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] strict kwarg in TDModule #1234

Merged
merged 1 commit into from
Feb 24, 2025
Merged

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 24, 2025

Stack from ghstack (oldest at bottom):

[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 24, 2025
ghstack-source-id: ced22130bf45945e2671fa9c2e776d482fcd8b15
Pull Request resolved: #1234
@facebook-github-bot facebook-github-bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Feb 24, 2025
@vmoens vmoens merged commit 8086c7c into gh/vmoens/48/base Feb 24, 2025
15 of 28 checks passed
vmoens added a commit that referenced this pull request Feb 24, 2025
ghstack-source-id: ced22130bf45945e2671fa9c2e776d482fcd8b15
Pull Request resolved: #1234
@vmoens vmoens deleted the gh/vmoens/48/head branch February 24, 2025 18:56
Copy link

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}21$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 44.5440μs 20.8115μs 48.0503 KOps/s 46.6387 KOps/s $\color{#35bf28}+3.03\%$
test_plain_set_stack_nested 44.5230μs 21.3142μs 46.9170 KOps/s 47.0141 KOps/s $\color{#d91a1a}-0.21\%$
test_plain_set_nested_inplace 62.7880μs 23.0799μs 43.3277 KOps/s 42.7302 KOps/s $\color{#35bf28}+1.40\%$
test_plain_set_stack_nested_inplace 0.1144ms 22.6772μs 44.0971 KOps/s 43.2913 KOps/s $\color{#35bf28}+1.86\%$
test_items 40.1350μs 4.1001μs 243.8940 KOps/s 242.4979 KOps/s $\color{#35bf28}+0.58\%$
test_items_nested 0.8783ms 0.4039ms 2.4760 KOps/s 2.4864 KOps/s $\color{#d91a1a}-0.42\%$
test_items_nested_locked 0.6842ms 0.4038ms 2.4763 KOps/s 2.4921 KOps/s $\color{#d91a1a}-0.63\%$
test_items_nested_leaf 0.1572ms 77.8227μs 12.8497 KOps/s 13.1644 KOps/s $\color{#d91a1a}-2.39\%$
test_items_stack_nested 0.5334ms 0.4028ms 2.4828 KOps/s 2.4949 KOps/s $\color{#d91a1a}-0.48\%$
test_items_stack_nested_leaf 0.1709ms 77.3402μs 12.9299 KOps/s 13.0209 KOps/s $\color{#d91a1a}-0.70\%$
test_items_stack_nested_locked 0.6530ms 0.4032ms 2.4803 KOps/s 2.4879 KOps/s $\color{#d91a1a}-0.31\%$
test_keys 40.5960μs 3.4739μs 287.8610 KOps/s 288.9440 KOps/s $\color{#d91a1a}-0.37\%$
test_keys_nested 0.3025ms 0.1628ms 6.1409 KOps/s 6.0901 KOps/s $\color{#35bf28}+0.83\%$
test_keys_nested_locked 1.7431ms 0.1693ms 5.9061 KOps/s 5.8763 KOps/s $\color{#35bf28}+0.51\%$
test_keys_nested_leaf 0.1951ms 0.1425ms 7.0163 KOps/s 6.9829 KOps/s $\color{#35bf28}+0.48\%$
test_keys_stack_nested 0.3970ms 0.1633ms 6.1226 KOps/s 6.1152 KOps/s $\color{#35bf28}+0.12\%$
test_keys_stack_nested_leaf 0.2253ms 0.1437ms 6.9583 KOps/s 6.9786 KOps/s $\color{#d91a1a}-0.29\%$
test_keys_stack_nested_locked 0.3098ms 0.1694ms 5.9039 KOps/s 5.8883 KOps/s $\color{#35bf28}+0.26\%$
test_values 12.9522μs 1.0346μs 966.5619 KOps/s 965.6817 KOps/s $\color{#35bf28}+0.09\%$
test_values_nested 0.1251ms 62.4432μs 16.0145 KOps/s 16.0757 KOps/s $\color{#d91a1a}-0.38\%$
test_values_nested_locked 0.1084ms 61.9900μs 16.1316 KOps/s 14.9192 KOps/s $\textbf{\color{#35bf28}+8.13\%}$
test_values_nested_leaf 0.1422ms 71.2718μs 14.0308 KOps/s 14.1030 KOps/s $\color{#d91a1a}-0.51\%$
test_values_stack_nested 0.1518ms 63.9454μs 15.6383 KOps/s 16.1065 KOps/s $\color{#d91a1a}-2.91\%$
test_values_stack_nested_leaf 0.1729ms 70.3673μs 14.2112 KOps/s 14.0668 KOps/s $\color{#35bf28}+1.03\%$
test_values_stack_nested_locked 0.1242ms 62.1500μs 16.0901 KOps/s 15.8967 KOps/s $\color{#35bf28}+1.22\%$
test_membership 28.1630μs 0.8533μs 1.1719 MOps/s 1.3872 MOps/s $\textbf{\color{#d91a1a}-15.52\%}$
test_membership_nested 41.5880μs 2.8894μs 346.0930 KOps/s 332.9129 KOps/s $\color{#35bf28}+3.96\%$
test_membership_nested_leaf 30.1160μs 2.9113μs 343.4894 KOps/s 325.0490 KOps/s $\textbf{\color{#35bf28}+5.67\%}$
test_membership_stacked_nested 57.3970μs 3.0027μs 333.0304 KOps/s 335.0742 KOps/s $\color{#d91a1a}-0.61\%$
test_membership_stacked_nested_leaf 38.3020μs 2.8788μs 347.3689 KOps/s 326.4315 KOps/s $\textbf{\color{#35bf28}+6.41\%}$
test_membership_nested_last 41.5580μs 4.3468μs 230.0560 KOps/s 224.2451 KOps/s $\color{#35bf28}+2.59\%$
test_membership_nested_leaf_last 59.7620μs 4.3374μs 230.5504 KOps/s 223.4468 KOps/s $\color{#35bf28}+3.18\%$
test_membership_stacked_nested_last 35.8670μs 4.2944μs 232.8635 KOps/s 227.1440 KOps/s $\color{#35bf28}+2.52\%$
test_membership_stacked_nested_leaf_last 40.3650μs 4.3417μs 230.3219 KOps/s 225.0434 KOps/s $\color{#35bf28}+2.35\%$
test_nested_getleaf 35.2960μs 10.6503μs 93.8938 KOps/s 94.1559 KOps/s $\color{#d91a1a}-0.28\%$
test_nested_get 50.3840μs 10.0540μs 99.4625 KOps/s 98.4708 KOps/s $\color{#35bf28}+1.01\%$
test_stacked_getleaf 47.6990μs 10.6758μs 93.6697 KOps/s 92.0810 KOps/s $\color{#35bf28}+1.73\%$
test_stacked_get 34.8750μs 9.9750μs 100.2511 KOps/s 96.2009 KOps/s $\color{#35bf28}+4.21\%$
test_nested_getitemleaf 69.6990μs 11.2515μs 88.8767 KOps/s 87.0460 KOps/s $\color{#35bf28}+2.10\%$
test_nested_getitem 50.9290μs 10.6513μs 93.8853 KOps/s 91.6991 KOps/s $\color{#35bf28}+2.38\%$
test_stacked_getitemleaf 54.4020μs 11.2661μs 88.7618 KOps/s 87.8199 KOps/s $\color{#35bf28}+1.07\%$
test_stacked_getitem 55.2640μs 10.7914μs 92.6665 KOps/s 89.7925 KOps/s $\color{#35bf28}+3.20\%$
test_lock_nested 0.6641ms 0.4156ms 2.4060 KOps/s 2.4137 KOps/s $\color{#d91a1a}-0.32\%$
test_lock_stack_nested 0.4985ms 0.4240ms 2.3583 KOps/s 2.3439 KOps/s $\color{#35bf28}+0.61\%$
test_unlock_nested 0.4727ms 0.3370ms 2.9673 KOps/s 2.9308 KOps/s $\color{#35bf28}+1.24\%$
test_unlock_stack_nested 0.5193ms 0.3424ms 2.9210 KOps/s 2.8541 KOps/s $\color{#35bf28}+2.34\%$
test_flatten_speed 0.1568ms 0.1002ms 9.9845 KOps/s 10.0611 KOps/s $\color{#d91a1a}-0.76\%$
test_unflatten_speed 0.6516ms 0.5179ms 1.9309 KOps/s 1.9077 KOps/s $\color{#35bf28}+1.21\%$
test_common_ops 4.7876ms 0.8142ms 1.2281 KOps/s 1.1963 KOps/s $\color{#35bf28}+2.66\%$
test_creation 38.4020μs 2.4804μs 403.1681 KOps/s 386.2634 KOps/s $\color{#35bf28}+4.38\%$
test_creation_empty 38.1820μs 12.9328μs 77.3227 KOps/s 72.7890 KOps/s $\textbf{\color{#35bf28}+6.23\%}$
