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.

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

[BugFix] Fix truncated normal #2147

Merged
merged 2 commits into from
May 1, 2024
Merged

[BugFix] Fix truncated normal #2147

merged 2 commits into from
May 1, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented May 1, 2024

No description provided.

Copy link

pytorch-bot bot commented May 1, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2147

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

❌ 2 New Failures, 2 Unrelated Failures

As of commit c6c491a with merge base 711a4ee (image):

NEW FAILURES - The following jobs have failed:

BROKEN TRUNK - The following jobs 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 May 1, 2024
@vmoens vmoens linked an issue May 1, 2024 that may be closed by this pull request
3 tasks
@vmoens vmoens added the bug Something isn't working label May 1, 2024
Copy link

github-actions bot commented May 1, 2024

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

Total Benchmarks: 91. Improved: $\large\color{#35bf28}6$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 56.8709ms 54.7157ms 18.2763 Ops/s 18.4193 Ops/s $\color{#d91a1a}-0.78\%$
test_sync 32.2412ms 30.2211ms 33.0894 Ops/s 32.7638 Ops/s $\color{#35bf28}+0.99\%$
test_async 59.6299ms 28.5986ms 34.9668 Ops/s 35.4260 Ops/s $\color{#d91a1a}-1.30\%$
test_simple 0.4092s 0.3553s 2.8146 Ops/s 2.9214 Ops/s $\color{#d91a1a}-3.65\%$
test_transformed 0.4958s 0.4942s 2.0233 Ops/s 1.9843 Ops/s $\color{#35bf28}+1.97\%$
test_serial 1.2870s 1.2300s 0.8130 Ops/s 0.8184 Ops/s $\color{#d91a1a}-0.65\%$
test_parallel 1.0749s 1.0281s 0.9727 Ops/s 0.9965 Ops/s $\color{#d91a1a}-2.39\%$
test_step_mdp_speed[True-True-True-True-True] 0.1757ms 21.0849μs 47.4274 KOps/s 47.3219 KOps/s $\color{#35bf28}+0.22\%$
test_step_mdp_speed[True-True-True-True-False] 37.0890μs 12.8792μs 77.6443 KOps/s 77.5629 KOps/s $\color{#35bf28}+0.10\%$
test_step_mdp_speed[True-True-True-False-True] 41.8980μs 12.4087μs 80.5885 KOps/s 80.7832 KOps/s $\color{#d91a1a}-0.24\%$
test_step_mdp_speed[True-True-True-False-False] 30.5070μs 7.4439μs 134.3382 KOps/s 134.4111 KOps/s $\color{#d91a1a}-0.05\%$
test_step_mdp_speed[True-True-False-True-True] 66.1830μs 22.4992μs 44.4461 KOps/s 44.7563 KOps/s $\color{#d91a1a}-0.69\%$
test_step_mdp_speed[True-True-False-True-False] 42.2900μs 14.2180μs 70.3335 KOps/s 70.9096 KOps/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[True-True-False-False-True] 0.1332ms 14.2298μs 70.2750 KOps/s 73.4852 KOps/s $\color{#d91a1a}-4.37\%$
test_step_mdp_speed[True-True-False-False-False] 86.5620μs 8.8093μs 113.5164 KOps/s 115.5536 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[True-False-True-True-True] 56.7460μs 24.0126μs 41.6447 KOps/s 42.1268 KOps/s $\color{#d91a1a}-1.14\%$
test_step_mdp_speed[True-False-True-True-False] 43.9730μs 15.5816μs 64.1784 KOps/s 64.7193 KOps/s $\color{#d91a1a}-0.84\%$
