Reduce chance of disconnected components in random graph test #1672
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Code Climate has analyzed commit aca5990 and detected 0 issues on this pull request. View more on Code Climate. |
kjun9
approved these changes
Jun 12, 2020
huonw
added a commit
that referenced
this pull request
Jun 12, 2020
This adds an xfail marker `pytest.mark.xfail` https://docs.pytest.org/en/5.4.3/skipping.html#xfail-mark-test-functions-as-expected-to-fail to several tests that are flaky. These tests fail semi-regularly on CI requiring a lot of retried builds, and, for the most part, don't catch much. The mark is added using a function that requires that several things are specified: - issue numbers which are linked in the `reason`; which makes it easier to keep track of why/how things are flaky - an exception class which is the form of the xfail, so that the test still catches basic problems like invalid code (e.g. renaming functions or properties) Issues for flaky tests (found by https://github.com/stellargraph/stellargraph/issues?q=is%3Aissue+is%3Aopen+flaky): - #585: `tests/data/test_edge_splitter.py`: `test_split_data_by_edge_type_and_attribute`, `test_split_data_by_edge_type` - #970: `tests/reproducibility/test_graphsage.py`: `test_link_prediction[True]` - #990: `tests/reproducibility/test_graphsage.py`: `test_link_prediction[False]` - #1115: `tests/reproducibility/test_graphsage.py`: `test_unsupervised[False]`, `test_unsupervised[True]`, `test_nai[True]`, `test_nai[False]` - #1160: `tests/core/test_utils.py`:`test_normalized_laplacian` - #1623: `tests/layer/test_knowledge_graph.py`: `test_rotate`, `test_model_rankings[RotatE]` - #1675: `tests/layer/test_knowledge_graph.py`: `test_model_rankings[RotH]` This PR doesn't include #1569 (`tests/mapper/test_node_mappers.py`: `test_nodemapper_isolated_nodes`), because that's fixed, not xfailed, independently in #1672.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This
test_nodemapper_isolated_nodestest creates a graph with 5 nodes (0, 1, 2, 3, 4) being connected by 20 edges and a single isolated node (5). The test checks that the graph has 2 connected components, that is, the first 5 nodes are connected and the last node is separate. With only 20 edges, this fails very occasionally, because the randomness will sometimes miss a node or have separate clumps.This PR ups the number of edges to 1000 to exponentially reduce the chance of the graph having more than 2 connected components.
See: #1569