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

[Doc] Better doc for distributed RBs #2378

Open
wants to merge 39 commits into
base: main
Choose a base branch
from
Open

[Doc] Better doc for distributed RBs #2378

wants to merge 39 commits into from

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Aug 7, 2024 •
edited
Loading

No description provided.

@pytorch-bot PyTorch Bot
Copy link

pytorch-bot bot commented Aug 7, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2378

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

❌ 3 New Failures, 5 Unrelated Failures

As of commit 5f58811 with merge base a27514c (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 Aug 7, 2024
@github-actions GitHub Actions
Copy link

github-actions bot commented Aug 7, 2024 •
edited
Loading

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.4130s 0.4115s 2.4304 Ops/s 2.3712 Ops/s $\color{#35bf28}+2.50\%$
test_transformed 0.6741s 0.5959s 1.6783 Ops/s 1.6714 Ops/s $\color{#35bf28}+0.41\%$
test_serial 1.4078s 1.3318s 0.7509 Ops/s 0.7567 Ops/s $\color{#d91a1a}-0.77\%$
test_parallel 1.3799s 1.3149s 0.7605 Ops/s 0.7542 Ops/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[True-True-True-True-True] 0.2160ms 28.6088μs 34.9542 KOps/s 35.0860 KOps/s $\color{#d91a1a}-0.38\%$
test_step_mdp_speed[True-True-True-True-False] 54.0200μs 17.1089μs 58.4490 KOps/s 57.9018 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[True-True-True-False-True] 70.0600μs 15.9166μs 62.8273 KOps/s 63.6753 KOps/s $\color{#d91a1a}-1.33\%$
test_step_mdp_speed[True-True-True-False-False] 35.2860μs 9.3775μs 106.6388 KOps/s 105.7594 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[True-True-False-True-True] 69.8700μs 31.0269μs 32.2301 KOps/s 32.4535 KOps/s $\color{#d91a1a}-0.69\%$
test_step_mdp_speed[True-True-False-True-False] 48.9110μs 19.4045μs 51.5345 KOps/s 51.2991 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[True-True-False-False-True] 71.1230μs 18.0932μs 55.2695 KOps/s 54.9641 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[True-True-False-False-False] 35.4460μs 11.6447μs 85.8760 KOps/s 85.6487 KOps/s $\color{#35bf28}+0.27\%$
test_step_mdp_speed[True-False-True-True-True] 88.4540μs 33.2985μs 30.0314 KOps/s 30.0732 KOps/s $\color{#d91a1a}-0.14\%$
test_step_mdp_speed[True-False-True-True-False] 48.0290μs 21.4115μs 46.7039 KOps/s 46.8653 KOps/s $\color{#d91a1a}-0.34\%$
test_step_mdp_speed[True-False-True-False-True] 0.6162ms 18.1692μs 55.0383 KOps/s 56.0990 KOps/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[True-False-True-False-False] 42.0080μs 11.3906μs 87.7917 KOps/s 85.8967 KOps/s $\color{#35bf28}+2.21\%$
test_step_mdp_speed[True-False-False-True-True] 0.1040ms 34.8392μs 28.7033 KOps/s 28.5854 KOps/s $\color{#35bf28}+0.41\%$
test_step_mdp_speed[True-False-False-True-False] 68.3470μs 23.1485μs 43.1993 KOps/s 42.9534 KOps/s $\color{#35bf28}+0.57\%$
test_step_mdp_speed[True-False-False-False-True] 92.8520μs 19.8358μs 50.4138 KOps/s 49.8346 KOps/s $\color{#35bf28}+1.16\%$
test_step_mdp_speed[True-False-False-False-False] 42.2280μs 13.4890μs 74.1347 KOps/s 72.8610 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[False-True-True-True-True] 78.8570μs 33.1067μs 30.2054 KOps/s 29.9072 KOps/s $\color{#35bf28}+1.00\%$
test_step_mdp_speed[False-True-True-True-False] 71.3620μs 21.5907μs 46.3163 KOps/s 46.2299 KOps/s $\color{#35bf28}+0.19\%$
test_step_mdp_speed[False-True-True-False-True] 45.3140μs 21.3959μs 46.7380 KOps/s 45.3683 KOps/s $\color{#35bf28}+3.02\%$
test_step_mdp_speed[False-True-True-False-False] 54.3310μs 13.2864μs 75.2649 KOps/s 74.8511 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[False-True-False-True-True] 79.8180μs 35.7209μs 27.9948 KOps/s 28.2246 KOps/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[False-True-False-True-False] 66.0120μs 23.5491μs 42.4645 KOps/s 43.0266 KOps/s $\color{#d91a1a}-1.31\%$
test_step_mdp_speed[False-True-False-False-True] 2.8611ms 23.6736μs 42.2412 KOps/s 43.0202 KOps/s $\color{#d91a1a}-1.81\%$
test_step_mdp_speed[False-True-False-False-False] 39.8650μs 15.3003μs 65.3581 KOps/s 65.0976 KOps/s $\color{#35bf28}+0.40\%$
test_step_mdp_speed[False-False-True-True-True] 91.2390μs 36.9552μs 27.0598 KOps/s 26.6681 KOps/s $\color{#35bf28}+1.47\%$