test_step_mdp_speed[True-False-True-False-True] 34.8350μs 13.8506μs 72.1992 KOps/s 74.2080 KOps/s $\color{#d91a1a}-2.71\%$
test_step_mdp_speed[True-False-True-False-False] 34.7560μs 8.8208μs 113.3686 KOps/s 115.1029 KOps/s $\color{#d91a1a}-1.51\%$
test_step_mdp_speed[True-False-False-True-True] 57.0070μs 25.4901μs 39.2310 KOps/s 39.9086 KOps/s $\color{#d91a1a}-1.70\%$
test_step_mdp_speed[True-False-False-True-False] 60.5930μs 16.9697μs 58.9286 KOps/s 60.3153 KOps/s $\color{#d91a1a}-2.30\%$
test_step_mdp_speed[True-False-False-False-True] 59.5410μs 14.9691μs 66.8043 KOps/s 68.1381 KOps/s $\color{#d91a1a}-1.96\%$
test_step_mdp_speed[True-False-False-False-False] 63.8730μs 10.0963μs 99.0461 KOps/s 101.5392 KOps/s $\color{#d91a1a}-2.46\%$
test_step_mdp_speed[False-True-True-True-True] 0.1873ms 24.2142μs 41.2981 KOps/s 42.5323 KOps/s $\color{#d91a1a}-2.90\%$
test_step_mdp_speed[False-True-True-True-False] 81.7430μs 15.6903μs 63.7337 KOps/s 64.7718 KOps/s $\color{#d91a1a}-1.60\%$
test_step_mdp_speed[False-True-True-False-True] 42.8600μs 16.0270μs 62.3947 KOps/s 63.8053 KOps/s $\color{#d91a1a}-2.21\%$
test_step_mdp_speed[False-True-True-False-False] 30.7780μs 10.1483μs 98.5387 KOps/s 99.3288 KOps/s $\color{#d91a1a}-0.80\%$
test_step_mdp_speed[False-True-False-True-True] 52.1170μs 25.2301μs 39.6352 KOps/s 39.9878 KOps/s $\color{#d91a1a}-0.88\%$
test_step_mdp_speed[False-True-False-True-False] 53.5600μs 16.8352μs 59.3994 KOps/s 58.5425 KOps/s $\color{#35bf28}+1.46\%$
test_step_mdp_speed[False-True-False-False-True] 0.1007ms 17.2666μs 57.9153 KOps/s 58.6724 KOps/s $\color{#d91a1a}-1.29\%$
test_step_mdp_speed[False-True-False-False-False] 29.4750μs 11.3121μs 88.4012 KOps/s 90.9488 KOps/s $\color{#d91a1a}-2.80\%$
test_step_mdp_speed[False-False-True-True-True] 76.0330μs 26.7805μs 37.3405 KOps/s 38.2409 KOps/s $\color{#d91a1a}-2.35\%$
test_step_mdp_speed[False-False-True-True-False] 38.7030μs 18.3403μs 54.5247 KOps/s 56.0064 KOps/s $\color{#d91a1a}-2.65\%$
test_step_mdp_speed[False-False-True-False-True] 40.8160μs 17.3604μs 57.6022 KOps/s 58.9210 KOps/s $\color{#d91a1a}-2.24\%$
test_step_mdp_speed[False-False-True-False-False] 35.8380μs 11.4732μs 87.1593 KOps/s 90.4726 KOps/s $\color{#d91a1a}-3.66\%$
test_step_mdp_speed[False-False-False-True-True] 47.6190μs 28.3057μs 35.3286 KOps/s 36.4770 KOps/s $\color{#d91a1a}-3.15\%$
test_step_mdp_speed[False-False-False-True-False] 63.4990μs 19.3550μs 51.6661 KOps/s 53.0038 KOps/s $\color{#d91a1a}-2.52\%$
test_step_mdp_speed[False-False-False-False-True] 44.3430μs 18.2130μs 54.9057 KOps/s 55.7724 KOps/s $\color{#d91a1a}-1.55\%$
test_step_mdp_speed[False-False-False-False-False] 39.9250μs 12.4748μs 80.1615 KOps/s 82.7272 KOps/s $\color{#d91a1a}-3.10\%$
test_values[generalized_advantage_estimate-True-True] 9.8411ms 9.6424ms 103.7081 Ops/s 105.5673 Ops/s $\color{#d91a1a}-1.76\%$
