[ https://issues.apache.org/jira/browse/MADLIB-1413?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17051724#comment-17051724 ]
Frank McQuillan commented on MADLIB-1413: ----------------------------------------- {code} DROP TABLE IF EXISTS abalone_summary ; SELECT madlib.summary ( 'abalone_encoded', -- source_table, 'abalone_summary', -- output_table, NULL, -- target_cols, NULL -- grouping_cols ); {code} produces {code} group_by | group_by_value | target_column | age column_number | 10 data_type | numeric row_count | 4177 distinct_values | 28 missing_values | 0 blank_values | fraction_missing | 0 fraction_blank | positive_values | 4177 negative_values | 0 zero_values | 0 mean | 11.4336844625329 variance | 10.3952659473471 confidence_interval | {11.3359063530521,11.5314625720137} min | 2.5 max | 30.5 first_quartile | 9.5 median | 10.5 third_quartile | 12.5 most_frequent_values | {10.5,11.5,9.5,12.5,8.5,13.5,7.5,14.5,15.5,6.5} mfv_frequencies | {689,634,568,487,391,267,259,203,126,115} {code} LGTM > Last optional param in summary errors when NULL > ----------------------------------------------- > > Key: MADLIB-1413 > URL: https://issues.apache.org/jira/browse/MADLIB-1413 > Project: Apache MADlib > Issue Type: Improvement > Components: Module: Descriptive Statistics > Reporter: Frank McQuillan > Assignee: Orhan Kislal > Priority: Minor > Fix For: v1.17 > > > {code} > DROP TABLE IF EXISTS abalone_summary_exact; > SELECT madlib.summary ( > 'abalone_encoded', -- source_table > 'abalone_summary_exact', -- output_table > NULL, -- target_cols > NULL, -- grouping_cols > TRUE, -- get_distinct > TRUE, -- get_quartiles > NULL, -- quantile_array > 10, -- how_many_mfv > FALSE, -- get_estimate > NULL -- n_cols_per_run > ); > {code} > produces this error > {code} > ERROR: plpy.Error: Summary - Invalid parameter: Number of columns per run > should be positive (plpython.c:5038) > CONTEXT: Traceback (most recent call last): > PL/Python function "summary", line 24, in <module> > get_estimates, n_cols_per_run) > PL/Python function "summary", line 67, in summary > PL/Python function "summary", line 388, in run > PL/Python function "summary", line 105, in _validate_params > PL/Python function "summary", line 117, in _assert > PL/Python function "summary" > {code} > which seems wrong since the last param is optional. > The following does work: > {code} > DROP TABLE IF EXISTS abalone_summary_exact; > SELECT madlib.summary ( > 'abalone_encoded', -- source_table > 'abalone_summary_exact', -- output_table > NULL, -- target_cols > NULL, -- grouping_cols > TRUE, -- get_distinct > TRUE, -- get_quartiles > NULL, -- quantile_array > 10, -- how_many_mfv > FALSE, -- get_estimate > 15 -- n_cols_per_run > ); > {code} > and so does this: > {code} > DROP TABLE IF EXISTS abalone_summary_exact; > SELECT madlib.summary ( > 'abalone_encoded', -- source_table > 'abalone_summary_exact', -- output_table > NULL, -- target_cols > NULL, -- grouping_cols > TRUE, -- get_distinct > TRUE, -- get_quartiles > NULL, -- quantile_array > 10, -- how_many_mfv > FALSE -- get_estimate > ); > {code} -- This message was sent by Atlassian Jira (v8.3.4#803005)