[BugFix] Consolidate lazy stacks of non-tensors [EDIT]

[vmoens]

Stack from ghstack (oldest at bottom):

• [BugFix] Pass type directly during reduction #1225
• -> [BugFix] Consolidate lazy stacks of non-tensors [EDIT] #1224

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	82.4630μs	21.0865μs	47.4236 KOps/s	47.2165 KOps/s	$\color{#35bf28}+0.44\%$
test_plain_set_stack_nested	45.7460μs	21.2103μs	47.1470 KOps/s	46.5496 KOps/s	$\color{#35bf28}+1.28\%$
test_plain_set_nested_inplace	60.5530μs	22.7574μs	43.9418 KOps/s	42.7514 KOps/s	$\color{#35bf28}+2.78\%$
test_plain_set_stack_nested_inplace	90.4680μs	22.7549μs	43.9467 KOps/s	42.5656 KOps/s	$\color{#35bf28}+3.24\%$
test_items	63.6990μs	4.2981μs	232.6621 KOps/s	239.0922 KOps/s	$\color{#d91a1a}-2.69\%$
test_items_nested	0.6487ms	0.4181ms	2.3917 KOps/s	2.4323 KOps/s	$\color{#d91a1a}-1.67\%$
test_items_nested_locked	0.5034ms	0.3982ms	2.5115 KOps/s	2.4411 KOps/s	$\color{#35bf28}+2.89\%$
test_items_nested_leaf	0.1316ms	77.7667μs	12.8590 KOps/s	12.9141 KOps/s	$\color{#d91a1a}-0.43\%$
test_items_stack_nested	0.8318ms	0.4017ms	2.4893 KOps/s	2.4025 KOps/s	$\color{#35bf28}+3.61\%$
test_items_stack_nested_leaf	0.1793ms	77.2070μs	12.9522 KOps/s	12.6480 KOps/s	$\color{#35bf28}+2.40\%$
test_items_stack_nested_locked	0.5886ms	0.4012ms	2.4925 KOps/s	2.4185 KOps/s	$\color{#35bf28}+3.06\%$
test_keys	23.4340μs	3.6231μs	276.0073 KOps/s	281.7361 KOps/s	$\color{#d91a1a}-2.03\%$
test_keys_nested	0.2607ms	0.1642ms	6.0885 KOps/s	6.0467 KOps/s	$\color{#35bf28}+0.69\%$
test_keys_nested_locked	0.7230ms	0.1701ms	5.8799 KOps/s	5.8635 KOps/s	$\color{#35bf28}+0.28\%$
test_keys_nested_leaf	0.2363ms	0.1428ms	7.0006 KOps/s	6.9104 KOps/s	$\color{#35bf28}+1.30\%$
test_keys_stack_nested	0.2537ms	0.1630ms	6.1367 KOps/s	6.0809 KOps/s	$\color{#35bf28}+0.92\%$
test_keys_stack_nested_leaf	0.2248ms	0.1431ms	6.9857 KOps/s	6.9415 KOps/s	$\color{#35bf28}+0.64\%$
test_keys_stack_nested_locked	0.2683ms	0.1681ms	5.9486 KOps/s	5.9060 KOps/s	$\color{#35bf28}+0.72\%$
test_values	8.6340μs	1.0539μs	948.8470 KOps/s	931.9904 KOps/s	$\color{#35bf28}+1.81\%$
test_values_nested	0.1095ms	61.4310μs	16.2784 KOps/s	16.0460 KOps/s	$\color{#35bf28}+1.45\%$
test_values_nested_locked	0.1268ms	61.6657μs	16.2165 KOps/s	16.1329 KOps/s	$\color{#35bf28}+0.52\%$
test_values_nested_leaf	0.1453ms	70.9685μs	14.0908 KOps/s	13.9352 KOps/s	$\color{#35bf28}+1.12\%$
test_values_stack_nested	0.1130ms	62.0003μs	16.1290 KOps/s	15.4073 KOps/s	$\color{#35bf28}+4.68\%$
test_values_stack_nested_leaf	0.1332ms	70.9605μs	14.0924 KOps/s	13.8987 KOps/s	$\color{#35bf28}+1.39\%$
test_values_stack_nested_locked	0.1941ms	62.7352μs	15.9400 KOps/s	15.8732 KOps/s	$\color{#35bf28}+0.42\%$
test_membership	83.1650μs	0.8880μs	1.1261 MOps/s	1.0928 MOps/s	$\color{#35bf28}+3.05\%$
test_membership_nested	19.0050μs	2.8596μs	349.7012 KOps/s	337.3472 KOps/s	$\color{#35bf28}+3.66\%$
test_membership_nested_leaf	24.0750μs	2.8800μs	347.2175 KOps/s	329.7488 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_membership_stacked_nested	22.7920μs	2.8786μs	347.3971 KOps/s	332.5313 KOps/s	$\color{#35bf28}+4.47\%$
test_membership_stacked_nested_leaf	45.8350μs	2.8799μs	347.2330 KOps/s	339.8966 KOps/s	$\color{#35bf28}+2.16\%$
test_membership_nested_last	22.9830μs	4.2681μs	234.2990 KOps/s	225.1568 KOps/s	$\color{#35bf28}+4.06\%$
test_membership_nested_leaf_last	50.6350μs	4.2918μs	233.0033 KOps/s	223.5508 KOps/s	$\color{#35bf28}+4.23\%$
test_membership_stacked_nested_last	25.2270μs	4.2539μs	235.0791 KOps/s	176.9916 KOps/s	$\textbf{\color{#35bf28}+32.82\%}$
test_membership_stacked_nested_leaf_last	47.2080μs	4.2997μs	232.5733 KOps/s	174.1117 KOps/s	$\textbf{\color{#35bf28}+33.58\%}$
test_nested_getleaf	77.1840μs	10.5409μs	94.8684 KOps/s	92.1813 KOps/s	$\color{#35bf28}+2.92\%$
test_nested_get	29.5750μs	10.0552μs	99.4508 KOps/s	97.6473 KOps/s	$\color{#35bf28}+1.85\%$
test_stacked_getleaf	58.2380μs	10.3622μs	96.5048 KOps/s	92.8452 KOps/s	$\color{#35bf28}+3.94\%$
test_stacked_get	53.5000μs	9.9363μs	100.6410 KOps/s	98.3182 KOps/s	$\color{#35bf28}+2.36\%$
test_nested_getitemleaf	33.5030μs	11.0905μs	90.1671 KOps/s	87.3221 KOps/s	$\color{#35bf28}+3.26\%$
test_nested_getitem	49.3910μs	10.5163μs	95.0907 KOps/s	91.0306 KOps/s	$\color{#35bf28}+4.46\%$
test_stacked_getitemleaf	46.5170μs	11.0967μs	90.1166 KOps/s	88.2706 KOps/s	$\color{#35bf28}+2.09\%$
test_stacked_getitem	47.1780μs	10.4019μs	96.1365 KOps/s	93.9209 KOps/s	$\color{#35bf28}+2.36\%$
test_lock_nested	7.3809ms	0.4163ms	2.4020 KOps/s	2.4156 KOps/s	$\color{#d91a1a}-0.57\%$
test_lock_stack_nested	0.5416ms	0.4229ms	2.3644 KOps/s	2.3587 KOps/s	$\color{#35bf28}+0.24\%$
test_unlock_nested	0.7279ms	0.3395ms	2.9456 KOps/s	2.9867 KOps/s	$\color{#d91a1a}-1.38\%$
test_unlock_stack_nested	0.5199ms	0.3421ms	2.9235 KOps/s	2.9190 KOps/s	$\color{#35bf28}+0.15\%$
test_flatten_speed	0.2089ms	0.1014ms	9.8663 KOps/s	9.8595 KOps/s	$\color{#35bf28}+0.07\%$
test_unflatten_speed	0.9781ms	0.5211ms	1.9189 KOps/s	1.8846 KOps/s	$\color{#35bf28}+1.82\%$
test_common_ops	1.0447ms	0.8335ms	1.1998 KOps/s	1.2173 KOps/s	$\color{#d91a1a}-1.43\%$
test_creation	32.6610μs	2.6163μs	382.2257 KOps/s	394.5776 KOps/s	$\color{#d91a1a}-3.13\%$
test_creation_empty	36.7990μs	12.6643μs	78.9622 KOps/s	77.5212 KOps/s	$\color{#35bf28}+1.86\%$
test_creation_nested_1	78.7770μs	15.3764μs	65.0349 KOps/s	63.2115 KOps/s	$\color{#35bf28}+2.88\%$
test_creation_nested_2	74.3690μs	20.4473μs	48.9063 KOps/s	48.8544 KOps/s	$\color{#35bf28}+0.11\%$
test_clone	67.3480μs	13.3371μs	74.9785 KOps/s	72.8953 KOps/s	$\color{#35bf28}+2.86\%$
test_getitem[int]	0.9629ms	12.5038μs	79.9759 KOps/s	77.0490 KOps/s	$\color{#35bf28}+3.80\%$
test_getitem[slice_int]	0.1616ms	23.7087μs	42.1785 KOps/s	41.0078 KOps/s	$\color{#35bf28}+2.85\%$
test_getitem[range]	0.1698ms	51.0256μs	19.5980 KOps/s	19.4746 KOps/s	$\color{#35bf28}+0.63\%$
