ray/release/xgboost_tests
Balaji Veeramani 7f1bacc7dc
[CI] Format Python code with Black (#21975)
See #21316 and #21311 for the motivation behind these changes.
2022-01-29 18:41:57 -08:00
..
workloads [CI] Format Python code with Black (#21975) 2022-01-29 18:41:57 -08:00
app_config.yaml [release/xgboost] xgboost release test fixes via app config (#20325) 2021-11-15 10:03:21 -08:00
app_config_gpu.yaml [release/xgboost] xgboost release test fixes via app config (#20325) 2021-11-15 10:03:21 -08:00
create_test_data.py [CI] Format Python code with Black (#21975) 2022-01-29 18:41:57 -08:00
README.rst [xgboost] Update XGBoost release test configs (#13941) 2021-02-17 23:00:49 +01:00
tpl_cpu_moderate.yaml [release] Move xgboost tune small + microbenchmark release test to new release automation (#15619) 2021-05-08 20:38:39 +01:00
tpl_cpu_small.yaml [release] Move xgboost tune small + microbenchmark release test to new release automation (#15619) 2021-05-08 20:38:39 +01:00
tpl_gpu_small.yaml [release] Move xgboost tune small + microbenchmark release test to new release automation (#15619) 2021-05-08 20:38:39 +01:00
wait_cluster.py [CI] Format Python code with Black (#21975) 2022-01-29 18:41:57 -08:00
xgboost_tests.yaml [Test Infra] Unrevert team col (#21700) 2022-01-19 13:29:53 -08:00

XGBoost on Ray tests
====================

This directory contains various XGBoost on Ray release tests.

You should run these tests with the `releaser <https://github.com/ray-project/releaser>`_ tool.

Overview
--------
There are four kinds of tests:

1. ``distributed_api_test`` - checks general API functionality and should finish very quickly (< 1 minute)
2. ``train_*`` - checks single trial training on different setups.
3. ``tune_*`` - checks multi trial training via Ray Tune.
4. ``ft_*`` - checks fault tolerance.

Generally the releaser tool will run all tests in parallel, but if you do
it sequentially, be sure to do it in the order above. If ``train_*`` fails,
``tune_*`` will fail, too.

Acceptance criteria
-------------------
These tests are considered passing when they throw no error at the end of
the output log.