rl
rl copied to clipboard
[Example] Comprehensive dataset rendering examples
:link: Helpful Links
:test_tube: See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2141
- :page_facing_up: Preview Python docs built from this PR
Note: Links to docs will display an error until the docs builds have been completed.
:x: 8 New Failures, 2 Unrelated Failures
As of commit 739eac6c01563886b5fc48a960d85fe8b2a5dda4 with merge base f613eefb0f44b6a9ccd79f82703bb309aa165009 ():
NEW FAILURES - The following jobs have failed:
- Continuous Benchmark (PR) / CPU Pytest benchmark (gh)
Process completed with exit code 4. - Examples Tests on Linux / tests (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t d47edf0cd5a4e9b63c1f63ad6b95fc831e289f134eb4c501e79def9bae64acb3 /exec failed with exit code 1 - Habitat Tests on Linux / tests (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t 488abbc22adc2ad46a290102eb07f12ede71815df9f1e3e68c4ce0ff5a1e22e6 /exec failed with exit code 1 - Libs Tests on Linux / unittests-sklearn (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t ace67acce4592ebab5151a232e4732ef4f469fe67a56b80e1062aa870271ba70 /exec failed with exit code 1 - Lint / python-source-and-configs / linux-job (gh)
torchrl/envs/libs/jumanji.py:12:1: F401 'torchvision.transforms.v2.functional' imported but unused - RLHF Tests on Linux / unittests (3.9, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t f73dd8f0d5b235d032150c69c75a53ea6ed99bc5f5b7cce989c61f067f7d05f7 /exec failed with exit code 1 - Unit-tests on Linux / tests-optdeps (3.10, 12.1) / linux-job (gh)
RuntimeError: Command docker exec -t 467ae0239552ce7ea887fde2899c9061f6a515a7f6e6380bde16362028e2d59f /exec failed with exit code 1 - Unit-tests on Windows / unittests-cpu / windows-job (gh)
ModuleNotFoundError: No module named 'torchvision'
BROKEN TRUNK - The following jobs failed but were present on the merge base:
👉 Rebase onto the `viable/strict` branch to avoid these failures
- Libs Tests on Linux / unittests-gym (3.9, 12.1) / linux-job (gh) (trunk failure)
##[error]The operation was canceled. - Unit-tests on Linux / tests-olddeps (3.8, 11.6) / linux-job (gh) (trunk failure)
##[error]The operation was canceled.
This comment was automatically generated by Dr. CI and updates every 15 minutes.
$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests
Total Benchmarks: 94. Improved: $\large\color{#35bf28}16$. Worsened: $\large\color{#d91a1a}4$.
Expand to view detailed results
| Name | Max | Mean | Ops | Ops on Repo HEAD |
Change |
|---|---|---|---|---|---|
| test_single | 0.1170s | 0.1159s | 8.6246 Ops/s | 8.5174 Ops/s | $\color{#35bf28}+1.26\%$ |
| test_sync | 99.0960ms | 97.5079ms | 10.2556 Ops/s | 9.7162 Ops/s | $\textbf{\color{#35bf28}+5.55\%}$ |
| test_async | 0.2709s | 97.4224ms | 10.2646 Ops/s | 10.3690 Ops/s | $\color{#d91a1a}-1.01\%$ |
| test_single_pixels | 0.1257s | 0.1256s | 7.9642 Ops/s | 7.7250 Ops/s | $\color{#35bf28}+3.10\%$ |
| test_sync_pixels | 83.4182ms | 79.0104ms | 12.6566 Ops/s | 11.8392 Ops/s | $\textbf{\color{#35bf28}+6.90\%}$ |
| test_async_pixels | 0.1594s | 79.2651ms | 12.6159 Ops/s | 14.4170 Ops/s | $\textbf{\color{#d91a1a}-12.49\%}$ |
| test_simple | 0.7894s | 0.7887s | 1.2679 Ops/s | 1.2315 Ops/s | $\color{#35bf28}+2.95\%$ |
| test_transformed | 1.1345s | 1.0638s | 0.9400 Ops/s | 0.9354 Ops/s | $\color{#35bf28}+0.49\%$ |
| test_serial | 2.5046s | 2.4335s | 0.4109 Ops/s | 0.4078 Ops/s | $\color{#35bf28}+0.78\%$ |
| test_parallel | 2.4176s | 2.3460s | 0.4263 Ops/s | 0.4252 Ops/s | $\color{#35bf28}+0.25\%$ |
| test_step_mdp_speed[True-True-True-True-True] | 0.1064ms | 32.4380μs | 30.8280 KOps/s | 29.7107 KOps/s | $\color{#35bf28}+3.76\%$ |
| test_step_mdp_speed[True-True-True-True-False] | 0.1237ms | 19.5323μs | 51.1972 KOps/s | 49.3577 KOps/s | $\color{#35bf28}+3.73\%$ |
| test_step_mdp_speed[True-True-True-False-True] | 90.6810μs | 18.7669μs | 53.2852 KOps/s | 50.7808 KOps/s | $\color{#35bf28}+4.93\%$ |
| test_step_mdp_speed[True-True-True-False-False] | 30.1100μs | 11.2464μs | 88.9174 KOps/s | 87.5264 KOps/s | $\color{#35bf28}+1.59\%$ |
| test_step_mdp_speed[True-True-False-True-True] | 59.5600μs | 34.7016μs | 28.8172 KOps/s | 27.7614 KOps/s | $\color{#35bf28}+3.80\%$ |
| test_step_mdp_speed[True-True-False-True-False] | 46.6610μs | 21.4262μs | 46.6719 KOps/s | 44.6534 KOps/s | $\color{#35bf28}+4.52\%$ |
| test_step_mdp_speed[True-True-False-False-True] | 50.6210μs | 20.9149μs | 47.8128 KOps/s | 47.6979 KOps/s | $\color{#35bf28}+0.24\%$ |
| test_step_mdp_speed[True-True-False-False-False] | 32.7100μs | 13.1431μs | 76.0856 KOps/s | 75.9784 KOps/s | $\color{#35bf28}+0.14\%$ |
| test_step_mdp_speed[True-False-True-True-True] | 61.0900μs | 36.4040μs | 27.4695 KOps/s | 26.2877 KOps/s | $\color{#35bf28}+4.50\%$ |
| test_step_mdp_speed[True-False-True-True-False] | 51.5710μs | 23.5960μs | 42.3801 KOps/s | 40.6707 KOps/s | $\color{#35bf28}+4.20\%$ |
| test_step_mdp_speed[True-False-True-False-True] | 41.5900μs | 20.3799μs | 49.0680 KOps/s | 46.6760 KOps/s | $\textbf{\color{#35bf28}+5.12\%}$ |
| test_step_mdp_speed[True-False-True-False-False] | 33.9200μs | 13.1435μs | 76.0832 KOps/s | 75.1566 KOps/s | $\color{#35bf28}+1.23\%$ |
| test_step_mdp_speed[True-False-False-True-True] | 62.6510μs | 37.5701μs | 26.6169 KOps/s | 25.2507 KOps/s | $\textbf{\color{#35bf28}+5.41\%}$ |
| test_step_mdp_speed[True-False-False-True-False] | 41.5700μs | 24.7918μs | 40.3359 KOps/s | 38.1581 KOps/s | $\textbf{\color{#35bf28}+5.71\%}$ |
| test_step_mdp_speed[True-False-False-False-True] | 42.7910μs | 21.7247μs | 46.0306 KOps/s | 43.7932 KOps/s | $\textbf{\color{#35bf28}+5.11\%}$ |
| test_step_mdp_speed[True-False-False-False-False] | 0.1069ms | 14.6664μs | 68.1830 KOps/s | 65.1782 KOps/s | $\color{#35bf28}+4.61\%$ |
| test_step_mdp_speed[False-True-True-True-True] | 58.2410μs | 35.8097μs | 27.9254 KOps/s | 26.5517 KOps/s | $\textbf{\color{#35bf28}+5.17\%}$ |
| test_step_mdp_speed[False-True-True-True-False] | 84.7200μs | 22.7840μs | 43.8904 KOps/s | 41.1994 KOps/s | $\textbf{\color{#35bf28}+6.53\%}$ |
| test_step_mdp_speed[False-True-True-False-True] | 0.1282ms | 24.2608μs | 41.2187 KOps/s | 39.3858 KOps/s | $\color{#35bf28}+4.65\%$ |
| test_step_mdp_speed[False-True-True-False-False] | 41.5710μs | 14.7468μs | 67.8114 KOps/s | 65.4272 KOps/s | $\color{#35bf28}+3.64\%$ |
| test_step_mdp_speed[False-True-False-True-True] | 56.4900μs | 37.8820μs | 26.3978 KOps/s | 25.5729 KOps/s | $\color{#35bf28}+3.23\%$ |
| test_step_mdp_speed[False-True-False-True-False] | 47.5800μs | 25.3483μs | 39.4503 KOps/s | 37.6466 KOps/s | $\color{#35bf28}+4.79\%$ |
| test_step_mdp_speed[False-True-False-False-True] | 66.8810μs | 26.1663μs | 38.2171 KOps/s | 36.3780 KOps/s | $\textbf{\color{#35bf28}+5.06\%}$ |
| test_step_mdp_speed[False-True-False-False-False] | 31.2300μs | 16.2982μs | 61.3564 KOps/s | 58.0893 KOps/s | $\textbf{\color{#35bf28}+5.62\%}$ |
| test_step_mdp_speed[False-False-True-True-True] | 56.5400μs | 39.3035μs | 25.4430 KOps/s | 24.1983 KOps/s | $\textbf{\color{#35bf28}+5.14\%}$ |
| test_step_mdp_speed[False-False-True-True-False] | 43.7300μs | 26.4066μs | 37.8693 KOps/s | 35.4968 KOps/s | $\textbf{\color{#35bf28}+6.68\%}$ |
| test_step_mdp_speed[False-False-True-False-True] | 56.4110μs | 26.0817μs | 38.3411 KOps/s | 36.3104 KOps/s | $\textbf{\color{#35bf28}+5.59\%}$ |
| test_step_mdp_speed[False-False-True-False-False] | 36.4810μs | 16.5882μs | 60.2838 KOps/s | 57.9273 KOps/s | $\color{#35bf28}+4.07\%$ |
| test_step_mdp_speed[False-False-False-True-True] | 0.1825ms | 42.7206μs | 23.4079 KOps/s | 23.3529 KOps/s | $\color{#35bf28}+0.24\%$ |
| test_step_mdp_speed[False-False-False-True-False] | 0.1217ms | 28.8071μs | 34.7136 KOps/s | 33.5451 KOps/s | $\color{#35bf28}+3.48\%$ |
| test_step_mdp_speed[False-False-False-False-True] | 52.0610μs | 28.0056μs | 35.7071 KOps/s | 34.9452 KOps/s | $\color{#35bf28}+2.18\%$ |
| test_step_mdp_speed[False-False-False-False-False] | 0.1971ms | 18.4241μs | 54.2768 KOps/s | 52.8411 KOps/s | $\color{#35bf28}+2.72\%$ |
| test_values[generalized_advantage_estimate-True-True] | 24.4727ms | 23.3542ms | 42.8188 Ops/s | 42.6754 Ops/s | $\color{#35bf28}+0.34\%$ |
| test_values[vec_generalized_advantage_estimate-True-True] | 92.2848ms | 2.7307ms | 366.2121 Ops/s | 375.2731 Ops/s | $\color{#d91a1a}-2.41\%$ |
| test_values[td0_return_estimate-False-False] | 90.8310μs | 65.0734μs | 15.3673 KOps/s | 15.5863 KOps/s | $\color{#d91a1a}-1.41\%$ |
| test_values[td1_return_estimate-False-False] | 54.2131ms | 52.7562ms | 18.9551 Ops/s | 18.9347 Ops/s | $\color{#35bf28}+0.11\%$ |
| test_values[vec_td1_return_estimate-False-False] | 1.4111ms | 1.0718ms | 933.0153 Ops/s | 927.6805 Ops/s | $\color{#35bf28}+0.58\%$ |
| test_values[td_lambda_return_estimate-True-False] | 84.6032ms | 83.2762ms | 12.0082 Ops/s | 11.7166 Ops/s | $\color{#35bf28}+2.49\%$ |
| test_values[vec_td_lambda_return_estimate-True-False] | 1.3990ms | 1.0671ms | 937.1552 Ops/s | 928.6635 Ops/s | $\color{#35bf28}+0.91\%$ |
| test_gae_speed[generalized_advantage_estimate-False-1-512] | 24.2472ms | 23.5024ms | 42.5488 Ops/s | 42.7474 Ops/s | $\color{#d91a1a}-0.46\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-1-512] | 0.9163ms | 0.6953ms | 1.4383 KOps/s | 1.4234 KOps/s | $\color{#35bf28}+1.05\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-1-512] | 0.8123ms | 0.6508ms | 1.5366 KOps/s | 1.4928 KOps/s | $\color{#35bf28}+2.93\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-True-32-512] | 1.6198ms | 1.4499ms | 689.7118 Ops/s | 687.1193 Ops/s | $\color{#35bf28}+0.38\%$ |
| test_gae_speed[vec_generalized_advantage_estimate-False-32-512] | 0.8418ms | 0.6662ms | 1.5011 KOps/s | 1.4615 KOps/s | $\color{#35bf28}+2.72\%$ |
| test_dqn_speed | 7.7513ms | 1.4397ms | 694.5680 Ops/s | 702.1763 Ops/s | $\color{#d91a1a}-1.08\%$ |
| test_ddpg_speed | 3.1641ms | 2.9184ms | 342.6496 Ops/s | 344.9976 Ops/s | $\color{#d91a1a}-0.68\%$ |
| test_sac_speed | 8.7112ms | 8.3793ms | 119.3412 Ops/s | 119.5778 Ops/s | $\color{#d91a1a}-0.20\%$ |
| test_redq_speed | 12.8073ms | 10.5608ms | 94.6900 Ops/s | 93.4101 Ops/s | $\color{#35bf28}+1.37\%$ |
| test_redq_deprec_speed | 13.0302ms | 11.5029ms | 86.9345 Ops/s | 86.1911 Ops/s | $\color{#35bf28}+0.86\%$ |
| test_td3_speed | 8.5364ms | 8.3265ms | 120.0988 Ops/s | 120.0865 Ops/s | $\color{#35bf28}+0.01\%$ |
| test_cql_speed | 27.1495ms | 25.7460ms | 38.8409 Ops/s | 38.8829 Ops/s | $\color{#d91a1a}-0.11\%$ |
| test_a2c_speed | 5.8376ms | 5.5744ms | 179.3930 Ops/s | 174.9330 Ops/s | $\color{#35bf28}+2.55\%$ |
| test_ppo_speed | 7.1036ms | 6.0009ms | 166.6403 Ops/s | 164.4680 Ops/s | $\color{#35bf28}+1.32\%$ |
| test_reinforce_speed | 5.4270ms | 4.5945ms | 217.6539 Ops/s | 214.5364 Ops/s | $\color{#35bf28}+1.45\%$ |
| test_iql_speed | 20.2648ms | 19.6420ms | 50.9113 Ops/s | 50.1146 Ops/s | $\color{#35bf28}+1.59\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 4.9081ms | 4.7635ms | 209.9284 Ops/s | 209.2085 Ops/s | $\color{#35bf28}+0.34\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 1.4299ms | 0.5949ms | 1.6809 KOps/s | 1.6809 KOps/s | $-0.00\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7622ms | 0.5695ms | 1.7558 KOps/s | 1.7435 KOps/s | $\color{#35bf28}+0.71\%$ |
| test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.5265ms | 4.6993ms | 212.7984 Ops/s | 210.2700 Ops/s | $\color{#35bf28}+1.20\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 1.3018ms | 0.5894ms | 1.6967 KOps/s | 1.7071 KOps/s | $\color{#d91a1a}-0.61\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7405ms | 0.5642ms | 1.7723 KOps/s | 1.4986 KOps/s | $\textbf{\color{#35bf28}+18.27\%}$ |
| test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] | 5.2799ms | 2.0556ms | 486.4802 Ops/s | 484.6462 Ops/s | $\color{#35bf28}+0.38\%$ |
| test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] | 2.1100ms | 1.9526ms | 512.1293 Ops/s | 509.5730 Ops/s | $\color{#35bf28}+0.50\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.1717ms | 4.8933ms | 204.3601 Ops/s | 203.0581 Ops/s | $\color{#35bf28}+0.64\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9873ms | 0.7248ms | 1.3798 KOps/s | 1.3967 KOps/s | $\color{#d91a1a}-1.21\%$ |
| test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 4.9508ms | 0.7007ms | 1.4271 KOps/s | 1.4474 KOps/s | $\color{#d91a1a}-1.40\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] | 4.9704ms | 4.7629ms | 209.9580 Ops/s | 208.3933 Ops/s | $\color{#35bf28}+0.75\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] | 0.7913ms | 0.5942ms | 1.6829 KOps/s | 1.6771 KOps/s | $\color{#35bf28}+0.35\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] | 0.7528ms | 0.5721ms | 1.7481 KOps/s | 1.7369 KOps/s | $\color{#35bf28}+0.64\%$ |
| test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] | 5.0030ms | 4.7392ms | 211.0064 Ops/s | 210.4316 Ops/s | $\color{#35bf28}+0.27\%$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] | 0.1359s | 0.7860ms | 1.2723 KOps/s | 1.6967 KOps/s | $\textbf{\color{#d91a1a}-25.01\%}$ |
| test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] | 0.7276ms | 0.5633ms | 1.7753 KOps/s | 1.7672 KOps/s | $\color{#35bf28}+0.46\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] | 5.0727ms | 4.8808ms | 204.8838 Ops/s | 202.1973 Ops/s | $\color{#35bf28}+1.33\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] | 0.9227ms | 0.7210ms | 1.3870 KOps/s | 1.3954 KOps/s | $\color{#d91a1a}-0.60\%$ |
| test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] | 0.8791ms | 0.6973ms | 1.4342 KOps/s | 1.4363 KOps/s | $\color{#d91a1a}-0.15\%$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] | 0.1341s | 7.7214ms | 129.5098 Ops/s | 134.7574 Ops/s | $\color{#d91a1a}-3.89\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] | 18.3377ms | 15.9248ms | 62.7952 Ops/s | 62.4851 Ops/s | $\color{#35bf28}+0.50\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] | 2.2647ms | 1.2766ms | 783.3240 Ops/s | 760.5122 Ops/s | $\color{#35bf28}+3.00\%$ |
| test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] | 0.1266s | 9.8572ms | 101.4487 Ops/s | 140.5986 Ops/s | $\textbf{\color{#d91a1a}-27.85\%}$ |
| test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] | 18.4528ms | 16.0723ms | 62.2189 Ops/s | 62.4100 Ops/s | $\color{#d91a1a}-0.31\%$ |
| test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] | 2.3703ms | 1.3026ms | 767.7066 Ops/s | 699.1779 Ops/s | $\textbf{\color{#35bf28}+9.80\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] | 0.1264s | 7.6828ms | 130.1613 Ops/s | 108.1123 Ops/s | $\textbf{\color{#35bf28}+20.39\%}$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] | 19.0859ms | 16.2393ms | 61.5789 Ops/s | 61.9049 Ops/s | $\color{#d91a1a}-0.53\%$ |
| test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] | 7.5234ms | 1.5721ms | 636.1032 Ops/s | 677.9601 Ops/s | $\textbf{\color{#d91a1a}-6.17\%}$ |