test_creation_nested_1 53.7500μs 15.4750μs 64.6202 KOps/s 60.3236 KOps/s $\textbf{\color{#35bf28}+7.12\%}$
test_creation_nested_2 72.3260μs 20.1258μs 49.6875 KOps/s 46.4840 KOps/s $\textbf{\color{#35bf28}+6.89\%}$
test_clone 59.1600μs 13.3650μs 74.8223 KOps/s 73.4621 KOps/s $\color{#35bf28}+1.85\%$
test_getitem[int] 1.0310ms 12.6048μs 79.3352 KOps/s 76.4370 KOps/s $\color{#35bf28}+3.79\%$
test_getitem[slice_int] 0.1308ms 24.0047μs 41.6585 KOps/s 39.2667 KOps/s $\textbf{\color{#35bf28}+6.09\%}$
test_getitem[range] 0.1637ms 48.6543μs 20.5532 KOps/s 19.3250 KOps/s $\textbf{\color{#35bf28}+6.36\%}$
test_getitem[tuple] 0.1331ms 20.1763μs 49.5631 KOps/s 48.4410 KOps/s $\color{#35bf28}+2.32\%$
test_getitem[list] 0.1583ms 44.7449μs 22.3489 KOps/s 21.4225 KOps/s $\color{#35bf28}+4.32\%$
test_setitem_dim[int] 53.9010μs 25.2455μs 39.6109 KOps/s 38.8467 KOps/s $\color{#35bf28}+1.97\%$
test_setitem_dim[slice_int] 0.1204ms 49.6730μs 20.1317 KOps/s 19.8285 KOps/s $\color{#35bf28}+1.53\%$
test_setitem_dim[range] 0.1007ms 73.8611μs 13.5389 KOps/s 12.9165 KOps/s $\color{#35bf28}+4.82\%$
test_setitem_dim[tuple] 96.6710μs 40.9478μs 24.4214 KOps/s 24.3002 KOps/s $\color{#35bf28}+0.50\%$
test_setitem 69.0890μs 20.9258μs 47.7879 KOps/s 45.9325 KOps/s $\color{#35bf28}+4.04\%$
test_set 0.2744ms 20.1926μs 49.5230 KOps/s 47.9798 KOps/s $\color{#35bf28}+3.22\%$
test_set_shared 4.0521ms 0.1826ms 5.4752 KOps/s 5.3289 KOps/s $\color{#35bf28}+2.74\%$
test_update 0.1031ms 23.6508μs 42.2818 KOps/s 40.2226 KOps/s $\textbf{\color{#35bf28}+5.12\%}$
test_update_nested 97.2120μs 34.0352μs 29.3814 KOps/s 27.6113 KOps/s $\textbf{\color{#35bf28}+6.41\%}$
test_update__nested 0.4194ms 33.7506μs 29.6291 KOps/s 29.4523 KOps/s $\color{#35bf28}+0.60\%$
test_set_nested 64.5010μs 22.4249μs 44.5932 KOps/s 42.0006 KOps/s $\textbf{\color{#35bf28}+6.17\%}$
test_set_nested_new 0.1019ms 26.9452μs 37.1124 KOps/s 35.0887 KOps/s $\textbf{\color{#35bf28}+5.77\%}$
test_select 95.8190μs 42.5939μs 23.4776 KOps/s 22.4960 KOps/s $\color{#35bf28}+4.36\%$
test_select_nested 0.1247ms 62.2262μs 16.0704 KOps/s 16.1039 KOps/s $\color{#d91a1a}-0.21\%$
test_exclude_nested 0.1861ms 80.0994μs 12.4845 KOps/s 12.3054 KOps/s $\color{#35bf28}+1.46\%$
test_empty[True] 0.5905ms 0.4093ms 2.4434 KOps/s 2.4473 KOps/s $\color{#d91a1a}-0.16\%$
test_empty[False] 11.0858μs 1.3902μs 719.3449 KOps/s 739.1336 KOps/s $\color{#d91a1a}-2.68\%$
test_unbind_speed 0.6323ms 0.2713ms 3.6862 KOps/s 3.5803 KOps/s $\color{#35bf28}+2.96\%$
test_unbind_speed_stack0 0.3883ms 0.2674ms 3.7396 KOps/s 3.6020 KOps/s $\color{#35bf28}+3.82\%$
test_unbind_speed_stack1 0.1015s 0.7317ms 1.3667 KOps/s 1.2290 KOps/s $\textbf{\color{#35bf28}+11.21\%}$
test_split 0.1007s 1.7479ms 572.1204 Ops/s 543.1077 Ops/s $\textbf{\color{#35bf28}+5.34\%}$
test_chunk 0.1008s 1.7475ms 572.2501 Ops/s 611.5674 Ops/s $\textbf{\color{#d91a1a}-6.43\%}$
test_consolidate_njt[False-None] 12.1605ms 8.1845ms 122.1815 Ops/s 109.6394 Ops/s $\textbf{\color{#35bf28}+11.44\%}$
test_creation[device0] 0.2293ms 89.1934μs 11.2116 KOps/s 10.7767 KOps/s $\color{#35bf28}+4.04\%$
test_creation_from_tensor 3.4611ms 93.8429μs 10.6561 KOps/s 10.6227 KOps/s $\color{#35bf28}+0.31\%$
test_add_one[memmap_tensor0] 0.1312ms 4.7477μs 210.6289 KOps/s 202.8342 KOps/s $\color{#35bf28}+3.84\%$
test_contiguous[memmap_tensor0] 20.8690μs 0.5040μs 1.9842 MOps/s 1.9479 MOps/s $\color{#35bf28}+1.87\%$
test_stack[memmap_tensor0] 25.5470μs 3.3360μs 299.7586 KOps/s 293.4258 KOps/s $\color{#35bf28}+2.16\%$
test_memmaptd_index 0.3152ms 0.2315ms 4.3197 KOps/s 4.4246 KOps/s $\color{#d91a1a}-2.37\%$
test_memmaptd_index_astensor 1.0517ms 0.3160ms 3.1641 KOps/s 3.2145 KOps/s $\color{#d91a1a}-1.57\%$
test_memmaptd_index_op 0.8244ms 0.5920ms 1.6893 KOps/s 1.6646 KOps/s $\color{#35bf28}+1.49\%$
test_serialize_model 0.2081s 0.1305s 7.6638 Ops/s 8.5426 Ops/s $\textbf{\color{#d91a1a}-10.29\%}$
test_serialize_model_pickle 0.4433s 0.3884s 2.5750 Ops/s 2.5359 Ops/s $\color{#35bf28}+1.54\%$
test_serialize_weights 0.1335s 0.1138s 8.7896 Ops/s 8.7133 Ops/s $\color{#35bf28}+0.88\%$
test_serialize_weights_returnearly 0.1892s 0.1611s 6.2063 Ops/s 6.1198 Ops/s $\color{#35bf28}+1.41\%$
test_serialize_weights_pickle 0.4901s 0.4039s 2.4757 Ops/s 1.1921 Ops/s $\textbf{\color{#35bf28}+107.67\%}$
test_serialize_weights_filesystem 0.2401s 0.1602s 6.2414 Ops/s 6.7484 Ops/s $\textbf{\color{#d91a1a}-7.51\%}$
test_serialize_model_filesystem 0.1526s 0.1451s 6.8918 Ops/s 6.9043 Ops/s $\color{#d91a1a}-0.18\%$
test_reshape_pytree 60.9040μs 26.2602μs 38.0804 KOps/s 37.1824 KOps/s $\color{#35bf28}+2.42\%$
test_reshape_td 77.3650μs 32.9513μs 30.3479 KOps/s 28.6111 KOps/s $\textbf{\color{#35bf28}+6.07\%}$
test_view_pytree 72.8460μs 26.1660μs 38.2176 KOps/s 37.5929 KOps/s $\color{#35bf28}+1.66\%$
test_view_td 83.9070μs 41.7050μs 23.9779 KOps/s 23.9365 KOps/s $\color{#35bf28}+0.17\%$
test_unbind_pytree 0.1278ms 29.8568μs 33.4932 KOps/s 33.9263 KOps/s $\color{#d91a1a}-1.28\%$
test_unbind_td 0.3381ms 40.3522μs 24.7818 KOps/s 24.3762 KOps/s $\color{#35bf28}+1.66\%$
test_split_pytree 89.8380μs 29.0724μs 34.3969 KOps/s 34.6651 KOps/s $\color{#d91a1a}-0.77\%$
test_split_td 0.2051ms 45.8796μs 21.7962 KOps/s 21.1169 KOps/s $\color{#35bf28}+3.22\%$
test_add_pytree 82.8960μs 35.6728μs 28.0326 KOps/s 27.1783 KOps/s $\color{#35bf28}+3.14\%$
test_add_td 0.1254ms 59.0464μs 16.9358 KOps/s 16.4129 KOps/s $\color{#35bf28}+3.19\%$
test_compile_add_one_nested[tensordict-compile] 0.1304ms 66.8269μs 14.9640 KOps/s 14.8552 KOps/s $\color{#35bf28}+0.73\%$
test_compile_add_one_nested[tensordict-eager] 1.0931ms 0.1731ms 5.7760 KOps/s 5.8314 KOps/s $\color{#d91a1a}-0.95\%$
test_compile_add_one_nested[pytree-compile] 0.1070ms 45.5789μs 21.9400 KOps/s 21.8390 KOps/s $\color{#35bf28}+0.46\%$
test_compile_add_one_nested[pytree-eager] 0.3249ms 0.1201ms 8.3264 KOps/s 8.4396 KOps/s $\color{#d91a1a}-1.34\%$
