Skip to content

Measuring runtimes #1031

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mfeurer merged 19 commits into from
Mar 12, 2021
Merged

Measuring runtimes #1031

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
19 commits
Select commit Hold shift + click to select a range
c3bc8a5
[skip ci] addressing #248
Neeratyoy Feb 23, 2021
bfacd7f
Unit test to test existence of refit time
Neeratyoy Feb 24, 2021
b5d0190
Refactoring unit test
Neeratyoy Feb 25, 2021
ee66338
Fixing unit test failures
Neeratyoy Mar 2, 2021
909d560
Unit test fixing + removing redundant parameter
Neeratyoy Mar 4, 2021
d246fa4
Debugging stochastic failure of test_joblib_backends unit test
Neeratyoy Mar 4, 2021
b8f906f
Unit test fix with decorators
Neeratyoy Mar 8, 2021
1e279ea
Flaky for failing unit test
Neeratyoy Mar 8, 2021
cd6053f
Adding flaky reruns for unit tests
Neeratyoy Mar 8, 2021
9b4ccbe
Fixing setup big
Neeratyoy Mar 9, 2021
f74f3dc
pytest rerun debug
Neeratyoy Mar 9, 2021
a767cea
Fixing coverage failure
Neeratyoy Mar 9, 2021
f8fd620
Debugging coverage failure
Neeratyoy Mar 9, 2021
e8cf215
Debugging coverage failure
Neeratyoy Mar 9, 2021
0991c4a
Adding __init__ files in test/ for pytest-cov
Neeratyoy Mar 9, 2021
bde4cc9
Debugging coverage failure
Neeratyoy Mar 10, 2021
8d15c7d
Debugging lean unit test
Neeratyoy Mar 12, 2021
0814a64
Debugging loky failure in unit tests
Neeratyoy Mar 12, 2021
11d1677
Clean up of debugging stuff
Neeratyoy Mar 12, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Unit test fixing + removing redundant parameter
  • Loading branch information
@Neeratyoy
Neeratyoy committed Mar 4, 2021
commit 909d560fe82fe3b48b750bf8135ae0f8aae74c66
17 changes: 7 additions & 10 deletions openml/runs/functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -271,7 +271,6 @@ def run_flow_on_task(

# execute the run
res = _run_task_get_arffcontent(
flow=flow,
model=flow.model,
task=task,
extension=flow.extension,
Expand Down Expand Up @@ -432,7 +431,6 @@ def run_exists(task_id: int, setup_id: int) -> Set[int]:


def _run_task_get_arffcontent(
flow: OpenMLFlow,
model: Any,
task: OpenMLTask,
extension: "Extension",
Expand Down Expand Up @@ -476,7 +474,6 @@ def _run_task_get_arffcontent(
job_rvals = Parallel(verbose=0, n_jobs=n_jobs)(
delayed(_run_task_get_arffcontent_parallel_helper)(
extension=extension,
flow=flow,
fold_no=fold_no,
model=model,
rep_no=rep_no,
Expand Down Expand Up @@ -613,7 +610,6 @@ def _calculate_local_measure(sklearn_fn, openml_name):

def _run_task_get_arffcontent_parallel_helper(
extension: "Extension",
flow: OpenMLFlow,
fold_no: int,
model: Any,
rep_no: int,
Expand Down Expand Up @@ -661,12 +657,13 @@ def _run_task_get_arffcontent_parallel_helper(
else:
raise NotImplementedError(task.task_type)
config.logger.info(
"Going to execute flow '%s' on task %d for repeat %d fold %d sample %d.",
str(model),
task.task_id,
rep_no,
fold_no,
sample_no,
"Going to run model {} on dataset {} for repeat {} fold {} sample {}".format(
str(model),
openml.datasets.get_dataset(task.dataset_id).name,
rep_no,
fold_no,
sample_no,
)
)
pred_y, proba_y, user_defined_measures_fold, trace, = extension._run_model_on_fold(
model=model,
Expand Down
13 changes: 2 additions & 11 deletions tests/test_runs/test_run_functions.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@
import time
import sys
import ast
import pytest
import unittest.mock

import numpy as np
Expand Down Expand Up @@ -1188,13 +1189,10 @@ def test__run_task_get_arffcontent(self):
num_folds = 10
num_repeats = 1

flow = unittest.mock.Mock()
flow.name = "dummy"
clf = make_pipeline(
OneHotEncoder(handle_unknown="ignore"), SGDClassifier(loss="log", random_state=1)
)
res = openml.runs.functions._run_task_get_arffcontent(
flow=flow,
extension=self.extension,
model=clf,
task=task,
Expand Down Expand Up @@ -1405,8 +1403,6 @@ def test_run_on_dataset_with_missing_labels_dataframe(self):
# Check that _run_task_get_arffcontent works when one of the class
# labels only declared in the arff file, but is not present in the
# actual data
flow = unittest.mock.Mock()
flow.name = "dummy"
task = openml.tasks.get_task(2) # anneal; crossvalidation

from sklearn.compose import ColumnTransformer
Expand All @@ -1421,7 +1417,6 @@ def test_run_on_dataset_with_missing_labels_dataframe(self):
) # build a sklearn classifier

data_content, _, _, _ = _run_task_get_arffcontent(
flow=flow,
model=model,
task=task,
extension=self.extension,
Expand All @@ -1443,8 +1438,6 @@ def test_run_on_dataset_with_missing_labels_array(self):
# Check that _run_task_get_arffcontent works when one of the class
# labels only declared in the arff file, but is not present in the
# actual data
flow = unittest.mock.Mock()
flow.name = "dummy"
task = openml.tasks.get_task(2) # anneal; crossvalidation
# task_id=2 on test server has 38 columns with 6 numeric columns
cont_idx = [3, 4, 8, 32, 33, 34]
Expand All @@ -1466,7 +1459,6 @@ def test_run_on_dataset_with_missing_labels_array(self):
) # build a sklearn classifier

data_content, _, _, _ = _run_task_get_arffcontent(
flow=flow,
model=model,
task=task,
extension=self.extension,
Expand Down Expand Up @@ -1594,7 +1586,6 @@ def test__run_task_get_arffcontent_2(self, parallel_mock):
backend = "loky" if LooseVersion(joblib.__version__) > "0.11" else "multiprocessing"
with parallel_backend(backend, n_jobs=n_jobs):
res = openml.runs.functions._run_task_get_arffcontent(
flow=None,
extension=self.extension,
model=clf,
task=task,
Expand Down Expand Up @@ -1630,6 +1621,7 @@ def test__run_task_get_arffcontent_2(self, parallel_mock):
scores = [v for k, v in res[2]["predictive_accuracy"][0].items()]
self.assertSequenceEqual(scores, expected_scores, seq_type=list)

@pytest.mark.flaky() # appears to fail stochastically on test server
@unittest.skipIf(
LooseVersion(sklearn.__version__) < "0.21",
reason="couldn't perform local tests successfully w/o bloating RAM",
Expand Down Expand Up @@ -1670,7 +1662,6 @@ def test_joblib_backends(self, parallel_mock):
)
with parallel_backend(backend, n_jobs=n_jobs):
res = openml.runs.functions._run_task_get_arffcontent(
flow=None,
extension=self.extension,
model=clf,
task=task,
Expand Down