Kai Fricke 6c5229295e [ci/release] Support running tests with different python versions (#24843) 2 年之前
..
learning_tests fb2915d26a [RLlib] Replay Buffer API and Ape-X. (#24506) 2 年之前
multi_gpu_learning_tests fb2915d26a [RLlib] Replay Buffer API and Ape-X. (#24506) 2 年之前
multi_gpu_with_attention_learning_tests 7f1bacc7dc [CI] Format Python code with Black (#21975) 2 年之前
multi_gpu_with_lstm_learning_tests 7f1bacc7dc [CI] Format Python code with Black (#21975) 2 年之前
stress_tests 7f1bacc7dc [CI] Format Python code with Black (#21975) 2 年之前
unit_gpu_tests 65d9a410f7 [ci] Clean up ci/ directory (refactor ci/travis) (#23866) 2 年之前
12gpus_192cpus.yaml d8a61f801f [RLlib] Create a set of performance benchmark tests to run nightly. (#19945) 2 年之前
1gpu_4cpus.yaml 2c1fa459d4 [RLlib] Add an RLlib Tune experiment to UserTest suite. (#19807) 3 年之前
2gpus_32cpus.yaml 2c1fa459d4 [RLlib] Add an RLlib Tune experiment to UserTest suite. (#19807) 3 年之前
4gpus_544_cpus.yaml ab15dfd478 [Tune release test] Set 500G disk space for rllib_tests. (#19730) 3 年之前
4gpus_64cpus.yaml 2c1fa459d4 [RLlib] Add an RLlib Tune experiment to UserTest suite. (#19807) 3 年之前
8gpus_64cpus.yaml ab15dfd478 [Tune release test] Set 500G disk space for rllib_tests. (#19730) 3 年之前
8gpus_96cpus.yaml ab15dfd478 [Tune release test] Set 500G disk space for rllib_tests. (#19730) 3 年之前
README.rst c9d220bcda [RLlib] Upgrade RLlib regression test scripts to new testing tool - RLlib release logs for 1.4. (#16080) 3 年之前
app_config.yaml 6c5229295e [ci/release] Support running tests with different python versions (#24843) 2 年之前

README.rst

RLlib Tests
===========

This directory contains various RLlib release tests.

You should run these tests with the `releaser `_ tool.

Overview
--------
Currently, there are 3 RLlib tests:

1. ``learning_tests`` - Tests, whether major algos (tf+torch) can learn in Atari or PyBullet envs in ~30-60min.
1. ``stress_tests`` - Runs 4 IMPALA Atari jobs, each one using 1GPU and 128CPUs (needs autoscaling to succeed).
1. ``unit_gpu_tests`` - Tests, whether all of RLlib's example scripts can be run on a GPU.

Generally the releaser tool will run all tests in parallel.

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