test_compile_copy_nested[tensordict-compile] 96.1200μs 27.9435μs 35.7865 KOps/s 34.9946 KOps/s $\color{#35bf28}+2.26\%$
test_compile_copy_nested[tensordict-eager] 0.1295ms 58.1864μs 17.1861 KOps/s 17.1641 KOps/s $\color{#35bf28}+0.13\%$
test_compile_copy_nested[pytree-compile] 0.1623ms 79.1142μs 12.6400 KOps/s 12.2583 KOps/s $\color{#35bf28}+3.11\%$
test_compile_copy_nested[pytree-eager] 0.1289ms 66.6430μs 15.0053 KOps/s 14.7371 KOps/s $\color{#35bf28}+1.82\%$
test_compile_add_one_flat[tensordict-compile] 0.2203ms 0.1064ms 9.4011 KOps/s 9.4772 KOps/s $\color{#d91a1a}-0.80\%$
test_compile_add_one_flat[tensordict-eager] 0.4169ms 0.2142ms 4.6680 KOps/s 4.6777 KOps/s $\color{#d91a1a}-0.21\%$
test_compile_add_one_flat[tensorclass-compile] 0.1244ms 47.7511μs 20.9419 KOps/s 20.9979 KOps/s $\color{#d91a1a}-0.27\%$
test_compile_add_one_flat[tensorclass-eager] 0.1957ms 66.4847μs 15.0411 KOps/s 14.8497 KOps/s $\color{#35bf28}+1.29\%$
test_compile_add_one_flat[pytree-compile] 0.1832ms 0.1013ms 9.8753 KOps/s 10.0566 KOps/s $\color{#d91a1a}-1.80\%$
test_compile_add_one_flat[pytree-eager] 0.3662ms 0.2035ms 4.9146 KOps/s 4.8537 KOps/s $\color{#35bf28}+1.25\%$
test_compile_add_self_flat[tensordict-eager] 0.3958ms 0.2317ms 4.3157 KOps/s 4.3130 KOps/s $\color{#35bf28}+0.06\%$
test_compile_add_self_flat[tensordict-compile] 0.2957ms 0.1106ms 9.0386 KOps/s 9.4469 KOps/s $\color{#d91a1a}-4.32\%$
test_compile_add_self_flat[tensorclass-eager] 0.1442ms 62.6257μs 15.9679 KOps/s 15.9220 KOps/s $\color{#35bf28}+0.29\%$
test_compile_add_self_flat[tensorclass-compile] 97.7530μs 48.4434μs 20.6426 KOps/s 21.2489 KOps/s $\color{#d91a1a}-2.85\%$
test_compile_add_self_flat[pytree-eager] 0.2200ms 0.1564ms 6.3927 KOps/s 6.3669 KOps/s $\color{#35bf28}+0.41\%$
test_compile_add_self_flat[pytree-compile] 0.2388ms 0.1004ms 9.9560 KOps/s 10.0352 KOps/s $\color{#d91a1a}-0.79\%$
test_compile_copy_flat[tensordict-compile] 0.1074ms 21.2449μs 47.0701 KOps/s 46.8028 KOps/s $\color{#35bf28}+0.57\%$
test_compile_copy_flat[tensordict-eager] 0.1146ms 67.9247μs 14.7222 KOps/s 15.3273 KOps/s $\color{#d91a1a}-3.95\%$
test_compile_copy_flat[pytree-compile] 0.1976ms 81.8733μs 12.2140 KOps/s 12.1018 KOps/s $\color{#35bf28}+0.93\%$
test_compile_copy_flat[pytree-eager] 0.1158ms 67.3871μs 14.8396 KOps/s 14.7384 KOps/s $\color{#35bf28}+0.69\%$
test_compile_assign_and_add[tensordict-compile] 0.3058ms 0.2166ms 4.6179 KOps/s 4.7883 KOps/s $\color{#d91a1a}-3.56\%$
test_compile_assign_and_add[tensordict-eager] 1.8284ms 1.3787ms 725.3471 Ops/s 730.1003 Ops/s $\color{#d91a1a}-0.65\%$
test_compile_assign_and_add[pytree-compile] 0.4391ms 0.2088ms 4.7890 KOps/s 4.7610 KOps/s $\color{#35bf28}+0.59\%$
test_compile_assign_and_add[pytree-eager] 0.9293ms 0.8207ms 1.2185 KOps/s 1.1952 KOps/s $\color{#35bf28}+1.94\%$
test_compile_assign_and_add_stack[compile] 0.9151ms 0.4676ms 2.1384 KOps/s 2.2137 KOps/s $\color{#d91a1a}-3.40\%$
test_compile_assign_and_add_stack[eager] 4.6469ms 2.7047ms 369.7282 Ops/s 356.6375 Ops/s $\color{#35bf28}+3.67\%$
test_compile_indexing[tensor-tensordict-compile] 0.1125ms 38.8452μs 25.7432 KOps/s 25.8316 KOps/s $\color{#d91a1a}-0.34\%$
test_compile_indexing[tensor-tensordict-eager] 0.5931ms 34.1409μs 29.2903 KOps/s 30.2454 KOps/s $\color{#d91a1a}-3.16\%$
test_compile_indexing[tensor-tensorclass-compile] 78.8580μs 31.5978μs 31.6478 KOps/s 31.9479 KOps/s $\color{#d91a1a}-0.94\%$
test_compile_indexing[tensor-tensorclass-eager] 66.7750μs 23.2873μs 42.9419 KOps/s 43.4573 KOps/s $\color{#d91a1a}-1.19\%$
test_compile_indexing[tensor-pytree-compile] 75.6920μs 32.1440μs 31.1100 KOps/s 31.1025 KOps/s $\color{#35bf28}+0.02\%$
test_compile_indexing[tensor-pytree-eager] 78.2260μs 23.3250μs 42.8724 KOps/s 43.3952 KOps/s $\color{#d91a1a}-1.20\%$
test_compile_indexing[slice-tensordict-compile] 0.1517ms 52.8080μs 18.9365 KOps/s 19.0086 KOps/s $\color{#d91a1a}-0.38\%$
test_compile_indexing[slice-tensordict-eager] 0.3735ms 20.6492μs 48.4279 KOps/s 48.6860 KOps/s $\color{#d91a1a}-0.53\%$
test_compile_indexing[slice-tensorclass-compile] 0.1072ms 45.3038μs 22.0732 KOps/s 22.8213 KOps/s $\color{#d91a1a}-3.28\%$
test_compile_indexing[slice-tensorclass-eager] 53.6800μs 18.8960μs 52.9211 KOps/s 53.7793 KOps/s $\color{#d91a1a}-1.60\%$
test_compile_indexing[slice-pytree-compile] 96.3810μs 46.5255μs 21.4936 KOps/s 21.8498 KOps/s $\color{#d91a1a}-1.63\%$
test_compile_indexing[slice-pytree-eager] 68.6680μs 19.1143μs 52.3167 KOps/s 53.1473 KOps/s $\color{#d91a1a}-1.56\%$
test_compile_indexing[int-tensordict-compile] 0.1120ms 53.9937μs 18.5207 KOps/s 18.7632 KOps/s $\color{#d91a1a}-1.29\%$
test_compile_indexing[int-tensordict-eager] 0.9398ms 20.3641μs 49.1061 KOps/s 49.1169 KOps/s $\color{#d91a1a}-0.02\%$
test_compile_indexing[int-tensorclass-compile] 0.1343ms 46.3956μs 21.5538 KOps/s 22.0435 KOps/s $\color{#d91a1a}-2.22\%$
test_compile_indexing[int-tensorclass-eager] 65.1030μs 18.8740μs 52.9830 KOps/s 53.5426 KOps/s $\color{#d91a1a}-1.05\%$
test_compile_indexing[int-pytree-compile] 99.0360μs 46.8130μs 21.3616 KOps/s 21.7792 KOps/s $\color{#d91a1a}-1.92\%$
test_compile_indexing[int-pytree-eager] 53.2300μs 18.7768μs 53.2571 KOps/s 54.1892 KOps/s $\color{#d91a1a}-1.72\%$
test_mod_add[eager] 88.6060μs 37.3474μs 26.7757 KOps/s 26.3189 KOps/s $\color{#35bf28}+1.74\%$
test_mod_add[compile] 0.1488ms 66.0907μs 15.1307 KOps/s 15.6001 KOps/s $\color{#d91a1a}-3.01\%$
test_mod_add[compile-overhead] 0.1236ms 65.0591μs 15.3706 KOps/s 15.7040 KOps/s $\color{#d91a1a}-2.12\%$
test_mod_wrap[eager] 0.3998ms 0.2191ms 4.5635 KOps/s 4.5218 KOps/s $\color{#35bf28}+0.92\%$
test_mod_wrap[compile] 1.4634ms 0.2283ms 4.3806 KOps/s 4.4562 KOps/s $\color{#d91a1a}-1.70\%$
test_mod_wrap[compile-overhead] 0.3639ms 0.2204ms 4.5363 KOps/s 4.5385 KOps/s $\color{#d91a1a}-0.05\%$
test_mod_wrap_and_backward[eager] 12.1285ms 10.5679ms 94.6263 Ops/s 92.9685 Ops/s $\color{#35bf28}+1.78\%$
test_mod_wrap_and_backward[compile] 12.0632ms 10.4727ms 95.4860 Ops/s 92.9316 Ops/s $\color{#35bf28}+2.75\%$
test_mod_wrap_and_backward[compile-overhead] 12.9523ms 10.5412ms 94.8662 Ops/s 92.8246 Ops/s $\color{#35bf28}+2.20\%$
test_seq_add[eager] 0.2119ms 0.1218ms 8.2082 KOps/s 8.1322 KOps/s $\color{#35bf28}+0.93\%$
test_seq_add[compile] 0.1937ms 79.1617μs 12.6324 KOps/s 12.8798 KOps/s $\color{#d91a1a}-1.92\%$
test_seq_add[compile-overhead] 0.1323ms 76.3581μs 13.0962 KOps/s 13.1605 KOps/s $\color{#d91a1a}-0.49\%$
test_seq_wrap[eager] 0.7465ms 0.4524ms 2.2105 KOps/s 2.1845 KOps/s $\color{#35bf28}+1.19\%$
test_seq_wrap[compile] 0.4363ms 0.2410ms 4.1502 KOps/s 4.0831 KOps/s $\color{#35bf28}+1.64\%$
test_seq_wrap[compile-overhead] 0.4426ms 0.2410ms 4.1495 KOps/s 4.1347 KOps/s $\color{#35bf28}+0.36\%$
test_func_call_runtime[False-eager] 1.0711ms 0.5349ms 1.8695 KOps/s 1.8818 KOps/s $\color{#d91a1a}-0.65\%$
test_func_call_runtime[False-compile] 0.9254ms 0.4456ms 2.2440 KOps/s 2.2541 KOps/s $\color{#d91a1a}-0.45\%$
test_func_call_runtime[False-compile-overhead] 0.8527ms 0.4423ms 2.2607 KOps/s 2.2782 KOps/s $\color{#d91a1a}-0.77\%$
test_func_call_runtime[True-eager] 0.9123ms 0.7501ms 1.3331 KOps/s 1.3459 KOps/s $\color{#d91a1a}-0.95\%$
test_func_call_runtime[True-compile] 0.5750ms 0.4628ms 2.1605 KOps/s 2.1580 KOps/s $\color{#35bf28}+0.12\%$
test_func_call_runtime[True-compile-overhead] 0.8317ms 0.4646ms 2.1526 KOps/s 2.1528 KOps/s $-0.01\%$
test_func_call_cm_runtime[False-eager] 0.9420ms 0.5224ms 1.9143 KOps/s 1.8980 KOps/s $\color{#35bf28}+0.86\%$
test_func_call_cm_runtime[False-compile] 0.5551ms 0.4414ms 2.2655 KOps/s 2.2543 KOps/s $\color{#35bf28}+0.50\%$
test_func_call_cm_runtime[False-compile-overhead] 0.6030ms 0.4407ms 2.2689 KOps/s 2.2512 KOps/s $\color{#35bf28}+0.79\%$
test_func_call_cm_runtime[True-eager] 1.4347ms 0.8956ms 1.1166 KOps/s 1.1250 KOps/s $\color{#d91a1a}-0.75\%$
test_func_call_cm_runtime[True-compile] 0.9980ms 0.7934ms 1.2605 KOps/s 1.2684 KOps/s $\color{#d91a1a}-0.62\%$
test_func_call_cm_runtime[True-compile-overhead] 0.9495ms 0.8020ms 1.2469 KOps/s 1.2727 KOps/s $\color{#d91a1a}-2.03\%$
test_vmap_func_call_cm_runtime[eager] 2.6142ms 1.8689ms 535.0793 Ops/s 527.3117 Ops/s $\color{#35bf28}+1.47\%$
test_vmap_func_call_cm_runtime[compile] 0.7806ms 0.5381ms 1.8584 KOps/s 1.8461 KOps/s $\color{#35bf28}+0.67\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.8861ms 0.5423ms 1.8442 KOps/s 1.8527 KOps/s $\color{#d91a1a}-0.46\%$
test_distributed 0.2577ms 0.1244ms 8.0396 KOps/s 7.7491 KOps/s $\color{#35bf28}+3.75\%$
test_tdmodule 91.2710μs 28.7735μs 34.7542 KOps/s 34.8733 KOps/s $\color{#d91a1a}-0.34\%$
test_tdmodule_dispatch 94.2670μs 52.1385μs 19.1797 KOps/s 19.0083 KOps/s $\color{#35bf28}+0.90\%$
test_tdseq 46.1870μs 29.6325μs 33.7467 KOps/s 32.2536 KOps/s $\color{#35bf28}+4.63\%$
test_tdseq_dispatch 81.4030μs 55.7838μs 17.9264 KOps/s 17.2037 KOps/s $\color{#35bf28}+4.20\%$
test_instantiation_functorch 1.7376ms 1.5404ms 649.1679 Ops/s 633.8720 Ops/s $\color{#35bf28}+2.41\%$
test_exec_functorch 0.3980ms 0.1742ms 5.7390 KOps/s 5.4896 KOps/s $\color{#35bf28}+4.54\%$
test_exec_functional_call 0.3317ms 0.1698ms 5.8910 KOps/s 5.8110 KOps/s $\color{#35bf28}+1.38\%$
test_exec_td_decorator 0.4834ms 0.2299ms 4.3502 KOps/s 4.3166 KOps/s $\color{#35bf28}+0.78\%$
test_vmap_mlp_speed_decorator[True-True] 1.1650ms 0.6498ms 1.5389 KOps/s 1.5429 KOps/s $\color{#d91a1a}-0.26\%$
test_vmap_mlp_speed_decorator[True-False] 0.9855ms 0.6472ms 1.5450 KOps/s 1.5388 KOps/s $\color{#35bf28}+0.41\%$
test_vmap_mlp_speed_decorator[False-True] 0.8441ms 0.5176ms 1.9322 KOps/s 1.9174 KOps/s $\color{#35bf28}+0.77\%$
test_vmap_mlp_speed_decorator[False-False] 0.8356ms 0.5189ms 1.9270 KOps/s 1.8965 KOps/s $\color{#35bf28}+1.61\%$
test_to_module_speed[True] 1.8518ms 1.3265ms 753.8595 Ops/s 747.4724 Ops/s $\color{#35bf28}+0.85\%$
test_to_module_speed[False] 2.1315ms 1.3155ms 760.1704 Ops/s 767.5310 Ops/s $\color{#d91a1a}-0.96\%$
test_tc_init 97.4330μs 50.0532μs 19.9788 KOps/s 20.5039 KOps/s $\color{#d91a1a}-2.56\%$
test_tc_init_nested 0.1839ms 99.7387μs 10.0262 KOps/s 10.1571 KOps/s $\color{#d91a1a}-1.29\%$
test_tc_first_layer_tensor 19.7670μs 1.5196μs 658.0749 KOps/s 656.8847 KOps/s $\color{#35bf28}+0.18\%$
test_tc_first_layer_nontensor 18.9960μs 4.7556μs 210.2803 KOps/s 208.5129 KOps/s $\color{#35bf28}+0.85\%$
test_tc_second_layer_tensor 32.0800μs 2.9062μs 344.0928 KOps/s 350.1679 KOps/s $\color{#d91a1a}-1.73\%$
test_tc_second_layer_nontensor 0.1046ms 6.1528μs 162.5273 KOps/s 158.2746 KOps/s $\color{#35bf28}+2.69\%$
test_unbind 0.2239s 12.9393ms 77.2839 Ops/s 76.4886 Ops/s $\color{#35bf28}+1.04\%$
test_full_like 9.1917ms 8.1671ms 122.4418 Ops/s 130.4145 Ops/s $\textbf{\color{#d91a1a}-6.11\%}$
test_zeros_like 5.0078ms 2.6395ms 378.8626 Ops/s 225.2808 Ops/s $\textbf{\color{#35bf28}+68.17\%}$
test_ones_like 4.6174ms 3.0909ms 323.5316 Ops/s 306.8165 Ops/s $\textbf{\color{#35bf28}+5.45\%}$
test_clone 7.6575ms 6.5349ms 153.0251 Ops/s 204.6715 Ops/s $\textbf{\color{#d91a1a}-25.23\%}$
test_squeeze 65.2720μs 12.6001μs 79.3643 KOps/s 79.7371 KOps/s $\color{#d91a1a}-0.47\%$
test_unsqueeze 0.3373ms 93.3148μs 10.7164 KOps/s 10.6398 KOps/s $\color{#35bf28}+0.72\%$
test_split 0.3115ms 0.1910ms 5.2346 KOps/s 5.1777 KOps/s $\color{#35bf28}+1.10\%$
test_permute 0.3311ms 0.1983ms 5.0434 KOps/s 4.9848 KOps/s $\color{#35bf28}+1.18\%$
test_stack 27.1478ms 23.7409ms 42.1214 Ops/s 39.2739 Ops/s $\textbf{\color{#35bf28}+7.25\%}$
test_cat 33.1466ms 23.9652ms 41.7271 Ops/s 39.0569 Ops/s $\textbf{\color{#35bf28}+6.84\%}$

Copy link

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}16$. Worsened: $\large\color{#d91a1a}24$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 34.8110μs 13.5406μs 73.8521 KOps/s 79.0865 KOps/s $\textbf{\color{#d91a1a}-6.62\%}$
test_plain_set_stack_nested 53.0730μs 13.5991μs 73.5343 KOps/s 78.8174 KOps/s $\textbf{\color{#d91a1a}-6.70\%}$
test_plain_set_nested_inplace 40.2330μs 14.6567μs 68.2282 KOps/s 72.9424 KOps/s $\textbf{\color{#d91a1a}-6.46\%}$
test_plain_set_stack_nested_inplace 53.6330μs 14.5704μs 68.6324 KOps/s 73.7387 KOps/s $\textbf{\color{#d91a1a}-6.92\%}$
test_items 28.2520μs 2.8663μs 348.8797 KOps/s 343.8510 KOps/s $\color{#35bf28}+1.46\%$
test_items_nested 0.4353ms 0.3857ms 2.5928 KOps/s 2.5715 KOps/s $\color{#35bf28}+0.83\%$
test_items_nested_locked 0.4304ms 0.3851ms 2.5967 KOps/s 2.5723 KOps/s $\color{#35bf28}+0.95\%$
test_items_nested_leaf 87.2450μs 60.5822μs 16.5065 KOps/s 16.5686 KOps/s $\color{#d91a1a}-0.38\%$
test_items_stack_nested 0.4507ms 0.3837ms 2.6063 KOps/s 2.6160 KOps/s $\color{#d91a1a}-0.37\%$
test_items_stack_nested_leaf 98.1750μs 60.8444μs 16.4354 KOps/s 16.5862 KOps/s $\color{#d91a1a}-0.91\%$
test_items_stack_nested_locked 0.4389ms 0.3834ms 2.6079 KOps/s 2.5886 KOps/s $\color{#35bf28}+0.75\%$
test_keys 31.4120μs 3.4261μs 291.8798 KOps/s 292.6072 KOps/s $\color{#d91a1a}-0.25\%$
test_keys_nested 0.1234ms 87.2273μs 11.4643 KOps/s 11.3949 KOps/s $\color{#35bf28}+0.61\%$
test_keys_nested_locked 0.7392ms 93.2496μs 10.7239 KOps/s 10.6863 KOps/s $\color{#35bf28}+0.35\%$
test_keys_nested_leaf 0.1192ms 78.5763μs 12.7265 KOps/s 12.7090 KOps/s $\color{#35bf28}+0.14\%$
test_keys_stack_nested 0.1226ms 87.1790μs 11.4706 KOps/s 11.4373 KOps/s $\color{#35bf28}+0.29\%$
test_keys_stack_nested_leaf 0.1139ms 78.2428μs 12.7807 KOps/s 12.7205 KOps/s $\color{#35bf28}+0.47\%$
test_keys_stack_nested_locked 0.1249ms 92.8696μs 10.7678 KOps/s 10.6616 KOps/s $\color{#35bf28}+1.00\%$
test_values 5.7453μs 0.8570μs 1.1669 MOps/s 1.1693 MOps/s $\color{#d91a1a}-0.20\%$
test_values_nested 60.9230μs 36.9072μs 27.0950 KOps/s 26.9716 KOps/s $\color{#35bf28}+0.46\%$
test_values_nested_locked 64.2630μs 38.7052μs 25.8363 KOps/s 25.8865 KOps/s $\color{#d91a1a}-0.19\%$
test_values_nested_leaf 77.8650μs 42.0307μs 23.7921 KOps/s 23.8799 KOps/s $\color{#d91a1a}-0.37\%$
test_values_stack_nested 76.6240μs 37.1673μs 26.9054 KOps/s 26.8633 KOps/s $\color{#35bf28}+0.16\%$
test_values_stack_nested_leaf 72.3140μs 42.0129μs 23.8022 KOps/s 23.7856 KOps/s $\color{#35bf28}+0.07\%$
test_values_stack_nested_locked 63.0340μs 38.8104μs 25.7663 KOps/s 25.9317 KOps/s $\color{#d91a1a}-0.64\%$
test_membership 2.5866μs 0.4976μs 2.0096 MOps/s 1.6843 MOps/s $\textbf{\color{#35bf28}+19.31\%}$
test_membership_nested 14.5605μs 2.0988μs 476.4730 KOps/s 476.2476 KOps/s $\color{#35bf28}+0.05\%$
test_membership_nested_leaf 16.1655μs 2.0977μs 476.7225 KOps/s 483.5888 KOps/s $\color{#d91a1a}-1.42\%$
test_membership_stacked_nested 42.5020μs 2.1807μs 458.5655 KOps/s 467.1513 KOps/s $\color{#d91a1a}-1.84\%$
test_membership_stacked_nested_leaf 29.1820μs 2.1765μs 459.4427 KOps/s 475.1123 KOps/s $\color{#d91a1a}-3.30\%$
test_membership_nested_last 37.7920μs 3.0717μs 325.5473 KOps/s 321.6525 KOps/s $\color{#35bf28}+1.21\%$
test_membership_nested_leaf_last 28.7720μs 3.0855μs 324.1005 KOps/s 320.5106 KOps/s $\color{#35bf28}+1.12\%$
test_membership_stacked_nested_last 34.2720μs 3.1084μs 321.7071 KOps/s 324.5700 KOps/s $\color{#d91a1a}-0.88\%$
test_membership_stacked_nested_leaf_last 25.6520μs 3.1110μs 321.4351 KOps/s 324.1604 KOps/s $\color{#d91a1a}-0.84\%$
test_nested_getleaf 60.9230μs 6.3081μs 158.5261 KOps/s 160.1775 KOps/s $\color{#d91a1a}-1.03\%$
test_nested_get 44.5220μs 6.0102μs 166.3840 KOps/s 166.0512 KOps/s $\color{#35bf28}+0.20\%$
test_stacked_getleaf 37.9420μs 6.3686μs 157.0195 KOps/s 161.6873 KOps/s $\color{#d91a1a}-2.89\%$
test_stacked_get 37.4720μs 5.9979μs 166.7251 KOps/s 172.0951 KOps/s $\color{#d91a1a}-3.12\%$
test_nested_getitemleaf 32.7820μs 6.4805μs 154.3097 KOps/s 153.8696 KOps/s $\color{#35bf28}+0.29\%$
test_nested_getitem 28.6210μs 6.2059μs 161.1357 KOps/s 161.9520 KOps/s $\color{#d91a1a}-0.50\%$
test_stacked_getitemleaf 33.2120μs 6.4820μs 154.2723 KOps/s 155.6120 KOps/s $\color{#d91a1a}-0.86\%$
test_stacked_getitem 27.0820μs 6.0808μs 164.4514 KOps/s 165.4990 KOps/s $\color{#d91a1a}-0.63\%$
test_lock_nested 8.7670ms 0.3433ms 2.9132 KOps/s 2.9853 KOps/s $\color{#d91a1a}-2.42\%$
test_lock_stack_nested 0.3966ms 0.3417ms 2.9269 KOps/s 2.9337 KOps/s $\color{#d91a1a}-0.23\%$
test_unlock_nested 0.3431ms 0.2823ms 3.5418 KOps/s 3.6027 KOps/s $\color{#d91a1a}-1.69\%$
test_unlock_stack_nested 0.3356ms 0.2825ms 3.5403 KOps/s 3.5818 KOps/s $\color{#d91a1a}-1.16\%$
test_flatten_speed 0.1094ms 77.5894μs 12.8884 KOps/s 12.9578 KOps/s $\color{#d91a1a}-0.54\%$
test_unflatten_speed 0.3759ms 0.3231ms 3.0953 KOps/s 3.0621 KOps/s $\color{#35bf28}+1.08\%$
test_common_ops 0.7647ms 0.6401ms 1.5623 KOps/s 1.6484 KOps/s $\textbf{\color{#d91a1a}-5.22\%}$
test_creation 72.0140μs 1.7287μs 578.4709 KOps/s 584.8597 KOps/s $\color{#d91a1a}-1.09\%$
test_creation_empty 38.1220μs 10.3154μs 96.9429 KOps/s 119.5345 KOps/s $\textbf{\color{#d91a1a}-18.90\%}$
test_creation_nested_1 37.3220μs 11.9654μs 83.5745 KOps/s 99.6293 KOps/s $\textbf{\color{#d91a1a}-16.11\%}$
test_creation_nested_2 36.8920μs 14.6952μs 68.0494 KOps/s 78.1235 KOps/s $\textbf{\color{#d91a1a}-12.90\%}$
test_clone 42.6320μs 10.3911μs 96.2364 KOps/s 96.7032 KOps/s $\color{#d91a1a}-0.48\%$
test_getitem[int] 1.1809ms 10.3322μs 96.7847 KOps/s 94.5644 KOps/s $\color{#35bf28}+2.35\%$
test_getitem[slice_int] 0.1148ms 20.3745μs 49.0809 KOps/s 47.7842 KOps/s $\color{#35bf28}+2.71\%$
test_getitem[range] 0.1284ms 37.4150μs 26.7273 KOps/s 25.2869 KOps/s $\textbf{\color{#35bf28}+5.70\%}$
test_getitem[tuple] 0.1069ms 17.3020μs 57.7969 KOps/s 55.4257 KOps/s $\color{#35bf28}+4.28\%$
test_getitem[list] 0.1315ms 32.9934μs 30.3091 KOps/s 30.0520 KOps/s $\color{#35bf28}+0.86\%$
test_setitem_dim[int] 38.0420μs 19.0305μs 52.5472 KOps/s 52.3028 KOps/s $\color{#35bf28}+0.47\%$
test_setitem_dim[slice_int] 63.1830μs 38.5877μs 25.9150 KOps/s 25.0532 KOps/s $\color{#35bf28}+3.44\%$
test_setitem_dim[range] 79.8140μs 52.1061μs 19.1916 KOps/s 18.0080 KOps/s $\textbf{\color{#35bf28}+6.57\%}$
test_setitem_dim[tuple] 42.9020μs 31.8588μs 31.3885 KOps/s 28.6588 KOps/s $\textbf{\color{#35bf28}+9.52\%}$
test_setitem 0.1362ms 16.0192μs 62.4251 KOps/s 67.3543 KOps/s $\textbf{\color{#d91a1a}-7.32\%}$
test_set 46.8730μs 15.5989μs 64.1072 KOps/s 69.8080 KOps/s $\textbf{\color{#d91a1a}-8.17\%}$
test_set_shared 0.5153ms 0.1567ms 6.3830 KOps/s 6.3081 KOps/s $\color{#35bf28}+1.19\%$
test_update 0.3397ms 19.4174μs 51.5003 KOps/s 57.2876 KOps/s $\textbf{\color{#d91a1a}-10.10\%}$
test_update_nested 58.7130μs 24.7966μs 40.3281 KOps/s 43.8481 KOps/s $\textbf{\color{#d91a1a}-8.03\%}$
test_update__nested 0.5056ms 24.7187μs 40.4552 KOps/s 39.3150 KOps/s $\color{#35bf28}+2.90\%$
test_set_nested 50.1830μs 16.8132μs 59.4770 KOps/s 64.2760 KOps/s $\textbf{\color{#d91a1a}-7.47\%}$
test_set_nested_new 54.2530μs 19.8188μs 50.4571 KOps/s 56.8035 KOps/s $\textbf{\color{#d91a1a}-11.17\%}$
test_select 58.2740μs 30.0298μs 33.3003 KOps/s 34.3073 KOps/s $\color{#d91a1a}-2.94\%$
test_select_nested 85.4050μs 43.2348μs 23.1295 KOps/s 23.1323 KOps/s $\color{#d91a1a}-0.01\%$
test_exclude_nested 0.1008ms 63.1311μs 15.8400 KOps/s 15.7843 KOps/s $\color{#35bf28}+0.35\%$
test_empty[True] 0.3545ms 0.3097ms 3.2286 KOps/s 3.2330 KOps/s $\color{#d91a1a}-0.13\%$
test_empty[False] 3.8592μs 0.8267μs 1.2097 MOps/s 1.2205 MOps/s $\color{#d91a1a}-0.88\%$
test_to 87.8950μs 55.7431μs 17.9394 KOps/s 16.2473 KOps/s $\textbf{\color{#35bf28}+10.41\%}$
test_to_nonblocking 97.9250μs 48.8256μs 20.4811 KOps/s 21.2703 KOps/s $\color{#d91a1a}-3.71\%$
test_unbind_speed 0.2664ms 0.2392ms 4.1798 KOps/s 4.2296 KOps/s $\color{#d91a1a}-1.18\%$
test_unbind_speed_stack0 0.3005ms 0.2338ms 4.2777 KOps/s 4.2814 KOps/s $\color{#d91a1a}-0.09\%$
test_unbind_speed_stack1 93.2221ms 0.7379ms 1.3552 KOps/s 1.3734 KOps/s $\color{#d91a1a}-1.32\%$
test_split 94.2489ms 1.5805ms 632.7112 Ops/s 630.0358 Ops/s $\color{#35bf28}+0.42\%$
test_chunk 96.1053ms 1.5912ms 628.4537 Ops/s 628.9411 Ops/s $\color{#d91a1a}-0.08\%$
test_consolidate[False-None] 2.8367ms 2.7131ms 368.5861 Ops/s 331.5470 Ops/s $\textbf{\color{#35bf28}+11.17\%}$
test_consolidate[default-None] 1.9236ms 1.7410ms 574.3747 Ops/s 577.6204 Ops/s $\color{#d91a1a}-0.56\%$
test_consolidate[reduce-overhead-None] 1.8566ms 1.7843ms 560.4576 Ops/s 568.1061 Ops/s $\color{#d91a1a}-1.35\%$
test_consolidate_njt[False-None] 0.2980s 8.5523ms 116.9280 Ops/s 155.5041 Ops/s $\textbf{\color{#d91a1a}-24.81\%}$
test_to[False-False-None] 1.7981ms 1.7366ms 575.8214 Ops/s 580.3141 Ops/s $\color{#d91a1a}-0.77\%$
test_to[True-False-None] 1.5363ms 1.3644ms 732.9225 Ops/s 733.8494 Ops/s $\color{#d91a1a}-0.13\%$
test_to[within-False-None] 4.5732ms 4.1664ms 240.0143 Ops/s 171.0588 Ops/s $\textbf{\color{#35bf28}+40.31\%}$
test_to[True-default-None] 5.4782ms 5.1442ms 194.3935 Ops/s 193.9438 Ops/s $\color{#35bf28}+0.23\%$
test_to_njt[False-False-None] 7.0966ms 6.9371ms 144.1530 Ops/s 143.4420 Ops/s $\color{#35bf28}+0.50\%$
test_to_njt[True-False-None] 5.6497ms 5.4271ms 184.2608 Ops/s 181.2478 Ops/s $\color{#35bf28}+1.66\%$
test_to_njt[within-False-None] 12.4571ms 12.0102ms 83.2624 Ops/s 81.6822 Ops/s $\color{#35bf28}+1.93\%$
test_creation[device0] 0.4548ms 84.7465μs 11.7999 KOps/s 12.4594 KOps/s $\textbf{\color{#d91a1a}-5.29\%}$
test_creation_from_tensor 0.5105ms 88.1628μs 11.3427 KOps/s 11.5813 KOps/s $\color{#d91a1a}-2.06\%$
test_add_one[memmap_tensor0] 0.4194ms 6.8281μs 146.4531 KOps/s 147.5917 KOps/s $\color{#d91a1a}-0.77\%$
test_contiguous[memmap_tensor0] 1.7001μs 0.4349μs 2.2992 MOps/s 2.2990 MOps/s $\color{#35bf28}+0.01\%$
test_stack[memmap_tensor0] 21.4120μs 4.3391μs 230.4640 KOps/s 228.4929 KOps/s $\color{#35bf28}+0.86\%$
test_memmaptd_index 0.4517ms 0.2385ms 4.1927 KOps/s 4.1355 KOps/s $\color{#35bf28}+1.38\%$
test_memmaptd_index_astensor 0.4475ms 0.3023ms 3.3079 KOps/s 3.3426 KOps/s $\color{#d91a1a}-1.04\%$
test_memmaptd_index_op 0.7532ms 0.6034ms 1.6573 KOps/s 1.7379 KOps/s $\color{#d91a1a}-4.64\%$
test_serialize_model 0.1317s 0.1310s 7.6313 Ops/s 7.6170 Ops/s $\color{#35bf28}+0.19\%$
test_serialize_model_pickle 1.3522s 1.2156s 0.8226 Ops/s 0.8218 Ops/s $\color{#35bf28}+0.10\%$
test_serialize_weights 0.1316s 0.1305s 7.6630 Ops/s 7.6754 Ops/s $\color{#d91a1a}-0.16\%$
test_serialize_weights_returnearly 0.3340s 55.8421ms 17.9076 Ops/s 14.9459 Ops/s $\textbf{\color{#35bf28}+19.82\%}$
test_serialize_weights_pickle 1.3753s 1.2158s 0.8225 Ops/s 0.8427 Ops/s $\color{#d91a1a}-2.39\%$
test_reshape_pytree 53.4330μs 21.9842μs 45.4872 KOps/s 45.0871 KOps/s $\color{#35bf28}+0.89\%$
test_reshape_td 51.7130μs 26.4532μs 37.8026 KOps/s 37.1707 KOps/s $\color{#35bf28}+1.70\%$
test_view_pytree 46.1330μs 21.8724μs 45.7196 KOps/s 45.3615 KOps/s $\color{#35bf28}+0.79\%$
test_view_td 60.4930μs 31.3964μs 31.8508 KOps/s 30.5846 KOps/s $\color{#35bf28}+4.14\%$
test_unbind_pytree 58.9430μs 27.5533μs 36.2933 KOps/s 36.3331 KOps/s $\color{#d91a1a}-0.11\%$
test_unbind_td 0.6184ms 35.7525μs 27.9701 KOps/s 27.5250 KOps/s $\color{#35bf28}+1.62\%$
test_split_pytree 70.1240μs 29.4163μs 33.9948 KOps/s 34.3054 KOps/s $\color{#d91a1a}-0.91\%$
test_split_td 0.7861ms 37.7089μs 26.5189 KOps/s 25.8152 KOps/s $\color{#35bf28}+2.73\%$
test_add_pytree 70.3340μs 35.2855μs 28.3402 KOps/s 29.0191 KOps/s $\color{#d91a1a}-2.34\%$
test_add_td 89.0840μs 49.7343μs 20.1068 KOps/s 20.8958 KOps/s $\color{#d91a1a}-3.78\%$
test_compile_add_one_nested[tensordict-compile] 0.1899ms 0.1217ms 8.2158 KOps/s 7.9822 KOps/s $\color{#35bf28}+2.93\%$
test_compile_add_one_nested[tensordict-eager] 0.5304ms 0.1313ms 7.6175 KOps/s 7.6096 KOps/s $\color{#35bf28}+0.10\%$
test_compile_add_one_nested[pytree-compile] 0.1894ms 94.6527μs 10.5649 KOps/s 10.3062 KOps/s $\color{#35bf28}+2.51\%$
test_compile_add_one_nested[pytree-eager] 1.4381ms 0.1495ms 6.6900 KOps/s 6.7336 KOps/s $\color{#d91a1a}-0.65\%$
test_compile_copy_nested[tensordict-compile] 0.4507ms 23.2684μs 42.9767 KOps/s 37.8378 KOps/s $\textbf{\color{#35bf28}+13.58\%}$
test_compile_copy_nested[tensordict-eager] 0.4333ms 30.0533μs 33.2743 KOps/s 33.5914 KOps/s $\color{#d91a1a}-0.94\%$
test_compile_copy_nested[pytree-compile] 0.4980ms 63.2236μs 15.8169 KOps/s 15.6000 KOps/s $\color{#35bf28}+1.39\%$
test_compile_copy_nested[pytree-eager] 0.4586ms 48.9435μs 20.4317 KOps/s 20.5399 KOps/s $\color{#d91a1a}-0.53\%$
test_compile_add_one_flat[tensordict-compile] 0.1891ms 0.1415ms 7.0695 KOps/s 7.1325 KOps/s $\color{#d91a1a}-0.88\%$
test_compile_add_one_flat[tensordict-eager] 0.6660ms 0.2171ms 4.6059 KOps/s 4.6292 KOps/s $\color{#d91a1a}-0.50\%$
test_compile_add_one_flat[tensorclass-compile] 0.1446ms 98.8237μs 10.1190 KOps/s 10.3720 KOps/s $\color{#d91a1a}-2.44\%$
test_compile_add_one_flat[tensorclass-eager] 0.4702ms 56.7620μs 17.6174 KOps/s 18.0396 KOps/s $\color{#d91a1a}-2.34\%$
test_compile_add_one_flat[pytree-compile] 0.2862ms 0.1353ms 7.3901 KOps/s 7.2869 KOps/s $\color{#35bf28}+1.42\%$
test_compile_add_one_flat[pytree-eager] 0.8898ms 0.4839ms 2.0665 KOps/s 2.0687 KOps/s $\color{#d91a1a}-0.11\%$
test_compile_add_self_flat[tensordict-eager] 0.6600ms 0.2606ms 3.8377 KOps/s 3.8479 KOps/s $\color{#d91a1a}-0.26\%$
test_compile_add_self_flat[tensordict-compile] 0.2038ms 0.1420ms 7.0407 KOps/s 7.0812 KOps/s $\color{#d91a1a}-0.57\%$
test_compile_add_self_flat[tensorclass-eager] 0.4765ms 66.5284μs 15.0312 KOps/s 14.4500 KOps/s $\color{#35bf28}+4.02\%$
test_compile_add_self_flat[tensorclass-compile] 0.1455ms 98.2833μs 10.1747 KOps/s 10.2870 KOps/s $\color{#d91a1a}-1.09\%$
test_compile_add_self_flat[pytree-eager] 0.8203ms 0.4123ms 2.4254 KOps/s 2.4617 KOps/s $\color{#d91a1a}-1.47\%$
test_compile_add_self_flat[pytree-compile] 0.1921ms 0.1327ms 7.5381 KOps/s 7.4238 KOps/s $\color{#35bf28}+1.54\%$
test_compile_copy_flat[tensordict-compile] 79.1550μs 18.4556μs 54.1840 KOps/s 57.5890 KOps/s $\textbf{\color{#d91a1a}-5.91\%}$
test_compile_copy_flat[tensordict-eager] 0.4156ms 32.4838μs 30.7846 KOps/s 31.6423 KOps/s $\color{#d91a1a}-2.71\%$
test_compile_copy_flat[pytree-compile] 0.4638ms 69.2695μs 14.4364 KOps/s 14.5035 KOps/s $\color{#d91a1a}-0.46\%$
test_compile_copy_flat[pytree-eager] 0.4259ms 51.5018μs 19.4168 KOps/s 19.3081 KOps/s $\color{#35bf28}+0.56\%$
test_compile_assign_and_add[tensordict-compile] 1.6814ms 0.4006ms 2.4966 KOps/s 2.1747 KOps/s $\textbf{\color{#35bf28}+14.80\%}$
test_compile_assign_and_add[tensordict-eager] 2.8299ms 2.6394ms 378.8736 Ops/s 374.8231 Ops/s $\color{#35bf28}+1.08\%$
test_compile_assign_and_add[pytree-compile] 1.6382ms 0.4410ms 2.2676 KOps/s 2.2449 KOps/s $\color{#35bf28}+1.01\%$
test_compile_assign_and_add[pytree-eager] 2.8058ms 2.6472ms 377.7521 Ops/s 360.2391 Ops/s $\color{#35bf28}+4.86\%$
test_compile_indexing[tensor-tensordict-compile] 0.2451ms 0.1155ms 8.6544 KOps/s 8.4249 KOps/s $\color{#35bf28}+2.72\%$
test_compile_indexing[tensor-tensordict-eager] 0.5811ms 78.8332μs 12.6850 KOps/s 12.3728 KOps/s $\color{#35bf28}+2.52\%$
test_compile_indexing[tensor-tensorclass-compile] 0.5192ms 0.1090ms 9.1778 KOps/s 9.2786 KOps/s $\color{#d91a1a}-1.09\%$
test_compile_indexing[tensor-tensorclass-eager] 0.1880ms 70.5501μs 14.1743 KOps/s 13.5368 KOps/s $\color{#35bf28}+4.71\%$
test_compile_indexing[tensor-pytree-compile] 0.1710ms 0.1154ms 8.6631 KOps/s 8.7189 KOps/s $\color{#d91a1a}-0.64\%$
test_compile_indexing[tensor-pytree-eager] 0.2013ms 72.6224μs 13.7699 KOps/s 13.5603 KOps/s $\color{#35bf28}+1.55\%$
test_compile_indexing[slice-tensordict-compile] 0.2395ms 0.1007ms 9.9280 KOps/s 9.9136 KOps/s $\color{#35bf28}+0.15\%$
test_compile_indexing[slice-tensordict-eager] 0.1453ms 16.9178μs 59.1092 KOps/s 57.7900 KOps/s $\color{#35bf28}+2.28\%$
test_compile_indexing[slice-tensorclass-compile] 0.1541ms 95.4692μs 10.4746 KOps/s 10.4492 KOps/s $\color{#35bf28}+0.24\%$
test_compile_indexing[slice-tensorclass-eager] 50.7030μs 15.4058μs 64.9106 KOps/s 60.6710 KOps/s $\textbf{\color{#35bf28}+6.99\%}$
test_compile_indexing[slice-pytree-compile] 0.1626ms 99.4483μs 10.0555 KOps/s 10.3303 KOps/s $\color{#d91a1a}-2.66\%$
test_compile_indexing[slice-pytree-eager] 86.7450μs 15.2413μs 65.6113 KOps/s 64.8205 KOps/s $\color{#35bf28}+1.22\%$
test_compile_indexing[int-tensordict-compile] 0.1739ms 0.1023ms 9.7706 KOps/s 9.8039 KOps/s $\color{#d91a1a}-0.34\%$
test_compile_indexing[int-tensordict-eager] 0.6180ms 17.1001μs 58.4791 KOps/s 57.4300 KOps/s $\color{#35bf28}+1.83\%$
test_compile_indexing[int-tensorclass-compile] 0.1519ms 96.2268μs 10.3921 KOps/s 10.3406 KOps/s $\color{#35bf28}+0.50\%$
test_compile_indexing[int-tensorclass-eager] 52.8530μs 15.3776μs 65.0295 KOps/s 64.8755 KOps/s $\color{#35bf28}+0.24\%$
test_compile_indexing[int-pytree-compile] 0.1614ms 0.1009ms 9.9132 KOps/s 10.2796 KOps/s $\color{#d91a1a}-3.56\%$
test_compile_indexing[int-pytree-eager] 36.0620μs 15.7473μs 63.5031 KOps/s 65.3527 KOps/s $\color{#d91a1a}-2.83\%$
test_mod_add[eager] 0.1437ms 41.5936μs 24.0421 KOps/s 24.3270 KOps/s $\color{#d91a1a}-1.17\%$
test_mod_add[compile] 0.3632ms 80.7311μs 12.3868 KOps/s 12.1634 KOps/s $\color{#35bf28}+1.84\%$
test_mod_add[compile-overhead] 0.3319ms 0.1701ms 5.8804 KOps/s 5.5994 KOps/s $\textbf{\color{#35bf28}+5.02\%}$
test_mod_wrap[eager] 0.3333ms 0.2526ms 3.9584 KOps/s 3.6704 KOps/s $\textbf{\color{#35bf28}+7.85\%}$
test_mod_wrap[compile] 0.3725ms 0.2911ms 3.4349 KOps/s 3.4326 KOps/s $\color{#35bf28}+0.07\%$
test_mod_wrap[compile-overhead] 7.6721ms 3.8896ms 257.0929 Ops/s 266.7582 Ops/s $\color{#d91a1a}-3.62\%$
test_mod_wrap_and_backward[eager] 1.4967ms 1.3898ms 719.5322 Ops/s 678.5885 Ops/s $\textbf{\color{#35bf28}+6.03\%}$
test_mod_wrap_and_backward[compile] 1.4135ms 1.2783ms 782.2745 Ops/s 716.5147 Ops/s $\textbf{\color{#35bf28}+9.18\%}$
test_mod_wrap_and_backward[compile-overhead] 1.3775ms 0.9336ms 1.0712 KOps/s 948.1243 Ops/s $\textbf{\color{#35bf28}+12.98\%}$
test_seq_add[eager] 0.1648ms 0.1183ms 8.4527 KOps/s 8.2651 KOps/s $\color{#35bf28}+2.27\%$
test_seq_add[compile] 0.1321ms 89.2883μs 11.1997 KOps/s 11.2810 KOps/s $\color{#d91a1a}-0.72\%$
test_seq_add[compile-overhead] 0.1730ms 0.1301ms 7.6884 KOps/s 7.7127 KOps/s $\color{#d91a1a}-0.31\%$
test_seq_wrap[eager] 0.5045ms 0.4358ms 2.2945 KOps/s 2.2674 KOps/s $\color{#35bf28}+1.19\%$
test_seq_wrap[compile] 0.3953ms 0.3052ms 3.2770 KOps/s 3.2143 KOps/s $\color{#35bf28}+1.95\%$
test_seq_wrap[compile-overhead] 0.2760ms 0.2272ms 4.4024 KOps/s 4.3503 KOps/s $\color{#35bf28}+1.20\%$
test_func_call_runtime[False-eager] 0.8953ms 0.7952ms 1.2575 KOps/s 1.3003 KOps/s $\color{#d91a1a}-3.29\%$
test_func_call_runtime[False-compile] 0.9065ms 0.7508ms 1.3319 KOps/s 1.3097 KOps/s $\color{#35bf28}+1.69\%$
test_func_call_runtime[False-compile-overhead] 0.4406ms 0.3654ms 2.7366 KOps/s 2.7264 KOps/s $\color{#35bf28}+0.37\%$
test_func_call_runtime[True-eager] 1.0637ms 0.9126ms 1.0957 KOps/s 1.0630 KOps/s $\color{#35bf28}+3.08\%$
test_func_call_runtime[True-compile] 0.8733ms 0.7782ms 1.2850 KOps/s 1.2806 KOps/s $\color{#35bf28}+0.34\%$
test_func_call_runtime[True-compile-overhead] 0.4781ms 0.3874ms 2.5815 KOps/s 2.5625 KOps/s $\color{#35bf28}+0.74\%$
test_func_call_cm_runtime[False-eager] 0.8591ms 0.7387ms 1.3538 KOps/s 1.3149 KOps/s $\color{#35bf28}+2.96\%$
test_func_call_cm_runtime[False-compile] 0.8589ms 0.7571ms 1.3209 KOps/s 1.3160 KOps/s $\color{#35bf28}+0.37\%$
test_func_call_cm_runtime[False-compile-overhead] 0.4294ms 0.3657ms 2.7344 KOps/s 2.7076 KOps/s $\color{#35bf28}+0.99\%$
test_func_call_cm_runtime[True-eager] 1.1067ms 1.0072ms 992.8429 Ops/s 957.7774 Ops/s $\color{#35bf28}+3.66\%$
test_func_call_cm_runtime[True-compile] 1.1372ms 1.0054ms 994.5864 Ops/s 978.6435 Ops/s $\color{#35bf28}+1.63\%$
test_func_call_cm_runtime[True-compile-overhead] 1.0905ms 1.0002ms 999.7792 Ops/s 964.3644 Ops/s $\color{#35bf28}+3.67\%$
test_vmap_func_call_cm_runtime[eager] 2.5035ms 2.1023ms 475.6739 Ops/s 465.2727 Ops/s $\color{#35bf28}+2.24\%$
test_vmap_func_call_cm_runtime[compile] 0.9288ms 0.8134ms 1.2293 KOps/s 1.2022 KOps/s $\color{#35bf28}+2.25\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.4714ms 0.4171ms 2.3975 KOps/s 2.3712 KOps/s $\color{#35bf28}+1.11\%$
test_distributed 0.6737ms 0.1624ms 6.1566 KOps/s 8.6665 KOps/s $\textbf{\color{#d91a1a}-28.96\%}$
test_tdmodule 0.1566ms 22.1867μs 45.0720 KOps/s 47.8395 KOps/s $\textbf{\color{#d91a1a}-5.78\%}$
test_tdmodule_dispatch 65.6830μs 38.0073μs 26.3108 KOps/s 27.1378 KOps/s $\color{#d91a1a}-3.05\%$
test_tdseq 68.6240μs 21.9819μs 45.4919 KOps/s 47.9057 KOps/s $\textbf{\color{#d91a1a}-5.04\%}$
test_tdseq_dispatch 71.2040μs 41.1096μs 24.3252 KOps/s 25.7817 KOps/s $\textbf{\color{#d91a1a}-5.65\%}$
test_instantiation_functorch 1.6262ms 1.5381ms 650.1725 Ops/s 648.6645 Ops/s $\color{#35bf28}+0.23\%$
test_exec_functorch 0.1845ms 0.1425ms 7.0175 KOps/s 6.9399 KOps/s $\color{#35bf28}+1.12\%$
test_exec_functional_call 0.1767ms 0.1359ms 7.3572 KOps/s 7.2918 KOps/s $\color{#35bf28}+0.90\%$
test_exec_td_decorator 0.3700ms 0.1860ms 5.3757 KOps/s 5.3272 KOps/s $\color{#35bf28}+0.91\%$
test_vmap_mlp_speed_decorator[True-True] 0.8203ms 0.6952ms 1.4385 KOps/s 1.4372 KOps/s $\color{#35bf28}+0.09\%$
test_vmap_mlp_speed_decorator[True-False] 0.8452ms 0.6956ms 1.4376 KOps/s 1.4353 KOps/s $\color{#35bf28}+0.16\%$
test_vmap_mlp_speed_decorator[False-True] 0.7296ms 0.6000ms 1.6668 KOps/s 1.6584 KOps/s $\color{#35bf28}+0.50\%$
test_vmap_mlp_speed_decorator[False-False] 0.7293ms 0.6006ms 1.6650 KOps/s 1.6591 KOps/s $\color{#35bf28}+0.36\%$
test_vmap_transformer_speed_decorator[True-True] 19.6200ms 19.4279ms 51.4723 Ops/s 51.5031 Ops/s $\color{#d91a1a}-0.06\%$
test_vmap_transformer_speed_decorator[True-False] 19.7516ms 19.4465ms 51.4232 Ops/s 51.5155 Ops/s $\color{#d91a1a}-0.18\%$
test_vmap_transformer_speed_decorator[False-True] 19.4395ms 19.2602ms 51.9207 Ops/s 51.7859 Ops/s $\color{#35bf28}+0.26\%$
test_vmap_transformer_speed_decorator[False-False] 19.4052ms 19.2845ms 51.8552 Ops/s 51.9762 Ops/s $\color{#d91a1a}-0.23\%$
test_to_module_speed[True] 1.5321ms 0.9674ms 1.0337 KOps/s 1.0276 KOps/s $\color{#35bf28}+0.60\%$
test_to_module_speed[False] 1.0141ms 0.9504ms 1.0521 KOps/s 1.0397 KOps/s $\color{#35bf28}+1.19\%$
test_tc_init 66.5840μs 38.6766μs 25.8554 KOps/s 27.6216 KOps/s $\textbf{\color{#d91a1a}-6.39\%}$
test_tc_init_nested 0.1132ms 78.0859μs 12.8064 KOps/s 13.8434 KOps/s $\textbf{\color{#d91a1a}-7.49\%}$
test_tc_first_layer_tensor 30.6310μs 0.7866μs 1.2713 MOps/s 1.2383 MOps/s $\color{#35bf28}+2.66\%$
test_tc_first_layer_nontensor 94.3150μs 2.2375μs 446.9327 KOps/s 446.2132 KOps/s $\color{#35bf28}+0.16\%$
test_tc_second_layer_tensor 12.7182μs 1.4154μs 706.5234 KOps/s 703.5868 KOps/s $\color{#35bf28}+0.42\%$
test_tc_second_layer_nontensor 22.3520μs 2.9566μs 338.2259 KOps/s 335.2744 KOps/s $\color{#35bf28}+0.88\%$
test_unbind 0.2375s 10.2693ms 97.3779 Ops/s 142.9291 Ops/s $\textbf{\color{#d91a1a}-31.87\%}$
test_full_like 10.8579ms 9.1438ms 109.3632 Ops/s 107.2402 Ops/s $\color{#35bf28}+1.98\%$
test_zeros_like 4.8601ms 4.3190ms 231.5360 Ops/s 234.8250 Ops/s $\color{#d91a1a}-1.40\%$
test_ones_like 4.5204ms 4.3251ms 231.2075 Ops/s 237.4955 Ops/s $\color{#d91a1a}-2.65\%$
test_clone 11.3049ms 9.0850ms 110.0719 Ops/s 110.6623 Ops/s $\color{#d91a1a}-0.53\%$
test_squeeze 59.5630μs 9.7374μs 102.6968 KOps/s 101.3356 KOps/s $\color{#35bf28}+1.34\%$
test_unsqueeze 0.2025ms 72.3898μs 13.8141 KOps/s 13.9825 KOps/s $\color{#d91a1a}-1.20\%$
test_split 0.2544ms 0.1570ms 6.3701 KOps/s 6.2831 KOps/s $\color{#35bf28}+1.38\%$
test_permute 0.2347ms 0.1786ms 5.5986 KOps/s 5.6328 KOps/s $\color{#d91a1a}-0.61\%$
test_stack 50.4184ms 50.1609ms 19.9358 Ops/s 20.0021 Ops/s $\color{#d91a1a}-0.33\%$
test_cat 50.4769ms 50.1170ms 19.9533 Ops/s 20.0353 Ops/s $\color{#d91a1a}-0.41\%$

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.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants