Skip to content

Conversation

vmoens
Copy link
Collaborator

@vmoens vmoens commented Jul 6, 2023

Fixes operations in lazy stack tds:

  • unsqueeze
  • squeeze
  • print
  • __len__
  • __eq__
  • __ne__
  • project
  • type_check
  • __delitem__
  • __iter__
  • keys
  • __setitem__

@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 Jul 6, 2023
@vmoens vmoens added the bug Something isn't working label Jul 6, 2023
@vmoens vmoens linked an issue Jul 6, 2023 that may be closed by this pull request
3 tasks
@vmoens vmoens marked this pull request as draft July 6, 2023 15:06
@github-actions
Copy link

github-actions bot commented Jul 6, 2023

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}23$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.2585s 0.2498s 4.0031 Ops/s 3.8543 Ops/s $\color{#35bf28}+3.86\%$
test_sync 0.1412s 0.1319s 7.5797 Ops/s 7.5830 Ops/s $\color{#d91a1a}-0.04\%$
test_async 0.2491s 0.1316s 7.5989 Ops/s 7.4508 Ops/s $\color{#35bf28}+1.99\%$
test_simple 1.0819s 1.0133s 0.9869 Ops/s 1.0010 Ops/s $\color{#d91a1a}-1.41\%$
test_transformed 2.6013s 2.4922s 0.4013 Ops/s 0.4007 Ops/s $\color{#35bf28}+0.14\%$
test_serial 3.1187s 3.0580s 0.3270 Ops/s 0.3219 Ops/s $\color{#35bf28}+1.58\%$
test_parallel 2.5797s 2.4636s 0.4059 Ops/s 0.4111 Ops/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[True-True-True-True-True] 0.8468ms 58.8181μs 17.0016 KOps/s 16.6437 KOps/s $\color{#35bf28}+2.15\%$
test_step_mdp_speed[True-True-True-True-False] 0.2493ms 31.4013μs 31.8458 KOps/s 29.0680 KOps/s $\textbf{\color{#35bf28}+9.56\%}$
test_step_mdp_speed[True-True-True-False-True] 0.4344ms 42.9698μs 23.2722 KOps/s 22.7007 KOps/s $\color{#35bf28}+2.52\%$
test_step_mdp_speed[True-True-True-False-False] 3.1710ms 26.5929μs 37.6041 KOps/s 41.4906 KOps/s $\textbf{\color{#d91a1a}-9.37\%}$
test_step_mdp_speed[True-True-False-True-True] 0.4446ms 59.3753μs 16.8420 KOps/s 16.4246 KOps/s $\color{#35bf28}+2.54\%$
test_step_mdp_speed[True-True-False-True-False] 0.3462ms 34.2896μs 29.1634 KOps/s 28.3311 KOps/s $\color{#35bf28}+2.94\%$
test_step_mdp_speed[True-True-False-False-True] 1.3128ms 45.3182μs 22.0662 KOps/s 21.4281 KOps/s $\color{#35bf28}+2.98\%$
test_step_mdp_speed[True-True-False-False-False] 0.3052ms 25.8929μs 38.6206 KOps/s 38.3827 KOps/s $\color{#35bf28}+0.62\%$
test_step_mdp_speed[True-False-True-True-True] 1.9050ms 61.6492μs 16.2208 KOps/s 16.5196 KOps/s $\color{#d91a1a}-1.81\%$
test_step_mdp_speed[True-False-True-True-False] 3.0511ms 36.6641μs 27.2747 KOps/s 27.5549 KOps/s $\color{#d91a1a}-1.02\%$
test_step_mdp_speed[True-False-True-False-True] 4.3248ms 47.5864μs 21.0144 KOps/s 20.9938 KOps/s $\color{#35bf28}+0.10\%$
test_step_mdp_speed[True-False-True-False-False] 5.1165ms 25.9922μs 38.4731 KOps/s 39.2105 KOps/s $\color{#d91a1a}-1.88\%$
