[Feature] Expose WrapModule

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] Expose WrapModule #1118
• [Feature] ProbabilisticTensorDictModule.num_samples #1117
• [Feature] Make ProbabilisticTensorDictSequential account for more than one distribution #1114

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}22$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.2820μs	17.5673μs	56.9241 KOps/s	59.0513 KOps/s	$\color{#d91a1a}-3.60\%$
test_plain_set_stack_nested	53.9210μs	17.9519μs	55.7046 KOps/s	57.6049 KOps/s	$\color{#d91a1a}-3.30\%$
test_plain_set_nested_inplace	47.6600μs	19.3355μs	51.7182 KOps/s	53.1901 KOps/s	$\color{#d91a1a}-2.77\%$
test_plain_set_stack_nested_inplace	47.4980μs	19.4093μs	51.5217 KOps/s	52.1501 KOps/s	$\color{#d91a1a}-1.20\%$
test_items	26.1390μs	4.2594μs	234.7726 KOps/s	236.7745 KOps/s	$\color{#d91a1a}-0.85\%$
test_items_nested	0.6761ms	0.4058ms	2.4643 KOps/s	2.4818 KOps/s	$\color{#d91a1a}-0.71\%$
test_items_nested_locked	0.5937ms	0.4058ms	2.4640 KOps/s	2.4189 KOps/s	$\color{#35bf28}+1.87\%$
test_items_nested_leaf	0.1765ms	72.2290μs	13.8449 KOps/s	13.9092 KOps/s	$\color{#d91a1a}-0.46\%$
test_items_stack_nested	0.6414ms	0.4102ms	2.4379 KOps/s	2.4507 KOps/s	$\color{#d91a1a}-0.52\%$
test_items_stack_nested_leaf	0.1249ms	74.9491μs	13.3424 KOps/s	13.3466 KOps/s	$\color{#d91a1a}-0.03\%$
test_items_stack_nested_locked	0.5214ms	0.4083ms	2.4492 KOps/s	2.4559 KOps/s	$\color{#d91a1a}-0.27\%$
test_keys	31.1080μs	3.5411μs	282.3993 KOps/s	246.7501 KOps/s	$\textbf{\color{#35bf28}+14.45\%}$
test_keys_nested	0.2219ms	0.1375ms	7.2721 KOps/s	7.2766 KOps/s	$\color{#d91a1a}-0.06\%$
test_keys_nested_locked	1.6880ms	0.1438ms	6.9562 KOps/s	6.9019 KOps/s	$\color{#35bf28}+0.79\%$
test_keys_nested_leaf	0.1983ms	0.1161ms	8.6159 KOps/s	8.2555 KOps/s	$\color{#35bf28}+4.36\%$
test_keys_stack_nested	0.2340ms	0.1359ms	7.3577 KOps/s	7.1717 KOps/s	$\color{#35bf28}+2.59\%$
test_keys_stack_nested_leaf	0.7961ms	0.1236ms	8.0888 KOps/s	8.2226 KOps/s	$\color{#d91a1a}-1.63\%$
test_keys_stack_nested_locked	0.2640ms	0.1409ms	7.0980 KOps/s	6.9122 KOps/s	$\color{#35bf28}+2.69\%$
test_values	5.8390μs	1.0370μs	964.3465 KOps/s	945.3625 KOps/s	$\color{#35bf28}+2.01\%$
test_values_nested	0.1089ms	55.3965μs	18.0517 KOps/s	17.8031 KOps/s	$\color{#35bf28}+1.40\%$
test_values_nested_locked	0.1069ms	55.2511μs	18.0992 KOps/s	17.9020 KOps/s	$\color{#35bf28}+1.10\%$
test_values_nested_leaf	0.1122ms	60.0540μs	16.6517 KOps/s	16.3475 KOps/s	$\color{#35bf28}+1.86\%$
test_values_stack_nested	0.1108ms	57.2152μs	17.4779 KOps/s	17.2526 KOps/s	$\color{#35bf28}+1.31\%$
test_values_stack_nested_leaf	0.4421ms	62.9550μs	15.8844 KOps/s	16.2551 KOps/s	$\color{#d91a1a}-2.28\%$
test_values_stack_nested_locked	0.1930ms	57.8647μs	17.2817 KOps/s	17.3432 KOps/s	$\color{#d91a1a}-0.35\%$
test_membership	7.1881μs	0.7495μs	1.3342 MOps/s	1.0521 MOps/s	$\textbf{\color{#35bf28}+26.82\%}$
test_membership_nested	19.6470μs	2.9613μs	337.6897 KOps/s	326.8228 KOps/s	$\color{#35bf28}+3.33\%$
test_membership_nested_leaf	40.9970μs	2.9579μs	338.0784 KOps/s	335.1323 KOps/s	$\color{#35bf28}+0.88\%$
test_membership_stacked_nested	19.7470μs	2.9532μs	338.6117 KOps/s	339.3087 KOps/s	$\color{#d91a1a}-0.21\%$
test_membership_stacked_nested_leaf	23.4640μs	2.9330μs	340.9529 KOps/s	333.2816 KOps/s	$\color{#35bf28}+2.30\%$
test_membership_nested_last	24.3360μs	4.1673μs	239.9653 KOps/s	236.6469 KOps/s	$\color{#35bf28}+1.40\%$
test_membership_nested_leaf_last	24.7360μs	4.1754μs	239.4983 KOps/s	228.2722 KOps/s	$\color{#35bf28}+4.92\%$
test_membership_stacked_nested_last	22.1920μs	4.1723μs	239.6778 KOps/s	181.5077 KOps/s	$\textbf{\color{#35bf28}+32.05\%}$
test_membership_stacked_nested_leaf_last	18.9360μs	4.1770μs	239.4077 KOps/s	182.2488 KOps/s	$\textbf{\color{#35bf28}+31.36\%}$
test_nested_getleaf	37.5400μs	10.4974μs	95.2614 KOps/s	91.7633 KOps/s	$\color{#35bf28}+3.81\%$
test_nested_get	38.9430μs	9.9790μs	100.2104 KOps/s	95.6148 KOps/s	$\color{#35bf28}+4.81\%$
test_stacked_getleaf	35.6370μs	10.5183μs	95.0723 KOps/s	92.3099 KOps/s	$\color{#35bf28}+2.99\%$
test_stacked_get	38.1220μs	9.9261μs	100.7447 KOps/s	95.6268 KOps/s	$\textbf{\color{#35bf28}+5.35\%}$
test_nested_getitemleaf	41.0170μs	10.9255μs	91.5287 KOps/s	88.0834 KOps/s	$\color{#35bf28}+3.91\%$
test_nested_getitem	37.2800μs	10.1421μs	98.5987 KOps/s	93.8168 KOps/s	$\textbf{\color{#35bf28}+5.10\%}$
test_stacked_getitemleaf	33.2020μs	10.8986μs	91.7548 KOps/s	87.6285 KOps/s	$\color{#35bf28}+4.71\%$
test_stacked_getitem	33.3020μs	10.2670μs	97.3994 KOps/s	93.9322 KOps/s	$\color{#35bf28}+3.69\%$
test_lock_nested	3.2900ms	0.4517ms	2.2139 KOps/s	2.2219 KOps/s	$\color{#d91a1a}-0.36\%$
test_lock_stack_nested	0.7964ms	0.4122ms	2.4259 KOps/s	2.4547 KOps/s	$\color{#d91a1a}-1.18\%$
test_unlock_nested	1.0361ms	0.3598ms	2.7790 KOps/s	2.7535 KOps/s	$\color{#35bf28}+0.93\%$
test_unlock_stack_nested	0.7214ms	0.3300ms	3.0306 KOps/s	3.0914 KOps/s	$\color{#d91a1a}-1.97\%$
test_flatten_speed	0.1915ms	94.6084μs	10.5699 KOps/s	10.5566 KOps/s	$\color{#35bf28}+0.13\%$
test_unflatten_speed	0.9801ms	0.4906ms	2.0385 KOps/s	2.0035 KOps/s	$\color{#35bf28}+1.74\%$
test_common_ops	6.8170ms	0.7768ms	1.2874 KOps/s	1.3465 KOps/s	$\color{#d91a1a}-4.39\%$
test_creation	41.6380μs	2.0731μs	482.3608 KOps/s	466.1474 KOps/s	$\color{#35bf28}+3.48\%$
test_creation_empty	42.8910μs	10.7104μs	93.3674 KOps/s	110.7458 KOps/s	$\textbf{\color{#d91a1a}-15.69\%}$
test_creation_nested_1	39.8650μs	14.0890μs	70.9771 KOps/s	83.1877 KOps/s	$\textbf{\color{#d91a1a}-14.68\%}$
test_creation_nested_2	1.4018ms	18.1343μs	55.1440 KOps/s	62.6853 KOps/s	$\textbf{\color{#d91a1a}-12.03\%}$
test_clone	70.1210μs	12.9808μs	77.0367 KOps/s	75.6326 KOps/s	$\color{#35bf28}+1.86\%$
test_getitem[int]	0.8538ms	12.5823μs	79.4767 KOps/s	78.1959 KOps/s	$\color{#35bf28}+1.64\%$
