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

[Doc] Fix wrong default value for flatten_tensordicts in ReplayBufferTrainer #2502

Merged
merged 1 commit into from
Oct 18, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Oct 18, 2024

No description provided.

Copy link

pytorch-bot bot commented Oct 18, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2502

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

❌ 18 New Failures, 4 Unrelated Failures

As of commit 7a132dc with merge base 815eece (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 Oct 18, 2024
@vmoens vmoens linked an issue Oct 18, 2024 that may be closed by this pull request
3 tasks
@vmoens vmoens added the documentation Improvements or additions to documentation label Oct 18, 2024
@vmoens vmoens merged commit 30df6d9 into main Oct 18, 2024
16 of 28 checks passed
@vmoens vmoens deleted the vmoens-patch-1 branch October 18, 2024 11:54
Copy link

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}3$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.4096s 0.4085s 2.4480 Ops/s 2.3388 Ops/s $\color{#35bf28}+4.67\%$
test_transformed 0.6721s 0.5993s 1.6685 Ops/s 1.6658 Ops/s $\color{#35bf28}+0.16\%$
test_serial 1.4102s 1.3460s 0.7429 Ops/s 0.7489 Ops/s $\color{#d91a1a}-0.80\%$
test_parallel 1.3829s 1.3122s 0.7621 Ops/s 0.7474 Ops/s $\color{#35bf28}+1.96\%$
test_step_mdp_speed[True-True-True-True-True] 0.1962ms 29.1753μs 34.2756 KOps/s 34.7622 KOps/s $\color{#d91a1a}-1.40\%$
test_step_mdp_speed[True-True-True-True-False] 41.0770μs 17.2032μs 58.1286 KOps/s 57.8379 KOps/s $\color{#35bf28}+0.50\%$
test_step_mdp_speed[True-True-True-False-True] 64.2300μs 16.1616μs 61.8749 KOps/s 61.7817 KOps/s $\color{#35bf28}+0.15\%$
test_step_mdp_speed[True-True-True-False-False] 46.8880μs 9.3169μs 107.3316 KOps/s 104.1111 KOps/s $\color{#35bf28}+3.09\%$
test_step_mdp_speed[True-True-False-True-True] 70.4720μs 31.2954μs 31.9536 KOps/s 32.0494 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[True-True-False-True-False] 51.9770μs 19.3406μs 51.7046 KOps/s 51.6846 KOps/s $\color{#35bf28}+0.04\%$
test_step_mdp_speed[True-True-False-False-True] 52.7990μs 18.2831μs 54.6954 KOps/s 55.1662 KOps/s $\color{#d91a1a}-0.85\%$
test_step_mdp_speed[True-True-False-False-False] 42.4500μs 11.6764μs 85.6431 KOps/s 84.9218 KOps/s $\color{#35bf28}+0.85\%$
test_step_mdp_speed[True-False-True-True-True] 73.8580μs 32.9518μs 30.3473 KOps/s 30.3810 KOps/s $\color{#d91a1a}-0.11\%$
test_step_mdp_speed[True-False-True-True-False] 58.7200μs 21.4940μs 46.5246 KOps/s 46.8486 KOps/s $\color{#d91a1a}-0.69\%$
test_step_mdp_speed[True-False-True-False-True] 51.0660μs 17.8813μs 55.9245 KOps/s 55.0873 KOps/s $\color{#35bf28}+1.52\%$
test_step_mdp_speed[True-False-True-False-False] 41.2570μs 11.6942μs 85.5125 KOps/s 85.5835 KOps/s $\color{#d91a1a}-0.08\%$
test_step_mdp_speed[True-False-False-True-True] 78.3370μs 35.0655μs 28.5181 KOps/s 28.5889 KOps/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[True-False-False-True-False] 74.3490μs 23.5590μs 42.4466 KOps/s 43.4030 KOps/s $\color{#d91a1a}-2.20\%$
test_step_mdp_speed[True-False-False-False-True] 51.0850μs 19.9554μs 50.1116 KOps/s 49.0097 KOps/s $\color{#35bf28}+2.25\%$
test_step_mdp_speed[True-False-False-False-False] 41.6580μs 13.6455μs 73.2841 KOps/s 74.2532 KOps/s $\color{#d91a1a}-1.31\%$
test_step_mdp_speed[False-True-True-True-True] 88.1960μs 33.3640μs 29.9724 KOps/s 30.1995 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[False-True-True-True-False] 63.5790μs 21.3701μs 46.7943 KOps/s 46.9617 KOps/s $\color{#d91a1a}-0.36\%$
test_step_mdp_speed[False-True-True-False-True] 56.6960μs 21.4580μs 46.6026 KOps/s 46.5453 KOps/s $\color{#35bf28}+0.12\%$
test_step_mdp_speed[False-True-True-False-False] 44.1720μs 13.2453μs 75.4982 KOps/s 74.9543 KOps/s $\color{#35bf28}+0.73\%$
test_step_mdp_speed[False-True-False-True-True] 96.1300μs 34.4052μs 29.0654 KOps/s 28.2140 KOps/s $\color{#35bf28}+3.02\%$
test_step_mdp_speed[False-True-False-True-False] 52.4890μs 23.0923μs 43.3045 KOps/s 42.5891 KOps/s $\color{#35bf28}+1.68\%$
test_step_mdp_speed[False-True-False-False-True] 2.6189ms 23.2794μs 42.9565 KOps/s 42.6859 KOps/s $\color{#35bf28}+0.63\%$
test_step_mdp_speed[False-True-False-False-False] 39.7740μs 15.1080μs 66.1901 KOps/s 65.0150 KOps/s $\color{#35bf28}+1.81\%$
test_step_mdp_speed[False-False-True-True-True] 0.1006ms 36.5665μs 27.3474 KOps/s 27.0000 KOps/s $\color{#35bf28}+1.29\%$
test_step_mdp_speed[False-False-True-True-False] 62.6080μs 25.4955μs 39.2227 KOps/s 39.4221 KOps/s $\color{#d91a1a}-0.51\%$
test_step_mdp_speed[False-False-True-False-True] 50.4350μs 23.1388μs 43.2175 KOps/s 43.3242 KOps/s $\color{#d91a1a}-0.25\%$
test_step_mdp_speed[False-False-True-False-False] 51.8770μs 15.2715μs 65.4813 KOps/s 64.8715 KOps/s $\color{#35bf28}+0.94\%$
test_step_mdp_speed[False-False-False-True-True] 82.5540μs 38.1639μs 26.2028 KOps/s 24.9623 KOps/s $\color{#35bf28}+4.97\%$
test_step_mdp_speed[False-False-False-True-False] 0.1065ms 27.1114μs 36.8848 KOps/s 36.6781 KOps/s $\color{#35bf28}+0.56\%$
test_step_mdp_speed[False-False-False-False-True] 54.6220μs 24.5273μs 40.7708 KOps/s 40.0611 KOps/s $\color{#35bf28}+1.77\%$
test_step_mdp_speed[False-False-False-False-False] 64.9430μs 16.9088μs 59.1407 KOps/s 58.1944 KOps/s $\color{#35bf28}+1.63\%$
test_values[generalized_advantage_estimate-True-True] 13.4911ms 9.5899ms 104.2762 Ops/s 105.0743 Ops/s $\color{#d91a1a}-0.76\%$
test_values[vec_generalized_advantage_estimate-True-True] 36.1072ms 33.4620ms 29.8847 Ops/s 29.7849 Ops/s $\color{#35bf28}+0.33\%$
test_values[td0_return_estimate-False-False] 0.2240ms 0.1659ms 6.0293 KOps/s 5.4312 KOps/s $\textbf{\color{#35bf28}+11.01\%}$
test_values[td1_return_estimate-False-False] 26.6052ms 23.7203ms 42.1579 Ops/s 40.9289 Ops/s $\color{#35bf28}+3.00\%$
test_values[vec_td1_return_estimate-False-False] 36.1159ms 33.5625ms 29.7952 Ops/s 29.8209 Ops/s $\color{#d91a1a}-0.09\%$
test_values[td_lambda_return_estimate-True-False] 42.3612ms 34.4325ms 29.0423 Ops/s 28.4941 Ops/s $\color{#35bf28}+1.92\%$
test_values[vec_td_lambda_return_estimate-True-False] 36.0294ms 33.6657ms 29.7039 Ops/s 29.8300 Ops/s $\color{#d91a1a}-0.42\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 9.1392ms 8.3260ms 120.1057 Ops/s 118.7361 Ops/s $\color{#35bf28}+1.15\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4520ms 1.9434ms 514.5688 Ops/s 556.7032 Ops/s $\textbf{\color{#d91a1a}-7.57\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.4874ms 0.3539ms 2.8256 KOps/s 2.7816 KOps/s $\color{#35bf28}+1.58\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 46.2431ms 43.5344ms 22.9703 Ops/s 25.3788 Ops/s $\textbf{\color{#d91a1a}-9.49\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 3.8397ms 3.0386ms 329.0960 Ops/s 328.3424 Ops/s $\color{#35bf28}+0.23\%$
test_dqn_speed[False-None] 6.0834ms 1.3685ms 730.7454 Ops/s 713.7064 Ops/s $\color{#35bf28}+2.39\%$
test_dqn_speed[False-backward] 1.9431ms 1.8163ms 550.5547 Ops/s 545.1812 Ops/s $\color{#35bf28}+0.99\%$
test_dqn_speed[True-None] 0.7715ms 0.4586ms 2.1804 KOps/s 2.1537 KOps/s $\color{#35bf28}+1.24\%$
test_dqn_speed[True-backward] 0.9145ms 0.8671ms 1.1533 KOps/s 1.1058 KOps/s $\color{#35bf28}+4.29\%$
test_dqn_speed[reduce-overhead-None] 0.7696ms 0.4830ms 2.0704 KOps/s 2.1270 KOps/s $\color{#d91a1a}-2.66\%$
test_dqn_speed[reduce-overhead-backward] 1.2882ms 0.9384ms 1.0656 KOps/s 1.1305 KOps/s $\textbf{\color{#d91a1a}-5.74\%}$
test_ddpg_speed[False-None] 3.4621ms 2.7837ms 359.2303 Ops/s 356.2576 Ops/s $\color{#35bf28}+0.83\%$
test_ddpg_speed[False-backward] 4.2042ms 3.9007ms 256.3624 Ops/s 254.5787 Ops/s $\color{#35bf28}+0.70\%$
test_ddpg_speed[True-None] 1.5060ms 0.9952ms 1.0049 KOps/s 1.0000 KOps/s $\color{#35bf28}+0.48\%$
test_ddpg_speed[True-backward] 2.0447ms 1.8907ms 528.8954 Ops/s 527.1637 Ops/s $\color{#35bf28}+0.33\%$
test_ddpg_speed[reduce-overhead-None] 1.6856ms 0.9954ms 1.0046 KOps/s 978.0424 Ops/s $\color{#35bf28}+2.72\%$
test_ddpg_speed[reduce-overhead-backward] 1.9819ms 1.8737ms 533.6968 Ops/s 524.2670 Ops/s $\color{#35bf28}+1.80\%$
test_sac_speed[False-None] 11.5792ms 7.9989ms 125.0176 Ops/s 125.9825 Ops/s $\color{#d91a1a}-0.77\%$
test_sac_speed[False-backward] 11.0169ms 10.6439ms 93.9510 Ops/s 93.6933 Ops/s $\color{#35bf28}+0.27\%$
test_sac_speed[True-None] 2.1281ms 1.8281ms 547.0258 Ops/s 531.4039 Ops/s $\color{#35bf28}+2.94\%$
test_sac_speed[True-backward] 3.5741ms 3.4829ms 287.1168 Ops/s 281.3841 Ops/s $\color{#35bf28}+2.04\%$
test_sac_speed[reduce-overhead-None] 2.1053ms 1.8559ms 538.8128 Ops/s 538.9060 Ops/s $\color{#d91a1a}-0.02\%$