test_step_mdp_speed[True-False-False-True-True] 0.6059ms 64.5844μs 15.4836 KOps/s 15.5099 KOps/s $\color{#d91a1a}-0.17\%$
test_step_mdp_speed[True-False-False-True-False] 2.6531ms 39.0465μs 25.6105 KOps/s 25.9561 KOps/s $\color{#d91a1a}-1.33\%$
test_step_mdp_speed[True-False-False-False-True] 0.9350ms 46.6905μs 21.4176 KOps/s 20.7759 KOps/s $\color{#35bf28}+3.09\%$
test_step_mdp_speed[True-False-False-False-False] 1.1738ms 30.0570μs 33.2701 KOps/s 35.5306 KOps/s $\textbf{\color{#d91a1a}-6.36\%}$
test_step_mdp_speed[False-True-True-True-True] 0.8437ms 64.1561μs 15.5870 KOps/s 15.8195 KOps/s $\color{#d91a1a}-1.47\%$
test_step_mdp_speed[False-True-True-True-False] 0.9282ms 39.4205μs 25.3675 KOps/s 27.1901 KOps/s $\textbf{\color{#d91a1a}-6.70\%}$
test_step_mdp_speed[False-True-True-False-True] 0.5360ms 52.2606μs 19.1349 KOps/s 19.1365 KOps/s $-0.01\%$
test_step_mdp_speed[False-True-True-False-False] 0.2702ms 27.8690μs 35.8821 KOps/s 34.4709 KOps/s $\color{#35bf28}+4.09\%$
test_step_mdp_speed[False-True-False-True-True] 8.4403ms 63.5331μs 15.7398 KOps/s 15.6484 KOps/s $\color{#35bf28}+0.58\%$
test_step_mdp_speed[False-True-False-True-False] 5.8093ms 38.3772μs 26.0572 KOps/s 26.1631 KOps/s $\color{#d91a1a}-0.40\%$
test_step_mdp_speed[False-True-False-False-True] 1.4964ms 53.6401μs 18.6428 KOps/s 18.4485 KOps/s $\color{#35bf28}+1.05\%$
test_step_mdp_speed[False-True-False-False-False] 0.3210ms 29.7944μs 33.5633 KOps/s 30.0483 KOps/s $\textbf{\color{#35bf28}+11.70\%}$
test_step_mdp_speed[False-False-True-True-True] 4.5215ms 65.6631μs 15.2293 KOps/s 15.2757 KOps/s $\color{#d91a1a}-0.30\%$
test_step_mdp_speed[False-False-True-True-False] 0.4020ms 40.1113μs 24.9307 KOps/s 25.7637 KOps/s $\color{#d91a1a}-3.23\%$
test_step_mdp_speed[False-False-True-False-True] 0.9039ms 64.2782μs 15.5574 KOps/s 19.8516 KOps/s $\textbf{\color{#d91a1a}-21.63\%}$
test_step_mdp_speed[False-False-True-False-False] 0.5094ms 29.4754μs 33.9266 KOps/s 33.2393 KOps/s $\color{#35bf28}+2.07\%$
test_step_mdp_speed[False-False-False-True-True] 0.4468ms 63.6450μs 15.7122 KOps/s 15.0016 KOps/s $\color{#35bf28}+4.74\%$
test_step_mdp_speed[False-False-False-True-False] 6.4640ms 42.2749μs 23.6547 KOps/s 23.6699 KOps/s $\color{#d91a1a}-0.06\%$
test_step_mdp_speed[False-False-False-False-True] 0.3553ms 52.3508μs 19.1019 KOps/s 18.2088 KOps/s $\color{#35bf28}+4.91\%$
test_step_mdp_speed[False-False-False-False-False] 3.9422ms 31.7676μs 31.4786 KOps/s 30.5733 KOps/s $\color{#35bf28}+2.96\%$
test_values[generalized_advantage_estimate-True-True] 27.9291ms 19.8045ms 50.4935 Ops/s 52.6452 Ops/s $\color{#d91a1a}-4.09\%$
test_values[vec_generalized_advantage_estimate-True-True] 0.1007s 71.3140ms 14.0225 Ops/s 12.9969 Ops/s $\textbf{\color{#35bf28}+7.89\%}$
test_values[td0_return_estimate-False-False] 1.5055ms 0.3502ms 2.8552 KOps/s 2.9448 KOps/s $\color{#d91a1a}-3.04\%$
