Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Feature] Immutable writer for datasets #1781

Merged
merged 16 commits into from
Jan 9, 2024
Merged

[Feature] Immutable writer for datasets #1781

merged 16 commits into from
Jan 9, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Jan 8, 2024 •
edited
Loading

No description provided.

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Jan 8, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/1781

Note: Links to docs will display an error until the docs builds have been completed.

✅ You can merge normally! (5 Unrelated Failures)

As of commit 839483b with merge base 11a82c3 (image):

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

BROKEN TRUNK - The following job failed but were present on the merge base:

👉 Rebase onto the `viable/strict` branch to avoid these failures

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Jan 8, 2024
@vmoens vmoens added the enhancement New feature or request label Jan 8, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Jan 8, 2024 •
edited
Loading

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of CPU Benchmark Tests

Total Benchmarks: 89. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 65.4079ms 64.9443ms 15.3978 Ops/s 15.2837 Ops/s $\color{#35bf28}+0.75\%$
test_sync 40.2590ms 35.9048ms 27.8514 Ops/s 29.1381 Ops/s $\color{#d91a1a}-4.42\%$
test_async 0.1033s 34.5448ms 28.9479 Ops/s 29.6686 Ops/s $\color{#d91a1a}-2.43\%$
test_simple 0.5027s 0.4461s 2.2417 Ops/s 2.1915 Ops/s $\color{#35bf28}+2.29\%$
test_transformed 0.6723s 0.6183s 1.6173 Ops/s 1.5835 Ops/s $\color{#35bf28}+2.13\%$
test_serial 1.4408s 1.3895s 0.7197 Ops/s 0.7080 Ops/s $\color{#35bf28}+1.65\%$
test_parallel 1.4610s 1.3916s 0.7186 Ops/s 0.7482 Ops/s $\color{#d91a1a}-3.96\%$
test_step_mdp_speed[True-True-True-True-True] 0.1406ms 21.3074μs 46.9321 KOps/s 45.1987 KOps/s $\color{#35bf28}+3.84\%$
test_step_mdp_speed[True-True-True-True-False] 54.0010μs 12.9753μs 77.0698 KOps/s 75.3780 KOps/s $\color{#35bf28}+2.24\%$
test_step_mdp_speed[True-True-True-False-True] 49.2530μs 12.6266μs 79.1976 KOps/s 75.8655 KOps/s $\color{#35bf28}+4.39\%$
test_step_mdp_speed[True-True-True-False-False] 32.9620μs 7.9752μs 125.3891 KOps/s 126.0721 KOps/s $\color{#d91a1a}-0.54\%$
test_step_mdp_speed[True-True-False-True-True] 69.4200μs 23.2691μs 42.9755 KOps/s 42.4706 KOps/s $\color{#35bf28}+1.19\%$
test_step_mdp_speed[True-True-False-True-False] 48.9310μs 14.4669μs 69.1235 KOps/s 69.2216 KOps/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[True-True-False-False-True] 61.9250μs 14.0318μs 71.2669 KOps/s 70.0448 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[True-True-False-False-False] 29.7050μs 8.9647μs 111.5484 KOps/s 98.0101 KOps/s $\textbf{\color{#35bf28}+13.81\%}$
test_step_mdp_speed[True-False-True-True-True] 71.6240μs 24.2678μs 41.2069 KOps/s 36.3411 KOps/s $\textbf{\color{#35bf28}+13.39\%}$
test_step_mdp_speed[True-False-True-True-False] 59.3510μs 15.6272μs 63.9911 KOps/s 62.2989 KOps/s $\color{#35bf28}+2.72\%$
test_step_mdp_speed[True-False-True-False-True] 45.9050μs 14.1632μs 70.6054 KOps/s 69.8433 KOps/s $\color{#35bf28}+1.09\%$
test_step_mdp_speed[True-False-True-False-False] 35.7270μs 8.9909μs 111.2241 KOps/s 109.4902 KOps/s $\color{#35bf28}+1.58\%$
test_step_mdp_speed[True-False-False-True-True] 67.8370μs 25.6537μs 38.9807 KOps/s 38.3702 KOps/s $\color{#35bf28}+1.59\%$
test_step_mdp_speed[True-False-False-True-False] 60.9640μs 16.8171μs 59.4633 KOps/s 58.1217 KOps/s $\color{#35bf28}+2.31\%$
test_step_mdp_speed[True-False-False-False-True] 44.4930μs 15.2114μs 65.7404 KOps/s 65.6022 KOps/s $\color{#35bf28}+0.21\%$
test_step_mdp_speed[True-False-False-False-False] 54.4810μs 10.0969μs 99.0406 KOps/s 96.1767 KOps/s $\color{#35bf28}+2.98\%$
test_step_mdp_speed[False-True-True-True-True] 67.5360μs 24.1835μs 41.3506 KOps/s 40.2089 KOps/s $\color{#35bf28}+2.84\%$
test_step_mdp_speed[False-True-True-True-False] 47.2280μs 15.7935μs 63.3172 KOps/s 63.2816 KOps/s $\color{#35bf28}+0.06\%$
test_step_mdp_speed[False-True-True-False-True] 63.8590μs 16.1929μs 61.7555 KOps/s 60.1951 KOps/s $\color{#35bf28}+2.59\%$
test_step_mdp_speed[False-True-True-False-False] 31.7090μs 10.2829μs 97.2487 KOps/s 96.7110 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-True-False-True-True] 70.0900μs 25.3155μs 39.5015 KOps/s 38.5590 KOps/s $\color{#35bf28}+2.44\%$
test_step_mdp_speed[False-True-False-True-False] 38.7920μs 16.8356μs 59.3980 KOps/s 57.8100 KOps/s $\color{#35bf28}+2.75\%$
test_step_mdp_speed[False-True-False-False-True] 46.2870μs 17.4711μs 57.2373 KOps/s 56.4954 KOps/s $\color{#35bf28}+1.31\%$
test_step_mdp_speed[False-True-False-False-False] 37.2700μs 11.2354μs 89.0042 KOps/s 85.4781 KOps/s $\color{#35bf28}+4.13\%$
test_step_mdp_speed[False-False-True-True-True] 58.7490μs 26.4442μs 37.8154 KOps/s 36.8842 KOps/s $\color{#35bf28}+2.52\%$
test_step_mdp_speed[False-False-True-True-False] 63.6790μs 18.1188μs 55.1914 KOps/s 53.4412 KOps/s $\color{#35bf28}+3.27\%$
test_step_mdp_speed[False-False-True-False-True] 51.6570μs 17.3445μs 57.6550 KOps/s 56.2627 KOps/s $\color{#35bf28}+2.47\%$
test_step_mdp_speed[False-False-True-False-False] 58.8990μs 11.5341μs 86.6996 KOps/s 86.3196 KOps/s $\color{#35bf28}+0.44\%$
test_step_mdp_speed[False-False-False-True-True] 79.8090μs 27.8242μs 35.9400 KOps/s 35.3562 KOps/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[False-False-False-True-False] 49.8530μs 19.2289μs 52.0051 KOps/s 50.9202 KOps/s $\color{#35bf28}+2.13\%$
test_step_mdp_speed[False-False-False-False-True] 64.9010μs 18.3618μs 54.4609 KOps/s 53.6763 KOps/s $\color{#35bf28}+1.46\%$
test_step_mdp_speed[False-False-False-False-False] 31.8700μs 12.3251μs 81.1350 KOps/s 79.3044 KOps/s $\color{#35bf28}+2.31\%$
test_values[generalized_advantage_estimate-True-True] 15.8653ms 12.3822ms 80.7612 Ops/s 81.5500 Ops/s $\color{#d91a1a}-0.97\%$
test_values[vec_generalized_advantage_estimate-True-True] 34.2946ms 27.4009ms 36.4952 Ops/s 35.7582 Ops/s $\color{#35bf28}+2.06\%$
test_values[td0_return_estimate-False-False] 0.2218ms 0.1758ms 5.6884 KOps/s 4.1830 KOps/s $\textbf{\color{#35bf28}+35.99\%}$
test_values[td1_return_estimate-False-False] 28.8092ms 26.0241ms 38.4260 Ops/s 38.6625 Ops/s $\color{#d91a1a}-0.61\%$
test_values[vec_td1_return_estimate-False-False] 34.6242ms 27.4271ms 36.4602 Ops/s 36.1067 Ops/s $\color{#35bf28}+0.98\%$
test_values[td_lambda_return_estimate-True-False] 36.5102ms 36.2551ms 27.5823 Ops/s 27.7347 Ops/s $\color{#d91a1a}-0.55\%$
test_values[vec_td_lambda_return_estimate-True-False] 35.0264ms 27.6329ms 36.1888 Ops/s 36.4332 Ops/s $\color{#d91a1a}-0.67\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.2777ms 8.1860ms 122.1602 Ops/s 121.7513 Ops/s $\color{#35bf28}+0.34\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.1811ms 1.9239ms 519.7663 Ops/s 548.8073 Ops/s $\textbf{\color{#d91a1a}-5.29\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 8.2986ms 0.4345ms 2.3014 KOps/s 2.3252 KOps/s $\color{#d91a1a}-1.03\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 47.6689ms 40.0175ms 24.9891 Ops/s 29.3266 Ops/s $\textbf{\color{#d91a1a}-14.79\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 10.5354ms 2.6486ms 377.5521 Ops/s 378.5394 Ops/s $\color{#d91a1a}-0.26\%$
test_dqn_speed 77.6996ms 8.2675ms 120.9548 Ops/s 120.3438 Ops/s $\color{#35bf28}+0.51\%$
test_ddpg_speed 22.6792ms 14.7548ms 67.7746 Ops/s 67.3186 Ops/s $\color{#35bf28}+0.68\%$
test_sac_speed 41.3402ms 29.9572ms 33.3809 Ops/s 33.8922 Ops/s $\color{#d91a1a}-1.51\%$
test_redq_speed 37.2404ms 35.6443ms 28.0549 Ops/s 27.8058 Ops/s $\color{#35bf28}+0.90\%$
test_redq_deprec_speed 36.6122ms 25.8127ms 38.7406 Ops/s 39.0160 Ops/s $\color{#d91a1a}-0.71\%$