test_step_mdp_speed[False-False-True-True-False] 53.1690μs 25.6357μs 39.0081 KOps/s 39.3439 KOps/s $\color{#d91a1a}-0.85\%$
test_step_mdp_speed[False-False-True-False-True] 76.8420μs 23.6334μs 42.3129 KOps/s 42.0336 KOps/s $\color{#35bf28}+0.66\%$
test_step_mdp_speed[False-False-True-False-False] 42.3790μs 15.3108μs 65.3135 KOps/s 65.1233 KOps/s $\color{#35bf28}+0.29\%$
test_step_mdp_speed[False-False-False-True-True] 98.6030μs 39.0374μs 25.6165 KOps/s 25.5000 KOps/s $\color{#35bf28}+0.46\%$
test_step_mdp_speed[False-False-False-True-False] 0.6140ms 27.2025μs 36.7614 KOps/s 36.2448 KOps/s $\color{#35bf28}+1.43\%$
test_step_mdp_speed[False-False-False-False-True] 74.7350μs 24.4808μs 40.8483 KOps/s 40.2023 KOps/s $\color{#35bf28}+1.61\%$
test_step_mdp_speed[False-False-False-False-False] 46.4860μs 17.0920μs 58.5070 KOps/s 57.5758 KOps/s $\color{#35bf28}+1.62\%$
test_values[generalized_advantage_estimate-True-True] 9.8652ms 9.6334ms 103.8052 Ops/s 104.4035 Ops/s $\color{#d91a1a}-0.57\%$
test_values[vec_generalized_advantage_estimate-True-True] 36.4466ms 33.5134ms 29.8388 Ops/s 29.8619 Ops/s $\color{#d91a1a}-0.08\%$
test_values[td0_return_estimate-False-False] 0.2369ms 0.1826ms 5.4758 KOps/s 5.9314 KOps/s $\textbf{\color{#d91a1a}-7.68\%}$
test_values[td1_return_estimate-False-False] 27.1035ms 24.1498ms 41.4082 Ops/s 41.5090 Ops/s $\color{#d91a1a}-0.24\%$
test_values[vec_td1_return_estimate-False-False] 35.1540ms 33.5238ms 29.8295 Ops/s 29.6573 Ops/s $\color{#35bf28}+0.58\%$
test_values[td_lambda_return_estimate-True-False] 38.1050ms 34.4067ms 29.0641 Ops/s 29.1645 Ops/s $\color{#d91a1a}-0.34\%$
test_values[vec_td_lambda_return_estimate-True-False] 36.2384ms 33.5567ms 29.8003 Ops/s 29.7775 Ops/s $\color{#35bf28}+0.08\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.8873ms 8.4964ms 117.6970 Ops/s 119.2384 Ops/s $\color{#d91a1a}-1.29\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4846ms 1.8195ms 549.6026 Ops/s 559.2466 Ops/s $\color{#d91a1a}-1.72\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4577ms 0.3558ms 2.8103 KOps/s 2.8038 KOps/s $\color{#35bf28}+0.23\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 42.5963ms 39.3179ms 25.4337 Ops/s 24.0725 Ops/s $\textbf{\color{#35bf28}+5.65\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.8793ms 3.0438ms 328.5368 Ops/s 328.4269 Ops/s $\color{#35bf28}+0.03\%$
test_dqn_speed[False-None] 5.9809ms 1.3617ms 734.3646 Ops/s 731.1581 Ops/s $\color{#35bf28}+0.44\%$
test_dqn_speed[False-backward] 2.0340ms 1.8348ms 545.0124 Ops/s 540.3308 Ops/s $\color{#35bf28}+0.87\%$
test_dqn_speed[True-None] 0.6965ms 0.4657ms 2.1474 KOps/s 2.1564 KOps/s $\color{#d91a1a}-0.42\%$
test_dqn_speed[True-backward] 0.9256ms 0.8735ms 1.1448 KOps/s 1.1191 KOps/s $\color{#35bf28}+2.30\%$
test_dqn_speed[reduce-overhead-None] 0.5673ms 0.4632ms 2.1587 KOps/s 2.1336 KOps/s $\color{#35bf28}+1.18\%$
test_dqn_speed[reduce-overhead-backward] 1.1003ms 0.9067ms 1.1029 KOps/s 1.1259 KOps/s $\color{#d91a1a}-2.04\%$
test_ddpg_speed[False-None] 3.5800ms 2.8035ms 356.7019 Ops/s 353.8323 Ops/s $\color{#35bf28}+0.81\%$
test_ddpg_speed[False-backward] 4.5779ms 3.9490ms 253.2305 Ops/s 250.1600 Ops/s $\color{#35bf28}+1.23\%$
test_ddpg_speed[True-None] 1.4693ms 0.9977ms 1.0023 KOps/s 994.6602 Ops/s $\color{#35bf28}+0.77\%$
test_ddpg_speed[True-backward] 2.1250ms 1.8923ms 528.4653 Ops/s 527.3681 Ops/s $\color{#35bf28}+0.21\%$
test_ddpg_speed[reduce-overhead-None] 1.4743ms 0.9989ms 1.0011 KOps/s 989.9629 Ops/s $\color{#35bf28}+1.12\%$
test_ddpg_speed[reduce-overhead-backward] 2.1110ms 1.8976ms 526.9840 Ops/s 517.9208 Ops/s $\color{#35bf28}+1.75\%$
test_sac_speed[False-None] 10.2125ms 7.9584ms 125.6541 Ops/s 124.6786 Ops/s $\color{#35bf28}+0.78\%$
test_sac_speed[False-backward] 11.0435ms 10.7025ms 93.4364 Ops/s 91.1779 Ops/s $\color{#35bf28}+2.48\%$
test_sac_speed[True-None] 2.4566ms 1.8496ms 540.6645 Ops/s 535.9623 Ops/s $\color{#35bf28}+0.88\%$
test_sac_speed[True-backward] 4.5384ms 3.5744ms 279.7673 Ops/s 253.2242 Ops/s $\textbf{\color{#35bf28}+10.48\%}$
test_sac_speed[reduce-overhead-None] 2.3690ms 1.8499ms 540.5828 Ops/s 535.1084 Ops/s $\color{#35bf28}+1.02\%$
test_sac_speed[reduce-overhead-backward] 3.6536ms 3.5527ms 281.4800 Ops/s 278.9912 Ops/s $\color{#35bf28}+0.89\%$
test_redq_speed[False-None] 14.4664ms 12.8708ms 77.6951 Ops/s 78.2482 Ops/s $\color{#d91a1a}-0.71\%$