test_getitem[slice_int]	0.1406ms	24.8527μs	40.2371 KOps/s	40.6779 KOps/s	$\color{#d91a1a}-1.08\%$
test_getitem[range]	0.1734ms	48.6461μs	20.5566 KOps/s	20.5759 KOps/s	$\color{#d91a1a}-0.09\%$
test_getitem[tuple]	0.1533ms	20.1108μs	49.7244 KOps/s	48.3454 KOps/s	$\color{#35bf28}+2.85\%$
test_getitem[list]	0.2631ms	43.9203μs	22.7685 KOps/s	22.7502 KOps/s	$\color{#35bf28}+0.08\%$
test_setitem_dim[int]	51.2870μs	24.7330μs	40.4318 KOps/s	40.1950 KOps/s	$\color{#35bf28}+0.59\%$
test_setitem_dim[slice_int]	95.0090μs	50.5961μs	19.7644 KOps/s	19.3771 KOps/s	$\color{#35bf28}+2.00\%$
test_setitem_dim[range]	0.1122ms	73.7231μs	13.5643 KOps/s	13.6739 KOps/s	$\color{#d91a1a}-0.80\%$
test_setitem_dim[tuple]	77.2850μs	39.9485μs	25.0322 KOps/s	24.6696 KOps/s	$\color{#35bf28}+1.47\%$
test_setitem	0.1250ms	19.7713μs	50.5785 KOps/s	51.0585 KOps/s	$\color{#d91a1a}-0.94\%$
test_set	90.6200μs	19.5537μs	51.1411 KOps/s	52.7968 KOps/s	$\color{#d91a1a}-3.14\%$
test_set_shared	1.1701ms	0.1702ms	5.8754 KOps/s	5.8391 KOps/s	$\color{#35bf28}+0.62\%$
test_update	0.1690ms	21.8613μs	45.7430 KOps/s	48.7064 KOps/s	$\textbf{\color{#d91a1a}-6.08\%}$
test_update_nested	0.1158ms	32.0870μs	31.1652 KOps/s	32.6266 KOps/s	$\color{#d91a1a}-4.48\%$
test_update__nested	0.5891ms	32.1070μs	31.1458 KOps/s	29.2169 KOps/s	$\textbf{\color{#35bf28}+6.60\%}$
test_set_nested	89.8790μs	21.0834μs	47.4306 KOps/s	48.1025 KOps/s	$\color{#d91a1a}-1.40\%$
test_set_nested_new	77.1040μs	25.5947μs	39.0706 KOps/s	38.0520 KOps/s	$\color{#35bf28}+2.68\%$
test_select	96.1610μs	40.8172μs	24.4995 KOps/s	23.9059 KOps/s	$\color{#35bf28}+2.48\%$
test_select_nested	0.1205ms	59.2728μs	16.8712 KOps/s	16.3615 KOps/s	$\color{#35bf28}+3.11\%$
test_exclude_nested	0.1457ms	77.3120μs	12.9346 KOps/s	12.6130 KOps/s	$\color{#35bf28}+2.55\%$
test_empty[True]	0.6965ms	0.3832ms	2.6097 KOps/s	2.5433 KOps/s	$\color{#35bf28}+2.61\%$
test_empty[False]	10.7050μs	1.2571μs	795.4841 KOps/s	813.8327 KOps/s	$\color{#d91a1a}-2.25\%$
test_unbind_speed	0.3738ms	0.2589ms	3.8632 KOps/s	3.7260 KOps/s	$\color{#35bf28}+3.68\%$
test_unbind_speed_stack0	0.5661ms	0.2576ms	3.8816 KOps/s	3.9151 KOps/s	$\color{#d91a1a}-0.86\%$
test_unbind_speed_stack1	0.1091s	0.7730ms	1.2936 KOps/s	1.4439 KOps/s	$\textbf{\color{#d91a1a}-10.41\%}$
test_split	1.7540ms	1.5664ms	638.4210 Ops/s	577.7624 Ops/s	$\textbf{\color{#35bf28}+10.50\%}$
test_chunk	0.1372s	2.0013ms	499.6771 Ops/s	580.8788 Ops/s	$\textbf{\color{#d91a1a}-13.98\%}$
test_consolidate_njt[False-None]	10.9117ms	8.3357ms	119.9661 Ops/s	118.3285 Ops/s	$\color{#35bf28}+1.38\%$
test_creation[device0]	4.7255ms	94.6477μs	10.5655 KOps/s	10.8092 KOps/s	$\color{#d91a1a}-2.25\%$
test_creation_from_tensor	0.2747ms	95.8137μs	10.4369 KOps/s	10.3733 KOps/s	$\color{#35bf28}+0.61\%$
test_add_one[memmap_tensor0]	0.1772ms	4.7428μs	210.8453 KOps/s	206.7204 KOps/s	$\color{#35bf28}+2.00\%$
test_contiguous[memmap_tensor0]	16.7810μs	0.5231μs	1.9115 MOps/s	1.8787 MOps/s	$\color{#35bf28}+1.75\%$
test_stack[memmap_tensor0]	45.0940μs	3.4809μs	287.2855 KOps/s	285.9437 KOps/s	$\color{#35bf28}+0.47\%$
test_memmaptd_index	0.4648ms	0.2357ms	4.2427 KOps/s	4.2643 KOps/s	$\color{#d91a1a}-0.51\%$
test_memmaptd_index_astensor	0.7214ms	0.3154ms	3.1701 KOps/s	3.2045 KOps/s	$\color{#d91a1a}-1.07\%$
test_memmaptd_index_op	1.0967ms	0.5678ms	1.7612 KOps/s	1.8749 KOps/s	$\textbf{\color{#d91a1a}-6.07\%}$
test_serialize_model	0.1242s	0.1161s	8.6101 Ops/s	7.6101 Ops/s	$\textbf{\color{#35bf28}+13.14\%}$
test_serialize_model_pickle	0.4612s	0.3899s	2.5650 Ops/s	2.5478 Ops/s	$\color{#35bf28}+0.68\%$
test_serialize_weights	0.1165s	0.1114s	8.9799 Ops/s	8.8102 Ops/s	$\color{#35bf28}+1.93\%$
test_serialize_weights_returnearly	0.1686s	0.1582s	6.3217 Ops/s	6.3263 Ops/s	$\color{#d91a1a}-0.07\%$
test_serialize_weights_pickle	0.4811s	0.4158s	2.4050 Ops/s	2.4565 Ops/s	$\color{#d91a1a}-2.10\%$
test_serialize_weights_filesystem	0.1450s	0.1396s	7.1611 Ops/s	6.3750 Ops/s	$\textbf{\color{#35bf28}+12.33\%}$
test_serialize_model_filesystem	0.1590s	0.1495s	6.6872 Ops/s	6.7395 Ops/s	$\color{#d91a1a}-0.78\%$
test_reshape_pytree	65.1120μs	27.2427μs	36.7070 KOps/s	35.2676 KOps/s	$\color{#35bf28}+4.08\%$
test_reshape_td	79.9100μs	32.7476μs	30.5365 KOps/s	28.7128 KOps/s	$\textbf{\color{#35bf28}+6.35\%}$
test_view_pytree	63.4590μs	26.8320μs	37.2689 KOps/s	35.4408 KOps/s	$\textbf{\color{#35bf28}+5.16\%}$
test_view_td	84.5280μs	38.5172μs	25.9624 KOps/s	25.0621 KOps/s	$\color{#35bf28}+3.59\%$
test_unbind_pytree	70.0020μs	29.7853μs	33.5736 KOps/s	32.5127 KOps/s	$\color{#35bf28}+3.26\%$
test_unbind_td	0.3603ms	37.8814μs	26.3982 KOps/s	24.8315 KOps/s	$\textbf{\color{#35bf28}+6.31\%}$
test_split_pytree	84.8090μs	29.6777μs	33.6954 KOps/s	32.9880 KOps/s	$\color{#35bf28}+2.14\%$
test_split_td	0.5340ms	43.7754μs	22.8439 KOps/s	21.9723 KOps/s	$\color{#35bf28}+3.97\%$
test_add_pytree	80.7610μs	35.4619μs	28.1993 KOps/s	27.5441 KOps/s	$\color{#35bf28}+2.38\%$
test_add_td	0.1239ms	54.2282μs	18.4406 KOps/s	18.6896 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_add_one_nested[tensordict-compile]	0.1123ms	63.0661μs	15.8564 KOps/s	15.9386 KOps/s	$\color{#d91a1a}-0.52\%$
test_compile_add_one_nested[tensordict-eager]	0.3980ms	0.1624ms	6.1582 KOps/s	6.1148 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_add_one_nested[pytree-compile]	0.1077ms	46.0642μs	21.7088 KOps/s	21.6378 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_add_one_nested[pytree-eager]	0.2219ms	0.1177ms	8.4965 KOps/s	8.3394 KOps/s	$\color{#35bf28}+1.88\%$
test_compile_copy_nested[tensordict-compile]	60.8040μs	27.0506μs	36.9677 KOps/s	38.1991 KOps/s	$\color{#d91a1a}-3.22\%$
test_compile_copy_nested[tensordict-eager]	0.1164ms	54.4116μs	18.3784 KOps/s	18.2916 KOps/s	$\color{#35bf28}+0.47\%$
test_compile_copy_nested[pytree-compile]	0.1430ms	78.7136μs	12.7043 KOps/s	12.2951 KOps/s	$\color{#35bf28}+3.33\%$