test_td3_speed 28.7581ms 20.4481ms 48.9044 Ops/s 49.1400 Ops/s $\color{#d91a1a}-0.48\%$
test_cql_speed 98.0427ms 89.3576ms 11.1910 Ops/s 11.2507 Ops/s $\color{#d91a1a}-0.53\%$
test_a2c_speed 35.3127ms 27.6836ms 36.1225 Ops/s 36.3527 Ops/s $\color{#d91a1a}-0.63\%$
test_ppo_speed 32.3249ms 27.6983ms 36.1032 Ops/s 35.6908 Ops/s $\color{#35bf28}+1.16\%$
test_reinforce_speed 49.2406ms 27.6820ms 36.1245 Ops/s 37.4016 Ops/s $\color{#d91a1a}-3.41\%$
test_iql_speed 72.2676ms 63.7019ms 15.6981 Ops/s 15.3950 Ops/s $\color{#35bf28}+1.97\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.7758ms 1.4379ms 695.4488 Ops/s 660.1523 Ops/s $\textbf{\color{#35bf28}+5.35\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.2545ms 0.5153ms 1.9407 KOps/s 1.9282 KOps/s $\color{#35bf28}+0.65\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 9.4220ms 0.5065ms 1.9742 KOps/s 2.0033 KOps/s $\color{#d91a1a}-1.45\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.7766ms 1.4072ms 710.6434 Ops/s 695.8917 Ops/s $\color{#35bf28}+2.12\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 8.9896ms 0.5160ms 1.9381 KOps/s 1.9250 KOps/s $\color{#35bf28}+0.68\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 8.6606ms 0.5028ms 1.9887 KOps/s 1.9908 KOps/s $\color{#d91a1a}-0.10\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.4116ms 1.6214ms 616.7643 Ops/s 585.8163 Ops/s $\textbf{\color{#35bf28}+5.28\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 10.0531ms 0.6573ms 1.5213 KOps/s 1.5185 KOps/s $\color{#35bf28}+0.19\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.7149ms 0.6399ms 1.5626 KOps/s 1.5456 KOps/s $\color{#35bf28}+1.10\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.1255ms 1.4602ms 684.8435 Ops/s 697.7534 Ops/s $\color{#d91a1a}-1.85\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 8.8845ms 0.5247ms 1.9059 KOps/s 1.9005 KOps/s $\color{#35bf28}+0.28\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 6.9089ms 0.5128ms 1.9501 KOps/s 2.0060 KOps/s $\color{#d91a1a}-2.79\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 1.6550ms 1.4111ms 708.6526 Ops/s 717.7636 Ops/s $\color{#d91a1a}-1.27\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 10.0424ms 0.5244ms 1.9068 KOps/s 1.9190 KOps/s $\color{#d91a1a}-0.63\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.1261s 0.6527ms 1.5321 KOps/s 2.0201 KOps/s $\textbf{\color{#d91a1a}-24.16\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.4660ms 1.6079ms 621.9338 Ops/s 588.8300 Ops/s $\textbf{\color{#35bf28}+5.62\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 8.9187ms 0.6633ms 1.5075 KOps/s 1.5111 KOps/s $\color{#d91a1a}-0.24\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 8.6638ms 0.6470ms 1.5457 KOps/s 1.5682 KOps/s $\color{#d91a1a}-1.44\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1165s 16.6373ms 60.1059 Ops/s 57.4500 Ops/s $\color{#35bf28}+4.62\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 14.3922ms 12.1697ms 82.1713 Ops/s 79.5251 Ops/s $\color{#35bf28}+3.33\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.2792ms 1.5947ms 627.0583 Ops/s 588.4476 Ops/s $\textbf{\color{#35bf28}+6.56\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1074s 16.2856ms 61.4039 Ops/s 58.4488 Ops/s $\textbf{\color{#35bf28}+5.06\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 14.8264ms 12.1890ms 82.0415 Ops/s 81.0327 Ops/s $\color{#35bf28}+1.24\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.0953ms 1.5025ms 665.5441 Ops/s 627.3242 Ops/s $\textbf{\color{#35bf28}+6.09\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1122s 16.6940ms 59.9017 Ops/s 56.7365 Ops/s $\textbf{\color{#35bf28}+5.58\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 14.9686ms 12.4000ms 80.6449 Ops/s 78.7033 Ops/s $\color{#35bf28}+2.47\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.3199ms 1.7021ms 587.5150 Ops/s 598.5666 Ops/s $\color{#d91a1a}-1.85\%$

@github-actions GitHub Actions
Copy link

github-actions bot commented Jan 8, 2024 •
edited
Loading

$\color{#D29922}\textsf{\Large⚠\kern{0.2cm}\normalsize Warning}$ Result of GPU Benchmark Tests

Total Benchmarks: 92. Improved: $\large\color{#35bf28}2$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1242s 0.1236s 8.0895 Ops/s 8.4058 Ops/s $\color{#d91a1a}-3.76\%$
test_sync 0.1797s 0.1104s 9.0605 Ops/s 9.1740 Ops/s $\color{#d91a1a}-1.24\%$
test_async 0.2772s 0.1014s 9.8615 Ops/s 10.1548 Ops/s $\color{#d91a1a}-2.89\%$
test_single_pixels 0.1469s 0.1466s 6.8194 Ops/s 6.9475 Ops/s $\color{#d91a1a}-1.84\%$
test_sync_pixels 95.4839ms 95.1359ms 10.5113 Ops/s 10.5534 Ops/s $\color{#d91a1a}-0.40\%$
test_async_pixels 0.2462s 90.9655ms 10.9932 Ops/s 10.9676 Ops/s $\color{#35bf28}+0.23\%$
test_simple 0.9547s 0.8836s 1.1317 Ops/s 1.1355 Ops/s $\color{#d91a1a}-0.33\%$
test_transformed 1.2051s 1.1390s 0.8779 Ops/s 0.9099 Ops/s $\color{#d91a1a}-3.52\%$
test_serial 2.5510s 2.4833s 0.4027 Ops/s 0.4041 Ops/s $\color{#d91a1a}-0.35\%$
test_parallel 2.5808s 2.5164s 0.3974 Ops/s 0.4074 Ops/s $\color{#d91a1a}-2.46\%$
test_step_mdp_speed[True-True-True-True-True] 0.2647ms 33.6465μs 29.7208 KOps/s 30.8408 KOps/s $\color{#d91a1a}-3.63\%$
test_step_mdp_speed[True-True-True-True-False] 38.2310μs 19.7268μs 50.6925 KOps/s 52.5409 KOps/s $\color{#d91a1a}-3.52\%$
test_step_mdp_speed[True-True-True-False-True] 0.1840ms 19.5336μs 51.1938 KOps/s 53.5937 KOps/s $\color{#d91a1a}-4.48\%$
test_step_mdp_speed[True-True-True-False-False] 37.7800μs 11.4111μs 87.6341 KOps/s 90.7870 KOps/s $\color{#d91a1a}-3.47\%$
test_step_mdp_speed[True-True-False-True-True] 0.2057ms 34.6459μs 28.8634 KOps/s 29.4229 KOps/s $\color{#d91a1a}-1.90\%$
test_step_mdp_speed[True-True-False-True-False] 0.2430ms 21.5377μs 46.4302 KOps/s 48.1140 KOps/s $\color{#d91a1a}-3.50\%$
test_step_mdp_speed[True-True-False-False-True] 51.3300μs 20.8273μs 48.0139 KOps/s 48.6878 KOps/s $\color{#d91a1a}-1.38\%$
test_step_mdp_speed[True-True-False-False-False] 0.2092ms 13.2745μs 75.3322 KOps/s 77.9747 KOps/s $\color{#d91a1a}-3.39\%$
test_step_mdp_speed[True-False-True-True-True] 71.8900μs 36.7348μs 27.2221 KOps/s 27.6675 KOps/s $\color{#d91a1a}-1.61\%$
test_step_mdp_speed[True-False-True-True-False] 0.2137ms 23.6767μs 42.2356 KOps/s 43.9523 KOps/s $\color{#d91a1a}-3.91\%$
test_step_mdp_speed[True-False-True-False-True] 42.4310μs 20.7450μs 48.2043 KOps/s 48.9276 KOps/s $\color{#d91a1a}-1.48\%$
test_step_mdp_speed[True-False-True-False-False] 0.2098ms 13.3700μs 74.7943 KOps/s 77.6688 KOps/s $\color{#d91a1a}-3.70\%$
test_step_mdp_speed[True-False-False-True-True] 0.2151ms 38.5334μs 25.9515 KOps/s 26.6219 KOps/s $\color{#d91a1a}-2.52\%$
test_step_mdp_speed[True-False-False-True-False] 45.7100μs 25.4997μs 39.2162 KOps/s 40.1506 KOps/s $\color{#d91a1a}-2.33\%$
test_step_mdp_speed[True-False-False-False-True] 46.1710μs 22.7903μs 43.8783 KOps/s 44.7121 KOps/s $\color{#d91a1a}-1.86\%$
test_step_mdp_speed[True-False-False-False-False] 0.2232ms 15.1438μs 66.0335 KOps/s 68.2052 KOps/s $\color{#d91a1a}-3.18\%$
test_step_mdp_speed[False-True-True-True-True] 69.8700μs 36.9875μs 27.0362 KOps/s 27.6533 KOps/s $\color{#d91a1a}-2.23\%$
test_step_mdp_speed[False-True-True-True-False] 48.1410μs 23.7894μs 42.0355 KOps/s 43.9613 KOps/s $\color{#d91a1a}-4.38\%$
test_step_mdp_speed[False-True-True-False-True] 44.5610μs 25.2796μs 39.5577 KOps/s 41.2688 KOps/s $\color{#d91a1a}-4.15\%$
test_step_mdp_speed[False-True-True-False-False] 32.7300μs 15.3692μs 65.0651 KOps/s 68.1962 KOps/s $\color{#d91a1a}-4.59\%$
test_step_mdp_speed[False-True-False-True-True] 66.7910μs 38.6303μs 25.8864 KOps/s 26.3602 KOps/s $\color{#d91a1a}-1.80\%$
test_step_mdp_speed[False-True-False-True-False] 49.6400μs 25.3718μs 39.4139 KOps/s 40.6550 KOps/s $\color{#d91a1a}-3.05\%$
test_step_mdp_speed[False-True-False-False-True] 62.1600μs 27.1772μs 36.7955 KOps/s 39.0437 KOps/s $\textbf{\color{#d91a1a}-5.76\%}$
test_step_mdp_speed[False-True-False-False-False] 35.7100μs 17.0663μs 58.5949 KOps/s 60.8684 KOps/s $\color{#d91a1a}-3.74\%$
test_step_mdp_speed[False-False-True-True-True] 69.7400μs 40.6208μs 24.6179 KOps/s 25.4356 KOps/s $\color{#d91a1a}-3.21\%$
test_step_mdp_speed[False-False-True-True-False] 0.1962ms 27.5202μs 36.3370 KOps/s 37.7840 KOps/s $\color{#d91a1a}-3.83\%$
test_step_mdp_speed[False-False-True-False-True] 49.8010μs 27.0189μs 37.0112 KOps/s 38.2847 KOps/s $\color{#d91a1a}-3.33\%$
test_step_mdp_speed[False-False-True-False-False] 33.9500μs 16.9882μs 58.8644 KOps/s 60.3150 KOps/s $\color{#d91a1a}-2.41\%$
test_step_mdp_speed[False-False-False-True-True] 0.2424ms 41.8439μs 23.8983 KOps/s 24.4252 KOps/s $\color{#d91a1a}-2.16\%$
test_step_mdp_speed[False-False-False-True-False] 0.1960ms 29.2312μs 34.2100 KOps/s 35.4630 KOps/s $\color{#d91a1a}-3.53\%$
test_step_mdp_speed[False-False-False-False-True] 0.2097ms 28.4160μs 35.1915 KOps/s 36.3392 KOps/s $\color{#d91a1a}-3.16\%$
test_step_mdp_speed[False-False-False-False-False] 0.2158ms 18.6752μs 53.5469 KOps/s 54.8050 KOps/s $\color{#d91a1a}-2.30\%$
test_values[generalized_advantage_estimate-True-True] 24.1188ms 23.4705ms 42.6066 Ops/s 43.0686 Ops/s $\color{#d91a1a}-1.07\%$
test_values[vec_generalized_advantage_estimate-True-True] 99.4518ms 3.5466ms 281.9633 Ops/s 288.7075 Ops/s $\color{#d91a1a}-2.34\%$
test_values[td0_return_estimate-False-False] 96.0710μs 61.8565μs 16.1664 KOps/s 16.8076 KOps/s $\color{#d91a1a}-3.81\%$
test_values[td1_return_estimate-False-False] 51.8024ms 50.5887ms 19.7672 Ops/s 20.1762 Ops/s $\color{#d91a1a}-2.03\%$
test_values[vec_td1_return_estimate-False-False] 2.0807ms 1.7581ms 568.7915 Ops/s 577.5975 Ops/s $\color{#d91a1a}-1.52\%$
test_values[td_lambda_return_estimate-True-False] 84.1505ms 80.9794ms 12.3488 Ops/s 12.5837 Ops/s $\color{#d91a1a}-1.87\%$
test_values[vec_td_lambda_return_estimate-True-False] 2.0292ms 1.7508ms 571.1545 Ops/s 576.8104 Ops/s $\color{#d91a1a}-0.98\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 22.5258ms 22.2048ms 45.0352 Ops/s 46.0801 Ops/s $\color{#d91a1a}-2.27\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.8996ms 0.6938ms 1.4414 KOps/s 1.4885 KOps/s $\color{#d91a1a}-3.16\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7892ms 0.6441ms 1.5525 KOps/s 1.5924 KOps/s $\color{#d91a1a}-2.51\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6496ms 1.4516ms 688.9086 Ops/s 697.7347 Ops/s $\color{#d91a1a}-1.26\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.9228ms 0.6683ms 1.4964 KOps/s 1.5309 KOps/s $\color{#d91a1a}-2.26\%$
test_dqn_speed 14.3490ms 7.7022ms 129.8338 Ops/s 137.9689 Ops/s $\textbf{\color{#d91a1a}-5.90\%}$
test_ddpg_speed 15.8877ms 14.9992ms 66.6703 Ops/s 70.5206 Ops/s $\textbf{\color{#d91a1a}-5.46\%}$
test_sac_speed 30.9587ms 30.0213ms 33.3097 Ops/s 35.0537 Ops/s $\color{#d91a1a}-4.98\%$
test_redq_speed 37.4443ms 36.1161ms 27.6885 Ops/s 29.0388 Ops/s $\color{#d91a1a}-4.65\%$
test_redq_deprec_speed 26.1935ms 24.9923ms 40.0123 Ops/s 41.6142 Ops/s $\color{#d91a1a}-3.85\%$
test_td3_speed 29.4615ms 20.3482ms 49.1444 Ops/s 51.4974 Ops/s $\color{#d91a1a}-4.57\%$
test_cql_speed 88.0924ms 87.1274ms 11.4774 Ops/s 11.9852 Ops/s $\color{#d91a1a}-4.24\%$
test_a2c_speed 29.2512ms 27.8520ms 35.9041 Ops/s 37.6122 Ops/s $\color{#d91a1a}-4.54\%$
test_ppo_speed 0.1302s 30.9026ms 32.3597 Ops/s 37.1457 Ops/s $\textbf{\color{#d91a1a}-12.88\%}$
test_reinforce_speed 27.4689ms 26.9343ms 37.1274 Ops/s 38.8597 Ops/s $\color{#d91a1a}-4.46\%$
test_iql_speed 60.8286ms 59.6377ms 16.7679 Ops/s 17.4054 Ops/s $\color{#d91a1a}-3.66\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 0.1070s 2.1644ms 462.0299 Ops/s 529.1425 Ops/s $\textbf{\color{#d91a1a}-12.68\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.2313ms 0.8507ms 1.1755 KOps/s 1.1957 KOps/s $\color{#d91a1a}-1.69\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 1.0901ms 0.8396ms 1.1911 KOps/s 1.2126 KOps/s $\color{#d91a1a}-1.77\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.4698ms 1.9025ms 525.6363 Ops/s 539.1541 Ops/s $\color{#d91a1a}-2.51\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.2296ms 0.8377ms 1.1937 KOps/s 1.2105 KOps/s $\color{#d91a1a}-1.39\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 1.0064ms 0.8260ms 1.2106 KOps/s 1.2254 KOps/s $\color{#d91a1a}-1.21\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.9438ms 2.2245ms 449.5385 Ops/s 463.4309 Ops/s $\color{#d91a1a}-3.00\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.5782ms 0.9709ms 1.0300 KOps/s 882.7316 Ops/s $\textbf{\color{#35bf28}+16.68\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.1627ms 0.9608ms 1.0408 KOps/s 1.0576 KOps/s $\color{#d91a1a}-1.59\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.9006ms 1.9450ms 514.1338 Ops/s 532.3770 Ops/s $\color{#d91a1a}-3.43\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0519ms 0.8476ms 1.1797 KOps/s 1.1942 KOps/s $\color{#d91a1a}-1.21\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 5.2471ms 0.8452ms 1.1832 KOps/s 1.1995 KOps/s $\color{#d91a1a}-1.37\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.3210ms 1.9094ms 523.7376 Ops/s 542.7990 Ops/s $\color{#d91a1a}-3.51\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.3925ms 0.8410ms 1.1891 KOps/s 1.2086 KOps/s $\color{#d91a1a}-1.62\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 1.0595ms 0.8295ms 1.2055 KOps/s 1.2248 KOps/s $\color{#d91a1a}-1.58\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.9839ms 2.2290ms 448.6347 Ops/s 459.9640 Ops/s $\color{#d91a1a}-2.46\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 3.7424ms 0.9763ms 1.0242 KOps/s 1.0417 KOps/s $\color{#d91a1a}-1.68\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.2297ms 0.9618ms 1.0397 KOps/s 1.0560 KOps/s $\color{#d91a1a}-1.54\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1691s 18.8426ms 53.0712 Ops/s 55.5856 Ops/s $\color{#d91a1a}-4.52\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 15.6025ms 12.3963ms 80.6691 Ops/s 83.7500 Ops/s $\color{#d91a1a}-3.68\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.6251ms 1.9197ms 520.9223 Ops/s 535.8055 Ops/s $\color{#d91a1a}-2.78\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1302s 18.1294ms 55.1589 Ops/s 65.8577 Ops/s $\textbf{\color{#d91a1a}-16.25\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 15.2706ms 12.4380ms 80.3985 Ops/s 70.5485 Ops/s $\textbf{\color{#35bf28}+13.96\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.8731ms 1.9303ms 518.0446 Ops/s 525.1935 Ops/s $\color{#d91a1a}-1.36\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1304s 18.2776ms 54.7117 Ops/s 65.4353 Ops/s $\textbf{\color{#d91a1a}-16.39\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 15.9473ms 12.5777ms 79.5059 Ops/s 83.4138 Ops/s $\color{#d91a1a}-4.68\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.3357ms 2.1198ms 471.7373 Ops/s 501.0221 Ops/s $\textbf{\color{#d91a1a}-5.85\%}$

@vmoens vmoens added the Data Data-related PR, will launch data-related jobs label Jan 9, 2024
@vmoens vmoens merged commit 8194565 into main Jan 9, 2024
59 of 64 checks passed
@vmoens vmoens deleted the fix-writers-datasets branch January 9, 2024 20:51
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Data Data-related PR, will launch data-related jobs enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@vmoens @facebook-github-bot