test_sac_speed[reduce-overhead-backward] 3.6121ms 3.4985ms 285.8372 Ops/s 278.8452 Ops/s $\color{#35bf28}+2.51\%$
test_redq_speed[False-None] 14.2806ms 12.6070ms 79.3208 Ops/s 76.6994 Ops/s $\color{#35bf28}+3.42\%$
test_redq_speed[False-backward] 23.3685ms 21.8761ms 45.7119 Ops/s 44.9710 Ops/s $\color{#35bf28}+1.65\%$
test_redq_speed[True-None] 5.4719ms 4.5499ms 219.7861 Ops/s 217.1833 Ops/s $\color{#35bf28}+1.20\%$
test_redq_speed[True-backward] 12.9128ms 11.9792ms 83.4784 Ops/s 80.4311 Ops/s $\color{#35bf28}+3.79\%$
test_redq_speed[reduce-overhead-None] 5.4777ms 4.5490ms 219.8271 Ops/s 215.6387 Ops/s $\color{#35bf28}+1.94\%$
test_redq_speed[reduce-overhead-backward] 12.7602ms 12.0706ms 82.8461 Ops/s 83.2949 Ops/s $\color{#d91a1a}-0.54\%$
test_redq_deprec_speed[False-None] 15.1468ms 12.7445ms 78.4651 Ops/s 78.5494 Ops/s $\color{#d91a1a}-0.11\%$
test_redq_deprec_speed[False-backward] 19.3068ms 18.5426ms 53.9299 Ops/s 54.2677 Ops/s $\color{#d91a1a}-0.62\%$
test_redq_deprec_speed[True-None] 4.1022ms 3.5442ms 282.1537 Ops/s 279.8555 Ops/s $\color{#35bf28}+0.82\%$
test_redq_deprec_speed[True-backward] 9.7175ms 8.3434ms 119.8557 Ops/s 123.7023 Ops/s $\color{#d91a1a}-3.11\%$
test_redq_deprec_speed[reduce-overhead-None] 3.9026ms 3.5478ms 281.8680 Ops/s 278.7280 Ops/s $\color{#35bf28}+1.13\%$
test_redq_deprec_speed[reduce-overhead-backward] 9.0176ms 8.0087ms 124.8643 Ops/s 125.0445 Ops/s $\color{#d91a1a}-0.14\%$
test_td3_speed[False-None] 8.1434ms 7.8222ms 127.8415 Ops/s 125.8283 Ops/s $\color{#35bf28}+1.60\%$
test_td3_speed[False-backward] 11.3940ms 10.1954ms 98.0835 Ops/s 47.6007 Ops/s $\textbf{\color{#35bf28}+106.05\%}$
test_td3_speed[True-None] 1.8407ms 1.7321ms 577.3238 Ops/s 568.5040 Ops/s $\color{#35bf28}+1.55\%$
test_td3_speed[True-backward] 3.6357ms 3.3500ms 298.5106 Ops/s 301.0729 Ops/s $\color{#d91a1a}-0.85\%$
test_td3_speed[reduce-overhead-None] 1.9480ms 1.7312ms 577.6295 Ops/s 573.3818 Ops/s $\color{#35bf28}+0.74\%$
test_td3_speed[reduce-overhead-backward] 3.4562ms 3.3421ms 299.2132 Ops/s 298.8831 Ops/s $\color{#35bf28}+0.11\%$
test_cql_speed[False-None] 38.3692ms 35.9311ms 27.8310 Ops/s 27.8276 Ops/s $\color{#35bf28}+0.01\%$
test_cql_speed[False-backward] 54.5231ms 47.1225ms 21.2213 Ops/s 21.7977 Ops/s $\color{#d91a1a}-2.64\%$
test_cql_speed[True-None] 16.9371ms 15.5359ms 64.3670 Ops/s 62.6967 Ops/s $\color{#35bf28}+2.66\%$
test_cql_speed[True-backward] 23.3214ms 22.0976ms 45.2538 Ops/s 44.7810 Ops/s $\color{#35bf28}+1.06\%$
test_cql_speed[reduce-overhead-None] 16.2978ms 15.5161ms 64.4492 Ops/s 64.3943 Ops/s $\color{#35bf28}+0.09\%$
test_cql_speed[reduce-overhead-backward] 25.8068ms 22.1619ms 45.1225 Ops/s 44.5036 Ops/s $\color{#35bf28}+1.39\%$
test_a2c_speed[False-None] 9.2049ms 7.0950ms 140.9443 Ops/s 138.1590 Ops/s $\color{#35bf28}+2.02\%$
test_a2c_speed[False-backward] 14.5055ms 14.0666ms 71.0906 Ops/s 70.0394 Ops/s $\color{#35bf28}+1.50\%$
test_a2c_speed[True-None] 4.5171ms 3.3563ms 297.9489 Ops/s 300.2118 Ops/s $\color{#d91a1a}-0.75\%$
test_a2c_speed[True-backward] 10.1760ms 9.7477ms 102.5880 Ops/s 101.7349 Ops/s $\color{#35bf28}+0.84\%$
test_a2c_speed[reduce-overhead-None] 3.6054ms 3.3094ms 302.1736 Ops/s 298.2480 Ops/s $\color{#35bf28}+1.32\%$
test_a2c_speed[reduce-overhead-backward] 10.4621ms 9.7374ms 102.6971 Ops/s 102.9955 Ops/s $\color{#d91a1a}-0.29\%$
test_ppo_speed[False-None] 8.4435ms 7.3427ms 136.1900 Ops/s 133.3334 Ops/s $\color{#35bf28}+2.14\%$
test_ppo_speed[False-backward] 14.8800ms 14.5430ms 68.7616 Ops/s 68.4162 Ops/s $\color{#35bf28}+0.50\%$
test_ppo_speed[True-None] 4.0272ms 3.7222ms 268.6616 Ops/s 268.6816 Ops/s $-0.01\%$
test_ppo_speed[True-backward] 9.9764ms 9.5877ms 104.3001 Ops/s 98.2937 Ops/s $\textbf{\color{#35bf28}+6.11\%}$
test_ppo_speed[reduce-overhead-None] 3.8050ms 3.7006ms 270.2294 Ops/s 268.7693 Ops/s $\color{#35bf28}+0.54\%$
test_ppo_speed[reduce-overhead-backward] 10.1101ms 9.6175ms 103.9773 Ops/s 102.6462 Ops/s $\color{#35bf28}+1.30\%$
test_reinforce_speed[False-None] 7.7305ms 6.4556ms 154.9044 Ops/s 152.8900 Ops/s $\color{#35bf28}+1.32\%$
test_reinforce_speed[False-backward] 10.0657ms 9.6892ms 103.2073 Ops/s 101.9205 Ops/s $\color{#35bf28}+1.26\%$
test_reinforce_speed[True-None] 3.0052ms 2.6360ms 379.3575 Ops/s 367.6727 Ops/s $\color{#35bf28}+3.18\%$
test_reinforce_speed[True-backward] 8.8495ms 8.5390ms 117.1103 Ops/s 115.3133 Ops/s $\color{#35bf28}+1.56\%$
test_reinforce_speed[reduce-overhead-None] 3.2477ms 2.6564ms 376.4489 Ops/s 374.4341 Ops/s $\color{#35bf28}+0.54\%$
test_reinforce_speed[reduce-overhead-backward] 9.0390ms 8.5656ms 116.7457 Ops/s 115.8605 Ops/s $\color{#35bf28}+0.76\%$
test_iql_speed[False-None] 33.8042ms 31.9655ms 31.2837 Ops/s 30.8805 Ops/s $\color{#35bf28}+1.31\%$
test_iql_speed[False-backward] 47.1977ms 45.1039ms 22.1710 Ops/s 22.0709 Ops/s $\color{#35bf28}+0.45\%$
test_iql_speed[True-None] 11.9646ms 10.5209ms 95.0487 Ops/s 93.5218 Ops/s $\color{#35bf28}+1.63\%$
test_iql_speed[True-backward] 22.4154ms 21.5801ms 46.3391 Ops/s 45.7881 Ops/s $\color{#35bf28}+1.20\%$
test_iql_speed[reduce-overhead-None] 11.6260ms 10.5889ms 94.4388 Ops/s 91.4622 Ops/s $\color{#35bf28}+3.25\%$
test_iql_speed[reduce-overhead-backward] 22.8151ms 21.8580ms 45.7498 Ops/s 45.6894 Ops/s $\color{#35bf28}+0.13\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 5.0957ms 4.7155ms 212.0685 Ops/s 211.3871 Ops/s $\color{#35bf28}+0.32\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 2.4916ms 0.4837ms 2.0672 KOps/s 2.0932 KOps/s $\color{#d91a1a}-1.24\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6920ms 0.4551ms 2.1971 KOps/s 2.1687 KOps/s $\color{#35bf28}+1.31\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.4418ms 4.6197ms 216.4623 Ops/s 214.2494 Ops/s $\color{#35bf28}+1.03\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1110ms 0.4731ms 2.1139 KOps/s 2.1156 KOps/s $\color{#d91a1a}-0.08\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7004ms 0.4543ms 2.2012 KOps/s 2.2008 KOps/s $\color{#35bf28}+0.02\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 2.2297ms 1.6073ms 622.1485 Ops/s 627.9330 Ops/s $\color{#d91a1a}-0.92\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.8014ms 1.5503ms 645.0193 Ops/s 645.9962 Ops/s $\color{#d91a1a}-0.15\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 12.0021ms 4.9097ms 203.6795 Ops/s 206.8107 Ops/s $\color{#d91a1a}-1.51\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.0990ms 0.6197ms 1.6138 KOps/s 1.6294 KOps/s $\color{#d91a1a}-0.96\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8092ms 0.5876ms 1.7017 KOps/s 1.6956 KOps/s $\color{#35bf28}+0.36\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.1501ms 4.6744ms 213.9325 Ops/s 214.8964 Ops/s $\color{#d91a1a}-0.45\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.6938ms 0.4805ms 2.0813 KOps/s 2.0795 KOps/s $\color{#35bf28}+0.09\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 7.5665ms 0.4714ms 2.1214 KOps/s 2.1454 KOps/s $\color{#d91a1a}-1.12\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.7931ms 4.6458ms 215.2461 Ops/s 215.8664 Ops/s $\color{#d91a1a}-0.29\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1312ms 0.4801ms 2.0831 KOps/s 2.1293 KOps/s $\color{#d91a1a}-2.17\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.7102ms 0.4539ms 2.2030 KOps/s 2.1839 KOps/s $\color{#35bf28}+0.87\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.4815ms 4.8196ms 207.4873 Ops/s 205.3509 Ops/s $\color{#35bf28}+1.04\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.4521ms 0.6161ms 1.6230 KOps/s 1.6143 KOps/s $\color{#35bf28}+0.54\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8524ms 0.5948ms 1.6813 KOps/s 1.6736 KOps/s $\color{#35bf28}+0.46\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.4165ms 4.1796ms 239.2594 Ops/s 235.2211 Ops/s $\color{#35bf28}+1.72\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.8796ms 2.3485ms 425.8084 Ops/s 431.5716 Ops/s $\color{#d91a1a}-1.34\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.4612ms 1.3232ms 755.7565 Ops/s 830.2541 Ops/s $\textbf{\color{#d91a1a}-8.97\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.3901s 11.9601ms 83.6117 Ops/s 235.5197 Ops/s $\textbf{\color{#d91a1a}-64.50\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 5.1974ms 2.2684ms 440.8362 Ops/s 450.2660 Ops/s $\color{#d91a1a}-2.09\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 6.4755ms 1.3161ms 759.8001 Ops/s 740.1821 Ops/s $\color{#35bf28}+2.65\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 5.8841ms 4.3425ms 230.2796 Ops/s 223.8764 Ops/s $\color{#35bf28}+2.86\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 6.7573ms 2.4668ms 405.3903 Ops/s 404.3139 Ops/s $\color{#35bf28}+0.27\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 12.5995ms 1.6155ms 618.9986 Ops/s 664.9875 Ops/s $\textbf{\color{#d91a1a}-6.92\%}$

