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] Atari DQN dataset #1815

Merged
merged 18 commits into from
Jan 22, 2024
Merged

[Feature] Atari DQN dataset #1815

merged 18 commits into from
Jan 22, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Jan 18, 2024

Docstrings can be read here

cc @agarwl: do you think we could add this in https://github.com/mila-iqia/SGI/blob/master/src/offline_dataset.py after the release (next week)?

@vmoens vmoens added the Data Data-related PR, will launch data-related jobs label Jan 18, 2024
Copy link

pytorch-bot bot commented Jan 18, 2024

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/rl/1815

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

⏳ 1 Pending, 1 Unrelated Failure

As of commit 0e0982d with merge base a10cdbf (image):

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.

@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 Jan 18, 2024
Copy link

github-actions bot commented Jan 18, 2024

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

Total Benchmarks: 92. Improved: $\large\color{#35bf28}1$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1171s 0.1163s 8.6011 Ops/s 8.6407 Ops/s $\color{#d91a1a}-0.46\%$
test_sync 0.1717s 0.1026s 9.7429 Ops/s 9.7170 Ops/s $\color{#35bf28}+0.27\%$
test_async 0.1819s 91.9587ms 10.8744 Ops/s 10.8712 Ops/s $\color{#35bf28}+0.03\%$
test_single_pixels 0.1292s 0.1271s 7.8649 Ops/s 7.1455 Ops/s $\textbf{\color{#35bf28}+10.07\%}$
test_sync_pixels 79.0428ms 76.6967ms 13.0384 Ops/s 12.6919 Ops/s $\color{#35bf28}+2.73\%$
test_async_pixels 0.1374s 72.5443ms 13.7847 Ops/s 13.8686 Ops/s $\color{#d91a1a}-0.61\%$
test_simple 0.9140s 0.8439s 1.1850 Ops/s 1.2036 Ops/s $\color{#d91a1a}-1.54\%$
test_transformed 1.0640s 1.0633s 0.9405 Ops/s 0.9349 Ops/s $\color{#35bf28}+0.59\%$
test_serial 2.3872s 2.3229s 0.4305 Ops/s 0.4281 Ops/s $\color{#35bf28}+0.57\%$
test_parallel 2.0803s 1.9059s 0.5247 Ops/s 0.5348 Ops/s $\color{#d91a1a}-1.89\%$
test_step_mdp_speed[True-True-True-True-True] 75.1410μs 33.1167μs 30.1962 KOps/s 30.5142 KOps/s $\color{#d91a1a}-1.04\%$
test_step_mdp_speed[True-True-True-True-False] 38.5710μs 19.9424μs 50.1444 KOps/s 50.2110 KOps/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[True-True-True-False-True] 41.2800μs 18.5237μs 53.9850 KOps/s 53.6888 KOps/s $\color{#35bf28}+0.55\%$
test_step_mdp_speed[True-True-True-False-False] 36.0210μs 11.1926μs 89.3445 KOps/s 89.5992 KOps/s $\color{#d91a1a}-0.28\%$
test_step_mdp_speed[True-True-False-True-True] 54.8910μs 34.9665μs 28.5988 KOps/s 29.0366 KOps/s $\color{#d91a1a}-1.51\%$
test_step_mdp_speed[True-True-False-True-False] 39.6200μs 21.6689μs 46.1491 KOps/s 46.7448 KOps/s $\color{#d91a1a}-1.27\%$
test_step_mdp_speed[True-True-False-False-True] 49.1610μs 20.4192μs 48.9735 KOps/s 49.3891 KOps/s $\color{#d91a1a}-0.84\%$
test_step_mdp_speed[True-True-False-False-False] 32.3610μs 13.0812μs 76.4458 KOps/s 76.3410 KOps/s $\color{#35bf28}+0.14\%$