test_values[vec_generalized_advantage_estimate-True-True] 41.0403ms 35.6816ms 28.0256 Ops/s 28.6596 Ops/s $\color{#d91a1a}-2.21\%$
test_values[td0_return_estimate-False-False] 0.2252ms 0.1947ms 5.1367 KOps/s 6.0632 KOps/s $\textbf{\color{#d91a1a}-15.28\%}$
test_values[td1_return_estimate-False-False] 27.8283ms 24.2643ms 41.2128 Ops/s 43.1026 Ops/s $\color{#d91a1a}-4.38\%$
test_values[vec_td1_return_estimate-False-False] 38.5542ms 35.5490ms 28.1302 Ops/s 27.8653 Ops/s $\color{#35bf28}+0.95\%$
test_values[td_lambda_return_estimate-True-False] 35.5833ms 33.9988ms 29.4128 Ops/s 29.3811 Ops/s $\color{#35bf28}+0.11\%$
test_values[vec_td_lambda_return_estimate-True-False] 36.6234ms 35.3080ms 28.3222 Ops/s 27.8736 Ops/s $\color{#35bf28}+1.61\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 10.3013ms 8.3349ms 119.9775 Ops/s 122.7651 Ops/s $\color{#d91a1a}-2.27\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.3174ms 2.0075ms 498.1266 Ops/s 555.0813 Ops/s $\textbf{\color{#d91a1a}-10.26\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4621ms 0.3520ms 2.8412 KOps/s 2.8043 KOps/s $\color{#35bf28}+1.32\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 49.1863ms 47.9197ms 20.8682 Ops/s 21.2557 Ops/s $\color{#d91a1a}-1.82\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.6251ms 3.0390ms 329.0522 Ops/s 329.6733 Ops/s $\color{#d91a1a}-0.19\%$
test_dqn_speed 7.9990ms 1.3876ms 720.6884 Ops/s 722.5503 Ops/s $\color{#d91a1a}-0.26\%$
test_ddpg_speed 3.5992ms 2.9094ms 343.7150 Ops/s 344.0354 Ops/s $\color{#d91a1a}-0.09\%$
test_sac_speed 10.1094ms 8.5045ms 117.5851 Ops/s 117.2321 Ops/s $\color{#35bf28}+0.30\%$
test_redq_speed 15.5620ms 13.0854ms 76.4210 Ops/s 70.4923 Ops/s $\textbf{\color{#35bf28}+8.41\%}$
test_redq_deprec_speed 90.5497ms 14.2262ms 70.2930 Ops/s 74.9556 Ops/s $\textbf{\color{#d91a1a}-6.22\%}$
test_td3_speed 9.1022ms 8.4641ms 118.1454 Ops/s 117.1192 Ops/s $\color{#35bf28}+0.88\%$
test_cql_speed 38.2228ms 37.1118ms 26.9456 Ops/s 27.0020 Ops/s $\color{#d91a1a}-0.21\%$
test_a2c_speed 8.6901ms 7.5798ms 131.9300 Ops/s 133.2996 Ops/s $\color{#d91a1a}-1.03\%$
test_ppo_speed 8.5623ms 7.8615ms 127.2016 Ops/s 129.4375 Ops/s $\color{#d91a1a}-1.73\%$
test_reinforce_speed 7.7569ms 6.8510ms 145.9643 Ops/s 141.3465 Ops/s $\color{#35bf28}+3.27\%$
test_iql_speed 35.7127ms 33.5832ms 29.7768 Ops/s 29.9011 Ops/s $\color{#d91a1a}-0.42\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.6600ms 2.2656ms 441.3787 Ops/s 442.1417 Ops/s $\color{#d91a1a}-0.17\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0760ms 0.5084ms 1.9670 KOps/s 1.9440 KOps/s $\color{#35bf28}+1.19\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7826ms 0.4820ms 2.0749 KOps/s 2.0541 KOps/s $\color{#35bf28}+1.01\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.2749ms 2.1582ms 463.3518 Ops/s 462.5873 Ops/s $\color{#35bf28}+0.17\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0360ms 0.4971ms 2.0118 KOps/s 1.9911 KOps/s $\color{#35bf28}+1.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7963ms 0.4736ms 2.1116 KOps/s 2.0979 KOps/s $\color{#35bf28}+0.65\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.9375ms 1.2492ms 800.5348 Ops/s 800.8654 Ops/s $\color{#d91a1a}-0.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.9844ms 1.1894ms 840.7811 Ops/s 833.9467 Ops/s $\color{#35bf28}+0.82\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.6531ms 2.2823ms 438.1504 Ops/s 420.1486 Ops/s $\color{#35bf28}+4.28\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0547ms 0.6210ms 1.6104 KOps/s 1.5855 KOps/s $\color{#35bf28}+1.57\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9074ms 0.5953ms 1.6797 KOps/s 1.6462 KOps/s $\color{#35bf28}+2.04\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.3896ms 2.0610ms 485.1906 Ops/s 442.5776 Ops/s $\textbf{\color{#35bf28}+9.63\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.0161ms 0.5067ms 1.9737 KOps/s 1.9417 KOps/s $\color{#35bf28}+1.65\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 3.4960ms 0.4792ms 2.0869 KOps/s 2.0624 KOps/s $\color{#35bf28}+1.19\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.1553ms 2.0616ms 485.0508 Ops/s 452.6071 Ops/s $\textbf{\color{#35bf28}+7.17\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.5766ms 0.4939ms 2.0247 KOps/s 2.0111 KOps/s $\color{#35bf28}+0.68\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7065ms 0.4729ms 2.1145 KOps/s 2.0640 KOps/s $\color{#35bf28}+2.45\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.8449ms 2.2577ms 442.9328 Ops/s 444.3314 Ops/s $\color{#d91a1a}-0.31\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0854ms 0.6200ms 1.6130 KOps/s 1.5955 KOps/s $\color{#35bf28}+1.10\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8260ms 0.5924ms 1.6882 KOps/s 1.6718 KOps/s $\color{#35bf28}+0.98\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1082s 7.7474ms 129.0754 Ops/s 117.4537 Ops/s $\textbf{\color{#35bf28}+9.89\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 14.2200ms 12.0269ms 83.1473 Ops/s 80.6387 Ops/s $\color{#35bf28}+3.11\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 1.7790ms 1.0389ms 962.5849 Ops/s 940.1949 Ops/s $\color{#35bf28}+2.38\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1052s 5.6153ms 178.0845 Ops/s 171.7920 Ops/s $\color{#35bf28}+3.66\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1064s 13.8497ms 72.2039 Ops/s 80.1263 Ops/s $\textbf{\color{#d91a1a}-9.89\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.5033ms 1.0335ms 967.5952 Ops/s 901.2650 Ops/s $\textbf{\color{#35bf28}+7.36\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1047s 6.0073ms 166.4641 Ops/s 118.8393 Ops/s $\textbf{\color{#35bf28}+40.07\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 15.6446ms 12.4969ms 80.0197 Ops/s 79.4188 Ops/s $\color{#35bf28}+0.76\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.0302ms 1.3727ms 728.4676 Ops/s 695.8080 Ops/s $\color{#35bf28}+4.69\%$