test_values[td1_return_estimate-False-False] 21.2390ms 18.5198ms 53.9963 Ops/s 52.3476 Ops/s $\color{#35bf28}+3.15\%$
test_values[vec_td1_return_estimate-False-False] 84.6597ms 71.6533ms 13.9561 Ops/s 13.0057 Ops/s $\textbf{\color{#35bf28}+7.31\%}$
test_values[td_lambda_return_estimate-True-False] 66.7262ms 54.1161ms 18.4788 Ops/s 19.2982 Ops/s $\color{#d91a1a}-4.25\%$
test_values[vec_td_lambda_return_estimate-True-False] 86.3648ms 73.7850ms 13.5529 Ops/s 13.0549 Ops/s $\color{#35bf28}+3.81\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 26.6303ms 16.7161ms 59.8227 Ops/s 67.1761 Ops/s $\textbf{\color{#d91a1a}-10.95\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 12.8881ms 5.0148ms 199.4116 Ops/s 220.9321 Ops/s $\textbf{\color{#d91a1a}-9.74\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 7.7826ms 0.7514ms 1.3309 KOps/s 1.3395 KOps/s $\color{#d91a1a}-0.64\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 87.0230ms 77.7937ms 12.8545 Ops/s 12.4101 Ops/s $\color{#35bf28}+3.58\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 15.6116ms 6.5323ms 153.0860 Ops/s 156.7526 Ops/s $\color{#d91a1a}-2.34\%$
test_dqn_speed 8.4013ms 2.7131ms 368.5806 Ops/s 372.9092 Ops/s $\color{#d91a1a}-1.16\%$
test_ddpg_speed 7.1399ms 5.0736ms 197.0995 Ops/s 204.1247 Ops/s $\color{#d91a1a}-3.44\%$
test_sac_speed 0.1049s 15.4756ms 64.6179 Ops/s 73.6051 Ops/s $\textbf{\color{#d91a1a}-12.21\%}$
test_redq_speed 40.4366ms 26.0910ms 38.3274 Ops/s 34.6467 Ops/s $\textbf{\color{#35bf28}+10.62\%}$
test_redq_deprec_speed 27.6348ms 21.4826ms 46.5493 Ops/s 45.7337 Ops/s $\color{#35bf28}+1.78\%$
test_td3_speed 22.5442ms 18.3857ms 54.3901 Ops/s 52.1747 Ops/s $\color{#35bf28}+4.25\%$
test_cql_speed 74.3574ms 54.1979ms 18.4509 Ops/s 17.4853 Ops/s $\textbf{\color{#35bf28}+5.52\%}$
test_a2c_speed 15.7669ms 11.4573ms 87.2809 Ops/s 80.4796 Ops/s $\textbf{\color{#35bf28}+8.45\%}$
test_ppo_speed 19.3251ms 12.2332ms 81.7451 Ops/s 77.2631 Ops/s $\textbf{\color{#35bf28}+5.80\%}$
test_reinforce_speed 19.0955ms 9.6621ms 103.4970 Ops/s 101.3159 Ops/s $\color{#35bf28}+2.15\%$
test_iql_speed 58.4779ms 49.6871ms 20.1260 Ops/s 19.8543 Ops/s $\color{#35bf28}+1.37\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 8.1929ms 5.3080ms 188.3933 Ops/s 167.3093 Ops/s $\textbf{\color{#35bf28}+12.60\%}$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 7.5435ms 5.7166ms 174.9303 Ops/s 186.8291 Ops/s $\textbf{\color{#d91a1a}-6.37\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 10.4442ms 5.4417ms 183.7656 Ops/s 147.3323 Ops/s $\textbf{\color{#35bf28}+24.73\%}$