test_step_mdp_speed[True-False-True-True-True] 57.5310μs 36.8670μs 27.1245 KOps/s 27.2284 KOps/s $\color{#d91a1a}-0.38\%$
test_step_mdp_speed[True-False-True-True-False] 45.9210μs 23.6148μs 42.3463 KOps/s 42.2890 KOps/s $\color{#35bf28}+0.14\%$
test_step_mdp_speed[True-False-True-False-True] 48.3510μs 20.2450μs 49.3949 KOps/s 48.7374 KOps/s $\color{#35bf28}+1.35\%$
test_step_mdp_speed[True-False-True-False-False] 32.8500μs 13.0943μs 76.3693 KOps/s 75.1811 KOps/s $\color{#35bf28}+1.58\%$
test_step_mdp_speed[True-False-False-True-True] 64.2510μs 38.8013μs 25.7723 KOps/s 26.1435 KOps/s $\color{#d91a1a}-1.42\%$
test_step_mdp_speed[True-False-False-True-False] 48.6700μs 25.5758μs 39.0995 KOps/s 39.2905 KOps/s $\color{#d91a1a}-0.49\%$
test_step_mdp_speed[True-False-False-False-True] 46.0310μs 22.2690μs 44.9055 KOps/s 45.1409 KOps/s $\color{#d91a1a}-0.52\%$
test_step_mdp_speed[True-False-False-False-False] 32.9800μs 14.9190μs 67.0285 KOps/s 67.4081 KOps/s $\color{#d91a1a}-0.56\%$
test_step_mdp_speed[False-True-True-True-True] 77.2920μs 37.1599μs 26.9107 KOps/s 27.1295 KOps/s $\color{#d91a1a}-0.81\%$
test_step_mdp_speed[False-True-True-True-False] 53.5010μs 23.5118μs 42.5318 KOps/s 42.6264 KOps/s $\color{#d91a1a}-0.22\%$
test_step_mdp_speed[False-True-True-False-True] 45.8410μs 23.9668μs 41.7243 KOps/s 41.0519 KOps/s $\color{#35bf28}+1.64\%$
test_step_mdp_speed[False-True-True-False-False] 41.4210μs 14.8866μs 67.1745 KOps/s 66.4624 KOps/s $\color{#35bf28}+1.07\%$
test_step_mdp_speed[False-True-False-True-True] 79.3410μs 38.6827μs 25.8513 KOps/s 26.2776 KOps/s $\color{#d91a1a}-1.62\%$
test_step_mdp_speed[False-True-False-True-False] 56.6810μs 25.9702μs 38.5057 KOps/s 39.6142 KOps/s $\color{#d91a1a}-2.80\%$
test_step_mdp_speed[False-True-False-False-True] 51.0610μs 26.2515μs 38.0931 KOps/s 39.3235 KOps/s $\color{#d91a1a}-3.13\%$
test_step_mdp_speed[False-True-False-False-False] 57.0310μs 16.8067μs 59.5002 KOps/s 60.2476 KOps/s $\color{#d91a1a}-1.24\%$
test_step_mdp_speed[False-False-True-True-True] 64.6510μs 40.4296μs 24.7343 KOps/s 25.1762 KOps/s $\color{#d91a1a}-1.76\%$
test_step_mdp_speed[False-False-True-True-False] 46.5000μs 27.4158μs 36.4753 KOps/s 37.2471 KOps/s $\color{#d91a1a}-2.07\%$
test_step_mdp_speed[False-False-True-False-True] 51.5310μs 25.5285μs 39.1719 KOps/s 38.9293 KOps/s $\color{#35bf28}+0.62\%$
test_step_mdp_speed[False-False-True-False-False] 35.5910μs 16.6657μs 60.0035 KOps/s 59.7161 KOps/s $\color{#35bf28}+0.48\%$
test_step_mdp_speed[False-False-False-True-True] 67.1010μs 42.0520μs 23.7801 KOps/s 23.7930 KOps/s $\color{#d91a1a}-0.05\%$
test_step_mdp_speed[False-False-False-True-False] 49.7910μs 28.9818μs 34.5044 KOps/s 34.8339 KOps/s $\color{#d91a1a}-0.95\%$
test_step_mdp_speed[False-False-False-False-True] 49.3510μs 27.5321μs 36.3212 KOps/s 37.0445 KOps/s $\color{#d91a1a}-1.95\%$
