979  0.392550

Note

Adding a column to a DataFrame is relatively fast. However, adding a row requires a copy, and may be expensive. We recommend passing a pre-built list of records to the DataFrame constructor instead of building a DataFrame by iteratively appending records to it.

Join

SQL style merges. See the Database style joining section.

In [51]: left = md.DataFrame({'key': ['foo', 'foo'], 'lval': [1, 2]})

In [52]: right = md.DataFrame({'key': ['foo', 'foo'], 'rval': [4, 5]})

In [53]: left.execute()
Out[53]: 
   key  lval
0  foo     1
1  foo     2

In [54]: right.execute()
Out[54]: 
   key  rval
0  foo     4
1  foo     5

In [55]: md.merge(left, right, on='key').execute()
Out[55]: 
   key  lval  rval
0  foo     1     4
1  foo     1     5
2  foo     2     4
3  foo     2     5

Another example that can be given is:

In [56]: left = md.DataFrame({'key': ['foo', 'bar'], 'lval': [1, 2]})

In [57]: right = md.DataFrame({'key': ['foo', 'bar'], 'rval': [4, 5]})

In [58]: left.execute()
Out[58]: 
   key  lval
0  foo     1
1  bar     2

In [59]: right.execute()
Out[59]: 
   key  rval
0  foo     4
1  bar     5

In [60]: md.merge(left, right, on='key').execute()
Out[60]: 
   key  lval  rval
0  foo     1     4
1  bar     2     5

Grouping

By “group by” we are referring to a process involving one or more of the following steps:

  • Splitting the data into groups based on some criteria

  • Applying a function to each group independently

  • Combining the results into a data structure

In [61]: df = md.DataFrame({'A': ['foo', 'bar', 'foo', 'bar',
   ....:                          'foo', 'bar', 'foo', 'foo'],
   ....:                    'B': ['one', 'one', 'two', 'three',
   ....:                          'two', 'two', 'one', 'three'],
   ....:                    'C': mt.random.randn(8),
   ....:                    'D': mt.random.randn(8)})
   ....: 

In [62]: df.execute()
Out[62]: 
     A      B         C         D
0  foo    one  1.214069  0.058317
1  bar    one  0.348770 -0.747681
2  foo    two  1.204847 -0.550463
3  bar  three -0.750572  0.537047
4  foo    two -1.489774 -1.829142
5  bar    two -0.896199 -2.055427
6  foo    one  0.702748  0.535140
7  foo  three -0.781145  1.228572

Grouping and then applying the sum() function to the resulting groups.

In [63]: df.groupby('A').sum().execute()
Out[63]: 
            C         D
A                      
bar -1.298001 -2.266061
foo  0.850746 -0.557578

Grouping by multiple columns forms a hierarchical index, and again we can apply the sum function.

In [64]: df.groupby(['A', 'B']).sum().execute()
Out[64]: 
                  C         D
A   B                        
foo one    1.916817  0.593457
    two   -0.284926 -2.379606
    three -0.781145  1.228572
bar one    0.348770 -0.747681
    two   -0.896199 -2.055427
    three -0.750572  0.537047

Plotting

We use the standard convention for referencing the matplotlib API:

In [65]: import matplotlib.pyplot as plt

In [66]: plt.close('all')
In [67]: ts = md.Series(mt.random.randn(1000),
   ....:                index=md.date_range('1/1/2000', periods=1000))
   ....: 

In [68]: ts = ts.cumsum()

In [69]: ts.plot()
Out[69]: <AxesSubplot:>
../../_images/series_plot_basic.png

On a DataFrame, the plot() method is a convenience to plot all of the columns with labels:

In [70]: df = md.DataFrame(mt.random.randn(1000, 4), index=ts.index,
   ....:                   columns=['A', 'B', 'C', 'D'])
   ....: 

In [71]: df = df.cumsum()

In [72]: plt.figure()
Out[72]: <Figure size 640x480 with 0 Axes>

In [73]: df.plot()
Out[73]: <AxesSubplot:>

In [74]: plt.legend(loc='best')
Out[74]: <matplotlib.legend.Legend at 0x7f55f8a66890>
../../_images/frame_plot_basic.png

Getting data in/out

CSV

In [75]: df.to_csv('foo.csv').execute()
Out[75]: 
Empty DataFrame
Columns: []
Index: []

Reading from a csv file.

In [76]: md.read_csv('foo.csv').execute()
Out[76]: 
     Unnamed: 0          A          B          C          D
0    2000-01-01   3.538462   0.708814   0.867920  -0.681444
1    2000-01-02   3.327751   0.060214   0.170646  -1.239853
2    2000-01-03   3.736594  -0.790232  -0.955341   0.777303
3    2000-01-04   3.507770  -0.902764   0.019091   0.185178
4    2000-01-05   2.706285  -0.937815   1.130297  -0.302031
..          ...        ...        ...        ...        ...
995  2002-09-22  14.431740 -58.937363  20.424355  13.882051
996  2002-09-23  13.428467 -58.334534  19.236940  14.002610
997  2002-09-24  15.259785 -58.713229  19.039831  13.710882
998  2002-09-25  16.104898 -59.547079  17.827294  14.131420
999  2002-09-26  16.785207 -59.716472  18.613674  14.646230

[1000 rows x 5 columns]