This is an automated email from the ASF dual-hosted git repository.
njayaram pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/madlib.git
The following commit(s) were added to refs/heads/master by this push:
new c2a68c1 DL: Add preprocessor specific error message
c2a68c1 is described below
commit c2a68c1fab2d3a64c3cca66ae928c0f0c4c023aa
Author: Ekta Khanna <[email protected]>
AuthorDate: Mon Jun 24 12:23:52 2019 -0700
DL: Add preprocessor specific error message
JIRA: MADLIB-1364
This commit adds more informative error messages if user forgets to
preprocess source tables.
In addition to the error messages listed out in the above JIRA, this
commit also updates an error message for `validation_preprocessor_dl()`,
along with relevant unit tests.
Closes #416
Co-authored-by: Nandish Jayaram <[email protected]>
---
.../deep_learning/input_data_preprocessor.py_in | 6 +++++-
.../deep_learning/madlib_keras_validator.py_in | 20 ++++++++++++++------
.../test/unit_tests/test_validate_args.py_in | 13 +++++++++++++
.../postgres/modules/utilities/validate_args.py_in | 17 ++++++++++-------
4 files changed, 42 insertions(+), 14 deletions(-)
diff --git
a/src/ports/postgres/modules/deep_learning/input_data_preprocessor.py_in
b/src/ports/postgres/modules/deep_learning/input_data_preprocessor.py_in
index 5cdd638..82bec97 100644
--- a/src/ports/postgres/modules/deep_learning/input_data_preprocessor.py_in
+++ b/src/ports/postgres/modules/deep_learning/input_data_preprocessor.py_in
@@ -370,7 +370,11 @@ class
ValidationDataPreprocessorDL(InputDataPreprocessorDL):
input_tbl_valid(self.training_preprocessor_table, self.module_name)
training_summary_table = add_postfix(
self.training_preprocessor_table, "_summary")
- input_tbl_valid(training_summary_table, self.module_name)
+ input_tbl_valid(training_summary_table, self.module_name,
+ error_suffix_str="Please ensure that table '{0}' "
+ "has been preprocessed using "
+ "training_preprocessor_dl()."
+
.format(self.training_preprocessor_table))
summary_table = plpy.execute("SELECT * FROM {0} LIMIT 1".format(
training_summary_table))[0]
_assert(NORMALIZING_CONST_COLNAME in summary_table,
diff --git
a/src/ports/postgres/modules/deep_learning/madlib_keras_validator.py_in
b/src/ports/postgres/modules/deep_learning/madlib_keras_validator.py_in
index 56b4f2f..b111fc4 100644
--- a/src/ports/postgres/modules/deep_learning/madlib_keras_validator.py_in
+++ b/src/ports/postgres/modules/deep_learning/madlib_keras_validator.py_in
@@ -157,7 +157,10 @@ class EvaluateInputValidator(InputValidator):
output_table, module_name)
def _validate_input_args(self):
- input_tbl_valid(self.test_summary_table, self.module_name)
+ input_tbl_valid(self.test_summary_table, self.module_name,
+ error_suffix_str="Please ensure that the test table
({0}) "
+ "has been preprocessed by "
+ "the image
preprocessor.".format(self.test_table))
self._validate_test_summary_tbl_cols()
InputValidator._validate_input_args(self)
validate_dependent_var_for_minibatch(self.test_table,
@@ -236,16 +239,18 @@ class FitInputValidator:
def _validate_input_table(self, table):
_assert(is_var_valid(table, self.independent_varname),
"{module_name}: invalid independent_varname "
- "('{independent_varname}') for table "
- "({table}).".format(
+ "('{independent_varname}') for table ({table}). "
+ "Please ensure that the input table ({table}) "
+ "has been preprocessed by the image preprocessor.".format(
module_name=self.module_name,
independent_varname=self.independent_varname,
table=table))
_assert(is_var_valid(table, self.dependent_varname),
"{module_name}: invalid dependent_varname "
- "('{dependent_varname}') for table "
- "({table}).".format(
+ "('{dependent_varname}') for table ({table}). "
+ "Please ensure that the input table ({table}) "
+ "has been preprocessed by the image preprocessor.".format(
module_name=self.module_name,
dependent_varname=self.dependent_varname,
table=table))
@@ -262,7 +267,10 @@ class FitInputValidator:
"{0}: metrics_compute_frequency must be in the range (1 -
{1}).".format(
self.module_name, self.num_iterations))
input_tbl_valid(self.source_table, self.module_name)
- input_tbl_valid(self.source_summary_table, self.module_name)
+ input_tbl_valid(self.source_summary_table, self.module_name,
+ error_suffix_str="Please ensure that the source table
({0}) "
+ "has been preprocessed by "
+ "the image
preprocessor.".format(self.source_table))
cols_in_tbl_valid(self.source_summary_table, [CLASS_VALUES_COLNAME,
NORMALIZING_CONST_COLNAME, DEPENDENT_VARTYPE_COLNAME,
'dependent_varname', 'independent_varname'], self.module_name)
diff --git
a/src/ports/postgres/modules/utilities/test/unit_tests/test_validate_args.py_in
b/src/ports/postgres/modules/utilities/test/unit_tests/test_validate_args.py_in
index f94ee03..a3f2539 100644
---
a/src/ports/postgres/modules/utilities/test/unit_tests/test_validate_args.py_in
+++
b/src/ports/postgres/modules/utilities/test/unit_tests/test_validate_args.py_in
@@ -100,5 +100,18 @@ class ValidateArgsTestCase(unittest.TestCase):
self.subject.table_exists = Mock(return_value=False)
self.subject.output_tbl_valid("foo", "unittest_module")
+ def test_input_tbl_valid_table_not_exists_raises_custom_error(self):
+ self.subject.table_exists = Mock(return_value=False)
+ with self.assertRaises(plpy.PLPYException) as error:
+ self.subject.input_tbl_valid("foo", "unittest_module",
+ error_suffix_str="custom exception")
+ self.assertIn('custom exception', str(error.exception))
+
+ def test_input_tbl_valid_table_not_exists_raises_custom_error_none(self):
+ self.subject.table_exists = Mock(return_value=False)
+ with self.assertRaises(plpy.PLPYException) as error:
+ self.subject.input_tbl_valid("foo", "unittest_module")
+ self.assertNotIn('custom exception', str(error.exception))
+
if __name__ == '__main__':
unittest.main()
diff --git a/src/ports/postgres/modules/utilities/validate_args.py_in
b/src/ports/postgres/modules/utilities/validate_args.py_in
index 41e9a11..e0758d3 100644
--- a/src/ports/postgres/modules/utilities/validate_args.py_in
+++ b/src/ports/postgres/modules/utilities/validate_args.py_in
@@ -662,18 +662,21 @@ def is_var_valid(tbl, var, order_by=None):
# -------------------------------------------------------------------------
-def input_tbl_valid(tbl, module, check_empty=True):
+def input_tbl_valid(tbl, module, check_empty=True, error_suffix_str=None):
+ # Add a space in the beginning here instead of at the error message for
+ # backward compatibility.
+ error_suffix_str = ' {0}'.format(error_suffix_str) if error_suffix_str
else ''
if tbl is None or tbl.strip() == '':
- plpy.error(
- "{module} error: NULL/empty input table name!".format(**locals()))
+ plpy.error("{0} error: NULL/empty input table name!{1}".format(
+ module, error_suffix_str))
if not table_exists(tbl):
- plpy.error(
- "{module} error: Input table '{tbl}' does not
exist".format(**locals()))
+ plpy.error("{0} error: Input table '{1}' does not exist.{2}".format(
+ module, tbl, error_suffix_str))
if check_empty and table_is_empty(tbl):
- plpy.error(
- "{module} error: Input table '{tbl}' is empty!".format(**locals()))
+ plpy.error("{0} error: Input table '{1}' is empty!{2}".format(
+ module, tbl, error_suffix_str))
# -------------------------------------------------------------------------