test_step_mdp_speed[False-False-False-False-False] 45.6610μs 18.7604μs 53.3038 KOps/s 54.2605 KOps/s $\color{#d91a1a}-1.76\%$
test_values[generalized_advantage_estimate-True-True] 28.2515ms 27.0823ms 36.9244 Ops/s 38.6746 Ops/s $\color{#d91a1a}-4.53\%$
test_values[vec_generalized_advantage_estimate-True-True] 88.5038ms 3.3392ms 299.4754 Ops/s 307.7486 Ops/s $\color{#d91a1a}-2.69\%$
test_values[td0_return_estimate-False-False] 91.0110μs 63.4353μs 15.7641 KOps/s 16.1252 KOps/s $\color{#d91a1a}-2.24\%$
test_values[td1_return_estimate-False-False] 59.0149ms 58.0128ms 17.2376 Ops/s 18.6384 Ops/s $\textbf{\color{#d91a1a}-7.52\%}$
test_values[vec_td1_return_estimate-False-False] 2.1249ms 1.7882ms 559.2217 Ops/s 552.6371 Ops/s $\color{#35bf28}+1.19\%$
test_values[td_lambda_return_estimate-True-False] 95.1051ms 93.1536ms 10.7350 Ops/s 11.6355 Ops/s $\textbf{\color{#d91a1a}-7.74\%}$
test_values[vec_td_lambda_return_estimate-True-False] 2.0601ms 1.7969ms 556.4991 Ops/s 566.0408 Ops/s $\color{#d91a1a}-1.69\%$
test_gae_speed[generalized_advantage_estimate-False-1-512] 25.4344ms 24.9123ms 40.1407 Ops/s 42.5736 Ops/s $\textbf{\color{#d91a1a}-5.71\%}$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 0.9339ms 0.7125ms 1.4034 KOps/s 1.3968 KOps/s $\color{#35bf28}+0.47\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 0.7393ms 0.6873ms 1.4550 KOps/s 1.5191 KOps/s $\color{#d91a1a}-4.22\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 1.5401ms 1.4827ms 674.4316 Ops/s 683.8876 Ops/s $\color{#d91a1a}-1.38\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 0.9631ms 0.7068ms 1.4149 KOps/s 1.4718 KOps/s $\color{#d91a1a}-3.86\%$
test_dqn_speed 7.8688ms 7.3441ms 136.1629 Ops/s 137.1443 Ops/s $\color{#d91a1a}-0.72\%$
test_ddpg_speed 0.1010s 15.8178ms 63.2199 Ops/s 69.5576 Ops/s $\textbf{\color{#d91a1a}-9.11\%}$
test_sac_speed 30.0986ms 29.4181ms 33.9927 Ops/s 34.7642 Ops/s $\color{#d91a1a}-2.22\%$
test_redq_speed 50.4758ms 48.4779ms 20.6279 Ops/s 21.1746 Ops/s $\color{#d91a1a}-2.58\%$
test_redq_deprec_speed 25.0346ms 24.0688ms 41.5475 Ops/s 41.9742 Ops/s $\color{#d91a1a}-1.02\%$
test_td3_speed 29.2146ms 19.5965ms 51.0295 Ops/s 50.9837 Ops/s $\color{#35bf28}+0.09\%$
test_cql_speed 83.1340ms 81.7258ms 12.2360 Ops/s 12.1952 Ops/s $\color{#35bf28}+0.33\%$
test_a2c_speed 27.3600ms 26.0970ms 38.3186 Ops/s 38.0223 Ops/s $\color{#35bf28}+0.78\%$
test_ppo_speed 27.2443ms 26.4091ms 37.8657 Ops/s 37.3927 Ops/s $\color{#35bf28}+1.27\%$
test_reinforce_speed 26.1080ms 25.1572ms 39.7501 Ops/s 39.3985 Ops/s $\color{#35bf28}+0.89\%$
test_iql_speed 56.8452ms 56.1122ms 17.8214 Ops/s 17.6454 Ops/s $\color{#35bf28}+1.00\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.2566ms 1.8694ms 534.9343 Ops/s 539.4562 Ops/s $\color{#d91a1a}-0.84\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9629ms 0.8421ms 1.1875 KOps/s 1.1843 KOps/s $\color{#35bf28}+0.27\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9425ms 0.8184ms 1.2219 KOps/s 1.2203 KOps/s $\color{#35bf28}+0.13\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.7303ms 1.8268ms 547.4069 Ops/s 550.9535 Ops/s $\color{#d91a1a}-0.64\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.9822ms 0.8312ms 1.2030 KOps/s 1.2031 KOps/s $-0.01\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9524ms 0.8086ms 1.2367 KOps/s 1.2354 KOps/s $\color{#35bf28}+0.11\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 0.1242s 2.3496ms 425.6040 Ops/s 482.2052 Ops/s $\textbf{\color{#d91a1a}-11.74\%}$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.0853ms 0.9571ms 1.0448 KOps/s 1.0372 KOps/s $\color{#35bf28}+0.74\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0505ms 0.9355ms 1.0690 KOps/s 1.0637 KOps/s $\color{#35bf28}+0.49\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.5905ms 1.8522ms 539.8947 Ops/s 543.4935 Ops/s $\color{#d91a1a}-0.66\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 0.9631ms 0.8418ms 1.1879 KOps/s 1.1840 KOps/s $\color{#35bf28}+0.33\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.9493ms 0.8190ms 1.2210 KOps/s 1.2178 KOps/s $\color{#35bf28}+0.26\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.6086ms 1.8228ms 548.6161 Ops/s 553.4344 Ops/s $\color{#d91a1a}-0.87\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 0.1131s 0.9861ms 1.0141 KOps/s 1.1997 KOps/s $\textbf{\color{#d91a1a}-15.48\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.9386ms 0.8074ms 1.2386 KOps/s 1.2335 KOps/s $\color{#35bf28}+0.41\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.9236ms 2.0793ms 480.9298 Ops/s 483.6811 Ops/s $\color{#d91a1a}-0.57\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 1.1269ms 0.9627ms 1.0387 KOps/s 1.0402 KOps/s $\color{#d91a1a}-0.14\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 1.0676ms 0.9389ms 1.0651 KOps/s 1.0623 KOps/s $\color{#35bf28}+0.26\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1159s 9.8583ms 101.4371 Ops/s 99.3270 Ops/s $\color{#35bf28}+2.12\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 16.7491ms 14.3622ms 69.6272 Ops/s 70.5309 Ops/s $\color{#d91a1a}-1.28\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.0508ms 3.4441ms 290.3530 Ops/s 288.7905 Ops/s $\color{#35bf28}+0.54\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1168s 9.9153ms 100.8541 Ops/s 100.2807 Ops/s $\color{#35bf28}+0.57\%$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 16.7631ms 14.3094ms 69.8842 Ops/s 69.1079 Ops/s $\color{#35bf28}+1.12\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 5.5394ms 3.4297ms 291.5711 Ops/s 287.4130 Ops/s $\color{#35bf28}+1.45\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1201s 12.3997ms 80.6471 Ops/s 80.9013 Ops/s $\color{#d91a1a}-0.31\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 16.5371ms 14.4739ms 69.0900 Ops/s 69.2372 Ops/s $\color{#d91a1a}-0.21\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.0232ms 3.6278ms 275.6519 Ops/s 272.2597 Ops/s $\color{#35bf28}+1.25\%$