test_redq_speed[False-backward] 22.8797ms 21.8397ms 45.7882 Ops/s 45.2151 Ops/s $\color{#35bf28}+1.27\%$
test_redq_speed[True-None] 5.8100ms 4.6961ms 212.9440 Ops/s 210.8216 Ops/s $\color{#35bf28}+1.01\%$
test_redq_speed[True-backward] 12.9216ms 12.0149ms 83.2297 Ops/s 82.1106 Ops/s $\color{#35bf28}+1.36\%$
test_redq_speed[reduce-overhead-None] 5.7370ms 4.6287ms 216.0441 Ops/s 197.0886 Ops/s $\textbf{\color{#35bf28}+9.62\%}$
test_redq_speed[reduce-overhead-backward] 13.2324ms 12.3192ms 81.1743 Ops/s 81.4544 Ops/s $\color{#d91a1a}-0.34\%$
test_redq_deprec_speed[False-None] 15.3792ms 12.8159ms 78.0280 Ops/s 78.5192 Ops/s $\color{#d91a1a}-0.63\%$
test_redq_deprec_speed[False-backward] 20.6852ms 18.5695ms 53.8517 Ops/s 53.5662 Ops/s $\color{#35bf28}+0.53\%$
test_redq_deprec_speed[True-None] 4.9104ms 3.6299ms 275.4903 Ops/s 276.7463 Ops/s $\color{#d91a1a}-0.45\%$
test_redq_deprec_speed[True-backward] 8.1991ms 7.9963ms 125.0585 Ops/s 123.3936 Ops/s $\color{#35bf28}+1.35\%$
test_redq_deprec_speed[reduce-overhead-None] 4.4079ms 3.7666ms 265.4933 Ops/s 277.3027 Ops/s $\color{#d91a1a}-4.26\%$
test_redq_deprec_speed[reduce-overhead-backward] 9.2390ms 8.0775ms 123.8010 Ops/s 124.6403 Ops/s $\color{#d91a1a}-0.67\%$
test_td3_speed[False-None] 8.1405ms 7.8397ms 127.5567 Ops/s 126.0724 Ops/s $\color{#35bf28}+1.18\%$
test_td3_speed[False-backward] 10.7648ms 10.3359ms 96.7499 Ops/s 95.0501 Ops/s $\color{#35bf28}+1.79\%$
test_td3_speed[True-None] 1.9666ms 1.7356ms 576.1737 Ops/s 567.2135 Ops/s $\color{#35bf28}+1.58\%$
test_td3_speed[True-backward] 3.4000ms 3.3330ms 300.0264 Ops/s 296.9688 Ops/s $\color{#35bf28}+1.03\%$
test_td3_speed[reduce-overhead-None] 1.9467ms 1.7537ms 570.2293 Ops/s 572.1140 Ops/s $\color{#d91a1a}-0.33\%$
test_td3_speed[reduce-overhead-backward] 4.2569ms 3.3685ms 296.8703 Ops/s 295.6560 Ops/s $\color{#35bf28}+0.41\%$
test_cql_speed[False-None] 38.9572ms 36.0589ms 27.7324 Ops/s 27.9941 Ops/s $\color{#d91a1a}-0.94\%$
test_cql_speed[False-backward] 49.8229ms 45.3934ms 22.0296 Ops/s 21.5078 Ops/s $\color{#35bf28}+2.43\%$
test_cql_speed[True-None] 16.9696ms 15.7007ms 63.6913 Ops/s 63.4945 Ops/s $\color{#35bf28}+0.31\%$
test_cql_speed[True-backward] 24.4249ms 22.2432ms 44.9575 Ops/s 43.2380 Ops/s $\color{#35bf28}+3.98\%$
test_cql_speed[reduce-overhead-None] 16.5496ms 15.6514ms 63.8919 Ops/s 62.5377 Ops/s $\color{#35bf28}+2.17\%$
test_cql_speed[reduce-overhead-backward] 23.4613ms 22.2759ms 44.8915 Ops/s 44.1579 Ops/s $\color{#35bf28}+1.66\%$
test_a2c_speed[False-None] 7.9658ms 7.0561ms 141.7215 Ops/s 135.8277 Ops/s $\color{#35bf28}+4.34\%$
test_a2c_speed[False-backward] 15.7132ms 14.2825ms 70.0156 Ops/s 65.2245 Ops/s $\textbf{\color{#35bf28}+7.35\%}$
test_a2c_speed[True-None] 3.9323ms 3.3043ms 302.6346 Ops/s 293.1360 Ops/s $\color{#35bf28}+3.24\%$
test_a2c_speed[True-backward] 10.3884ms 9.8100ms 101.9366 Ops/s 101.7219 Ops/s $\color{#35bf28}+0.21\%$
test_a2c_speed[reduce-overhead-None] 4.1359ms 3.3307ms 300.2405 Ops/s 297.6613 Ops/s $\color{#35bf28}+0.87\%$
test_a2c_speed[reduce-overhead-backward] 10.5495ms 9.9032ms 100.9771 Ops/s 101.8496 Ops/s $\color{#d91a1a}-0.86\%$
test_ppo_speed[False-None] 8.0178ms 7.3370ms 136.2953 Ops/s 134.1746 Ops/s $\color{#35bf28}+1.58\%$
test_ppo_speed[False-backward] 17.0022ms 14.9563ms 66.8617 Ops/s 68.0303 Ops/s $\color{#d91a1a}-1.72\%$
test_ppo_speed[True-None] 3.9614ms 3.6919ms 270.8661 Ops/s 264.2355 Ops/s $\color{#35bf28}+2.51\%$
test_ppo_speed[True-backward] 10.7102ms 9.6756ms 103.3529 Ops/s 103.4583 Ops/s $\color{#d91a1a}-0.10\%$
test_ppo_speed[reduce-overhead-None] 4.0960ms 3.6942ms 270.6948 Ops/s 264.4663 Ops/s $\color{#35bf28}+2.36\%$
test_ppo_speed[reduce-overhead-backward] 9.9373ms 9.6427ms 103.7052 Ops/s 103.3969 Ops/s $\color{#35bf28}+0.30\%$
test_reinforce_speed[False-None] 6.9912ms 6.4347ms 155.4085 Ops/s 154.4273 Ops/s $\color{#35bf28}+0.64\%$
test_reinforce_speed[False-backward] 11.2209ms 9.6923ms 103.1749 Ops/s 101.9406 Ops/s $\color{#35bf28}+1.21\%$
test_reinforce_speed[True-None] 3.2433ms 2.6411ms 378.6265 Ops/s 370.3256 Ops/s $\color{#35bf28}+2.24\%$
test_reinforce_speed[True-backward] 9.3376ms 8.6167ms 116.0533 Ops/s 114.1809 Ops/s $\color{#35bf28}+1.64\%$
test_reinforce_speed[reduce-overhead-None] 2.8759ms 2.6481ms 377.6249 Ops/s 372.0830 Ops/s $\color{#35bf28}+1.49\%$
