[
https://issues.apache.org/jira/browse/CLIMATE-657?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14693819#comment-14693819
]
ASF GitHub Bot commented on CLIMATE-657:
----------------------------------------
Github user MJJoyce commented on a diff in the pull request:
https://github.com/apache/climate/pull/221#discussion_r36883786
--- Diff: ocw/tests/test_metrics.py ---
@@ -208,41 +229,6 @@ def test_function_run(self):
expected_result.fill(-300)
np.testing.assert_array_equal(self.mean_bias.run(self.target_dataset,self.reference_dataset),
expected_result)
- def test_function_run_abs(self):
- '''Test mean bias function between reference dataset and target
dataset with abs as True.'''
- expected_result = np.zeros((5, 5), dtype=np.int)
- expected_result.fill(300)
-
np.testing.assert_array_equal(self.mean_bias.run(self.reference_dataset,
self.target_dataset, True), expected_result)
-
-
-class TestSpatialMeanOfTemporalMeanBias(unittest.TestCase):
--- End diff --
Hey @huikyole, so if we don't need the test then we don't need the metric
that it's testing I gather? This sounds like something that should be under
another ticket so it can be discussed outside of the scope of this ticket which
seems to me to be changing the call structure of metrics so they use isolated
functions that take arrays instead of OCW dataset objects. Thoughts?
> Adding functions to calculate metrics
> -------------------------------------
>
> Key: CLIMATE-657
> URL: https://issues.apache.org/jira/browse/CLIMATE-657
> Project: Apache Open Climate Workbench
> Issue Type: Improvement
> Components: metrics
> Affects Versions: 1.0.0
> Reporter: Huikyo Lee
> Assignee: Huikyo Lee
>
> Currently, all the metrics defined as Class can only accept OCW dataset
> objects as input. Functions calculating metrics with input arrays will be
> added.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