Copy link

github-actions bot commented Jan 22, 2024

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

Total Benchmarks: 89. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_single 0.1303s 66.0120ms 15.1488 Ops/s 16.5016 Ops/s $\textbf{\color{#d91a1a}-8.20\%}$
test_sync 40.1081ms 34.2975ms 29.1566 Ops/s 30.6741 Ops/s $\color{#d91a1a}-4.95\%$
test_async 61.6728ms 31.7146ms 31.5313 Ops/s 30.6917 Ops/s $\color{#35bf28}+2.74\%$
test_simple 0.4948s 0.4370s 2.2886 Ops/s 2.3595 Ops/s $\color{#d91a1a}-3.00\%$
test_transformed 0.6521s 0.5975s 1.6735 Ops/s 1.7255 Ops/s $\color{#d91a1a}-3.01\%$
test_serial 1.4350s 1.3849s 0.7221 Ops/s 0.7479 Ops/s $\color{#d91a1a}-3.45\%$
test_parallel 1.2753s 1.2069s 0.8286 Ops/s 0.8296 Ops/s $\color{#d91a1a}-0.13\%$
test_step_mdp_speed[True-True-True-True-True] 0.1047ms 20.9746μs 47.6767 KOps/s 46.9021 KOps/s $\color{#35bf28}+1.65\%$
test_step_mdp_speed[True-True-True-True-False] 36.3480μs 12.9139μs 77.4361 KOps/s 76.3551 KOps/s $\color{#35bf28}+1.42\%$
test_step_mdp_speed[True-True-True-False-True] 31.8900μs 12.5326μs 79.7921 KOps/s 80.5061 KOps/s $\color{#d91a1a}-0.89\%$
test_step_mdp_speed[True-True-True-False-False] 34.4150μs 7.5697μs 132.1055 KOps/s 130.2931 KOps/s $\color{#35bf28}+1.39\%$
test_step_mdp_speed[True-True-False-True-True] 75.2240μs 22.2306μs 44.9831 KOps/s 43.9782 KOps/s $\color{#35bf28}+2.29\%$
test_step_mdp_speed[True-True-False-True-False] 41.1770μs 14.0841μs 71.0022 KOps/s 69.5247 KOps/s $\color{#35bf28}+2.13\%$
test_step_mdp_speed[True-True-False-False-True] 54.8930μs 13.7111μs 72.9334 KOps/s 73.0800 KOps/s $\color{#d91a1a}-0.20\%$
test_step_mdp_speed[True-True-False-False-False] 36.4280μs 8.8066μs 113.5508 KOps/s 112.4581 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-False-True-True-True] 59.5510μs 23.6443μs 42.2935 KOps/s 41.7714 KOps/s $\color{#35bf28}+1.25\%$
test_step_mdp_speed[True-False-True-True-False] 41.6880μs 15.4183μs 64.8582 KOps/s 63.3498 KOps/s $\color{#35bf28}+2.38\%$
test_step_mdp_speed[True-False-True-False-True] 60.8340μs 13.6015μs 73.5211 KOps/s 73.0628 KOps/s $\color{#35bf28}+0.63\%$
test_step_mdp_speed[True-False-True-False-False] 29.8260μs 8.8152μs 113.4410 KOps/s 112.3460 KOps/s $\color{#35bf28}+0.97\%$
test_step_mdp_speed[True-False-False-True-True] 51.7660μs 25.0091μs 39.9854 KOps/s 39.6065 KOps/s $\color{#35bf28}+0.96\%$
test_step_mdp_speed[True-False-False-True-False] 41.8390μs 16.5958μs 60.2561 KOps/s 59.2113 KOps/s $\color{#35bf28}+1.76\%$
test_step_mdp_speed[True-False-False-False-True] 70.5830μs 14.9194μs 67.0270 KOps/s 67.9320 KOps/s $\color{#d91a1a}-1.33\%$
test_step_mdp_speed[True-False-False-False-False] 40.7260μs 10.0112μs 99.8879 KOps/s 98.8388 KOps/s $\color{#35bf28}+1.06\%$
test_step_mdp_speed[False-True-True-True-True] 54.6420μs 23.8328μs 41.9590 KOps/s 41.1077 KOps/s $\color{#35bf28}+2.07\%$
test_step_mdp_speed[False-True-True-True-False] 47.8900μs 15.4180μs 64.8591 KOps/s 63.0976 KOps/s $\color{#35bf28}+2.79\%$
test_step_mdp_speed[False-True-True-False-True] 46.0270μs 15.9760μs 62.5939 KOps/s 62.5402 KOps/s $\color{#35bf28}+0.09\%$
test_step_mdp_speed[False-True-True-False-False] 32.2610μs 9.8759μs 101.2564 KOps/s 98.0063 KOps/s $\color{#35bf28}+3.32\%$
test_step_mdp_speed[False-True-False-True-True] 57.2470μs 25.1832μs 39.7089 KOps/s 39.5302 KOps/s $\color{#35bf28}+0.45\%$
test_step_mdp_speed[False-True-False-True-False] 45.0040μs 16.6129μs 60.1940 KOps/s 58.7225 KOps/s $\color{#35bf28}+2.51\%$
test_step_mdp_speed[False-True-False-False-True] 54.4320μs 17.0382μs 58.6916 KOps/s 57.7316 KOps/s $\color{#35bf28}+1.66\%$
test_step_mdp_speed[False-True-False-False-False] 40.8560μs 11.1643μs 89.5711 KOps/s 87.7370 KOps/s $\color{#35bf28}+2.09\%$
test_step_mdp_speed[False-False-True-True-True] 76.1990μs 26.0245μs 38.4254 KOps/s 37.3664 KOps/s $\color{#35bf28}+2.83\%$
test_step_mdp_speed[False-False-True-True-False] 44.2030μs 17.7019μs 56.4912 KOps/s 54.8474 KOps/s $\color{#35bf28}+3.00\%$
test_step_mdp_speed[False-False-True-False-True] 67.2560μs 16.9294μs 59.0688 KOps/s 57.7222 KOps/s $\color{#35bf28}+2.33\%$
test_step_mdp_speed[False-False-True-False-False] 38.0610μs 11.1351μs 89.8060 KOps/s 87.6674 KOps/s $\color{#35bf28}+2.44\%$
test_step_mdp_speed[False-False-False-True-True] 55.8640μs 27.1925μs 36.7749 KOps/s 36.0471 KOps/s $\color{#35bf28}+2.02\%$
test_step_mdp_speed[False-False-False-True-False] 47.5290μs 18.8648μs 53.0089 KOps/s 52.5737 KOps/s $\color{#35bf28}+0.83\%$
test_step_mdp_speed[False-False-False-False-True] 49.7930μs 18.2210μs 54.8818 KOps/s 55.2736 KOps/s $\color{#d91a1a}-0.71\%$
test_step_mdp_speed[False-False-False-False-False] 35.8570μs 12.1139μs 82.5495 KOps/s 81.7056 KOps/s $\color{#35bf28}+1.03\%$
test_values[generalized_advantage_estimate-True-True] 16.4830ms 12.1095ms 82.5800 Ops/s 81.7796 Ops/s $\color{#35bf28}+0.98\%$
test_values[vec_generalized_advantage_estimate-True-True] 37.7674ms 28.3193ms 35.3116 Ops/s 36.6170 Ops/s $\color{#d91a1a}-3.57\%$
test_values[td0_return_estimate-False-False] 0.2056ms 0.1820ms 5.4931 KOps/s 5.6341 KOps/s $\color{#d91a1a}-2.50\%$
test_values[td1_return_estimate-False-False] 28.0902ms 25.0775ms 39.8764 Ops/s 38.8665 Ops/s $\color{#35bf28}+2.60\%$
test_values[vec_td1_return_estimate-False-False] 28.4587ms 28.0601ms 35.6378 Ops/s 36.0401 Ops/s $\color{#d91a1a}-1.12\%$
test_values[td_lambda_return_estimate-True-False] 38.5493ms 35.5211ms 28.1523 Ops/s 27.9403 Ops/s $\color{#35bf28}+0.76\%$
test_values[vec_td_lambda_return_estimate-True-False] 0.1493s 38.1208ms 26.2324 Ops/s 36.2730 Ops/s $\textbf{\color{#d91a1a}-27.68\%}$
test_gae_speed[generalized_advantage_estimate-False-1-512] 8.0901ms 8.0061ms 124.9041 Ops/s 122.3874 Ops/s $\color{#35bf28}+2.06\%$
test_gae_speed[vec_generalized_advantage_estimate-True-1-512] 2.4988ms 1.9514ms 512.4505 Ops/s 509.3936 Ops/s $\color{#35bf28}+0.60\%$
test_gae_speed[vec_generalized_advantage_estimate-False-1-512] 8.3508ms 0.4438ms 2.2532 KOps/s 2.2839 KOps/s $\color{#d91a1a}-1.34\%$
test_gae_speed[vec_generalized_advantage_estimate-True-32-512] 43.6868ms 39.2161ms 25.4997 Ops/s 24.8750 Ops/s $\color{#35bf28}+2.51\%$
test_gae_speed[vec_generalized_advantage_estimate-False-32-512] 11.2909ms 2.6585ms 376.1454 Ops/s 377.9833 Ops/s $\color{#d91a1a}-0.49\%$
test_dqn_speed 80.1323ms 8.2362ms 121.4147 Ops/s 124.7089 Ops/s $\color{#d91a1a}-2.64\%$
test_ddpg_speed 20.8531ms 14.8601ms 67.2941 Ops/s 69.0003 Ops/s $\color{#d91a1a}-2.47\%$
test_sac_speed 31.1917ms 29.7558ms 33.6069 Ops/s 33.9616 Ops/s $\color{#d91a1a}-1.04\%$
test_redq_speed 49.0064ms 45.9436ms 21.7658 Ops/s 21.9401 Ops/s $\color{#d91a1a}-0.79\%$
test_redq_deprec_speed 35.7270ms 26.9117ms 37.1585 Ops/s 37.5907 Ops/s $\color{#d91a1a}-1.15\%$
test_td3_speed 29.6798ms 21.0695ms 47.4619 Ops/s 48.2935 Ops/s $\color{#d91a1a}-1.72\%$
test_cql_speed 91.4037ms 87.2607ms 11.4599 Ops/s 11.1571 Ops/s $\color{#35bf28}+2.71\%$
test_a2c_speed 34.3708ms 26.9526ms 37.1022 Ops/s 36.8013 Ops/s $\color{#35bf28}+0.82\%$
test_ppo_speed 32.6513ms 27.1185ms 36.8751 Ops/s 36.7407 Ops/s $\color{#35bf28}+0.37\%$
test_reinforce_speed 27.8506ms 25.7362ms 38.8558 Ops/s 38.2748 Ops/s $\color{#35bf28}+1.52\%$
test_iql_speed 0.1073s 68.8107ms 14.5326 Ops/s 15.5026 Ops/s $\textbf{\color{#d91a1a}-6.26\%}$
test_rb_sample[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 2.1418ms 1.3963ms 716.2037 Ops/s 727.8516 Ops/s $\color{#d91a1a}-1.60\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 9.5611ms 0.5221ms 1.9154 KOps/s 1.9008 KOps/s $\color{#35bf28}+0.77\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 9.8178ms 0.4948ms 2.0209 KOps/s 2.0570 KOps/s $\color{#d91a1a}-1.75\%$
test_rb_sample[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 2.1054ms 1.4020ms 713.2453 Ops/s 699.9461 Ops/s $\color{#35bf28}+1.90\%$
test_rb_sample[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 8.7682ms 0.5191ms 1.9266 KOps/s 1.9592 KOps/s $\color{#d91a1a}-1.67\%$
test_rb_sample[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 9.1900ms 0.4968ms 2.0130 KOps/s 2.0137 KOps/s $\color{#d91a1a}-0.04\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.2960ms 1.5896ms 629.1037 Ops/s 599.6369 Ops/s $\color{#35bf28}+4.91\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.8799ms 0.6496ms 1.5394 KOps/s 1.5204 KOps/s $\color{#35bf28}+1.25\%$
test_rb_sample[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 8.5162ms 0.6372ms 1.5694 KOps/s 1.5867 KOps/s $\color{#d91a1a}-1.09\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-RandomSampler-4000] 1.7643ms 1.4574ms 686.1324 Ops/s 668.6514 Ops/s $\color{#35bf28}+2.61\%$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-10000] 8.9815ms 0.5329ms 1.8765 KOps/s 1.9299 KOps/s $\color{#d91a1a}-2.77\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-10000] 0.6088ms 0.4901ms 2.0406 KOps/s 2.0343 KOps/s $\color{#35bf28}+0.31\%$
test_rb_iterate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-4000] 1.7384ms 1.3689ms 730.5106 Ops/s 677.2337 Ops/s $\textbf{\color{#35bf28}+7.87\%}$
test_rb_iterate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-10000] 1.5725ms 0.5178ms 1.9313 KOps/s 1.9035 KOps/s $\color{#35bf28}+1.46\%$
test_rb_iterate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-10000] 0.5395ms 0.4885ms 2.0470 KOps/s 2.0599 KOps/s $\color{#d91a1a}-0.62\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-ListStorage-None-4000] 2.2864ms 1.5622ms 640.1332 Ops/s 584.3619 Ops/s $\textbf{\color{#35bf28}+9.54\%}$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-10000] 0.7924ms 0.6525ms 1.5326 KOps/s 1.4957 KOps/s $\color{#35bf28}+2.47\%$
test_rb_iterate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-10000] 9.0578ms 0.6386ms 1.5660 KOps/s 1.6003 KOps/s $\color{#d91a1a}-2.14\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-RandomSampler-400] 0.1164s 12.0215ms 83.1842 Ops/s 78.5964 Ops/s $\textbf{\color{#35bf28}+5.84\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-RandomSampler-400] 22.8638ms 13.8512ms 72.1961 Ops/s 73.1702 Ops/s $\color{#d91a1a}-1.33\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-RandomSampler-400] 5.3649ms 3.2604ms 306.7140 Ops/s 302.5646 Ops/s $\color{#35bf28}+1.37\%$
test_rb_populate[TensorDictReplayBuffer-ListStorage-SamplerWithoutReplacement-400] 0.1003s 9.9691ms 100.3104 Ops/s 80.4613 Ops/s $\textbf{\color{#35bf28}+24.67\%}$
test_rb_populate[TensorDictReplayBuffer-LazyMemmapStorage-SamplerWithoutReplacement-400] 15.7114ms 13.4657ms 74.2626 Ops/s 73.4180 Ops/s $\color{#35bf28}+1.15\%$
test_rb_populate[TensorDictReplayBuffer-LazyTensorStorage-SamplerWithoutReplacement-400] 11.2826ms 3.4428ms 290.4654 Ops/s 285.8706 Ops/s $\color{#35bf28}+1.61\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-ListStorage-None-400] 0.1096s 12.5201ms 79.8717 Ops/s 92.3303 Ops/s $\textbf{\color{#d91a1a}-13.49\%}$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyMemmapStorage-None-400] 16.1925ms 13.8197ms 72.3605 Ops/s 71.2571 Ops/s $\color{#35bf28}+1.55\%$
test_rb_populate[TensorDictPrioritizedReplayBuffer-LazyTensorStorage-None-400] 6.4365ms 3.5387ms 282.5875 Ops/s 281.5792 Ops/s $\color{#35bf28}+0.36\%$

@agarwl
Copy link

agarwl commented Jan 22, 2024

I think that's the repository of @MaxASchwarzer and you need to ask him about this.

I can put a link to the pytorch dataset in the original offline RL repo.

@vmoens vmoens merged commit 6769fee into main Jan 22, 2024
63 of 64 checks passed
@vmoens vmoens deleted the atari-dataset branch January 22, 2024 17:58
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. Data Data-related PR, will launch data-related jobs
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants