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

[Feature] Add MCTSForest/Tree.to_string #2794

Open
wants to merge 1 commit into
base: gh/kurtamohler/3/base
Choose a base branch
from

Conversation

kurtamohler
Copy link
Collaborator

@kurtamohler kurtamohler commented Feb 19, 2025

[ghstack-poisoned]
@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 Feb 19, 2025
Copy link

pytorch-bot bot commented Feb 19, 2025

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/2794

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

❌ 9 New Failures, 1 Unrelated Failure

As of commit 5816aeb with merge base 76aa9bc (image):

NEW FAILURES - The following jobs have failed:

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

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

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

kurtamohler added a commit to kurtamohler/torchrl that referenced this pull request Feb 19, 2025
ghstack-source-id: ef146b5f851ce47495cb729c3b675170698856ff
Pull Request resolved: pytorch#2794
@kurtamohler kurtamohler requested a review from vmoens February 19, 2025 21:39
kurtamohler added a commit to kurtamohler/torchrl that referenced this pull request Feb 19, 2025
ghstack-source-id: ef146b5f851ce47495cb729c3b675170698856ff
Pull Request resolved: pytorch#2794
Copy link

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}32$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.6032s 0.5153s 1.9405 Ops/s 1.9038 Ops/s $\color{#35bf28}+1.93\%$
test_transformed 1.1128s 1.0180s 0.9823 Ops/s 0.9625 Ops/s $\color{#35bf28}+2.05\%$
test_serial 1.6116s 1.5191s 0.6583 Ops/s 0.6457 Ops/s $\color{#35bf28}+1.95\%$
test_parallel 1.3850s 1.2920s 0.7740 Ops/s 0.7676 Ops/s $\color{#35bf28}+0.84\%$
test_step_mdp_speed[True-True-True-True-True] 0.2281ms 29.6448μs 33.7327 KOps/s 31.4595 KOps/s $\textbf{\color{#35bf28}+7.23\%}$
test_step_mdp_speed[True-True-True-True-False] 0.1110ms 18.2153μs 54.8990 KOps/s 54.1851 KOps/s $\color{#35bf28}+1.32\%$
test_step_mdp_speed[True-True-True-False-True] 62.3160μs 16.8341μs 59.4032 KOps/s 56.2168 KOps/s $\textbf{\color{#35bf28}+5.67\%}$
test_step_mdp_speed[True-True-True-False-False] 37.0080μs 9.8549μs 101.4724 KOps/s 95.7456 KOps/s $\textbf{\color{#35bf28}+5.98\%}$
test_step_mdp_speed[True-True-False-True-True] 80.9190μs 31.4707μs 31.7756 KOps/s 30.1071 KOps/s $\textbf{\color{#35bf28}+5.54\%}$
test_step_mdp_speed[True-True-False-True-False] 56.7360μs 19.2245μs 52.0170 KOps/s 49.7271 KOps/s $\color{#35bf28}+4.60\%$
test_step_mdp_speed[True-True-False-False-True] 62.2080μs 18.4915μs 54.0788 KOps/s 51.0496 KOps/s $\textbf{\color{#35bf28}+5.93\%}$
test_step_mdp_speed[True-True-False-False-False] 49.4020μs 11.5765μs 86.3821 KOps/s 81.4330 KOps/s $\textbf{\color{#35bf28}+6.08\%}$
test_step_mdp_speed[True-False-True-True-True] 74.1080μs 33.8560μs 29.5368 KOps/s 28.4767 KOps/s $\color{#35bf28}+3.72\%$
test_step_mdp_speed[True-False-True-True-False] 0.5952ms 21.1669μs 47.2436 KOps/s 45.3662 KOps/s $\color{#35bf28}+4.14\%$
test_step_mdp_speed[True-False-True-False-True] 44.7030μs 18.5380μs 53.9433 KOps/s 51.3857 KOps/s $\color{#35bf28}+4.98\%$
test_step_mdp_speed[True-False-True-False-False] 49.9130μs 11.5789μs 86.3642 KOps/s 81.6146 KOps/s $\textbf{\color{#35bf28}+5.82\%}$
test_step_mdp_speed[True-False-False-True-True] 84.0060μs 34.6443μs 28.8647 KOps/s 27.2469 KOps/s $\textbf{\color{#35bf28}+5.94\%}$
test_step_mdp_speed[True-False-False-True-False] 99.8250μs 22.7228μs 44.0087 KOps/s 41.1848 KOps/s $\textbf{\color{#35bf28}+6.86\%}$
test_step_mdp_speed[True-False-False-False-True] 68.9580μs 20.1596μs 49.6042 KOps/s 46.2743 KOps/s $\textbf{\color{#35bf28}+7.20\%}$
test_step_mdp_speed[True-False-False-False-False] 52.8270μs 13.2660μs 75.3809 KOps/s 71.0691 KOps/s $\textbf{\color{#35bf28}+6.07\%}$
test_step_mdp_speed[False-True-True-True-True] 86.7790μs 33.4231μs 29.9194 KOps/s 28.5835 KOps/s $\color{#35bf28}+4.67\%$
test_step_mdp_speed[False-True-True-True-False] 59.1900μs 21.0661μs 47.4696 KOps/s 45.1129 KOps/s $\textbf{\color{#35bf28}+5.22\%}$
test_step_mdp_speed[False-True-True-False-True] 62.8880μs 21.0327μs 47.5451 KOps/s 44.0465 KOps/s $\textbf{\color{#35bf28}+7.94\%}$
test_step_mdp_speed[False-True-True-False-False] 46.3260μs 12.9381μs 77.2914 KOps/s 72.7375 KOps/s $\textbf{\color{#35bf28}+6.26\%}$
test_step_mdp_speed[False-True-False-True-True] 70.4410μs 34.7363μs 28.7883 KOps/s 27.3360 KOps/s $\textbf{\color{#35bf28}+5.31\%}$
test_step_mdp_speed[False-True-False-True-False] 69.6190μs 22.6621μs 44.1266 KOps/s 41.5752 KOps/s $\textbf{\color{#35bf28}+6.14\%}$
test_step_mdp_speed[False-True-False-False-True] 2.6939ms 22.9227μs 43.6249 KOps/s 40.9156 KOps/s $\textbf{\color{#35bf28}+6.62\%}$
