[Feature] Optional in_keys for WrapModule

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] Force log_prob to return a tensordict when kwargs are passed to ProbabilisticTensorDictSequential.log_prob #1146
• -> [Feature] Optional in_keys for WrapModule #1145

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}23$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	53.4200μs	21.2080μs	47.1521 KOps/s	49.8487 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_plain_set_stack_nested	62.1960μs	21.0640μs	47.4744 KOps/s	49.0533 KOps/s	$\color{#d91a1a}-3.22\%$
test_plain_set_nested_inplace	58.3590μs	22.9718μs	43.5316 KOps/s	45.8462 KOps/s	$\textbf{\color{#d91a1a}-5.05\%}$
test_plain_set_stack_nested_inplace	58.0890μs	23.0160μs	43.4480 KOps/s	45.6165 KOps/s	$\color{#d91a1a}-4.75\%$
test_items	34.2740μs	4.1728μs	239.6463 KOps/s	237.6867 KOps/s	$\color{#35bf28}+0.82\%$
test_items_nested	0.6252ms	0.4030ms	2.4812 KOps/s	2.4505 KOps/s	$\color{#35bf28}+1.26\%$
test_items_nested_locked	0.5623ms	0.4054ms	2.4664 KOps/s	2.4532 KOps/s	$\color{#35bf28}+0.54\%$
test_items_nested_leaf	0.1462ms	77.2844μs	12.9392 KOps/s	12.7552 KOps/s	$\color{#35bf28}+1.44\%$
test_items_stack_nested	0.5856ms	0.4134ms	2.4188 KOps/s	2.4149 KOps/s	$\color{#35bf28}+0.16\%$
test_items_stack_nested_leaf	0.1740ms	81.5213μs	12.2667 KOps/s	12.4101 KOps/s	$\color{#d91a1a}-1.16\%$
test_items_stack_nested_locked	0.7953ms	0.4124ms	2.4249 KOps/s	2.4052 KOps/s	$\color{#35bf28}+0.82\%$
test_keys	27.5620μs	3.4968μs	285.9737 KOps/s	277.2323 KOps/s	$\color{#35bf28}+3.15\%$
test_keys_nested	0.2290ms	0.1659ms	6.0289 KOps/s	5.9484 KOps/s	$\color{#35bf28}+1.35\%$
test_keys_nested_locked	1.9598ms	0.1738ms	5.7538 KOps/s	5.6985 KOps/s	$\color{#35bf28}+0.97\%$
test_keys_nested_leaf	0.2101ms	0.1446ms	6.9159 KOps/s	6.7777 KOps/s	$\color{#35bf28}+2.04\%$
test_keys_stack_nested	0.2290ms	0.1639ms	6.1006 KOps/s	5.9592 KOps/s	$\color{#35bf28}+2.37\%$
test_keys_stack_nested_leaf	0.1972ms	0.1398ms	7.1552 KOps/s	6.8665 KOps/s	$\color{#35bf28}+4.21\%$
test_keys_stack_nested_locked	0.2344ms	0.1694ms	5.9021 KOps/s	5.8107 KOps/s	$\color{#35bf28}+1.57\%$
test_values	11.1728μs	1.0473μs	954.8002 KOps/s	964.7260 KOps/s	$\color{#d91a1a}-1.03\%$
test_values_nested	0.1356ms	63.1094μs	15.8455 KOps/s	15.7405 KOps/s	$\color{#35bf28}+0.67\%$
test_values_nested_locked	0.1175ms	62.8360μs	15.9144 KOps/s	15.9428 KOps/s	$\color{#d91a1a}-0.18\%$
test_values_nested_leaf	0.1329ms	72.7534μs	13.7451 KOps/s	13.7526 KOps/s	$\color{#d91a1a}-0.06\%$
test_values_stack_nested	0.1227ms	64.1778μs	15.5817 KOps/s	15.8333 KOps/s	$\color{#d91a1a}-1.59\%$
test_values_stack_nested_leaf	0.1360ms	71.1261μs	14.0595 KOps/s	13.8056 KOps/s	$\color{#35bf28}+1.84\%$
test_values_stack_nested_locked	0.1358ms	64.5571μs	15.4902 KOps/s	15.7785 KOps/s	$\color{#d91a1a}-1.83\%$
test_membership	20.1580μs	0.8899μs	1.1237 MOps/s	1.1683 MOps/s	$\color{#d91a1a}-3.82\%$
test_membership_nested	48.1000μs	2.9116μs	343.4531 KOps/s	343.7016 KOps/s	$\color{#d91a1a}-0.07\%$
test_membership_nested_leaf	65.8950μs	2.9693μs	336.7836 KOps/s	338.4976 KOps/s	$\color{#d91a1a}-0.51\%$
test_membership_stacked_nested	44.5840μs	2.9189μs	342.5953 KOps/s	345.8450 KOps/s	$\color{#d91a1a}-0.94\%$
test_membership_stacked_nested_leaf	25.3380μs	2.9388μs	340.2803 KOps/s	349.6691 KOps/s	$\color{#d91a1a}-2.69\%$
test_membership_nested_last	68.2260μs	4.4303μs	225.7195 KOps/s	230.6200 KOps/s	$\color{#d91a1a}-2.12\%$
test_membership_nested_leaf_last	42.6900μs	4.3800μs	228.3127 KOps/s	223.0888 KOps/s	$\color{#35bf28}+2.34\%$
test_membership_stacked_nested_last	69.7610μs	13.5264μs	73.9296 KOps/s	142.2617 KOps/s	$\textbf{\color{#d91a1a}-48.03\%}$
test_membership_stacked_nested_leaf_last	42.8900μs	13.7015μs	72.9845 KOps/s	142.4671 KOps/s	$\textbf{\color{#d91a1a}-48.77\%}$
test_nested_getleaf	61.5750μs	10.9446μs	91.3693 KOps/s	91.3074 KOps/s	$\color{#35bf28}+0.07\%$
test_nested_get	66.5550μs	10.3330μs	96.7774 KOps/s	96.2999 KOps/s	$\color{#35bf28}+0.50\%$
test_stacked_getleaf	39.6140μs	10.5214μs	95.0441 KOps/s	91.7525 KOps/s	$\color{#35bf28}+3.59\%$
test_stacked_get	69.7110μs	10.1120μs	98.8922 KOps/s	96.7135 KOps/s	$\color{#35bf28}+2.25\%$
test_nested_getitemleaf	62.4770μs	11.3207μs	88.3338 KOps/s	84.9023 KOps/s	$\color{#35bf28}+4.04\%$
test_nested_getitem	51.9870μs	10.5773μs	94.5421 KOps/s	93.5466 KOps/s	$\color{#35bf28}+1.06\%$
test_stacked_getitemleaf	62.4970μs	11.2121μs	89.1890 KOps/s	88.1993 KOps/s	$\color{#35bf28}+1.12\%$
test_stacked_getitem	83.5660μs	10.7589μs	92.9466 KOps/s	93.6851 KOps/s	$\color{#d91a1a}-0.79\%$
test_lock_nested	4.2331ms	0.4598ms	2.1748 KOps/s	2.1520 KOps/s	$\color{#35bf28}+1.06\%$
test_lock_stack_nested	0.8224ms	0.4136ms	2.4176 KOps/s	2.3572 KOps/s	$\color{#35bf28}+2.56\%$
test_unlock_nested	1.1670ms	0.3752ms	2.6649 KOps/s	2.6284 KOps/s	$\color{#35bf28}+1.39\%$
test_unlock_stack_nested	0.4240ms	0.3317ms	3.0145 KOps/s	2.8941 KOps/s	$\color{#35bf28}+4.16\%$
test_flatten_speed	0.1809ms	99.5365μs	10.0466 KOps/s	9.9624 KOps/s	$\color{#35bf28}+0.85\%$
test_unflatten_speed	0.7485ms	0.5222ms	1.9150 KOps/s	1.8627 KOps/s	$\color{#35bf28}+2.81\%$
test_common_ops	1.8182ms	0.8152ms	1.2267 KOps/s	1.3385 KOps/s	$\textbf{\color{#d91a1a}-8.36\%}$
test_creation	27.9020μs	2.4800μs	403.2288 KOps/s	402.1582 KOps/s	$\color{#35bf28}+0.27\%$
test_creation_empty	62.3770μs	12.4490μs	80.3279 KOps/s	102.8608 KOps/s	$\textbf{\color{#d91a1a}-21.91\%}$
test_creation_nested_1	48.1000μs	15.6082μs	64.0688 KOps/s	79.6168 KOps/s	$\textbf{\color{#d91a1a}-19.53\%}$
test_creation_nested_2	88.3760μs	19.9924μs	50.0190 KOps/s	58.5564 KOps/s	$\textbf{\color{#d91a1a}-14.58\%}$
test_clone	0.1026ms	13.1977μs	75.7708 KOps/s	75.5678 KOps/s	$\color{#35bf28}+0.27\%$
test_getitem[int]	1.2997ms	12.8127μs	78.0477 KOps/s	78.2177 KOps/s	$\color{#d91a1a}-0.22\%$
test_getitem[slice_int]	0.1593ms	23.8181μs	41.9848 KOps/s	39.3122 KOps/s	$\textbf{\color{#35bf28}+6.80\%}$
test_getitem[range]	0.1912ms	47.2779μs	21.1515 KOps/s	20.4098 KOps/s	$\color{#35bf28}+3.63\%$
test_getitem[tuple]	0.1566ms	20.1929μs	49.5223 KOps/s	48.8507 KOps/s	$\color{#35bf28}+1.37\%$
test_getitem[list]	0.3165ms	42.3696μs	23.6018 KOps/s	22.4169 KOps/s	$\textbf{\color{#35bf28}+5.29\%}$
test_setitem_dim[int]	47.1590μs	24.3308μs	41.1002 KOps/s	40.0613 KOps/s	$\color{#35bf28}+2.59\%$
test_setitem_dim[slice_int]	89.5880μs	50.2568μs	19.8978 KOps/s	19.4487 KOps/s	$\color{#35bf28}+2.31\%$
test_setitem_dim[range]	0.1209ms	71.4498μs	13.9958 KOps/s	13.6354 KOps/s	$\color{#35bf28}+2.64\%$
test_setitem_dim[tuple]	77.2150μs	39.8297μs	25.1069 KOps/s	24.4687 KOps/s	$\color{#35bf28}+2.61\%$
test_setitem	0.1189ms	20.5987μs	48.5467 KOps/s	52.6495 KOps/s	$\textbf{\color{#d91a1a}-7.79\%}$
test_set	0.1358ms	20.2074μs	49.4868 KOps/s	51.8705 KOps/s	$\color{#d91a1a}-4.60\%$
test_set_shared	2.5452ms	0.1751ms	5.7122 KOps/s	5.9207 KOps/s	$\color{#d91a1a}-3.52\%$
test_update	0.2549ms	23.8592μs	41.9125 KOps/s	49.3226 KOps/s	$\textbf{\color{#d91a1a}-15.02\%}$
test_update_nested	0.1436ms	34.0149μs	29.3989 KOps/s	33.2086 KOps/s	$\textbf{\color{#d91a1a}-11.47\%}$
test_update__nested	0.4199ms	33.7299μs	29.6473 KOps/s	30.1091 KOps/s	$\color{#d91a1a}-1.53\%$
test_set_nested	0.1358ms	22.6482μs	44.1535 KOps/s	48.2192 KOps/s	$\textbf{\color{#d91a1a}-8.43\%}$
test_set_nested_new	0.1483ms	26.8116μs	37.2972 KOps/s	40.2961 KOps/s	$\textbf{\color{#d91a1a}-7.44\%}$
test_select	0.2157ms	44.0481μs	22.7025 KOps/s	24.1732 KOps/s	$\textbf{\color{#d91a1a}-6.08\%}$
test_select_nested	0.1313ms	63.9699μs	15.6324 KOps/s	15.7195 KOps/s	$\color{#d91a1a}-0.55\%$
test_exclude_nested	0.2828ms	85.0184μs	11.7622 KOps/s	12.0393 KOps/s	$\color{#d91a1a}-2.30\%$
test_empty[True]	1.0505ms	0.4226ms	2.3661 KOps/s	2.4021 KOps/s	$\color{#d91a1a}-1.50\%$
test_empty[False]	42.8527μs	1.3889μs	719.9804 KOps/s	699.4656 KOps/s	$\color{#35bf28}+2.93\%$
test_unbind_speed	0.3876ms	0.2695ms	3.7112 KOps/s	3.7441 KOps/s	$\color{#d91a1a}-0.88\%$
test_unbind_speed_stack0	0.3662ms	0.2571ms	3.8889 KOps/s	3.8129 KOps/s	$\color{#35bf28}+1.99\%$
test_unbind_speed_stack1	0.1167s	0.7904ms	1.2652 KOps/s	1.3656 KOps/s	$\textbf{\color{#d91a1a}-7.35\%}$
test_split	0.1134s	1.7580ms	568.8325 Ops/s	555.2450 Ops/s	$\color{#35bf28}+2.45\%$
test_chunk	0.1156s	1.7810ms	561.4812 Ops/s	559.0630 Ops/s	$\color{#35bf28}+0.43\%$
test_consolidate_njt[False-None]	8.6018ms	8.0850ms	123.6853 Ops/s	121.0131 Ops/s	$\color{#35bf28}+2.21\%$
test_creation[device0]	0.2682ms	91.1498μs	10.9709 KOps/s	10.8524 KOps/s	$\color{#35bf28}+1.09\%$
test_creation_from_tensor	0.3200ms	94.0119μs	10.6370 KOps/s	10.3472 KOps/s	$\color{#35bf28}+2.80\%$
test_add_one[memmap_tensor0]	0.2997ms	4.7073μs	212.4341 KOps/s	212.7025 KOps/s	$\color{#d91a1a}-0.13\%$
test_contiguous[memmap_tensor0]	39.9140μs	0.5127μs	1.9503 MOps/s	1.9862 MOps/s	$\color{#d91a1a}-1.81\%$
test_stack[memmap_tensor0]	46.0970μs	3.2904μs	303.9150 KOps/s	305.8107 KOps/s	$\color{#d91a1a}-0.62\%$
test_memmaptd_index	0.4904ms	0.2347ms	4.2604 KOps/s	4.3193 KOps/s	$\color{#d91a1a}-1.36\%$
test_memmaptd_index_astensor	0.7669ms	0.3221ms	3.1042 KOps/s	3.1202 KOps/s	$\color{#d91a1a}-0.51\%$
test_memmaptd_index_op	1.0248ms	0.5972ms	1.6744 KOps/s	1.8288 KOps/s	$\textbf{\color{#d91a1a}-8.44\%}$
test_serialize_model	0.1358s	0.1179s	8.4830 Ops/s	8.4269 Ops/s	$\color{#35bf28}+0.67\%$
test_serialize_model_pickle	0.4988s	0.4004s	2.4977 Ops/s	2.4571 Ops/s	$\color{#35bf28}+1.65\%$
test_serialize_weights	0.2413s	0.1349s	7.4125 Ops/s	7.5447 Ops/s	$\color{#d91a1a}-1.75\%$
test_serialize_weights_returnearly	0.2635s	0.1774s	5.6383 Ops/s	6.4035 Ops/s	$\textbf{\color{#d91a1a}-11.95\%}$
test_serialize_weights_pickle	0.4951s	0.4083s	2.4493 Ops/s	2.1850 Ops/s	$\textbf{\color{#35bf28}+12.10\%}$
test_serialize_weights_filesystem	0.1523s	0.1444s	6.9251 Ops/s	7.0915 Ops/s	$\color{#d91a1a}-2.35\%$
test_serialize_model_filesystem	0.2657s	0.1729s	5.7833 Ops/s	5.9112 Ops/s	$\color{#d91a1a}-2.16\%$
test_reshape_pytree	66.0140μs	26.4369μs	37.8260 KOps/s	37.7521 KOps/s	$\color{#35bf28}+0.20\%$
test_reshape_td	82.1540μs	32.4669μs	30.8006 KOps/s	29.6762 KOps/s	$\color{#35bf28}+3.79\%$
test_view_pytree	58.4390μs	26.8560μs	37.2357 KOps/s	38.0254 KOps/s	$\color{#d91a1a}-2.08\%$
test_view_td	81.5930μs	37.7093μs	26.5186 KOps/s	26.4128 KOps/s	$\color{#35bf28}+0.40\%$
test_unbind_pytree	72.1450μs	29.7949μs	33.5627 KOps/s	34.0393 KOps/s	$\color{#d91a1a}-1.40\%$
test_unbind_td	0.3442ms	39.3971μs	25.3826 KOps/s	25.3867 KOps/s	$\color{#d91a1a}-0.02\%$
test_split_pytree	78.6170μs	29.2448μs	34.1941 KOps/s	34.3820 KOps/s	$\color{#d91a1a}-0.55\%$
test_split_td	0.5630ms	44.5585μs	22.4424 KOps/s	22.2997 KOps/s	$\color{#35bf28}+0.64\%$
test_add_pytree	79.4480μs	35.3429μs	28.2942 KOps/s	28.9898 KOps/s	$\color{#d91a1a}-2.40\%$
test_add_td	0.1205ms	56.6126μs	17.6639 KOps/s	17.9471 KOps/s	$\color{#d91a1a}-1.58\%$
test_compile_add_one_nested[tensordict-compile]	0.1599ms	61.7590μs	16.1920 KOps/s	16.5064 KOps/s	$\color{#d91a1a}-1.90\%$
test_compile_add_one_nested[tensordict-eager]	0.4717ms	0.1683ms	5.9435 KOps/s	5.9144 KOps/s	$\color{#35bf28}+0.49\%$
test_compile_add_one_nested[pytree-compile]	0.1366ms	45.2507μs	22.0991 KOps/s	22.2272 KOps/s	$\color{#d91a1a}-0.58\%$
test_compile_add_one_nested[pytree-eager]	0.2394ms	0.1176ms	8.5044 KOps/s	8.6131 KOps/s	$\color{#d91a1a}-1.26\%$
test_compile_copy_nested[tensordict-compile]	91.4400μs	26.3004μs	38.0222 KOps/s	38.8517 KOps/s	$\color{#d91a1a}-2.14\%$
test_compile_copy_nested[tensordict-eager]	0.1298ms	59.1190μs	16.9150 KOps/s	16.9982 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_copy_nested[pytree-compile]	0.2035ms	78.1504μs	12.7958 KOps/s	12.7145 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_copy_nested[pytree-eager]	0.1515ms	68.0467μs	14.6958 KOps/s	14.7625 KOps/s	$\color{#d91a1a}-0.45\%$
test_compile_add_one_flat[tensordict-compile]	0.2311ms	0.1041ms	9.6039 KOps/s	9.7020 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_add_one_flat[tensordict-eager]	0.4297ms	0.2137ms	4.6785 KOps/s	4.6819 KOps/s	$\color{#d91a1a}-0.07\%$
test_compile_add_one_flat[tensorclass-compile]	0.1344ms	44.3086μs	22.5690 KOps/s	22.1905 KOps/s	$\color{#35bf28}+1.71\%$
test_compile_add_one_flat[tensorclass-eager]	0.4955ms	63.6951μs	15.6998 KOps/s	15.3937 KOps/s	$\color{#35bf28}+1.99\%$
test_compile_add_one_flat[pytree-compile]	0.1915ms	0.1030ms	9.7048 KOps/s	9.7944 KOps/s	$\color{#d91a1a}-0.91\%$
test_compile_add_one_flat[pytree-eager]	0.4351ms	0.2022ms	4.9453 KOps/s	5.0299 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_add_self_flat[tensordict-eager]	0.4624ms	0.2297ms	4.3541 KOps/s	4.3353 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_add_self_flat[tensordict-compile]	0.1996ms	0.1043ms	9.5867 KOps/s	9.4623 KOps/s	$\color{#35bf28}+1.32\%$
test_compile_add_self_flat[tensorclass-eager]	0.1207ms	59.9178μs	16.6895 KOps/s	16.9525 KOps/s	$\color{#d91a1a}-1.55\%$
test_compile_add_self_flat[tensorclass-compile]	0.1296ms	44.3126μs	22.5669 KOps/s	21.9457 KOps/s	$\color{#35bf28}+2.83\%$
test_compile_add_self_flat[pytree-eager]	1.4078ms	0.1617ms	6.1840 KOps/s	6.3751 KOps/s	$\color{#d91a1a}-3.00\%$
test_compile_add_self_flat[pytree-compile]	0.2489ms	0.1053ms	9.4935 KOps/s	9.8004 KOps/s	$\color{#d91a1a}-3.13\%$
test_compile_copy_flat[tensordict-compile]	93.3050μs	21.4703μs	46.5759 KOps/s	47.7177 KOps/s	$\color{#d91a1a}-2.39\%$
test_compile_copy_flat[tensordict-eager]	0.3393ms	67.8520μs	14.7380 KOps/s	15.2028 KOps/s	$\color{#d91a1a}-3.06\%$
test_compile_copy_flat[pytree-compile]	0.1933ms	81.7681μs	12.2297 KOps/s	12.4404 KOps/s	$\color{#d91a1a}-1.69\%$
test_compile_copy_flat[pytree-eager]	0.1248ms	69.1185μs	14.4679 KOps/s	14.7696 KOps/s	$\color{#d91a1a}-2.04\%$
test_compile_assign_and_add[tensordict-compile]	0.4353ms	0.2085ms	4.7962 KOps/s	4.8748 KOps/s	$\color{#d91a1a}-1.61\%$
test_compile_assign_and_add[tensordict-eager]	2.3799ms	1.2835ms	779.1273 Ops/s	726.8870 Ops/s	$\textbf{\color{#35bf28}+7.19\%}$
test_compile_assign_and_add[pytree-compile]	0.4010ms	0.2049ms	4.8795 KOps/s	4.9539 KOps/s	$\color{#d91a1a}-1.50\%$
test_compile_assign_and_add[pytree-eager]	1.2315ms	0.7676ms	1.3028 KOps/s	1.3150 KOps/s	$\color{#d91a1a}-0.93\%$
test_compile_assign_and_add_stack[compile]	0.9899ms	0.4574ms	2.1865 KOps/s	2.1951 KOps/s	$\color{#d91a1a}-0.39\%$
test_compile_assign_and_add_stack[eager]	3.5690ms	2.6607ms	375.8456 Ops/s	376.7318 Ops/s	$\color{#d91a1a}-0.24\%$
test_compile_indexing[tensor-tensordict-compile]	98.0820μs	36.1920μs	27.6304 KOps/s	27.9012 KOps/s	$\color{#d91a1a}-0.97\%$
test_compile_indexing[tensor-tensordict-eager]	0.6123ms	32.9214μs	30.3754 KOps/s	30.6141 KOps/s	$\color{#d91a1a}-0.78\%$
test_compile_indexing[tensor-tensorclass-compile]	73.5480μs	28.5865μs	34.9815 KOps/s	34.4768 KOps/s	$\color{#35bf28}+1.46\%$
test_compile_indexing[tensor-tensorclass-eager]	64.2000μs	23.5319μs	42.4955 KOps/s	43.3073 KOps/s	$\color{#d91a1a}-1.87\%$
test_compile_indexing[tensor-pytree-compile]	76.8940μs	28.9954μs	34.4882 KOps/s	33.3232 KOps/s	$\color{#35bf28}+3.50\%$
test_compile_indexing[tensor-pytree-eager]	0.1112ms	23.1329μs	43.2284 KOps/s	43.8897 KOps/s	$\color{#d91a1a}-1.51\%$
test_compile_indexing[slice-tensordict-compile]	0.1002ms	50.4939μs	19.8044 KOps/s	19.7367 KOps/s	$\color{#35bf28}+0.34\%$
test_compile_indexing[slice-tensordict-eager]	0.5990ms	20.2211μs	49.4532 KOps/s	50.2748 KOps/s	$\color{#d91a1a}-1.63\%$
test_compile_indexing[slice-tensorclass-compile]	88.9160μs	42.4064μs	23.5814 KOps/s	22.5720 KOps/s	$\color{#35bf28}+4.47\%$
test_compile_indexing[slice-tensorclass-eager]	60.0620μs	18.9051μs	52.8958 KOps/s	54.4394 KOps/s	$\color{#d91a1a}-2.84\%$
test_compile_indexing[slice-pytree-compile]	90.4300μs	43.3736μs	23.0555 KOps/s	22.1510 KOps/s	$\color{#35bf28}+4.08\%$
test_compile_indexing[slice-pytree-eager]	46.4570μs	18.9363μs	52.8087 KOps/s	54.4491 KOps/s	$\color{#d91a1a}-3.01\%$
test_compile_indexing[int-tensordict-compile]	0.1223ms	51.6560μs	19.3588 KOps/s	19.1422 KOps/s	$\color{#35bf28}+1.13\%$
test_compile_indexing[int-tensordict-eager]	0.9550ms	19.7789μs	50.5588 KOps/s	50.2451 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_indexing[int-tensorclass-compile]	96.7710μs	44.1712μs	22.6392 KOps/s	22.0850 KOps/s	$\color{#35bf28}+2.51\%$
test_compile_indexing[int-tensorclass-eager]	0.1759ms	18.7777μs	53.2547 KOps/s	54.5953 KOps/s	$\color{#d91a1a}-2.46\%$
test_compile_indexing[int-pytree-compile]	0.5741ms	44.3112μs	22.5676 KOps/s	22.1901 KOps/s	$\color{#35bf28}+1.70\%$
test_compile_indexing[int-pytree-eager]	59.5620μs	18.6954μs	53.4890 KOps/s	54.4149 KOps/s	$\color{#d91a1a}-1.70\%$
test_mod_add[eager]	90.2790μs	35.3536μs	28.2857 KOps/s	29.4170 KOps/s	$\color{#d91a1a}-3.85\%$
test_mod_add[compile]	99.7470μs	47.2762μs	21.1523 KOps/s	20.8499 KOps/s	$\color{#35bf28}+1.45\%$
test_mod_add[compile-overhead]	0.1030ms	47.3700μs	21.1104 KOps/s	20.4734 KOps/s	$\color{#35bf28}+3.11\%$
test_mod_wrap[eager]	0.4289ms	0.2251ms	4.4431 KOps/s	4.4733 KOps/s	$\color{#d91a1a}-0.68\%$
test_mod_wrap[compile]	0.2976ms	0.2019ms	4.9521 KOps/s	4.8512 KOps/s	$\color{#35bf28}+2.08\%$
test_mod_wrap[compile-overhead]	0.4146ms	0.2020ms	4.9505 KOps/s	4.8253 KOps/s	$\color{#35bf28}+2.59\%$
test_mod_wrap_and_backward[eager]	18.7171ms	12.1022ms	82.6295 Ops/s	86.5498 Ops/s	$\color{#d91a1a}-4.53\%$
test_mod_wrap_and_backward[compile]	19.6460ms	12.8310ms	77.9361 Ops/s	72.7191 Ops/s	$\textbf{\color{#35bf28}+7.17\%}$
test_mod_wrap_and_backward[compile-overhead]	14.1073ms	12.2432ms	81.6782 Ops/s	70.9170 Ops/s	$\textbf{\color{#35bf28}+15.17\%}$
test_seq_add[eager]	0.2827ms	0.1151ms	8.6848 KOps/s	8.7375 KOps/s	$\color{#d91a1a}-0.60\%$
test_seq_add[compile]	0.1128ms	62.3187μs	16.0465 KOps/s	15.9751 KOps/s	$\color{#35bf28}+0.45\%$
test_seq_add[compile-overhead]	0.1661ms	60.5202μs	16.5234 KOps/s	16.2302 KOps/s	$\color{#35bf28}+1.81\%$
test_seq_wrap[eager]	0.7255ms	0.4427ms	2.2589 KOps/s	2.2342 KOps/s	$\color{#35bf28}+1.11\%$
test_seq_wrap[compile]	0.3294ms	0.2214ms	4.5164 KOps/s	4.2747 KOps/s	$\textbf{\color{#35bf28}+5.65\%}$
test_seq_wrap[compile-overhead]	0.4344ms	0.2228ms	4.4876 KOps/s	4.3329 KOps/s	$\color{#35bf28}+3.57\%$
test_func_call_runtime[False-eager]	0.8051ms	0.5386ms	1.8567 KOps/s	1.8385 KOps/s	$\color{#35bf28}+0.99\%$
test_func_call_runtime[False-compile]	0.5343ms	0.4159ms	2.4041 KOps/s	2.3609 KOps/s	$\color{#35bf28}+1.83\%$
test_func_call_runtime[False-compile-overhead]	0.7599ms	0.4160ms	2.4036 KOps/s	2.3399 KOps/s	$\color{#35bf28}+2.73\%$
test_func_call_runtime[True-eager]	1.2850ms	0.7560ms	1.3228 KOps/s	1.3061 KOps/s	$\color{#35bf28}+1.28\%$
test_func_call_runtime[True-compile]	0.6980ms	0.4566ms	2.1900 KOps/s	2.1576 KOps/s	$\color{#35bf28}+1.50\%$
test_func_call_runtime[True-compile-overhead]	0.6077ms	0.4548ms	2.1990 KOps/s	2.1474 KOps/s	$\color{#35bf28}+2.40\%$
test_func_call_cm_runtime[False-eager]	1.0017ms	0.5455ms	1.8333 KOps/s	1.8447 KOps/s	$\color{#d91a1a}-0.62\%$
test_func_call_cm_runtime[False-compile]	0.5768ms	0.4182ms	2.3915 KOps/s	2.3463 KOps/s	$\color{#35bf28}+1.93\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5451ms	0.4186ms	2.3889 KOps/s	2.3591 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_cm_runtime[True-eager]	1.4533ms	0.9192ms	1.0879 KOps/s	1.0944 KOps/s	$\color{#d91a1a}-0.59\%$
test_func_call_cm_runtime[True-compile]	0.6273ms	0.4833ms	2.0689 KOps/s	2.0365 KOps/s	$\color{#35bf28}+1.59\%$
test_func_call_cm_runtime[True-compile-overhead]	1.2017ms	0.5041ms	1.9838 KOps/s	1.9903 KOps/s	$\color{#d91a1a}-0.33\%$
test_vmap_func_call_cm_runtime[eager]	2.5878ms	1.9352ms	516.7475 Ops/s	512.7829 Ops/s	$\color{#35bf28}+0.77\%$
test_vmap_func_call_cm_runtime[compile]	1.5275ms	0.5193ms	1.9255 KOps/s	1.9400 KOps/s	$\color{#d91a1a}-0.75\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.5188ms	0.5313ms	1.8822 KOps/s	1.8848 KOps/s	$\color{#d91a1a}-0.14\%$
test_distributed	0.3776ms	0.1257ms	7.9542 KOps/s	7.6744 KOps/s	$\color{#35bf28}+3.65\%$
test_tdmodule	0.1203ms	26.9669μs	37.0825 KOps/s	39.1526 KOps/s	$\textbf{\color{#d91a1a}-5.29\%}$
test_tdmodule_dispatch	88.4460μs	49.3089μs	20.2803 KOps/s	21.4834 KOps/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_tdseq	69.7210μs	29.6396μs	33.7386 KOps/s	35.4333 KOps/s	$\color{#d91a1a}-4.78\%$
test_tdseq_dispatch	0.1168ms	57.4108μs	17.4183 KOps/s	19.0826 KOps/s	$\textbf{\color{#d91a1a}-8.72\%}$
test_instantiation_functorch	1.6892ms	1.5256ms	655.4821 Ops/s	641.0368 Ops/s	$\color{#35bf28}+2.25\%$
test_exec_functorch	0.2757ms	0.1784ms	5.6057 KOps/s	5.5373 KOps/s	$\color{#35bf28}+1.23\%$
test_exec_functional_call	0.4116ms	0.1754ms	5.7009 KOps/s	5.7787 KOps/s	$\color{#d91a1a}-1.35\%$
test_exec_td_decorator	0.5184ms	0.2323ms	4.3051 KOps/s	4.2848 KOps/s	$\color{#35bf28}+0.47\%$
test_vmap_mlp_speed_decorator[True-True]	0.9625ms	0.6621ms	1.5103 KOps/s	1.5157 KOps/s	$\color{#d91a1a}-0.35\%$
test_vmap_mlp_speed_decorator[True-False]	1.2526ms	0.6572ms	1.5215 KOps/s	1.5315 KOps/s	$\color{#d91a1a}-0.65\%$
test_vmap_mlp_speed_decorator[False-True]	0.8782ms	0.5277ms	1.8950 KOps/s	1.8740 KOps/s	$\color{#35bf28}+1.12\%$
test_vmap_mlp_speed_decorator[False-False]	0.7400ms	0.5250ms	1.9049 KOps/s	1.8883 KOps/s	$\color{#35bf28}+0.88\%$
test_to_module_speed[True]	2.1531ms	1.3692ms	730.3585 Ops/s	722.7623 Ops/s	$\color{#35bf28}+1.05\%$
test_to_module_speed[False]	1.8102ms	1.3251ms	754.6636 Ops/s	732.2177 Ops/s	$\color{#35bf28}+3.07\%$
test_tc_init	96.4410μs	48.7065μs	20.5311 KOps/s	23.3502 KOps/s	$\textbf{\color{#d91a1a}-12.07\%}$
test_tc_init_nested	0.2159ms	99.4313μs	10.0572 KOps/s	11.0159 KOps/s	$\textbf{\color{#d91a1a}-8.70\%}$
test_tc_first_layer_tensor	18.9450μs	1.5005μs	666.4252 KOps/s	654.8910 KOps/s	$\color{#35bf28}+1.76\%$
test_tc_first_layer_nontensor	0.1271ms	4.9962μs	200.1510 KOps/s	212.1105 KOps/s	$\textbf{\color{#d91a1a}-5.64\%}$
test_tc_second_layer_tensor	53.0890μs	2.7755μs	360.3007 KOps/s	344.5175 KOps/s	$\color{#35bf28}+4.58\%$
test_tc_second_layer_nontensor	45.5250μs	5.9593μs	167.8060 KOps/s	164.2559 KOps/s	$\color{#35bf28}+2.16\%$
test_unbind	0.2416s	15.0722ms	66.3472 Ops/s	62.4524 Ops/s	$\textbf{\color{#35bf28}+6.24\%}$
test_full_like	10.3570ms	8.9001ms	112.3583 Ops/s	108.2412 Ops/s	$\color{#35bf28}+3.80\%$
test_zeros_like	5.3324ms	3.4826ms	287.1378 Ops/s	290.5799 Ops/s	$\color{#d91a1a}-1.18\%$
test_ones_like	4.7822ms	4.1580ms	240.4987 Ops/s	248.2286 Ops/s	$\color{#d91a1a}-3.11\%$
test_clone	7.2043ms	6.1457ms	162.7142 Ops/s	165.1478 Ops/s	$\color{#d91a1a}-1.47\%$
test_squeeze	71.9450μs	12.4035μs	80.6226 KOps/s	82.0601 KOps/s	$\color{#d91a1a}-1.75\%$
test_unsqueeze	0.2444ms	90.7924μs	11.0141 KOps/s	11.0525 KOps/s	$\color{#d91a1a}-0.35\%$
test_split	0.5109ms	0.1931ms	5.1797 KOps/s	5.0325 KOps/s	$\color{#35bf28}+2.93\%$
test_permute	0.3296ms	0.2063ms	4.8462 KOps/s	4.8000 KOps/s	$\color{#35bf28}+0.96\%$
test_stack	28.6262ms	26.0624ms	38.3695 Ops/s	36.4605 Ops/s	$\textbf{\color{#35bf28}+5.24\%}$
test_cat	30.9192ms	25.9670ms	38.5104 Ops/s	36.4342 Ops/s	$\textbf{\color{#35bf28}+5.70\%}$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}48$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	42.6000μs	12.0328μs	83.1062 KOps/s	76.0947 KOps/s	$\textbf{\color{#35bf28}+9.21\%}$
test_plain_set_stack_nested	37.0800μs	12.2793μs	81.4376 KOps/s	75.3392 KOps/s	$\textbf{\color{#35bf28}+8.09\%}$
test_plain_set_nested_inplace	54.9210μs	13.1649μs	75.9598 KOps/s	70.0096 KOps/s	$\textbf{\color{#35bf28}+8.50\%}$
test_plain_set_stack_nested_inplace	79.9120μs	13.1903μs	75.8131 KOps/s	69.9717 KOps/s	$\textbf{\color{#35bf28}+8.35\%}$
test_items	33.5000μs	2.9058μs	344.1433 KOps/s	341.6565 KOps/s	$\color{#35bf28}+0.73\%$
test_items_nested	0.4179ms	0.3747ms	2.6685 KOps/s	2.6713 KOps/s	$\color{#d91a1a}-0.10\%$
test_items_nested_locked	0.4191ms	0.3810ms	2.6247 KOps/s	2.6480 KOps/s	$\color{#d91a1a}-0.88\%$
test_items_nested_leaf	0.1074ms	57.4288μs	17.4129 KOps/s	17.0185 KOps/s	$\color{#35bf28}+2.32\%$
test_items_stack_nested	0.4185ms	0.3782ms	2.6444 KOps/s	2.6162 KOps/s	$\color{#35bf28}+1.08\%$
test_items_stack_nested_leaf	0.1050ms	57.9859μs	17.2456 KOps/s	16.6847 KOps/s	$\color{#35bf28}+3.36\%$
test_items_stack_nested_locked	0.4314ms	0.3821ms	2.6169 KOps/s	2.6483 KOps/s	$\color{#d91a1a}-1.18\%$
test_keys	33.2010μs	3.4930μs	286.2841 KOps/s	284.8428 KOps/s	$\color{#35bf28}+0.51\%$
test_keys_nested	0.1151ms	82.8487μs	12.0702 KOps/s	12.1643 KOps/s	$\color{#d91a1a}-0.77\%$
test_keys_nested_locked	0.7844ms	88.7879μs	11.2628 KOps/s	11.3528 KOps/s	$\color{#d91a1a}-0.79\%$
test_keys_nested_leaf	0.1037ms	73.5893μs	13.5889 KOps/s	13.6700 KOps/s	$\color{#d91a1a}-0.59\%$
test_keys_stack_nested	0.1249ms	82.8320μs	12.0726 KOps/s	11.8538 KOps/s	$\color{#35bf28}+1.85\%$
test_keys_stack_nested_leaf	0.1134ms	74.1097μs	13.4935 KOps/s	13.3266 KOps/s	$\color{#35bf28}+1.25\%$
test_keys_stack_nested_locked	0.1369ms	88.9669μs	11.2401 KOps/s	11.0536 KOps/s	$\color{#35bf28}+1.69\%$
test_values	5.5552μs	0.8472μs	1.1804 MOps/s	1.1755 MOps/s	$\color{#35bf28}+0.41\%$
test_values_nested	83.3720μs	35.8695μs	27.8788 KOps/s	28.1362 KOps/s	$\color{#d91a1a}-0.91\%$
test_values_nested_locked	63.9010μs	38.3686μs	26.0630 KOps/s	26.8423 KOps/s	$\color{#d91a1a}-2.90\%$
test_values_nested_leaf	74.5420μs	40.5375μs	24.6685 KOps/s	24.9033 KOps/s	$\color{#d91a1a}-0.94\%$
test_values_stack_nested	71.7310μs	35.8587μs	27.8872 KOps/s	27.6775 KOps/s	$\color{#35bf28}+0.76\%$
test_values_stack_nested_leaf	67.3810μs	40.8215μs	24.4969 KOps/s	24.4323 KOps/s	$\color{#35bf28}+0.26\%$
test_values_stack_nested_locked	69.6710μs	38.6295μs	25.8869 KOps/s	26.4178 KOps/s	$\color{#d91a1a}-2.01\%$
test_membership	1.6725μs	0.5330μs	1.8762 MOps/s	1.8470 MOps/s	$\color{#35bf28}+1.58\%$
test_membership_nested	15.7055μs	2.0862μs	479.3304 KOps/s	455.8850 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_membership_nested_leaf	36.9755μs	2.1315μs	469.1452 KOps/s	474.7715 KOps/s	$\color{#d91a1a}-1.19\%$
test_membership_stacked_nested	43.9210μs	2.1655μs	461.7803 KOps/s	465.5412 KOps/s	$\color{#d91a1a}-0.81\%$
test_membership_stacked_nested_leaf	26.5710μs	2.2140μs	451.6772 KOps/s	464.8341 KOps/s	$\color{#d91a1a}-2.83\%$
test_membership_nested_last	36.1300μs	3.1808μs	314.3910 KOps/s	310.6257 KOps/s	$\color{#35bf28}+1.21\%$
test_membership_nested_leaf_last	27.4100μs	3.2171μs	310.8369 KOps/s	308.9280 KOps/s	$\color{#35bf28}+0.62\%$
test_membership_stacked_nested_last	19.1410μs	3.2046μs	312.0479 KOps/s	270.7138 KOps/s	$\textbf{\color{#35bf28}+15.27\%}$
test_membership_stacked_nested_leaf_last	32.0610μs	3.2108μs	311.4481 KOps/s	270.9397 KOps/s	$\textbf{\color{#35bf28}+14.95\%}$
test_nested_getleaf	37.3210μs	6.2125μs	160.9657 KOps/s	162.5282 KOps/s	$\color{#d91a1a}-0.96\%$
test_nested_get	35.2510μs	5.8983μs	169.5393 KOps/s	172.0359 KOps/s	$\color{#d91a1a}-1.45\%$
test_stacked_getleaf	34.6100μs	6.3056μs	158.5889 KOps/s	161.3757 KOps/s	$\color{#d91a1a}-1.73\%$
test_stacked_get	35.0010μs	5.8784μs	170.1148 KOps/s	169.7752 KOps/s	$\color{#35bf28}+0.20\%$
test_nested_getitemleaf	36.4810μs	6.2263μs	160.6087 KOps/s	159.3792 KOps/s	$\color{#35bf28}+0.77\%$
test_nested_getitem	27.3910μs	6.0023μs	166.6031 KOps/s	167.0447 KOps/s	$\color{#d91a1a}-0.26\%$
test_stacked_getitemleaf	42.2310μs	6.2603μs	159.7374 KOps/s	157.2917 KOps/s	$\color{#35bf28}+1.55\%$
test_stacked_getitem	27.6000μs	6.0465μs	165.3843 KOps/s	166.3426 KOps/s	$\color{#d91a1a}-0.58\%$
test_lock_nested	9.4184ms	0.3919ms	2.5520 KOps/s	2.5408 KOps/s	$\color{#35bf28}+0.44\%$
test_lock_stack_nested	0.4047ms	0.3526ms	2.8364 KOps/s	2.7991 KOps/s	$\color{#35bf28}+1.33\%$
test_unlock_nested	0.6263ms	0.3119ms	3.2059 KOps/s	3.1772 KOps/s	$\color{#35bf28}+0.90\%$
test_unlock_stack_nested	0.3578ms	0.2822ms	3.5435 KOps/s	3.4917 KOps/s	$\color{#35bf28}+1.48\%$
test_flatten_speed	0.1284ms	75.2303μs	13.2925 KOps/s	13.0152 KOps/s	$\color{#35bf28}+2.13\%$
test_unflatten_speed	0.3783ms	0.3266ms	3.0616 KOps/s	3.0835 KOps/s	$\color{#d91a1a}-0.71\%$
test_common_ops	1.6419ms	0.5984ms	1.6712 KOps/s	1.5325 KOps/s	$\textbf{\color{#35bf28}+9.05\%}$
test_creation	96.9020μs	1.8243μs	548.1469 KOps/s	570.3690 KOps/s	$\color{#d91a1a}-3.90\%$
test_creation_empty	30.9110μs	7.9108μs	126.4091 KOps/s	101.5735 KOps/s	$\textbf{\color{#35bf28}+24.45\%}$
test_creation_nested_1	39.8300μs	9.5059μs	105.1980 KOps/s	86.8270 KOps/s	$\textbf{\color{#35bf28}+21.16\%}$
test_creation_nested_2	40.2510μs	12.3010μs	81.2944 KOps/s	70.3414 KOps/s	$\textbf{\color{#35bf28}+15.57\%}$
test_clone	48.8010μs	10.5814μs	94.5059 KOps/s	94.2201 KOps/s	$\color{#35bf28}+0.30\%$
test_getitem[int]	93.0000ms	16.1461μs	61.9346 KOps/s	93.7571 KOps/s	$\textbf{\color{#d91a1a}-33.94\%}$
test_getitem[slice_int]	0.1103ms	21.1528μs	47.2750 KOps/s	48.0388 KOps/s	$\color{#d91a1a}-1.59\%$
test_getitem[range]	0.1257ms	37.7445μs	26.4939 KOps/s	26.2987 KOps/s	$\color{#35bf28}+0.74\%$
test_getitem[tuple]	0.1025ms	17.9890μs	55.5895 KOps/s	53.9146 KOps/s	$\color{#35bf28}+3.11\%$
test_getitem[list]	0.2248ms	33.4694μs	29.8780 KOps/s	29.4334 KOps/s	$\color{#35bf28}+1.51\%$
test_setitem_dim[int]	39.0210μs	18.1485μs	55.1011 KOps/s	52.2733 KOps/s	$\textbf{\color{#35bf28}+5.41\%}$
test_setitem_dim[slice_int]	70.7310μs	38.5597μs	25.9338 KOps/s	25.5863 KOps/s	$\color{#35bf28}+1.36\%$
test_setitem_dim[range]	84.3320μs	52.9694μs	18.8788 KOps/s	18.8947 KOps/s	$\color{#d91a1a}-0.08\%$
test_setitem_dim[tuple]	59.7210μs	31.4503μs	31.7962 KOps/s	31.0823 KOps/s	$\color{#35bf28}+2.30\%$
test_setitem	77.4820μs	14.6652μs	68.1885 KOps/s	61.9954 KOps/s	$\textbf{\color{#35bf28}+9.99\%}$
test_set	88.9320μs	14.9665μs	66.8159 KOps/s	63.3103 KOps/s	$\textbf{\color{#35bf28}+5.54\%}$
test_set_shared	1.4655ms	0.1505ms	6.6452 KOps/s	6.6313 KOps/s	$\color{#35bf28}+0.21\%$
test_update	0.5555ms	17.3230μs	57.7268 KOps/s	51.8771 KOps/s	$\textbf{\color{#35bf28}+11.28\%}$
test_update_nested	83.3410μs	23.1920μs	43.1183 KOps/s	41.0542 KOps/s	$\textbf{\color{#35bf28}+5.03\%}$
test_update__nested	0.3165ms	26.8831μs	37.1981 KOps/s	39.7907 KOps/s	$\textbf{\color{#d91a1a}-6.52\%}$
test_set_nested	87.0320μs	16.4041μs	60.9604 KOps/s	54.2908 KOps/s	$\textbf{\color{#35bf28}+12.28\%}$
test_set_nested_new	77.0710μs	18.9099μs	52.8824 KOps/s	48.1089 KOps/s	$\textbf{\color{#35bf28}+9.92\%}$
test_select	0.2109ms	30.5749μs	32.7066 KOps/s	32.7673 KOps/s	$\color{#d91a1a}-0.19\%$
test_select_nested	67.9210μs	44.2299μs	22.6091 KOps/s	22.7558 KOps/s	$\color{#d91a1a}-0.64\%$
test_exclude_nested	0.1115ms	63.1560μs	15.8338 KOps/s	15.5434 KOps/s	$\color{#35bf28}+1.87\%$
test_empty[True]	0.3546ms	0.2960ms	3.3780 KOps/s	3.3813 KOps/s	$\color{#d91a1a}-0.10\%$
test_empty[False]	2.9931μs	0.8567μs	1.1672 MOps/s	1.1933 MOps/s	$\color{#d91a1a}-2.19\%$
test_to	0.1156ms	58.6800μs	17.0416 KOps/s	17.5918 KOps/s	$\color{#d91a1a}-3.13\%$
test_to_nonblocking	91.6010μs	48.9004μs	20.4497 KOps/s	19.9310 KOps/s	$\color{#35bf28}+2.60\%$
test_unbind_speed	0.2855ms	0.2362ms	4.2340 KOps/s	4.2791 KOps/s	$\color{#d91a1a}-1.05\%$
test_unbind_speed_stack0	0.2882ms	0.2337ms	4.2787 KOps/s	4.1867 KOps/s	$\color{#35bf28}+2.20\%$
test_unbind_speed_stack1	93.7397ms	0.6713ms	1.4896 KOps/s	1.4795 KOps/s	$\color{#35bf28}+0.69\%$
test_split	92.9169ms	1.5770ms	634.1096 Ops/s	629.3105 Ops/s	$\color{#35bf28}+0.76\%$
test_chunk	97.4347ms	1.7254ms	579.5662 Ops/s	574.3179 Ops/s	$\color{#35bf28}+0.91\%$
test_consolidate[False-None]	2.8168ms	2.6861ms	372.2897 Ops/s	369.0313 Ops/s	$\color{#35bf28}+0.88\%$
test_consolidate[default-None]	1.7824ms	1.6880ms	592.4071 Ops/s	561.7200 Ops/s	$\textbf{\color{#35bf28}+5.46\%}$
test_consolidate[reduce-overhead-None]	1.7811ms	1.7161ms	582.7230 Ops/s	572.2419 Ops/s	$\color{#35bf28}+1.83\%$
test_consolidate_njt[False-None]	6.9614ms	6.6364ms	150.6836 Ops/s	150.3001 Ops/s	$\color{#35bf28}+0.26\%$
test_to[False-False-None]	1.7994ms	1.6738ms	597.4544 Ops/s	567.6426 Ops/s	$\textbf{\color{#35bf28}+5.25\%}$
test_to[True-False-None]	1.5478ms	1.2975ms	770.6971 Ops/s	755.9525 Ops/s	$\color{#35bf28}+1.95\%$
test_to[within-False-None]	4.3162ms	4.0999ms	243.9101 Ops/s	240.8796 Ops/s	$\color{#35bf28}+1.26\%$
test_to[True-default-None]	5.7231ms	5.2948ms	188.8630 Ops/s	187.1163 Ops/s	$\color{#35bf28}+0.93\%$
test_to_njt[False-False-None]	7.0548ms	6.8961ms	145.0103 Ops/s	143.7741 Ops/s	$\color{#35bf28}+0.86\%$
test_to_njt[True-False-None]	5.6947ms	5.4687ms	182.8594 Ops/s	179.0001 Ops/s	$\color{#35bf28}+2.16\%$
test_to_njt[within-False-None]	12.3182ms	11.9992ms	83.3387 Ops/s	80.7501 Ops/s	$\color{#35bf28}+3.21\%$
test_creation[device0]	0.4644ms	79.2031μs	12.6258 KOps/s	12.4499 KOps/s	$\color{#35bf28}+1.41\%$
test_creation_from_tensor	0.5195ms	85.3115μs	11.7217 KOps/s	11.4987 KOps/s	$\color{#35bf28}+1.94\%$
test_add_one[memmap_tensor0]	0.4212ms	6.6082μs	151.3264 KOps/s	137.8660 KOps/s	$\textbf{\color{#35bf28}+9.76\%}$
test_contiguous[memmap_tensor0]	1.9176μs	0.4309μs	2.3208 MOps/s	2.2790 MOps/s	$\color{#35bf28}+1.83\%$
test_stack[memmap_tensor0]	38.6110μs	4.3664μs	229.0232 KOps/s	232.5074 KOps/s	$\color{#d91a1a}-1.50\%$
test_memmaptd_index	1.7621ms	0.2471ms	4.0466 KOps/s	4.0621 KOps/s	$\color{#d91a1a}-0.38\%$
test_memmaptd_index_astensor	0.5804ms	0.3082ms	3.2449 KOps/s	3.2538 KOps/s	$\color{#d91a1a}-0.27\%$
test_memmaptd_index_op	0.9959ms	0.5641ms	1.7729 KOps/s	1.6660 KOps/s	$\textbf{\color{#35bf28}+6.42\%}$
test_serialize_model	0.1321s	0.1311s	7.6281 Ops/s	7.6996 Ops/s	$\color{#d91a1a}-0.93\%$
test_serialize_model_pickle	1.3498s	1.2139s	0.8238 Ops/s	0.8216 Ops/s	$\color{#35bf28}+0.26\%$
test_serialize_weights	0.1309s	0.1294s	7.7267 Ops/s	7.6870 Ops/s	$\color{#35bf28}+0.52\%$
test_serialize_weights_returnearly	0.3112s	54.1192ms	18.4777 Ops/s	11.1624 Ops/s	$\textbf{\color{#35bf28}+65.54\%}$
test_serialize_weights_pickle	1.3553s	1.2178s	0.8211 Ops/s	0.8201 Ops/s	$\color{#35bf28}+0.13\%$
test_reshape_pytree	53.5910μs	21.8648μs	45.7357 KOps/s	43.9917 KOps/s	$\color{#35bf28}+3.96\%$
test_reshape_td	68.1920μs	26.2999μs	38.0230 KOps/s	36.6873 KOps/s	$\color{#35bf28}+3.64\%$
test_view_pytree	60.2210μs	22.1881μs	45.0692 KOps/s	44.4510 KOps/s	$\color{#35bf28}+1.39\%$
test_view_td	59.2710μs	30.7731μs	32.4959 KOps/s	31.4276 KOps/s	$\color{#35bf28}+3.40\%$
test_unbind_pytree	51.7210μs	27.9693μs	35.7535 KOps/s	35.6399 KOps/s	$\color{#35bf28}+0.32\%$
test_unbind_td	0.8113ms	36.8798μs	27.1151 KOps/s	27.6933 KOps/s	$\color{#d91a1a}-2.09\%$
test_split_pytree	63.5320μs	29.7243μs	33.6425 KOps/s	32.7069 KOps/s	$\color{#35bf28}+2.86\%$
test_split_td	0.9338ms	38.0275μs	26.2968 KOps/s	26.1384 KOps/s	$\color{#35bf28}+0.61\%$
test_add_pytree	66.1410μs	33.9396μs	29.4641 KOps/s	29.3187 KOps/s	$\color{#35bf28}+0.50\%$
test_add_td	75.3720μs	47.1000μs	21.2314 KOps/s	20.5823 KOps/s	$\color{#35bf28}+3.15\%$
test_compile_add_one_nested[tensordict-compile]	0.1842ms	0.1261ms	7.9308 KOps/s	7.8829 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_add_one_nested[tensordict-eager]	0.2228ms	0.1300ms	7.6937 KOps/s	7.6579 KOps/s	$\color{#35bf28}+0.47\%$
test_compile_add_one_nested[pytree-compile]	0.1369ms	98.0935μs	10.1944 KOps/s	10.2162 KOps/s	$\color{#d91a1a}-0.21\%$
test_compile_add_one_nested[pytree-eager]	1.5546ms	0.1483ms	6.7419 KOps/s	6.6226 KOps/s	$\color{#35bf28}+1.80\%$
test_compile_copy_nested[tensordict-compile]	82.7320μs	21.9339μs	45.5916 KOps/s	46.0911 KOps/s	$\color{#d91a1a}-1.08\%$
test_compile_copy_nested[tensordict-eager]	56.7710μs	28.8897μs	34.6145 KOps/s	34.2075 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_copy_nested[pytree-compile]	0.3419ms	64.2505μs	15.5641 KOps/s	15.3149 KOps/s	$\color{#35bf28}+1.63\%$
test_compile_copy_nested[pytree-eager]	0.1414ms	50.2615μs	19.8959 KOps/s	20.0466 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_add_one_flat[tensordict-compile]	0.2173ms	0.1435ms	6.9706 KOps/s	7.0650 KOps/s	$\color{#d91a1a}-1.34\%$
test_compile_add_one_flat[tensordict-eager]	0.3034ms	0.2130ms	4.6958 KOps/s	4.6838 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_add_one_flat[tensorclass-compile]	0.1359ms	98.8872μs	10.1125 KOps/s	9.7759 KOps/s	$\color{#35bf28}+3.44\%$
test_compile_add_one_flat[tensorclass-eager]	0.1653ms	53.1424μs	18.8174 KOps/s	17.9520 KOps/s	$\color{#35bf28}+4.82\%$
test_compile_add_one_flat[pytree-compile]	0.2052ms	0.1372ms	7.2863 KOps/s	7.3450 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_add_one_flat[pytree-eager]	0.5975ms	0.4709ms	2.1234 KOps/s	2.0688 KOps/s	$\color{#35bf28}+2.64\%$
test_compile_add_self_flat[tensordict-eager]	0.3732ms	0.2546ms	3.9277 KOps/s	3.9084 KOps/s	$\color{#35bf28}+0.49\%$
test_compile_add_self_flat[tensordict-compile]	0.1903ms	0.1452ms	6.8855 KOps/s	6.9887 KOps/s	$\color{#d91a1a}-1.48\%$
test_compile_add_self_flat[tensorclass-eager]	0.1628ms	64.6075μs	15.4781 KOps/s	14.6692 KOps/s	$\textbf{\color{#35bf28}+5.51\%}$
test_compile_add_self_flat[tensorclass-compile]	0.2030ms	0.1003ms	9.9697 KOps/s	10.1492 KOps/s	$\color{#d91a1a}-1.77\%$
test_compile_add_self_flat[pytree-eager]	0.4666ms	0.4025ms	2.4846 KOps/s	2.4074 KOps/s	$\color{#35bf28}+3.21\%$
test_compile_add_self_flat[pytree-compile]	0.1941ms	0.1376ms	7.2655 KOps/s	7.2450 KOps/s	$\color{#35bf28}+0.28\%$
test_compile_copy_flat[tensordict-compile]	50.4300μs	18.3793μs	54.4089 KOps/s	58.5871 KOps/s	$\textbf{\color{#d91a1a}-7.13\%}$
test_compile_copy_flat[tensordict-eager]	62.9720μs	31.5562μs	31.6895 KOps/s	32.7719 KOps/s	$\color{#d91a1a}-3.30\%$
test_compile_copy_flat[pytree-compile]	0.1925ms	69.9758μs	14.2907 KOps/s	14.2629 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_copy_flat[pytree-eager]	89.1520μs	51.0801μs	19.5771 KOps/s	19.7170 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_assign_and_add[tensordict-compile]	1.6208ms	0.3898ms	2.5652 KOps/s	2.1906 KOps/s	$\textbf{\color{#35bf28}+17.10\%}$
test_compile_assign_and_add[tensordict-eager]	2.6041ms	2.5015ms	399.7665 Ops/s	366.1800 Ops/s	$\textbf{\color{#35bf28}+9.17\%}$
test_compile_assign_and_add[pytree-compile]	1.5783ms	0.4284ms	2.3342 KOps/s	2.2797 KOps/s	$\color{#35bf28}+2.39\%$
test_compile_assign_and_add[pytree-eager]	2.8927ms	2.5380ms	394.0059 Ops/s	374.8968 Ops/s	$\textbf{\color{#35bf28}+5.10\%}$
test_compile_indexing[tensor-tensordict-compile]	0.1689ms	0.1139ms	8.7793 KOps/s	8.6341 KOps/s	$\color{#35bf28}+1.68\%$
test_compile_indexing[tensor-tensordict-eager]	0.5962ms	78.9599μs	12.6647 KOps/s	12.2422 KOps/s	$\color{#35bf28}+3.45\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2059ms	0.1029ms	9.7198 KOps/s	8.9858 KOps/s	$\textbf{\color{#35bf28}+8.17\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.1129ms	66.1290μs	15.1219 KOps/s	14.2734 KOps/s	$\textbf{\color{#35bf28}+5.95\%}$
test_compile_indexing[tensor-pytree-compile]	0.1599ms	0.1035ms	9.6655 KOps/s	9.2013 KOps/s	$\textbf{\color{#35bf28}+5.05\%}$
test_compile_indexing[tensor-pytree-eager]	0.2172ms	66.5419μs	15.0281 KOps/s	14.3896 KOps/s	$\color{#35bf28}+4.44\%$
test_compile_indexing[slice-tensordict-compile]	0.1459ms	0.1003ms	9.9656 KOps/s	9.4995 KOps/s	$\color{#35bf28}+4.91\%$
test_compile_indexing[slice-tensordict-eager]	0.1411ms	17.2473μs	57.9801 KOps/s	56.0351 KOps/s	$\color{#35bf28}+3.47\%$
test_compile_indexing[slice-tensorclass-compile]	0.1275ms	96.1108μs	10.4047 KOps/s	10.3863 KOps/s	$\color{#35bf28}+0.18\%$
test_compile_indexing[slice-tensorclass-eager]	47.1810μs	15.6911μs	63.7305 KOps/s	63.5114 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_indexing[slice-pytree-compile]	0.1416ms	96.3322μs	10.3807 KOps/s	10.2846 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_indexing[slice-pytree-eager]	72.1310μs	15.4813μs	64.5941 KOps/s	62.3645 KOps/s	$\color{#35bf28}+3.58\%$
test_compile_indexing[int-tensordict-compile]	0.1650ms	0.1017ms	9.8363 KOps/s	9.4940 KOps/s	$\color{#35bf28}+3.61\%$
test_compile_indexing[int-tensordict-eager]	0.6028ms	16.6892μs	59.9188 KOps/s	55.2788 KOps/s	$\textbf{\color{#35bf28}+8.39\%}$
test_compile_indexing[int-tensorclass-compile]	0.2260ms	96.9857μs	10.3108 KOps/s	9.6736 KOps/s	$\textbf{\color{#35bf28}+6.59\%}$
test_compile_indexing[int-tensorclass-eager]	54.0210μs	15.5816μs	64.1781 KOps/s	63.3929 KOps/s	$\color{#35bf28}+1.24\%$
test_compile_indexing[int-pytree-compile]	0.2487ms	97.8753μs	10.2171 KOps/s	9.7963 KOps/s	$\color{#35bf28}+4.30\%$
test_compile_indexing[int-pytree-eager]	75.4010μs	15.4833μs	64.5855 KOps/s	63.8775 KOps/s	$\color{#35bf28}+1.11\%$
test_mod_add[eager]	0.1828ms	36.9148μs	27.0894 KOps/s	25.2394 KOps/s	$\textbf{\color{#35bf28}+7.33\%}$
test_mod_add[compile]	0.1332ms	83.8831μs	11.9214 KOps/s	12.1045 KOps/s	$\color{#d91a1a}-1.51\%$
test_mod_add[compile-overhead]	0.3188ms	0.1676ms	5.9662 KOps/s	5.6717 KOps/s	$\textbf{\color{#35bf28}+5.19\%}$
test_mod_wrap[eager]	0.3876ms	0.2446ms	4.0886 KOps/s	3.9370 KOps/s	$\color{#35bf28}+3.85\%$
test_mod_wrap[compile]	0.3519ms	0.2845ms	3.5147 KOps/s	3.3107 KOps/s	$\textbf{\color{#35bf28}+6.16\%}$
test_mod_wrap[compile-overhead]	7.0764ms	3.7479ms	266.8128 Ops/s	272.1873 Ops/s	$\color{#d91a1a}-1.97\%$
test_mod_wrap_and_backward[eager]	1.5335ms	1.3554ms	737.7906 Ops/s	674.6063 Ops/s	$\textbf{\color{#35bf28}+9.37\%}$
test_mod_wrap_and_backward[compile]	1.3602ms	1.2672ms	789.1132 Ops/s	717.9024 Ops/s	$\textbf{\color{#35bf28}+9.92\%}$
test_mod_wrap_and_backward[compile-overhead]	1.4113ms	0.9287ms	1.0768 KOps/s	945.5372 Ops/s	$\textbf{\color{#35bf28}+13.88\%}$
test_seq_add[eager]	0.2618ms	0.1139ms	8.7834 KOps/s	8.3609 KOps/s	$\textbf{\color{#35bf28}+5.05\%}$
test_seq_add[compile]	0.4821ms	88.1949μs	11.3385 KOps/s	11.4216 KOps/s	$\color{#d91a1a}-0.73\%$
test_seq_add[compile-overhead]	0.2793ms	0.1298ms	7.7057 KOps/s	7.4208 KOps/s	$\color{#35bf28}+3.84\%$
test_seq_wrap[eager]	0.5565ms	0.4119ms	2.4276 KOps/s	2.3215 KOps/s	$\color{#35bf28}+4.57\%$
test_seq_wrap[compile]	0.3669ms	0.2972ms	3.3644 KOps/s	3.2947 KOps/s	$\color{#35bf28}+2.12\%$
test_seq_wrap[compile-overhead]	0.3023ms	0.2260ms	4.4248 KOps/s	4.3787 KOps/s	$\color{#35bf28}+1.05\%$
test_func_call_runtime[False-eager]	0.8210ms	0.7131ms	1.4024 KOps/s	1.3377 KOps/s	$\color{#35bf28}+4.83\%$
test_func_call_runtime[False-compile]	0.8544ms	0.7415ms	1.3487 KOps/s	1.3331 KOps/s	$\color{#35bf28}+1.17\%$
test_func_call_runtime[False-compile-overhead]	0.4577ms	0.3674ms	2.7218 KOps/s	2.7019 KOps/s	$\color{#35bf28}+0.74\%$
test_func_call_runtime[True-eager]	0.9473ms	0.8877ms	1.1265 KOps/s	1.0928 KOps/s	$\color{#35bf28}+3.08\%$
test_func_call_runtime[True-compile]	0.8619ms	0.7638ms	1.3093 KOps/s	1.2973 KOps/s	$\color{#35bf28}+0.93\%$
test_func_call_runtime[True-compile-overhead]	0.5314ms	0.3886ms	2.5734 KOps/s	2.5734 KOps/s	$-0.00\%$
test_func_call_cm_runtime[False-eager]	0.8073ms	0.7162ms	1.3963 KOps/s	1.3509 KOps/s	$\color{#35bf28}+3.36\%$
test_func_call_cm_runtime[False-compile]	0.7970ms	0.7440ms	1.3441 KOps/s	1.3253 KOps/s	$\color{#35bf28}+1.42\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4143ms	0.3685ms	2.7137 KOps/s	2.7007 KOps/s	$\color{#35bf28}+0.48\%$
test_func_call_cm_runtime[True-eager]	1.1033ms	0.9937ms	1.0063 KOps/s	983.7119 Ops/s	$\color{#35bf28}+2.30\%$
test_func_call_cm_runtime[True-compile]	0.8452ms	0.7872ms	1.2703 KOps/s	1.2515 KOps/s	$\color{#35bf28}+1.51\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4693ms	0.4135ms	2.4184 KOps/s	2.3838 KOps/s	$\color{#35bf28}+1.45\%$
test_vmap_func_call_cm_runtime[eager]	2.5467ms	2.0921ms	477.9864 Ops/s	464.7711 Ops/s	$\color{#35bf28}+2.84\%$
test_vmap_func_call_cm_runtime[compile]	0.9473ms	0.8054ms	1.2416 KOps/s	1.2240 KOps/s	$\color{#35bf28}+1.43\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4671ms	0.4177ms	2.3939 KOps/s	2.3792 KOps/s	$\color{#35bf28}+0.62\%$
test_distributed	6.1231ms	0.2145ms	4.6613 KOps/s	8.3477 KOps/s	$\textbf{\color{#d91a1a}-44.16\%}$
test_tdmodule	54.4220μs	19.2860μs	51.8512 KOps/s	43.7563 KOps/s	$\textbf{\color{#35bf28}+18.50\%}$
test_tdmodule_dispatch	0.1661ms	35.7434μs	27.9772 KOps/s	25.4910 KOps/s	$\textbf{\color{#35bf28}+9.75\%}$
test_tdseq	39.4910μs	20.4598μs	48.8764 KOps/s	44.5270 KOps/s	$\textbf{\color{#35bf28}+9.77\%}$
test_tdseq_dispatch	60.3810μs	38.1327μs	26.2242 KOps/s	23.4311 KOps/s	$\textbf{\color{#35bf28}+11.92\%}$
test_instantiation_functorch	1.6427ms	1.5646ms	639.1318 Ops/s	623.4008 Ops/s	$\color{#35bf28}+2.52\%$
test_exec_functorch	0.1975ms	0.1419ms	7.0476 KOps/s	6.8227 KOps/s	$\color{#35bf28}+3.30\%$
test_exec_functional_call	0.2137ms	0.1334ms	7.4962 KOps/s	7.0058 KOps/s	$\textbf{\color{#35bf28}+7.00\%}$
test_exec_td_decorator	0.3843ms	0.1830ms	5.4659 KOps/s	5.3843 KOps/s	$\color{#35bf28}+1.52\%$
test_vmap_mlp_speed_decorator[True-True]	0.8554ms	0.6839ms	1.4622 KOps/s	1.4118 KOps/s	$\color{#35bf28}+3.57\%$
test_vmap_mlp_speed_decorator[True-False]	0.8716ms	0.6810ms	1.4684 KOps/s	1.3994 KOps/s	$\color{#35bf28}+4.93\%$
test_vmap_mlp_speed_decorator[False-True]	0.7214ms	0.5960ms	1.6779 KOps/s	1.5725 KOps/s	$\textbf{\color{#35bf28}+6.70\%}$
test_vmap_mlp_speed_decorator[False-False]	0.7424ms	0.5949ms	1.6809 KOps/s	1.5829 KOps/s	$\textbf{\color{#35bf28}+6.20\%}$
test_vmap_transformer_speed_decorator[True-True]	19.3669ms	19.2375ms	51.9819 Ops/s	51.0031 Ops/s	$\color{#35bf28}+1.92\%$
test_vmap_transformer_speed_decorator[True-False]	19.9999ms	19.2878ms	51.8462 Ops/s	50.7364 Ops/s	$\color{#35bf28}+2.19\%$
test_vmap_transformer_speed_decorator[False-True]	19.2398ms	19.1530ms	52.2112 Ops/s	51.5272 Ops/s	$\color{#35bf28}+1.33\%$
test_vmap_transformer_speed_decorator[False-False]	19.2597ms	19.1358ms	52.2581 Ops/s	51.4043 Ops/s	$\color{#35bf28}+1.66\%$
test_to_module_speed[True]	1.0805ms	0.9757ms	1.0249 KOps/s	1.0168 KOps/s	$\color{#35bf28}+0.79\%$
test_to_module_speed[False]	1.3472ms	0.9552ms	1.0470 KOps/s	1.0362 KOps/s	$\color{#35bf28}+1.03\%$
test_tc_init	71.3720μs	38.2802μs	26.1232 KOps/s	25.6036 KOps/s	$\color{#35bf28}+2.03\%$
test_tc_init_nested	0.1743ms	76.7088μs	13.0363 KOps/s	13.0757 KOps/s	$\color{#d91a1a}-0.30\%$
test_tc_first_layer_tensor	28.7400μs	0.8108μs	1.2334 MOps/s	1.4458 MOps/s	$\textbf{\color{#d91a1a}-14.69\%}$
test_tc_first_layer_nontensor	37.6400μs	2.3502μs	425.4964 KOps/s	427.2682 KOps/s	$\color{#d91a1a}-0.41\%$
test_tc_second_layer_tensor	23.9805μs	1.4479μs	690.6477 KOps/s	717.1851 KOps/s	$\color{#d91a1a}-3.70\%$
test_tc_second_layer_nontensor	0.1003ms	3.0828μs	324.3801 KOps/s	324.9702 KOps/s	$\color{#d91a1a}-0.18\%$
test_unbind	0.2270s	10.2269ms	97.7818 Ops/s	140.0149 Ops/s	$\textbf{\color{#d91a1a}-30.16\%}$
test_full_like	11.6167ms	9.5806ms	104.3778 Ops/s	105.3411 Ops/s	$\color{#d91a1a}-0.91\%$
test_zeros_like	6.0272ms	4.3697ms	228.8472 Ops/s	233.1153 Ops/s	$\color{#d91a1a}-1.83\%$
test_ones_like	5.0353ms	4.3653ms	229.0785 Ops/s	229.0816 Ops/s	$-0.00\%$
test_clone	7.0522ms	6.6487ms	150.4059 Ops/s	106.6790 Ops/s	$\textbf{\color{#35bf28}+40.99\%}$
test_squeeze	56.1710μs	9.9395μs	100.6084 KOps/s	92.2083 KOps/s	$\textbf{\color{#35bf28}+9.11\%}$
test_unsqueeze	0.1323ms	75.0888μs	13.3176 KOps/s	13.0064 KOps/s	$\color{#35bf28}+2.39\%$
test_split	0.4051ms	0.1627ms	6.1473 KOps/s	5.9063 KOps/s	$\color{#35bf28}+4.08\%$
test_permute	0.2408ms	0.1877ms	5.3269 KOps/s	5.2130 KOps/s	$\color{#35bf28}+2.19\%$
test_stack	51.7170ms	51.3288ms	19.4822 Ops/s	19.4331 Ops/s	$\color{#35bf28}+0.25\%$
test_cat	52.0423ms	51.2501ms	19.5122 Ops/s	19.5030 Ops/s	$\color{#35bf28}+0.05\%$