Copy link

github-actions bot commented May 1, 2024

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

Total Benchmarks: 94. Improved: $\large\color{#35bf28}2$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1141s 0.1139s 8.7762 Ops/s 8.8072 Ops/s $\color{#d91a1a}-0.35\%$
test_sync 0.1005s 0.1001s 9.9887 Ops/s 9.9729 Ops/s $\color{#35bf28}+0.16\%$
test_async 0.1902s 95.4638ms 10.4752 Ops/s 10.3986 Ops/s $\color{#35bf28}+0.74\%$
test_single_pixels 0.1961s 0.1328s 7.5290 Ops/s 8.0332 Ops/s $\textbf{\color{#d91a1a}-6.28\%}$
test_sync_pixels 0.1390s 84.8655ms 11.7834 Ops/s 11.7348 Ops/s $\color{#35bf28}+0.41\%$
test_async_pixels 0.1583s 80.2518ms 12.4608 Ops/s 12.7842 Ops/s $\color{#d91a1a}-2.53\%$
test_simple 0.8239s 0.7640s 1.3089 Ops/s 1.3414 Ops/s $\color{#d91a1a}-2.42\%$
test_transformed 1.0638s 1.0087s 0.9914 Ops/s 0.9953 Ops/s $\color{#d91a1a}-0.39\%$
test_serial 2.4632s 2.4153s 0.4140 Ops/s 0.4169 Ops/s $\color{#d91a1a}-0.70\%$
test_parallel 2.3769s 2.3152s 0.4319 Ops/s 0.4355 Ops/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[True-True-True-True-True] 0.1052ms 33.3463μs 29.9883 KOps/s 30.1675 KOps/s $\color{#d91a1a}-0.59\%$
test_step_mdp_speed[True-True-True-True-False] 48.4510μs 19.4984μs 51.2863 KOps/s 50.4115 KOps/s $\color{#35bf28}+1.74\%$
test_step_mdp_speed[True-True-True-False-True] 42.9800μs 18.8176μs 53.1417 KOps/s 53.1061 KOps/s $\color{#35bf28}+0.07\%$
test_step_mdp_speed[True-True-True-False-False] 42.8410μs 11.2970μs 88.5187 KOps/s 88.1545 KOps/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[True-True-False-True-True] 62.3410μs 35.0080μs 28.5649 KOps/s 28.4979 KOps/s $\color{#35bf28}+0.23\%$
test_step_mdp_speed[True-True-False-True-False] 42.5610μs 21.7410μs 45.9960 KOps/s 46.1198 KOps/s $\color{#d91a1a}-0.27\%$
test_step_mdp_speed[True-True-False-False-True] 45.3610μs 20.9059μs 47.8334 KOps/s 48.3853 KOps/s $\color{#d91a1a}-1.14\%$
test_step_mdp_speed[True-True-False-False-False] 29.7910μs 13.2232μs 75.6248 KOps/s 75.3527 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[True-False-True-True-True] 62.7020μs 36.4096μs 27.4653 KOps/s 26.7767 KOps/s $\color{#35bf28}+2.57\%$
test_step_mdp_speed[True-False-True-True-False] 44.5810μs 23.5711μs 42.4249 KOps/s 42.1904 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[True-False-True-False-True] 45.6210μs 20.7461μs 48.2018 KOps/s 47.9112 KOps/s $\color{#35bf28}+0.61\%$
test_step_mdp_speed[True-False-True-False-False] 30.2000μs 13.1635μs 75.9678 KOps/s 75.2594 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[True-False-False-True-True] 73.6920μs 37.9155μs 26.3744 KOps/s 25.6611 KOps/s $\color{#35bf28}+2.78\%$
test_step_mdp_speed[True-False-False-True-False] 76.7610μs 25.3073μs 39.5142 KOps/s 39.5930 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[True-False-False-False-True] 52.7410μs 22.5202μs 44.4046 KOps/s 44.8124 KOps/s $\color{#d91a1a}-0.91\%$
test_step_mdp_speed[True-False-False-False-False] 37.3200μs 14.9602μs 66.8440 KOps/s 66.6363 KOps/s $\color{#35bf28}+0.31\%$
test_step_mdp_speed[False-True-True-True-True] 63.1910μs 37.2094μs 26.8749 KOps/s 27.0455 KOps/s $\color{#d91a1a}-0.63\%$
test_step_mdp_speed[False-True-True-True-False] 45.9800μs 23.5508μs 42.4613 KOps/s 41.6267 KOps/s $\color{#35bf28}+2.01\%$
test_step_mdp_speed[False-True-True-False-True] 48.5810μs 24.8046μs 40.3151 KOps/s 40.0347 KOps/s $\color{#35bf28}+0.70\%$
test_step_mdp_speed[False-True-True-False-False] 40.0710μs 14.7357μs 67.8623 KOps/s 65.3780 KOps/s $\color{#35bf28}+3.80\%$
test_step_mdp_speed[False-True-False-True-True] 58.9910μs 38.3610μs 26.0681 KOps/s 25.4044 KOps/s $\color{#35bf28}+2.61\%$
test_step_mdp_speed[False-True-False-True-False] 46.6210μs 25.6023μs 39.0590 KOps/s 38.9063 KOps/s $\color{#35bf28}+0.39\%$
test_step_mdp_speed[False-True-False-False-True] 52.4810μs 26.8312μs 37.2701 KOps/s 37.4454 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[False-True-False-False-False] 43.0210μs 16.6857μs 59.9314 KOps/s 58.3655 KOps/s $\color{#35bf28}+2.68\%$
test_step_mdp_speed[False-False-True-True-True] 65.1520μs 40.8149μs 24.5009 KOps/s 24.3714 KOps/s $\color{#35bf28}+0.53\%$
test_step_mdp_speed[False-False-True-True-False] 55.1920μs 27.5334μs 36.3196 KOps/s 36.3473 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[False-False-True-False-True] 49.5010μs 26.8663μs 37.2214 KOps/s 37.9194 KOps/s $\color{#d91a1a}-1.84\%$
test_step_mdp_speed[False-False-True-False-False] 36.8110μs 16.7585μs 59.6711 KOps/s 58.7972 KOps/s $\color{#35bf28}+1.49\%$
test_step_mdp_speed[False-False-False-True-True] 65.7910μs 42.6197μs 23.4633 KOps/s 23.2185 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[False-False-False-True-False] 52.7820μs 29.3783μs 34.0387 KOps/s 33.8411 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[False-False-False-False-True] 51.5510μs 28.3149μs 35.3171 KOps/s 35.5730 KOps/s $\color{#d91a1a}-0.72\%$
test_step_mdp_speed[False-False-False-False-False] 47.1310μs 18.5117μs 54.0200 KOps/s 53.3428 KOps/s $\color{#35bf28}+1.27\%$
test_values[generalized_advantage_estimate-True-True] 24.2397ms 23.7820ms 42.0487 Ops/s 42.7244 Ops/s $\color{#d91a1a}-1.58\%$
test_values[vec_generalized_advantage_estimate-True-True] 82.5002ms 3.2061ms 311.9082 Ops/s 304.0961 Ops/s $\color{#35bf28}+2.57\%$
test_values[td0_return_estimate-False-False] 87.4320μs 64.6288μs 15.4730 KOps/s 15.9459 KOps/s $\color{#d91a1a}-2.97\%$
test_values[td1_return_estimate-False-False] 51.7651ms 50.5760ms 19.7722 Ops/s 19.8539 Ops/s $\color{#d91a1a}-0.41\%$
test_values[vec_td1_return_estimate-False-False] 2.0860ms 1.7521ms 570.7517 Ops/s 573.4591 Ops/s $\color{#d91a1a}-0.47\%$
test_values[td_lambda_return_estimate-True-False] 83.2584ms 81.0841ms 12.3329 Ops/s 12.4491 Ops/s $\color{#d91a1a}-0.93\%$
test_values[vec_td_lambda_return_estimate-True-False] 2.1070ms 1.7574ms 569.0120 Ops/s 573.7414 Ops/s $\color{#d91a1a}-0.82\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 22.8621ms 22.3993ms 44.6442 Ops/s 44.4069 Ops/s $\color{#35bf28}+0.53\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.8863ms 0.6920ms 1.4452 KOps/s 1.4760 KOps/s $\color{#d91a1a}-2.09\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7594ms 0.6509ms 1.5363 KOps/s 1.5811 KOps/s $\color{#d91a1a}-2.83\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5468ms 1.4448ms 692.1156 Ops/s 696.2118 Ops/s $\color{#d91a1a}-0.59\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.9147ms 0.6634ms 1.5075 KOps/s 1.5442 KOps/s $\color{#d91a1a}-2.38\%$
test_dqn_speed 1.6307ms 1.4120ms 708.1958 Ops/s 690.3628 Ops/s $\color{#35bf28}+2.58\%$
test_ddpg_speed 3.1967ms 2.9616ms 337.6506 Ops/s 340.1495 Ops/s $\color{#d91a1a}-0.73\%$
test_sac_speed 8.6821ms 8.3641ms 119.5588 Ops/s 119.9967 Ops/s $\color{#d91a1a}-0.36\%$
test_redq_speed 11.1413ms 10.2807ms 97.2693 Ops/s 97.7214 Ops/s $\color{#d91a1a}-0.46\%$
test_redq_deprec_speed 12.2519ms 11.6795ms 85.6204 Ops/s 80.9236 Ops/s $\textbf{\color{#35bf28}+5.80\%}$
test_td3_speed 8.4864ms 8.3394ms 119.9126 Ops/s 120.5095 Ops/s $\color{#d91a1a}-0.50\%$
test_cql_speed 26.8070ms 25.8862ms 38.6306 Ops/s 39.3758 Ops/s $\color{#d91a1a}-1.89\%$
test_a2c_speed 6.6730ms 5.7630ms 173.5208 Ops/s 177.0566 Ops/s $\color{#d91a1a}-2.00\%$
test_ppo_speed 7.0347ms 6.0075ms 166.4575 Ops/s 169.2251 Ops/s $\color{#d91a1a}-1.64\%$
test_reinforce_speed 4.9139ms 4.7273ms 211.5365 Ops/s 213.4993 Ops/s $\color{#d91a1a}-0.92\%$
test_iql_speed 20.3834ms 19.8700ms 50.3272 Ops/s 50.9812 Ops/s $\color{#d91a1a}-1.28\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 3.0653ms 2.8088ms 356.0203 Ops/s 357.4146 Ops/s $\color{#d91a1a}-0.39\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8009ms 0.5961ms 1.6777 KOps/s 1.6709 KOps/s $\color{#35bf28}+0.41\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 4.4124ms 0.5756ms 1.7372 KOps/s 1.7463 KOps/s $\color{#d91a1a}-0.52\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.0269ms 2.7916ms 358.2128 Ops/s 357.0079 Ops/s $\color{#35bf28}+0.34\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7362ms 0.5882ms 1.7002 KOps/s 1.6949 KOps/s $\color{#35bf28}+0.31\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 4.4502ms 0.5655ms 1.7682 KOps/s 1.7533 KOps/s $\color{#35bf28}+0.85\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.6687ms 1.5189ms 658.3804 Ops/s 670.4798 Ops/s $\color{#d91a1a}-1.80\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.6407ms 1.4488ms 690.2247 Ops/s 690.0810 Ops/s $\color{#35bf28}+0.02\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.0407ms 2.9285ms 341.4755 Ops/s 342.2534 Ops/s $\color{#d91a1a}-0.23\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.3878ms 0.7238ms 1.3817 KOps/s 1.3769 KOps/s $\color{#35bf28}+0.34\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8691ms 0.6975ms 1.4337 KOps/s 1.4329 KOps/s $\color{#35bf28}+0.05\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.8994ms 2.7433ms 364.5227 Ops/s 358.4341 Ops/s $\color{#35bf28}+1.70\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6930ms 0.5971ms 1.6749 KOps/s 1.6704 KOps/s $\color{#35bf28}+0.27\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 4.5894ms 0.5760ms 1.7361 KOps/s 1.7478 KOps/s $\color{#d91a1a}-0.67\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 3.0224ms 2.8084ms 356.0727 Ops/s 355.7696 Ops/s $\color{#35bf28}+0.09\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.7021ms 0.5910ms 1.6921 KOps/s 1.6985 KOps/s $\color{#d91a1a}-0.38\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7363ms 0.5664ms 1.7656 KOps/s 1.7654 KOps/s $+0.01\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 3.0412ms 2.9204ms 342.4219 Ops/s 341.0329 Ops/s $\color{#35bf28}+0.41\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8399ms 0.7233ms 1.3825 KOps/s 1.3755 KOps/s $\color{#35bf28}+0.51\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 4.6309ms 0.7030ms 1.4225 KOps/s 1.4182 KOps/s $\color{#35bf28}+0.30\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1244s 9.5208ms 105.0328 Ops/s 134.8686 Ops/s $\textbf{\color{#d91a1a}-22.12\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 16.8634ms 14.8392ms 67.3889 Ops/s 57.3244 Ops/s $\textbf{\color{#35bf28}+17.56\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.2396ms 1.1817ms 846.2520 Ops/s 846.0772 Ops/s $\color{#35bf28}+0.02\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1154s 7.1269ms 140.3136 Ops/s 141.8565 Ops/s $\color{#d91a1a}-1.09\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 17.1922ms 14.8871ms 67.1725 Ops/s 65.7165 Ops/s $\color{#35bf28}+2.22\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.3016ms 1.2091ms 827.0700 Ops/s 814.8400 Ops/s $\color{#35bf28}+1.50\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1182s 9.7723ms 102.3299 Ops/s 133.0230 Ops/s $\textbf{\color{#d91a1a}-23.07\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 17.5057ms 15.2621ms 65.5217 Ops/s 65.4520 Ops/s $\color{#35bf28}+0.11\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.7993ms 1.6386ms 610.2734 Ops/s 652.2743 Ops/s $\textbf{\color{#d91a1a}-6.44\%}$

@vmoens vmoens merged commit 69a6cb1 into main May 1, 2024
53 of 57 checks passed
@vmoens vmoens deleted the fix-trunc-normal branch May 1, 2024 16:08
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[BUG] TruncatedNormal returns wrong values in case of asymmetric bounds
2 participants