test_step_mdp_speed[False-True-False-False-False] 43.5410μs 14.6542μs 68.2400 KOps/s 64.2587 KOps/s $\textbf{\color{#35bf28}+6.20\%}$
test_step_mdp_speed[False-False-True-True-True] 89.5650μs 36.8567μs 27.1321 KOps/s 25.7481 KOps/s $\textbf{\color{#35bf28}+5.38\%}$
test_step_mdp_speed[False-False-True-True-False] 0.7229ms 24.9109μs 40.1430 KOps/s 38.4355 KOps/s $\color{#35bf28}+4.44\%$
test_step_mdp_speed[False-False-True-False-True] 53.9110μs 22.9403μs 43.5915 KOps/s 41.1386 KOps/s $\textbf{\color{#35bf28}+5.96\%}$
test_step_mdp_speed[False-False-True-False-False] 60.8430μs 14.5730μs 68.6200 KOps/s 63.9432 KOps/s $\textbf{\color{#35bf28}+7.31\%}$
test_step_mdp_speed[False-False-False-True-True] 84.7380μs 38.2445μs 26.1476 KOps/s 24.9670 KOps/s $\color{#35bf28}+4.73\%$
test_step_mdp_speed[False-False-False-True-False] 67.5560μs 25.8598μs 38.6701 KOps/s 36.6504 KOps/s $\textbf{\color{#35bf28}+5.51\%}$
test_step_mdp_speed[False-False-False-False-True] 63.2880μs 24.5227μs 40.7786 KOps/s 38.6905 KOps/s $\textbf{\color{#35bf28}+5.40\%}$
test_step_mdp_speed[False-False-False-False-False] 42.4190μs 16.3402μs 61.1987 KOps/s 58.1953 KOps/s $\textbf{\color{#35bf28}+5.16\%}$
test_values[generalized_advantage_estimate-True-True] 10.0468ms 9.7374ms 102.6971 Ops/s 101.6654 Ops/s $\color{#35bf28}+1.01\%$
test_values[vec_generalized_advantage_estimate-True-True] 35.1240ms 26.2937ms 38.0320 Ops/s 41.3796 Ops/s $\textbf{\color{#d91a1a}-8.09\%}$
test_values[td0_return_estimate-False-False] 0.2259ms 0.1748ms 5.7192 KOps/s 5.6283 KOps/s $\color{#35bf28}+1.62\%$
test_values[td1_return_estimate-False-False] 27.3373ms 23.8494ms 41.9298 Ops/s 41.8605 Ops/s $\color{#35bf28}+0.17\%$
test_values[vec_td1_return_estimate-False-False] 28.0955ms 26.1360ms 38.2614 Ops/s 41.3123 Ops/s $\textbf{\color{#d91a1a}-7.38\%}$
test_values[td_lambda_return_estimate-True-False] 38.7583ms 34.5015ms 28.9843 Ops/s 29.3499 Ops/s $\color{#d91a1a}-1.25\%$
test_values[vec_td_lambda_return_estimate-True-False] 28.2163ms 26.1631ms 38.2217 Ops/s 41.2723 Ops/s $\textbf{\color{#d91a1a}-7.39\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.9806ms 8.4642ms 118.1453 Ops/s 117.1986 Ops/s $\color{#35bf28}+0.81\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 3.4087ms 1.9633ms 509.3573 Ops/s 504.0344 Ops/s $\color{#35bf28}+1.06\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.6051ms 0.3568ms 2.8024 KOps/s 2.6941 KOps/s $\color{#35bf28}+4.02\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 49.0634ms 45.3515ms 22.0500 Ops/s 22.8442 Ops/s $\color{#d91a1a}-3.48\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.2294ms 3.4276ms 291.7488 Ops/s 290.4492 Ops/s $\color{#35bf28}+0.45\%$
test_dqn_speed[False-None] 5.7277ms 1.3892ms 719.8440 Ops/s 694.2061 Ops/s $\color{#35bf28}+3.69\%$
test_dqn_speed[False-backward] 1.9157ms 1.8365ms 544.5287 Ops/s 517.3809 Ops/s $\textbf{\color{#35bf28}+5.25\%}$
test_dqn_speed[True-None] 0.6055ms 0.4865ms 2.0553 KOps/s 1.9784 KOps/s $\color{#35bf28}+3.89\%$
test_dqn_speed[True-backward] 1.0180ms 0.9109ms 1.0978 KOps/s 1.0726 KOps/s $\color{#35bf28}+2.35\%$
test_dqn_speed[reduce-overhead-None] 0.6530ms 0.4863ms 2.0563 KOps/s 1.9865 KOps/s $\color{#35bf28}+3.51\%$
test_dqn_speed[reduce-overhead-backward] 0.9387ms 0.9014ms 1.1094 KOps/s 1.0864 KOps/s $\color{#35bf28}+2.11\%$
test_ddpg_speed[False-None] 3.6439ms 2.8611ms 349.5116 Ops/s 344.0874 Ops/s $\color{#35bf28}+1.58\%$
test_ddpg_speed[False-backward] 4.1007ms 3.9742ms 251.6256 Ops/s 247.5278 Ops/s $\color{#35bf28}+1.66\%$
test_ddpg_speed[True-None] 2.0649ms 1.2380ms 807.7461 Ops/s 790.1211 Ops/s $\color{#35bf28}+2.23\%$
test_ddpg_speed[True-backward] 2.2274ms 2.1232ms 470.9800 Ops/s 447.1820 Ops/s $\textbf{\color{#35bf28}+5.32\%}$
test_ddpg_speed[reduce-overhead-None] 2.0637ms 1.2366ms 808.6990 Ops/s 786.6177 Ops/s $\color{#35bf28}+2.81\%$
test_ddpg_speed[reduce-overhead-backward] 2.2591ms 2.1508ms 464.9427 Ops/s 459.5075 Ops/s $\color{#35bf28}+1.18\%$
test_sac_speed[False-None] 8.4795ms 7.9405ms 125.9359 Ops/s 122.6373 Ops/s $\color{#35bf28}+2.69\%$
test_sac_speed[False-backward] 11.1410ms 10.5787ms 94.5295 Ops/s 92.0665 Ops/s $\color{#35bf28}+2.68\%$
test_sac_speed[True-None] 3.5248ms 2.1742ms 459.9359 Ops/s 462.6727 Ops/s $\color{#d91a1a}-0.59\%$
test_sac_speed[True-backward] 3.8871ms 3.7833ms 264.3177 Ops/s 250.0600 Ops/s $\textbf{\color{#35bf28}+5.70\%}$
test_sac_speed[reduce-overhead-None] 2.7613ms 2.1196ms 471.7799 Ops/s 461.6921 Ops/s $\color{#35bf28}+2.18\%$
test_sac_speed[reduce-overhead-backward] 3.8938ms 3.8140ms 262.1916 Ops/s 259.4128 Ops/s $\color{#35bf28}+1.07\%$
test_redq_speed[False-None] 14.9726ms 13.1697ms 75.9321 Ops/s 77.0482 Ops/s $\color{#d91a1a}-1.45\%$
test_redq_speed[False-backward] 30.5324ms 23.4334ms 42.6742 Ops/s 45.4836 Ops/s $\textbf{\color{#d91a1a}-6.18\%}$
test_redq_speed[True-None] 5.9422ms 4.7648ms 209.8716 Ops/s 207.3242 Ops/s $\color{#35bf28}+1.23\%$
test_redq_speed[True-backward] 13.6319ms 12.2571ms 81.5852 Ops/s 81.2930 Ops/s $\color{#35bf28}+0.36\%$
test_redq_speed[reduce-overhead-None] 6.2475ms 4.8897ms 204.5129 Ops/s 205.4092 Ops/s $\color{#d91a1a}-0.44\%$
test_redq_speed[reduce-overhead-backward] 13.4920ms 12.2837ms 81.4088 Ops/s 81.2537 Ops/s $\color{#35bf28}+0.19\%$
test_redq_deprec_speed[False-None] 14.7850ms 12.7394ms 78.4964 Ops/s 77.4557 Ops/s $\color{#35bf28}+1.34\%$
test_redq_deprec_speed[False-backward] 19.3228ms 18.3850ms 54.3920 Ops/s 53.7333 Ops/s $\color{#35bf28}+1.23\%$
test_redq_deprec_speed[True-None] 4.6139ms 3.8393ms 260.4640 Ops/s 257.2728 Ops/s $\color{#35bf28}+1.24\%$
test_redq_deprec_speed[True-backward] 8.9774ms 8.2138ms 121.7470 Ops/s 121.3086 Ops/s $\color{#35bf28}+0.36\%$
test_redq_deprec_speed[reduce-overhead-None] 4.3943ms 3.8450ms 260.0803 Ops/s 257.5297 Ops/s $\color{#35bf28}+0.99\%$
test_redq_deprec_speed[reduce-overhead-backward] 8.4069ms 8.2547ms 121.1433 Ops/s 120.4895 Ops/s $\color{#35bf28}+0.54\%$
test_td3_speed[False-None] 8.6169ms 7.8891ms 126.7569 Ops/s 123.0693 Ops/s $\color{#35bf28}+3.00\%$
test_td3_speed[False-backward] 12.2665ms 10.3064ms 97.0274 Ops/s 94.8468 Ops/s $\color{#35bf28}+2.30\%$
test_td3_speed[True-None] 2.0022ms 1.8302ms 546.3841 Ops/s 537.9093 Ops/s $\color{#35bf28}+1.58\%$
test_td3_speed[True-backward] 4.2155ms 3.4613ms 288.9084 Ops/s 280.0209 Ops/s $\color{#35bf28}+3.17\%$
test_td3_speed[reduce-overhead-None] 1.9023ms 1.8174ms 550.2309 Ops/s 543.0062 Ops/s $\color{#35bf28}+1.33\%$
test_td3_speed[reduce-overhead-backward] 3.5020ms 3.4230ms 292.1436 Ops/s 290.1773 Ops/s $\color{#35bf28}+0.68\%$
test_cql_speed[False-None] 38.9389ms 36.4296ms 27.4502 Ops/s 27.2741 Ops/s $\color{#35bf28}+0.65\%$
test_cql_speed[False-backward] 50.0242ms 46.6460ms 21.4381 Ops/s 21.3046 Ops/s $\color{#35bf28}+0.63\%$
test_cql_speed[True-None] 16.9582ms 15.9521ms 62.6878 Ops/s 62.2079 Ops/s $\color{#35bf28}+0.77\%$
test_cql_speed[True-backward] 31.9729ms 22.8234ms 43.8147 Ops/s 43.8047 Ops/s $\color{#35bf28}+0.02\%$
test_cql_speed[reduce-overhead-None] 16.8524ms 16.1329ms 61.9853 Ops/s 62.2469 Ops/s $\color{#d91a1a}-0.42\%$
test_cql_speed[reduce-overhead-backward] 23.7921ms 22.8266ms 43.8085 Ops/s 43.0109 Ops/s $\color{#35bf28}+1.85\%$
test_a2c_speed[False-None] 7.9476ms 7.1948ms 138.9886 Ops/s 137.6318 Ops/s $\color{#35bf28}+0.99\%$
test_a2c_speed[False-backward] 14.3363ms 13.8503ms 72.2004 Ops/s 69.8534 Ops/s $\color{#35bf28}+3.36\%$
test_a2c_speed[True-None] 4.1202ms 3.7487ms 266.7610 Ops/s 265.1989 Ops/s $\color{#35bf28}+0.59\%$
test_a2c_speed[True-backward] 16.7461ms 10.4753ms 95.4623 Ops/s 97.4102 Ops/s $\color{#d91a1a}-2.00\%$
test_a2c_speed[reduce-overhead-None] 4.0260ms 3.7179ms 268.9719 Ops/s 263.6314 Ops/s $\color{#35bf28}+2.03\%$
test_a2c_speed[reduce-overhead-backward] 11.9196ms 10.5013ms 95.2260 Ops/s 97.8483 Ops/s $\color{#d91a1a}-2.68\%$
test_ppo_speed[False-None] 9.1138ms 7.4412ms 134.3866 Ops/s 132.1050 Ops/s $\color{#35bf28}+1.73\%$
test_ppo_speed[False-backward] 16.0686ms 14.7088ms 67.9863 Ops/s 67.8951 Ops/s $\color{#35bf28}+0.13\%$
test_ppo_speed[True-None] 4.6591ms 4.0926ms 244.3420 Ops/s 238.3123 Ops/s $\color{#35bf28}+2.53\%$
test_ppo_speed[True-backward] 10.3276ms 9.9498ms 100.5042 Ops/s 99.2635 Ops/s $\color{#35bf28}+1.25\%$
test_ppo_speed[reduce-overhead-None] 4.9078ms 4.1009ms 243.8499 Ops/s 240.7119 Ops/s $\color{#35bf28}+1.30\%$
test_ppo_speed[reduce-overhead-backward] 10.2657ms 9.9447ms 100.5562 Ops/s 99.3983 Ops/s $\color{#35bf28}+1.16\%$
test_reinforce_speed[False-None] 7.7744ms 6.5320ms 153.0917 Ops/s 152.1662 Ops/s $\color{#35bf28}+0.61\%$
test_reinforce_speed[False-backward] 11.0072ms 9.8028ms 102.0117 Ops/s 101.3204 Ops/s $\color{#35bf28}+0.68\%$
test_reinforce_speed[True-None] 3.8779ms 3.0835ms 324.3047 Ops/s 320.0685 Ops/s $\color{#35bf28}+1.32\%$
test_reinforce_speed[True-backward] 9.6751ms 9.1560ms 109.2184 Ops/s 110.1890 Ops/s $\color{#d91a1a}-0.88\%$
test_reinforce_speed[reduce-overhead-None] 3.4430ms 3.0534ms 327.5059 Ops/s 310.5354 Ops/s $\textbf{\color{#35bf28}+5.46\%}$
test_reinforce_speed[reduce-overhead-backward] 11.0483ms 9.0723ms 110.2262 Ops/s 109.6678 Ops/s $\color{#35bf28}+0.51\%$
test_iql_speed[False-None] 35.1128ms 32.5368ms 30.7345 Ops/s 30.6405 Ops/s $\color{#35bf28}+0.31\%$
test_iql_speed[False-backward] 0.3576s 51.8417ms 19.2895 Ops/s 21.9952 Ops/s $\textbf{\color{#d91a1a}-12.30\%}$
test_iql_speed[True-None] 12.1841ms 10.9849ms 91.0344 Ops/s 88.9568 Ops/s $\color{#35bf28}+2.34\%$
test_iql_speed[True-backward] 28.3319ms 22.5041ms 44.4364 Ops/s 45.1741 Ops/s $\color{#d91a1a}-1.63\%$
test_iql_speed[reduce-overhead-None] 12.7949ms 11.7114ms 85.3869 Ops/s 88.0936 Ops/s $\color{#d91a1a}-3.07\%$
test_iql_speed[reduce-overhead-backward] 22.7761ms 21.6422ms 46.2060 Ops/s 45.1165 Ops/s $\color{#35bf28}+2.41\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.2877ms 4.7172ms 211.9916 Ops/s 204.9431 Ops/s $\color{#35bf28}+3.44\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 3.1417ms 0.5380ms 1.8587 KOps/s 1.8710 KOps/s $\color{#d91a1a}-0.66\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.7942ms 0.5094ms 1.9632 KOps/s 1.9562 KOps/s $\color{#35bf28}+0.36\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 4.9186ms 4.5407ms 220.2304 Ops/s 214.4392 Ops/s $\color{#35bf28}+2.70\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 2.6561ms 0.5195ms 1.9250 KOps/s 1.8942 KOps/s $\color{#35bf28}+1.62\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9046ms 0.5066ms 1.9739 KOps/s 2.0152 KOps/s $\color{#d91a1a}-2.05\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 3.6259ms 1.8105ms 552.3436 Ops/s 583.0890 Ops/s $\textbf{\color{#d91a1a}-5.27\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 2.0363ms 1.6047ms 623.1807 Ops/s 613.0011 Ops/s $\color{#35bf28}+1.66\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 7.3454ms 4.7625ms 209.9754 Ops/s 209.3213 Ops/s $\color{#35bf28}+0.31\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0438ms 0.6694ms 1.4940 KOps/s 1.4831 KOps/s $\color{#35bf28}+0.73\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.8868ms 0.6435ms 1.5540 KOps/s 1.5524 KOps/s $\color{#35bf28}+0.10\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.1371ms 4.6674ms 214.2532 Ops/s 210.4863 Ops/s $\color{#35bf28}+1.79\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.1697ms 0.5359ms 1.8661 KOps/s 1.8699 KOps/s $\color{#d91a1a}-0.20\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.8386ms 0.5177ms 1.9315 KOps/s 1.9670 KOps/s $\color{#d91a1a}-1.81\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 5.0533ms 4.6151ms 216.6794 Ops/s 214.2278 Ops/s $\color{#35bf28}+1.14\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.1234ms 0.5282ms 1.8931 KOps/s 1.8709 KOps/s $\color{#35bf28}+1.18\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.8619ms 0.5100ms 1.9610 KOps/s 2.0032 KOps/s $\color{#d91a1a}-2.11\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.2119ms 4.8608ms 205.7278 Ops/s 207.1947 Ops/s $\color{#d91a1a}-0.71\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0021ms 0.6761ms 1.4790 KOps/s 1.4553 KOps/s $\color{#35bf28}+1.63\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0178ms 0.6552ms 1.5263 KOps/s 1.5573 KOps/s $\color{#d91a1a}-1.99\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 5.6693ms 4.2651ms 234.4598 Ops/s 235.1957 Ops/s $\color{#d91a1a}-0.31\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 8.6326ms 2.4096ms 415.0138 Ops/s 451.2482 Ops/s $\textbf{\color{#d91a1a}-8.03\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 4.8625ms 1.4012ms 713.6784 Ops/s 772.5981 Ops/s $\textbf{\color{#d91a1a}-7.63\%}$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 9.2774ms 4.4512ms 224.6597 Ops/s 177.7864 Ops/s $\textbf{\color{#35bf28}+26.36\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.4543s 11.4638ms 87.2311 Ops/s 365.4302 Ops/s $\textbf{\color{#d91a1a}-76.13\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 1.9847ms 1.3307ms 751.4596 Ops/s 749.6034 Ops/s $\color{#35bf28}+0.25\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 6.2907ms 4.5794ms 218.3702 Ops/s 32.9476 Ops/s $\textbf{\color{#35bf28}+562.78\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 5.7528ms 2.4510ms 407.9934 Ops/s 384.2574 Ops/s $\textbf{\color{#35bf28}+6.18\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.3074ms 1.5234ms 656.4419 Ops/s 638.4055 Ops/s $\color{#35bf28}+2.83\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 14.3639ms 11.5956ms 86.2395 Ops/s 81.0197 Ops/s $\textbf{\color{#35bf28}+6.44\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 15.8612ms 14.1992ms 70.4263 Ops/s 69.6572 Ops/s $\color{#35bf28}+1.10\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 21.4169ms 20.4097ms 48.9963 Ops/s 47.7777 Ops/s $\color{#35bf28}+2.55\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 16.1284ms 14.3550ms 69.6622 Ops/s 67.6880 Ops/s $\color{#35bf28}+2.92\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 20.9778ms 20.2563ms 49.3674 Ops/s 48.3566 Ops/s $\color{#35bf28}+2.09\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 16.6262ms 15.5988ms 64.1074 Ops/s 62.3447 Ops/s $\color{#35bf28}+2.83\%$

Copy link

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

Total Benchmarks: 149. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_simple 0.9143s 0.8274s 1.2086 Ops/s 1.1949 Ops/s $\color{#35bf28}+1.15\%$
test_transformed 1.5356s 1.4474s 0.6909 Ops/s 0.6744 Ops/s $\color{#35bf28}+2.45\%$
test_serial 2.4284s 2.3403s 0.4273 Ops/s 0.4213 Ops/s $\color{#35bf28}+1.43\%$
test_parallel 2.1166s 1.9564s 0.5111 Ops/s 0.4904 Ops/s $\color{#35bf28}+4.23\%$
test_step_mdp_speed[True-True-True-True-True] 0.2012ms 40.9192μs 24.4384 KOps/s 24.9441 KOps/s $\color{#d91a1a}-2.03\%$
test_step_mdp_speed[True-True-True-True-False] 47.0110μs 23.4385μs 42.6648 KOps/s 42.3094 KOps/s $\color{#35bf28}+0.84\%$
test_step_mdp_speed[True-True-True-False-True] 51.5710μs 22.5206μs 44.4039 KOps/s 44.5536 KOps/s $\color{#d91a1a}-0.34\%$
test_step_mdp_speed[True-True-True-False-False] 51.8110μs 12.9996μs 76.9254 KOps/s 75.9474 KOps/s $\color{#35bf28}+1.29\%$
test_step_mdp_speed[True-True-False-True-True] 76.0710μs 42.7453μs 23.3944 KOps/s 23.3556 KOps/s $\color{#35bf28}+0.17\%$
test_step_mdp_speed[True-True-False-True-False] 57.9910μs 25.7469μs 38.8397 KOps/s 38.1831 KOps/s $\color{#35bf28}+1.72\%$
test_step_mdp_speed[True-True-False-False-True] 60.6810μs 24.7190μs 40.4546 KOps/s 40.0624 KOps/s $\color{#35bf28}+0.98\%$
test_step_mdp_speed[True-True-False-False-False] 43.8400μs 15.5363μs 64.3655 KOps/s 63.7805 KOps/s $\color{#35bf28}+0.92\%$
test_step_mdp_speed[True-False-True-True-True] 0.1173ms 44.0215μs 22.7162 KOps/s 21.9610 KOps/s $\color{#35bf28}+3.44\%$
test_step_mdp_speed[True-False-True-True-False] 58.3510μs 28.1353μs 35.5425 KOps/s 34.8280 KOps/s $\color{#35bf28}+2.05\%$
test_step_mdp_speed[True-False-True-False-True] 59.4710μs 24.7403μs 40.4198 KOps/s 40.0003 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[True-False-True-False-False] 40.1310μs 15.4561μs 64.6994 KOps/s 64.1479 KOps/s $\color{#35bf28}+0.86\%$
test_step_mdp_speed[True-False-False-True-True] 76.3210μs 47.5026μs 21.0515 KOps/s 20.9444 KOps/s $\color{#35bf28}+0.51\%$
test_step_mdp_speed[True-False-False-True-False] 59.0910μs 29.8724μs 33.4757 KOps/s 32.1795 KOps/s $\color{#35bf28}+4.03\%$
test_step_mdp_speed[True-False-False-False-True] 53.7810μs 26.9340μs 37.1278 KOps/s 37.3402 KOps/s $\color{#d91a1a}-0.57\%$
test_step_mdp_speed[True-False-False-False-False] 46.5400μs 17.5579μs 56.9544 KOps/s 55.4746 KOps/s $\color{#35bf28}+2.67\%$
test_step_mdp_speed[False-True-True-True-True] 76.8510μs 45.2391μs 22.1048 KOps/s 22.3309 KOps/s $\color{#d91a1a}-1.01\%$
test_step_mdp_speed[False-True-True-True-False] 56.2810μs 28.0017μs 35.7121 KOps/s 35.1479 KOps/s $\color{#35bf28}+1.61\%$
test_step_mdp_speed[False-True-True-False-True] 62.5510μs 28.9012μs 34.6007 KOps/s 34.9592 KOps/s $\color{#d91a1a}-1.03\%$
test_step_mdp_speed[False-True-True-False-False] 44.0910μs 17.1215μs 58.4062 KOps/s 57.5243 KOps/s $\color{#35bf28}+1.53\%$
test_step_mdp_speed[False-True-False-True-True] 78.6510μs 47.7334μs 20.9497 KOps/s 20.8746 KOps/s $\color{#35bf28}+0.36\%$
test_step_mdp_speed[False-True-False-True-False] 60.9110μs 30.5444μs 32.7392 KOps/s 31.4790 KOps/s $\color{#35bf28}+4.00\%$
test_step_mdp_speed[False-True-False-False-True] 3.1952ms 31.5896μs 31.6560 KOps/s 31.6242 KOps/s $\color{#35bf28}+0.10\%$
test_step_mdp_speed[False-True-False-False-False] 46.4000μs 19.5722μs 51.0929 KOps/s 50.9020 KOps/s $\color{#35bf28}+0.37\%$
test_step_mdp_speed[False-False-True-True-True] 84.6110μs 49.6913μs 20.1242 KOps/s 19.9453 KOps/s $\color{#35bf28}+0.90\%$
test_step_mdp_speed[False-False-True-True-False] 62.1400μs 32.9788μs 30.3226 KOps/s 30.0266 KOps/s $\color{#35bf28}+0.99\%$
test_step_mdp_speed[False-False-True-False-True] 62.6810μs 31.4896μs 31.7565 KOps/s 32.3923 KOps/s $\color{#d91a1a}-1.96\%$
test_step_mdp_speed[False-False-True-False-False] 48.0010μs 19.5851μs 51.0592 KOps/s 51.3607 KOps/s $\color{#d91a1a}-0.59\%$
test_step_mdp_speed[False-False-False-True-True] 87.2610μs 51.8953μs 19.2696 KOps/s 19.6641 KOps/s $\color{#d91a1a}-2.01\%$
test_step_mdp_speed[False-False-False-True-False] 65.9110μs 35.2781μs 28.3462 KOps/s 28.1519 KOps/s $\color{#35bf28}+0.69\%$
test_step_mdp_speed[False-False-False-False-True] 66.0110μs 33.4742μs 29.8738 KOps/s 31.1633 KOps/s $\color{#d91a1a}-4.14\%$
test_step_mdp_speed[False-False-False-False-False] 56.7810μs 21.5564μs 46.3900 KOps/s 44.6548 KOps/s $\color{#35bf28}+3.89\%$
test_values[generalized_advantage_estimate-True-True] 25.5361ms 25.1758ms 39.7207 Ops/s 39.7746 Ops/s $\color{#d91a1a}-0.14\%$
test_values[vec_generalized_advantage_estimate-True-True] 99.8712ms 2.9071ms 343.9860 Ops/s 323.1822 Ops/s $\textbf{\color{#35bf28}+6.44\%}$
test_values[td0_return_estimate-False-False] 0.1063ms 81.7962μs 12.2255 KOps/s 12.4221 KOps/s $\color{#d91a1a}-1.58\%$
test_values[td1_return_estimate-False-False] 60.6929ms 57.8254ms 17.2934 Ops/s 17.1528 Ops/s $\color{#35bf28}+0.82\%$
test_values[vec_td1_return_estimate-False-False] 1.3324ms 1.0962ms 912.2190 Ops/s 919.6098 Ops/s $\color{#d91a1a}-0.80\%$
test_values[td_lambda_return_estimate-True-False] 96.7263ms 92.4695ms 10.8144 Ops/s 11.0599 Ops/s $\color{#d91a1a}-2.22\%$
test_values[vec_td_lambda_return_estimate-True-False] 1.2461ms 1.0899ms 917.5564 Ops/s 918.7483 Ops/s $\color{#d91a1a}-0.13\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.8834ms 26.6938ms 37.4619 Ops/s 40.3869 Ops/s $\textbf{\color{#d91a1a}-7.24\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 1.0541ms 0.7841ms 1.2753 KOps/s 1.3238 KOps/s $\color{#d91a1a}-3.66\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8159ms 0.7044ms 1.4196 KOps/s 1.4910 KOps/s $\color{#d91a1a}-4.79\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5271ms 1.4944ms 669.1551 Ops/s 673.9243 Ops/s $\color{#d91a1a}-0.71\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.7354ms 0.6945ms 1.4398 KOps/s 1.4495 KOps/s $\color{#d91a1a}-0.67\%$
test_dqn_speed[False-None] 7.0916ms 1.5401ms 649.2986 Ops/s 624.9691 Ops/s $\color{#35bf28}+3.89\%$
test_dqn_speed[False-backward] 2.3603ms 2.1490ms 465.3311 Ops/s 463.3680 Ops/s $\color{#35bf28}+0.42\%$
test_dqn_speed[True-None] 0.7126ms 0.5733ms 1.7443 KOps/s 1.6556 KOps/s $\textbf{\color{#35bf28}+5.36\%}$
test_dqn_speed[True-backward] 1.2330ms 1.1548ms 865.9829 Ops/s 854.7152 Ops/s $\color{#35bf28}+1.32\%$
test_dqn_speed[reduce-overhead-None] 0.9922ms 0.5955ms 1.6792 KOps/s 1.6780 KOps/s $\color{#35bf28}+0.07\%$
test_dqn_speed[reduce-overhead-backward] 1.0731ms 0.9864ms 1.0138 KOps/s 1.0015 KOps/s $\color{#35bf28}+1.23\%$
test_ddpg_speed[False-None] 3.3085ms 2.9151ms 343.0431 Ops/s 343.6315 Ops/s $\color{#d91a1a}-0.17\%$
test_ddpg_speed[False-backward] 4.6547ms 4.1888ms 238.7334 Ops/s 240.5379 Ops/s $\color{#d91a1a}-0.75\%$
test_ddpg_speed[True-None] 1.7798ms 1.3834ms 722.8803 Ops/s 720.9534 Ops/s $\color{#35bf28}+0.27\%$
test_ddpg_speed[True-backward] 2.6150ms 2.5090ms 398.5652 Ops/s 398.7227 Ops/s $\color{#d91a1a}-0.04\%$
test_ddpg_speed[reduce-overhead-None] 1.4855ms 1.3928ms 718.0009 Ops/s 715.5542 Ops/s $\color{#35bf28}+0.34\%$
test_ddpg_speed[reduce-overhead-backward] 2.0765ms 1.9459ms 513.8892 Ops/s 511.7945 Ops/s $\color{#35bf28}+0.41\%$
test_sac_speed[False-None] 8.6294ms 8.1340ms 122.9406 Ops/s 121.3445 Ops/s $\color{#35bf28}+1.32\%$
test_sac_speed[False-backward] 11.6086ms 11.1082ms 90.0237 Ops/s 89.1502 Ops/s $\color{#35bf28}+0.98\%$
test_sac_speed[True-None] 1.9717ms 1.9080ms 524.1186 Ops/s 520.3045 Ops/s $\color{#35bf28}+0.73\%$
test_sac_speed[True-backward] 3.9749ms 3.8829ms 257.5424 Ops/s 259.8970 Ops/s $\color{#d91a1a}-0.91\%$
test_sac_speed[reduce-overhead-None] 20.7059ms 11.7733ms 84.9377 Ops/s 84.4365 Ops/s $\color{#35bf28}+0.59\%$
test_sac_speed[reduce-overhead-backward] 1.8140ms 1.7718ms 564.4121 Ops/s 534.1104 Ops/s $\textbf{\color{#35bf28}+5.67\%}$
test_redq_speed[False-None] 8.0158ms 7.6012ms 131.5578 Ops/s 129.5677 Ops/s $\color{#35bf28}+1.54\%$
test_redq_speed[False-backward] 12.0989ms 11.7896ms 84.8206 Ops/s 83.5860 Ops/s $\color{#35bf28}+1.48\%$
test_redq_speed[True-None] 2.7966ms 2.3901ms 418.3855 Ops/s 410.2407 Ops/s $\color{#35bf28}+1.99\%$
test_redq_speed[True-backward] 4.5465ms 4.1354ms 241.8159 Ops/s 230.9391 Ops/s $\color{#35bf28}+4.71\%$
test_redq_speed[reduce-overhead-None] 2.8146ms 2.4192ms 413.3657 Ops/s 409.0730 Ops/s $\color{#35bf28}+1.05\%$
test_redq_speed[reduce-overhead-backward] 4.5116ms 4.1897ms 238.6826 Ops/s 230.0911 Ops/s $\color{#35bf28}+3.73\%$
test_redq_deprec_speed[False-None] 9.6718ms 9.1348ms 109.4716 Ops/s 108.8120 Ops/s $\color{#35bf28}+0.61\%$
test_redq_deprec_speed[False-backward] 12.8863ms 12.2233ms 81.8112 Ops/s 80.1370 Ops/s $\color{#35bf28}+2.09\%$
test_redq_deprec_speed[True-None] 2.9640ms 2.7525ms 363.3036 Ops/s 360.8938 Ops/s $\color{#35bf28}+0.67\%$
test_redq_deprec_speed[True-backward] 4.7640ms 4.4453ms 224.9542 Ops/s 215.7036 Ops/s $\color{#35bf28}+4.29\%$
test_redq_deprec_speed[reduce-overhead-None] 2.8995ms 2.7229ms 367.2542 Ops/s 365.5632 Ops/s $\color{#35bf28}+0.46\%$
test_redq_deprec_speed[reduce-overhead-backward] 4.8914ms 4.6867ms 213.3684 Ops/s 216.8714 Ops/s $\color{#d91a1a}-1.62\%$
test_td3_speed[False-None] 8.3534ms 8.0706ms 123.9061 Ops/s 122.6423 Ops/s $\color{#35bf28}+1.03\%$
test_td3_speed[False-backward] 11.0107ms 10.7173ms 93.3068 Ops/s 94.1989 Ops/s $\color{#d91a1a}-0.95\%$
test_td3_speed[True-None] 1.7984ms 1.7233ms 580.2797 Ops/s 578.5516 Ops/s $\color{#35bf28}+0.30\%$
test_td3_speed[True-backward] 3.3939ms 3.2624ms 306.5237 Ops/s 289.8040 Ops/s $\textbf{\color{#35bf28}+5.77\%}$
test_td3_speed[reduce-overhead-None] 51.0978ms 26.2722ms 38.0630 Ops/s 37.6354 Ops/s $\color{#35bf28}+1.14\%$
test_td3_speed[reduce-overhead-backward] 1.4609ms 1.3936ms 717.5616 Ops/s 617.8325 Ops/s $\textbf{\color{#35bf28}+16.14\%}$
test_cql_speed[False-None] 17.8000ms 17.0614ms 58.6120 Ops/s 58.3787 Ops/s $\color{#35bf28}+0.40\%$
test_cql_speed[False-backward] 22.7073ms 22.1737ms 45.0984 Ops/s 44.2370 Ops/s $\color{#35bf28}+1.95\%$
test_cql_speed[True-None] 3.5960ms 3.4259ms 291.8960 Ops/s 292.9438 Ops/s $\color{#d91a1a}-0.36\%$
test_cql_speed[True-backward] 6.1856ms 5.6934ms 175.6417 Ops/s 169.7229 Ops/s $\color{#35bf28}+3.49\%$
test_cql_speed[reduce-overhead-None] 20.6015ms 12.9800ms 77.0414 Ops/s 75.4706 Ops/s $\color{#35bf28}+2.08\%$
test_cql_speed[reduce-overhead-backward] 2.0483ms 1.8877ms 529.7402 Ops/s 523.6325 Ops/s $\color{#35bf28}+1.17\%$
test_a2c_speed[False-None] 3.6916ms 3.2480ms 307.8777 Ops/s 303.2223 Ops/s $\color{#35bf28}+1.54\%$
test_a2c_speed[False-backward] 7.2483ms 6.1780ms 161.8645 Ops/s 159.6737 Ops/s $\color{#35bf28}+1.37\%$
test_a2c_speed[True-None] 1.5819ms 1.4080ms 710.2189 Ops/s 716.4519 Ops/s $\color{#d91a1a}-0.87\%$
test_a2c_speed[True-backward] 3.0206ms 2.9694ms 336.7714 Ops/s 332.2376 Ops/s $\color{#35bf28}+1.36\%$
test_a2c_speed[reduce-overhead-None] 15.6808ms 8.8511ms 112.9809 Ops/s 112.3532 Ops/s $\color{#35bf28}+0.56\%$
test_a2c_speed[reduce-overhead-backward] 1.5707ms 1.4933ms 669.6798 Ops/s 660.6451 Ops/s $\color{#35bf28}+1.37\%$
test_ppo_speed[False-None] 4.1471ms 3.8461ms 260.0062 Ops/s 262.7534 Ops/s $\color{#d91a1a}-1.05\%$
test_ppo_speed[False-backward] 7.3990ms 6.9186ms 144.5381 Ops/s 143.2954 Ops/s $\color{#35bf28}+0.87\%$
test_ppo_speed[True-None] 1.5518ms 1.4563ms 686.6529 Ops/s 678.2625 Ops/s $\color{#35bf28}+1.24\%$
test_ppo_speed[True-backward] 3.2478ms 3.1166ms 320.8673 Ops/s 295.2075 Ops/s $\textbf{\color{#35bf28}+8.69\%}$
test_ppo_speed[reduce-overhead-None] 1.1204ms 1.0015ms 998.5390 Ops/s 1.0026 KOps/s $\color{#d91a1a}-0.41\%$
test_ppo_speed[reduce-overhead-backward] 1.5822ms 1.4452ms 691.9571 Ops/s 604.6980 Ops/s $\textbf{\color{#35bf28}+14.43\%}$
test_reinforce_speed[False-None] 2.9786ms 2.3175ms 431.4954 Ops/s 428.5671 Ops/s $\color{#35bf28}+0.68\%$
test_reinforce_speed[False-backward] 3.9065ms 3.3240ms 300.8450 Ops/s 288.0574 Ops/s $\color{#35bf28}+4.44\%$
test_reinforce_speed[True-None] 1.4332ms 1.3412ms 745.5968 Ops/s 737.0868 Ops/s $\color{#35bf28}+1.15\%$
test_reinforce_speed[True-backward] 3.1113ms 2.9890ms 334.5631 Ops/s 318.5829 Ops/s $\textbf{\color{#35bf28}+5.02\%}$
test_reinforce_speed[reduce-overhead-None] 17.5753ms 9.8087ms 101.9502 Ops/s 100.1324 Ops/s $\color{#35bf28}+1.82\%$
test_reinforce_speed[reduce-overhead-backward] 1.7873ms 1.6781ms 595.9218 Ops/s 582.0798 Ops/s $\color{#35bf28}+2.38\%$
test_iql_speed[False-None] 9.8922ms 9.3330ms 107.1463 Ops/s 102.6088 Ops/s $\color{#35bf28}+4.42\%$
test_iql_speed[False-backward] 13.7958ms 13.2647ms 75.3883 Ops/s 74.4065 Ops/s $\color{#35bf28}+1.32\%$
test_iql_speed[True-None] 2.4867ms 2.3105ms 432.7995 Ops/s 422.6521 Ops/s $\color{#35bf28}+2.40\%$
test_iql_speed[True-backward] 5.1702ms 5.0339ms 198.6549 Ops/s 194.7285 Ops/s $\color{#35bf28}+2.02\%$
test_iql_speed[reduce-overhead-None] 0.4824s 12.7784ms 78.2573 Ops/s 89.9866 Ops/s $\textbf{\color{#d91a1a}-13.03\%}$
test_iql_speed[reduce-overhead-backward] 2.1710ms 2.1274ms 470.0591 Ops/s 455.6710 Ops/s $\color{#35bf28}+3.16\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 7.7815ms 6.2805ms 159.2226 Ops/s 154.5545 Ops/s $\color{#35bf28}+3.02\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.5660ms 0.3384ms 2.9548 KOps/s 2.8667 KOps/s $\color{#35bf28}+3.07\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.4756ms 0.2782ms 3.5949 KOps/s 3.6061 KOps/s $\color{#d91a1a}-0.31\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.2692ms 5.9801ms 167.2204 Ops/s 164.2828 Ops/s $\color{#35bf28}+1.79\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.0355ms 0.3054ms 3.2741 KOps/s 3.2573 KOps/s $\color{#35bf28}+0.52\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6143ms 0.3107ms 3.2184 KOps/s 3.2722 KOps/s $\color{#d91a1a}-1.64\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-sampler6-10000] 1.7732ms 1.4696ms 680.4700 Ops/s 731.1073 Ops/s $\textbf{\color{#d91a1a}-6.93\%}$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-sampler7-10000] 1.5507ms 1.3415ms 745.4608 Ops/s 802.0942 Ops/s $\textbf{\color{#d91a1a}-7.06\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.2837ms 6.1697ms 162.0835 Ops/s 158.6810 Ops/s $\color{#35bf28}+2.14\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.7239ms 0.4335ms 2.3067 KOps/s 2.2892 KOps/s $\color{#35bf28}+0.77\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6926ms 0.4749ms 2.1059 KOps/s 2.4072 KOps/s $\textbf{\color{#d91a1a}-12.52\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 6.2657ms 6.0791ms 164.4986 Ops/s 162.9591 Ops/s $\color{#35bf28}+0.94\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 1.8961ms 0.3166ms 3.1581 KOps/s 3.7062 KOps/s $\textbf{\color{#d91a1a}-14.79\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6750ms 0.3157ms 3.1677 KOps/s 3.9839 KOps/s $\textbf{\color{#d91a1a}-20.49\%}$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 9.0118ms 5.9992ms 166.6877 Ops/s 163.4765 Ops/s $\color{#35bf28}+1.96\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.7601ms 0.3383ms 2.9558 KOps/s 3.2147 KOps/s $\textbf{\color{#d91a1a}-8.05\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.6021ms 0.3172ms 3.1524 KOps/s 3.5117 KOps/s $\textbf{\color{#d91a1a}-10.23\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 6.4053ms 6.2359ms 160.3611 Ops/s 159.2585 Ops/s $\color{#35bf28}+0.69\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7480ms 0.4300ms 2.3255 KOps/s 2.4117 KOps/s $\color{#d91a1a}-3.57\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 0.6838ms 0.4243ms 2.3569 KOps/s 2.3758 KOps/s $\color{#d91a1a}-0.79\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 7.0066ms 5.4689ms 182.8521 Ops/s 173.5605 Ops/s $\textbf{\color{#35bf28}+5.35\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 6.2441ms 1.9998ms 500.0437 Ops/s 426.9833 Ops/s $\textbf{\color{#35bf28}+17.11\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 8.7830ms 1.3054ms 766.0252 Ops/s 755.0800 Ops/s $\color{#35bf28}+1.45\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.4487s 14.6878ms 68.0838 Ops/s 180.3001 Ops/s $\textbf{\color{#d91a1a}-62.24\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 9.1081ms 2.0590ms 485.6800 Ops/s 423.2588 Ops/s $\textbf{\color{#35bf28}+14.75\%}$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 7.5480ms 1.2508ms 799.4831 Ops/s 750.3488 Ops/s $\textbf{\color{#35bf28}+6.55\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 7.5664ms 5.7151ms 174.9743 Ops/s 31.1474 Ops/s $\textbf{\color{#35bf28}+461.76\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 9.0180ms 2.2206ms 450.3360 Ops/s 436.0419 Ops/s $\color{#35bf28}+3.28\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 7.1447ms 1.4120ms 708.2004 Ops/s 719.2612 Ops/s $\color{#d91a1a}-1.54\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-True] 13.9998ms 13.3258ms 75.0421 Ops/s 70.5523 Ops/s $\textbf{\color{#35bf28}+6.36\%}$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-10000-10000-100-False] 20.0565ms 17.0140ms 58.7751 Ops/s 57.8974 Ops/s $\color{#35bf28}+1.52\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-True] 18.7247ms 17.9000ms 55.8660 Ops/s 54.1800 Ops/s $\color{#35bf28}+3.11\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-100000-10000-100-False] 18.5830ms 16.9477ms 59.0050 Ops/s 57.8845 Ops/s $\color{#35bf28}+1.94\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-True] 18.0510ms 17.8324ms 56.0777 Ops/s 54.1029 Ops/s $\color{#35bf28}+3.65\%$
test_rb_extend_sample[ReplayBuffer-LazyTensorStorage-RandomSampler-1000000-10000-100-False] 19.3773ms 18.2225ms 54.8772 Ops/s 53.6575 Ops/s $\color{#35bf28}+2.27\%$

kurtamohler added a commit to kurtamohler/torchrl that referenced this pull request Feb 19, 2025
ghstack-source-id: ef146b5f851ce47495cb729c3b675170698856ff
Pull Request resolved: pytorch#2794
@vmoens vmoens added the enhancement New feature or request label Feb 20, 2025
Copy link
Contributor

@vmoens vmoens left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Looks good! I would just consider either a default value and / or a simple example in the docstrings (you can just copy paste the test, that will be good enough)

@@ -604,6 +604,42 @@ def plot(
f"Unknown plotting backend {backend} with figure {figure}."
)

def to_string(self, node_format_fn):
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

we don't want a default for this?

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. enhancement New feature or request
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants