[BugFix] Do not unlock td if it's not locked in TDParams (for compile compat)

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Do not unlock td if it's not locked in TDParams (for compile compat) #1125
• [BugFix] Add checks and to_module to tensorclass accepted methods #1124

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}25$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.9590μs	17.8971μs	55.8750 KOps/s	51.2743 KOps/s	$\textbf{\color{#35bf28}+8.97\%}$
test_plain_set_stack_nested	42.2790μs	18.3172μs	54.5934 KOps/s	50.7668 KOps/s	$\textbf{\color{#35bf28}+7.54\%}$
test_plain_set_nested_inplace	46.5280μs	19.9735μs	50.0663 KOps/s	47.0619 KOps/s	$\textbf{\color{#35bf28}+6.38\%}$
test_plain_set_stack_nested_inplace	73.4180μs	19.7642μs	50.5965 KOps/s	47.4179 KOps/s	$\textbf{\color{#35bf28}+6.70\%}$
test_items	16.7320μs	4.2435μs	235.6559 KOps/s	238.2722 KOps/s	$\color{#d91a1a}-1.10\%$
test_items_nested	0.7107ms	0.4042ms	2.4740 KOps/s	2.4613 KOps/s	$\color{#35bf28}+0.52\%$
test_items_nested_locked	0.8414ms	0.4039ms	2.4758 KOps/s	2.4430 KOps/s	$\color{#35bf28}+1.34\%$
test_items_nested_leaf	0.1278ms	71.2121μs	14.0425 KOps/s	13.9226 KOps/s	$\color{#35bf28}+0.86\%$
test_items_stack_nested	0.7276ms	0.4091ms	2.4442 KOps/s	2.4282 KOps/s	$\color{#35bf28}+0.66\%$
test_items_stack_nested_leaf	0.1387ms	73.5699μs	13.5925 KOps/s	13.6515 KOps/s	$\color{#d91a1a}-0.43\%$
test_items_stack_nested_locked	0.5122ms	0.4082ms	2.4498 KOps/s	2.4474 KOps/s	$\color{#35bf28}+0.10\%$
test_keys	35.7470μs	3.7569μs	266.1742 KOps/s	285.0655 KOps/s	$\textbf{\color{#d91a1a}-6.63\%}$
test_keys_nested	0.2285ms	0.1359ms	7.3606 KOps/s	7.4519 KOps/s	$\color{#d91a1a}-1.23\%$
test_keys_nested_locked	1.7829ms	0.1424ms	7.0243 KOps/s	7.0035 KOps/s	$\color{#35bf28}+0.30\%$
test_keys_nested_leaf	0.2180ms	0.1156ms	8.6515 KOps/s	8.4954 KOps/s	$\color{#35bf28}+1.84\%$
test_keys_stack_nested	0.2341ms	0.1353ms	7.3895 KOps/s	7.3209 KOps/s	$\color{#35bf28}+0.94\%$
test_keys_stack_nested_leaf	0.2039ms	0.1155ms	8.6578 KOps/s	8.5326 KOps/s	$\color{#35bf28}+1.47\%$
test_keys_stack_nested_locked	0.2357ms	0.1404ms	7.1226 KOps/s	7.0100 KOps/s	$\color{#35bf28}+1.61\%$
test_values	5.0996μs	1.0409μs	960.7440 KOps/s	934.7764 KOps/s	$\color{#35bf28}+2.78\%$
test_values_nested	97.0520μs	54.7870μs	18.2525 KOps/s	18.3091 KOps/s	$\color{#d91a1a}-0.31\%$
test_values_nested_locked	0.1088ms	54.9583μs	18.1956 KOps/s	18.1470 KOps/s	$\color{#35bf28}+0.27\%$
test_values_nested_leaf	0.1080ms	59.3426μs	16.8513 KOps/s	16.9725 KOps/s	$\color{#d91a1a}-0.71\%$
test_values_stack_nested	0.1035ms	56.2108μs	17.7902 KOps/s	17.2884 KOps/s	$\color{#35bf28}+2.90\%$
test_values_stack_nested_leaf	0.1072ms	59.6684μs	16.7593 KOps/s	16.5222 KOps/s	$\color{#35bf28}+1.44\%$
test_values_stack_nested_locked	0.1006ms	56.5957μs	17.6692 KOps/s	17.8830 KOps/s	$\color{#d91a1a}-1.20\%$
test_membership	17.5430μs	0.8713μs	1.1477 MOps/s	1.0919 MOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_membership_nested	28.7640μs	2.9761μs	336.0128 KOps/s	336.0028 KOps/s	$+0.00\%$
test_membership_nested_leaf	25.9580μs	2.9890μs	334.5614 KOps/s	327.4718 KOps/s	$\color{#35bf28}+2.16\%$
test_membership_stacked_nested	37.6000μs	2.9235μs	342.0579 KOps/s	339.5888 KOps/s	$\color{#35bf28}+0.73\%$
test_membership_stacked_nested_leaf	26.8710μs	2.9298μs	341.3181 KOps/s	336.5030 KOps/s	$\color{#35bf28}+1.43\%$
test_membership_nested_last	28.9040μs	4.1754μs	239.4972 KOps/s	233.3659 KOps/s	$\color{#35bf28}+2.63\%$
test_membership_nested_leaf_last	32.9620μs	4.2084μs	237.6209 KOps/s	226.0014 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_membership_stacked_nested_last	25.8380μs	6.7857μs	147.3695 KOps/s	235.5626 KOps/s	$\textbf{\color{#d91a1a}-37.44\%}$
test_membership_stacked_nested_leaf_last	30.8080μs	6.9143μs	144.6285 KOps/s	233.7929 KOps/s	$\textbf{\color{#d91a1a}-38.14\%}$
test_nested_getleaf	45.9960μs	10.7500μs	93.0231 KOps/s	90.4860 KOps/s	$\color{#35bf28}+2.80\%$
test_nested_get	33.3120μs	10.2840μs	97.2386 KOps/s	95.3273 KOps/s	$\color{#35bf28}+2.00\%$
test_stacked_getleaf	38.2820μs	10.6563μs	93.8412 KOps/s	93.8100 KOps/s	$\color{#35bf28}+0.03\%$
test_stacked_get	33.3920μs	10.1648μs	98.3791 KOps/s	93.5739 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_nested_getitemleaf	39.4740μs	11.1137μs	89.9792 KOps/s	86.1600 KOps/s	$\color{#35bf28}+4.43\%$
test_nested_getitem	34.7350μs	10.4146μs	96.0191 KOps/s	92.8552 KOps/s	$\color{#35bf28}+3.41\%$
test_stacked_getitemleaf	39.1930μs	11.1275μs	89.8673 KOps/s	87.4239 KOps/s	$\color{#35bf28}+2.79\%$
test_stacked_getitem	36.2280μs	10.4203μs	95.9668 KOps/s	92.1939 KOps/s	$\color{#35bf28}+4.09\%$
test_lock_nested	4.3624ms	0.4473ms	2.2358 KOps/s	2.2333 KOps/s	$\color{#35bf28}+0.11\%$
test_lock_stack_nested	0.6418ms	0.4134ms	2.4191 KOps/s	2.3968 KOps/s	$\color{#35bf28}+0.93\%$
test_unlock_nested	0.8000ms	0.3649ms	2.7404 KOps/s	2.7310 KOps/s	$\color{#35bf28}+0.35\%$
test_unlock_stack_nested	0.6013ms	0.3333ms	3.0003 KOps/s	2.9823 KOps/s	$\color{#35bf28}+0.60\%$
test_flatten_speed	0.2075ms	95.8269μs	10.4355 KOps/s	10.6504 KOps/s	$\color{#d91a1a}-2.02\%$
test_unflatten_speed	0.6268ms	0.4962ms	2.0152 KOps/s	1.9790 KOps/s	$\color{#35bf28}+1.83\%$
test_common_ops	4.9016ms	0.7925ms	1.2618 KOps/s	1.2115 KOps/s	$\color{#35bf28}+4.16\%$
test_creation	21.0900μs	2.0811μs	480.5252 KOps/s	476.0072 KOps/s	$\color{#35bf28}+0.95\%$
test_creation_empty	36.2580μs	11.6071μs	86.1543 KOps/s	75.9211 KOps/s	$\textbf{\color{#35bf28}+13.48\%}$
test_creation_nested_1	40.0760μs	14.5256μs	68.8438 KOps/s	61.8194 KOps/s	$\textbf{\color{#35bf28}+11.36\%}$
test_creation_nested_2	58.0090μs	18.6396μs	53.6492 KOps/s	48.9421 KOps/s	$\textbf{\color{#35bf28}+9.62\%}$
test_clone	99.8670μs	13.8774μs	72.0595 KOps/s	75.1946 KOps/s	$\color{#d91a1a}-4.17\%$
test_getitem[int]	1.1568ms	12.5146μs	79.9065 KOps/s	81.0308 KOps/s	$\color{#d91a1a}-1.39\%$
test_getitem[slice_int]	0.1760ms	24.4122μs	40.9631 KOps/s	41.5725 KOps/s	$\color{#d91a1a}-1.47\%$
test_getitem[range]	0.1708ms	47.8416μs	20.9023 KOps/s	21.6778 KOps/s	$\color{#d91a1a}-3.58\%$
test_getitem[tuple]	0.1340ms	20.7147μs	48.2749 KOps/s	49.5560 KOps/s	$\color{#d91a1a}-2.59\%$
test_getitem[list]	0.2007ms	43.9880μs	22.7335 KOps/s	23.7884 KOps/s	$\color{#d91a1a}-4.43\%$
test_setitem_dim[int]	50.8950μs	26.1142μs	38.2933 KOps/s	40.1998 KOps/s	$\color{#d91a1a}-4.74\%$
test_setitem_dim[slice_int]	91.4720μs	51.9541μs	19.2477 KOps/s	19.5756 KOps/s	$\color{#d91a1a}-1.67\%$
test_setitem_dim[range]	99.4560μs	73.6257μs	13.5822 KOps/s	14.2111 KOps/s	$\color{#d91a1a}-4.43\%$
test_setitem_dim[tuple]	88.7260μs	41.4241μs	24.1405 KOps/s	24.2472 KOps/s	$\color{#d91a1a}-0.44\%$
test_setitem	76.1530μs	20.9564μs	47.7182 KOps/s	46.7075 KOps/s	$\color{#35bf28}+2.16\%$
test_set	81.2530μs	20.3888μs	49.0464 KOps/s	46.8730 KOps/s	$\color{#35bf28}+4.64\%$
test_set_shared	4.4097ms	0.1691ms	5.9142 KOps/s	5.9632 KOps/s	$\color{#d91a1a}-0.82\%$
test_update	0.1286ms	23.5929μs	42.3857 KOps/s	40.0529 KOps/s	$\textbf{\color{#35bf28}+5.82\%}$
test_update_nested	0.1032ms	33.8551μs	29.5377 KOps/s	28.2230 KOps/s	$\color{#35bf28}+4.66\%$
test_update__nested	0.8264ms	32.7737μs	30.5123 KOps/s	31.3144 KOps/s	$\color{#d91a1a}-2.56\%$
test_set_nested	78.0660μs	23.0289μs	43.4237 KOps/s	43.3717 KOps/s	$\color{#35bf28}+0.12\%$
test_set_nested_new	95.2890μs	27.4834μs	36.3856 KOps/s	36.0470 KOps/s	$\color{#35bf28}+0.94\%$
test_select	0.2711ms	44.2581μs	22.5948 KOps/s	22.5319 KOps/s	$\color{#35bf28}+0.28\%$
test_select_nested	0.1453ms	58.8664μs	16.9876 KOps/s	16.7523 KOps/s	$\color{#35bf28}+1.40\%$
test_exclude_nested	0.1667ms	78.3535μs	12.7627 KOps/s	12.7145 KOps/s	$\color{#35bf28}+0.38\%$
test_empty[True]	0.5904ms	0.3873ms	2.5817 KOps/s	2.5958 KOps/s	$\color{#d91a1a}-0.54\%$
test_empty[False]	12.8490μs	1.2430μs	804.5075 KOps/s	837.7783 KOps/s	$\color{#d91a1a}-3.97\%$
test_unbind_speed	0.5986ms	0.2621ms	3.8152 KOps/s	3.8309 KOps/s	$\color{#d91a1a}-0.41\%$
test_unbind_speed_stack0	0.5122ms	0.2568ms	3.8944 KOps/s	3.8580 KOps/s	$\color{#35bf28}+0.95\%$
test_unbind_speed_stack1	0.1092s	0.7642ms	1.3085 KOps/s	1.4210 KOps/s	$\textbf{\color{#d91a1a}-7.92\%}$
test_split	1.6386ms	1.5481ms	645.9541 Ops/s	574.8659 Ops/s	$\textbf{\color{#35bf28}+12.37\%}$
test_chunk	0.1096s	1.8941ms	527.9537 Ops/s	572.3047 Ops/s	$\textbf{\color{#d91a1a}-7.75\%}$
test_consolidate_njt[False-None]	8.6357ms	8.1298ms	123.0044 Ops/s	121.8489 Ops/s	$\color{#35bf28}+0.95\%$
test_creation[device0]	0.2371ms	91.4886μs	10.9303 KOps/s	10.9658 KOps/s	$\color{#d91a1a}-0.32\%$
test_creation_from_tensor	4.4997ms	97.3505μs	10.2722 KOps/s	10.3882 KOps/s	$\color{#d91a1a}-1.12\%$
test_add_one[memmap_tensor0]	0.1405ms	5.0669μs	197.3576 KOps/s	205.5693 KOps/s	$\color{#d91a1a}-3.99\%$
test_contiguous[memmap_tensor0]	14.9180μs	0.5235μs	1.9104 MOps/s	1.9362 MOps/s	$\color{#d91a1a}-1.33\%$
test_stack[memmap_tensor0]	44.0730μs	3.4084μs	293.3967 KOps/s	284.3373 KOps/s	$\color{#35bf28}+3.19\%$
test_memmaptd_index	1.0386ms	0.2373ms	4.2142 KOps/s	4.1794 KOps/s	$\color{#35bf28}+0.83\%$
test_memmaptd_index_astensor	0.6787ms	0.3173ms	3.1520 KOps/s	3.1569 KOps/s	$\color{#d91a1a}-0.15\%$
test_memmaptd_index_op	0.9982ms	0.5965ms	1.6765 KOps/s	1.6302 KOps/s	$\color{#35bf28}+2.84\%$
test_serialize_model	0.1207s	0.1153s	8.6748 Ops/s	7.5010 Ops/s	$\textbf{\color{#35bf28}+15.65\%}$
test_serialize_model_pickle	0.4871s	0.4013s	2.4916 Ops/s	2.4864 Ops/s	$\color{#35bf28}+0.21\%$
test_serialize_weights	0.2256s	0.1308s	7.6452 Ops/s	8.7468 Ops/s	$\textbf{\color{#d91a1a}-12.59\%}$
test_serialize_weights_returnearly	0.1909s	0.1632s	6.1288 Ops/s	6.3627 Ops/s	$\color{#d91a1a}-3.68\%$
test_serialize_weights_pickle	0.4481s	0.3933s	2.5425 Ops/s	2.2686 Ops/s	$\textbf{\color{#35bf28}+12.07\%}$
test_serialize_weights_filesystem	0.1515s	0.1420s	7.0421 Ops/s	6.2110 Ops/s	$\textbf{\color{#35bf28}+13.38\%}$
test_serialize_model_filesystem	0.1611s	0.1538s	6.5011 Ops/s	6.5208 Ops/s	$\color{#d91a1a}-0.30\%$
test_reshape_pytree	72.0850μs	26.9611μs	37.0904 KOps/s	36.5179 KOps/s	$\color{#35bf28}+1.57\%$
test_reshape_td	75.5020μs	32.4394μs	30.8267 KOps/s	30.7617 KOps/s	$\color{#35bf28}+0.21\%$
test_view_pytree	56.1150μs	26.6688μs	37.4970 KOps/s	36.9710 KOps/s	$\color{#35bf28}+1.42\%$
test_view_td	77.1340μs	38.1032μs	26.2445 KOps/s	25.6076 KOps/s	$\color{#35bf28}+2.49\%$
test_unbind_pytree	62.6270μs	29.6476μs	33.7295 KOps/s	33.3026 KOps/s	$\color{#35bf28}+1.28\%$
test_unbind_td	0.3538ms	38.9610μs	25.6667 KOps/s	26.5377 KOps/s	$\color{#d91a1a}-3.28\%$
test_split_pytree	91.1310μs	30.0959μs	33.2271 KOps/s	33.0591 KOps/s	$\color{#35bf28}+0.51\%$
test_split_td	0.2128ms	43.7852μs	22.8388 KOps/s	22.6969 KOps/s	$\color{#35bf28}+0.63\%$
test_add_pytree	89.6280μs	36.7092μs	27.2412 KOps/s	27.6737 KOps/s	$\color{#d91a1a}-1.56\%$
test_add_td	0.1185ms	56.1654μs	17.8046 KOps/s	17.3861 KOps/s	$\color{#35bf28}+2.41\%$
test_compile_add_one_nested[tensordict-compile]	0.1245ms	62.1849μs	16.0811 KOps/s	16.1000 KOps/s	$\color{#d91a1a}-0.12\%$
test_compile_add_one_nested[tensordict-eager]	0.4195ms	0.1612ms	6.2040 KOps/s	6.1892 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_add_one_nested[pytree-compile]	92.7940μs	44.3620μs	22.5418 KOps/s	22.1088 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_add_one_nested[pytree-eager]	0.1952ms	0.1193ms	8.3792 KOps/s	8.3808 KOps/s	$\color{#d91a1a}-0.02\%$
test_compile_copy_nested[tensordict-compile]	94.7950μs	26.4795μs	37.7650 KOps/s	39.1374 KOps/s	$\color{#d91a1a}-3.51\%$
test_compile_copy_nested[tensordict-eager]	0.1137ms	54.4612μs	18.3617 KOps/s	18.2116 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_copy_nested[pytree-compile]	0.1799ms	77.9489μs	12.8289 KOps/s	12.6373 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_copy_nested[pytree-eager]	0.1273ms	67.1146μs	14.8999 KOps/s	14.6264 KOps/s	$\color{#35bf28}+1.87\%$
test_compile_add_one_flat[tensordict-compile]	0.1863ms	0.1035ms	9.6645 KOps/s	9.4474 KOps/s	$\color{#35bf28}+2.30\%$
test_compile_add_one_flat[tensordict-eager]	0.3660ms	0.1989ms	5.0287 KOps/s	5.0505 KOps/s	$\color{#d91a1a}-0.43\%$
test_compile_add_one_flat[tensorclass-compile]	0.1354ms	44.3201μs	22.5631 KOps/s	22.4469 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_add_one_flat[tensorclass-eager]	0.5326ms	61.8628μs	16.1648 KOps/s	16.3234 KOps/s	$\color{#d91a1a}-0.97\%$
test_compile_add_one_flat[pytree-compile]	0.1987ms	0.1020ms	9.8085 KOps/s	9.6638 KOps/s	$\color{#35bf28}+1.50\%$
test_compile_add_one_flat[pytree-eager]	0.3770ms	0.2044ms	4.8928 KOps/s	4.9616 KOps/s	$\color{#d91a1a}-1.39\%$
test_compile_add_self_flat[tensordict-eager]	0.3728ms	0.2097ms	4.7697 KOps/s	4.7357 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_add_self_flat[tensordict-compile]	0.3418ms	0.1051ms	9.5104 KOps/s	9.3759 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_add_self_flat[tensorclass-eager]	0.1654ms	57.5962μs	17.3623 KOps/s	18.2886 KOps/s	$\textbf{\color{#d91a1a}-5.07\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1045ms	44.3251μs	22.5606 KOps/s	21.6808 KOps/s	$\color{#35bf28}+4.06\%$
test_compile_add_self_flat[pytree-eager]	0.6388ms	0.1591ms	6.2860 KOps/s	6.2855 KOps/s	$+0.01\%$
test_compile_add_self_flat[pytree-compile]	0.2706ms	0.1025ms	9.7564 KOps/s	9.6401 KOps/s	$\color{#35bf28}+1.21\%$
test_compile_copy_flat[tensordict-compile]	0.2288ms	21.0799μs	47.4386 KOps/s	46.8727 KOps/s	$\color{#35bf28}+1.21\%$
test_compile_copy_flat[tensordict-eager]	0.1562ms	58.8894μs	16.9810 KOps/s	16.7309 KOps/s	$\color{#35bf28}+1.49\%$
test_compile_copy_flat[pytree-compile]	0.1864ms	80.5628μs	12.4127 KOps/s	12.1999 KOps/s	$\color{#35bf28}+1.74\%$
test_compile_copy_flat[pytree-eager]	0.1252ms	68.5558μs	14.5867 KOps/s	14.1724 KOps/s	$\color{#35bf28}+2.92\%$
test_compile_assign_and_add[tensordict-compile]	0.3095ms	0.2057ms	4.8614 KOps/s	4.7123 KOps/s	$\color{#35bf28}+3.16\%$
test_compile_assign_and_add[tensordict-eager]	1.5751ms	1.2949ms	772.2555 Ops/s	758.7459 Ops/s	$\color{#35bf28}+1.78\%$
test_compile_assign_and_add[pytree-compile]	0.3647ms	0.2041ms	4.8996 KOps/s	4.8533 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_assign_and_add[pytree-eager]	0.9689ms	0.7974ms	1.2541 KOps/s	1.2371 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_assign_and_add_stack[compile]	0.8249ms	0.4527ms	2.2089 KOps/s	2.1335 KOps/s	$\color{#35bf28}+3.53\%$
test_compile_assign_and_add_stack[eager]	2.9708ms	2.6901ms	371.7346 Ops/s	360.2048 Ops/s	$\color{#35bf28}+3.20\%$
test_compile_indexing[tensor-tensordict-compile]	0.1080ms	34.7184μs	28.8031 KOps/s	27.9468 KOps/s	$\color{#35bf28}+3.06\%$
test_compile_indexing[tensor-tensordict-eager]	0.5908ms	33.1439μs	30.1715 KOps/s	31.5407 KOps/s	$\color{#d91a1a}-4.34\%$
test_compile_indexing[tensor-tensorclass-compile]	99.6570μs	29.0521μs	34.4209 KOps/s	34.2954 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_indexing[tensor-tensorclass-eager]	63.3780μs	23.4834μs	42.5833 KOps/s	41.9079 KOps/s	$\color{#35bf28}+1.61\%$
test_compile_indexing[tensor-pytree-compile]	0.1012ms	29.7082μs	33.6607 KOps/s	33.3214 KOps/s	$\color{#35bf28}+1.02\%$
test_compile_indexing[tensor-pytree-eager]	0.1056ms	23.5160μs	42.5243 KOps/s	42.6001 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_indexing[slice-tensordict-compile]	0.1146ms	50.2657μs	19.8943 KOps/s	19.4095 KOps/s	$\color{#35bf28}+2.50\%$
test_compile_indexing[slice-tensordict-eager]	0.6194ms	20.2284μs	49.4355 KOps/s	49.1582 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_indexing[slice-tensorclass-compile]	0.1167ms	43.0568μs	23.2251 KOps/s	22.8380 KOps/s	$\color{#35bf28}+1.70\%$
test_compile_indexing[slice-tensorclass-eager]	55.9850μs	18.7329μs	53.3821 KOps/s	51.9839 KOps/s	$\color{#35bf28}+2.69\%$
test_compile_indexing[slice-pytree-compile]	0.1360ms	43.7141μs	22.8759 KOps/s	22.5788 KOps/s	$\color{#35bf28}+1.32\%$
test_compile_indexing[slice-pytree-eager]	61.6460μs	19.0133μs	52.5947 KOps/s	51.9315 KOps/s	$\color{#35bf28}+1.28\%$
test_compile_indexing[int-tensordict-compile]	0.1238ms	51.5194μs	19.4102 KOps/s	19.3145 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_indexing[int-tensordict-eager]	1.0254ms	20.0495μs	49.8765 KOps/s	49.7810 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_indexing[int-tensorclass-compile]	0.1417ms	43.8522μs	22.8039 KOps/s	22.4324 KOps/s	$\color{#35bf28}+1.66\%$
test_compile_indexing[int-tensorclass-eager]	78.4570μs	18.9449μs	52.7847 KOps/s	52.9021 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_indexing[int-pytree-compile]	0.1406ms	44.0744μs	22.6889 KOps/s	22.4790 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_indexing[int-pytree-eager]	62.5970μs	18.7880μs	53.2254 KOps/s	52.2878 KOps/s	$\color{#35bf28}+1.79\%$
test_mod_add[eager]	97.5930μs	34.4855μs	28.9977 KOps/s	27.2345 KOps/s	$\textbf{\color{#35bf28}+6.47\%}$
test_mod_add[compile]	0.1724ms	46.8483μs	21.3455 KOps/s	21.1219 KOps/s	$\color{#35bf28}+1.06\%$
test_mod_add[compile-overhead]	0.1299ms	46.4167μs	21.5440 KOps/s	21.0480 KOps/s	$\color{#35bf28}+2.36\%$
test_mod_wrap[eager]	0.4359ms	0.2260ms	4.4255 KOps/s	4.3594 KOps/s	$\color{#35bf28}+1.52\%$
test_mod_wrap[compile]	0.3061ms	0.2040ms	4.9026 KOps/s	4.7362 KOps/s	$\color{#35bf28}+3.51\%$
test_mod_wrap[compile-overhead]	0.3794ms	0.2034ms	4.9161 KOps/s	4.6939 KOps/s	$\color{#35bf28}+4.73\%$
test_mod_wrap_and_backward[eager]	16.2453ms	11.6811ms	85.6081 Ops/s	86.8469 Ops/s	$\color{#d91a1a}-1.43\%$
test_mod_wrap_and_backward[compile]	27.4175ms	13.7730ms	72.6060 Ops/s	90.0263 Ops/s	$\textbf{\color{#d91a1a}-19.35\%}$
test_mod_wrap_and_backward[compile-overhead]	13.5952ms	12.3054ms	81.2652 Ops/s	89.4400 Ops/s	$\textbf{\color{#d91a1a}-9.14\%}$
test_seq_add[eager]	0.2003ms	0.1119ms	8.9343 KOps/s	8.4117 KOps/s	$\textbf{\color{#35bf28}+6.21\%}$
test_seq_add[compile]	0.1285ms	59.6120μs	16.7751 KOps/s	16.3163 KOps/s	$\color{#35bf28}+2.81\%$
test_seq_add[compile-overhead]	0.1410ms	58.4990μs	17.0943 KOps/s	16.5111 KOps/s	$\color{#35bf28}+3.53\%$
test_seq_wrap[eager]	0.5621ms	0.4337ms	2.3060 KOps/s	2.0659 KOps/s	$\textbf{\color{#35bf28}+11.62\%}$
test_seq_wrap[compile]	0.3367ms	0.2230ms	4.4837 KOps/s	4.3232 KOps/s	$\color{#35bf28}+3.71\%$
test_seq_wrap[compile-overhead]	0.3312ms	0.2228ms	4.4887 KOps/s	4.2872 KOps/s	$\color{#35bf28}+4.70\%$
test_func_call_runtime[False-eager]	0.6943ms	0.5313ms	1.8821 KOps/s	1.7546 KOps/s	$\textbf{\color{#35bf28}+7.27\%}$
test_func_call_runtime[False-compile]	0.7864ms	0.4214ms	2.3729 KOps/s	2.3227 KOps/s	$\color{#35bf28}+2.16\%$
test_func_call_runtime[False-compile-overhead]	0.7290ms	0.4197ms	2.3828 KOps/s	2.3027 KOps/s	$\color{#35bf28}+3.48\%$
test_func_call_runtime[True-eager]	1.0235ms	0.7448ms	1.3426 KOps/s	1.2862 KOps/s	$\color{#35bf28}+4.39\%$
test_func_call_runtime[True-compile]	0.5506ms	0.4578ms	2.1842 KOps/s	2.1272 KOps/s	$\color{#35bf28}+2.68\%$
test_func_call_runtime[True-compile-overhead]	0.8543ms	0.4657ms	2.1475 KOps/s	2.1335 KOps/s	$\color{#35bf28}+0.66\%$
test_func_call_cm_runtime[False-eager]	0.8989ms	0.5396ms	1.8534 KOps/s	1.7712 KOps/s	$\color{#35bf28}+4.64\%$
test_func_call_cm_runtime[False-compile]	1.0970ms	0.4308ms	2.3214 KOps/s	2.3331 KOps/s	$\color{#d91a1a}-0.50\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6263ms	0.4200ms	2.3811 KOps/s	2.3219 KOps/s	$\color{#35bf28}+2.55\%$
test_func_call_cm_runtime[True-eager]	1.0917ms	0.8721ms	1.1467 KOps/s	1.0928 KOps/s	$\color{#35bf28}+4.93\%$
test_func_call_cm_runtime[True-compile]	1.0529ms	0.4995ms	2.0021 KOps/s	2.0209 KOps/s	$\color{#d91a1a}-0.93\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5863ms	0.4840ms	2.0663 KOps/s	2.0283 KOps/s	$\color{#35bf28}+1.87\%$
test_vmap_func_call_cm_runtime[eager]	3.0564ms	1.8608ms	537.4017 Ops/s	528.1735 Ops/s	$\color{#35bf28}+1.75\%$
test_vmap_func_call_cm_runtime[compile]	1.4023ms	0.6966ms	1.4356 KOps/s	1.9071 KOps/s	$\textbf{\color{#d91a1a}-24.73\%}$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9878ms	0.5177ms	1.9315 KOps/s	1.9069 KOps/s	$\color{#35bf28}+1.29\%$
test_distributed	0.2453ms	0.1254ms	7.9759 KOps/s	7.8273 KOps/s	$\color{#35bf28}+1.90\%$
test_tdmodule	94.0360μs	25.9080μs	38.5982 KOps/s	36.8010 KOps/s	$\color{#35bf28}+4.88\%$
test_tdmodule_dispatch	83.9470μs	47.5835μs	21.0157 KOps/s	19.6240 KOps/s	$\textbf{\color{#35bf28}+7.09\%}$
test_tdseq	66.3250μs	26.3243μs	37.9878 KOps/s	36.0445 KOps/s	$\textbf{\color{#35bf28}+5.39\%}$
test_tdseq_dispatch	83.2260μs	50.5048μs	19.8001 KOps/s	18.1386 KOps/s	$\textbf{\color{#35bf28}+9.16\%}$
test_instantiation_functorch	1.7272ms	1.5321ms	652.6888 Ops/s	635.6138 Ops/s	$\color{#35bf28}+2.69\%$
test_exec_functorch	0.4167ms	0.1824ms	5.4836 KOps/s	5.4825 KOps/s	$\color{#35bf28}+0.02\%$
test_exec_functional_call	0.2701ms	0.1685ms	5.9350 KOps/s	5.7463 KOps/s	$\color{#35bf28}+3.28\%$
test_exec_td_decorator	0.4825ms	0.2279ms	4.3871 KOps/s	4.3983 KOps/s	$\color{#d91a1a}-0.26\%$
test_vmap_mlp_speed_decorator[True-True]	0.9331ms	0.6452ms	1.5500 KOps/s	1.4734 KOps/s	$\textbf{\color{#35bf28}+5.20\%}$
test_vmap_mlp_speed_decorator[True-False]	1.1571ms	0.6464ms	1.5470 KOps/s	1.5029 KOps/s	$\color{#35bf28}+2.94\%$
test_vmap_mlp_speed_decorator[False-True]	1.1514ms	0.5325ms	1.8779 KOps/s	1.8801 KOps/s	$\color{#d91a1a}-0.12\%$
test_vmap_mlp_speed_decorator[False-False]	0.7908ms	0.5212ms	1.9187 KOps/s	1.8706 KOps/s	$\color{#35bf28}+2.57\%$
test_to_module_speed[True]	1.4353ms	1.2611ms	792.9462 Ops/s	775.8576 Ops/s	$\color{#35bf28}+2.20\%$
test_to_module_speed[False]	2.0984ms	1.2556ms	796.4088 Ops/s	784.6562 Ops/s	$\color{#35bf28}+1.50\%$
test_tc_init	86.9330μs	47.3574μs	21.1160 KOps/s	19.5691 KOps/s	$\textbf{\color{#35bf28}+7.90\%}$
test_tc_init_nested	0.3994ms	92.6959μs	10.7880 KOps/s	10.0850 KOps/s	$\textbf{\color{#35bf28}+6.97\%}$
test_tc_first_layer_tensor	19.8970μs	1.5233μs	656.4779 KOps/s	658.2538 KOps/s	$\color{#d91a1a}-0.27\%$
test_tc_first_layer_nontensor	50.7250μs	4.7304μs	211.4003 KOps/s	217.8521 KOps/s	$\color{#d91a1a}-2.96\%$
test_tc_second_layer_tensor	33.3430μs	2.8292μs	353.4571 KOps/s	359.6773 KOps/s	$\color{#d91a1a}-1.73\%$
test_tc_second_layer_nontensor	49.3530μs	6.0789μs	164.5044 KOps/s	168.5055 KOps/s	$\color{#d91a1a}-2.37\%$
test_unbind	0.2386s	13.6232ms	73.4040 Ops/s	77.6447 Ops/s	$\textbf{\color{#d91a1a}-5.46\%}$
test_full_like	17.5165ms	12.9892ms	76.9871 Ops/s	79.2393 Ops/s	$\color{#d91a1a}-2.84\%$
test_zeros_like	14.4381ms	7.6269ms	131.1150 Ops/s	135.6396 Ops/s	$\color{#d91a1a}-3.34\%$
test_ones_like	13.1655ms	8.5864ms	116.4631 Ops/s	123.3355 Ops/s	$\textbf{\color{#d91a1a}-5.57\%}$
test_clone	16.0085ms	10.3341ms	96.7670 Ops/s	98.4681 Ops/s	$\color{#d91a1a}-1.73\%$
test_squeeze	70.3720μs	12.0977μs	82.6601 KOps/s	83.6696 KOps/s	$\color{#d91a1a}-1.21\%$
test_unsqueeze	0.1873ms	88.8992μs	11.2487 KOps/s	11.1233 KOps/s	$\color{#35bf28}+1.13\%$
test_split	0.3781ms	0.1942ms	5.1486 KOps/s	5.2346 KOps/s	$\color{#d91a1a}-1.64\%$
test_permute	0.3609ms	0.2184ms	4.5798 KOps/s	4.5058 KOps/s	$\color{#35bf28}+1.64\%$
test_stack	31.4993ms	25.4088ms	39.3565 Ops/s	39.8716 Ops/s	$\color{#d91a1a}-1.29\%$
test_cat	31.1507ms	25.0807ms	39.8713 Ops/s	40.5947 Ops/s	$\color{#d91a1a}-1.78\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}28$. Worsened: $\large\color{#d91a1a}18$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	43.8010μs	10.8803μs	91.9091 KOps/s	97.3820 KOps/s	$\textbf{\color{#d91a1a}-5.62\%}$
test_plain_set_stack_nested	31.0910μs	10.8148μs	92.4663 KOps/s	97.0621 KOps/s	$\color{#d91a1a}-4.73\%$
test_plain_set_nested_inplace	36.1910μs	11.6867μs	85.5673 KOps/s	89.8763 KOps/s	$\color{#d91a1a}-4.79\%$
test_plain_set_stack_nested_inplace	33.9000μs	11.6549μs	85.8010 KOps/s	90.5466 KOps/s	$\textbf{\color{#d91a1a}-5.24\%}$
test_items	44.3410μs	2.8932μs	345.6345 KOps/s	346.6423 KOps/s	$\color{#d91a1a}-0.29\%$
test_items_nested	0.3927ms	0.3544ms	2.8219 KOps/s	2.8146 KOps/s	$\color{#35bf28}+0.26\%$
test_items_nested_locked	0.3958ms	0.3564ms	2.8055 KOps/s	2.8117 KOps/s	$\color{#d91a1a}-0.22\%$
test_items_nested_leaf	0.1089ms	61.2737μs	16.3202 KOps/s	16.4471 KOps/s	$\color{#d91a1a}-0.77\%$
test_items_stack_nested	0.3999ms	0.3588ms	2.7873 KOps/s	2.8058 KOps/s	$\color{#d91a1a}-0.66\%$
test_items_stack_nested_leaf	92.4110μs	62.4529μs	16.0121 KOps/s	16.0484 KOps/s	$\color{#d91a1a}-0.23\%$
test_items_stack_nested_locked	0.4060ms	0.3567ms	2.8031 KOps/s	2.8111 KOps/s	$\color{#d91a1a}-0.29\%$
test_keys	27.3100μs	3.4736μs	287.8863 KOps/s	288.1896 KOps/s	$\color{#d91a1a}-0.11\%$
test_keys_nested	0.1292ms	71.3803μs	14.0095 KOps/s	14.0584 KOps/s	$\color{#d91a1a}-0.35\%$
test_keys_nested_locked	0.8121ms	76.5667μs	13.0605 KOps/s	13.0135 KOps/s	$\color{#35bf28}+0.36\%$
test_keys_nested_leaf	0.1007ms	62.1496μs	16.0902 KOps/s	16.1299 KOps/s	$\color{#d91a1a}-0.25\%$
test_keys_stack_nested	0.1046ms	70.4246μs	14.1996 KOps/s	14.1432 KOps/s	$\color{#35bf28}+0.40\%$
test_keys_stack_nested_leaf	90.7710μs	62.2615μs	16.0613 KOps/s	16.1766 KOps/s	$\color{#d91a1a}-0.71\%$
test_keys_stack_nested_locked	0.1085ms	76.4461μs	13.0811 KOps/s	12.9595 KOps/s	$\color{#35bf28}+0.94\%$
test_values	9.4583μs	0.8590μs	1.1641 MOps/s	1.1259 MOps/s	$\color{#35bf28}+3.39\%$
test_values_nested	57.0010μs	31.7898μs	31.4566 KOps/s	31.8364 KOps/s	$\color{#d91a1a}-1.19\%$
test_values_nested_locked	63.4110μs	33.1552μs	30.1612 KOps/s	30.2845 KOps/s	$\color{#d91a1a}-0.41\%$
test_values_nested_leaf	63.9310μs	34.0183μs	29.3959 KOps/s	29.6014 KOps/s	$\color{#d91a1a}-0.69\%$
test_values_stack_nested	62.0110μs	32.1445μs	31.1095 KOps/s	31.2960 KOps/s	$\color{#d91a1a}-0.60\%$
test_values_stack_nested_leaf	68.0510μs	34.6186μs	28.8862 KOps/s	29.1686 KOps/s	$\color{#d91a1a}-0.97\%$
test_values_stack_nested_locked	68.3410μs	33.7027μs	29.6713 KOps/s	29.7607 KOps/s	$\color{#d91a1a}-0.30\%$
test_membership	1.6420μs	0.5098μs	1.9617 MOps/s	1.9500 MOps/s	$\color{#35bf28}+0.60\%$
test_membership_nested	23.6200μs	2.0010μs	499.7600 KOps/s	486.5670 KOps/s	$\color{#35bf28}+2.71\%$
test_membership_nested_leaf	15.9950μs	1.9948μs	501.3122 KOps/s	487.4238 KOps/s	$\color{#35bf28}+2.85\%$
test_membership_stacked_nested	33.6310μs	2.0858μs	479.4381 KOps/s	474.8514 KOps/s	$\color{#35bf28}+0.97\%$
test_membership_stacked_nested_leaf	32.9210μs	2.0809μs	480.5645 KOps/s	479.0693 KOps/s	$\color{#35bf28}+0.31\%$
test_membership_nested_last	30.9000μs	2.9437μs	339.7037 KOps/s	336.0648 KOps/s	$\color{#35bf28}+1.08\%$
test_membership_nested_leaf_last	32.5700μs	2.9582μs	338.0387 KOps/s	331.6938 KOps/s	$\color{#35bf28}+1.91\%$
test_membership_stacked_nested_last	39.5310μs	3.5843μs	278.9960 KOps/s	126.9636 KOps/s	$\textbf{\color{#35bf28}+119.74\%}$
test_membership_stacked_nested_leaf_last	25.9800μs	3.6109μs	276.9359 KOps/s	125.6250 KOps/s	$\textbf{\color{#35bf28}+120.45\%}$
test_nested_getleaf	37.2300μs	6.1115μs	163.6252 KOps/s	162.7256 KOps/s	$\color{#35bf28}+0.55\%$
test_nested_get	26.8500μs	5.8033μs	172.3170 KOps/s	170.4617 KOps/s	$\color{#35bf28}+1.09\%$
test_stacked_getleaf	62.3610μs	6.1261μs	163.2350 KOps/s	163.4221 KOps/s	$\color{#d91a1a}-0.11\%$
test_stacked_get	26.6110μs	5.8301μs	171.5249 KOps/s	171.4546 KOps/s	$\color{#35bf28}+0.04\%$
test_nested_getitemleaf	37.4210μs	6.2434μs	160.1701 KOps/s	164.3473 KOps/s	$\color{#d91a1a}-2.54\%$
test_nested_getitem	25.7800μs	5.8875μs	169.8528 KOps/s	170.2617 KOps/s	$\color{#d91a1a}-0.24\%$
test_stacked_getitemleaf	32.8510μs	6.1929μs	161.4755 KOps/s	160.5391 KOps/s	$\color{#35bf28}+0.58\%$
test_stacked_getitem	27.2110μs	5.8724μs	170.2889 KOps/s	170.4852 KOps/s	$\color{#d91a1a}-0.12\%$
test_lock_nested	9.9921ms	0.3821ms	2.6171 KOps/s	2.5864 KOps/s	$\color{#35bf28}+1.19\%$
test_lock_stack_nested	0.3702ms	0.3452ms	2.8970 KOps/s	2.8922 KOps/s	$\color{#35bf28}+0.17\%$
test_unlock_nested	0.6452ms	0.3051ms	3.2778 KOps/s	3.2159 KOps/s	$\color{#35bf28}+1.92\%$
test_unlock_stack_nested	0.3114ms	0.2736ms	3.6552 KOps/s	3.6529 KOps/s	$\color{#35bf28}+0.06\%$
test_flatten_speed	0.1113ms	78.2994μs	12.7715 KOps/s	12.7808 KOps/s	$\color{#d91a1a}-0.07\%$
test_unflatten_speed	0.3734ms	0.3081ms	3.2461 KOps/s	3.2433 KOps/s	$\color{#35bf28}+0.09\%$
test_common_ops	1.7694ms	0.5896ms	1.6962 KOps/s	1.7432 KOps/s	$\color{#d91a1a}-2.70\%$
test_creation	0.1719ms	1.4967μs	668.1266 KOps/s	670.3203 KOps/s	$\color{#d91a1a}-0.33\%$
test_creation_empty	31.5900μs	7.9836μs	125.2575 KOps/s	149.4506 KOps/s	$\textbf{\color{#d91a1a}-16.19\%}$
test_creation_nested_1	31.8410μs	9.6430μs	103.7017 KOps/s	120.8563 KOps/s	$\textbf{\color{#d91a1a}-14.19\%}$
test_creation_nested_2	47.5910μs	12.1825μs	82.0851 KOps/s	92.2831 KOps/s	$\textbf{\color{#d91a1a}-11.05\%}$
test_clone	71.7910μs	9.7973μs	102.0692 KOps/s	97.0152 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_getitem[int]	1.6508ms	10.7609μs	92.9287 KOps/s	89.0781 KOps/s	$\color{#35bf28}+4.32\%$
test_getitem[slice_int]	0.1130ms	20.7598μs	48.1701 KOps/s	43.8751 KOps/s	$\textbf{\color{#35bf28}+9.79\%}$
test_getitem[range]	0.1322ms	35.6211μs	28.0733 KOps/s	25.8394 KOps/s	$\textbf{\color{#35bf28}+8.65\%}$
test_getitem[tuple]	0.1087ms	18.2429μs	54.8159 KOps/s	52.8762 KOps/s	$\color{#35bf28}+3.67\%$
test_getitem[list]	0.2631ms	31.8500μs	31.3971 KOps/s	31.4773 KOps/s	$\color{#d91a1a}-0.25\%$
test_setitem_dim[int]	37.2600μs	17.0978μs	58.4869 KOps/s	54.1124 KOps/s	$\textbf{\color{#35bf28}+8.08\%}$
test_setitem_dim[slice_int]	58.3110μs	37.0601μs	26.9832 KOps/s	26.2916 KOps/s	$\color{#35bf28}+2.63\%$
test_setitem_dim[range]	93.8710μs	52.1673μs	19.1691 KOps/s	19.1303 KOps/s	$\color{#35bf28}+0.20\%$
test_setitem_dim[tuple]	59.9110μs	31.9178μs	31.3305 KOps/s	31.0454 KOps/s	$\color{#35bf28}+0.92\%$
test_setitem	83.8910μs	14.2077μs	70.3843 KOps/s	71.3248 KOps/s	$\color{#d91a1a}-1.32\%$
test_set	91.1910μs	14.3033μs	69.9138 KOps/s	75.0236 KOps/s	$\textbf{\color{#d91a1a}-6.81\%}$
test_set_shared	1.6226ms	0.1441ms	6.9382 KOps/s	6.8018 KOps/s	$\color{#35bf28}+2.01\%$
test_update	0.3188ms	16.9747μs	58.9113 KOps/s	63.9741 KOps/s	$\textbf{\color{#d91a1a}-7.91\%}$
test_update_nested	92.0720μs	22.1285μs	45.1906 KOps/s	50.8973 KOps/s	$\textbf{\color{#d91a1a}-11.21\%}$
test_update__nested	0.5206ms	22.8282μs	43.8054 KOps/s	42.1285 KOps/s	$\color{#35bf28}+3.98\%$
test_set_nested	79.6010μs	14.6033μs	68.4775 KOps/s	70.0645 KOps/s	$\color{#d91a1a}-2.27\%$
test_set_nested_new	0.1256ms	16.9170μs	59.1123 KOps/s	60.5676 KOps/s	$\color{#d91a1a}-2.40\%$
test_select	0.1041ms	29.4187μs	33.9920 KOps/s	35.2319 KOps/s	$\color{#d91a1a}-3.52\%$
test_select_nested	81.1310μs	44.1285μs	22.6611 KOps/s	22.6478 KOps/s	$\color{#35bf28}+0.06\%$
test_exclude_nested	0.1039ms	63.2825μs	15.8022 KOps/s	15.7793 KOps/s	$\color{#35bf28}+0.14\%$
test_empty[True]	0.3310ms	0.2804ms	3.5658 KOps/s	3.5435 KOps/s	$\color{#35bf28}+0.63\%$
test_empty[False]	3.8270μs	0.7540μs	1.3263 MOps/s	1.3367 MOps/s	$\color{#d91a1a}-0.78\%$
test_to	92.3810μs	54.7280μs	18.2722 KOps/s	16.8105 KOps/s	$\textbf{\color{#35bf28}+8.70\%}$
test_to_nonblocking	86.8210μs	45.1703μs	22.1385 KOps/s	21.1796 KOps/s	$\color{#35bf28}+4.53\%$
test_unbind_speed	0.2773ms	0.2286ms	4.3751 KOps/s	4.3003 KOps/s	$\color{#35bf28}+1.74\%$
test_unbind_speed_stack0	0.2829ms	0.2295ms	4.3566 KOps/s	4.3268 KOps/s	$\color{#35bf28}+0.69\%$
test_unbind_speed_stack1	0.1018s	0.6394ms	1.5640 KOps/s	1.5388 KOps/s	$\color{#35bf28}+1.64\%$
test_split	0.1062s	1.6019ms	624.2628 Ops/s	565.8253 Ops/s	$\textbf{\color{#35bf28}+10.33\%}$
test_chunk	99.4837ms	1.5974ms	626.0079 Ops/s	676.5325 Ops/s	$\textbf{\color{#d91a1a}-7.47\%}$
test_consolidate[False-None]	0.1011s	2.9064ms	344.0671 Ops/s	343.0686 Ops/s	$\color{#35bf28}+0.29\%$
test_consolidate[default-None]	1.7657ms	1.6736ms	597.5267 Ops/s	588.1606 Ops/s	$\color{#35bf28}+1.59\%$
test_consolidate[reduce-overhead-None]	1.8440ms	1.7166ms	582.5544 Ops/s	577.3076 Ops/s	$\color{#35bf28}+0.91\%$
test_consolidate_njt[False-None]	7.1904ms	6.7937ms	147.1961 Ops/s	148.7703 Ops/s	$\color{#d91a1a}-1.06\%$
test_to[False-False-None]	1.7869ms	1.7001ms	588.2138 Ops/s	590.7492 Ops/s	$\color{#d91a1a}-0.43\%$
test_to[True-False-None]	0.3208s	1.7780ms	562.4210 Ops/s	758.3630 Ops/s	$\textbf{\color{#d91a1a}-25.84\%}$
test_to[within-False-None]	4.1353ms	4.0111ms	249.3098 Ops/s	244.0777 Ops/s	$\color{#35bf28}+2.14\%$
test_to[True-default-None]	5.4435ms	5.2071ms	192.0470 Ops/s	191.3753 Ops/s	$\color{#35bf28}+0.35\%$
test_to_njt[False-False-None]	7.1907ms	7.0615ms	141.6127 Ops/s	141.5102 Ops/s	$\color{#35bf28}+0.07\%$
test_to_njt[True-False-None]	6.0905ms	5.6548ms	176.8419 Ops/s	181.9529 Ops/s	$\color{#d91a1a}-2.81\%$
test_to_njt[within-False-None]	12.5378ms	12.4128ms	80.5623 Ops/s	80.4840 Ops/s	$\color{#35bf28}+0.10\%$
test_creation[device0]	0.5383ms	79.8472μs	12.5239 KOps/s	12.6119 KOps/s	$\color{#d91a1a}-0.70\%$
test_creation_from_tensor	0.6138ms	83.4106μs	11.9889 KOps/s	12.1050 KOps/s	$\color{#d91a1a}-0.96\%$
test_add_one[memmap_tensor0]	0.2251ms	6.1176μs	163.4632 KOps/s	154.1545 KOps/s	$\textbf{\color{#35bf28}+6.04\%}$
test_contiguous[memmap_tensor0]	2.0810μs	0.4089μs	2.4453 MOps/s	2.4238 MOps/s	$\color{#35bf28}+0.89\%$
test_stack[memmap_tensor0]	23.9900μs	4.4367μs	225.3939 KOps/s	214.2760 KOps/s	$\textbf{\color{#35bf28}+5.19\%}$
test_memmaptd_index	2.1919ms	0.2428ms	4.1194 KOps/s	3.9456 KOps/s	$\color{#35bf28}+4.40\%$
test_memmaptd_index_astensor	0.5642ms	0.3010ms	3.3226 KOps/s	3.1457 KOps/s	$\textbf{\color{#35bf28}+5.62\%}$
test_memmaptd_index_op	0.9559ms	0.5520ms	1.8114 KOps/s	1.7853 KOps/s	$\color{#35bf28}+1.47\%$
test_serialize_model	0.1325s	0.1315s	7.6048 Ops/s	7.6127 Ops/s	$\color{#d91a1a}-0.10\%$
test_serialize_model_pickle	1.3514s	1.2121s	0.8250 Ops/s	0.8418 Ops/s	$\color{#d91a1a}-2.00\%$
test_serialize_weights	0.2971s	0.1541s	6.4873 Ops/s	5.3509 Ops/s	$\textbf{\color{#35bf28}+21.24\%}$
test_serialize_weights_returnearly	0.3693s	55.5439ms	18.0038 Ops/s	23.2590 Ops/s	$\textbf{\color{#d91a1a}-22.59\%}$
test_serialize_weights_pickle	1.9406s	1.4565s	0.6866 Ops/s	0.8246 Ops/s	$\textbf{\color{#d91a1a}-16.74\%}$
test_reshape_pytree	84.1310μs	22.4470μs	44.5493 KOps/s	42.4750 KOps/s	$\color{#35bf28}+4.88\%$
test_reshape_td	64.1710μs	26.1558μs	38.2324 KOps/s	36.5482 KOps/s	$\color{#35bf28}+4.61\%$
test_view_pytree	54.5110μs	22.0676μs	45.3153 KOps/s	44.3262 KOps/s	$\color{#35bf28}+2.23\%$
test_view_td	71.7510μs	30.9387μs	32.3220 KOps/s	34.0278 KOps/s	$\textbf{\color{#d91a1a}-5.01\%}$
test_unbind_pytree	60.4610μs	27.6648μs	36.1470 KOps/s	35.3034 KOps/s	$\color{#35bf28}+2.39\%$
test_unbind_td	0.6449ms	35.1848μs	28.4214 KOps/s	27.7745 KOps/s	$\color{#35bf28}+2.33\%$
test_split_pytree	67.7820μs	29.8962μs	33.4491 KOps/s	32.7976 KOps/s	$\color{#35bf28}+1.99\%$
test_split_td	0.7576ms	38.8214μs	25.7590 KOps/s	25.0897 KOps/s	$\color{#35bf28}+2.67\%$
test_add_pytree	64.7010μs	32.5385μs	30.7328 KOps/s	29.7318 KOps/s	$\color{#35bf28}+3.37\%$
test_add_td	88.6520μs	43.6917μs	22.8877 KOps/s	23.5658 KOps/s	$\color{#d91a1a}-2.88\%$
test_compile_add_one_nested[tensordict-compile]	0.1722ms	0.1216ms	8.2258 KOps/s	7.8903 KOps/s	$\color{#35bf28}+4.25\%$
test_compile_add_one_nested[tensordict-eager]	0.2418ms	0.1268ms	7.8884 KOps/s	7.8579 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_add_one_nested[pytree-compile]	0.1435ms	98.9938μs	10.1016 KOps/s	9.7193 KOps/s	$\color{#35bf28}+3.93\%$
test_compile_add_one_nested[pytree-eager]	1.0954ms	0.1501ms	6.6616 KOps/s	6.6688 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_copy_nested[tensordict-compile]	71.7020μs	23.1023μs	43.2857 KOps/s	44.8304 KOps/s	$\color{#d91a1a}-3.45\%$
test_compile_copy_nested[tensordict-eager]	66.2010μs	28.0932μs	35.5958 KOps/s	35.4414 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_copy_nested[pytree-compile]	0.4949ms	65.8060μs	15.1962 KOps/s	14.9448 KOps/s	$\color{#35bf28}+1.68\%$
test_compile_copy_nested[pytree-eager]	85.1720μs	50.3372μs	19.8660 KOps/s	19.8717 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_add_one_flat[tensordict-compile]	0.2860ms	0.1463ms	6.8339 KOps/s	6.7786 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_add_one_flat[tensordict-eager]	0.2964ms	0.2085ms	4.7956 KOps/s	4.7746 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_one_flat[tensorclass-compile]	0.1433ms	99.9028μs	10.0097 KOps/s	9.8276 KOps/s	$\color{#35bf28}+1.85\%$
test_compile_add_one_flat[tensorclass-eager]	0.1083ms	52.0849μs	19.1994 KOps/s	18.9203 KOps/s	$\color{#35bf28}+1.47\%$
test_compile_add_one_flat[pytree-compile]	0.1814ms	0.1395ms	7.1696 KOps/s	7.0984 KOps/s	$\color{#35bf28}+1.00\%$
test_compile_add_one_flat[pytree-eager]	0.5398ms	0.4864ms	2.0560 KOps/s	2.0805 KOps/s	$\color{#d91a1a}-1.18\%$
test_compile_add_self_flat[tensordict-eager]	0.3666ms	0.2490ms	4.0165 KOps/s	4.0010 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_add_self_flat[tensordict-compile]	0.1923ms	0.1469ms	6.8079 KOps/s	6.7368 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_add_self_flat[tensorclass-eager]	0.1499ms	64.8459μs	15.4212 KOps/s	15.8700 KOps/s	$\color{#d91a1a}-2.83\%$
test_compile_add_self_flat[tensorclass-compile]	0.1525ms	0.1025ms	9.7530 KOps/s	9.8028 KOps/s	$\color{#d91a1a}-0.51\%$
test_compile_add_self_flat[pytree-eager]	0.4837ms	0.4131ms	2.4210 KOps/s	2.4518 KOps/s	$\color{#d91a1a}-1.26\%$
test_compile_add_self_flat[pytree-compile]	0.1892ms	0.1383ms	7.2324 KOps/s	7.1336 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_copy_flat[tensordict-compile]	0.1537ms	19.2213μs	52.0256 KOps/s	54.4503 KOps/s	$\color{#d91a1a}-4.45\%$
test_compile_copy_flat[tensordict-eager]	61.2610μs	29.0510μs	34.4222 KOps/s	34.8405 KOps/s	$\color{#d91a1a}-1.20\%$
test_compile_copy_flat[pytree-compile]	0.1103ms	70.7118μs	14.1419 KOps/s	13.9887 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_copy_flat[pytree-eager]	81.3310μs	52.1425μs	19.1782 KOps/s	19.0644 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_assign_and_add[tensordict-compile]	1.6588ms	0.4002ms	2.4987 KOps/s	2.1732 KOps/s	$\textbf{\color{#35bf28}+14.97\%}$
test_compile_assign_and_add[tensordict-eager]	2.7491ms	2.5498ms	392.1922 Ops/s	380.3744 Ops/s	$\color{#35bf28}+3.11\%$
test_compile_assign_and_add[pytree-compile]	1.6150ms	0.4389ms	2.2784 KOps/s	2.1603 KOps/s	$\textbf{\color{#35bf28}+5.47\%}$
test_compile_assign_and_add[pytree-eager]	2.7100ms	2.6601ms	375.9307 Ops/s	380.7869 Ops/s	$\color{#d91a1a}-1.28\%$
test_compile_indexing[tensor-tensordict-compile]	0.1563ms	0.1130ms	8.8471 KOps/s	8.4463 KOps/s	$\color{#35bf28}+4.75\%$
test_compile_indexing[tensor-tensordict-eager]	0.5763ms	77.1586μs	12.9603 KOps/s	12.2974 KOps/s	$\textbf{\color{#35bf28}+5.39\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.1460ms	0.1053ms	9.4944 KOps/s	9.1382 KOps/s	$\color{#35bf28}+3.90\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1864ms	66.2858μs	15.0862 KOps/s	14.5264 KOps/s	$\color{#35bf28}+3.85\%$
test_compile_indexing[tensor-pytree-compile]	0.1764ms	0.1063ms	9.4041 KOps/s	9.4329 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_indexing[tensor-pytree-eager]	0.1587ms	65.7386μs	15.2118 KOps/s	14.8844 KOps/s	$\color{#35bf28}+2.20\%$
test_compile_indexing[slice-tensordict-compile]	0.1518ms	0.1025ms	9.7538 KOps/s	9.7191 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_indexing[slice-tensordict-eager]	0.1605ms	17.2156μs	58.0867 KOps/s	48.0237 KOps/s	$\textbf{\color{#35bf28}+20.95\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1351ms	97.2105μs	10.2870 KOps/s	9.7722 KOps/s	$\textbf{\color{#35bf28}+5.27\%}$
test_compile_indexing[slice-tensorclass-eager]	45.8510μs	16.0230μs	62.4102 KOps/s	62.7001 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_indexing[slice-pytree-compile]	0.1518ms	98.5548μs	10.1466 KOps/s	10.1548 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_indexing[slice-pytree-eager]	44.7710μs	16.0040μs	62.4844 KOps/s	62.0126 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_indexing[int-tensordict-compile]	0.1473ms	0.1032ms	9.6867 KOps/s	9.2708 KOps/s	$\color{#35bf28}+4.49\%$
test_compile_indexing[int-tensordict-eager]	0.6290ms	17.0181μs	58.7608 KOps/s	56.2915 KOps/s	$\color{#35bf28}+4.39\%$
test_compile_indexing[int-tensorclass-compile]	0.1717ms	98.3171μs	10.1712 KOps/s	9.8650 KOps/s	$\color{#35bf28}+3.10\%$
test_compile_indexing[int-tensorclass-eager]	0.1675ms	16.0330μs	62.3715 KOps/s	62.5700 KOps/s	$\color{#d91a1a}-0.32\%$
test_compile_indexing[int-pytree-compile]	0.1383ms	97.4575μs	10.2609 KOps/s	9.7069 KOps/s	$\textbf{\color{#35bf28}+5.71\%}$
test_compile_indexing[int-pytree-eager]	58.0700μs	16.1298μs	61.9972 KOps/s	62.2814 KOps/s	$\color{#d91a1a}-0.46\%$
test_mod_add[eager]	75.1010μs	38.4083μs	26.0360 KOps/s	26.8146 KOps/s	$\color{#d91a1a}-2.90\%$
test_mod_add[compile]	0.1541ms	85.5020μs	11.6956 KOps/s	11.7950 KOps/s	$\color{#d91a1a}-0.84\%$
test_mod_add[compile-overhead]	0.3485ms	0.1795ms	5.5696 KOps/s	5.2586 KOps/s	$\textbf{\color{#35bf28}+5.91\%}$
test_mod_wrap[eager]	0.3297ms	0.2477ms	4.0366 KOps/s	3.8940 KOps/s	$\color{#35bf28}+3.66\%$
test_mod_wrap[compile]	0.4204ms	0.3008ms	3.3241 KOps/s	3.4353 KOps/s	$\color{#d91a1a}-3.24\%$
test_mod_wrap[compile-overhead]	7.1952ms	3.7621ms	265.8065 Ops/s	265.9550 Ops/s	$\color{#d91a1a}-0.06\%$
test_mod_wrap_and_backward[eager]	1.4688ms	1.3293ms	752.2875 Ops/s	700.2714 Ops/s	$\textbf{\color{#35bf28}+7.43\%}$
test_mod_wrap_and_backward[compile]	1.4351ms	1.2646ms	790.7690 Ops/s	723.0840 Ops/s	$\textbf{\color{#35bf28}+9.36\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3982ms	0.9203ms	1.0866 KOps/s	938.0537 Ops/s	$\textbf{\color{#35bf28}+15.83\%}$
test_seq_add[eager]	0.1953ms	0.1127ms	8.8739 KOps/s	8.9825 KOps/s	$\color{#d91a1a}-1.21\%$
test_seq_add[compile]	0.1595ms	89.5966μs	11.1611 KOps/s	11.2696 KOps/s	$\color{#d91a1a}-0.96\%$
test_seq_add[compile-overhead]	0.1847ms	0.1310ms	7.6343 KOps/s	7.2947 KOps/s	$\color{#35bf28}+4.66\%$
test_seq_wrap[eager]	0.5013ms	0.4160ms	2.4038 KOps/s	2.3724 KOps/s	$\color{#35bf28}+1.32\%$
test_seq_wrap[compile]	0.3732ms	0.3036ms	3.2943 KOps/s	3.0924 KOps/s	$\textbf{\color{#35bf28}+6.53\%}$
test_seq_wrap[compile-overhead]	0.2788ms	0.2247ms	4.4508 KOps/s	4.1839 KOps/s	$\textbf{\color{#35bf28}+6.38\%}$
test_func_call_runtime[False-eager]	0.8050ms	0.7256ms	1.3781 KOps/s	1.2644 KOps/s	$\textbf{\color{#35bf28}+8.99\%}$
test_func_call_runtime[False-compile]	0.8578ms	0.7512ms	1.3312 KOps/s	1.2580 KOps/s	$\textbf{\color{#35bf28}+5.81\%}$
test_func_call_runtime[False-compile-overhead]	0.4319ms	0.3638ms	2.7489 KOps/s	2.7080 KOps/s	$\color{#35bf28}+1.51\%$
test_func_call_runtime[True-eager]	0.9655ms	0.8922ms	1.1208 KOps/s	1.1101 KOps/s	$\color{#35bf28}+0.97\%$
test_func_call_runtime[True-compile]	0.8617ms	0.7789ms	1.2839 KOps/s	1.2886 KOps/s	$\color{#d91a1a}-0.37\%$
test_func_call_runtime[True-compile-overhead]	0.4448ms	0.3860ms	2.5904 KOps/s	2.5606 KOps/s	$\color{#35bf28}+1.17\%$
test_func_call_cm_runtime[False-eager]	0.7980ms	0.7234ms	1.3823 KOps/s	1.3842 KOps/s	$\color{#d91a1a}-0.14\%$
test_func_call_cm_runtime[False-compile]	0.8496ms	0.7575ms	1.3201 KOps/s	1.2490 KOps/s	$\textbf{\color{#35bf28}+5.69\%}$
test_func_call_cm_runtime[False-compile-overhead]	0.4240ms	0.3674ms	2.7216 KOps/s	2.7072 KOps/s	$\color{#35bf28}+0.53\%$
test_func_call_cm_runtime[True-eager]	1.1178ms	0.9946ms	1.0054 KOps/s	1.0005 KOps/s	$\color{#35bf28}+0.49\%$
test_func_call_cm_runtime[True-compile]	0.8989ms	0.8098ms	1.2348 KOps/s	1.2398 KOps/s	$\color{#d91a1a}-0.40\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4628ms	0.4098ms	2.4402 KOps/s	2.3913 KOps/s	$\color{#35bf28}+2.05\%$
test_vmap_func_call_cm_runtime[eager]	2.4819ms	2.0039ms	499.0225 Ops/s	488.8809 Ops/s	$\color{#35bf28}+2.07\%$
test_vmap_func_call_cm_runtime[compile]	0.9719ms	0.8124ms	1.2309 KOps/s	1.2047 KOps/s	$\color{#35bf28}+2.18\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5065ms	0.4150ms	2.4095 KOps/s	2.3786 KOps/s	$\color{#35bf28}+1.30\%$
test_distributed	2.5903ms	0.1703ms	5.8727 KOps/s	8.5017 KOps/s	$\textbf{\color{#d91a1a}-30.92\%}$
test_tdmodule	56.9710μs	18.9409μs	52.7957 KOps/s	52.8972 KOps/s	$\color{#d91a1a}-0.19\%$
test_tdmodule_dispatch	51.7610μs	32.9704μs	30.3302 KOps/s	31.0649 KOps/s	$\color{#d91a1a}-2.37\%$
test_tdseq	56.7110μs	18.9658μs	52.7266 KOps/s	54.6999 KOps/s	$\color{#d91a1a}-3.61\%$
test_tdseq_dispatch	67.2510μs	35.5444μs	28.1338 KOps/s	28.5129 KOps/s	$\color{#d91a1a}-1.33\%$
test_instantiation_functorch	1.6321ms	1.5259ms	655.3434 Ops/s	641.8670 Ops/s	$\color{#35bf28}+2.10\%$
test_exec_functorch	0.2162ms	0.1410ms	7.0930 KOps/s	6.9898 KOps/s	$\color{#35bf28}+1.48\%$
test_exec_functional_call	0.1878ms	0.1327ms	7.5352 KOps/s	7.4283 KOps/s	$\color{#35bf28}+1.44\%$
test_exec_td_decorator	0.3648ms	0.1810ms	5.5248 KOps/s	5.4727 KOps/s	$\color{#35bf28}+0.95\%$
test_vmap_mlp_speed_decorator[True-True]	0.8074ms	0.6692ms	1.4944 KOps/s	1.4839 KOps/s	$\color{#35bf28}+0.71\%$
test_vmap_mlp_speed_decorator[True-False]	0.8157ms	0.6677ms	1.4976 KOps/s	1.4905 KOps/s	$\color{#35bf28}+0.48\%$
test_vmap_mlp_speed_decorator[False-True]	0.6981ms	0.5755ms	1.7377 KOps/s	1.7081 KOps/s	$\color{#35bf28}+1.73\%$
test_vmap_mlp_speed_decorator[False-False]	0.6943ms	0.5782ms	1.7294 KOps/s	1.7196 KOps/s	$\color{#35bf28}+0.57\%$
test_vmap_transformer_speed_decorator[True-True]	18.8116ms	18.7417ms	53.3570 Ops/s	52.7266 Ops/s	$\color{#35bf28}+1.20\%$
test_vmap_transformer_speed_decorator[True-False]	18.8399ms	18.7830ms	53.2397 Ops/s	52.7926 Ops/s	$\color{#35bf28}+0.85\%$
test_vmap_transformer_speed_decorator[False-True]	18.8353ms	18.6401ms	53.6478 Ops/s	53.3654 Ops/s	$\color{#35bf28}+0.53\%$
test_vmap_transformer_speed_decorator[False-False]	19.3632ms	18.7389ms	53.3649 Ops/s	53.4428 Ops/s	$\color{#d91a1a}-0.15\%$
test_to_module_speed[True]	1.0824ms	0.9846ms	1.0156 KOps/s	1.0227 KOps/s	$\color{#d91a1a}-0.69\%$
test_to_module_speed[False]	1.4275ms	0.9672ms	1.0339 KOps/s	1.0426 KOps/s	$\color{#d91a1a}-0.84\%$
test_tc_init	72.1020μs	34.9824μs	28.5858 KOps/s	29.2603 KOps/s	$\color{#d91a1a}-2.31\%$
test_tc_init_nested	0.1132ms	71.8818μs	13.9117 KOps/s	14.6602 KOps/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_tc_first_layer_tensor	5.7059μs	0.7378μs	1.3554 MOps/s	1.3505 MOps/s	$\color{#35bf28}+0.36\%$
test_tc_first_layer_nontensor	36.3000μs	2.4875μs	402.0125 KOps/s	409.4965 KOps/s	$\color{#d91a1a}-1.83\%$
test_tc_second_layer_tensor	10.1303μs	1.5389μs	649.8173 KOps/s	661.8417 KOps/s	$\color{#d91a1a}-1.82\%$
test_tc_second_layer_nontensor	31.0800μs	3.2769μs	305.1645 KOps/s	309.6473 KOps/s	$\color{#d91a1a}-1.45\%$
test_unbind	0.2375s	10.2095ms	97.9476 Ops/s	149.8714 Ops/s	$\textbf{\color{#d91a1a}-34.65\%}$
test_full_like	11.2616ms	10.4798ms	95.4219 Ops/s	92.5658 Ops/s	$\color{#35bf28}+3.09\%$
test_zeros_like	5.2558ms	4.5938ms	217.6830 Ops/s	132.9454 Ops/s	$\textbf{\color{#35bf28}+63.74\%}$
test_ones_like	5.9163ms	4.6583ms	214.6721 Ops/s	215.6067 Ops/s	$\color{#d91a1a}-0.43\%$
test_clone	12.2694ms	7.4291ms	134.6053 Ops/s	132.5738 Ops/s	$\color{#35bf28}+1.53\%$
test_squeeze	60.5510μs	9.7215μs	102.8646 KOps/s	106.0973 KOps/s	$\color{#d91a1a}-3.05\%$
test_unsqueeze	0.1204ms	75.6752μs	13.2144 KOps/s	14.1869 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_split	0.4151ms	0.1593ms	6.2783 KOps/s	6.1937 KOps/s	$\color{#35bf28}+1.37\%$
test_permute	0.2524ms	0.1858ms	5.3814 KOps/s	5.7622 KOps/s	$\textbf{\color{#d91a1a}-6.61\%}$
test_stack	54.2975ms	53.0058ms	18.8658 Ops/s	18.1714 Ops/s	$\color{#35bf28}+3.82\%$
test_cat	53.4074ms	52.6048ms	19.0097 Ops/s	18.3033 Ops/s	$\color{#35bf28}+3.86\%$