Copy link

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

Total Benchmarks: 143. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.7486s 0.7412s 1.3491 Ops/s 1.3669 Ops/s $\color{#d91a1a}-1.30\%$
test_transformed 1.0860s 1.0020s 0.9980 Ops/s 1.0290 Ops/s $\color{#d91a1a}-3.02\%$
test_serial 2.2500s 2.1670s 0.4615 Ops/s 0.4717 Ops/s $\color{#d91a1a}-2.16\%$
test_parallel 2.0801s 2.0214s 0.4947 Ops/s 0.5013 Ops/s $\color{#d91a1a}-1.30\%$
test_step_mdp_speed[True-True-True-True-True] 0.1377ms 39.7206μs 25.1759 KOps/s 24.9224 KOps/s $\color{#35bf28}+1.02\%$
test_step_mdp_speed[True-True-True-True-False] 0.1653ms 23.4363μs 42.6688 KOps/s 43.0491 KOps/s $\color{#d91a1a}-0.88\%$
test_step_mdp_speed[True-True-True-False-True] 53.0310μs 21.6303μs 46.2314 KOps/s 45.7983 KOps/s $\color{#35bf28}+0.95\%$
test_step_mdp_speed[True-True-True-False-False] 40.1410μs 12.8453μs 77.8496 KOps/s 80.0853 KOps/s $\color{#d91a1a}-2.79\%$
test_step_mdp_speed[True-True-False-True-True] 0.1052ms 43.5008μs 22.9881 KOps/s 23.0744 KOps/s $\color{#d91a1a}-0.37\%$
test_step_mdp_speed[True-True-False-True-False] 57.4920μs 25.5199μs 39.1852 KOps/s 38.5855 KOps/s $\color{#35bf28}+1.55\%$
test_step_mdp_speed[True-True-False-False-True] 54.7820μs 24.6214μs 40.6151 KOps/s 40.7843 KOps/s $\color{#d91a1a}-0.41\%$
test_step_mdp_speed[True-True-False-False-False] 66.4810μs 15.3829μs 65.0070 KOps/s 67.6139 KOps/s $\color{#d91a1a}-3.86\%$
test_step_mdp_speed[True-False-True-True-True] 0.2313ms 46.1352μs 21.6754 KOps/s 22.0867 KOps/s $\color{#d91a1a}-1.86\%$
test_step_mdp_speed[True-False-True-True-False] 0.2184ms 28.4366μs 35.1660 KOps/s 36.0190 KOps/s $\color{#d91a1a}-2.37\%$
test_step_mdp_speed[True-False-True-False-True] 0.2207ms 24.4706μs 40.8654 KOps/s 43.1361 KOps/s $\textbf{\color{#d91a1a}-5.26\%}$
test_step_mdp_speed[True-False-True-False-False] 75.4220μs 15.1797μs 65.8775 KOps/s 66.4017 KOps/s $\color{#d91a1a}-0.79\%$
test_step_mdp_speed[True-False-False-True-True] 0.1714ms 47.5451μs 21.0327 KOps/s 20.8191 KOps/s $\color{#35bf28}+1.03\%$
test_step_mdp_speed[True-False-False-True-False] 69.7620μs 30.9116μs 32.3503 KOps/s 32.3557 KOps/s $\color{#d91a1a}-0.02\%$
test_step_mdp_speed[True-False-False-False-True] 76.9620μs 27.2336μs 36.7194 KOps/s 37.4867 KOps/s $\color{#d91a1a}-2.05\%$
test_step_mdp_speed[True-False-False-False-False] 44.5120μs 17.8563μs 56.0028 KOps/s 56.5333 KOps/s $\color{#d91a1a}-0.94\%$
test_step_mdp_speed[False-True-True-True-True] 79.3120μs 45.1575μs 22.1447 KOps/s 22.0743 KOps/s $\color{#35bf28}+0.32\%$
test_step_mdp_speed[False-True-True-True-False] 72.3220μs 28.5658μs 35.0068 KOps/s 34.8955 KOps/s $\color{#35bf28}+0.32\%$
test_step_mdp_speed[False-True-True-False-True] 63.8020μs 29.9530μs 33.3856 KOps/s 34.2837 KOps/s $\color{#d91a1a}-2.62\%$
test_step_mdp_speed[False-True-True-False-False] 52.2310μs 17.9290μs 55.7756 KOps/s 55.1878 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[False-True-False-True-True] 78.8730μs 48.0474μs 20.8128 KOps/s 20.9151 KOps/s $\color{#d91a1a}-0.49\%$
test_step_mdp_speed[False-True-False-True-False] 0.1513ms 31.3711μs 31.8765 KOps/s 32.4780 KOps/s $\color{#d91a1a}-1.85\%$
test_step_mdp_speed[False-True-False-False-True] 3.4227ms 32.4611μs 30.8061 KOps/s 30.9780 KOps/s $\color{#d91a1a}-0.55\%$
test_step_mdp_speed[False-True-False-False-False] 49.6620μs 20.6766μs 48.3637 KOps/s 49.5647 KOps/s $\color{#d91a1a}-2.42\%$
test_step_mdp_speed[False-False-True-True-True] 86.8140μs 50.0366μs 19.9854 KOps/s 19.7055 KOps/s $\color{#35bf28}+1.42\%$
test_step_mdp_speed[False-False-True-True-False] 0.1001ms 33.0326μs 30.2732 KOps/s 29.2718 KOps/s $\color{#35bf28}+3.42\%$
test_step_mdp_speed[False-False-True-False-True] 69.2320μs 32.0511μs 31.2002 KOps/s 32.0448 KOps/s $\color{#d91a1a}-2.64\%$
test_step_mdp_speed[False-False-True-False-False] 70.5920μs 20.5112μs 48.7539 KOps/s 49.2541 KOps/s $\color{#d91a1a}-1.02\%$
test_step_mdp_speed[False-False-False-True-True] 0.2335ms 51.7272μs 19.3322 KOps/s 18.7267 KOps/s $\color{#35bf28}+3.23\%$
test_step_mdp_speed[False-False-False-True-False] 68.1920μs 35.8407μs 27.9013 KOps/s 27.5778 KOps/s $\color{#35bf28}+1.17\%$
test_step_mdp_speed[False-False-False-False-True] 68.8720μs 33.5643μs 29.7936 KOps/s 30.0149 KOps/s $\color{#d91a1a}-0.74\%$
test_step_mdp_speed[False-False-False-False-False] 49.9110μs 22.5779μs 44.2912 KOps/s 44.1028 KOps/s $\color{#35bf28}+0.43\%$
test_values[generalized_advantage_estimate-True-True] 25.3555ms 24.8874ms 40.1810 Ops/s 40.8483 Ops/s $\color{#d91a1a}-1.63\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1153s 3.1920ms 313.2790 Ops/s 345.4327 Ops/s $\textbf{\color{#d91a1a}-9.31\%}$
test_values[td0_return_estimate-False-False] 86.7930μs 65.7861μs 15.2008 KOps/s 15.3220 KOps/s $\color{#d91a1a}-0.79\%$
test_values[td1_return_estimate-False-False] 59.4634ms 58.2570ms 17.1653 Ops/s 18.2129 Ops/s $\textbf{\color{#d91a1a}-5.75\%}$
test_values[vec_td1_return_estimate-False-False] 1.4106ms 1.0842ms 922.3146 Ops/s 937.7863 Ops/s $\color{#d91a1a}-1.65\%$
test_values[td_lambda_return_estimate-True-False] 93.4082ms 91.4953ms 10.9295 Ops/s 11.6115 Ops/s $\textbf{\color{#d91a1a}-5.87\%}$
test_values[vec_td_lambda_return_estimate-True-False] 1.2823ms 1.0677ms 936.5685 Ops/s 943.4596 Ops/s $\color{#d91a1a}-0.73\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.2214ms 25.7737ms 38.7993 Ops/s 41.3948 Ops/s $\textbf{\color{#d91a1a}-6.27\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0651ms 0.7450ms 1.3422 KOps/s 1.3517 KOps/s $\color{#d91a1a}-0.70\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8047ms 0.6634ms 1.5074 KOps/s 1.5381 KOps/s $\color{#d91a1a}-1.99\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.6794ms 1.4821ms 674.7185 Ops/s 682.0710 Ops/s $\color{#d91a1a}-1.08\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.8628ms 0.6916ms 1.4460 KOps/s 1.4899 KOps/s $\color{#d91a1a}-2.95\%$
test_dqn_speed[False-None] 6.6362ms 1.3505ms 740.4522 Ops/s 666.3010 Ops/s $\textbf{\color{#35bf28}+11.13\%}$
test_dqn_speed[False-backward] 2.0447ms 1.8895ms 529.2275 Ops/s 525.7765 Ops/s $\color{#35bf28}+0.66\%$
test_dqn_speed[True-None] 1.3838ms 0.5695ms 1.7560 KOps/s 1.7043 KOps/s $\color{#35bf28}+3.03\%$
test_dqn_speed[True-backward] 1.0463ms 1.0085ms 991.5311 Ops/s 891.9150 Ops/s $\textbf{\color{#35bf28}+11.17\%}$
test_dqn_speed[reduce-overhead-None] 0.9065ms 0.5677ms 1.7614 KOps/s 1.6948 KOps/s $\color{#35bf28}+3.93\%$
test_dqn_speed[reduce-overhead-backward] 1.0823ms 1.0246ms 976.0195 Ops/s 975.5069 Ops/s $\color{#35bf28}+0.05\%$
test_ddpg_speed[False-None] 3.0312ms 2.7375ms 365.2928 Ops/s 362.0809 Ops/s $\color{#35bf28}+0.89\%$
test_ddpg_speed[False-backward] 4.1088ms 3.9759ms 251.5129 Ops/s 250.7107 Ops/s $\color{#35bf28}+0.32\%$
test_ddpg_speed[True-None] 1.4145ms 1.2542ms 797.3223 Ops/s 752.9979 Ops/s $\textbf{\color{#35bf28}+5.89\%}$
test_ddpg_speed[True-backward] 2.3685ms 2.2506ms 444.3309 Ops/s 438.5687 Ops/s $\color{#35bf28}+1.31\%$
test_ddpg_speed[reduce-overhead-None] 1.6257ms 1.2676ms 788.8686 Ops/s 772.3550 Ops/s $\color{#35bf28}+2.14\%$
test_ddpg_speed[reduce-overhead-backward] 2.5314ms 2.2454ms 445.3470 Ops/s 441.0526 Ops/s $\color{#35bf28}+0.97\%$
test_sac_speed[False-None] 8.0675ms 7.6865ms 130.0984 Ops/s 126.3838 Ops/s $\color{#35bf28}+2.94\%$
test_sac_speed[False-backward] 11.3112ms 10.9140ms 91.6254 Ops/s 89.8428 Ops/s $\color{#35bf28}+1.98\%$
test_sac_speed[True-None] 2.3442ms 2.0696ms 483.1966 Ops/s 472.6181 Ops/s $\color{#35bf28}+2.24\%$
test_sac_speed[True-backward] 4.1756ms 3.9882ms 250.7407 Ops/s 250.3665 Ops/s $\color{#35bf28}+0.15\%$
test_sac_speed[reduce-overhead-None] 2.4601ms 2.0793ms 480.9348 Ops/s 482.7917 Ops/s $\color{#d91a1a}-0.38\%$
test_sac_speed[reduce-overhead-backward] 4.2030ms 4.0187ms 248.8355 Ops/s 250.1777 Ops/s $\color{#d91a1a}-0.54\%$
test_redq_speed[False-None] 11.3329ms 10.1353ms 98.6651 Ops/s 95.2549 Ops/s $\color{#35bf28}+3.58\%$
test_redq_speed[False-backward] 24.0476ms 17.9069ms 55.8443 Ops/s 56.3402 Ops/s $\color{#d91a1a}-0.88\%$
test_redq_speed[True-None] 4.0144ms 3.6543ms 273.6499 Ops/s 261.3523 Ops/s $\color{#35bf28}+4.71\%$
test_redq_speed[True-backward] 9.3137ms 8.8443ms 113.0670 Ops/s 111.7516 Ops/s $\color{#35bf28}+1.18\%$
test_redq_speed[reduce-overhead-None] 3.9377ms 3.6587ms 273.3183 Ops/s 272.4217 Ops/s $\color{#35bf28}+0.33\%$
test_redq_speed[reduce-overhead-backward] 9.4090ms 8.8142ms 113.4535 Ops/s 113.9118 Ops/s $\color{#d91a1a}-0.40\%$
test_redq_deprec_speed[False-None] 12.2709ms 10.7541ms 92.9877 Ops/s 93.0755 Ops/s $\color{#d91a1a}-0.09\%$
test_redq_deprec_speed[False-backward] 16.2257ms 15.6670ms 63.8282 Ops/s 64.3848 Ops/s $\color{#d91a1a}-0.86\%$
test_redq_deprec_speed[True-None] 3.5038ms 3.2781ms 305.0511 Ops/s 303.6524 Ops/s $\color{#35bf28}+0.46\%$
test_redq_deprec_speed[True-backward] 7.5118ms 7.1935ms 139.0142 Ops/s 139.7152 Ops/s $\color{#d91a1a}-0.50\%$
test_redq_deprec_speed[reduce-overhead-None] 3.6194ms 3.2560ms 307.1223 Ops/s 305.2446 Ops/s $\color{#35bf28}+0.62\%$
test_redq_deprec_speed[reduce-overhead-backward] 7.5525ms 7.2329ms 138.2563 Ops/s 139.4889 Ops/s $\color{#d91a1a}-0.88\%$
test_td3_speed[False-None] 7.8530ms 7.6373ms 130.9366 Ops/s 128.2895 Ops/s $\color{#35bf28}+2.06\%$
test_td3_speed[False-backward] 10.9497ms 10.4517ms 95.6779 Ops/s 91.3469 Ops/s $\color{#35bf28}+4.74\%$
test_td3_speed[True-None] 1.9814ms 1.9316ms 517.7029 Ops/s 516.6548 Ops/s $\color{#35bf28}+0.20\%$
test_td3_speed[True-backward] 3.9329ms 3.7649ms 265.6112 Ops/s 266.6381 Ops/s $\color{#d91a1a}-0.39\%$
test_td3_speed[reduce-overhead-None] 1.9645ms 1.9305ms 518.0060 Ops/s 513.2370 Ops/s $\color{#35bf28}+0.93\%$
test_td3_speed[reduce-overhead-backward] 3.9386ms 3.7170ms 269.0354 Ops/s 264.5810 Ops/s $\color{#35bf28}+1.68\%$
test_cql_speed[False-None] 28.1304ms 25.5310ms 39.1681 Ops/s 39.3346 Ops/s $\color{#d91a1a}-0.42\%$
test_cql_speed[False-backward] 39.7601ms 35.5635ms 28.1187 Ops/s 28.2576 Ops/s $\color{#d91a1a}-0.49\%$
test_cql_speed[True-None] 11.7367ms 11.0880ms 90.1879 Ops/s 89.6145 Ops/s $\color{#35bf28}+0.64\%$
test_cql_speed[True-backward] 17.1794ms 16.7268ms 59.7842 Ops/s 59.9793 Ops/s $\color{#d91a1a}-0.33\%$
test_cql_speed[reduce-overhead-None] 11.5414ms 11.0797ms 90.2548 Ops/s 90.7508 Ops/s $\color{#d91a1a}-0.55\%$
test_cql_speed[reduce-overhead-backward] 17.4004ms 16.9623ms 58.9544 Ops/s 59.2632 Ops/s $\color{#d91a1a}-0.52\%$
test_a2c_speed[False-None] 7.4342ms 5.3485ms 186.9700 Ops/s 182.9488 Ops/s $\color{#35bf28}+2.20\%$
test_a2c_speed[False-backward] 12.3631ms 11.8827ms 84.1559 Ops/s 84.0415 Ops/s $\color{#35bf28}+0.14\%$
test_a2c_speed[True-None] 3.3939ms 3.0519ms 327.6650 Ops/s 323.1685 Ops/s $\color{#35bf28}+1.39\%$
test_a2c_speed[True-backward] 8.9623ms 8.6367ms 115.7849 Ops/s 106.8438 Ops/s $\textbf{\color{#35bf28}+8.37\%}$
test_a2c_speed[reduce-overhead-None] 3.6744ms 3.0897ms 323.6559 Ops/s 322.3255 Ops/s $\color{#35bf28}+0.41\%$
test_a2c_speed[reduce-overhead-backward] 11.3778ms 8.9779ms 111.3845 Ops/s 116.9732 Ops/s $\color{#d91a1a}-4.78\%$
test_ppo_speed[False-None] 5.9857ms 5.7426ms 174.1367 Ops/s 171.4890 Ops/s $\color{#35bf28}+1.54\%$
test_ppo_speed[False-backward] 12.7534ms 12.4042ms 80.6179 Ops/s 80.0850 Ops/s $\color{#35bf28}+0.67\%$
test_ppo_speed[True-None] 3.8353ms 3.4738ms 287.8718 Ops/s 288.3975 Ops/s $\color{#d91a1a}-0.18\%$
test_ppo_speed[True-backward] 8.7444ms 8.3711ms 119.4586 Ops/s 120.2385 Ops/s $\color{#d91a1a}-0.65\%$
test_ppo_speed[reduce-overhead-None] 3.7716ms 3.4718ms 288.0329 Ops/s 284.8852 Ops/s $\color{#35bf28}+1.10\%$
test_ppo_speed[reduce-overhead-backward] 8.6895ms 8.4171ms 118.8051 Ops/s 118.4970 Ops/s $\color{#35bf28}+0.26\%$
test_reinforce_speed[False-None] 6.0862ms 4.4610ms 224.1672 Ops/s 215.7049 Ops/s $\color{#35bf28}+3.92\%$
test_reinforce_speed[False-backward] 7.9759ms 7.3743ms 135.6057 Ops/s 132.4988 Ops/s $\color{#35bf28}+2.34\%$
test_reinforce_speed[True-None] 2.5488ms 2.2893ms 436.8142 Ops/s 442.5158 Ops/s $\color{#d91a1a}-1.29\%$
test_reinforce_speed[True-backward] 7.6509ms 7.3219ms 136.5767 Ops/s 138.4329 Ops/s $\color{#d91a1a}-1.34\%$
test_reinforce_speed[reduce-overhead-None] 2.6313ms 2.2789ms 438.8085 Ops/s 436.6925 Ops/s $\color{#35bf28}+0.48\%$
test_reinforce_speed[reduce-overhead-backward] 7.4447ms 7.1750ms 139.3736 Ops/s 140.0885 Ops/s $\color{#d91a1a}-0.51\%$
test_iql_speed[False-None] 21.1884ms 19.7440ms 50.6484 Ops/s 51.5455 Ops/s $\color{#d91a1a}-1.74\%$
test_iql_speed[False-backward] 31.5866ms 30.6269ms 32.6510 Ops/s 33.6128 Ops/s $\color{#d91a1a}-2.86\%$
test_iql_speed[True-None] 7.2784ms 6.8695ms 145.5709 Ops/s 145.4867 Ops/s $\color{#35bf28}+0.06\%$
test_iql_speed[True-backward] 16.3677ms 15.7276ms 63.5824 Ops/s 63.4746 Ops/s $\color{#35bf28}+0.17\%$
test_iql_speed[reduce-overhead-None] 7.4370ms 6.9125ms 144.6659 Ops/s 146.2347 Ops/s $\color{#d91a1a}-1.07\%$
test_iql_speed[reduce-overhead-backward] 16.5980ms 15.7265ms 63.5871 Ops/s 64.5217 Ops/s $\color{#d91a1a}-1.45\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.6043ms 6.3160ms 158.3293 Ops/s 158.9457 Ops/s $\color{#d91a1a}-0.39\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9307ms 0.2482ms 4.0289 KOps/s 3.3775 KOps/s $\textbf{\color{#35bf28}+19.28\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.5904ms 0.2891ms 3.4585 KOps/s 3.6976 KOps/s $\textbf{\color{#d91a1a}-6.47\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.6761ms 6.2184ms 160.8126 Ops/s 161.5051 Ops/s $\color{#d91a1a}-0.43\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9138ms 0.2533ms 3.9474 KOps/s 3.0706 KOps/s $\textbf{\color{#35bf28}+28.56\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6651ms 0.2658ms 3.7629 KOps/s 4.1630 KOps/s $\textbf{\color{#d91a1a}-9.61\%}$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.4564ms 1.2293ms 813.4836 Ops/s 715.3492 Ops/s $\textbf{\color{#35bf28}+13.72\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.3858ms 1.1764ms 850.0504 Ops/s 750.4696 Ops/s $\textbf{\color{#35bf28}+13.27\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6692ms 6.3760ms 156.8375 Ops/s 157.4175 Ops/s $\color{#d91a1a}-0.37\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1303ms 0.4775ms 2.0943 KOps/s 2.3179 KOps/s $\textbf{\color{#d91a1a}-9.65\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6135ms 0.4222ms 2.3684 KOps/s 2.3923 KOps/s $\color{#d91a1a}-1.00\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.4775ms 6.1421ms 162.8102 Ops/s 161.6015 Ops/s $\color{#35bf28}+0.75\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.8774ms 0.3347ms 2.9874 KOps/s 4.0532 KOps/s $\textbf{\color{#d91a1a}-26.30\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7347ms 0.3200ms 3.1247 KOps/s 4.4419 KOps/s $\textbf{\color{#d91a1a}-29.65\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.6196ms 6.1889ms 161.5791 Ops/s 160.7413 Ops/s $\color{#35bf28}+0.52\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9889ms 0.3351ms 2.9842 KOps/s 2.9473 KOps/s $\color{#35bf28}+1.25\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5075ms 0.3152ms 3.1731 KOps/s 3.1081 KOps/s $\color{#35bf28}+2.09\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.6754ms 6.3782ms 156.7837 Ops/s 155.4904 Ops/s $\color{#35bf28}+0.83\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 2.5032ms 0.4790ms 2.0875 KOps/s 2.0582 KOps/s $\color{#35bf28}+1.42\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.7793ms 0.4579ms 2.1840 KOps/s 2.1041 KOps/s $\color{#35bf28}+3.80\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 6.8688ms 5.2714ms 189.7036 Ops/s 186.9028 Ops/s $\color{#35bf28}+1.50\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 7.9458ms 2.2199ms 450.4732 Ops/s 490.3702 Ops/s $\textbf{\color{#d91a1a}-8.14\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 7.6747ms 1.2371ms 808.3271 Ops/s 790.4630 Ops/s $\color{#35bf28}+2.26\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4345s 13.9209ms 71.8344 Ops/s 184.8677 Ops/s $\textbf{\color{#d91a1a}-61.14\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.1195ms 2.0511ms 487.5499 Ops/s 486.7594 Ops/s $\color{#35bf28}+0.16\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.0768ms 1.2302ms 812.8979 Ops/s 804.3249 Ops/s $\color{#35bf28}+1.07\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.1261ms 5.4647ms 182.9922 Ops/s 178.1464 Ops/s $\color{#35bf28}+2.72\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.3861s 9.8633ms 101.3856 Ops/s 473.9401 Ops/s $\textbf{\color{#d91a1a}-78.61\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 2.7159ms 1.2499ms 800.0427 Ops/s 732.8727 Ops/s $\textbf{\color{#35bf28}+9.17\%}$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. documentation Improvements or additions to documentation
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[BUG]flatten_tensordicts default to False
2 participants