test_reinforce_speed[reduce-overhead-backward] 9.4527ms 8.6107ms 116.1342 Ops/s 114.1541 Ops/s $\color{#35bf28}+1.73\%$
test_iql_speed[False-None] 33.1021ms 31.5747ms 31.6709 Ops/s 31.0027 Ops/s $\color{#35bf28}+2.16\%$
test_iql_speed[False-backward] 45.8896ms 44.4841ms 22.4799 Ops/s 22.0343 Ops/s $\color{#35bf28}+2.02\%$
test_iql_speed[True-None] 11.4996ms 10.7536ms 92.9921 Ops/s 91.7268 Ops/s $\color{#35bf28}+1.38\%$
test_iql_speed[True-backward] 22.4041ms 21.6844ms 46.1162 Ops/s 35.2143 Ops/s $\textbf{\color{#35bf28}+30.96\%}$
test_iql_speed[reduce-overhead-None] 11.8676ms 10.7181ms 93.2998 Ops/s 92.2704 Ops/s $\color{#35bf28}+1.12\%$
test_iql_speed[reduce-overhead-backward] 22.9244ms 22.2671ms 44.9092 Ops/s 45.2271 Ops/s $\color{#d91a1a}-0.70\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.7147ms 4.8730ms 205.2142 Ops/s 207.5778 Ops/s $\color{#d91a1a}-1.14\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.4234ms 0.4763ms 2.0993 KOps/s 2.0674 KOps/s $\color{#35bf28}+1.54\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7182ms 0.4558ms 2.1939 KOps/s 2.1671 KOps/s $\color{#35bf28}+1.24\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.6944ms 4.8178ms 207.5657 Ops/s 209.8620 Ops/s $\color{#d91a1a}-1.09\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.4635ms 0.4770ms 2.0963 KOps/s 2.0869 KOps/s $\color{#35bf28}+0.45\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7322ms 0.4634ms 2.1582 KOps/s 2.2121 KOps/s $\color{#d91a1a}-2.44\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.8090ms 1.5710ms 636.5321 Ops/s 625.4340 Ops/s $\color{#35bf28}+1.77\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.8411ms 1.5220ms 657.0441 Ops/s 639.2254 Ops/s $\color{#35bf28}+2.79\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.5028ms 4.9023ms 203.9864 Ops/s 196.6357 Ops/s $\color{#35bf28}+3.74\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.7746ms 0.6157ms 1.6242 KOps/s 1.5899 KOps/s $\color{#35bf28}+2.16\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.9453ms 0.5873ms 1.7028 KOps/s 1.6756 KOps/s $\color{#35bf28}+1.62\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0780ms 4.7361ms 211.1442 Ops/s 204.6247 Ops/s $\color{#35bf28}+3.19\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9959ms 0.4901ms 2.0404 KOps/s 2.0762 KOps/s $\color{#d91a1a}-1.72\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6884ms 0.4713ms 2.1216 KOps/s 2.1894 KOps/s $\color{#d91a1a}-3.09\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.0442ms 4.7018ms 212.6863 Ops/s 205.6146 Ops/s $\color{#35bf28}+3.44\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.5518ms 0.4692ms 2.1314 KOps/s 2.0598 KOps/s $\color{#35bf28}+3.48\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7445ms 0.4512ms 2.2163 KOps/s 2.1847 KOps/s $\color{#35bf28}+1.45\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.4143ms 4.8026ms 208.2186 Ops/s 205.8811 Ops/s $\color{#35bf28}+1.14\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.5542ms 0.6091ms 1.6417 KOps/s 1.5943 KOps/s $\color{#35bf28}+2.97\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7934ms 0.5870ms 1.7035 KOps/s 1.6178 KOps/s $\textbf{\color{#35bf28}+5.30\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.3919ms 4.1170ms 242.8958 Ops/s 35.9760 Ops/s $\textbf{\color{#35bf28}+575.16\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.6227ms 2.2337ms 447.6880 Ops/s 450.0810 Ops/s $\color{#d91a1a}-0.53\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.5949ms 1.3180ms 758.7082 Ops/s 734.9873 Ops/s $\color{#35bf28}+3.23\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.3849s 11.8184ms 84.6140 Ops/s 220.9076 Ops/s $\textbf{\color{#d91a1a}-61.70\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 8.8155ms 2.3220ms 430.6614 Ops/s 441.5124 Ops/s $\color{#d91a1a}-2.46\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 2.6621ms 1.2641ms 791.0716 Ops/s 790.0191 Ops/s $\color{#35bf28}+0.13\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.0721ms 4.2755ms 233.8924 Ops/s 223.2888 Ops/s $\color{#35bf28}+4.75\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 5.2069ms 2.4013ms 416.4375 Ops/s 409.5404 Ops/s $\color{#35bf28}+1.68\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.6724ms 1.4450ms 692.0347 Ops/s 736.7981 Ops/s $\textbf{\color{#d91a1a}-6.08\%}$

@github-actions GitHub Actions
Copy link

github-actions bot commented Aug 7, 2024 •
edited
Loading

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7384s 0.7319s 1.3664 Ops/s 1.3905 Ops/s $\color{#d91a1a}-1.73\%$
test_transformed 1.0519s 0.9751s 1.0255 Ops/s 1.0453 Ops/s $\color{#d91a1a}-1.89\%$
test_serial 2.2019s 2.1110s 0.4737 Ops/s 0.4797 Ops/s $\color{#d91a1a}-1.25\%$
test_parallel 2.0552s 1.9857s 0.5036 Ops/s 0.4922 Ops/s $\color{#35bf28}+2.32\%$
test_step_mdp_speed[True-True-True-True-True] 0.1821ms 37.3802μs 26.7522 KOps/s 25.5325 KOps/s $\color{#35bf28}+4.78\%$
test_step_mdp_speed[True-True-True-True-False] 0.1454ms 22.3990μs 44.6449 KOps/s 44.4311 KOps/s $\color{#35bf28}+0.48\%$
test_step_mdp_speed[True-True-True-False-True] 86.8510μs 20.7735μs 48.1383 KOps/s 47.6768 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-True-True-False-False] 0.1299ms 12.0699μs 82.8510 KOps/s 82.4314 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-True-False-True-True] 0.1176ms 40.3252μs 24.7984 KOps/s 23.9301 KOps/s $\color{#35bf28}+3.63\%$
test_step_mdp_speed[True-True-False-True-False] 0.1079ms 24.3282μs 41.1046 KOps/s 39.9416 KOps/s $\color{#35bf28}+2.91\%$
test_step_mdp_speed[True-True-False-False-True] 67.5410μs 23.4164μs 42.7051 KOps/s 41.2892 KOps/s $\color{#35bf28}+3.43\%$
test_step_mdp_speed[True-True-False-False-False] 40.1310μs 14.4352μs 69.2749 KOps/s 68.3327 KOps/s $\color{#35bf28}+1.38\%$
test_step_mdp_speed[True-False-True-True-True] 0.1351ms 42.7751μs 23.3781 KOps/s 22.6845 KOps/s $\color{#35bf28}+3.06\%$
test_step_mdp_speed[True-False-True-True-False] 69.8020μs 27.0707μs 36.9402 KOps/s 36.3232 KOps/s $\color{#35bf28}+1.70\%$
test_step_mdp_speed[True-False-True-False-True] 69.9510μs 23.1594μs 43.1790 KOps/s 41.6687 KOps/s $\color{#35bf28}+3.62\%$
test_step_mdp_speed[True-False-True-False-False] 89.5810μs 14.6042μs 68.4735 KOps/s 68.0447 KOps/s $\color{#35bf28}+0.63\%$
test_step_mdp_speed[True-False-False-True-True] 94.4720μs 46.0675μs 21.7073 KOps/s 21.3681 KOps/s $\color{#35bf28}+1.59\%$
test_step_mdp_speed[True-False-False-True-False] 0.1164ms 29.8601μs 33.4895 KOps/s 33.9843 KOps/s $\color{#d91a1a}-1.46\%$
test_step_mdp_speed[True-False-False-False-True] 0.1227ms 26.2362μs 38.1153 KOps/s 37.3399 KOps/s $\color{#35bf28}+2.08\%$
test_step_mdp_speed[True-False-False-False-False] 86.4710μs 17.0557μs 58.6313 KOps/s 57.8209 KOps/s $\color{#35bf28}+1.40\%$
test_step_mdp_speed[False-True-True-True-True] 0.1128ms 41.9785μs 23.8217 KOps/s 22.6062 KOps/s $\textbf{\color{#35bf28}+5.38\%}$
test_step_mdp_speed[False-True-True-True-False] 90.3520μs 26.7171μs 37.4292 KOps/s 36.7856 KOps/s $\color{#35bf28}+1.75\%$
test_step_mdp_speed[False-True-True-False-True] 67.1810μs 27.8045μs 35.9654 KOps/s 35.4403 KOps/s $\color{#35bf28}+1.48\%$
test_step_mdp_speed[False-True-True-False-False] 55.8010μs 16.1424μs 61.9488 KOps/s 58.1378 KOps/s $\textbf{\color{#35bf28}+6.55\%}$
test_step_mdp_speed[False-True-False-True-True] 0.1690ms 44.3024μs 22.5721 KOps/s 20.9986 KOps/s $\textbf{\color{#35bf28}+7.49\%}$
test_step_mdp_speed[False-True-False-True-False] 67.3710μs 29.0301μs 34.4470 KOps/s 33.2189 KOps/s $\color{#35bf28}+3.70\%$
test_step_mdp_speed[False-True-False-False-True] 3.4825ms 30.7156μs 32.5568 KOps/s 31.9316 KOps/s $\color{#35bf28}+1.96\%$
test_step_mdp_speed[False-True-False-False-False] 0.1838ms 19.3035μs 51.8040 KOps/s 51.4095 KOps/s $\color{#35bf28}+0.77\%$
test_step_mdp_speed[False-False-True-True-True] 87.0120μs 48.0564μs 20.8089 KOps/s 20.0283 KOps/s $\color{#35bf28}+3.90\%$
test_step_mdp_speed[False-False-True-True-False] 0.2261ms 32.2831μs 30.9759 KOps/s 30.5622 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[False-False-True-False-True] 0.2273ms 29.5238μs 33.8710 KOps/s 32.5439 KOps/s $\color{#35bf28}+4.08\%$
test_step_mdp_speed[False-False-True-False-False] 0.1938ms 19.0586μs 52.4697 KOps/s 51.1975 KOps/s $\color{#35bf28}+2.48\%$
test_step_mdp_speed[False-False-False-True-True] 0.2486ms 49.7156μs 20.1144 KOps/s 19.2573 KOps/s $\color{#35bf28}+4.45\%$
test_step_mdp_speed[False-False-False-True-False] 71.9610μs 34.0698μs 29.3515 KOps/s 28.3892 KOps/s $\color{#35bf28}+3.39\%$
test_step_mdp_speed[False-False-False-False-True] 71.0220μs 31.4042μs 31.8428 KOps/s 31.0713 KOps/s $\color{#35bf28}+2.48\%$
test_step_mdp_speed[False-False-False-False-False] 0.1572ms 21.2208μs 47.1235 KOps/s 45.6432 KOps/s $\color{#35bf28}+3.24\%$
test_values[generalized_advantage_estimate-True-True] 24.6224ms 24.3107ms 41.1342 Ops/s 40.2633 Ops/s $\color{#35bf28}+2.16\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1017s 2.9238ms 342.0163 Ops/s 340.5523 Ops/s $\color{#35bf28}+0.43\%$
test_values[td0_return_estimate-False-False] 0.2357ms 68.9119μs 14.5113 KOps/s 15.0241 KOps/s $\color{#d91a1a}-3.41\%$
test_values[td1_return_estimate-False-False] 55.0657ms 54.6372ms 18.3025 Ops/s 17.8649 Ops/s $\color{#35bf28}+2.45\%$
test_values[vec_td1_return_estimate-False-False] 1.4029ms 1.0757ms 929.5969 Ops/s 930.3422 Ops/s $\color{#d91a1a}-0.08\%$
test_values[td_lambda_return_estimate-True-False] 87.6141ms 87.2086ms 11.4668 Ops/s 11.5359 Ops/s $\color{#d91a1a}-0.60\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2556ms 1.0636ms 940.1753 Ops/s 934.1178 Ops/s $\color{#35bf28}+0.65\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 24.3683ms 24.0349ms 41.6062 Ops/s 41.6626 Ops/s $\color{#d91a1a}-0.14\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0381ms 0.7580ms 1.3192 KOps/s 1.3551 KOps/s $\color{#d91a1a}-2.65\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7914ms 0.6581ms 1.5194 KOps/s 1.5165 KOps/s $\color{#35bf28}+0.19\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6154ms 1.4650ms 682.6054 Ops/s 682.3020 Ops/s $\color{#35bf28}+0.04\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8641ms 0.6771ms 1.4769 KOps/s 1.4841 KOps/s $\color{#d91a1a}-0.49\%$
test_dqn_speed[False-None] 6.8029ms 1.3098ms 763.4950 Ops/s 755.0016 Ops/s $\color{#35bf28}+1.12\%$
test_dqn_speed[False-backward] 2.0172ms 1.8475ms 541.2650 Ops/s 538.3317 Ops/s $\color{#35bf28}+0.54\%$
test_dqn_speed[True-None] 0.9412ms 0.5484ms 1.8234 KOps/s 1.8266 KOps/s $\color{#d91a1a}-0.18\%$
test_dqn_speed[True-backward] 1.2229ms 1.0366ms 964.6565 Ops/s 815.3966 Ops/s $\textbf{\color{#35bf28}+18.31\%}$
test_dqn_speed[reduce-overhead-None] 0.9097ms 0.5638ms 1.7738 KOps/s 1.7149 KOps/s $\color{#35bf28}+3.44\%$
test_dqn_speed[reduce-overhead-backward] 1.4799ms 1.2158ms 822.4701 Ops/s 996.3429 Ops/s $\textbf{\color{#d91a1a}-17.45\%}$
test_ddpg_speed[False-None] 3.4293ms 2.7248ms 367.0034 Ops/s 371.1226 Ops/s $\color{#d91a1a}-1.11\%$
test_ddpg_speed[False-backward] 4.2649ms 3.9651ms 252.2006 Ops/s 245.3271 Ops/s $\color{#35bf28}+2.80\%$
test_ddpg_speed[True-None] 1.4682ms 1.2480ms 801.3010 Ops/s 795.7473 Ops/s $\color{#35bf28}+0.70\%$
test_ddpg_speed[True-backward] 2.4316ms 2.2274ms 448.9572 Ops/s 448.0686 Ops/s $\color{#35bf28}+0.20\%$
test_ddpg_speed[reduce-overhead-None] 1.5901ms 1.2366ms 808.6836 Ops/s 800.8129 Ops/s $\color{#35bf28}+0.98\%$
test_ddpg_speed[reduce-overhead-backward] 2.4595ms 2.2757ms 439.4255 Ops/s 454.6855 Ops/s $\color{#d91a1a}-3.36\%$
test_sac_speed[False-None] 8.0964ms 7.5356ms 132.7039 Ops/s 132.7042 Ops/s $-0.00\%$
test_sac_speed[False-backward] 11.3246ms 10.8193ms 92.4278 Ops/s 92.9242 Ops/s $\color{#d91a1a}-0.53\%$
test_sac_speed[True-None] 2.4067ms 2.0457ms 488.8360 Ops/s 491.0682 Ops/s $\color{#d91a1a}-0.45\%$
test_sac_speed[True-backward] 4.3211ms 3.9604ms 252.4995 Ops/s 251.3279 Ops/s $\color{#35bf28}+0.47\%$
test_sac_speed[reduce-overhead-None] 2.3615ms 2.0421ms 489.7033 Ops/s 485.6772 Ops/s $\color{#35bf28}+0.83\%$
test_sac_speed[reduce-overhead-backward] 4.1557ms 3.9812ms 251.1781 Ops/s 254.1177 Ops/s $\color{#d91a1a}-1.16\%$
test_redq_speed[False-None] 15.4205ms 10.1405ms 98.6146 Ops/s 99.4427 Ops/s $\color{#d91a1a}-0.83\%$
test_redq_speed[False-backward] 18.3586ms 17.4721ms 57.2341 Ops/s 57.1784 Ops/s $\color{#35bf28}+0.10\%$
test_redq_speed[True-None] 3.9384ms 3.6257ms 275.8094 Ops/s 281.1269 Ops/s $\color{#d91a1a}-1.89\%$
test_redq_speed[True-backward] 9.1241ms 8.6261ms 115.9278 Ops/s 104.5482 Ops/s $\textbf{\color{#35bf28}+10.88\%}$
test_redq_speed[reduce-overhead-None] 3.9544ms 3.5483ms 281.8284 Ops/s 277.8919 Ops/s $\color{#35bf28}+1.42\%$
test_redq_speed[reduce-overhead-backward] 9.0222ms 8.7533ms 114.2423 Ops/s 116.4707 Ops/s $\color{#d91a1a}-1.91\%$
test_redq_deprec_speed[False-None] 12.4695ms 10.6086ms 94.2632 Ops/s 92.5268 Ops/s $\color{#35bf28}+1.88\%$
test_redq_deprec_speed[False-backward] 15.9966ms 15.4656ms 64.6598 Ops/s 63.7234 Ops/s $\color{#35bf28}+1.47\%$
test_redq_deprec_speed[True-None] 3.6301ms 3.3114ms 301.9856 Ops/s 304.0676 Ops/s $\color{#d91a1a}-0.68\%$
test_redq_deprec_speed[True-backward] 7.4852ms 7.1702ms 139.4666 Ops/s 138.9442 Ops/s $\color{#35bf28}+0.38\%$
test_redq_deprec_speed[reduce-overhead-None] 3.8711ms 3.2866ms 304.2695 Ops/s 307.2698 Ops/s $\color{#d91a1a}-0.98\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.5429ms 7.2627ms 137.6899 Ops/s 139.3152 Ops/s $\color{#d91a1a}-1.17\%$
test_td3_speed[False-None] 7.6663ms 7.4705ms 133.8601 Ops/s 132.1194 Ops/s $\color{#35bf28}+1.32\%$
test_td3_speed[False-backward] 10.8263ms 10.3916ms 96.2313 Ops/s 95.6452 Ops/s $\color{#35bf28}+0.61\%$
test_td3_speed[True-None] 1.9673ms 1.9297ms 518.2170 Ops/s 523.4361 Ops/s $\color{#d91a1a}-1.00\%$
test_td3_speed[True-backward] 4.0279ms 3.7731ms 265.0368 Ops/s 268.5672 Ops/s $\color{#d91a1a}-1.31\%$
test_td3_speed[reduce-overhead-None] 1.9998ms 1.9296ms 518.2322 Ops/s 521.3885 Ops/s $\color{#d91a1a}-0.61\%$
test_td3_speed[reduce-overhead-backward] 4.0234ms 3.7273ms 268.2923 Ops/s 264.1550 Ops/s $\color{#35bf28}+1.57\%$
test_cql_speed[False-None] 0.2762s 31.4303ms 31.8165 Ops/s 40.0651 Ops/s $\textbf{\color{#d91a1a}-20.59\%}$
test_cql_speed[False-backward] 38.9572ms 34.7904ms 28.7436 Ops/s 29.2101 Ops/s $\color{#d91a1a}-1.60\%$
test_cql_speed[True-None] 11.5737ms 11.0512ms 90.4881 Ops/s 91.2982 Ops/s $\color{#d91a1a}-0.89\%$
test_cql_speed[True-backward] 17.6352ms 16.9198ms 59.1025 Ops/s 61.1047 Ops/s $\color{#d91a1a}-3.28\%$
test_cql_speed[reduce-overhead-None] 11.6016ms 11.0661ms 90.3661 Ops/s 94.3316 Ops/s $\color{#d91a1a}-4.20\%$
test_cql_speed[reduce-overhead-backward] 17.9185ms 16.9525ms 58.9882 Ops/s 61.1745 Ops/s $\color{#d91a1a}-3.57\%$
test_a2c_speed[False-None] 7.5545ms 5.3337ms 187.4857 Ops/s 189.5507 Ops/s $\color{#d91a1a}-1.09\%$
test_a2c_speed[False-backward] 12.2133ms 11.8846ms 84.1422 Ops/s 85.0989 Ops/s $\color{#d91a1a}-1.12\%$
test_a2c_speed[True-None] 3.2392ms 3.0498ms 327.8916 Ops/s 322.2461 Ops/s $\color{#35bf28}+1.75\%$
test_a2c_speed[True-backward] 8.9040ms 8.6964ms 114.9900 Ops/s 116.3096 Ops/s $\color{#d91a1a}-1.13\%$
test_a2c_speed[reduce-overhead-None] 3.2957ms 3.0754ms 325.1563 Ops/s 328.7817 Ops/s $\color{#d91a1a}-1.10\%$
test_a2c_speed[reduce-overhead-backward] 9.0323ms 8.5919ms 116.3882 Ops/s 117.7570 Ops/s $\color{#d91a1a}-1.16\%$
test_ppo_speed[False-None] 5.9703ms 5.6464ms 177.1043 Ops/s 178.3613 Ops/s $\color{#d91a1a}-0.70\%$
test_ppo_speed[False-backward] 12.8115ms 12.3965ms 80.6676 Ops/s 82.7473 Ops/s $\color{#d91a1a}-2.51\%$
test_ppo_speed[True-None] 3.7471ms 3.5166ms 284.3695 Ops/s 285.2469 Ops/s $\color{#d91a1a}-0.31\%$
test_ppo_speed[True-backward] 9.2106ms 8.3290ms 120.0631 Ops/s 117.1113 Ops/s $\color{#35bf28}+2.52\%$
test_ppo_speed[reduce-overhead-None] 3.8194ms 3.4748ms 287.7869 Ops/s 288.7937 Ops/s $\color{#d91a1a}-0.35\%$
test_ppo_speed[reduce-overhead-backward] 8.6833ms 8.3258ms 120.1089 Ops/s 120.0783 Ops/s $\color{#35bf28}+0.03\%$
test_reinforce_speed[False-None] 4.8199ms 4.4786ms 223.2817 Ops/s 226.2963 Ops/s $\color{#d91a1a}-1.33\%$
test_reinforce_speed[False-backward] 8.9760ms 7.2939ms 137.1017 Ops/s 138.3477 Ops/s $\color{#d91a1a}-0.90\%$
test_reinforce_speed[True-None] 2.6263ms 2.2186ms 450.7343 Ops/s 459.0786 Ops/s $\color{#d91a1a}-1.82\%$
test_reinforce_speed[True-backward] 7.6423ms 7.1954ms 138.9768 Ops/s 140.9072 Ops/s $\color{#d91a1a}-1.37\%$
test_reinforce_speed[reduce-overhead-None] 2.4493ms 2.2297ms 448.4969 Ops/s 450.6393 Ops/s $\color{#d91a1a}-0.48\%$
test_reinforce_speed[reduce-overhead-backward] 7.3309ms 7.1375ms 140.1060 Ops/s 140.3408 Ops/s $\color{#d91a1a}-0.17\%$
test_iql_speed[False-None] 20.1466ms 19.5056ms 51.2674 Ops/s 52.0981 Ops/s $\color{#d91a1a}-1.59\%$
test_iql_speed[False-backward] 31.1439ms 30.3849ms 32.9111 Ops/s 33.6673 Ops/s $\color{#d91a1a}-2.25\%$
test_iql_speed[True-None] 7.7982ms 6.8066ms 146.9171 Ops/s 144.9194 Ops/s $\color{#35bf28}+1.38\%$
test_iql_speed[True-backward] 16.1596ms 15.5949ms 64.1236 Ops/s 62.4855 Ops/s $\color{#35bf28}+2.62\%$
test_iql_speed[reduce-overhead-None] 7.1577ms 6.8352ms 146.3016 Ops/s 148.1156 Ops/s $\color{#d91a1a}-1.22\%$
test_iql_speed[reduce-overhead-backward] 16.0026ms 15.6479ms 63.9062 Ops/s 63.3729 Ops/s $\color{#35bf28}+0.84\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2677ms 6.0114ms 166.3514 Ops/s 166.4202 Ops/s $\color{#d91a1a}-0.04\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.2354ms 0.2733ms 3.6584 KOps/s 3.2442 KOps/s $\textbf{\color{#35bf28}+12.77\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5085ms 0.2709ms 3.6921 KOps/s 3.2898 KOps/s $\textbf{\color{#35bf28}+12.23\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.3391ms 5.9025ms 169.4185 Ops/s 170.6336 Ops/s $\color{#d91a1a}-0.71\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.8045ms 0.2677ms 3.7353 KOps/s 3.6383 KOps/s $\color{#35bf28}+2.67\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6803ms 0.2554ms 3.9159 KOps/s 4.6876 KOps/s $\textbf{\color{#d91a1a}-16.46\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.5781ms 1.3484ms 741.5940 Ops/s 828.8831 Ops/s $\textbf{\color{#d91a1a}-10.53\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5754ms 1.2883ms 776.1991 Ops/s 778.1405 Ops/s $\color{#d91a1a}-0.25\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.3383ms 6.0892ms 164.2255 Ops/s 166.1483 Ops/s $\color{#d91a1a}-1.16\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.2701ms 0.4349ms 2.2993 KOps/s 2.5656 KOps/s $\textbf{\color{#d91a1a}-10.38\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6500ms 0.4155ms 2.4066 KOps/s 2.8078 KOps/s $\textbf{\color{#d91a1a}-14.29\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.0730ms 5.8870ms 169.8658 Ops/s 170.0150 Ops/s $\color{#d91a1a}-0.09\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8400ms 0.3174ms 3.1510 KOps/s 3.0706 KOps/s $\color{#35bf28}+2.62\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4671ms 0.2442ms 4.0956 KOps/s 4.7055 KOps/s $\textbf{\color{#d91a1a}-12.96\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.3064ms 5.9278ms 168.6980 Ops/s 172.0536 Ops/s $\color{#d91a1a}-1.95\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.4171s 0.6704ms 1.4916 KOps/s 676.3364 Ops/s $\textbf{\color{#35bf28}+120.54\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5000ms 0.2535ms 3.9445 KOps/s 2.9250 KOps/s $\textbf{\color{#35bf28}+34.85\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4241ms 6.1091ms 163.6891 Ops/s 161.6954 Ops/s $\color{#35bf28}+1.23\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.9392ms 0.3924ms 2.5481 KOps/s 1.8136 KOps/s $\textbf{\color{#35bf28}+40.50\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7735ms 0.3727ms 2.6828 KOps/s 2.0431 KOps/s $\textbf{\color{#35bf28}+31.31\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.1085ms 5.2417ms 190.7794 Ops/s 190.8388 Ops/s $\color{#d91a1a}-0.03\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 10.9463ms 2.0546ms 486.7149 Ops/s 473.8151 Ops/s $\color{#35bf28}+2.72\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 2.0926ms 1.0635ms 940.2980 Ops/s 941.0603 Ops/s $\color{#d91a1a}-0.08\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.3966s 13.1279ms 76.1736 Ops/s 190.0918 Ops/s $\textbf{\color{#d91a1a}-59.93\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 7.1953ms 1.9646ms 508.9979 Ops/s 506.6974 Ops/s $\color{#35bf28}+0.45\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 8.2081ms 1.2326ms 811.2742 Ops/s 810.7470 Ops/s $\color{#35bf28}+0.07\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 8.6131ms 5.4167ms 184.6158 Ops/s 182.6706 Ops/s $\color{#35bf28}+1.06\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.3382ms 2.1613ms 462.6799 Ops/s 463.0249 Ops/s $\color{#d91a1a}-0.07\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.6799ms 1.3240ms 755.3014 Ops/s 744.2585 Ops/s $\color{#35bf28}+1.48\%$

@vmoens vmoens changed the title [Doc] Better doc for ditributed RBs [Doc] Better doc for distributed RBs Oct 11, 2024
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.
Projects
None yet
Milestone
No milestone
Development

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