test_sample_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 7.3527ms 5.1716ms 193.3622 Ops/s 175.1796 Ops/s $\textbf{\color{#35bf28}+10.38\%}$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 12.0685ms 5.3546ms 186.7539 Ops/s 170.5653 Ops/s $\textbf{\color{#35bf28}+9.49\%}$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 9.7536ms 5.8528ms 170.8578 Ops/s 173.7140 Ops/s $\color{#d91a1a}-1.64\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 10.4226ms 5.0980ms 196.1535 Ops/s 158.4697 Ops/s $\textbf{\color{#35bf28}+23.78\%}$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 12.7131ms 5.5673ms 179.6213 Ops/s 181.9562 Ops/s $\color{#d91a1a}-1.28\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 9.5775ms 5.4053ms 185.0033 Ops/s 146.4281 Ops/s $\textbf{\color{#35bf28}+26.34\%}$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 9.9733ms 5.1068ms 195.8164 Ops/s 182.3267 Ops/s $\textbf{\color{#35bf28}+7.40\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 8.2416ms 5.3218ms 187.9062 Ops/s 178.0248 Ops/s $\textbf{\color{#35bf28}+5.55\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 6.8331ms 5.2653ms 189.9234 Ops/s 174.0384 Ops/s $\textbf{\color{#35bf28}+9.13\%}$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 6.6399ms 5.1633ms 193.6735 Ops/s 161.4583 Ops/s $\textbf{\color{#35bf28}+19.95\%}$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 11.0060ms 5.5624ms 179.7781 Ops/s 179.0172 Ops/s $\color{#35bf28}+0.43\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 7.3345ms 5.2273ms 191.3024 Ops/s 147.3562 Ops/s $\textbf{\color{#35bf28}+29.82\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 8.2081ms 5.1978ms 192.3881 Ops/s 188.5164 Ops/s $\color{#35bf28}+2.05\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 14.5615ms 5.5029ms 181.7215 Ops/s 173.6230 Ops/s $\color{#35bf28}+4.66\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 7.1235ms 5.2904ms 189.0208 Ops/s 172.7644 Ops/s $\textbf{\color{#35bf28}+9.41\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.3337s 50.4337ms 19.8280 Ops/s 18.0555 Ops/s $\textbf{\color{#35bf28}+9.82\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 0.2019s 45.2988ms 22.0757 Ops/s 21.0359 Ops/s $\color{#35bf28}+4.94\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 0.2021s 45.5284ms 21.9643 Ops/s 20.9816 Ops/s $\color{#35bf28}+4.68\%$
test_populate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1958s 45.0191ms 22.2128 Ops/s 20.5232 Ops/s $\textbf{\color{#35bf28}+8.23\%}$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1915s 44.6096ms 22.4167 Ops/s 21.8315 Ops/s $\color{#35bf28}+2.68\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.2018s 49.1573ms 20.3429 Ops/s 20.5517 Ops/s $\color{#d91a1a}-1.02\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.2003s 44.3932ms 22.5259 Ops/s 20.9966 Ops/s $\textbf{\color{#35bf28}+7.28\%}$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.2155s 45.1307ms 22.1578 Ops/s 21.8673 Ops/s $\color{#35bf28}+1.33\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 0.2037s 44.7663ms 22.3382 Ops/s 21.7710 Ops/s $\color{#35bf28}+2.61\%$

@github-actions
Copy link

github-actions bot commented Jul 6, 2023

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1642s 0.1641s 6.0934 Ops/s 5.9781 Ops/s $\color{#35bf28}+1.93\%$
test_sync 93.1942ms 88.6945ms 11.2747 Ops/s 11.3806 Ops/s $\color{#d91a1a}-0.93\%$
test_async 0.1861s 84.9233ms 11.7753 Ops/s 11.6996 Ops/s $\color{#35bf28}+0.65\%$
test_simple 0.7222s 0.6522s 1.5333 Ops/s 1.5216 Ops/s $\color{#35bf28}+0.77\%$
test_transformed 1.7368s 1.6823s 0.5944 Ops/s 0.5918 Ops/s $\color{#35bf28}+0.44\%$
test_serial 2.0645s 2.0056s 0.4986 Ops/s 0.4947 Ops/s $\color{#35bf28}+0.78\%$
test_parallel 1.6655s 1.5530s 0.6439 Ops/s 0.6401 Ops/s $\color{#35bf28}+0.60\%$
test_step_mdp_speed[True-True-True-True-True] 0.1964ms 36.2777μs 27.5652 KOps/s 27.6965 KOps/s $\color{#d91a1a}-0.47\%$
test_step_mdp_speed[True-True-True-True-False] 53.6000μs 20.6391μs 48.4517 KOps/s 49.4263 KOps/s $\color{#d91a1a}-1.97\%$
test_step_mdp_speed[True-True-True-False-True] 0.5169ms 25.6189μs 39.0336 KOps/s 39.0959 KOps/s $\color{#d91a1a}-0.16\%$
test_step_mdp_speed[True-True-True-False-False] 93.2010μs 14.1492μs 70.6756 KOps/s 71.5247 KOps/s $\color{#d91a1a}-1.19\%$
test_step_mdp_speed[True-True-False-True-True] 0.1275ms 37.6897μs 26.5325 KOps/s 26.8736 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[True-True-False-True-False] 72.5000μs 21.9274μs 45.6050 KOps/s 46.3827 KOps/s $\color{#d91a1a}-1.68\%$
test_step_mdp_speed[True-True-False-False-True] 49.9000μs 27.1492μs 36.8336 KOps/s 37.2038 KOps/s $\color{#d91a1a}-1.00\%$
test_step_mdp_speed[True-True-False-False-False] 0.1908ms 15.7430μs 63.5205 KOps/s 64.3394 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[True-False-True-True-True] 60.4010μs 39.2861μs 25.4543 KOps/s 25.8890 KOps/s $\color{#d91a1a}-1.68\%$
test_step_mdp_speed[True-False-True-True-False] 0.1180ms 23.0006μs 43.4772 KOps/s 43.1878 KOps/s $\color{#35bf28}+0.67\%$
test_step_mdp_speed[True-False-True-False-True] 87.4010μs 27.1724μs 36.8021 KOps/s 37.0784 KOps/s $\color{#d91a1a}-0.75\%$
test_step_mdp_speed[True-False-True-False-False] 41.9010μs 15.5177μs 64.4427 KOps/s 64.1274 KOps/s $\color{#35bf28}+0.49\%$
test_step_mdp_speed[True-False-False-True-True] 60.5000μs 40.1655μs 24.8970 KOps/s 24.8227 KOps/s $\color{#35bf28}+0.30\%$
test_step_mdp_speed[True-False-False-True-False] 0.1461ms 24.6116μs 40.6313 KOps/s 41.0143 KOps/s $\color{#d91a1a}-0.93\%$
test_step_mdp_speed[True-False-False-False-True] 0.1221ms 28.5783μs 34.9916 KOps/s 35.9820 KOps/s $\color{#d91a1a}-2.75\%$
test_step_mdp_speed[True-False-False-False-False] 51.5010μs 16.9530μs 58.9865 KOps/s 59.0509 KOps/s $\color{#d91a1a}-0.11\%$
test_step_mdp_speed[False-True-True-True-True] 65.9000μs 38.6422μs 25.8784 KOps/s 25.5237 KOps/s $\color{#35bf28}+1.39\%$
test_step_mdp_speed[False-True-True-True-False] 50.8000μs 23.0838μs 43.3203 KOps/s 43.3731 KOps/s $\color{#d91a1a}-0.12\%$
test_step_mdp_speed[False-True-True-False-True] 56.7000μs 31.1114μs 32.1426 KOps/s 31.6447 KOps/s $\color{#35bf28}+1.57\%$
test_step_mdp_speed[False-True-True-False-False] 61.9010μs 17.5200μs 57.0776 KOps/s 57.8104 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[False-True-False-True-True] 66.1010μs 39.9129μs 25.0545 KOps/s 24.6032 KOps/s $\color{#35bf28}+1.83\%$
test_step_mdp_speed[False-True-False-True-False] 54.9000μs 24.3628μs 41.0462 KOps/s 41.0227 KOps/s $\color{#35bf28}+0.06\%$
test_step_mdp_speed[False-True-False-False-True] 92.4000μs 32.4347μs 30.8312 KOps/s 30.1980 KOps/s $\color{#35bf28}+2.10\%$
test_step_mdp_speed[False-True-False-False-False] 49.7000μs 18.8048μs 53.1779 KOps/s 53.2172 KOps/s $\color{#d91a1a}-0.07\%$
test_step_mdp_speed[False-False-True-True-True] 0.1111ms 41.0853μs 24.3396 KOps/s 24.2506 KOps/s $\color{#35bf28}+0.37\%$
test_step_mdp_speed[False-False-True-True-False] 50.5000μs 26.0430μs 38.3980 KOps/s 38.8271 KOps/s $\color{#d91a1a}-1.11\%$
test_step_mdp_speed[False-False-True-False-True] 47.4000μs 33.4518μs 29.8938 KOps/s 30.0083 KOps/s $\color{#d91a1a}-0.38\%$
test_step_mdp_speed[False-False-True-False-False] 50.0010μs 18.5603μs 53.8783 KOps/s 54.5447 KOps/s $\color{#d91a1a}-1.22\%$
test_step_mdp_speed[False-False-False-True-True] 93.5000μs 42.7543μs 23.3895 KOps/s 23.5077 KOps/s $\color{#d91a1a}-0.50\%$
test_step_mdp_speed[False-False-False-True-False] 81.3010μs 27.1543μs 36.8266 KOps/s 36.8051 KOps/s $\color{#35bf28}+0.06\%$
test_step_mdp_speed[False-False-False-False-True] 60.7000μs 33.7928μs 29.5921 KOps/s 29.4904 KOps/s $\color{#35bf28}+0.34\%$
test_step_mdp_speed[False-False-False-False-False] 74.5000μs 19.7444μs 50.6473 KOps/s 50.9456 KOps/s $\color{#d91a1a}-0.59\%$
test_values[generalized_advantage_estimate-True-True] 13.8675ms 13.2473ms 75.4871 Ops/s 75.1282 Ops/s $\color{#35bf28}+0.48\%$
test_values[vec_generalized_advantage_estimate-True-True] 54.1172ms 41.3996ms 24.1548 Ops/s 24.1932 Ops/s $\color{#d91a1a}-0.16\%$
test_values[td0_return_estimate-False-False] 0.2884ms 0.2076ms 4.8162 KOps/s 4.5457 KOps/s $\textbf{\color{#35bf28}+5.95\%}$
test_values[td1_return_estimate-False-False] 13.3180ms 13.0340ms 76.7226 Ops/s 77.1074 Ops/s $\color{#d91a1a}-0.50\%$
test_values[vec_td1_return_estimate-False-False] 42.2230ms 40.9557ms 24.4166 Ops/s 24.5331 Ops/s $\color{#d91a1a}-0.47\%$
test_values[td_lambda_return_estimate-True-False] 31.6419ms 31.2560ms 31.9938 Ops/s 31.6459 Ops/s $\color{#35bf28}+1.10\%$
test_values[vec_td_lambda_return_estimate-True-False] 49.3956ms 41.1441ms 24.3048 Ops/s 24.4807 Ops/s $\color{#d91a1a}-0.72\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 11.6685ms 11.5402ms 86.6536 Ops/s 86.9332 Ops/s $\color{#d91a1a}-0.32\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 6.9852ms 3.5390ms 282.5674 Ops/s 291.8082 Ops/s $\color{#d91a1a}-3.17\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.8313ms 0.4764ms 2.0990 KOps/s 2.1262 KOps/s $\color{#d91a1a}-1.28\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 49.7106ms 44.1001ms 22.6757 Ops/s 19.2724 Ops/s $\textbf{\color{#35bf28}+17.66\%}$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 4.7054ms 2.9197ms 342.4981 Ops/s 359.9323 Ops/s $\color{#d91a1a}-4.84\%$
test_dqn_speed 2.5747ms 1.6494ms 606.2700 Ops/s 609.6605 Ops/s $\color{#d91a1a}-0.56\%$
test_ddpg_speed 3.0779ms 2.7270ms 366.6971 Ops/s 369.3497 Ops/s $\color{#d91a1a}-0.72\%$
test_sac_speed 8.9656ms 8.2052ms 121.8738 Ops/s 122.6482 Ops/s $\color{#d91a1a}-0.63\%$
test_redq_speed 99.0631ms 16.0560ms 62.2820 Ops/s 67.8486 Ops/s $\textbf{\color{#d91a1a}-8.20\%}$
test_redq_deprec_speed 13.2072ms 12.2420ms 81.6860 Ops/s 82.0070 Ops/s $\color{#d91a1a}-0.39\%$
test_td3_speed 17.5835ms 11.1325ms 89.8272 Ops/s 88.6361 Ops/s $\color{#35bf28}+1.34\%$
test_cql_speed 39.9105ms 26.7790ms 37.3427 Ops/s 36.6333 Ops/s $\color{#35bf28}+1.94\%$
test_a2c_speed 5.9837ms 5.1734ms 193.2973 Ops/s 171.8824 Ops/s $\textbf{\color{#35bf28}+12.46\%}$
test_ppo_speed 7.1278ms 5.6955ms 175.5765 Ops/s 159.0474 Ops/s $\textbf{\color{#35bf28}+10.39\%}$
test_reinforce_speed 4.6537ms 4.0842ms 244.8479 Ops/s 222.9669 Ops/s $\textbf{\color{#35bf28}+9.81\%}$
test_iql_speed 22.3393ms 21.3447ms 46.8500 Ops/s 45.0641 Ops/s $\color{#35bf28}+3.96\%$
test_sample_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 4.3109ms 3.6350ms 275.1052 Ops/s 273.7810 Ops/s $\color{#35bf28}+0.48\%$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 7.6553ms 3.7889ms 263.9291 Ops/s 260.6293 Ops/s $\color{#35bf28}+1.27\%$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 6.1363ms 3.7906ms 263.8136 Ops/s 226.4101 Ops/s $\textbf{\color{#35bf28}+16.52\%}$
test_sample_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 0.1255s 4.0990ms 243.9625 Ops/s 184.3589 Ops/s $\textbf{\color{#35bf28}+32.33\%}$
test_sample_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 6.6949ms 3.7758ms 264.8455 Ops/s 262.1238 Ops/s $\color{#35bf28}+1.04\%$
test_sample_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.1669s 4.3470ms 230.0458 Ops/s 259.7072 Ops/s $\textbf{\color{#d91a1a}-11.42\%}$
test_sample_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 5.0151ms 3.6065ms 277.2768 Ops/s 272.0347 Ops/s $\color{#35bf28}+1.93\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 6.4302ms 3.7488ms 266.7494 Ops/s 263.0281 Ops/s $\color{#35bf28}+1.41\%$
test_sample_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 6.1052ms 3.7899ms 263.8604 Ops/s 223.6586 Ops/s $\textbf{\color{#35bf28}+17.97\%}$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 4.9171ms 3.5605ms 280.8564 Ops/s 278.0300 Ops/s $\color{#35bf28}+1.02\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 5.9427ms 3.7121ms 269.3880 Ops/s 266.3203 Ops/s $\color{#35bf28}+1.15\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 6.9388ms 3.7848ms 264.2154 Ops/s 266.7480 Ops/s $\color{#d91a1a}-0.95\%$
test_iterate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 0.1242s 4.0096ms 249.4036 Ops/s 251.9758 Ops/s $\color{#d91a1a}-1.02\%$
test_iterate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 6.2350ms 3.7908ms 263.7998 Ops/s 267.3454 Ops/s $\color{#d91a1a}-1.33\%$
test_iterate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 5.8681ms 3.8204ms 261.7559 Ops/s 286.4464 Ops/s $\textbf{\color{#d91a1a}-8.62\%}$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 4.8577ms 3.6212ms 276.1497 Ops/s 283.5227 Ops/s $\color{#d91a1a}-2.60\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 7.8801ms 3.8095ms 262.5023 Ops/s 266.4490 Ops/s $\color{#d91a1a}-1.48\%$
test_iterate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 6.6848ms 3.7726ms 265.0716 Ops/s 263.3309 Ops/s $\color{#35bf28}+0.66\%$
test_populate_rb[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.2793s 35.4620ms 28.1992 Ops/s 29.6046 Ops/s $\color{#d91a1a}-4.75\%$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 0.1622s 30.4539ms 32.8365 Ops/s 33.0088 Ops/s $\color{#d91a1a}-0.52\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 0.1647s 30.9570ms 32.3029 Ops/s 29.8341 Ops/s $\textbf{\color{#35bf28}+8.28\%}$
test_populate_rb[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1746s 30.9346ms 32.3263 Ops/s 32.8966 Ops/s $\color{#d91a1a}-1.73\%$
test_populate_rb[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 0.1669s 30.5330ms 32.7515 Ops/s 33.1401 Ops/s $\color{#d91a1a}-1.17\%$
test_populate_rb[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 0.1582s 30.0883ms 33.2355 Ops/s 33.3453 Ops/s $\color{#d91a1a}-0.33\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1590s 30.2641ms 33.0425 Ops/s 33.8679 Ops/s $\color{#d91a1a}-2.44\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 0.1670s 30.9904ms 32.2681 Ops/s 31.3333 Ops/s $\color{#35bf28}+2.98\%$
test_populate_rb[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 0.1648s 30.7020ms 32.5711 Ops/s 33.7984 Ops/s $\color{#d91a1a}-3.63\%$

@facebook-github-bot
Copy link

Hi @vmoens!

Thank you for your pull request.

We require contributors to sign our Contributor License Agreement, and yours needs attention.

You currently have a record in our system, but the CLA is no longer valid, and will need to be resubmitted.

Process

In order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA.

Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with CLA signed. The tagging process may take up to 1 hour after signing. Please give it that time before contacting us about it.

If you have received this in error or have any questions, please contact us at [email protected]. Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[BUG] Operations on Spec of Stacked CompositeSpec with different keys crashes

2 participants