Teradata Package for Python Function Reference on VantageCloud Lake | 17.20 - count - Teradata Package for Python - Look here for syntax, methods and examples for the functions included in the Teradata Package for Python.

Teradata® Package for Python Function Reference on VantageCloud Lake

Deployment
VantageCloud
Edition
Lake
Product
Teradata Package for Python
Release Number
17.20
Publication Date
November 2022
Content Type
Programming Reference
Publication ID
B700-4500-112K
ft:locale
en-US
ft:lastEdition
2024-02-21
ft:sourceId
FTML
dita:id
TeradataPython_FxRef_Lake_1720
Product Category
Teradata Vantage
teradataml.dataframe.dataframe.DataFrame.count = count(self, distinct=False)
DESCRIPTION:
    Returns column-wise count of the dataframe.
 
PARAMETERS:
    distinct:
        Optional Argument.
        Specifies whether to exclude duplicate values while calculating the count.
        Default Value: False
        Types: bool
 
RETURNS:
    teradataml DataFrame object with count() operation
    performed.
 
RAISES:
    TeradataMLException
    1. EXECUTION_FAILED - If count() operation fails to
        generate the column-wise count of the dataframe.
 
        Possible error message:
        Failed to perform 'count'. (Followed by error message).
 
    2. TDMLDF_AGGREGATE_COMBINED_ERR - If the count() operation
        doesn't support all the columns in the dataframe.
 
        Possible error message:
        No results. Below is/are the error message(s):
        All selected columns [(col2 -  PERIOD_TIME), (col3 -
        BLOB)] is/are unsupported for 'count' operation.
 
EXAMPLES :
    # Load the data to run the example.
    >>> from teradataml.data.load_example_data import load_example_data
    >>> load_example_data("dataframe", ["employee_info"])
 
    # Create teradataml dataframe.
    >>> df1 = DataFrame("employee_info")
    >>> print(df1)
                first_name marks   dob joined_date
    employee_no
    101              abcde  None  None    02/12/05
    100               abcd  None  None        None
    112               None  None  None    18/12/05
    >>>
 
    # Select only subset of columns from the DataFrame.
    >>> df2 = df1.select(['employee_no', 'first_name', 'marks'])
 
    # Prints count of the values in all the selected columns
    # (excluding None types).
    >>> df2.count()
      count_employee_no count_first_name count_marks
    0                 3                2           0
    >>>
 
    #
    # Using count() as Time Series Aggregate.
    #
    >>> # Load the example datasets.
    ... load_example_data("dataframe", ["ocean_buoys_seq"])
    >>>
 
    #
    # Time Series Aggregate Example 1: Executing count() function on DataFrame created on
    #                                  sequenced PTI table. We will consider all rows for the
    #                                  columns while calculating the count.
    #
    >>> ocean_buoys_seq = DataFrame("ocean_buoys_seq")
    >>> ocean_buoys_seq.columns
    ['TD_TIMECODE', 'TD_SEQNO', 'buoyid', 'salinity', 'temperature', 'dates']
    >>> ocean_buoys_seq
                           TD_TIMECODE  TD_SEQNO  salinity  temperature       dates
    buoyid
    44      2014-01-06 10:00:25.122200         6        55           43  2014-06-06
    44      2014-01-06 10:01:25.122200         8        55           53  2014-08-08
    44      2014-01-06 10:01:25.122200        20        55           54  2015-08-20
    1       2014-01-06 09:01:25.122200        11        55           70  2014-11-11
    1       2014-01-06 09:02:25.122200        12        55           71  2014-12-12
    1       2014-01-06 09:02:25.122200        24        55           78  2015-12-24
    1       2014-01-06 09:03:25.122200        13        55           72  2015-01-13
    1       2014-01-06 09:03:25.122200        25        55           79  2016-01-25
    1       2014-01-06 09:01:25.122200        23        55           77  2015-11-23
    44      2014-01-06 10:00:26.122200         7        55           43  2014-07-07
    >>>
 
    # To use count() as Time Series Aggregate we must run groupby_time() first, followed by count().
    >>> ocean_buoys_grpby1 = ocean_buoys_seq.groupby_time(timebucket_duration="2cy", value_expression="buoyid", fill="NULLS")
    >>> ocean_buoys_grpby1.count().sort(["TIMECODE_RANGE", "buoyid"])
                                          TIMECODE_RANGE  GROUP BY TIME(CAL_YEARS(2))  buoyid  count_TD_TIMECODE  count_TD_SEQNO  count_salinity  count_temperature  count_dates
    0  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2       0                  5               5               5                  4            5
    1  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2       1                  6               6               6                  6            6
    2  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2       2                  3               3               3                  3            3
    3  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2      22                  1               1               1                  1            1
    4  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2      44                 13              13              13                 13           13
    >>>
 
    #
    # Time Series Aggregate Example 2: Executing count() function on DataFrame created on
    #                                  sequenced PTI table. We will consider DISTINCT rows for the
    #                                  columns while calculating the count.
    #
    # To use count() as Time Series Aggregate we must run groupby_time() first, followed by count().
    >>> ocean_buoys_grpby1 = ocean_buoys_seq.groupby_time(timebucket_duration="2cy",value_expression="buoyid", fill="NULLS")
    >>> ocean_buoys_grpby1.count(distinct = True).sort(["TIMECODE_RANGE", "buoyid"])
                                          TIMECODE_RANGE  GROUP BY TIME(CAL_YEARS(2))  buoyid  count_TD_TIMECODE  count_TD_SEQNO  count_salinity  count_temperature  count_dates
    0  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2       0                  4               5               1                  3            5
    1  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2       1                  3               6               1                  6            6
    2  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2       2                  3               3               1                  3            3
    3  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2      22                  1               1               1                  1            1
    4  ('2014-01-01 00:00:00.000000-00:00', '2016-01-...                            2      44                 10              13               1                  5           13
    >>>