test_getitem[tuple]	0.1651ms	20.0937μs	49.7669 KOps/s	48.7459 KOps/s	$\color{#35bf28}+2.09\%$
test_getitem[list]	0.1663ms	44.9448μs	22.2495 KOps/s	21.3282 KOps/s	$\color{#35bf28}+4.32\%$
test_setitem_dim[int]	47.0580μs	25.6688μs	38.9578 KOps/s	38.6969 KOps/s	$\color{#35bf28}+0.67\%$
test_setitem_dim[slice_int]	90.7990μs	50.9202μs	19.6386 KOps/s	19.1944 KOps/s	$\color{#35bf28}+2.31\%$
test_setitem_dim[range]	0.1689ms	77.4101μs	12.9182 KOps/s	13.0314 KOps/s	$\color{#d91a1a}-0.87\%$
test_setitem_dim[tuple]	84.8480μs	40.7397μs	24.5461 KOps/s	24.0542 KOps/s	$\color{#35bf28}+2.04\%$
test_setitem	80.0600μs	20.8298μs	48.0081 KOps/s	46.3571 KOps/s	$\color{#35bf28}+3.56\%$
test_set	75.7010μs	20.2127μs	49.4739 KOps/s	47.6689 KOps/s	$\color{#35bf28}+3.79\%$
test_set_shared	3.5387ms	0.1829ms	5.4680 KOps/s	5.4974 KOps/s	$\color{#d91a1a}-0.53\%$
test_update	0.1663ms	23.5092μs	42.5365 KOps/s	40.9509 KOps/s	$\color{#35bf28}+3.87\%$
test_update_nested	0.1144ms	34.0774μs	29.3449 KOps/s	28.4413 KOps/s	$\color{#35bf28}+3.18\%$
test_update__nested	0.4722ms	34.0385μs	29.3785 KOps/s	29.4282 KOps/s	$\color{#d91a1a}-0.17\%$
test_set_nested	75.6410μs	22.4759μs	44.4922 KOps/s	43.3813 KOps/s	$\color{#35bf28}+2.56\%$
test_set_nested_new	68.1670μs	27.2126μs	36.7477 KOps/s	36.3067 KOps/s	$\color{#35bf28}+1.21\%$
test_select	0.1230ms	43.6241μs	22.9231 KOps/s	22.1864 KOps/s	$\color{#35bf28}+3.32\%$
test_select_nested	0.1165ms	63.3639μs	15.7819 KOps/s	15.8221 KOps/s	$\color{#d91a1a}-0.25\%$
test_exclude_nested	0.1516ms	81.1762μs	12.3189 KOps/s	12.3289 KOps/s	$\color{#d91a1a}-0.08\%$
test_empty[True]	0.6969ms	0.4048ms	2.4706 KOps/s	2.4313 KOps/s	$\color{#35bf28}+1.62\%$
test_empty[False]	11.6040μs	1.3546μs	738.2208 KOps/s	736.8984 KOps/s	$\color{#35bf28}+0.18\%$
test_unbind_speed	0.5860ms	0.2685ms	3.7247 KOps/s	3.6497 KOps/s	$\color{#35bf28}+2.05\%$
test_unbind_speed_stack0	0.4651ms	0.2667ms	3.7497 KOps/s	3.6921 KOps/s	$\color{#35bf28}+1.56\%$
test_unbind_speed_stack1	0.7457ms	0.6617ms	1.5114 KOps/s	1.2302 KOps/s	$\textbf{\color{#35bf28}+22.86\%}$
test_split	0.1043s	1.8935ms	528.1172 Ops/s	569.7388 Ops/s	$\textbf{\color{#d91a1a}-7.31\%}$
test_chunk	2.5084ms	1.5705ms	636.7429 Ops/s	630.8728 Ops/s	$\color{#35bf28}+0.93\%$
test_consolidate_njt[False-None]	0.1082s	9.0311ms	110.7281 Ops/s	108.2703 Ops/s	$\color{#35bf28}+2.27\%$
test_creation[device0]	4.3265ms	93.5931μs	10.6846 KOps/s	10.8415 KOps/s	$\color{#d91a1a}-1.45\%$
test_creation_from_tensor	0.2459ms	95.0217μs	10.5239 KOps/s	10.4794 KOps/s	$\color{#35bf28}+0.42\%$
test_add_one[memmap_tensor0]	93.3540μs	5.0357μs	198.5817 KOps/s	185.0536 KOps/s	$\textbf{\color{#35bf28}+7.31\%}$
test_contiguous[memmap_tensor0]	10.6600μs	0.5070μs	1.9722 MOps/s	1.9067 MOps/s	$\color{#35bf28}+3.43\%$
test_stack[memmap_tensor0]	23.4030μs	3.3981μs	294.2843 KOps/s	275.7159 KOps/s	$\textbf{\color{#35bf28}+6.73\%}$
test_memmaptd_index	1.1629ms	0.2282ms	4.3814 KOps/s	4.2088 KOps/s	$\color{#35bf28}+4.10\%$
test_memmaptd_index_astensor	0.6253ms	0.3184ms	3.1411 KOps/s	3.0303 KOps/s	$\color{#35bf28}+3.66\%$
test_memmaptd_index_op	1.1309ms	0.6056ms	1.6512 KOps/s	1.5857 KOps/s	$\color{#35bf28}+4.13\%$
test_serialize_model	0.1214s	0.1143s	8.7499 Ops/s	8.5528 Ops/s	$\color{#35bf28}+2.30\%$
test_serialize_model_pickle	0.5032s	0.4054s	2.4664 Ops/s	2.4931 Ops/s	$\color{#d91a1a}-1.07\%$
test_serialize_weights	0.1283s	0.1139s	8.7810 Ops/s	8.6161 Ops/s	$\color{#35bf28}+1.91\%$
test_serialize_weights_returnearly	0.1782s	0.1619s	6.1782 Ops/s	6.4577 Ops/s	$\color{#d91a1a}-4.33\%$
test_serialize_weights_pickle	0.4992s	0.4230s	2.3638 Ops/s	2.2941 Ops/s	$\color{#35bf28}+3.04\%$
test_serialize_weights_filesystem	0.1513s	0.1426s	7.0131 Ops/s	7.0826 Ops/s	$\color{#d91a1a}-0.98\%$
test_serialize_model_filesystem	0.1544s	0.1442s	6.9346 Ops/s	6.4716 Ops/s	$\textbf{\color{#35bf28}+7.16\%}$
test_reshape_pytree	65.9630μs	26.0539μs	38.3819 KOps/s	37.9071 KOps/s	$\color{#35bf28}+1.25\%$
test_reshape_td	74.0080μs	32.6844μs	30.5957 KOps/s	30.0414 KOps/s	$\color{#35bf28}+1.84\%$
test_view_pytree	58.6900μs	25.9476μs	38.5392 KOps/s	38.2133 KOps/s	$\color{#35bf28}+0.85\%$
test_view_td	82.3540μs	40.0559μs	24.9651 KOps/s	24.7749 KOps/s	$\color{#35bf28}+0.77\%$
test_unbind_pytree	67.4560μs	29.0648μs	34.4059 KOps/s	33.6720 KOps/s	$\color{#35bf28}+2.18\%$
test_unbind_td	0.3800ms	39.6129μs	25.2443 KOps/s	24.6740 KOps/s	$\color{#35bf28}+2.31\%$
test_split_pytree	67.8960μs	28.7404μs	34.7943 KOps/s	34.0421 KOps/s	$\color{#35bf28}+2.21\%$
test_split_td	0.5025ms	45.3800μs	22.0361 KOps/s	21.9141 KOps/s	$\color{#35bf28}+0.56\%$
test_add_pytree	96.5800μs	35.8438μs	27.8989 KOps/s	26.8287 KOps/s	$\color{#35bf28}+3.99\%$
test_add_td	0.1593ms	58.2898μs	17.1557 KOps/s	16.2786 KOps/s	$\textbf{\color{#35bf28}+5.39\%}$
test_compile_add_one_nested[tensordict-compile]	0.1305ms	66.2430μs	15.0959 KOps/s	14.9286 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_add_one_nested[tensordict-eager]	0.5280ms	0.1710ms	5.8475 KOps/s	5.7295 KOps/s	$\color{#35bf28}+2.06\%$
test_compile_add_one_nested[pytree-compile]	0.1045ms	45.6390μs	21.9111 KOps/s	21.6286 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_add_one_nested[pytree-eager]	0.2190ms	0.1184ms	8.4429 KOps/s	8.0804 KOps/s	$\color{#35bf28}+4.49\%$
test_compile_copy_nested[tensordict-compile]	65.2310μs	28.3307μs	35.2974 KOps/s	35.9903 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_copy_nested[tensordict-eager]	0.1191ms	58.6565μs	17.0484 KOps/s	16.9660 KOps/s	$\color{#35bf28}+0.49\%$
test_compile_copy_nested[pytree-compile]	0.1512ms	79.0822μs	12.6451 KOps/s	12.4669 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_copy_nested[pytree-eager]	0.1514ms	66.8387μs	14.9614 KOps/s	14.8288 KOps/s	$\color{#35bf28}+0.89\%$
test_compile_add_one_flat[tensordict-compile]	0.1934ms	0.1075ms	9.3063 KOps/s	9.3162 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_add_one_flat[tensordict-eager]	0.4327ms	0.2171ms	4.6060 KOps/s	4.5933 KOps/s	$\color{#35bf28}+0.28\%$
test_compile_add_one_flat[tensorclass-compile]	0.1828ms	47.5467μs	21.0320 KOps/s	21.2629 KOps/s	$\color{#d91a1a}-1.09\%$
test_compile_add_one_flat[tensorclass-eager]	0.1556ms	68.3639μs	14.6276 KOps/s	14.5390 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_add_one_flat[pytree-compile]	0.1766ms	0.1020ms	9.8062 KOps/s	9.9329 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_add_one_flat[pytree-eager]	0.3399ms	0.2042ms	4.8968 KOps/s	4.8189 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_add_self_flat[tensordict-eager]	0.4098ms	0.2331ms	4.2901 KOps/s	4.2467 KOps/s	$\color{#35bf28}+1.02\%$
test_compile_add_self_flat[tensordict-compile]	0.2370ms	0.1118ms	8.9432 KOps/s	9.1892 KOps/s	$\color{#d91a1a}-2.68\%$
test_compile_add_self_flat[tensorclass-eager]	0.3611ms	63.4030μs	15.7721 KOps/s	15.8970 KOps/s	$\color{#d91a1a}-0.79\%$
test_compile_add_self_flat[tensorclass-compile]	0.3409ms	49.0053μs	20.4060 KOps/s	19.7038 KOps/s	$\color{#35bf28}+3.56\%$
test_compile_add_self_flat[pytree-eager]	0.3684ms	0.1642ms	6.0910 KOps/s	6.2098 KOps/s	$\color{#d91a1a}-1.91\%$
test_compile_add_self_flat[pytree-compile]	0.1752ms	0.1008ms	9.9214 KOps/s	9.8986 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_copy_flat[tensordict-compile]	60.9230μs	20.8340μs	47.9985 KOps/s	47.3309 KOps/s	$\color{#35bf28}+1.41\%$
test_compile_copy_flat[tensordict-eager]	0.1494ms	68.5625μs	14.5852 KOps/s	15.0306 KOps/s	$\color{#d91a1a}-2.96\%$
test_compile_copy_flat[pytree-compile]	0.2323ms	83.4707μs	11.9803 KOps/s	12.2345 KOps/s	$\color{#d91a1a}-2.08\%$
test_compile_copy_flat[pytree-eager]	0.1160ms	68.0617μs	14.6925 KOps/s	14.8208 KOps/s	$\color{#d91a1a}-0.87\%$
test_compile_assign_and_add[tensordict-compile]	0.5718ms	0.2180ms	4.5876 KOps/s	4.7023 KOps/s	$\color{#d91a1a}-2.44\%$
test_compile_assign_and_add[tensordict-eager]	1.6185ms	1.3687ms	730.6345 Ops/s	722.0444 Ops/s	$\color{#35bf28}+1.19\%$
test_compile_assign_and_add[pytree-compile]	0.3123ms	0.2112ms	4.7343 KOps/s	4.7446 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_assign_and_add[pytree-eager]	1.7398ms	0.8329ms	1.2007 KOps/s	1.1942 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_assign_and_add_stack[compile]	0.8237ms	0.4626ms	2.1617 KOps/s	2.1920 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_assign_and_add_stack[eager]	2.9123ms	2.6960ms	370.9247 Ops/s	354.0799 Ops/s	$\color{#35bf28}+4.76\%$
test_compile_indexing[tensor-tensordict-compile]	90.6690μs	38.9448μs	25.6773 KOps/s	25.9003 KOps/s	$\color{#d91a1a}-0.86\%$
test_compile_indexing[tensor-tensordict-eager]	0.5940ms	33.1436μs	30.1717 KOps/s	29.2257 KOps/s	$\color{#35bf28}+3.24\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1352ms	31.6795μs	31.5661 KOps/s	31.3385 KOps/s	$\color{#35bf28}+0.73\%$
test_compile_indexing[tensor-tensorclass-eager]	0.5797ms	23.0931μs	43.3029 KOps/s	40.8484 KOps/s	$\textbf{\color{#35bf28}+6.01\%}$
test_compile_indexing[tensor-pytree-compile]	92.7430μs	31.7059μs	31.5399 KOps/s	30.5264 KOps/s	$\color{#35bf28}+3.32\%$
test_compile_indexing[tensor-pytree-eager]	64.7300μs	22.9197μs	43.6307 KOps/s	41.9134 KOps/s	$\color{#35bf28}+4.10\%$
test_compile_indexing[slice-tensordict-compile]	0.1366ms	52.7919μs	18.9423 KOps/s	19.0129 KOps/s	$\color{#d91a1a}-0.37\%$
test_compile_indexing[slice-tensordict-eager]	0.3838ms	19.2303μs	52.0013 KOps/s	49.3713 KOps/s	$\textbf{\color{#35bf28}+5.33\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1221ms	46.1500μs	21.6685 KOps/s	21.2749 KOps/s	$\color{#35bf28}+1.85\%$
test_compile_indexing[slice-tensorclass-eager]	50.6150μs	18.5929μs	53.7839 KOps/s	53.6053 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_indexing[slice-pytree-compile]	98.9540μs	46.7660μs	21.3830 KOps/s	21.2672 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_indexing[slice-pytree-eager]	65.3720μs	18.7618μs	53.2998 KOps/s	53.6329 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_indexing[int-tensordict-compile]	0.1152ms	54.0010μs	18.5182 KOps/s	18.3499 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_indexing[int-tensordict-eager]	0.9827ms	19.1544μs	52.2072 KOps/s	49.4436 KOps/s	$\textbf{\color{#35bf28}+5.59\%}$
test_compile_indexing[int-tensorclass-compile]	0.1124ms	46.1708μs	21.6587 KOps/s	21.3371 KOps/s	$\color{#35bf28}+1.51\%$
test_compile_indexing[int-tensorclass-eager]	0.1287ms	19.0606μs	52.4643 KOps/s	53.5299 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_indexing[int-pytree-compile]	0.1119ms	46.2624μs	21.6158 KOps/s	21.1581 KOps/s	$\color{#35bf28}+2.16\%$
test_compile_indexing[int-pytree-eager]	73.9480μs	18.5569μs	53.8882 KOps/s	53.5920 KOps/s	$\color{#35bf28}+0.55\%$
test_mod_add[eager]	93.8240μs	35.6360μs	28.0615 KOps/s	27.1542 KOps/s	$\color{#35bf28}+3.34\%$
test_mod_add[compile]	0.1101ms	63.8104μs	15.6714 KOps/s	15.5863 KOps/s	$\color{#35bf28}+0.55\%$
test_mod_add[compile-overhead]	0.2042ms	64.0806μs	15.6053 KOps/s	15.5402 KOps/s	$\color{#35bf28}+0.42\%$
test_mod_wrap[eager]	0.9339ms	0.2254ms	4.4365 KOps/s	4.4596 KOps/s	$\color{#d91a1a}-0.52\%$
test_mod_wrap[compile]	1.9018ms	0.2299ms	4.3492 KOps/s	4.3505 KOps/s	$\color{#d91a1a}-0.03\%$
test_mod_wrap[compile-overhead]	0.4242ms	0.2264ms	4.4177 KOps/s	4.3822 KOps/s	$\color{#35bf28}+0.81\%$
test_mod_wrap_and_backward[eager]	15.9641ms	13.4876ms	74.1421 Ops/s	72.0827 Ops/s	$\color{#35bf28}+2.86\%$
test_mod_wrap_and_backward[compile]	14.5857ms	11.7592ms	85.0397 Ops/s	85.3565 Ops/s	$\color{#d91a1a}-0.37\%$
test_mod_wrap_and_backward[compile-overhead]	16.7307ms	11.9013ms	84.0245 Ops/s	87.4527 Ops/s	$\color{#d91a1a}-3.92\%$
test_seq_add[eager]	0.1880ms	0.1154ms	8.6621 KOps/s	8.2400 KOps/s	$\textbf{\color{#35bf28}+5.12\%}$
test_seq_add[compile]	0.1398ms	75.7865μs	13.1950 KOps/s	12.8365 KOps/s	$\color{#35bf28}+2.79\%$
test_seq_add[compile-overhead]	0.1384ms	74.9989μs	13.3335 KOps/s	13.3191 KOps/s	$\color{#35bf28}+0.11\%$
test_seq_wrap[eager]	0.5786ms	0.4380ms	2.2831 KOps/s	2.1509 KOps/s	$\textbf{\color{#35bf28}+6.14\%}$
test_seq_wrap[compile]	0.4534ms	0.2445ms	4.0896 KOps/s	4.0100 KOps/s	$\color{#35bf28}+1.98\%$
test_seq_wrap[compile-overhead]	0.3412ms	0.2427ms	4.1197 KOps/s	4.0278 KOps/s	$\color{#35bf28}+2.28\%$
test_func_call_runtime[False-eager]	0.7223ms	0.5300ms	1.8866 KOps/s	1.8483 KOps/s	$\color{#35bf28}+2.08\%$
test_func_call_runtime[False-compile]	0.5586ms	0.4438ms	2.2534 KOps/s	2.2153 KOps/s	$\color{#35bf28}+1.72\%$
test_func_call_runtime[False-compile-overhead]	0.5366ms	0.4429ms	2.2578 KOps/s	2.2218 KOps/s	$\color{#35bf28}+1.62\%$
test_func_call_runtime[True-eager]	1.1114ms	0.7417ms	1.3483 KOps/s	1.3263 KOps/s	$\color{#35bf28}+1.66\%$
test_func_call_runtime[True-compile]	0.5717ms	0.4624ms	2.1627 KOps/s	2.1236 KOps/s	$\color{#35bf28}+1.84\%$
test_func_call_runtime[True-compile-overhead]	0.6886ms	0.4639ms	2.1557 KOps/s	2.1266 KOps/s	$\color{#35bf28}+1.37\%$
test_func_call_cm_runtime[False-eager]	0.9641ms	0.5306ms	1.8845 KOps/s	1.8505 KOps/s	$\color{#35bf28}+1.84\%$
test_func_call_cm_runtime[False-compile]	0.6201ms	0.4385ms	2.2805 KOps/s	2.2049 KOps/s	$\color{#35bf28}+3.43\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6220ms	0.4407ms	2.2692 KOps/s	2.2187 KOps/s	$\color{#35bf28}+2.28\%$
test_func_call_cm_runtime[True-eager]	1.0202ms	0.8791ms	1.1375 KOps/s	1.1095 KOps/s	$\color{#35bf28}+2.53\%$
test_func_call_cm_runtime[True-compile]	1.1585ms	0.7833ms	1.2767 KOps/s	1.2448 KOps/s	$\color{#35bf28}+2.57\%$
test_func_call_cm_runtime[True-compile-overhead]	1.3268ms	0.7936ms	1.2601 KOps/s	1.2324 KOps/s	$\color{#35bf28}+2.24\%$
test_vmap_func_call_cm_runtime[eager]	3.9205ms	1.9193ms	521.0315 Ops/s	517.6806 Ops/s	$\color{#35bf28}+0.65\%$
test_vmap_func_call_cm_runtime[compile]	0.8637ms	0.5537ms	1.8061 KOps/s	1.8355 KOps/s	$\color{#d91a1a}-1.60\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.7810ms	0.5387ms	1.8563 KOps/s	1.8335 KOps/s	$\color{#35bf28}+1.24\%$
test_distributed	0.2775ms	0.1279ms	7.8198 KOps/s	7.8115 KOps/s	$\color{#35bf28}+0.11\%$
test_tdmodule	46.8070μs	27.0599μs	36.9551 KOps/s	36.6310 KOps/s	$\color{#35bf28}+0.88\%$
test_tdmodule_dispatch	0.1148ms	51.4228μs	19.4466 KOps/s	19.7018 KOps/s	$\color{#d91a1a}-1.30\%$
test_tdseq	48.8610μs	29.5049μs	33.8927 KOps/s	32.7639 KOps/s	$\color{#35bf28}+3.45\%$
test_tdseq_dispatch	0.1157ms	55.6928μs	17.9556 KOps/s	18.1166 KOps/s	$\color{#d91a1a}-0.89\%$
test_instantiation_functorch	2.0360ms	1.5378ms	650.2841 Ops/s	639.8023 Ops/s	$\color{#35bf28}+1.64\%$
test_exec_functorch	0.3884ms	0.1766ms	5.6611 KOps/s	5.6186 KOps/s	$\color{#35bf28}+0.76\%$
test_exec_functional_call	0.3989ms	0.1713ms	5.8387 KOps/s	5.8353 KOps/s	$\color{#35bf28}+0.06\%$
test_exec_td_decorator	0.5217ms	0.2331ms	4.2900 KOps/s	4.2743 KOps/s	$\color{#35bf28}+0.37\%$
test_vmap_mlp_speed_decorator[True-True]	0.9678ms	0.6575ms	1.5208 KOps/s	1.4939 KOps/s	$\color{#35bf28}+1.80\%$
test_vmap_mlp_speed_decorator[True-False]	0.8806ms	0.6531ms	1.5311 KOps/s	1.5001 KOps/s	$\color{#35bf28}+2.07\%$
test_vmap_mlp_speed_decorator[False-True]	0.8592ms	0.5301ms	1.8866 KOps/s	1.8675 KOps/s	$\color{#35bf28}+1.02\%$
test_vmap_mlp_speed_decorator[False-False]	0.8765ms	0.5334ms	1.8748 KOps/s	1.8564 KOps/s	$\color{#35bf28}+0.99\%$
test_to_module_speed[True]	1.9705ms	1.3308ms	751.4401 Ops/s	753.2216 Ops/s	$\color{#d91a1a}-0.24\%$
test_to_module_speed[False]	2.4274ms	1.3214ms	756.7445 Ops/s	771.0601 Ops/s	$\color{#d91a1a}-1.86\%$
test_tc_init	88.8260μs	49.3753μs	20.2530 KOps/s	20.2017 KOps/s	$\color{#35bf28}+0.25\%$
test_tc_init_nested	0.1824ms	98.6175μs	10.1402 KOps/s	10.1113 KOps/s	$\color{#35bf28}+0.29\%$
test_tc_first_layer_tensor	43.8520μs	1.6314μs	612.9637 KOps/s	639.3121 KOps/s	$\color{#d91a1a}-4.12\%$
test_tc_first_layer_nontensor	30.7970μs	4.9586μs	201.6687 KOps/s	209.5784 KOps/s	$\color{#d91a1a}-3.77\%$
test_tc_second_layer_tensor	22.2920μs	2.9403μs	340.0963 KOps/s	343.6534 KOps/s	$\color{#d91a1a}-1.04\%$
test_tc_second_layer_nontensor	51.9570μs	6.3697μs	156.9938 KOps/s	163.5150 KOps/s	$\color{#d91a1a}-3.99\%$
test_unbind	0.2491s	13.9931ms	71.4639 Ops/s	70.1015 Ops/s	$\color{#35bf28}+1.94\%$
test_full_like	9.2718ms	8.0533ms	124.1725 Ops/s	139.4272 Ops/s	$\textbf{\color{#d91a1a}-10.94\%}$
test_zeros_like	9.2510ms	4.6573ms	214.7161 Ops/s	359.8828 Ops/s	$\textbf{\color{#d91a1a}-40.34\%}$
test_ones_like	5.0855ms	3.3418ms	299.2442 Ops/s	313.3520 Ops/s	$\color{#d91a1a}-4.50\%$
test_clone	13.2095ms	7.4227ms	134.7225 Ops/s	187.3545 Ops/s	$\textbf{\color{#d91a1a}-28.09\%}$
test_squeeze	69.4400μs	12.6291μs	79.1822 KOps/s	79.5056 KOps/s	$\color{#d91a1a}-0.41\%$
test_unsqueeze	0.2755ms	95.0682μs	10.5188 KOps/s	10.3282 KOps/s	$\color{#35bf28}+1.84\%$
test_split	0.4317ms	0.1956ms	5.1122 KOps/s	5.0690 KOps/s	$\color{#35bf28}+0.85\%$
test_permute	0.3364ms	0.2030ms	4.9270 KOps/s	4.9264 KOps/s	$\color{#35bf28}+0.01\%$
test_stack	28.2078ms	25.2077ms	39.6704 Ops/s	40.2664 Ops/s	$\color{#d91a1a}-1.48\%$
test_cat	29.8918ms	25.2455ms	39.6111 Ops/s	40.7719 Ops/s	$\color{#d91a1a}-2.85\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}16$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	40.9700μs	12.8500μs	77.8213 KOps/s	81.3223 KOps/s	$\color{#d91a1a}-4.31\%$
test_plain_set_stack_nested	34.1700μs	12.9815μs	77.0327 KOps/s	80.1578 KOps/s	$\color{#d91a1a}-3.90\%$
test_plain_set_nested_inplace	43.4410μs	13.9801μs	71.5302 KOps/s	73.7923 KOps/s	$\color{#d91a1a}-3.07\%$
test_plain_set_stack_nested_inplace	45.8910μs	13.8753μs	72.0706 KOps/s	74.8436 KOps/s	$\color{#d91a1a}-3.71\%$
test_items	29.9310μs	2.8676μs	348.7220 KOps/s	342.5485 KOps/s	$\color{#35bf28}+1.80\%$
test_items_nested	0.4357ms	0.3702ms	2.7012 KOps/s	2.6730 KOps/s	$\color{#35bf28}+1.06\%$
test_items_nested_locked	0.4173ms	0.3715ms	2.6916 KOps/s	2.6617 KOps/s	$\color{#35bf28}+1.12\%$
test_items_nested_leaf	91.4410μs	60.0957μs	16.6401 KOps/s	16.5699 KOps/s	$\color{#35bf28}+0.42\%$
test_items_stack_nested	0.4035ms	0.3683ms	2.7155 KOps/s	2.6927 KOps/s	$\color{#35bf28}+0.85\%$
test_items_stack_nested_leaf	87.5810μs	60.5718μs	16.5093 KOps/s	16.3929 KOps/s	$\color{#35bf28}+0.71\%$
test_items_stack_nested_locked	0.4306ms	0.3720ms	2.6880 KOps/s	2.6653 KOps/s	$\color{#35bf28}+0.85\%$
test_keys	37.2610μs	3.4274μs	291.7690 KOps/s	291.9284 KOps/s	$\color{#d91a1a}-0.05\%$
test_keys_nested	0.1334ms	87.9063μs	11.3758 KOps/s	11.3327 KOps/s	$\color{#35bf28}+0.38\%$
test_keys_nested_locked	0.7633ms	94.4716μs	10.5852 KOps/s	10.6198 KOps/s	$\color{#d91a1a}-0.33\%$
test_keys_nested_leaf	0.1151ms	79.8926μs	12.5168 KOps/s	12.6384 KOps/s	$\color{#d91a1a}-0.96\%$
test_keys_stack_nested	0.1435ms	88.4761μs	11.3025 KOps/s	11.3579 KOps/s	$\color{#d91a1a}-0.49\%$
test_keys_stack_nested_leaf	0.1132ms	80.3783μs	12.4412 KOps/s	12.5539 KOps/s	$\color{#d91a1a}-0.90\%$
test_keys_stack_nested_locked	0.1325ms	94.9989μs	10.5264 KOps/s	10.5430 KOps/s	$\color{#d91a1a}-0.16\%$
test_values	6.0435μs	0.8584μs	1.1650 MOps/s	1.1740 MOps/s	$\color{#d91a1a}-0.77\%$
test_values_nested	77.9510μs	37.2826μs	26.8222 KOps/s	26.8376 KOps/s	$\color{#d91a1a}-0.06\%$
test_values_nested_locked	72.1010μs	39.6248μs	25.2367 KOps/s	25.6083 KOps/s	$\color{#d91a1a}-1.45\%$
test_values_nested_leaf	78.9510μs	42.4983μs	23.5304 KOps/s	23.7158 KOps/s	$\color{#d91a1a}-0.78\%$
test_values_stack_nested	65.7010μs	37.4495μs	26.7026 KOps/s	26.5897 KOps/s	$\color{#35bf28}+0.42\%$
test_values_stack_nested_leaf	75.4010μs	42.4132μs	23.5776 KOps/s	23.4314 KOps/s	$\color{#35bf28}+0.62\%$
test_values_stack_nested_locked	0.1033ms	39.9061μs	25.0588 KOps/s	25.5233 KOps/s	$\color{#d91a1a}-1.82\%$
test_membership	1.7071μs	0.5016μs	1.9936 MOps/s	1.9741 MOps/s	$\color{#35bf28}+0.99\%$
test_membership_nested	29.0600μs	2.0976μs	476.7342 KOps/s	478.7006 KOps/s	$\color{#d91a1a}-0.41\%$
test_membership_nested_leaf	21.9755μs	2.0089μs	497.7783 KOps/s	491.8793 KOps/s	$\color{#35bf28}+1.20\%$
test_membership_stacked_nested	32.7710μs	2.1207μs	471.5489 KOps/s	475.8986 KOps/s	$\color{#d91a1a}-0.91\%$
test_membership_stacked_nested_leaf	41.8710μs	2.1199μs	471.7275 KOps/s	462.3066 KOps/s	$\color{#35bf28}+2.04\%$
test_membership_nested_last	31.4910μs	3.0673μs	326.0244 KOps/s	322.5975 KOps/s	$\color{#35bf28}+1.06\%$
test_membership_nested_leaf_last	32.1400μs	3.0397μs	328.9748 KOps/s	323.6620 KOps/s	$\color{#35bf28}+1.64\%$
test_membership_stacked_nested_last	30.4610μs	3.1044μs	322.1272 KOps/s	324.5949 KOps/s	$\color{#d91a1a}-0.76\%$
test_membership_stacked_nested_leaf_last	34.6410μs	3.1018μs	322.3926 KOps/s	323.1095 KOps/s	$\color{#d91a1a}-0.22\%$
test_nested_getleaf	55.2410μs	6.1851μs	161.6790 KOps/s	159.1955 KOps/s	$\color{#35bf28}+1.56\%$
test_nested_get	34.0410μs	5.9774μs	167.2966 KOps/s	168.1455 KOps/s	$\color{#d91a1a}-0.50\%$
test_stacked_getleaf	35.5010μs	6.1623μs	162.2769 KOps/s	160.2197 KOps/s	$\color{#35bf28}+1.28\%$
test_stacked_get	33.0100μs	5.8616μs	170.6008 KOps/s	170.4022 KOps/s	$\color{#35bf28}+0.12\%$
test_nested_getitemleaf	30.0710μs	6.4790μs	154.3441 KOps/s	153.7289 KOps/s	$\color{#35bf28}+0.40\%$
test_nested_getitem	31.0900μs	6.1507μs	162.5819 KOps/s	162.9327 KOps/s	$\color{#d91a1a}-0.22\%$
test_stacked_getitemleaf	41.1610μs	6.4345μs	155.4115 KOps/s	155.5882 KOps/s	$\color{#d91a1a}-0.11\%$
test_stacked_getitem	35.1610μs	5.9939μs	166.8363 KOps/s	167.0478 KOps/s	$\color{#d91a1a}-0.13\%$
test_lock_nested	0.4003ms	0.3433ms	2.9130 KOps/s	2.9817 KOps/s	$\color{#d91a1a}-2.30\%$
test_lock_stack_nested	0.3837ms	0.3494ms	2.8622 KOps/s	2.9064 KOps/s	$\color{#d91a1a}-1.52\%$
test_unlock_nested	0.3827ms	0.2898ms	3.4506 KOps/s	3.5542 KOps/s	$\color{#d91a1a}-2.92\%$
test_unlock_stack_nested	0.3366ms	0.2898ms	3.4506 KOps/s	3.5404 KOps/s	$\color{#d91a1a}-2.53\%$
test_flatten_speed	0.1115ms	77.8701μs	12.8419 KOps/s	12.7541 KOps/s	$\color{#35bf28}+0.69\%$
test_unflatten_speed	0.4177ms	0.3261ms	3.0662 KOps/s	3.1084 KOps/s	$\color{#d91a1a}-1.36\%$
test_common_ops	0.7546ms	0.6212ms	1.6099 KOps/s	1.6362 KOps/s	$\color{#d91a1a}-1.61\%$
test_creation	78.4810μs	1.7575μs	569.0022 KOps/s	580.2079 KOps/s	$\color{#d91a1a}-1.93\%$
test_creation_empty	39.7000μs	9.0458μs	110.5483 KOps/s	124.5914 KOps/s	$\textbf{\color{#d91a1a}-11.27\%}$
test_creation_nested_1	42.6100μs	10.5729μs	94.5817 KOps/s	102.6839 KOps/s	$\textbf{\color{#d91a1a}-7.89\%}$
test_creation_nested_2	48.1910μs	13.3807μs	74.7344 KOps/s	80.9147 KOps/s	$\textbf{\color{#d91a1a}-7.64\%}$
test_clone	56.7210μs	11.1498μs	89.6876 KOps/s	92.8230 KOps/s	$\color{#d91a1a}-3.38\%$
test_getitem[int]	1.2048ms	10.9538μs	91.2927 KOps/s	95.6949 KOps/s	$\color{#d91a1a}-4.60\%$
test_getitem[slice_int]	0.1167ms	21.4230μs	46.6787 KOps/s	49.1492 KOps/s	$\textbf{\color{#d91a1a}-5.03\%}$
test_getitem[range]	0.1295ms	39.2133μs	25.5016 KOps/s	26.8209 KOps/s	$\color{#d91a1a}-4.92\%$
test_getitem[tuple]	0.1064ms	18.4766μs	54.1226 KOps/s	56.3739 KOps/s	$\color{#d91a1a}-3.99\%$
test_getitem[list]	0.1675ms	34.4229μs	29.0504 KOps/s	29.7768 KOps/s	$\color{#d91a1a}-2.44\%$
test_setitem_dim[int]	47.6910μs	20.0976μs	49.7571 KOps/s	50.4232 KOps/s	$\color{#d91a1a}-1.32\%$
test_setitem_dim[slice_int]	63.8010μs	39.1260μs	25.5585 KOps/s	25.7394 KOps/s	$\color{#d91a1a}-0.70\%$
test_setitem_dim[range]	0.1059ms	55.9077μs	17.8866 KOps/s	18.8388 KOps/s	$\textbf{\color{#d91a1a}-5.05\%}$
test_setitem_dim[tuple]	60.5710μs	32.6786μs	30.6011 KOps/s	29.6221 KOps/s	$\color{#35bf28}+3.30\%$
test_setitem	67.0110μs	16.0120μs	62.4533 KOps/s	66.8440 KOps/s	$\textbf{\color{#d91a1a}-6.57\%}$
test_set	75.7920μs	15.1916μs	65.8260 KOps/s	67.6725 KOps/s	$\color{#d91a1a}-2.73\%$
test_set_shared	0.5159ms	0.1599ms	6.2535 KOps/s	6.2965 KOps/s	$\color{#d91a1a}-0.68\%$
test_update	0.2313ms	18.6899μs	53.5049 KOps/s	56.7598 KOps/s	$\textbf{\color{#d91a1a}-5.73\%}$
test_update_nested	68.9410μs	24.3475μs	41.0721 KOps/s	42.8787 KOps/s	$\color{#d91a1a}-4.21\%$
test_update__nested	0.5101ms	25.5786μs	39.0952 KOps/s	38.8957 KOps/s	$\color{#35bf28}+0.51\%$
test_set_nested	64.0610μs	16.7630μs	59.6552 KOps/s	62.2903 KOps/s	$\color{#d91a1a}-4.23\%$
test_set_nested_new	76.2120μs	18.6645μs	53.5777 KOps/s	53.4038 KOps/s	$\color{#35bf28}+0.33\%$
test_select	67.1610μs	30.1491μs	33.1684 KOps/s	33.7563 KOps/s	$\color{#d91a1a}-1.74\%$
test_select_nested	74.3110μs	43.2636μs	23.1141 KOps/s	22.5378 KOps/s	$\color{#35bf28}+2.56\%$
test_exclude_nested	94.1510μs	63.1741μs	15.8293 KOps/s	15.6248 KOps/s	$\color{#35bf28}+1.31\%$
test_empty[True]	0.3604ms	0.2945ms	3.3957 KOps/s	3.3541 KOps/s	$\color{#35bf28}+1.24\%$
test_empty[False]	3.7130μs	0.8388μs	1.1922 MOps/s	1.1961 MOps/s	$\color{#d91a1a}-0.32\%$
test_to	88.2910μs	57.1923μs	17.4849 KOps/s	16.4618 KOps/s	$\textbf{\color{#35bf28}+6.21\%}$
test_to_nonblocking	91.3620μs	49.2259μs	20.3145 KOps/s	20.8840 KOps/s	$\color{#d91a1a}-2.73\%$
test_unbind_speed	0.2923ms	0.2480ms	4.0329 KOps/s	4.2837 KOps/s	$\textbf{\color{#d91a1a}-5.85\%}$
test_unbind_speed_stack0	0.3126ms	0.2474ms	4.0414 KOps/s	4.2288 KOps/s	$\color{#d91a1a}-4.43\%$
test_unbind_speed_stack1	93.5453ms	0.7503ms	1.3328 KOps/s	1.3467 KOps/s	$\color{#d91a1a}-1.03\%$
test_split	94.2802ms	1.6044ms	623.2778 Ops/s	627.8731 Ops/s	$\color{#d91a1a}-0.73\%$
test_chunk	96.3463ms	1.6143ms	619.4774 Ops/s	627.1133 Ops/s	$\color{#d91a1a}-1.22\%$
test_consolidate[False-None]	2.8077ms	2.7081ms	369.2684 Ops/s	336.0734 Ops/s	$\textbf{\color{#35bf28}+9.88\%}$
test_consolidate[default-None]	1.7968ms	1.7316ms	577.4926 Ops/s	584.7331 Ops/s	$\color{#d91a1a}-1.24\%$
test_consolidate[reduce-overhead-None]	1.8454ms	1.7534ms	570.3048 Ops/s	572.8247 Ops/s	$\color{#d91a1a}-0.44\%$
test_consolidate_njt[False-None]	6.7370ms	6.5330ms	153.0692 Ops/s	111.2031 Ops/s	$\textbf{\color{#35bf28}+37.65\%}$
test_to[False-False-None]	1.8194ms	1.7532ms	570.3847 Ops/s	564.3212 Ops/s	$\color{#35bf28}+1.07\%$
test_to[True-False-None]	1.4931ms	1.3763ms	726.6059 Ops/s	751.6225 Ops/s	$\color{#d91a1a}-3.33\%$
test_to[within-False-None]	4.4646ms	4.2387ms	235.9187 Ops/s	237.8136 Ops/s	$\color{#d91a1a}-0.80\%$
test_to[True-default-None]	5.5874ms	5.2450ms	190.6581 Ops/s	178.7626 Ops/s	$\textbf{\color{#35bf28}+6.65\%}$
test_to_njt[False-False-None]	7.1205ms	6.9804ms	143.2579 Ops/s	134.1229 Ops/s	$\textbf{\color{#35bf28}+6.81\%}$
test_to_njt[True-False-None]	5.7053ms	5.5317ms	180.7763 Ops/s	173.7753 Ops/s	$\color{#35bf28}+4.03\%$
test_to_njt[within-False-None]	12.4026ms	12.1236ms	82.4838 Ops/s	80.6630 Ops/s	$\color{#35bf28}+2.26\%$
test_creation[device0]	0.4579ms	79.3675μs	12.5996 KOps/s	12.6119 KOps/s	$\color{#d91a1a}-0.10\%$
test_creation_from_tensor	0.6225ms	83.3414μs	11.9988 KOps/s	11.8954 KOps/s	$\color{#35bf28}+0.87\%$
test_add_one[memmap_tensor0]	0.4339ms	7.0271μs	142.3069 KOps/s	148.1619 KOps/s	$\color{#d91a1a}-3.95\%$
test_contiguous[memmap_tensor0]	2.0430μs	0.4183μs	2.3905 MOps/s	2.2562 MOps/s	$\textbf{\color{#35bf28}+5.95\%}$
test_stack[memmap_tensor0]	38.1010μs	4.7791μs	209.2435 KOps/s	229.5432 KOps/s	$\textbf{\color{#d91a1a}-8.84\%}$
test_memmaptd_index	1.8343ms	0.2515ms	3.9766 KOps/s	4.1440 KOps/s	$\color{#d91a1a}-4.04\%$
test_memmaptd_index_astensor	0.4534ms	0.3110ms	3.2153 KOps/s	3.2908 KOps/s	$\color{#d91a1a}-2.29\%$
test_memmaptd_index_op	0.7522ms	0.6109ms	1.6369 KOps/s	1.7363 KOps/s	$\textbf{\color{#d91a1a}-5.72\%}$
test_serialize_model	0.1310s	0.1298s	7.7012 Ops/s	7.6913 Ops/s	$\color{#35bf28}+0.13\%$
test_serialize_model_pickle	1.3820s	1.2222s	0.8182 Ops/s	0.8206 Ops/s	$\color{#d91a1a}-0.29\%$
test_serialize_weights	0.1308s	0.1292s	7.7384 Ops/s	7.7419 Ops/s	$\color{#d91a1a}-0.04\%$
test_serialize_weights_returnearly	0.2951s	52.8193ms	18.9325 Ops/s	14.8062 Ops/s	$\textbf{\color{#35bf28}+27.87\%}$
test_serialize_weights_pickle	1.3779s	1.2203s	0.8195 Ops/s	0.7714 Ops/s	$\textbf{\color{#35bf28}+6.23\%}$
test_reshape_pytree	51.7400μs	22.0371μs	45.3781 KOps/s	45.0607 KOps/s	$\color{#35bf28}+0.70\%$
test_reshape_td	66.1910μs	27.2803μs	36.6564 KOps/s	36.3830 KOps/s	$\color{#35bf28}+0.75\%$
test_view_pytree	56.2200μs	22.1690μs	45.1081 KOps/s	45.4979 KOps/s	$\color{#d91a1a}-0.86\%$
test_view_td	65.5910μs	31.4571μs	31.7893 KOps/s	30.8832 KOps/s	$\color{#35bf28}+2.93\%$
test_unbind_pytree	58.0510μs	28.5857μs	34.9825 KOps/s	35.8853 KOps/s	$\color{#d91a1a}-2.52\%$
test_unbind_td	0.6173ms	37.6149μs	26.5852 KOps/s	27.1819 KOps/s	$\color{#d91a1a}-2.20\%$
test_split_pytree	65.5810μs	29.9188μs	33.4238 KOps/s	33.3854 KOps/s	$\color{#35bf28}+0.11\%$
test_split_td	0.7787ms	39.0661μs	25.5977 KOps/s	25.5585 KOps/s	$\color{#35bf28}+0.15\%$
test_add_pytree	66.4010μs	35.5357μs	28.1407 KOps/s	28.4188 KOps/s	$\color{#d91a1a}-0.98\%$
test_add_td	0.1032ms	53.7308μs	18.6113 KOps/s	20.9821 KOps/s	$\textbf{\color{#d91a1a}-11.30\%}$
test_compile_add_one_nested[tensordict-compile]	0.1744ms	0.1223ms	8.1735 KOps/s	7.9163 KOps/s	$\color{#35bf28}+3.25\%$
test_compile_add_one_nested[tensordict-eager]	0.2263ms	0.1320ms	7.5782 KOps/s	7.3398 KOps/s	$\color{#35bf28}+3.25\%$
test_compile_add_one_nested[pytree-compile]	0.2029ms	96.0188μs	10.4146 KOps/s	10.3462 KOps/s	$\color{#35bf28}+0.66\%$
test_compile_add_one_nested[pytree-eager]	1.1541ms	0.1511ms	6.6191 KOps/s	6.6733 KOps/s	$\color{#d91a1a}-0.81\%$
test_compile_copy_nested[tensordict-compile]	54.8110μs	24.3814μs	41.0149 KOps/s	41.5395 KOps/s	$\color{#d91a1a}-1.26\%$
test_compile_copy_nested[tensordict-eager]	95.8120μs	29.4006μs	34.0129 KOps/s	33.1532 KOps/s	$\color{#35bf28}+2.59\%$
test_compile_copy_nested[pytree-compile]	0.4255ms	63.8800μs	15.6544 KOps/s	15.2975 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_copy_nested[pytree-eager]	82.7710μs	48.6699μs	20.5466 KOps/s	20.2094 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_add_one_flat[tensordict-compile]	0.1965ms	0.1426ms	7.0124 KOps/s	7.0377 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_add_one_flat[tensordict-eager]	0.3140ms	0.2188ms	4.5703 KOps/s	4.5716 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_add_one_flat[tensorclass-compile]	0.2210ms	97.7165μs	10.2337 KOps/s	10.2954 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_add_one_flat[tensorclass-eager]	0.1160ms	55.8169μs	17.9157 KOps/s	17.8406 KOps/s	$\color{#35bf28}+0.42\%$
test_compile_add_one_flat[pytree-compile]	0.1858ms	0.1375ms	7.2706 KOps/s	7.3665 KOps/s	$\color{#d91a1a}-1.30\%$
test_compile_add_one_flat[pytree-eager]	0.5390ms	0.4920ms	2.0324 KOps/s	2.0490 KOps/s	$\color{#d91a1a}-0.81\%$
test_compile_add_self_flat[tensordict-eager]	0.3807ms	0.2639ms	3.7898 KOps/s	3.7699 KOps/s	$\color{#35bf28}+0.53\%$
test_compile_add_self_flat[tensordict-compile]	0.1985ms	0.1474ms	6.7821 KOps/s	7.0611 KOps/s	$\color{#d91a1a}-3.95\%$
test_compile_add_self_flat[tensorclass-eager]	0.1636ms	67.2128μs	14.8781 KOps/s	14.3054 KOps/s	$\color{#35bf28}+4.00\%$
test_compile_add_self_flat[tensorclass-compile]	0.1418ms	0.1003ms	9.9734 KOps/s	10.0912 KOps/s	$\color{#d91a1a}-1.17\%$
test_compile_add_self_flat[pytree-eager]	0.4800ms	0.4182ms	2.3912 KOps/s	2.4414 KOps/s	$\color{#d91a1a}-2.05\%$
test_compile_add_self_flat[pytree-compile]	0.1792ms	0.1381ms	7.2388 KOps/s	7.4232 KOps/s	$\color{#d91a1a}-2.49\%$
test_compile_copy_flat[tensordict-compile]	58.7710μs	18.9517μs	52.7657 KOps/s	46.4657 KOps/s	$\textbf{\color{#35bf28}+13.56\%}$
test_compile_copy_flat[tensordict-eager]	92.8320μs	31.3424μs	31.9056 KOps/s	31.4160 KOps/s	$\color{#35bf28}+1.56\%$
test_compile_copy_flat[pytree-compile]	0.2180ms	68.8135μs	14.5320 KOps/s	14.3578 KOps/s	$\color{#35bf28}+1.21\%$
test_compile_copy_flat[pytree-eager]	0.1384ms	51.4516μs	19.4357 KOps/s	19.2906 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_assign_and_add[tensordict-compile]	1.6371ms	0.4003ms	2.4983 KOps/s	2.2053 KOps/s	$\textbf{\color{#35bf28}+13.29\%}$
test_compile_assign_and_add[tensordict-eager]	2.9187ms	2.7465ms	364.1025 Ops/s	376.8972 Ops/s	$\color{#d91a1a}-3.39\%$
test_compile_assign_and_add[pytree-compile]	1.5984ms	0.4325ms	2.3124 KOps/s	2.2681 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_assign_and_add[pytree-eager]	2.8367ms	2.7249ms	366.9871 Ops/s	376.7115 Ops/s	$\color{#d91a1a}-2.58\%$
test_compile_indexing[tensor-tensordict-compile]	0.5111ms	0.1176ms	8.5031 KOps/s	8.5720 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_indexing[tensor-tensordict-eager]	0.5650ms	80.7123μs	12.3897 KOps/s	12.1327 KOps/s	$\color{#35bf28}+2.12\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5430ms	0.1128ms	8.8623 KOps/s	9.2199 KOps/s	$\color{#d91a1a}-3.88\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1290ms	69.8626μs	14.3138 KOps/s	14.2006 KOps/s	$\color{#35bf28}+0.80\%$
test_compile_indexing[tensor-pytree-compile]	0.2335ms	0.1139ms	8.7799 KOps/s	9.1360 KOps/s	$\color{#d91a1a}-3.90\%$
test_compile_indexing[tensor-pytree-eager]	0.1521ms	69.4133μs	14.4065 KOps/s	14.2129 KOps/s	$\color{#35bf28}+1.36\%$
test_compile_indexing[slice-tensordict-compile]	0.1480ms	0.1040ms	9.6187 KOps/s	9.8932 KOps/s	$\color{#d91a1a}-2.77\%$
test_compile_indexing[slice-tensordict-eager]	0.1480ms	17.2742μs	57.8897 KOps/s	57.1565 KOps/s	$\color{#35bf28}+1.28\%$
test_compile_indexing[slice-tensorclass-compile]	0.1439ms	97.0872μs	10.3000 KOps/s	10.3329 KOps/s	$\color{#d91a1a}-0.32\%$
test_compile_indexing[slice-tensorclass-eager]	84.8110μs	15.6353μs	63.9577 KOps/s	64.3575 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_indexing[slice-pytree-compile]	0.1671ms	97.5174μs	10.2546 KOps/s	10.2303 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_indexing[slice-pytree-eager]	47.1010μs	15.6956μs	63.7121 KOps/s	57.7693 KOps/s	$\textbf{\color{#35bf28}+10.29\%}$
test_compile_indexing[int-tensordict-compile]	0.1515ms	0.1028ms	9.7282 KOps/s	9.6683 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_indexing[int-tensordict-eager]	0.5919ms	17.0922μs	58.5063 KOps/s	57.5289 KOps/s	$\color{#35bf28}+1.70\%$
test_compile_indexing[int-tensorclass-compile]	0.1636ms	98.0224μs	10.2018 KOps/s	9.9624 KOps/s	$\color{#35bf28}+2.40\%$
test_compile_indexing[int-tensorclass-eager]	50.9010μs	15.7669μs	63.4239 KOps/s	64.5180 KOps/s	$\color{#d91a1a}-1.70\%$
test_compile_indexing[int-pytree-compile]	0.1591ms	97.6275μs	10.2430 KOps/s	10.1749 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_indexing[int-pytree-eager]	0.2485ms	17.5007μs	57.1406 KOps/s	64.6684 KOps/s	$\textbf{\color{#d91a1a}-11.64\%}$
test_mod_add[eager]	77.0410μs	39.5224μs	25.3021 KOps/s	26.1127 KOps/s	$\color{#d91a1a}-3.10\%$
test_mod_add[compile]	0.3082ms	82.0436μs	12.1886 KOps/s	12.1894 KOps/s	$-0.01\%$
test_mod_add[compile-overhead]	0.3390ms	0.1715ms	5.8308 KOps/s	5.6896 KOps/s	$\color{#35bf28}+2.48\%$
test_mod_wrap[eager]	0.3380ms	0.2559ms	3.9074 KOps/s	3.7725 KOps/s	$\color{#35bf28}+3.58\%$
test_mod_wrap[compile]	0.3592ms	0.2894ms	3.4552 KOps/s	3.4658 KOps/s	$\color{#d91a1a}-0.31\%$
test_mod_wrap[compile-overhead]	7.2202ms	3.8200ms	261.7777 Ops/s	268.9822 Ops/s	$\color{#d91a1a}-2.68\%$
test_mod_wrap_and_backward[eager]	1.5329ms	1.3913ms	718.7608 Ops/s	677.9732 Ops/s	$\textbf{\color{#35bf28}+6.02\%}$
test_mod_wrap_and_backward[compile]	1.3867ms	1.2899ms	775.2711 Ops/s	714.5655 Ops/s	$\textbf{\color{#35bf28}+8.50\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3848ms	0.9340ms	1.0707 KOps/s	954.8104 Ops/s	$\textbf{\color{#35bf28}+12.13\%}$
test_seq_add[eager]	0.1868ms	0.1201ms	8.3245 KOps/s	8.5295 KOps/s	$\color{#d91a1a}-2.40\%$
test_seq_add[compile]	0.1572ms	89.1346μs	11.2190 KOps/s	10.8283 KOps/s	$\color{#35bf28}+3.61\%$
test_seq_add[compile-overhead]	0.1697ms	0.1305ms	7.6603 KOps/s	7.6080 KOps/s	$\color{#35bf28}+0.69\%$
test_seq_wrap[eager]	0.5195ms	0.4337ms	2.3055 KOps/s	2.2236 KOps/s	$\color{#35bf28}+3.68\%$
test_seq_wrap[compile]	0.4825ms	0.3118ms	3.2073 KOps/s	3.1003 KOps/s	$\color{#35bf28}+3.45\%$
test_seq_wrap[compile-overhead]	0.3094ms	0.2269ms	4.4066 KOps/s	4.3613 KOps/s	$\color{#35bf28}+1.04\%$
test_func_call_runtime[False-eager]	0.8157ms	0.7500ms	1.3333 KOps/s	1.2357 KOps/s	$\textbf{\color{#35bf28}+7.90\%}$
test_func_call_runtime[False-compile]	0.8971ms	0.7634ms	1.3100 KOps/s	1.3111 KOps/s	$\color{#d91a1a}-0.09\%$
test_func_call_runtime[False-compile-overhead]	0.4105ms	0.3689ms	2.7110 KOps/s	2.7214 KOps/s	$\color{#d91a1a}-0.38\%$
test_func_call_runtime[True-eager]	0.9923ms	0.9187ms	1.0885 KOps/s	1.0705 KOps/s	$\color{#35bf28}+1.68\%$
test_func_call_runtime[True-compile]	0.8481ms	0.7842ms	1.2752 KOps/s	1.2902 KOps/s	$\color{#d91a1a}-1.16\%$
test_func_call_runtime[True-compile-overhead]	0.5562ms	0.3894ms	2.5682 KOps/s	2.5692 KOps/s	$\color{#d91a1a}-0.04\%$
test_func_call_cm_runtime[False-eager]	0.8293ms	0.7510ms	1.3316 KOps/s	1.3103 KOps/s	$\color{#35bf28}+1.62\%$
test_func_call_cm_runtime[False-compile]	1.0111ms	0.7614ms	1.3134 KOps/s	1.2879 KOps/s	$\color{#35bf28}+1.97\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4484ms	0.3688ms	2.7112 KOps/s	2.7081 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[True-eager]	1.1059ms	1.0215ms	978.9055 Ops/s	961.0818 Ops/s	$\color{#35bf28}+1.85\%$
test_func_call_cm_runtime[True-compile]	1.1188ms	1.0050ms	995.0678 Ops/s	975.3712 Ops/s	$\color{#35bf28}+2.02\%$
test_func_call_cm_runtime[True-compile-overhead]	1.0697ms	1.0064ms	993.6562 Ops/s	971.0163 Ops/s	$\color{#35bf28}+2.33\%$
test_vmap_func_call_cm_runtime[eager]	2.5238ms	2.1032ms	475.4685 Ops/s	467.7946 Ops/s	$\color{#35bf28}+1.64\%$
test_vmap_func_call_cm_runtime[compile]	0.9021ms	0.8268ms	1.2096 KOps/s	1.2106 KOps/s	$\color{#d91a1a}-0.09\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4649ms	0.4187ms	2.3885 KOps/s	2.3707 KOps/s	$\color{#35bf28}+0.75\%$
test_distributed	2.8979ms	0.2354ms	4.2482 KOps/s	8.7807 KOps/s	$\textbf{\color{#d91a1a}-51.62\%}$
test_tdmodule	36.2710μs	20.8044μs	48.0668 KOps/s	50.5049 KOps/s	$\color{#d91a1a}-4.83\%$
test_tdmodule_dispatch	83.2810μs	37.2086μs	26.8755 KOps/s	28.3575 KOps/s	$\textbf{\color{#d91a1a}-5.23\%}$
test_tdseq	31.0800μs	21.4020μs	46.7246 KOps/s	48.2386 KOps/s	$\color{#d91a1a}-3.14\%$
test_tdseq_dispatch	58.1910μs	38.6556μs	25.8694 KOps/s	25.8834 KOps/s	$\color{#d91a1a}-0.05\%$
test_instantiation_functorch	1.6976ms	1.5738ms	635.4113 Ops/s	649.6501 Ops/s	$\color{#d91a1a}-2.19\%$
test_exec_functorch	0.1873ms	0.1461ms	6.8451 KOps/s	7.0015 KOps/s	$\color{#d91a1a}-2.23\%$
test_exec_functional_call	0.1982ms	0.1411ms	7.0885 KOps/s	7.3345 KOps/s	$\color{#d91a1a}-3.35\%$
test_exec_td_decorator	0.3803ms	0.1909ms	5.2394 KOps/s	5.3129 KOps/s	$\color{#d91a1a}-1.38\%$
test_vmap_mlp_speed_decorator[True-True]	0.8366ms	0.6931ms	1.4428 KOps/s	1.4404 KOps/s	$\color{#35bf28}+0.17\%$
test_vmap_mlp_speed_decorator[True-False]	0.8281ms	0.6913ms	1.4465 KOps/s	1.4427 KOps/s	$\color{#35bf28}+0.27\%$
test_vmap_mlp_speed_decorator[False-True]	0.7136ms	0.6010ms	1.6639 KOps/s	1.6552 KOps/s	$\color{#35bf28}+0.52\%$
test_vmap_mlp_speed_decorator[False-False]	0.7155ms	0.6005ms	1.6653 KOps/s	1.6515 KOps/s	$\color{#35bf28}+0.84\%$
test_vmap_transformer_speed_decorator[True-True]	20.1005ms	19.4450ms	51.4272 Ops/s	51.1006 Ops/s	$\color{#35bf28}+0.64\%$
test_vmap_transformer_speed_decorator[True-False]	19.5093ms	19.4289ms	51.4696 Ops/s	51.2784 Ops/s	$\color{#35bf28}+0.37\%$
test_vmap_transformer_speed_decorator[False-True]	19.3391ms	19.2907ms	51.8385 Ops/s	51.6929 Ops/s	$\color{#35bf28}+0.28\%$
test_vmap_transformer_speed_decorator[False-False]	19.3845ms	19.2929ms	51.8326 Ops/s	51.6996 Ops/s	$\color{#35bf28}+0.26\%$
test_to_module_speed[True]	1.4302ms	0.9694ms	1.0316 KOps/s	1.0240 KOps/s	$\color{#35bf28}+0.74\%$
test_to_module_speed[False]	1.0211ms	0.9507ms	1.0519 KOps/s	1.0391 KOps/s	$\color{#35bf28}+1.23\%$
test_tc_init	75.6310μs	36.6059μs	27.3180 KOps/s	28.2507 KOps/s	$\color{#d91a1a}-3.30\%$
test_tc_init_nested	0.1178ms	73.0007μs	13.6985 KOps/s	14.4008 KOps/s	$\color{#d91a1a}-4.88\%$
test_tc_first_layer_tensor	24.7310μs	0.7970μs	1.2547 MOps/s	1.2590 MOps/s	$\color{#d91a1a}-0.35\%$
test_tc_first_layer_nontensor	31.6900μs	2.2046μs	453.5987 KOps/s	451.2249 KOps/s	$\color{#35bf28}+0.53\%$
test_tc_second_layer_tensor	9.4402μs	1.4078μs	710.3391 KOps/s	709.0221 KOps/s	$\color{#35bf28}+0.19\%$
test_tc_second_layer_nontensor	31.2910μs	2.9328μs	340.9676 KOps/s	335.3211 KOps/s	$\color{#35bf28}+1.68\%$
test_unbind	0.2164s	11.9649ms	83.5775 Ops/s	141.1119 Ops/s	$\textbf{\color{#d91a1a}-40.77\%}$
test_full_like	9.3135ms	9.1365ms	109.4505 Ops/s	108.2235 Ops/s	$\color{#35bf28}+1.13\%$
test_zeros_like	6.7772ms	4.3513ms	229.8182 Ops/s	234.7166 Ops/s	$\color{#d91a1a}-2.09\%$
test_ones_like	4.9591ms	4.3278ms	231.0635 Ops/s	235.4041 Ops/s	$\color{#d91a1a}-1.84\%$
test_clone	6.9333ms	6.3820ms	156.6903 Ops/s	109.8489 Ops/s	$\textbf{\color{#35bf28}+42.64\%}$
test_squeeze	53.4510μs	10.0993μs	99.0171 KOps/s	103.2357 KOps/s	$\color{#d91a1a}-4.09\%$
test_unsqueeze	0.1264ms	74.4285μs	13.4357 KOps/s	13.5746 KOps/s	$\color{#d91a1a}-1.02\%$
test_split	0.3711ms	0.1621ms	6.1703 KOps/s	6.1252 KOps/s	$\color{#35bf28}+0.74\%$
test_permute	0.2425ms	0.1859ms	5.3797 KOps/s	5.4420 KOps/s	$\color{#d91a1a}-1.14\%$
test_stack	50.8762ms	50.5157ms	19.7958 Ops/s	19.9554 Ops/s	$\color{#d91a1a}-0.80\%$
test_cat	50.6447ms	50.2292ms	19.9088 Ops/s	20.0065 Ops/s	$\color{#d91a1a}-0.49\%$