test_compile_copy_nested[pytree-eager]	0.1401ms	67.7101μs	14.7688 KOps/s	14.3879 KOps/s	$\color{#35bf28}+2.65\%$
test_compile_add_one_flat[tensordict-compile]	0.2120ms	0.1061ms	9.4224 KOps/s	9.5894 KOps/s	$\color{#d91a1a}-1.74\%$
test_compile_add_one_flat[tensordict-eager]	0.3936ms	0.1964ms	5.0909 KOps/s	4.7830 KOps/s	$\textbf{\color{#35bf28}+6.44\%}$
test_compile_add_one_flat[tensorclass-compile]	0.1092ms	45.1494μs	22.1487 KOps/s	20.8586 KOps/s	$\textbf{\color{#35bf28}+6.19\%}$
test_compile_add_one_flat[tensorclass-eager]	0.5038ms	60.7590μs	16.4585 KOps/s	15.9867 KOps/s	$\color{#35bf28}+2.95\%$
test_compile_add_one_flat[pytree-compile]	0.2252ms	0.1054ms	9.4850 KOps/s	9.7329 KOps/s	$\color{#d91a1a}-2.55\%$
test_compile_add_one_flat[pytree-eager]	0.4178ms	0.2011ms	4.9720 KOps/s	4.9100 KOps/s	$\color{#35bf28}+1.26\%$
test_compile_add_self_flat[tensordict-eager]	0.4213ms	0.2112ms	4.7344 KOps/s	4.5292 KOps/s	$\color{#35bf28}+4.53\%$
test_compile_add_self_flat[tensordict-compile]	0.5118ms	0.1055ms	9.4809 KOps/s	9.4558 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_add_self_flat[tensorclass-eager]	0.1484ms	56.3516μs	17.7457 KOps/s	18.0966 KOps/s	$\color{#d91a1a}-1.94\%$
test_compile_add_self_flat[tensorclass-compile]	0.1078ms	46.4425μs	21.5320 KOps/s	20.2482 KOps/s	$\textbf{\color{#35bf28}+6.34\%}$
test_compile_add_self_flat[pytree-eager]	0.6246ms	0.1580ms	6.3286 KOps/s	6.1990 KOps/s	$\color{#35bf28}+2.09\%$
test_compile_add_self_flat[pytree-compile]	0.1784ms	0.1076ms	9.2967 KOps/s	9.7115 KOps/s	$\color{#d91a1a}-4.27\%$
test_compile_copy_flat[tensordict-compile]	58.5690μs	21.0134μs	47.5887 KOps/s	48.0024 KOps/s	$\color{#d91a1a}-0.86\%$
test_compile_copy_flat[tensordict-eager]	0.1304ms	58.8361μs	16.9964 KOps/s	16.9372 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_copy_flat[pytree-compile]	0.1535ms	80.5863μs	12.4091 KOps/s	11.9847 KOps/s	$\color{#35bf28}+3.54\%$
test_compile_copy_flat[pytree-eager]	0.1550ms	68.4502μs	14.6092 KOps/s	14.2999 KOps/s	$\color{#35bf28}+2.16\%$
test_compile_assign_and_add[tensordict-compile]	0.2973ms	0.2063ms	4.8476 KOps/s	4.8455 KOps/s	$\color{#35bf28}+0.04\%$
test_compile_assign_and_add[tensordict-eager]	2.3187ms	1.2868ms	777.1200 Ops/s	780.1837 Ops/s	$\color{#d91a1a}-0.39\%$
test_compile_assign_and_add[pytree-compile]	0.2889ms	0.2047ms	4.8841 KOps/s	4.9173 KOps/s	$\color{#d91a1a}-0.67\%$
test_compile_assign_and_add[pytree-eager]	0.9736ms	0.7666ms	1.3044 KOps/s	1.2838 KOps/s	$\color{#35bf28}+1.60\%$
test_compile_assign_and_add_stack[compile]	0.8404ms	0.4588ms	2.1798 KOps/s	2.1905 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_assign_and_add_stack[eager]	3.4529ms	2.5380ms	394.0150 Ops/s	402.9492 Ops/s	$\color{#d91a1a}-2.22\%$
test_compile_indexing[tensor-tensordict-compile]	92.3530μs	37.6551μs	26.5568 KOps/s	27.5640 KOps/s	$\color{#d91a1a}-3.65\%$
test_compile_indexing[tensor-tensordict-eager]	0.4641ms	31.9286μs	31.3198 KOps/s	29.2526 KOps/s	$\textbf{\color{#35bf28}+7.07\%}$
test_compile_indexing[tensor-tensorclass-compile]	88.3130μs	29.8931μs	33.4526 KOps/s	33.5629 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1003ms	23.8088μs	42.0013 KOps/s	41.2118 KOps/s	$\color{#35bf28}+1.92\%$
test_compile_indexing[tensor-pytree-compile]	79.7490μs	31.3104μs	31.9383 KOps/s	33.4021 KOps/s	$\color{#d91a1a}-4.38\%$
test_compile_indexing[tensor-pytree-eager]	90.8440μs	22.9705μs	43.5340 KOps/s	42.2034 KOps/s	$\color{#35bf28}+3.15\%$
test_compile_indexing[slice-tensordict-compile]	0.1413ms	52.5882μs	19.0157 KOps/s	18.9894 KOps/s	$\color{#35bf28}+0.14\%$
test_compile_indexing[slice-tensordict-eager]	0.5064ms	18.9290μs	52.8289 KOps/s	47.8903 KOps/s	$\textbf{\color{#35bf28}+10.31\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1060ms	44.8911μs	22.2761 KOps/s	22.0622 KOps/s	$\color{#35bf28}+0.97\%$
test_compile_indexing[slice-tensorclass-eager]	52.8890μs	18.8106μs	53.1616 KOps/s	50.9766 KOps/s	$\color{#35bf28}+4.29\%$
test_compile_indexing[slice-pytree-compile]	0.1007ms	45.2132μs	22.1174 KOps/s	21.6304 KOps/s	$\color{#35bf28}+2.25\%$
test_compile_indexing[slice-pytree-eager]	88.7290μs	18.8591μs	53.0249 KOps/s	51.0477 KOps/s	$\color{#35bf28}+3.87\%$
test_compile_indexing[int-tensordict-compile]	0.1261ms	54.1619μs	18.4632 KOps/s	18.4483 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_indexing[int-tensordict-eager]	1.0628ms	19.5419μs	51.1721 KOps/s	48.0766 KOps/s	$\textbf{\color{#35bf28}+6.44\%}$
test_compile_indexing[int-tensorclass-compile]	95.7830μs	45.4368μs	22.0086 KOps/s	21.6267 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_indexing[int-tensorclass-eager]	53.0900μs	18.9474μs	52.7778 KOps/s	51.3739 KOps/s	$\color{#35bf28}+2.73\%$
test_compile_indexing[int-pytree-compile]	0.1026ms	45.3792μs	22.0365 KOps/s	21.6936 KOps/s	$\color{#35bf28}+1.58\%$
test_compile_indexing[int-pytree-eager]	54.5320μs	18.5241μs	53.9837 KOps/s	50.9326 KOps/s	$\textbf{\color{#35bf28}+5.99\%}$
test_mod_add[eager]	85.1990μs	35.0325μs	28.5449 KOps/s	30.0564 KOps/s	$\textbf{\color{#d91a1a}-5.03\%}$
test_mod_add[compile]	0.1851ms	48.6247μs	20.5657 KOps/s	20.4849 KOps/s	$\color{#35bf28}+0.39\%$
test_mod_add[compile-overhead]	96.2000μs	48.6416μs	20.5585 KOps/s	20.6692 KOps/s	$\color{#d91a1a}-0.54\%$
test_mod_wrap[eager]	0.4267ms	0.2236ms	4.4719 KOps/s	4.5486 KOps/s	$\color{#d91a1a}-1.69\%$
test_mod_wrap[compile]	0.4195ms	0.2075ms	4.8192 KOps/s	4.7638 KOps/s	$\color{#35bf28}+1.16\%$
test_mod_wrap[compile-overhead]	0.6244ms	0.2160ms	4.6304 KOps/s	4.8402 KOps/s	$\color{#d91a1a}-4.33\%$
test_mod_wrap_and_backward[eager]	16.2305ms	12.3105ms	81.2315 Ops/s	90.6588 Ops/s	$\textbf{\color{#d91a1a}-10.40\%}$
test_mod_wrap_and_backward[compile]	14.4814ms	13.1081ms	76.2888 Ops/s	92.5357 Ops/s	$\textbf{\color{#d91a1a}-17.56\%}$
test_mod_wrap_and_backward[compile-overhead]	14.7886ms	12.6445ms	79.0855 Ops/s	91.9302 Ops/s	$\textbf{\color{#d91a1a}-13.97\%}$
test_seq_add[eager]	0.1824ms	0.1106ms	9.0437 KOps/s	9.0032 KOps/s	$\color{#35bf28}+0.45\%$
test_seq_add[compile]	0.1420ms	63.3038μs	15.7968 KOps/s	15.4752 KOps/s	$\color{#35bf28}+2.08\%$
test_seq_add[compile-overhead]	0.1347ms	61.3361μs	16.3036 KOps/s	16.5818 KOps/s	$\color{#d91a1a}-1.68\%$
test_seq_wrap[eager]	0.8200ms	0.4446ms	2.2491 KOps/s	2.2989 KOps/s	$\color{#d91a1a}-2.17\%$
test_seq_wrap[compile]	0.3378ms	0.2296ms	4.3552 KOps/s	4.2937 KOps/s	$\color{#35bf28}+1.43\%$
test_seq_wrap[compile-overhead]	0.3589ms	0.2279ms	4.3884 KOps/s	4.3505 KOps/s	$\color{#35bf28}+0.87\%$
test_func_call_runtime[False-eager]	0.9526ms	0.5334ms	1.8747 KOps/s	1.8263 KOps/s	$\color{#35bf28}+2.65\%$
test_func_call_runtime[False-compile]	0.5935ms	0.4280ms	2.3367 KOps/s	2.3328 KOps/s	$\color{#35bf28}+0.16\%$
test_func_call_runtime[False-compile-overhead]	0.5481ms	0.4258ms	2.3485 KOps/s	2.3056 KOps/s	$\color{#35bf28}+1.86\%$
test_func_call_runtime[True-eager]	1.4821ms	0.7500ms	1.3333 KOps/s	1.3029 KOps/s	$\color{#35bf28}+2.33\%$
test_func_call_runtime[True-compile]	0.8678ms	0.4671ms	2.1409 KOps/s	2.1210 KOps/s	$\color{#35bf28}+0.94\%$
test_func_call_runtime[True-compile-overhead]	0.5629ms	0.4663ms	2.1443 KOps/s	2.1281 KOps/s	$\color{#35bf28}+0.76\%$
test_func_call_cm_runtime[False-eager]	1.2099ms	0.5329ms	1.8765 KOps/s	1.8313 KOps/s	$\color{#35bf28}+2.47\%$
test_func_call_cm_runtime[False-compile]	2.3404ms	0.4425ms	2.2599 KOps/s	2.3181 KOps/s	$\color{#d91a1a}-2.51\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7042ms	0.4270ms	2.3421 KOps/s	2.3354 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[True-eager]	1.0199ms	0.8841ms	1.1312 KOps/s	1.1082 KOps/s	$\color{#35bf28}+2.07\%$
test_func_call_cm_runtime[True-compile]	1.1926ms	0.4946ms	2.0220 KOps/s	2.0145 KOps/s	$\color{#35bf28}+0.37\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6515ms	0.4906ms	2.0385 KOps/s	2.0153 KOps/s	$\color{#35bf28}+1.15\%$
test_vmap_func_call_cm_runtime[eager]	2.5610ms	1.8827ms	531.1460 Ops/s	525.5197 Ops/s	$\color{#35bf28}+1.07\%$
test_vmap_func_call_cm_runtime[compile]	1.0086ms	0.5240ms	1.9085 KOps/s	1.9011 KOps/s	$\color{#35bf28}+0.39\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.8819ms	0.5277ms	1.8950 KOps/s	1.8704 KOps/s	$\color{#35bf28}+1.31\%$
test_distributed	0.2873ms	0.1276ms	7.8395 KOps/s	7.7147 KOps/s	$\color{#35bf28}+1.62\%$
test_tdmodule	44.9740μs	26.5220μs	37.7045 KOps/s	39.1729 KOps/s	$\color{#d91a1a}-3.75\%$
test_tdmodule_dispatch	78.6780μs	47.3055μs	21.1392 KOps/s	20.9383 KOps/s	$\color{#35bf28}+0.96\%$
test_tdseq	53.5710μs	26.4006μs	37.8779 KOps/s	38.5894 KOps/s	$\color{#d91a1a}-1.84\%$
test_tdseq_dispatch	81.0520μs	50.3826μs	19.8481 KOps/s	20.8071 KOps/s	$\color{#d91a1a}-4.61\%$
test_instantiation_functorch	2.0105ms	1.5238ms	656.2519 Ops/s	624.1480 Ops/s	$\textbf{\color{#35bf28}+5.14\%}$
test_exec_functorch	0.3177ms	0.1807ms	5.5328 KOps/s	5.4460 KOps/s	$\color{#35bf28}+1.59\%$
test_exec_functional_call	0.4739ms	0.1780ms	5.6186 KOps/s	5.7303 KOps/s	$\color{#d91a1a}-1.95\%$
test_exec_td_decorator	0.5122ms	0.2308ms	4.3320 KOps/s	4.3311 KOps/s	$\color{#35bf28}+0.02\%$
test_vmap_mlp_speed_decorator[True-True]	1.0307ms	0.6659ms	1.5017 KOps/s	1.5365 KOps/s	$\color{#d91a1a}-2.27\%$
test_vmap_mlp_speed_decorator[True-False]	1.1518ms	0.6511ms	1.5358 KOps/s	1.5014 KOps/s	$\color{#35bf28}+2.29\%$
test_vmap_mlp_speed_decorator[False-True]	0.7328ms	0.5214ms	1.9177 KOps/s	1.8736 KOps/s	$\color{#35bf28}+2.36\%$
test_vmap_mlp_speed_decorator[False-False]	0.9628ms	0.5198ms	1.9237 KOps/s	1.8754 KOps/s	$\color{#35bf28}+2.58\%$
test_to_module_speed[True]	2.1132ms	1.2853ms	778.0082 Ops/s	758.2307 Ops/s	$\color{#35bf28}+2.61\%$
test_to_module_speed[False]	1.6646ms	1.2564ms	795.9469 Ops/s	772.1191 Ops/s	$\color{#35bf28}+3.09\%$
test_tc_init	84.3780μs	45.9568μs	21.7596 KOps/s	22.1186 KOps/s	$\color{#d91a1a}-1.62\%$
test_tc_init_nested	0.2087ms	91.9134μs	10.8798 KOps/s	11.3385 KOps/s	$\color{#d91a1a}-4.05\%$
test_tc_first_layer_tensor	38.7480μs	1.5485μs	645.7697 KOps/s	648.6115 KOps/s	$\color{#d91a1a}-0.44\%$
test_tc_first_layer_nontensor	22.9120μs	4.7785μs	209.2697 KOps/s	211.7119 KOps/s	$\color{#d91a1a}-1.15\%$
test_tc_second_layer_tensor	20.4980μs	2.8649μs	349.0475 KOps/s	353.2499 KOps/s	$\color{#d91a1a}-1.19\%$
test_tc_second_layer_nontensor	59.1680μs	6.0248μs	165.9806 KOps/s	163.8880 KOps/s	$\color{#35bf28}+1.28\%$
test_unbind	0.2225s	12.9455ms	77.2469 Ops/s	79.8032 Ops/s	$\color{#d91a1a}-3.20\%$
test_full_like	16.5745ms	11.8527ms	84.3692 Ops/s	80.0977 Ops/s	$\textbf{\color{#35bf28}+5.33\%}$
test_zeros_like	13.0459ms	7.8517ms	127.3611 Ops/s	134.7831 Ops/s	$\textbf{\color{#d91a1a}-5.51\%}$
test_ones_like	12.4532ms	7.8739ms	127.0018 Ops/s	128.3639 Ops/s	$\color{#d91a1a}-1.06\%$
test_clone	16.7010ms	9.5218ms	105.0222 Ops/s	103.5484 Ops/s	$\color{#35bf28}+1.42\%$
test_squeeze	64.2210μs	12.5219μs	79.8598 KOps/s	81.6634 KOps/s	$\color{#d91a1a}-2.21\%$
test_unsqueeze	0.1476ms	89.2255μs	11.2076 KOps/s	11.0986 KOps/s	$\color{#35bf28}+0.98\%$
test_split	0.4811ms	0.1955ms	5.1147 KOps/s	5.0754 KOps/s	$\color{#35bf28}+0.77\%$
test_permute	0.3244ms	0.2236ms	4.4714 KOps/s	4.5257 KOps/s	$\color{#d91a1a}-1.20\%$
test_stack	29.3148ms	25.6389ms	39.0032 Ops/s	39.8665 Ops/s	$\color{#d91a1a}-2.17\%$
test_cat	30.5586ms	25.5284ms	39.1720 Ops/s	40.1367 Ops/s	$\color{#d91a1a}-2.40\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}17$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	26.0500μs	11.0223μs	90.7255 KOps/s	92.9009 KOps/s	$\color{#d91a1a}-2.34\%$
test_plain_set_stack_nested	33.7110μs	10.9487μs	91.3351 KOps/s	92.0098 KOps/s	$\color{#d91a1a}-0.73\%$
test_plain_set_nested_inplace	45.4010μs	11.8148μs	84.6398 KOps/s	84.8179 KOps/s	$\color{#d91a1a}-0.21\%$
test_plain_set_stack_nested_inplace	46.5610μs	11.7985μs	84.7569 KOps/s	86.2045 KOps/s	$\color{#d91a1a}-1.68\%$
test_items	33.8800μs	2.8963μs	345.2735 KOps/s	342.8511 KOps/s	$\color{#35bf28}+0.71\%$
test_items_nested	0.3873ms	0.3529ms	2.8339 KOps/s	2.8653 KOps/s	$\color{#d91a1a}-1.09\%$
test_items_nested_locked	0.3854ms	0.3515ms	2.8449 KOps/s	2.8662 KOps/s	$\color{#d91a1a}-0.74\%$
test_items_nested_leaf	87.4610μs	58.2732μs	17.1606 KOps/s	17.3396 KOps/s	$\color{#d91a1a}-1.03\%$
test_items_stack_nested	0.3911ms	0.3509ms	2.8502 KOps/s	2.8499 KOps/s	$+0.01\%$
test_items_stack_nested_leaf	91.8110μs	58.8856μs	16.9821 KOps/s	16.7199 KOps/s	$\color{#35bf28}+1.57\%$
test_items_stack_nested_locked	0.3848ms	0.3500ms	2.8569 KOps/s	2.8652 KOps/s	$\color{#d91a1a}-0.29\%$
test_keys	28.7500μs	3.4309μs	291.4698 KOps/s	290.7440 KOps/s	$\color{#35bf28}+0.25\%$
test_keys_nested	0.1010ms	69.9521μs	14.2955 KOps/s	14.3021 KOps/s	$\color{#d91a1a}-0.05\%$
test_keys_nested_locked	0.8127ms	75.3316μs	13.2746 KOps/s	13.2600 KOps/s	$\color{#35bf28}+0.11\%$
test_keys_nested_leaf	94.3010μs	61.0617μs	16.3769 KOps/s	16.2076 KOps/s	$\color{#35bf28}+1.04\%$
test_keys_stack_nested	0.1019ms	70.8065μs	14.1230 KOps/s	14.2093 KOps/s	$\color{#d91a1a}-0.61\%$
test_keys_stack_nested_leaf	96.9810μs	61.2943μs	16.3147 KOps/s	16.0995 KOps/s	$\color{#35bf28}+1.34\%$
test_keys_stack_nested_locked	0.1172ms	75.8249μs	13.1883 KOps/s	13.0900 KOps/s	$\color{#35bf28}+0.75\%$
test_values	4.6033μs	0.8489μs	1.1780 MOps/s	1.1480 MOps/s	$\color{#35bf28}+2.62\%$
test_values_nested	71.1410μs	31.2360μs	32.0143 KOps/s	32.2005 KOps/s	$\color{#d91a1a}-0.58\%$
test_values_nested_locked	71.4510μs	32.6532μs	30.6248 KOps/s	30.6686 KOps/s	$\color{#d91a1a}-0.14\%$
test_values_nested_leaf	59.1010μs	33.4272μs	29.9158 KOps/s	29.8841 KOps/s	$\color{#35bf28}+0.11\%$
test_values_stack_nested	56.4410μs	31.5443μs	31.7015 KOps/s	31.4382 KOps/s	$\color{#35bf28}+0.84\%$
test_values_stack_nested_leaf	69.7710μs	33.8385μs	29.5522 KOps/s	29.1959 KOps/s	$\color{#35bf28}+1.22\%$
test_values_stack_nested_locked	74.2510μs	33.0397μs	30.2666 KOps/s	30.1143 KOps/s	$\color{#35bf28}+0.51\%$
test_membership	2.0235μs	0.5095μs	1.9625 MOps/s	1.9910 MOps/s	$\color{#d91a1a}-1.43\%$
test_membership_nested	19.5900μs	1.9791μs	505.2928 KOps/s	531.4262 KOps/s	$\color{#d91a1a}-4.92\%$
test_membership_nested_leaf	18.6055μs	1.9714μs	507.2458 KOps/s	517.5170 KOps/s	$\color{#d91a1a}-1.98\%$
test_membership_stacked_nested	34.1800μs	2.0581μs	485.8855 KOps/s	496.9786 KOps/s	$\color{#d91a1a}-2.23\%$
test_membership_stacked_nested_leaf	48.3500μs	2.0922μs	477.9698 KOps/s	501.8514 KOps/s	$\color{#d91a1a}-4.76\%$
test_membership_nested_last	38.7810μs	2.9583μs	338.0322 KOps/s	348.0033 KOps/s	$\color{#d91a1a}-2.87\%$
test_membership_nested_leaf_last	26.0900μs	2.9570μs	338.1861 KOps/s	347.2838 KOps/s	$\color{#d91a1a}-2.62\%$
test_membership_stacked_nested_last	25.5100μs	2.9408μs	340.0404 KOps/s	297.2915 KOps/s	$\textbf{\color{#35bf28}+14.38\%}$
test_membership_stacked_nested_leaf_last	55.3110μs	2.9284μs	341.4820 KOps/s	301.4147 KOps/s	$\textbf{\color{#35bf28}+13.29\%}$
test_nested_getleaf	30.8100μs	6.0825μs	164.4068 KOps/s	164.5754 KOps/s	$\color{#d91a1a}-0.10\%$
test_nested_get	40.0910μs	5.7630μs	173.5201 KOps/s	173.0323 KOps/s	$\color{#35bf28}+0.28\%$
test_stacked_getleaf	40.1010μs	6.0971μs	164.0112 KOps/s	165.1822 KOps/s	$\color{#d91a1a}-0.71\%$
test_stacked_get	31.1810μs	5.7704μs	173.2990 KOps/s	173.1350 KOps/s	$\color{#35bf28}+0.09\%$
test_nested_getitemleaf	0.7511ms	6.1052μs	163.7952 KOps/s	162.5913 KOps/s	$\color{#35bf28}+0.74\%$
test_nested_getitem	37.3310μs	5.8377μs	171.3004 KOps/s	171.2466 KOps/s	$\color{#35bf28}+0.03\%$
test_stacked_getitemleaf	0.3885ms	6.1634μs	162.2486 KOps/s	162.2402 KOps/s	$+0.01\%$
test_stacked_getitem	27.9500μs	5.8428μs	171.1509 KOps/s	171.3648 KOps/s	$\color{#d91a1a}-0.12\%$
test_lock_nested	9.3991ms	0.3705ms	2.6992 KOps/s	2.7220 KOps/s	$\color{#d91a1a}-0.84\%$
test_lock_stack_nested	0.7214ms	0.3312ms	3.0191 KOps/s	3.0443 KOps/s	$\color{#d91a1a}-0.83\%$
test_unlock_nested	0.7194ms	0.3009ms	3.3229 KOps/s	3.3338 KOps/s	$\color{#d91a1a}-0.33\%$
test_unlock_stack_nested	0.6643ms	0.2709ms	3.6920 KOps/s	3.7282 KOps/s	$\color{#d91a1a}-0.97\%$
test_flatten_speed	0.1150ms	74.4722μs	13.4278 KOps/s	13.3838 KOps/s	$\color{#35bf28}+0.33\%$
test_unflatten_speed	0.6876ms	0.3030ms	3.3003 KOps/s	3.3534 KOps/s	$\color{#d91a1a}-1.58\%$
test_common_ops	1.6540ms	0.5950ms	1.6808 KOps/s	1.6943 KOps/s	$\color{#d91a1a}-0.80\%$
test_creation	0.1757ms	1.4162μs	706.1339 KOps/s	697.7189 KOps/s	$\color{#35bf28}+1.21\%$
test_creation_empty	37.4000μs	7.9120μs	126.3903 KOps/s	130.5294 KOps/s	$\color{#d91a1a}-3.17\%$
test_creation_nested_1	34.1210μs	9.5462μs	104.7534 KOps/s	108.7915 KOps/s	$\color{#d91a1a}-3.71\%$
test_creation_nested_2	36.7800μs	12.0786μs	82.7910 KOps/s	85.4482 KOps/s	$\color{#d91a1a}-3.11\%$
test_clone	57.5010μs	10.2455μs	97.6042 KOps/s	95.2751 KOps/s	$\color{#35bf28}+2.44\%$
test_getitem[int]	92.9929ms	15.5477μs	64.3182 KOps/s	95.7063 KOps/s	$\textbf{\color{#d91a1a}-32.80\%}$
test_getitem[slice_int]	0.1049ms	20.0667μs	49.8338 KOps/s	50.2216 KOps/s	$\color{#d91a1a}-0.77\%$
test_getitem[range]	0.1285ms	35.7686μs	27.9575 KOps/s	26.3236 KOps/s	$\textbf{\color{#35bf28}+6.21\%}$
test_getitem[tuple]	0.1046ms	17.1682μs	58.2473 KOps/s	56.7023 KOps/s	$\color{#35bf28}+2.72\%$
test_getitem[list]	0.2189ms	31.5775μs	31.6681 KOps/s	30.2904 KOps/s	$\color{#35bf28}+4.55\%$
test_setitem_dim[int]	37.0500μs	17.8681μs	55.9658 KOps/s	54.0842 KOps/s	$\color{#35bf28}+3.48\%$
test_setitem_dim[slice_int]	61.0010μs	36.7094μs	27.2410 KOps/s	26.9746 KOps/s	$\color{#35bf28}+0.99\%$
test_setitem_dim[range]	76.0710μs	50.8721μs	19.6571 KOps/s	19.1068 KOps/s	$\color{#35bf28}+2.88\%$
test_setitem_dim[tuple]	58.4910μs	30.3960μs	32.8990 KOps/s	31.8455 KOps/s	$\color{#35bf28}+3.31\%$
test_setitem	84.3420μs	14.6196μs	68.4013 KOps/s	66.5654 KOps/s	$\color{#35bf28}+2.76\%$
test_set	96.1910μs	14.3789μs	69.5465 KOps/s	69.3309 KOps/s	$\color{#35bf28}+0.31\%$
test_set_shared	1.4763ms	0.1444ms	6.9272 KOps/s	6.8823 KOps/s	$\color{#35bf28}+0.65\%$
test_update	0.6485ms	17.0839μs	58.5348 KOps/s	58.5338 KOps/s	$+0.00\%$
test_update_nested	89.1320μs	21.7210μs	46.0385 KOps/s	45.0854 KOps/s	$\color{#35bf28}+2.11\%$
test_update__nested	0.7610ms	23.6804μs	42.2290 KOps/s	41.2018 KOps/s	$\color{#35bf28}+2.49\%$
test_set_nested	85.0720μs	15.3814μs	65.0135 KOps/s	64.2435 KOps/s	$\color{#35bf28}+1.20\%$
test_set_nested_new	0.1015ms	17.4464μs	57.3183 KOps/s	54.9287 KOps/s	$\color{#35bf28}+4.35\%$
test_select	90.6120μs	29.5601μs	33.8293 KOps/s	33.2074 KOps/s	$\color{#35bf28}+1.87\%$
test_select_nested	91.0310μs	41.2689μs	24.2313 KOps/s	24.0493 KOps/s	$\color{#35bf28}+0.76\%$
test_exclude_nested	96.9420μs	59.2190μs	16.8865 KOps/s	16.9276 KOps/s	$\color{#d91a1a}-0.24\%$
test_empty[True]	0.3090ms	0.2685ms	3.7242 KOps/s	3.7025 KOps/s	$\color{#35bf28}+0.59\%$
test_empty[False]	5.3461μs	0.7389μs	1.3533 MOps/s	1.3503 MOps/s	$\color{#35bf28}+0.23\%$
test_to	85.4810μs	52.9633μs	18.8810 KOps/s	18.5573 KOps/s	$\color{#35bf28}+1.74\%$
test_to_nonblocking	76.9510μs	45.1119μs	22.1671 KOps/s	20.8234 KOps/s	$\textbf{\color{#35bf28}+6.45\%}$
test_unbind_speed	0.2587ms	0.2268ms	4.4089 KOps/s	4.4400 KOps/s	$\color{#d91a1a}-0.70\%$
test_unbind_speed_stack0	0.3304ms	0.2254ms	4.4371 KOps/s	4.3853 KOps/s	$\color{#35bf28}+1.18\%$
test_unbind_speed_stack1	92.5130ms	0.6434ms	1.5543 KOps/s	1.5607 KOps/s	$\color{#d91a1a}-0.41\%$
test_split	93.8196ms	1.6762ms	596.5977 Ops/s	584.8149 Ops/s	$\color{#35bf28}+2.01\%$
test_chunk	96.4195ms	1.5600ms	641.0322 Ops/s	693.1673 Ops/s	$\textbf{\color{#d91a1a}-7.52\%}$
test_consolidate[False-None]	2.6493ms	2.5695ms	389.1876 Ops/s	350.6690 Ops/s	$\textbf{\color{#35bf28}+10.98\%}$
test_consolidate[default-None]	1.8096ms	1.6510ms	605.6769 Ops/s	594.3022 Ops/s	$\color{#35bf28}+1.91\%$
test_consolidate[reduce-overhead-None]	1.7554ms	1.6914ms	591.2312 Ops/s	576.3324 Ops/s	$\color{#35bf28}+2.59\%$
test_consolidate_njt[False-None]	6.6730ms	6.3885ms	156.5322 Ops/s	156.8570 Ops/s	$\color{#d91a1a}-0.21\%$
test_to[False-False-None]	1.7701ms	1.6834ms	594.0210 Ops/s	577.0158 Ops/s	$\color{#35bf28}+2.95\%$
test_to[True-False-None]	1.5043ms	1.2637ms	791.3137 Ops/s	774.4353 Ops/s	$\color{#35bf28}+2.18\%$
test_to[within-False-None]	0.2916s	5.1795ms	193.0690 Ops/s	252.9487 Ops/s	$\textbf{\color{#d91a1a}-23.67\%}$
test_to[True-default-None]	5.3816ms	5.0941ms	196.3055 Ops/s	191.8540 Ops/s	$\color{#35bf28}+2.32\%$
test_to_njt[False-False-None]	6.9438ms	6.8324ms	146.3625 Ops/s	145.3686 Ops/s	$\color{#35bf28}+0.68\%$
test_to_njt[True-False-None]	6.0032ms	5.4224ms	184.4185 Ops/s	183.3660 Ops/s	$\color{#35bf28}+0.57\%$
test_to_njt[within-False-None]	11.9939ms	11.7700ms	84.9617 Ops/s	84.8540 Ops/s	$\color{#35bf28}+0.13\%$
test_creation[device0]	0.5800ms	78.5332μs	12.7335 KOps/s	12.4178 KOps/s	$\color{#35bf28}+2.54\%$
test_creation_from_tensor	0.4699ms	80.6563μs	12.3983 KOps/s	12.1472 KOps/s	$\color{#35bf28}+2.07\%$
test_add_one[memmap_tensor0]	0.2901ms	6.6587μs	150.1803 KOps/s	148.2092 KOps/s	$\color{#35bf28}+1.33\%$
test_contiguous[memmap_tensor0]	1.7695μs	0.3932μs	2.5434 MOps/s	2.5738 MOps/s	$\color{#d91a1a}-1.18\%$
test_stack[memmap_tensor0]	41.4110μs	4.1203μs	242.6994 KOps/s	229.3789 KOps/s	$\textbf{\color{#35bf28}+5.81\%}$
test_memmaptd_index	1.9910ms	0.2453ms	4.0772 KOps/s	4.0022 KOps/s	$\color{#35bf28}+1.87\%$
test_memmaptd_index_astensor	0.9649ms	0.2960ms	3.3784 KOps/s	3.2627 KOps/s	$\color{#35bf28}+3.55\%$
test_memmaptd_index_op	1.1045ms	0.5639ms	1.7732 KOps/s	1.7190 KOps/s	$\color{#35bf28}+3.16\%$
test_serialize_model	0.1315s	0.1300s	7.6931 Ops/s	7.7262 Ops/s	$\color{#d91a1a}-0.43\%$
test_serialize_model_pickle	1.3465s	1.1868s	0.8426 Ops/s	0.8252 Ops/s	$\color{#35bf28}+2.11\%$
test_serialize_weights	0.1305s	0.1298s	7.7058 Ops/s	7.7287 Ops/s	$\color{#d91a1a}-0.30\%$
test_serialize_weights_returnearly	48.9859ms	39.5732ms	25.2696 Ops/s	23.3636 Ops/s	$\textbf{\color{#35bf28}+8.16\%}$
test_serialize_weights_pickle	1.3500s	1.2128s	0.8245 Ops/s	0.8210 Ops/s	$\color{#35bf28}+0.43\%$
test_reshape_pytree	62.3710μs	21.4635μs	46.5908 KOps/s	46.0688 KOps/s	$\color{#35bf28}+1.13\%$
test_reshape_td	91.0510μs	25.9026μs	38.6061 KOps/s	38.5314 KOps/s	$\color{#35bf28}+0.19\%$
test_view_pytree	0.1589ms	21.4921μs	46.5287 KOps/s	46.6435 KOps/s	$\color{#d91a1a}-0.25\%$
test_view_td	60.9310μs	29.4466μs	33.9598 KOps/s	32.7994 KOps/s	$\color{#35bf28}+3.54\%$
test_unbind_pytree	51.9810μs	27.3246μs	36.5971 KOps/s	36.0964 KOps/s	$\color{#35bf28}+1.39\%$
test_unbind_td	0.5978ms	35.4813μs	28.1838 KOps/s	28.1339 KOps/s	$\color{#35bf28}+0.18\%$
test_split_pytree	56.1910μs	28.9911μs	34.4933 KOps/s	34.5595 KOps/s	$\color{#d91a1a}-0.19\%$
test_split_td	0.1843ms	36.7004μs	27.2476 KOps/s	26.4372 KOps/s	$\color{#35bf28}+3.07\%$
test_add_pytree	0.1194ms	33.1364μs	30.1783 KOps/s	29.5141 KOps/s	$\color{#35bf28}+2.25\%$
test_add_td	88.8310μs	44.7957μs	22.3236 KOps/s	23.3904 KOps/s	$\color{#d91a1a}-4.56\%$
test_compile_add_one_nested[tensordict-compile]	0.1707ms	0.1180ms	8.4730 KOps/s	8.0876 KOps/s	$\color{#35bf28}+4.77\%$
test_compile_add_one_nested[tensordict-eager]	0.2130ms	0.1215ms	8.2295 KOps/s	8.0471 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_add_one_nested[pytree-compile]	0.1326ms	94.5347μs	10.5781 KOps/s	10.2434 KOps/s	$\color{#35bf28}+3.27\%$
test_compile_add_one_nested[pytree-eager]	1.0952ms	0.1479ms	6.7615 KOps/s	6.6079 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_copy_nested[tensordict-compile]	55.8410μs	22.7213μs	44.0115 KOps/s	43.8482 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_copy_nested[tensordict-eager]	85.1220μs	25.8927μs	38.6209 KOps/s	37.7770 KOps/s	$\color{#35bf28}+2.23\%$
test_compile_copy_nested[pytree-compile]	0.4357ms	63.3612μs	15.7825 KOps/s	15.5856 KOps/s	$\color{#35bf28}+1.26\%$
test_compile_copy_nested[pytree-eager]	87.2710μs	48.9006μs	20.4496 KOps/s	20.1436 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_add_one_flat[tensordict-compile]	0.1876ms	0.1422ms	7.0339 KOps/s	6.7613 KOps/s	$\color{#35bf28}+4.03\%$
test_compile_add_one_flat[tensordict-eager]	0.3169ms	0.2088ms	4.7898 KOps/s	4.8140 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_add_one_flat[tensorclass-compile]	0.1431ms	99.8145μs	10.0186 KOps/s	10.2668 KOps/s	$\color{#d91a1a}-2.42\%$
test_compile_add_one_flat[tensorclass-eager]	0.1382ms	50.0036μs	19.9985 KOps/s	19.7863 KOps/s	$\color{#35bf28}+1.07\%$
test_compile_add_one_flat[pytree-compile]	0.1903ms	0.1360ms	7.3546 KOps/s	7.2883 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_add_one_flat[pytree-eager]	0.5817ms	0.4787ms	2.0891 KOps/s	2.0520 KOps/s	$\color{#35bf28}+1.81\%$
test_compile_add_self_flat[tensordict-eager]	0.3682ms	0.2452ms	4.0787 KOps/s	4.0493 KOps/s	$\color{#35bf28}+0.73\%$
test_compile_add_self_flat[tensordict-compile]	0.2476ms	0.1470ms	6.8010 KOps/s	6.9421 KOps/s	$\color{#d91a1a}-2.03\%$
test_compile_add_self_flat[tensorclass-eager]	0.1757ms	61.4333μs	16.2778 KOps/s	16.4469 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_add_self_flat[tensorclass-compile]	0.1493ms	98.2023μs	10.1831 KOps/s	10.1389 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_self_flat[pytree-eager]	0.4874ms	0.4054ms	2.4667 KOps/s	2.4174 KOps/s	$\color{#35bf28}+2.04\%$
test_compile_add_self_flat[pytree-compile]	0.1725ms	0.1370ms	7.3008 KOps/s	7.4126 KOps/s	$\color{#d91a1a}-1.51\%$
test_compile_copy_flat[tensordict-compile]	53.9710μs	18.2816μs	54.6998 KOps/s	42.5452 KOps/s	$\textbf{\color{#35bf28}+28.57\%}$
test_compile_copy_flat[tensordict-eager]	75.2110μs	26.7448μs	37.3904 KOps/s	37.8889 KOps/s	$\color{#d91a1a}-1.32\%$
test_compile_copy_flat[pytree-compile]	0.1020ms	69.2380μs	14.4429 KOps/s	14.6107 KOps/s	$\color{#d91a1a}-1.15\%$
test_compile_copy_flat[pytree-eager]	96.1120μs	50.7765μs	19.6942 KOps/s	19.5742 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_assign_and_add[tensordict-compile]	1.6007ms	0.3882ms	2.5761 KOps/s	2.2411 KOps/s	$\textbf{\color{#35bf28}+14.95\%}$
test_compile_assign_and_add[tensordict-eager]	2.6994ms	2.6022ms	384.2870 Ops/s	380.3197 Ops/s	$\color{#35bf28}+1.04\%$
test_compile_assign_and_add[pytree-compile]	1.5597ms	0.3744ms	2.6712 KOps/s	2.2721 KOps/s	$\textbf{\color{#35bf28}+17.56\%}$
test_compile_assign_and_add[pytree-eager]	2.7505ms	2.6580ms	376.2240 Ops/s	372.5106 Ops/s	$\color{#35bf28}+1.00\%$
test_compile_indexing[tensor-tensordict-compile]	0.6549ms	0.1132ms	8.8309 KOps/s	8.7207 KOps/s	$\color{#35bf28}+1.26\%$
test_compile_indexing[tensor-tensordict-eager]	0.5513ms	78.6618μs	12.7126 KOps/s	12.6444 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_indexing[tensor-tensorclass-compile]	0.3251ms	0.1064ms	9.3970 KOps/s	8.9838 KOps/s	$\color{#35bf28}+4.60\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1054ms	67.2123μs	14.8782 KOps/s	14.2384 KOps/s	$\color{#35bf28}+4.49\%$
test_compile_indexing[tensor-pytree-compile]	0.1668ms	0.1067ms	9.3745 KOps/s	8.9425 KOps/s	$\color{#35bf28}+4.83\%$
test_compile_indexing[tensor-pytree-eager]	0.1061ms	67.6097μs	14.7908 KOps/s	14.1630 KOps/s	$\color{#35bf28}+4.43\%$
test_compile_indexing[slice-tensordict-compile]	0.1394ms	99.1323μs	10.0875 KOps/s	9.9367 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_indexing[slice-tensordict-eager]	0.1420ms	16.5953μs	60.2580 KOps/s	59.0751 KOps/s	$\color{#35bf28}+2.00\%$
test_compile_indexing[slice-tensorclass-compile]	0.1379ms	94.2783μs	10.6069 KOps/s	10.4424 KOps/s	$\color{#35bf28}+1.58\%$
test_compile_indexing[slice-tensorclass-eager]	50.4310μs	15.4939μs	64.5417 KOps/s	64.1663 KOps/s	$\color{#35bf28}+0.59\%$
test_compile_indexing[slice-pytree-compile]	0.2237ms	95.7753μs	10.4411 KOps/s	10.4505 KOps/s	$\color{#d91a1a}-0.09\%$
test_compile_indexing[slice-pytree-eager]	51.2110μs	15.3505μs	65.1443 KOps/s	62.1145 KOps/s	$\color{#35bf28}+4.88\%$
test_compile_indexing[int-tensordict-compile]	0.1674ms	99.7366μs	10.0264 KOps/s	9.8966 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_indexing[int-tensordict-eager]	0.5985ms	16.4235μs	60.8885 KOps/s	58.9186 KOps/s	$\color{#35bf28}+3.34\%$
test_compile_indexing[int-tensorclass-compile]	0.1543ms	95.1127μs	10.5138 KOps/s	10.2986 KOps/s	$\color{#35bf28}+2.09\%$
test_compile_indexing[int-tensorclass-eager]	47.3210μs	15.3174μs	65.2852 KOps/s	64.7845 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_indexing[int-pytree-compile]	0.1367ms	95.5803μs	10.4624 KOps/s	10.4983 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_indexing[int-pytree-eager]	51.8610μs	15.2492μs	65.5774 KOps/s	64.7428 KOps/s	$\color{#35bf28}+1.29\%$
test_mod_add[eager]	78.5210μs	36.4018μs	27.4712 KOps/s	27.1484 KOps/s	$\color{#35bf28}+1.19\%$
test_mod_add[compile]	0.1611ms	84.7502μs	11.7994 KOps/s	12.5228 KOps/s	$\textbf{\color{#d91a1a}-5.78\%}$
test_mod_add[compile-overhead]	0.3309ms	0.1667ms	6.0006 KOps/s	5.7554 KOps/s	$\color{#35bf28}+4.26\%$
test_mod_wrap[eager]	0.3863ms	0.2466ms	4.0559 KOps/s	3.9688 KOps/s	$\color{#35bf28}+2.19\%$
test_mod_wrap[compile]	0.3456ms	0.2798ms	3.5734 KOps/s	3.4930 KOps/s	$\color{#35bf28}+2.30\%$
test_mod_wrap[compile-overhead]	7.0751ms	3.7633ms	265.7278 Ops/s	265.1473 Ops/s	$\color{#35bf28}+0.22\%$
test_mod_wrap_and_backward[eager]	1.4610ms	1.3574ms	736.6808 Ops/s	677.5424 Ops/s	$\textbf{\color{#35bf28}+8.73\%}$
test_mod_wrap_and_backward[compile]	1.3851ms	1.2544ms	797.1666 Ops/s	721.8783 Ops/s	$\textbf{\color{#35bf28}+10.43\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3576ms	0.9250ms	1.0811 KOps/s	981.8195 Ops/s	$\textbf{\color{#35bf28}+10.11\%}$
test_seq_add[eager]	0.1638ms	0.1144ms	8.7401 KOps/s	8.7650 KOps/s	$\color{#d91a1a}-0.28\%$
test_seq_add[compile]	0.1684ms	90.3926μs	11.0628 KOps/s	11.1204 KOps/s	$\color{#d91a1a}-0.52\%$
test_seq_add[compile-overhead]	0.1692ms	0.1274ms	7.8500 KOps/s	7.9276 KOps/s	$\color{#d91a1a}-0.98\%$
test_seq_wrap[eager]	0.4855ms	0.4091ms	2.4441 KOps/s	2.3315 KOps/s	$\color{#35bf28}+4.83\%$
test_seq_wrap[compile]	0.4015ms	0.2968ms	3.3687 KOps/s	3.3143 KOps/s	$\color{#35bf28}+1.64\%$
test_seq_wrap[compile-overhead]	0.2749ms	0.2225ms	4.4941 KOps/s	4.4408 KOps/s	$\color{#35bf28}+1.20\%$
test_func_call_runtime[False-eager]	0.8028ms	0.7386ms	1.3539 KOps/s	1.3103 KOps/s	$\color{#35bf28}+3.32\%$
test_func_call_runtime[False-compile]	1.0303ms	0.7318ms	1.3665 KOps/s	1.3304 KOps/s	$\color{#35bf28}+2.72\%$
test_func_call_runtime[False-compile-overhead]	0.4199ms	0.3558ms	2.8105 KOps/s	2.8011 KOps/s	$\color{#35bf28}+0.33\%$
test_func_call_runtime[True-eager]	0.9855ms	0.8990ms	1.1123 KOps/s	1.0939 KOps/s	$\color{#35bf28}+1.68\%$
test_func_call_runtime[True-compile]	0.8184ms	0.7467ms	1.3393 KOps/s	1.3004 KOps/s	$\color{#35bf28}+2.99\%$
test_func_call_runtime[True-compile-overhead]	0.4509ms	0.3758ms	2.6607 KOps/s	2.6618 KOps/s	$\color{#d91a1a}-0.04\%$
test_func_call_cm_runtime[False-eager]	0.8217ms	0.7662ms	1.3051 KOps/s	1.3504 KOps/s	$\color{#d91a1a}-3.35\%$
test_func_call_cm_runtime[False-compile]	0.7950ms	0.7318ms	1.3664 KOps/s	1.3302 KOps/s	$\color{#35bf28}+2.72\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4034ms	0.3574ms	2.7983 KOps/s	2.7942 KOps/s	$\color{#35bf28}+0.15\%$
test_func_call_cm_runtime[True-eager]	1.0974ms	0.9972ms	1.0028 KOps/s	988.8271 Ops/s	$\color{#35bf28}+1.41\%$
test_func_call_cm_runtime[True-compile]	0.9241ms	0.7798ms	1.2824 KOps/s	1.2512 KOps/s	$\color{#35bf28}+2.49\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4545ms	0.4015ms	2.4905 KOps/s	2.4800 KOps/s	$\color{#35bf28}+0.42\%$
test_vmap_func_call_cm_runtime[eager]	2.5265ms	2.0816ms	480.3954 Ops/s	477.3523 Ops/s	$\color{#35bf28}+0.64\%$
test_vmap_func_call_cm_runtime[compile]	0.8672ms	0.7946ms	1.2585 KOps/s	1.2291 KOps/s	$\color{#35bf28}+2.39\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4743ms	0.4037ms	2.4772 KOps/s	2.4675 KOps/s	$\color{#35bf28}+0.39\%$
test_distributed	2.8188ms	0.2105ms	4.7499 KOps/s	8.8704 KOps/s	$\textbf{\color{#d91a1a}-46.45\%}$
test_tdmodule	54.7310μs	19.3627μs	51.6457 KOps/s	51.6601 KOps/s	$\color{#d91a1a}-0.03\%$
test_tdmodule_dispatch	88.7610μs	34.9961μs	28.5746 KOps/s	29.3002 KOps/s	$\color{#d91a1a}-2.48\%$
test_tdseq	43.2710μs	19.2796μs	51.8683 KOps/s	52.0132 KOps/s	$\color{#d91a1a}-0.28\%$
test_tdseq_dispatch	60.6910μs	36.6242μs	27.3044 KOps/s	27.2952 KOps/s	$\color{#35bf28}+0.03\%$
test_instantiation_functorch	1.7101ms	1.4932ms	669.7001 Ops/s	648.6043 Ops/s	$\color{#35bf28}+3.25\%$
test_exec_functorch	0.1910ms	0.1375ms	7.2711 KOps/s	6.9379 KOps/s	$\color{#35bf28}+4.80\%$
test_exec_functional_call	0.1752ms	0.1308ms	7.6447 KOps/s	7.2072 KOps/s	$\textbf{\color{#35bf28}+6.07\%}$
test_exec_td_decorator	0.3614ms	0.1781ms	5.6138 KOps/s	5.4537 KOps/s	$\color{#35bf28}+2.94\%$
test_vmap_mlp_speed_decorator[True-True]	0.7795ms	0.7008ms	1.4269 KOps/s	1.4623 KOps/s	$\color{#d91a1a}-2.42\%$
test_vmap_mlp_speed_decorator[True-False]	0.8781ms	0.6818ms	1.4666 KOps/s	1.4651 KOps/s	$\color{#35bf28}+0.11\%$
test_vmap_mlp_speed_decorator[False-True]	0.7975ms	0.6019ms	1.6615 KOps/s	1.6761 KOps/s	$\color{#d91a1a}-0.87\%$
test_vmap_mlp_speed_decorator[False-False]	0.7282ms	0.6115ms	1.6354 KOps/s	1.6761 KOps/s	$\color{#d91a1a}-2.43\%$
test_vmap_transformer_speed_decorator[True-True]	20.1032ms	19.2628ms	51.9135 Ops/s	51.8116 Ops/s	$\color{#35bf28}+0.20\%$
test_vmap_transformer_speed_decorator[True-False]	20.2065ms	19.7039ms	50.7513 Ops/s	51.8385 Ops/s	$\color{#d91a1a}-2.10\%$
test_vmap_transformer_speed_decorator[False-True]	20.0857ms	19.3097ms	51.7874 Ops/s	52.1480 Ops/s	$\color{#d91a1a}-0.69\%$
test_vmap_transformer_speed_decorator[False-False]	20.1365ms	19.2304ms	52.0011 Ops/s	52.1280 Ops/s	$\color{#d91a1a}-0.24\%$
test_to_module_speed[True]	1.0406ms	0.9265ms	1.0793 KOps/s	1.0662 KOps/s	$\color{#35bf28}+1.23\%$
test_to_module_speed[False]	1.2143ms	0.9072ms	1.1023 KOps/s	1.0866 KOps/s	$\color{#35bf28}+1.45\%$
test_tc_init	67.1310μs	33.5903μs	29.7705 KOps/s	28.1818 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_tc_init_nested	0.1531ms	68.3522μs	14.6301 KOps/s	13.4406 KOps/s	$\textbf{\color{#35bf28}+8.85\%}$
test_tc_first_layer_tensor	3.8716μs	0.6881μs	1.4533 MOps/s	1.4434 MOps/s	$\color{#35bf28}+0.69\%$
test_tc_first_layer_nontensor	22.7010μs	2.2866μs	437.3248 KOps/s	435.1903 KOps/s	$\color{#35bf28}+0.49\%$
test_tc_second_layer_tensor	6.6100μs	1.3902μs	719.3031 KOps/s	708.7440 KOps/s	$\color{#35bf28}+1.49\%$
test_tc_second_layer_nontensor	20.9700μs	3.0327μs	329.7435 KOps/s	328.0304 KOps/s	$\color{#35bf28}+0.52\%$
test_unbind	0.2297s	10.0905ms	99.1031 Ops/s	151.8510 Ops/s	$\textbf{\color{#d91a1a}-34.74\%}$
test_full_like	9.3884ms	9.1746ms	108.9968 Ops/s	108.3733 Ops/s	$\color{#35bf28}+0.58\%$
test_zeros_like	4.8671ms	4.3224ms	231.3554 Ops/s	113.9843 Ops/s	$\textbf{\color{#35bf28}+102.97\%}$
test_ones_like	5.4575ms	4.3274ms	231.0874 Ops/s	235.6219 Ops/s	$\color{#d91a1a}-1.92\%$
test_clone	6.6878ms	6.4400ms	155.2798 Ops/s	155.2020 Ops/s	$\color{#35bf28}+0.05\%$
test_squeeze	56.7010μs	9.3882μs	106.5163 KOps/s	111.4314 KOps/s	$\color{#d91a1a}-4.41\%$
test_unsqueeze	0.2137ms	70.3517μs	14.2143 KOps/s	14.5195 KOps/s	$\color{#d91a1a}-2.10\%$
test_split	0.2594ms	0.1524ms	6.5638 KOps/s	6.4092 KOps/s	$\color{#35bf28}+2.41\%$
test_permute	0.2203ms	0.1721ms	5.8117 KOps/s	5.7378 KOps/s	$\color{#35bf28}+1.29\%$
test_stack	50.7144ms	50.4273ms	19.8305 Ops/s	19.8731 Ops/s	$\color{#d91a1a}-0.21\%$
test_cat	53.5734ms	50.8799ms	19.6541 Ops/s	19.9377 Ops/s	$\color{#d91a1a}-1.42\%$