Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Performance] Faster to #1073

Merged
merged 3 commits into from
Nov 5, 2024
Merged

[Performance] Faster to #1073

merged 3 commits into from
Nov 5, 2024

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Nov 5, 2024 •
edited
Loading

Stack from ghstack (oldest at bottom):

@vmoens
Update
9105f6d 
[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 Nov 5, 2024
vmoens added a commit that referenced this pull request Nov 5, 2024
@vmoens
[Performance] Faster to
b1339d1 
ghstack-source-id: 3ff1db59f081b75f24c34c5239f88b5c5de8dbe4
Pull Request resolved: #1073
@vmoens
Update
e008ee0 
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Nov 5, 2024
@vmoens
[Performance] Faster to
38038f4 
ghstack-source-id: 63222e1497d3be45d831003c387926fbfaade67d
Pull Request resolved: #1073
@github-actions GitHub Actions
Copy link

github-actions bot commented Nov 5, 2024 •
edited
Loading

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 42.2990μs 17.8200μs 56.1167 KOps/s 56.1628 KOps/s $\color{#d91a1a}-0.08\%$
test_plain_set_stack_nested 47.2490μs 17.7486μs 56.3423 KOps/s 54.9797 KOps/s $\color{#35bf28}+2.48\%$
test_plain_set_nested_inplace 44.3930μs 19.4382μs 51.4450 KOps/s 49.8002 KOps/s $\color{#35bf28}+3.30\%$
test_plain_set_stack_nested_inplace 43.5620μs 19.3934μs 51.5639 KOps/s 48.6930 KOps/s $\textbf{\color{#35bf28}+5.90\%}$
test_items 21.5410μs 4.1548μs 240.6866 KOps/s 239.4114 KOps/s $\color{#35bf28}+0.53\%$
test_items_nested 0.6309ms 0.3434ms 2.9122 KOps/s 2.9192 KOps/s $\color{#d91a1a}-0.24\%$
test_items_nested_locked 0.4906ms 0.3432ms 2.9137 KOps/s 2.9411 KOps/s $\color{#d91a1a}-0.93\%$
test_items_nested_leaf 0.1324ms 71.5837μs 13.9697 KOps/s 14.0722 KOps/s $\color{#d91a1a}-0.73\%$
test_items_stack_nested 0.6586ms 0.3457ms 2.8924 KOps/s 2.8903 KOps/s $\color{#35bf28}+0.07\%$
test_items_stack_nested_leaf 0.1340ms 70.9951μs 14.0855 KOps/s 13.6424 KOps/s $\color{#35bf28}+3.25\%$
test_items_stack_nested_locked 0.7924ms 0.3536ms 2.8281 KOps/s 2.9091 KOps/s $\color{#d91a1a}-2.79\%$
test_keys 44.2130μs 3.5082μs 285.0457 KOps/s 284.7310 KOps/s $\color{#35bf28}+0.11\%$
test_keys_nested 0.2257ms 0.1387ms 7.2081 KOps/s 7.2309 KOps/s $\color{#d91a1a}-0.32\%$
test_keys_nested_locked 0.9177ms 0.1419ms 7.0474 KOps/s 6.9348 KOps/s $\color{#35bf28}+1.62\%$
test_keys_nested_leaf 0.2047ms 0.1193ms 8.3791 KOps/s 8.5264 KOps/s $\color{#d91a1a}-1.73\%$
test_keys_stack_nested 0.2469ms 0.1391ms 7.1899 KOps/s 7.3426 KOps/s $\color{#d91a1a}-2.08\%$
test_keys_stack_nested_leaf 0.1776ms 0.1194ms 8.3727 KOps/s 8.5260 KOps/s $\color{#d91a1a}-1.80\%$
test_keys_stack_nested_locked 0.2668ms 0.1432ms 6.9852 KOps/s 7.0605 KOps/s $\color{#d91a1a}-1.07\%$
test_values 7.0212μs 1.0595μs 943.8252 KOps/s 966.4813 KOps/s $\color{#d91a1a}-2.34\%$
test_values_nested 0.1101ms 55.3433μs 18.0690 KOps/s 18.1599 KOps/s $\color{#d91a1a}-0.50\%$
test_values_nested_locked 0.1094ms 56.2113μs 17.7900 KOps/s 18.1499 KOps/s $\color{#d91a1a}-1.98\%$
test_values_nested_leaf 0.1119ms 60.4521μs 16.5420 KOps/s 16.0083 KOps/s $\color{#35bf28}+3.33\%$
test_values_stack_nested 0.1044ms 55.3497μs 18.0669 KOps/s 17.8017 KOps/s $\color{#35bf28}+1.49\%$
test_values_stack_nested_leaf 0.1169ms 60.9613μs 16.4038 KOps/s 16.4589 KOps/s $\color{#d91a1a}-0.33\%$
test_values_stack_nested_locked 90.4590μs 55.5665μs 17.9965 KOps/s 17.8069 KOps/s $\color{#35bf28}+1.06\%$
test_membership 23.7940μs 0.9062μs 1.1036 MOps/s 1.1441 MOps/s $\color{#d91a1a}-3.54\%$
test_membership_nested 26.0890μs 2.7585μs 362.5187 KOps/s 359.7531 KOps/s $\color{#35bf28}+0.77\%$
test_membership_nested_leaf 25.9780μs 2.7639μs 361.8013 KOps/s 362.4091 KOps/s $\color{#d91a1a}-0.17\%$
test_membership_stacked_nested 17.9840μs 2.7069μs 369.4254 KOps/s 363.5144 KOps/s $\color{#35bf28}+1.63\%$
test_membership_stacked_nested_leaf 22.3220μs 2.7457μs 364.2034 KOps/s 358.5334 KOps/s $\color{#35bf28}+1.58\%$
test_membership_nested_last 31.6090μs 4.1563μs 240.6006 KOps/s 239.5194 KOps/s $\color{#35bf28}+0.45\%$
test_membership_nested_leaf_last 29.5950μs 4.1653μs 240.0816 KOps/s 237.2419 KOps/s $\color{#35bf28}+1.20\%$
test_membership_stacked_nested_last 29.5750μs 4.1565μs 240.5845 KOps/s 207.1252 KOps/s $\textbf{\color{#35bf28}+16.15\%}$
test_membership_stacked_nested_leaf_last 38.6730μs 4.1579μs 240.5046 KOps/s 204.6098 KOps/s $\textbf{\color{#35bf28}+17.54\%}$
test_nested_getleaf 38.4630μs 10.8353μs 92.2909 KOps/s 94.2970 KOps/s $\color{#d91a1a}-2.13\%$
test_nested_get 41.3380μs 10.1073μs 98.9384 KOps/s 98.2556 KOps/s $\color{#35bf28}+0.69\%$
test_stacked_getleaf 40.2250μs 10.5488μs 94.7976 KOps/s 90.6993 KOps/s $\color{#35bf28}+4.52\%$
test_stacked_get 35.6270μs 10.1976μs 98.0624 KOps/s 94.9593 KOps/s $\color{#35bf28}+3.27\%$
test_nested_getitemleaf 36.4080μs 11.0517μs 90.4836 KOps/s 88.4393 KOps/s $\color{#35bf28}+2.31\%$
test_nested_getitem 35.2160μs 10.3491μs 96.6264 KOps/s 95.7829 KOps/s $\color{#35bf28}+0.88\%$
test_stacked_getitemleaf 35.1960μs 11.0328μs 90.6386 KOps/s 88.6585 KOps/s $\color{#35bf28}+2.23\%$
test_stacked_getitem 32.3710μs 10.2527μs 97.5355 KOps/s 96.6931 KOps/s $\color{#35bf28}+0.87\%$
test_lock_nested 0.8824ms 0.4344ms 2.3018 KOps/s 2.2647 KOps/s $\color{#35bf28}+1.64\%$
test_lock_stack_nested 0.6297ms 0.4085ms 2.4479 KOps/s 2.4165 KOps/s $\color{#35bf28}+1.30\%$
test_unlock_nested 0.7405ms 0.3503ms 2.8550 KOps/s 2.7929 KOps/s $\color{#35bf28}+2.22\%$
test_unlock_stack_nested 0.5797ms 0.3278ms 3.0506 KOps/s 3.0446 KOps/s $\color{#35bf28}+0.20\%$
test_flatten_speed 0.1709ms 92.4950μs 10.8114 KOps/s 10.7628 KOps/s $\color{#35bf28}+0.45\%$
test_unflatten_speed 0.6408ms 0.4780ms 2.0920 KOps/s 2.0711 KOps/s $\color{#35bf28}+1.01\%$
test_common_ops 3.9015ms 0.7705ms 1.2979 KOps/s 1.3131 KOps/s $\color{#d91a1a}-1.16\%$
test_creation 71.2230μs 2.1863μs 457.3941 KOps/s 473.2352 KOps/s $\color{#d91a1a}-3.35\%$
test_creation_empty 32.2800μs 10.5734μs 94.5766 KOps/s 89.4496 KOps/s $\textbf{\color{#35bf28}+5.73\%}$
test_creation_nested_1 67.2190μs 13.2542μs 75.4477 KOps/s 73.1560 KOps/s $\color{#35bf28}+3.13\%$
test_creation_nested_2 48.4610μs 17.8656μs 55.9734 KOps/s 56.1110 KOps/s $\color{#d91a1a}-0.25\%$
test_clone 0.1328ms 13.4580μs 74.3053 KOps/s 75.2244 KOps/s $\color{#d91a1a}-1.22\%$
test_getitem[int] 1.1134ms 12.7875μs 78.2016 KOps/s 80.0654 KOps/s $\color{#d91a1a}-2.33\%$
test_getitem[slice_int] 0.1420ms 24.7269μs 40.4418 KOps/s 42.0545 KOps/s $\color{#d91a1a}-3.83\%$
test_getitem[range] 0.3793ms 49.2948μs 20.2861 KOps/s 20.7958 KOps/s $\color{#d91a1a}-2.45\%$
test_getitem[tuple] 0.1408ms 19.8701μs 50.3269 KOps/s 51.2595 KOps/s $\color{#d91a1a}-1.82\%$
test_getitem[list] 0.3375ms 45.2292μs 22.1096 KOps/s 23.2676 KOps/s $\color{#d91a1a}-4.98\%$
test_setitem_dim[int] 45.6660μs 24.8899μs 40.1769 KOps/s 38.6312 KOps/s $\color{#35bf28}+4.00\%$
test_setitem_dim[slice_int] 90.7500μs 50.1290μs 19.9485 KOps/s 19.5636 KOps/s $\color{#35bf28}+1.97\%$
test_setitem_dim[range] 0.1366ms 73.8851μs 13.5345 KOps/s 13.7992 KOps/s $\color{#d91a1a}-1.92\%$
test_setitem_dim[tuple] 68.5990μs 39.4453μs 25.3516 KOps/s 24.2017 KOps/s $\color{#35bf28}+4.75\%$
test_setitem 0.1850ms 20.2834μs 49.3015 KOps/s 49.3389 KOps/s $\color{#d91a1a}-0.08\%$
test_set 0.1148ms 19.6692μs 50.8409 KOps/s 50.4272 KOps/s $\color{#35bf28}+0.82\%$
test_set_shared 1.1244ms 0.1681ms 5.9491 KOps/s 5.9586 KOps/s $\color{#d91a1a}-0.16\%$
test_update 0.1437ms 22.1836μs 45.0783 KOps/s 44.0594 KOps/s $\color{#35bf28}+2.31\%$
test_update_nested 0.1157ms 31.7233μs 31.5225 KOps/s 30.2316 KOps/s $\color{#35bf28}+4.27\%$
test_update__nested 0.5760ms 32.6600μs 30.6185 KOps/s 29.8200 KOps/s $\color{#35bf28}+2.68\%$
test_set_nested 0.1507ms 21.5254μs 46.4567 KOps/s 45.5103 KOps/s $\color{#35bf28}+2.08\%$
test_set_nested_new 0.1556ms 26.0534μs 38.3828 KOps/s 37.8636 KOps/s $\color{#35bf28}+1.37\%$
test_select 0.1335ms 41.8542μs 23.8925 KOps/s 23.8011 KOps/s $\color{#35bf28}+0.38\%$
test_select_nested 0.1205ms 60.0558μs 16.6512 KOps/s 16.9956 KOps/s $\color{#d91a1a}-2.03\%$
test_exclude_nested 0.1479ms 75.3266μs 13.2755 KOps/s 13.2264 KOps/s $\color{#35bf28}+0.37\%$
test_empty[True] 0.8190ms 0.3525ms 2.8371 KOps/s 2.8344 KOps/s $\color{#35bf28}+0.10\%$
test_empty[False] 6.1065μs 1.2100μs 826.4435 KOps/s 824.5927 KOps/s $\color{#35bf28}+0.22\%$
test_unbind_speed 0.4623ms 0.2582ms 3.8726 KOps/s 3.8686 KOps/s $\color{#35bf28}+0.10\%$
test_unbind_speed_stack0 0.4364ms 0.2567ms 3.8963 KOps/s 3.8522 KOps/s $\color{#35bf28}+1.15\%$
test_unbind_speed_stack1 97.9686ms 0.7531ms 1.3278 KOps/s 1.4432 KOps/s $\textbf{\color{#d91a1a}-8.00\%}$
test_split 0.1018s 1.7804ms 561.6864 Ops/s 578.0106 Ops/s $\color{#d91a1a}-2.82\%$
test_chunk 97.2289ms 1.7852ms 560.1719 Ops/s 578.1846 Ops/s $\color{#d91a1a}-3.12\%$
test_consolidate_njt[False-None] 8.3939ms 8.0624ms 124.0319 Ops/s 122.8318 Ops/s $\color{#35bf28}+0.98\%$
test_creation[device0] 3.3135ms 93.2586μs 10.7229 KOps/s 10.9147 KOps/s $\color{#d91a1a}-1.76\%$
test_creation_from_tensor 0.2296ms 93.9300μs 10.6462 KOps/s 10.4856 KOps/s $\color{#35bf28}+1.53\%$
test_add_one[memmap_tensor0] 0.2155ms 4.9572μs 201.7251 KOps/s 202.0930 KOps/s $\color{#d91a1a}-0.18\%$
test_contiguous[memmap_tensor0] 25.9480μs 0.5035μs 1.9862 MOps/s 1.9585 MOps/s $\color{#35bf28}+1.41\%$
test_stack[memmap_tensor0] 46.8780μs 3.4381μs 290.8581 KOps/s 310.5030 KOps/s $\textbf{\color{#d91a1a}-6.33\%}$
test_memmaptd_index 0.9030ms 0.2331ms 4.2899 KOps/s 4.3050 KOps/s $\color{#d91a1a}-0.35\%$
test_memmaptd_index_astensor 0.6107ms 0.3116ms 3.2092 KOps/s 3.2393 KOps/s $\color{#d91a1a}-0.93\%$
test_memmaptd_index_op 1.0679ms 0.5825ms 1.7168 KOps/s 1.7235 KOps/s $\color{#d91a1a}-0.39\%$
test_serialize_model 0.1310s 0.1149s 8.7065 Ops/s 7.6711 Ops/s $\textbf{\color{#35bf28}+13.50\%}$
test_serialize_model_pickle 0.4513s 0.3954s 2.5293 Ops/s 2.5263 Ops/s $\color{#35bf28}+0.12\%$
test_serialize_weights 0.2162s 0.1303s 7.6747 Ops/s 8.7639 Ops/s $\textbf{\color{#d91a1a}-12.43\%}$
test_serialize_weights_returnearly 0.1706s 0.1560s 6.4095 Ops/s 6.3970 Ops/s $\color{#35bf28}+0.20\%$
test_serialize_weights_pickle 1.0832s 0.7089s 1.4106 Ops/s 2.4931 Ops/s $\textbf{\color{#d91a1a}-43.42\%}$
test_serialize_weights_filesystem 0.1439s 0.1385s 7.2184 Ops/s 6.4490 Ops/s $\textbf{\color{#35bf28}+11.93\%}$
test_serialize_model_filesystem 0.2406s 0.1519s 6.5847 Ops/s 6.6488 Ops/s $\color{#d91a1a}-0.96\%$
test_reshape_pytree 57.2980μs 27.3174μs 36.6067 KOps/s 36.8492 KOps/s $\color{#d91a1a}-0.66\%$
test_reshape_td 67.5070μs 32.7840μs 30.5027 KOps/s 31.0518 KOps/s $\color{#d91a1a}-1.77\%$
test_view_pytree 62.3160μs 26.9517μs 37.1035 KOps/s 36.5710 KOps/s $\color{#35bf28}+1.46\%$
test_view_td 98.9360μs 38.4685μs 25.9953 KOps/s 26.5757 KOps/s $\color{#d91a1a}-2.18\%$
test_unbind_pytree 67.2860μs 29.8744μs 33.4735 KOps/s 33.4130 KOps/s $\color{#35bf28}+0.18\%$
test_unbind_td 0.3172ms 38.5889μs 25.9142 KOps/s 26.1629 KOps/s $\color{#d91a1a}-0.95\%$
test_split_pytree 0.1034ms 29.7083μs 33.6606 KOps/s 33.4438 KOps/s $\color{#35bf28}+0.65\%$
test_split_td 0.5381ms 44.9261μs 22.2588 KOps/s 22.3729 KOps/s $\color{#d91a1a}-0.51\%$
test_add_pytree 84.8080μs 36.8572μs 27.1317 KOps/s 27.2292 KOps/s $\color{#d91a1a}-0.36\%$
test_add_td 0.1553ms 57.4017μs 17.4211 KOps/s 17.9941 KOps/s $\color{#d91a1a}-3.18\%$
test_compile_add_one_nested[tensordict-compile] 0.1281ms 63.4179μs 15.7684 KOps/s 15.9544 KOps/s $\color{#d91a1a}-1.17\%$
test_compile_add_one_nested[tensordict-eager] 0.4826ms 0.1599ms 6.2544 KOps/s 6.2981 KOps/s $\color{#d91a1a}-0.69\%$
test_compile_add_one_nested[pytree-compile] 0.1331ms 47.2039μs 21.1847 KOps/s 21.5256 KOps/s $\color{#d91a1a}-1.58\%$
test_compile_add_one_nested[pytree-eager] 0.2285ms 0.1200ms 8.3356 KOps/s 8.5073 KOps/s $\color{#d91a1a}-2.02\%$
test_compile_copy_nested[tensordict-compile] 82.4640μs 26.3692μs 37.9231 KOps/s 38.5604 KOps/s $\color{#d91a1a}-1.65\%$
test_compile_copy_nested[tensordict-eager] 0.1145ms 54.8099μs 18.2449 KOps/s 18.7881 KOps/s $\color{#d91a1a}-2.89\%$
test_compile_copy_nested[pytree-compile] 0.1881ms 78.6969μs 12.7070 KOps/s 12.7077 KOps/s $-0.01\%$
test_compile_copy_nested[pytree-eager] 0.1426ms 68.8531μs 14.5237 KOps/s 14.8297 KOps/s $\color{#d91a1a}-2.06\%$
test_compile_add_one_flat[tensordict-compile] 0.2228ms 0.1061ms 9.4212 KOps/s 9.3660 KOps/s $\color{#35bf28}+0.59\%$
test_compile_add_one_flat[tensordict-eager] 0.3382ms 0.1983ms 5.0432 KOps/s 5.0629 KOps/s $\color{#d91a1a}-0.39\%$
test_compile_add_one_flat[tensorclass-compile] 0.1459ms 45.3763μs 22.0380 KOps/s 21.7243 KOps/s $\color{#35bf28}+1.44\%$
test_compile_add_one_flat[tensorclass-eager] 0.4868ms 61.5346μs 16.2510 KOps/s 16.0087 KOps/s $\color{#35bf28}+1.51\%$
test_compile_add_one_flat[pytree-compile] 0.1987ms 0.1033ms 9.6803 KOps/s 9.7822 KOps/s $\color{#d91a1a}-1.04\%$
test_compile_add_one_flat[pytree-eager] 0.4288ms 0.2031ms 4.9226 KOps/s 4.9738 KOps/s $\color{#d91a1a}-1.03\%$
test_compile_add_self_flat[tensordict-eager] 0.3860ms 0.2095ms 4.7737 KOps/s 4.7788 KOps/s $\color{#d91a1a}-0.11\%$
test_compile_add_self_flat[tensordict-compile] 0.1879ms 0.1055ms 9.4760 KOps/s 9.4921 KOps/s $\color{#d91a1a}-0.17\%$
test_compile_add_self_flat[tensorclass-eager] 0.2056ms 54.9300μs 18.2050 KOps/s 18.1000 KOps/s $\color{#35bf28}+0.58\%$
test_compile_add_self_flat[tensorclass-compile] 0.4249ms 47.8958μs 20.8787 KOps/s 20.7904 KOps/s $\color{#35bf28}+0.42\%$
test_compile_add_self_flat[pytree-eager] 0.2338ms 0.1600ms 6.2502 KOps/s 6.3361 KOps/s $\color{#d91a1a}-1.36\%$
test_compile_add_self_flat[pytree-compile] 0.5083ms 0.1065ms 9.3916 KOps/s 9.6113 KOps/s $\color{#d91a1a}-2.29\%$
test_compile_copy_flat[tensordict-compile] 67.8270μs 22.5115μs 44.4217 KOps/s 46.9900 KOps/s $\textbf{\color{#d91a1a}-5.47\%}$
test_compile_copy_flat[tensordict-eager] 0.1985ms 62.6575μs 15.9598 KOps/s 16.7792 KOps/s $\color{#d91a1a}-4.88\%$
test_compile_copy_flat[pytree-compile] 0.1676ms 81.1881μs 12.3171 KOps/s 12.2728 KOps/s $\color{#35bf28}+0.36\%$
test_compile_copy_flat[pytree-eager] 0.1302ms 69.3311μs 14.4235 KOps/s 14.6891 KOps/s $\color{#d91a1a}-1.81\%$
test_compile_assign_and_add[tensordict-compile] 0.3042ms 0.2114ms 4.7308 KOps/s 4.8012 KOps/s $\color{#d91a1a}-1.47\%$
test_compile_assign_and_add[tensordict-eager] 1.4144ms 1.2763ms 783.5191 Ops/s 770.4657 Ops/s $\color{#35bf28}+1.69\%$
test_compile_assign_and_add[pytree-compile] 0.3062ms 0.2059ms 4.8578 KOps/s 4.8872 KOps/s $\color{#d91a1a}-0.60\%$
test_compile_assign_and_add[pytree-eager] 1.3437ms 0.7864ms 1.2717 KOps/s 1.2816 KOps/s $\color{#d91a1a}-0.77\%$
test_compile_assign_and_add_stack[compile] 0.7030ms 0.4612ms 2.1681 KOps/s 2.1617 KOps/s $\color{#35bf28}+0.30\%$
test_compile_assign_and_add_stack[eager] 4.1288ms 2.6000ms 384.6108 Ops/s 382.4691 Ops/s $\color{#35bf28}+0.56\%$
test_compile_indexing[tensor-tensordict-compile] 0.1108ms 37.4728μs 26.6861 KOps/s 26.9911 KOps/s $\color{#d91a1a}-1.13\%$
test_compile_indexing[tensor-tensordict-eager] 0.4875ms 33.7251μs 29.6515 KOps/s 29.9533 KOps/s $\color{#d91a1a}-1.01\%$
test_compile_indexing[tensor-tensorclass-compile] 86.2420μs 29.8705μs 33.4779 KOps/s 33.9261 KOps/s $\color{#d91a1a}-1.32\%$
test_compile_indexing[tensor-tensorclass-eager] 72.3850μs 23.5050μs 42.5441 KOps/s 42.9644 KOps/s $\color{#d91a1a}-0.98\%$
test_compile_indexing[tensor-pytree-compile] 93.1040μs 30.4497μs 32.8411 KOps/s 33.3173 KOps/s $\color{#d91a1a}-1.43\%$
test_compile_indexing[tensor-pytree-eager] 68.2070μs 23.5981μs 42.3764 KOps/s 43.1632 KOps/s $\color{#d91a1a}-1.82\%$
test_compile_indexing[slice-tensordict-compile] 0.1151ms 52.1376μs 19.1800 KOps/s 19.0217 KOps/s $\color{#35bf28}+0.83\%$
test_compile_indexing[slice-tensordict-eager] 0.6453ms 20.5774μs 48.5970 KOps/s 48.4249 KOps/s $\color{#35bf28}+0.36\%$
test_compile_indexing[slice-tensorclass-compile] 0.1105ms 44.2860μs 22.5805 KOps/s 22.7310 KOps/s $\color{#d91a1a}-0.66\%$
test_compile_indexing[slice-tensorclass-eager] 78.0160μs 19.2471μs 51.9560 KOps/s 51.5905 KOps/s $\color{#35bf28}+0.71\%$
test_compile_indexing[slice-pytree-compile] 93.7160μs 45.2594μs 22.0949 KOps/s 22.2192 KOps/s $\color{#d91a1a}-0.56\%$
test_compile_indexing[slice-pytree-eager] 75.1810μs 19.0724μs 52.4319 KOps/s 51.4405 KOps/s $\color{#35bf28}+1.93\%$
test_compile_indexing[int-tensordict-compile] 0.1105ms 52.7182μs 18.9688 KOps/s 18.8541 KOps/s $\color{#35bf28}+0.61\%$
test_compile_indexing[int-tensordict-eager] 0.8565ms 19.9989μs 50.0027 KOps/s 50.8581 KOps/s $\color{#d91a1a}-1.68\%$
test_compile_indexing[int-tensorclass-compile] 95.6090μs 45.0318μs 22.2065 KOps/s 22.3106 KOps/s $\color{#d91a1a}-0.47\%$
test_compile_indexing[int-tensorclass-eager] 70.9430μs 19.0639μs 52.4552 KOps/s 52.9583 KOps/s $\color{#d91a1a}-0.95\%$
test_compile_indexing[int-pytree-compile] 0.1010ms 45.1441μs 22.1513 KOps/s 22.3579 KOps/s $\color{#d91a1a}-0.92\%$
test_compile_indexing[int-pytree-eager] 0.4547ms 19.2136μs 52.0464 KOps/s 52.5060 KOps/s $\color{#d91a1a}-0.88\%$
test_mod_add[eager] 63.4380μs 25.7472μs 38.8392 KOps/s 38.3506 KOps/s $\color{#35bf28}+1.27\%$
test_mod_add[compile] 85.5300μs 45.2368μs 22.1059 KOps/s 22.2236 KOps/s $\color{#d91a1a}-0.53\%$
test_mod_add[compile-overhead] 96.5300μs 44.9266μs 22.2585 KOps/s 22.3061 KOps/s $\color{#d91a1a}-0.21\%$
test_mod_wrap[eager] 0.4591ms 0.2134ms 4.6850 KOps/s 4.6366 KOps/s $\color{#35bf28}+1.04\%$
test_mod_wrap[compile] 1.5411ms 0.2107ms 4.7459 KOps/s 4.8365 KOps/s $\color{#d91a1a}-1.87\%$
test_mod_wrap[compile-overhead] 1.5693ms 0.2081ms 4.8051 KOps/s 4.8470 KOps/s $\color{#d91a1a}-0.86\%$
test_mod_wrap_and_backward[eager] 14.4694ms 12.6446ms 79.0853 Ops/s 83.3196 Ops/s $\textbf{\color{#d91a1a}-5.08\%}$
test_mod_wrap_and_backward[compile] 16.3591ms 12.7340ms 78.5299 Ops/s 79.0981 Ops/s $\color{#d91a1a}-0.72\%$
test_mod_wrap_and_backward[compile-overhead] 17.0178ms 12.7734ms 78.2876 Ops/s 83.1785 Ops/s $\textbf{\color{#d91a1a}-5.88\%}$
test_seq_add[eager] 0.2169ms 88.1651μs 11.3424 KOps/s 10.7916 KOps/s $\textbf{\color{#35bf28}+5.10\%}$
test_seq_add[compile] 0.1401ms 60.8539μs 16.4328 KOps/s 16.7219 KOps/s $\color{#d91a1a}-1.73\%$
test_seq_add[compile-overhead] 0.1181ms 59.1354μs 16.9103 KOps/s 16.7706 KOps/s $\color{#35bf28}+0.83\%$
test_seq_wrap[eager] 0.5478ms 0.3848ms 2.5990 KOps/s 2.5581 KOps/s $\color{#35bf28}+1.60\%$
test_seq_wrap[compile] 0.4271ms 0.2301ms 4.3464 KOps/s 4.3768 KOps/s $\color{#d91a1a}-0.69\%$
test_seq_wrap[compile-overhead] 0.4197ms 0.2273ms 4.3997 KOps/s 4.3984 KOps/s $\color{#35bf28}+0.03\%$
test_func_call_runtime[False-eager] 0.7829ms 0.5603ms 1.7847 KOps/s 1.8154 KOps/s $\color{#d91a1a}-1.70\%$
test_func_call_runtime[False-compile] 1.0595ms 0.4294ms 2.3288 KOps/s 2.3443 KOps/s $\color{#d91a1a}-0.66\%$
test_func_call_runtime[False-compile-overhead] 0.9045ms 0.4343ms 2.3028 KOps/s 2.3460 KOps/s $\color{#d91a1a}-1.84\%$
test_func_call_runtime[True-eager] 1.5158ms 0.7707ms 1.2976 KOps/s 1.2981 KOps/s $\color{#d91a1a}-0.04\%$
test_func_call_runtime[True-compile] 0.8386ms 0.4719ms 2.1192 KOps/s 2.1604 KOps/s $\color{#d91a1a}-1.91\%$
test_func_call_runtime[True-compile-overhead] 0.9125ms 0.4763ms 2.0994 KOps/s 2.1472 KOps/s $\color{#d91a1a}-2.23\%$
test_func_call_cm_runtime[False-eager] 1.0529ms 0.5587ms 1.7899 KOps/s 1.8141 KOps/s $\color{#d91a1a}-1.33\%$
test_func_call_cm_runtime[False-compile] 0.8634ms 0.4281ms 2.3361 KOps/s 2.3579 KOps/s $\color{#d91a1a}-0.93\%$
test_func_call_cm_runtime[False-compile-overhead] 0.9062ms 0.4307ms 2.3216 KOps/s 2.3632 KOps/s $\color{#d91a1a}-1.76\%$
test_func_call_cm_runtime[True-eager] 1.2821ms 0.9067ms 1.1029 KOps/s 1.1017 KOps/s $\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[True-compile] 0.6268ms 0.4961ms 2.0156 KOps/s 2.0566 KOps/s $\color{#d91a1a}-1.99\%$
test_func_call_cm_runtime[True-compile-overhead] 0.6746ms 0.4957ms 2.0173 KOps/s 2.0438 KOps/s $\color{#d91a1a}-1.30\%$
test_vmap_func_call_cm_runtime[eager] 2.6459ms 1.8928ms 528.3296 Ops/s 529.2865 Ops/s $\color{#d91a1a}-0.18\%$
test_vmap_func_call_cm_runtime[compile] 0.7094ms 0.5215ms 1.9177 KOps/s 1.9053 KOps/s $\color{#35bf28}+0.65\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.7367ms 0.5284ms 1.8924 KOps/s 1.9165 KOps/s $\color{#d91a1a}-1.26\%$
test_distributed 0.2873ms 0.1285ms 7.7814 KOps/s 7.7080 KOps/s $\color{#35bf28}+0.95\%$
test_tdmodule 45.2340μs 18.5054μs 54.0383 KOps/s 51.6002 KOps/s $\color{#35bf28}+4.72\%$
test_tdmodule_dispatch 71.8440μs 37.1797μs 26.8964 KOps/s 27.1718 KOps/s $\color{#d91a1a}-1.01\%$
test_tdseq 36.7480μs 20.7307μs 48.2376 KOps/s 44.5433 KOps/s $\textbf{\color{#35bf28}+8.29\%}$
test_tdseq_dispatch 68.3280μs 41.1711μs 24.2889 KOps/s 23.2586 KOps/s $\color{#35bf28}+4.43\%$
test_instantiation_functorch 2.1661ms 1.5504ms 645.0150 Ops/s 660.3664 Ops/s $\color{#d91a1a}-2.32\%$
test_exec_functorch 0.2860ms 0.1816ms 5.5066 KOps/s 5.6163 KOps/s $\color{#d91a1a}-1.95\%$
test_exec_functional_call 0.4429ms 0.1724ms 5.8020 KOps/s 5.6822 KOps/s $\color{#35bf28}+2.11\%$
test_exec_td_decorator 0.5232ms 0.2263ms 4.4198 KOps/s 4.3143 KOps/s $\color{#35bf28}+2.45\%$
test_vmap_mlp_speed_decorator[True-True] 0.8752ms 0.6355ms 1.5735 KOps/s 1.5804 KOps/s $\color{#d91a1a}-0.44\%$
test_vmap_mlp_speed_decorator[True-False] 0.8777ms 0.6347ms 1.5756 KOps/s 1.5365 KOps/s $\color{#35bf28}+2.54\%$
test_vmap_mlp_speed_decorator[False-True] 0.9235ms 0.5225ms 1.9140 KOps/s 1.9125 KOps/s $\color{#35bf28}+0.08\%$
test_vmap_mlp_speed_decorator[False-False] 0.8155ms 0.5227ms 1.9130 KOps/s 1.9091 KOps/s $\color{#35bf28}+0.20\%$
test_to_module_speed[True] 1.3765ms 1.2878ms 776.5270 Ops/s 766.7625 Ops/s $\color{#35bf28}+1.27\%$
test_to_module_speed[False] 1.7717ms 1.2637ms 791.3442 Ops/s 777.5382 Ops/s $\color{#35bf28}+1.78\%$
test_tc_init 86.8430μs 42.9718μs 23.2711 KOps/s 22.7022 KOps/s $\color{#35bf28}+2.51\%$
test_tc_init_nested 0.1554ms 87.8757μs 11.3797 KOps/s 11.2877 KOps/s $\color{#35bf28}+0.81\%$
test_tc_first_layer_tensor 22.8130μs 1.5419μs 648.5586 KOps/s 659.2786 KOps/s $\color{#d91a1a}-1.63\%$
test_tc_first_layer_nontensor 43.5610μs 4.8714μs 205.2785 KOps/s 212.3511 KOps/s $\color{#d91a1a}-3.33\%$
test_tc_second_layer_tensor 44.4030μs 2.7882μs 358.6495 KOps/s 357.6847 KOps/s $\color{#35bf28}+0.27\%$
test_tc_second_layer_nontensor 43.3210μs 6.1785μs 161.8514 KOps/s 167.6861 KOps/s $\color{#d91a1a}-3.48\%$
test_unbind 0.2127s 12.2958ms 81.3285 Ops/s 78.9360 Ops/s $\color{#35bf28}+3.03\%$
test_full_like 17.7212ms 11.6405ms 85.9068 Ops/s 85.3003 Ops/s $\color{#35bf28}+0.71\%$
test_zeros_like 9.8019ms 7.1102ms 140.6421 Ops/s 133.7339 Ops/s $\textbf{\color{#35bf28}+5.17\%}$
test_ones_like 13.0645ms 7.8133ms 127.9874 Ops/s 131.5689 Ops/s $\color{#d91a1a}-2.72\%$
test_clone 14.0065ms 9.4608ms 105.6988 Ops/s 109.0471 Ops/s $\color{#d91a1a}-3.07\%$
test_squeeze 60.0530μs 12.0184μs 83.2056 KOps/s 83.5152 KOps/s $\color{#d91a1a}-0.37\%$
test_unsqueeze 0.1653ms 87.3775μs 11.4446 KOps/s 11.2195 KOps/s $\color{#35bf28}+2.01\%$
test_split 0.5998ms 0.1929ms 5.1842 KOps/s 5.3098 KOps/s $\color{#d91a1a}-2.36\%$
test_permute 0.4274ms 0.2194ms 4.5578 KOps/s 4.5810 KOps/s $\color{#d91a1a}-0.51\%$
test_stack 28.3028ms 23.9160ms 41.8131 Ops/s 42.1023 Ops/s $\color{#d91a1a}-0.69\%$
test_cat 27.3812ms 23.5713ms 42.4246 Ops/s 41.8822 Ops/s $\color{#35bf28}+1.29\%$

@github-actions GitHub Actions
Copy link

github-actions bot commented Nov 5, 2024 •
edited
Loading

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 26.8710μs 10.5217μs 95.0421 KOps/s 93.8905 KOps/s $\color{#35bf28}+1.23\%$
test_plain_set_stack_nested 28.3500μs 10.6472μs 93.9213 KOps/s 93.0242 KOps/s $\color{#35bf28}+0.96\%$
test_plain_set_nested_inplace 0.1002ms 11.4753μs 87.1437 KOps/s 85.4243 KOps/s $\color{#35bf28}+2.01\%$
test_plain_set_stack_nested_inplace 57.2410μs 11.4953μs 86.9922 KOps/s 86.4631 KOps/s $\color{#35bf28}+0.61\%$
test_items 27.6110μs 2.8884μs 346.2156 KOps/s 344.1281 KOps/s $\color{#35bf28}+0.61\%$
test_items_nested 0.4581ms 0.3219ms 3.1061 KOps/s 3.1187 KOps/s $\color{#d91a1a}-0.40\%$
test_items_nested_locked 0.4806ms 0.3237ms 3.0894 KOps/s 3.1160 KOps/s $\color{#d91a1a}-0.85\%$
test_items_nested_leaf 0.1027ms 57.8740μs 17.2789 KOps/s 17.2580 KOps/s $\color{#35bf28}+0.12\%$
test_items_stack_nested 0.4341ms 0.3217ms 3.1087 KOps/s 3.0393 KOps/s $\color{#35bf28}+2.28\%$
test_items_stack_nested_leaf 92.6010μs 57.6787μs 17.3374 KOps/s 17.1618 KOps/s $\color{#35bf28}+1.02\%$
test_items_stack_nested_locked 0.4253ms 0.3249ms 3.0781 KOps/s 3.0941 KOps/s $\color{#d91a1a}-0.52\%$
test_keys 25.7900μs 3.4725μs 287.9769 KOps/s 290.3897 KOps/s $\color{#d91a1a}-0.83\%$
test_keys_nested 0.1315ms 70.4455μs 14.1954 KOps/s 14.1550 KOps/s $\color{#35bf28}+0.29\%$
test_keys_nested_locked 0.8795ms 75.8876μs 13.1774 KOps/s 13.0238 KOps/s $\color{#35bf28}+1.18\%$
test_keys_nested_leaf 0.1107ms 62.0557μs 16.1146 KOps/s 16.0852 KOps/s $\color{#35bf28}+0.18\%$
test_keys_stack_nested 0.1199ms 70.2233μs 14.2403 KOps/s 13.9370 KOps/s $\color{#35bf28}+2.18\%$
test_keys_stack_nested_leaf 91.0820μs 61.5284μs 16.2527 KOps/s 16.0384 KOps/s $\color{#35bf28}+1.34\%$
test_keys_stack_nested_locked 0.1227ms 75.7085μs 13.2086 KOps/s 13.0761 KOps/s $\color{#35bf28}+1.01\%$
test_values 7.2500μs 0.9565μs 1.0455 MOps/s 1.1818 MOps/s $\textbf{\color{#d91a1a}-11.53\%}$
test_values_nested 70.5210μs 31.3366μs 31.9116 KOps/s 32.0888 KOps/s $\color{#d91a1a}-0.55\%$
test_values_nested_locked 0.3976ms 32.6232μs 30.6530 KOps/s 30.3984 KOps/s $\color{#35bf28}+0.84\%$
test_values_nested_leaf 0.4183ms 33.6350μs 29.7309 KOps/s 29.6178 KOps/s $\color{#35bf28}+0.38\%$
test_values_stack_nested 65.5810μs 31.4300μs 31.8168 KOps/s 31.8268 KOps/s $\color{#d91a1a}-0.03\%$
test_values_stack_nested_leaf 94.2510μs 33.6985μs 29.6749 KOps/s 29.4258 KOps/s $\color{#35bf28}+0.85\%$
test_values_stack_nested_locked 0.4143ms 32.8537μs 30.4380 KOps/s 30.3574 KOps/s $\color{#35bf28}+0.27\%$
test_membership 18.9108μs 0.5204μs 1.9214 MOps/s 1.9742 MOps/s $\color{#d91a1a}-2.67\%$
test_membership_nested 0.1938ms 1.8762μs 533.0005 KOps/s 527.7844 KOps/s $\color{#35bf28}+0.99\%$
test_membership_nested_leaf 15.3655μs 1.9008μs 526.0910 KOps/s 534.9057 KOps/s $\color{#d91a1a}-1.65\%$
test_membership_stacked_nested 24.7810μs 1.9421μs 514.9099 KOps/s 503.5655 KOps/s $\color{#35bf28}+2.25\%$
test_membership_stacked_nested_leaf 17.0200μs 1.9254μs 519.3706 KOps/s 504.5934 KOps/s $\color{#35bf28}+2.93\%$
test_membership_nested_last 0.3864ms 2.8390μs 352.2412 KOps/s 355.2958 KOps/s $\color{#d91a1a}-0.86\%$
test_membership_nested_leaf_last 0.3814ms 2.8198μs 354.6384 KOps/s 352.7081 KOps/s $\color{#35bf28}+0.55\%$
test_membership_stacked_nested_last 30.4410μs 2.8300μs 353.3626 KOps/s 359.1944 KOps/s $\color{#d91a1a}-1.62\%$
test_membership_stacked_nested_leaf_last 43.3700μs 2.8224μs 354.3113 KOps/s 356.5444 KOps/s $\color{#d91a1a}-0.63\%$
test_nested_getleaf 0.3893ms 6.0453μs 165.4183 KOps/s 166.3351 KOps/s $\color{#d91a1a}-0.55\%$
test_nested_get 0.3912ms 5.7040μs 175.3140 KOps/s 175.0504 KOps/s $\color{#35bf28}+0.15\%$
test_stacked_getleaf 40.4410μs 6.0079μs 166.4474 KOps/s 166.7393 KOps/s $\color{#d91a1a}-0.18\%$
test_stacked_get 0.3876ms 5.7083μs 175.1833 KOps/s 175.0716 KOps/s $\color{#35bf28}+0.06\%$
test_nested_getitemleaf 33.9000μs 6.0865μs 164.2979 KOps/s 163.6521 KOps/s $\color{#35bf28}+0.39\%$
test_nested_getitem 0.3946ms 5.7711μs 173.2763 KOps/s 173.5886 KOps/s $\color{#d91a1a}-0.18\%$
test_stacked_getitemleaf 29.9900μs 6.1148μs 163.5375 KOps/s 164.0938 KOps/s $\color{#d91a1a}-0.34\%$
test_stacked_getitem 0.3859ms 5.7919μs 172.6563 KOps/s 173.1619 KOps/s $\color{#d91a1a}-0.29\%$
test_lock_nested 0.8724ms 0.3648ms 2.7411 KOps/s 2.6882 KOps/s $\color{#35bf28}+1.97\%$
test_lock_stack_nested 0.3683ms 0.3388ms 2.9513 KOps/s 2.9217 KOps/s $\color{#35bf28}+1.01\%$
test_unlock_nested 0.6587ms 0.3099ms 3.2266 KOps/s 3.2328 KOps/s $\color{#d91a1a}-0.19\%$
test_unlock_stack_nested 0.6511ms 0.2790ms 3.5847 KOps/s 3.5553 KOps/s $\color{#35bf28}+0.83\%$
test_flatten_speed 0.4677ms 72.2459μs 13.8416 KOps/s 13.9166 KOps/s $\color{#d91a1a}-0.54\%$
test_unflatten_speed 0.6550ms 0.2878ms 3.4751 KOps/s 3.4141 KOps/s $\color{#35bf28}+1.79\%$
test_common_ops 1.5473ms 0.5932ms 1.6858 KOps/s 1.6777 KOps/s $\color{#35bf28}+0.48\%$
test_creation 0.1810ms 1.4796μs 675.8383 KOps/s 664.5598 KOps/s $\color{#35bf28}+1.70\%$
test_creation_empty 28.4300μs 7.2499μs 137.9333 KOps/s 133.9479 KOps/s $\color{#35bf28}+2.98\%$
test_creation_nested_1 0.3906ms 8.8471μs 113.0312 KOps/s 109.8027 KOps/s $\color{#35bf28}+2.94\%$
test_creation_nested_2 34.2300μs 11.3687μs 87.9608 KOps/s 86.0556 KOps/s $\color{#35bf28}+2.21\%$
test_clone 37.8510μs 11.6057μs 86.1646 KOps/s 90.9333 KOps/s $\textbf{\color{#d91a1a}-5.24\%}$
test_getitem[int] 1.2811ms 11.0245μs 90.7072 KOps/s 89.7130 KOps/s $\color{#35bf28}+1.11\%$
test_getitem[slice_int] 0.4056ms 21.5146μs 46.4801 KOps/s 46.2337 KOps/s $\color{#35bf28}+0.53\%$
test_getitem[range] 0.1374ms 39.8716μs 25.0805 KOps/s 24.9862 KOps/s $\color{#35bf28}+0.38\%$
test_getitem[tuple] 0.1046ms 18.8382μs 53.0836 KOps/s 52.8734 KOps/s $\color{#35bf28}+0.40\%$
test_getitem[list] 0.4238ms 34.8759μs 28.6731 KOps/s 28.8505 KOps/s $\color{#d91a1a}-0.61\%$
test_setitem_dim[int] 39.9700μs 20.0659μs 49.8358 KOps/s 50.1193 KOps/s $\color{#d91a1a}-0.57\%$
test_setitem_dim[slice_int] 62.9210μs 38.7021μs 25.8384 KOps/s 25.6963 KOps/s $\color{#35bf28}+0.55\%$
test_setitem_dim[range] 77.5210μs 55.5497μs 18.0019 KOps/s 18.1281 KOps/s $\color{#d91a1a}-0.70\%$
test_setitem_dim[tuple] 55.2910μs 33.0631μs 30.2452 KOps/s 29.7989 KOps/s $\color{#35bf28}+1.50\%$
test_setitem 0.3991ms 15.9791μs 62.5819 KOps/s 64.6692 KOps/s $\color{#d91a1a}-3.23\%$
test_set 0.1207ms 15.0372μs 66.5016 KOps/s 66.8706 KOps/s $\color{#d91a1a}-0.55\%$
test_set_shared 1.8141ms 0.1486ms 6.7296 KOps/s 6.7358 KOps/s $\color{#d91a1a}-0.09\%$
test_update 1.2865ms 17.5105μs 57.1085 KOps/s 57.0048 KOps/s $\color{#35bf28}+0.18\%$
test_update_nested 0.4036ms 22.2373μs 44.9694 KOps/s 44.0000 KOps/s $\color{#35bf28}+2.20\%$
test_update__nested 0.1280ms 25.1719μs 39.7269 KOps/s 40.6120 KOps/s $\color{#d91a1a}-2.18\%$
test_set_nested 0.3999ms 16.4447μs 60.8100 KOps/s 57.7487 KOps/s $\textbf{\color{#35bf28}+5.30\%}$
test_set_nested_new 0.1165ms 18.6128μs 53.7264 KOps/s 49.7702 KOps/s $\textbf{\color{#35bf28}+7.95\%}$
test_select 0.4127ms 30.5320μs 32.7525 KOps/s 32.0495 KOps/s $\color{#35bf28}+2.19\%$
test_select_nested 76.9410μs 41.8571μs 23.8908 KOps/s 24.1279 KOps/s $\color{#d91a1a}-0.98\%$
test_exclude_nested 0.4404ms 59.4539μs 16.8197 KOps/s 16.9945 KOps/s $\color{#d91a1a}-1.03\%$
test_empty[True] 0.6283ms 0.2549ms 3.9231 KOps/s 3.9314 KOps/s $\color{#d91a1a}-0.21\%$
test_empty[False] 38.4136μs 0.7416μs 1.3484 MOps/s 1.3570 MOps/s $\color{#d91a1a}-0.63\%$
test_to 85.4320μs 55.9723μs 17.8660 KOps/s 16.8566 KOps/s $\textbf{\color{#35bf28}+5.99\%}$
test_to_nonblocking 0.4392ms 48.3636μs 20.6767 KOps/s 18.9736 KOps/s $\textbf{\color{#35bf28}+8.98\%}$
test_unbind_speed 0.2739ms 0.2352ms 4.2512 KOps/s 4.1930 KOps/s $\color{#35bf28}+1.39\%$
test_unbind_speed_stack0 0.6236ms 0.2353ms 4.2498 KOps/s 4.2043 KOps/s $\color{#35bf28}+1.08\%$
test_unbind_speed_stack1 92.3932ms 0.6543ms 1.5283 KOps/s 1.5028 KOps/s $\color{#35bf28}+1.70\%$
test_split 92.6383ms 1.6098ms 621.1975 Ops/s 575.0972 Ops/s $\textbf{\color{#35bf28}+8.02\%}$
test_chunk 94.9654ms 1.7403ms 574.6094 Ops/s 685.3991 Ops/s $\textbf{\color{#d91a1a}-16.16\%}$
test_consolidate[False-None] 3.2561ms 2.6545ms 376.7146 Ops/s 347.7775 Ops/s $\textbf{\color{#35bf28}+8.32\%}$
test_consolidate[default-None] 1.8207ms 1.7001ms 588.1983 Ops/s 598.8832 Ops/s $\color{#d91a1a}-1.78\%$
test_consolidate[reduce-overhead-None] 1.8401ms 1.6934ms 590.5216 Ops/s 584.9597 Ops/s $\color{#35bf28}+0.95\%$
test_consolidate_njt[False-None] 6.9644ms 6.6839ms 149.6134 Ops/s 148.5429 Ops/s $\color{#35bf28}+0.72\%$
test_to[False-False-None] 1.8091ms 1.7631ms 567.1864 Ops/s 482.8376 Ops/s $\textbf{\color{#35bf28}+17.47\%}$
test_to[True-False-None] 1.6258ms 1.3767ms 726.3558 Ops/s 715.5130 Ops/s $\color{#35bf28}+1.52\%$
test_to[within-False-None] 4.1863ms 4.0673ms 245.8608 Ops/s 241.2657 Ops/s $\color{#35bf28}+1.90\%$
test_to[True-default-None] 5.6343ms 5.2034ms 192.1813 Ops/s 187.6993 Ops/s $\color{#35bf28}+2.39\%$
test_to_njt[False-False-None] 7.2010ms 7.0748ms 141.3475 Ops/s 127.2648 Ops/s $\textbf{\color{#35bf28}+11.07\%}$
test_to_njt[True-False-None] 5.8147ms 5.6423ms 177.2341 Ops/s 166.0823 Ops/s $\textbf{\color{#35bf28}+6.71\%}$
test_to_njt[within-False-None] 12.5532ms 12.3906ms 80.7061 Ops/s 75.9551 Ops/s $\textbf{\color{#35bf28}+6.26\%}$
test_creation[device0] 0.3812ms 79.8014μs 12.5311 KOps/s 11.6946 KOps/s $\textbf{\color{#35bf28}+7.15\%}$
test_creation_from_tensor 0.6103ms 83.2098μs 12.0178 KOps/s 11.1977 KOps/s $\textbf{\color{#35bf28}+7.32\%}$
test_add_one[memmap_tensor0] 0.4208ms 7.2930μs 137.1171 KOps/s 137.9714 KOps/s $\color{#d91a1a}-0.62\%$
test_contiguous[memmap_tensor0] 1.7900μs 0.4346μs 2.3008 MOps/s 2.3042 MOps/s $\color{#d91a1a}-0.15\%$
test_stack[memmap_tensor0] 37.7810μs 4.8843μs 204.7377 KOps/s 201.2213 KOps/s $\color{#35bf28}+1.75\%$
test_memmaptd_index 1.9694ms 0.2585ms 3.8679 KOps/s 3.8556 KOps/s $\color{#35bf28}+0.32\%$
test_memmaptd_index_astensor 0.5986ms 0.3158ms 3.1662 KOps/s 3.1356 KOps/s $\color{#35bf28}+0.98\%$
test_memmaptd_index_op 1.0078ms 0.5940ms 1.6836 KOps/s 1.6822 KOps/s $\color{#35bf28}+0.09\%$
test_serialize_model 0.1305s 0.1295s 7.7245 Ops/s 7.6867 Ops/s $\color{#35bf28}+0.49\%$
test_serialize_model_pickle 1.3490s 1.2167s 0.8219 Ops/s 0.8419 Ops/s $\color{#d91a1a}-2.38\%$
test_serialize_weights 0.1304s 0.1291s 7.7455 Ops/s 7.7193 Ops/s $\color{#35bf28}+0.34\%$
test_serialize_weights_returnearly 0.3508s 65.7085ms 15.2187 Ops/s 23.3183 Ops/s $\textbf{\color{#d91a1a}-34.73\%}$
test_serialize_weights_pickle 1.3742s 1.2175s 0.8214 Ops/s 0.8399 Ops/s $\color{#d91a1a}-2.21\%$
test_reshape_pytree 53.3210μs 22.4613μs 44.5210 KOps/s 43.6607 KOps/s $\color{#35bf28}+1.97\%$
test_reshape_td 52.9800μs 27.3626μs 36.5463 KOps/s 36.9970 KOps/s $\color{#d91a1a}-1.22\%$
test_view_pytree 55.4710μs 22.1071μs 45.2343 KOps/s 43.9263 KOps/s $\color{#35bf28}+2.98\%$
test_view_td 58.6210μs 30.3173μs 32.9845 KOps/s 31.7099 KOps/s $\color{#35bf28}+4.02\%$
test_unbind_pytree 56.9210μs 28.1314μs 35.5475 KOps/s 35.2596 KOps/s $\color{#35bf28}+0.82\%$
test_unbind_td 0.8952ms 36.9144μs 27.0897 KOps/s 26.8170 KOps/s $\color{#35bf28}+1.02\%$
test_split_pytree 66.6210μs 30.3083μs 32.9942 KOps/s 32.9196 KOps/s $\color{#35bf28}+0.23\%$
test_split_td 0.1766ms 39.8841μs 25.0727 KOps/s 25.2286 KOps/s $\color{#d91a1a}-0.62\%$
test_add_pytree 76.8310μs 37.1182μs 26.9409 KOps/s 27.7704 KOps/s $\color{#d91a1a}-2.99\%$
test_add_td 86.9010μs 48.0388μs 20.8165 KOps/s 19.9195 KOps/s $\color{#35bf28}+4.50\%$
test_compile_add_one_nested[tensordict-compile] 0.1711ms 0.1244ms 8.0415 KOps/s 8.0422 KOps/s $-0.01\%$
test_compile_add_one_nested[tensordict-eager] 0.2701ms 0.1266ms 7.9006 KOps/s 7.7761 KOps/s $\color{#35bf28}+1.60\%$
test_compile_add_one_nested[pytree-compile] 0.1418ms 0.1037ms 9.6449 KOps/s 9.8664 KOps/s $\color{#d91a1a}-2.25\%$
test_compile_add_one_nested[pytree-eager] 0.5836ms 0.1611ms 6.2063 KOps/s 6.5024 KOps/s $\color{#d91a1a}-4.55\%$
test_compile_copy_nested[tensordict-compile] 68.3710μs 22.9738μs 43.5278 KOps/s 41.7793 KOps/s $\color{#35bf28}+4.19\%$
test_compile_copy_nested[tensordict-eager] 0.1081ms 26.8698μs 37.2165 KOps/s 36.4486 KOps/s $\color{#35bf28}+2.11\%$
test_compile_copy_nested[pytree-compile] 0.1421ms 63.3368μs 15.7886 KOps/s 15.3201 KOps/s $\color{#35bf28}+3.06\%$
test_compile_copy_nested[pytree-eager] 78.5610μs 49.1318μs 20.3534 KOps/s 20.0700 KOps/s $\color{#35bf28}+1.41\%$
test_compile_add_one_flat[tensordict-compile] 0.1819ms 0.1402ms 7.1343 KOps/s 6.8676 KOps/s $\color{#35bf28}+3.88\%$
test_compile_add_one_flat[tensordict-eager] 0.3590ms 0.2069ms 4.8329 KOps/s 4.7091 KOps/s $\color{#35bf28}+2.63\%$
test_compile_add_one_flat[tensorclass-compile] 0.1340ms 96.3000μs 10.3842 KOps/s 10.1660 KOps/s $\color{#35bf28}+2.15\%$
test_compile_add_one_flat[tensorclass-eager] 0.1353ms 53.0337μs 18.8559 KOps/s 18.9519 KOps/s $\color{#d91a1a}-0.51\%$
test_compile_add_one_flat[pytree-compile] 0.1791ms 0.1376ms 7.2662 KOps/s 6.9226 KOps/s $\color{#35bf28}+4.96\%$
test_compile_add_one_flat[pytree-eager] 0.5898ms 0.5219ms 1.9161 KOps/s 2.0243 KOps/s $\textbf{\color{#d91a1a}-5.35\%}$
test_compile_add_self_flat[tensordict-eager] 0.3544ms 0.2483ms 4.0280 KOps/s 4.0193 KOps/s $\color{#35bf28}+0.22\%$
test_compile_add_self_flat[tensordict-compile] 0.1822ms 0.1415ms 7.0651 KOps/s 6.9913 KOps/s $\color{#35bf28}+1.05\%$
test_compile_add_self_flat[tensorclass-eager] 0.1404ms 62.4288μs 16.0182 KOps/s 15.7654 KOps/s $\color{#35bf28}+1.60\%$
test_compile_add_self_flat[tensorclass-compile] 0.2317ms 97.2035μs 10.2877 KOps/s 9.8277 KOps/s $\color{#35bf28}+4.68\%$
test_compile_add_self_flat[pytree-eager] 0.4817ms 0.4388ms 2.2789 KOps/s 2.4126 KOps/s $\textbf{\color{#d91a1a}-5.54\%}$
test_compile_add_self_flat[pytree-compile] 0.1827ms 0.1381ms 7.2404 KOps/s 7.1308 KOps/s $\color{#35bf28}+1.54\%$
test_compile_copy_flat[tensordict-compile] 0.2022ms 27.6392μs 36.1805 KOps/s 52.3759 KOps/s $\textbf{\color{#d91a1a}-30.92\%}$
test_compile_copy_flat[tensordict-eager] 64.4710μs 27.0393μs 36.9832 KOps/s 36.6417 KOps/s $\color{#35bf28}+0.93\%$
test_compile_copy_flat[pytree-compile] 0.1283ms 69.3859μs 14.4121 KOps/s 14.3415 KOps/s $\color{#35bf28}+0.49\%$
test_compile_copy_flat[pytree-eager] 82.5610μs 51.6750μs 19.3517 KOps/s 19.4240 KOps/s $\color{#d91a1a}-0.37\%$
test_compile_assign_and_add[tensordict-compile] 1.6596ms 0.3963ms 2.5236 KOps/s 2.1927 KOps/s $\textbf{\color{#35bf28}+15.09\%}$
test_compile_assign_and_add[tensordict-eager] 3.2028ms 2.7944ms 357.8566 Ops/s 360.4011 Ops/s $\color{#d91a1a}-0.71\%$
test_compile_assign_and_add[pytree-compile] 1.6352ms 0.4407ms 2.2691 KOps/s 2.2206 KOps/s $\color{#35bf28}+2.18\%$
test_compile_assign_and_add[pytree-eager] 3.0330ms 2.8566ms 350.0643 Ops/s 354.7956 Ops/s $\color{#d91a1a}-1.33\%$
test_compile_indexing[tensor-tensordict-compile] 0.4100ms 0.1149ms 8.7055 KOps/s 8.1365 KOps/s $\textbf{\color{#35bf28}+6.99\%}$
test_compile_indexing[tensor-tensordict-eager] 0.5805ms 83.1589μs 12.0252 KOps/s 11.4177 KOps/s $\textbf{\color{#35bf28}+5.32\%}$
test_compile_indexing[tensor-tensorclass-compile] 0.3953ms 0.1119ms 8.9362 KOps/s 9.1326 KOps/s $\color{#d91a1a}-2.15\%$
test_compile_indexing[tensor-tensorclass-eager] 0.2099ms 70.0180μs 14.2820 KOps/s 14.1111 KOps/s $\color{#35bf28}+1.21\%$
test_compile_indexing[tensor-pytree-compile] 0.2066ms 0.1111ms 9.0022 KOps/s 9.2159 KOps/s $\color{#d91a1a}-2.32\%$
test_compile_indexing[tensor-pytree-eager] 0.1192ms 70.0232μs 14.2810 KOps/s 14.1374 KOps/s $\color{#35bf28}+1.02\%$
test_compile_indexing[slice-tensordict-compile] 0.2548ms 0.1075ms 9.2988 KOps/s 9.8400 KOps/s $\textbf{\color{#d91a1a}-5.50\%}$
test_compile_indexing[slice-tensordict-eager] 0.1538ms 18.0104μs 55.5235 KOps/s 54.5182 KOps/s $\color{#35bf28}+1.84\%$
test_compile_indexing[slice-tensorclass-compile] 0.2728ms 98.8945μs 10.1118 KOps/s 10.2490 KOps/s $\color{#d91a1a}-1.34\%$
test_compile_indexing[slice-tensorclass-eager] 0.1219ms 16.1665μs 61.8564 KOps/s 62.0558 KOps/s $\color{#d91a1a}-0.32\%$
test_compile_indexing[slice-pytree-compile] 0.2918ms 0.1023ms 9.7780 KOps/s 10.2091 KOps/s $\color{#d91a1a}-4.22\%$
test_compile_indexing[slice-pytree-eager] 48.6610μs 16.1136μs 62.0595 KOps/s 62.5605 KOps/s $\color{#d91a1a}-0.80\%$
test_compile_indexing[int-tensordict-compile] 0.2421ms 0.1063ms 9.4080 KOps/s 9.7531 KOps/s $\color{#d91a1a}-3.54\%$
test_compile_indexing[int-tensordict-eager] 0.5720ms 17.9233μs 55.7933 KOps/s 55.5097 KOps/s $\color{#35bf28}+0.51\%$
test_compile_indexing[int-tensorclass-compile] 0.2615ms 0.1032ms 9.6944 KOps/s 10.1572 KOps/s $\color{#d91a1a}-4.56\%$
test_compile_indexing[int-tensorclass-eager] 47.2610μs 16.0919μs 62.1432 KOps/s 62.0259 KOps/s $\color{#35bf28}+0.19\%$
test_compile_indexing[int-pytree-compile] 0.2018ms 99.2004μs 10.0806 KOps/s 10.1627 KOps/s $\color{#d91a1a}-0.81\%$
test_compile_indexing[int-pytree-eager] 46.1310μs 16.1806μs 61.8026 KOps/s 62.4001 KOps/s $\color{#d91a1a}-0.96\%$
test_mod_add[eager] 75.9210μs 31.9680μs 31.2813 KOps/s 30.7202 KOps/s $\color{#35bf28}+1.83\%$
test_mod_add[compile] 0.2473ms 77.7375μs 12.8638 KOps/s 12.7043 KOps/s $\color{#35bf28}+1.26\%$
test_mod_add[compile-overhead] 0.3106ms 0.1611ms 6.2068 KOps/s 5.8737 KOps/s $\textbf{\color{#35bf28}+5.67\%}$
test_mod_wrap[eager] 0.3262ms 0.2552ms 3.9188 KOps/s 3.8908 KOps/s $\color{#35bf28}+0.72\%$
test_mod_wrap[compile] 1.5851ms 0.2858ms 3.4990 KOps/s 3.4370 KOps/s $\color{#35bf28}+1.81\%$
test_mod_wrap[compile-overhead] 7.8839ms 4.0651ms 245.9975 Ops/s 255.6534 Ops/s $\color{#d91a1a}-3.78\%$
test_mod_wrap_and_backward[eager] 1.6328ms 1.5078ms 663.2340 Ops/s 682.0321 Ops/s $\color{#d91a1a}-2.76\%$
test_mod_wrap_and_backward[compile] 1.5596ms 1.4029ms 712.7888 Ops/s 719.0793 Ops/s $\color{#d91a1a}-0.87\%$
test_mod_wrap_and_backward[compile-overhead] 1.4938ms 1.0241ms 976.5118 Ops/s 968.8107 Ops/s $\color{#35bf28}+0.79\%$
test_seq_add[eager] 0.2399ms 96.2835μs 10.3860 KOps/s 9.8328 KOps/s $\textbf{\color{#35bf28}+5.63\%}$
test_seq_add[compile] 0.2189ms 92.1589μs 10.8508 KOps/s 10.7371 KOps/s $\color{#35bf28}+1.06\%$
test_seq_add[compile-overhead] 0.1698ms 0.1276ms 7.8385 KOps/s 7.7521 KOps/s $\color{#35bf28}+1.11\%$
test_seq_wrap[eager] 0.5133ms 0.3976ms 2.5149 KOps/s 2.4448 KOps/s $\color{#35bf28}+2.87\%$
test_seq_wrap[compile] 0.4351ms 0.3053ms 3.2758 KOps/s 3.2632 KOps/s $\color{#35bf28}+0.39\%$
test_seq_wrap[compile-overhead] 0.2950ms 0.2248ms 4.4483 KOps/s 4.4305 KOps/s $\color{#35bf28}+0.40\%$
test_func_call_runtime[False-eager] 0.8946ms 0.7535ms 1.3271 KOps/s 1.3058 KOps/s $\color{#35bf28}+1.63\%$
test_func_call_runtime[False-compile] 0.9822ms 0.7701ms 1.2986 KOps/s 1.2936 KOps/s $\color{#35bf28}+0.39\%$
test_func_call_runtime[False-compile-overhead] 0.5436ms 0.3639ms 2.7483 KOps/s 2.7270 KOps/s $\color{#35bf28}+0.78\%$
test_func_call_runtime[True-eager] 1.1954ms 0.9186ms 1.0886 KOps/s 1.0803 KOps/s $\color{#35bf28}+0.77\%$
test_func_call_runtime[True-compile] 0.8531ms 0.7839ms 1.2757 KOps/s 1.2612 KOps/s $\color{#35bf28}+1.14\%$
test_func_call_runtime[True-compile-overhead] 0.4628ms 0.3840ms 2.6042 KOps/s 2.5780 KOps/s $\color{#35bf28}+1.02\%$
test_func_call_cm_runtime[False-eager] 0.8326ms 0.7533ms 1.3276 KOps/s 1.3114 KOps/s $\color{#35bf28}+1.23\%$
test_func_call_cm_runtime[False-compile] 0.8386ms 0.7591ms 1.3174 KOps/s 1.3061 KOps/s $\color{#35bf28}+0.86\%$
test_func_call_cm_runtime[False-compile-overhead] 0.4370ms 0.3650ms 2.7397 KOps/s 2.7204 KOps/s $\color{#35bf28}+0.71\%$
test_func_call_cm_runtime[True-eager] 1.1903ms 1.0204ms 979.9999 Ops/s 970.9429 Ops/s $\color{#35bf28}+0.93\%$
test_func_call_cm_runtime[True-compile] 0.8930ms 0.8113ms 1.2326 KOps/s 1.2234 KOps/s $\color{#35bf28}+0.76\%$
test_func_call_cm_runtime[True-compile-overhead] 0.5348ms 0.4097ms 2.4407 KOps/s 2.4149 KOps/s $\color{#35bf28}+1.07\%$
test_vmap_func_call_cm_runtime[eager] 2.5776ms 2.1108ms 473.7562 Ops/s 472.4186 Ops/s $\color{#35bf28}+0.28\%$
test_vmap_func_call_cm_runtime[compile] 0.9731ms 0.8220ms 1.2165 KOps/s 1.1942 KOps/s $\color{#35bf28}+1.86\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.4852ms 0.4132ms 2.4203 KOps/s 2.4113 KOps/s $\color{#35bf28}+0.37\%$
test_distributed 3.7597ms 0.1341ms 7.4567 KOps/s 8.7914 KOps/s $\textbf{\color{#d91a1a}-15.18\%}$
test_tdmodule 0.3124ms 14.5291μs 68.8276 KOps/s 72.7571 KOps/s $\textbf{\color{#d91a1a}-5.40\%}$
test_tdmodule_dispatch 46.7700μs 27.3677μs 36.5395 KOps/s 37.1916 KOps/s $\color{#d91a1a}-1.75\%$
test_tdseq 38.6410μs 15.6519μs 63.8899 KOps/s 65.5575 KOps/s $\color{#d91a1a}-2.54\%$
test_tdseq_dispatch 53.4700μs 30.4383μs 32.8534 KOps/s 33.0635 KOps/s $\color{#d91a1a}-0.64\%$
test_instantiation_functorch 1.7186ms 1.5747ms 635.0537 Ops/s 630.3275 Ops/s $\color{#35bf28}+0.75\%$
test_exec_functorch 0.2545ms 0.1505ms 6.6456 KOps/s 6.7186 KOps/s $\color{#d91a1a}-1.09\%$
test_exec_functional_call 0.1922ms 0.1430ms 6.9953 KOps/s 7.0488 KOps/s $\color{#d91a1a}-0.76\%$
test_exec_td_decorator 0.3758ms 0.1925ms 5.1944 KOps/s 5.3703 KOps/s $\color{#d91a1a}-3.28\%$
test_vmap_mlp_speed_decorator[True-True] 0.8242ms 0.7014ms 1.4257 KOps/s 1.4088 KOps/s $\color{#35bf28}+1.20\%$
test_vmap_mlp_speed_decorator[True-False] 0.8253ms 0.6930ms 1.4431 KOps/s 1.4069 KOps/s $\color{#35bf28}+2.57\%$
test_vmap_mlp_speed_decorator[False-True] 0.7571ms 0.6091ms 1.6418 KOps/s 1.5981 KOps/s $\color{#35bf28}+2.73\%$
test_vmap_mlp_speed_decorator[False-False] 0.7117ms 0.5951ms 1.6804 KOps/s 1.6323 KOps/s $\color{#35bf28}+2.95\%$
test_vmap_transformer_speed_decorator[True-True] 20.0284ms 19.7176ms 50.7161 Ops/s 50.9890 Ops/s $\color{#d91a1a}-0.54\%$
test_vmap_transformer_speed_decorator[True-False] 20.4206ms 19.7829ms 50.5487 Ops/s 51.0459 Ops/s $\color{#d91a1a}-0.97\%$
test_vmap_transformer_speed_decorator[False-True] 19.7722ms 19.6211ms 50.9655 Ops/s 51.4541 Ops/s $\color{#d91a1a}-0.95\%$
test_vmap_transformer_speed_decorator[False-False] 19.6675ms 19.5746ms 51.0866 Ops/s 51.3989 Ops/s $\color{#d91a1a}-0.61\%$
test_to_module_speed[True] 2.1616ms 0.9405ms 1.0632 KOps/s 1.0560 KOps/s $\color{#35bf28}+0.69\%$
test_to_module_speed[False] 1.0108ms 0.9166ms 1.0909 KOps/s 1.0937 KOps/s $\color{#d91a1a}-0.25\%$
test_tc_init 77.0210μs 37.0087μs 27.0207 KOps/s 29.4483 KOps/s $\textbf{\color{#d91a1a}-8.24\%}$
test_tc_init_nested 0.1150ms 68.6509μs 14.5665 KOps/s 13.9524 KOps/s $\color{#35bf28}+4.40\%$
test_tc_first_layer_tensor 4.7229μs 0.6898μs 1.4498 MOps/s 1.4425 MOps/s $\color{#35bf28}+0.51\%$
test_tc_first_layer_nontensor 25.8100μs 2.3107μs 432.7720 KOps/s 429.0044 KOps/s $\color{#35bf28}+0.88\%$
test_tc_second_layer_tensor 8.6277μs 1.4112μs 708.5949 KOps/s 704.2830 KOps/s $\color{#35bf28}+0.61\%$
test_tc_second_layer_nontensor 25.2510μs 3.0671μs 326.0420 KOps/s 325.8264 KOps/s $\color{#35bf28}+0.07\%$
test_unbind 7.0541ms 6.7325ms 148.5324 Ops/s 148.5034 Ops/s $\color{#35bf28}+0.02\%$
test_full_like 10.8952ms 9.3820ms 106.5872 Ops/s 103.9159 Ops/s $\color{#35bf28}+2.57\%$
test_zeros_like 9.1767ms 7.1421ms 140.0139 Ops/s 113.9894 Ops/s $\textbf{\color{#35bf28}+22.83\%}$
test_ones_like 4.9481ms 4.2736ms 233.9934 Ops/s 231.5253 Ops/s $\color{#35bf28}+1.07\%$
test_clone 6.8602ms 6.4225ms 155.7035 Ops/s 155.5442 Ops/s $\color{#35bf28}+0.10\%$
test_squeeze 78.7710μs 10.0152μs 99.8487 KOps/s 91.4597 KOps/s $\textbf{\color{#35bf28}+9.17\%}$
test_unsqueeze 0.1282ms 77.1532μs 12.9612 KOps/s 13.5811 KOps/s $\color{#d91a1a}-4.56\%$
test_split 0.3077ms 0.1731ms 5.7766 KOps/s 5.9984 KOps/s $\color{#d91a1a}-3.70\%$
test_permute 0.2368ms 0.1894ms 5.2809 KOps/s 5.3346 KOps/s $\color{#d91a1a}-1.01\%$
test_stack 50.8373ms 50.5695ms 19.7748 Ops/s 19.7770 Ops/s $\color{#d91a1a}-0.01\%$
test_cat 50.8380ms 50.4965ms 19.8033 Ops/s 19.8564 Ops/s $\color{#d91a1a}-0.27\%$

@vmoens
Update
50c43f9 
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Nov 5, 2024
@vmoens
[Performance] Faster to
55f2a39 
ghstack-source-id: 3dfb0b66fae82dc8cf5ef2a14eccb1bec5237ebb
Pull Request resolved: #1073
@vmoens vmoens merged commit 50c43f9 into gh/vmoens/35/base Nov 5, 2024
50 of 51 checks passed
vmoens added a commit that referenced this pull request Nov 5, 2024
@vmoens
[Performance] Faster to
6272510 
ghstack-source-id: 3dfb0b66fae82dc8cf5ef2a14eccb1bec5237ebb
Pull Request resolved: #1073
@vmoens vmoens deleted the gh/vmoens/35/head branch November 5, 2024 15:28
vmoens added a commit that referenced this pull request Nov 14, 2024
@vmoens
[Performance] Faster to
f031bf2 
ghstack-source-id: 3dfb0b66fae82dc8cf5ef2a14eccb1bec5237ebb
Pull Request resolved: #1073

(cherry picked from commit 6272510)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. Performance
Projects
None yet
Milestone
No milestone
Development

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