jingyimei commented on a change in pull request #382: DL: Naming improvements
for dependent and independent varname
URL: https://github.com/apache/madlib/pull/382#discussion_r280642965
##########
File path: src/ports/postgres/modules/deep_learning/madlib_keras.py_in
##########
@@ -209,6 +211,10 @@ def fit(schema_madlib, source_table, model,
dependent_varname,
fit_validator.source_summary_table, NORMALIZING_CONST_COLNAME)
dep_vartype = plpy.execute("SELECT {0} AS dep FROM {1}".format(
DEPENDENT_VARTYPE_COLNAME,
fit_validator.source_summary_table))[0]['dep']
+ dependent_varname_in_source_table = plpy.execute("SELECT {0} FROM
{1}".format(
+ 'dependent_varname',
fit_validator.source_summary_table))[0]['dependent_varname']
+ independent_varname_in_source_table = plpy.execute("SELECT {0} FROM
{1}".format(
+ 'independent_varname',
fit_validator.source_summary_table))[0]['independent_varname']
Review comment:
I did look at here and try to make them variable. The problem here is in
`madlib_keras_helper.py_in`
https://github.com/apache/madlib/blob/master/src/ports/postgres/modules/deep_learning/madlib_keras_helper.py_in#L30,
we had a comment saying those are columns in model summary table, while here
it is actually from source summary table (summary table from minibatch
preprocessor). We already use those variables in a confusing way. Besides, we
don't have a global variable system for minibatch summary table(see
https://github.com/apache/madlib/blob/master/src/ports/postgres/modules/utilities/minibatch_preprocessing.py_in#L495),
and we use this kind of hard quote in many other places(example
https://github.com/apache/madlib/blob/master/src/ports/postgres/modules/deep_learning/madlib_keras.py_in#L211).
If we want to resolve this thoroughly, we need to build similar system to
minibatch summary table and remove those hard quote.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
