pytorch / pytorch/rl

[Flaky Tests] 5696 flaky test(s) detected

Open
#3,865 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

CI flaky-test-tracker
Dominant language
Python
Stars
3.6k
Forks
487
Avg merge
23h 50m
Merged PRs (30d)
209

Description

Flaky Test Report - 2026-09-09

Summary:

  • Flaky tests: 72
  • Newly flaky (last 7 days): 72
  • Resolved: 0

Flaky Test Report - 2026-09-09

Summary

  • Flaky tests: 72
  • Newly flaky (last 7 days): 72
  • Resolved: 0
  • Total tests analyzed: 31826
  • CI runs analyzed: 60

Flaky Tests

Test Failure Rate Failures Flaky Score Last Failed
test/objectives/test_dt.py::TestOnlineDT::test_odt[device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...bjectives/test_dt.py::TestOnlineDT::test_seq_odt[device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
test/objectives/test_dt.py::TestDT::test_dt[device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
test/objectives/test_dt.py::TestDT::test_seq_dt[device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...ectives/test_dt.py::TestGAIL::test_gail[0.1-True-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...ctives/test_dt.py::TestGAIL::test_gail[0.1-False-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...ectives/test_dt.py::TestGAIL::test_gail[1.0-True-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...ctives/test_dt.py::TestGAIL::test_gail[1.0-False-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...ves/test_dt.py::TestGAIL::test_seq_gail[0.1-True-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...es/test_dt.py::TestGAIL::test_seq_gail[0.1-False-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...ves/test_dt.py::TestGAIL::test_seq_gail[1.0-True-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...es/test_dt.py::TestGAIL::test_seq_gail[1.0-False-device1] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...omposite::test_device_cast[dest1-shape0-dtype0-None-True] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...mposite::test_device_cast[dest1-shape0-dtype0-None-False] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...osite::test_device_cast[dest1-shape0-dtype0-device1-True] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...site::test_device_cast[dest1-shape0-dtype0-device1-False] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...omposite::test_device_cast[dest1-shape0-dtype1-None-True] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...mposite::test_device_cast[dest1-shape0-dtype1-None-False] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...osite::test_device_cast[dest1-shape0-dtype1-device1-True] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
...site::test_device_cast[dest1-shape0-dtype1-device1-False] πŸ†• 40.0% (6/15) 6 0.80 2026-09-08
Newly Flaky Tests
  • test/objectives/test_dt.py::TestOnlineDT::test_odt[device1]
  • test/objectives/test_dt.py::TestOnlineDT::test_seq_odt[device1]
  • test/objectives/test_dt.py::TestDT::test_dt[device1]
  • test/objectives/test_dt.py::TestDT::test_seq_dt[device1]
  • test/objectives/test_dt.py::TestGAIL::test_gail[0.1-True-device1]
  • test/objectives/test_dt.py::TestGAIL::test_gail[0.1-False-device1]
  • test/objectives/test_dt.py::TestGAIL::test_gail[1.0-True-device1]
  • test/objectives/test_dt.py::TestGAIL::test_gail[1.0-False-device1]
  • test/objectives/test_dt.py::TestGAIL::test_seq_gail[0.1-True-device1]
  • test/objectives/test_dt.py::TestGAIL::test_seq_gail[0.1-False-device1]
  • test/objectives/test_dt.py::TestGAIL::test_seq_gail[1.0-True-device1]
  • test/objectives/test_dt.py::TestGAIL::test_seq_gail[1.0-False-device1]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype0-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype0-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype0-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype0-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype1-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype1-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype1-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype1-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype2-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype2-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype2-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-dtype2-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-None-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-None-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-None-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape0-None-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype0-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype0-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype0-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype0-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype1-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype1-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype1-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype1-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype2-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype2-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype2-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-dtype2-device1-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-None-None-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-None-None-False]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-None-device1-True]
  • test/test_specs.py::TestComposite::test_device_cast[dest1-shape1-None-device1-False]
  • test/objectives/test_dreamer_v3.py::test_dreamer_v3_native_replay_collection_smoke[cuda-async-True-True-True-frames]
  • test/objectives/test_dreamer_v3.py::test_dreamer_v3_native_replay_collection_smoke[cuda-sync-True-True-True-frames]
  • test/objectives/test_dreamer_v3.py::test_dreamer_v3_native_replay_collection_smoke[cuda-async-True-False-False-time]
  • test/objectives/test_dreamer_v3.py::test_dreamer_v3_native_replay_collection_smoke[cuda-async-True-False-False-warmup]
  • test/objectives/test_dreamer_v3.py::test_dreamer_v3_native_replay_collection_smoke[cuda-sync-True-False-False-reset_records]
  • test/modules/test_dreamer_components.py::TestDreamerV3Components::test_state_estimator_resets_and_posterior_rng[compile]
  • test/modules/test_dreamer_components.py::TestDreamerV3Components::test_discrete_actor[cpu-compile]
  • test/rb/test_prioritized.py::test_cuda_prioritized_replay_buffer_samples_on_cuda
  • test/rb/test_prioritized.py::test_tensordict_prioritized_replay_buffer_memmap_storage_cuda_sampler
  • test/rb/test_prioritized.py::test_cuda_prioritized_replay_buffer_weight_matches_cpu_formula
  • test/rb/test_rb_core.py::TestSequenceUnit::test_non_cpu_storage
  • test/rb/test_storages.py::TestStorages::test_storage_device[tensor-LazyMemmapStorage-device_data0-device_storage0]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tensor-LazyMemmapStorage-device_data3-auto]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tensor-LazyTensorStorage-device_data0-device_storage0]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tensor-LazyTensorStorage-device_data3-auto]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tc-LazyMemmapStorage-device_data0-device_storage0]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tc-LazyMemmapStorage-device_data3-auto]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tc-LazyTensorStorage-device_data0-device_storage0]
  • test/rb/test_storages.py::TestStorages::test_storage_device[tc-LazyTensorStorage-device_data3-auto]
  • test/rb/test_storages.py::TestStorages::test_storage_device[td-LazyMemmapStorage-device_data0-device_storage0]
  • test/rb/test_storages.py::TestStorages::test_storage_device[td-LazyMemmapStorage-device_data3-auto]
  • test/rb/test_storages.py::TestStorages::test_storage_device[td-LazyTensorStorage-device_data0-device_storage0]
  • test/rb/test_storages.py::TestStorages::test_storage_device[td-LazyTensorStorage-device_data3-auto]
  • test/rb/test_storages.py::TestSharedStorageInit::test_prioritized_memmap_cuda_sampler_after_multiprocess_writes
  • test/rb/test_writers.py::TestWriterGeneration::test_generation_cuda_data_into_cuda_storage
  • test/test_checkpoint.py::test_cuda_map_location_and_rng
  • test/test_custom_envs.py::TestCustomEnvs::test_financial_env_device
  • test/objectives/test_dreamer_v3.py::TestDreamerV3CompileStrategy::test_compiled_step_selects_the_scan_for_untouched_rollouts

Configuration

  • Minimum failure rate: 5%
  • Maximum failure rate: 95%
  • Minimum failures required: 2
  • Minimum executions required: 3

Generated at 2026-09-09T06:28:00.198636+00:00

This issue is automatically updated daily. See the full dashboard for more details.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up β€” it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the full flaky-test dashboard and reproduce the reported tests in test/objectives/test_dt.py and test/test_specs.py, then work through the other listed files such as test/objectives/test_dreamer_v3.py and test/rb/. Done means the reported flaky tests have diagnosed failure modes and no longer appear as flaky in subsequent reports.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
testing-qa
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Active
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.