[Refactor] Improve functional call efficiency

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 113. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.3920μs	16.3499μs	61.1625 KOps/s	60.8033 KOps/s	$\color{#35bf28}+0.59\%$
test_plain_set_stack_nested	0.2911ms	0.1485ms	6.7335 KOps/s	6.7729 KOps/s	$\color{#d91a1a}-0.58\%$
test_plain_set_nested_inplace	69.2480μs	19.3836μs	51.5900 KOps/s	51.4667 KOps/s	$\color{#35bf28}+0.24\%$
test_plain_set_stack_nested_inplace	0.4063ms	0.1757ms	5.6900 KOps/s	5.7172 KOps/s	$\color{#d91a1a}-0.47\%$
test_items	0.1627ms	2.6808μs	373.0212 KOps/s	405.0031 KOps/s	$\textbf{\color{#d91a1a}-7.90\%}$
test_items_nested	1.3161ms	0.2731ms	3.6617 KOps/s	3.6793 KOps/s	$\color{#d91a1a}-0.48\%$
test_items_nested_locked	0.3928ms	0.2697ms	3.7077 KOps/s	3.7163 KOps/s	$\color{#d91a1a}-0.23\%$
test_items_nested_leaf	0.8663ms	0.1736ms	5.7593 KOps/s	6.0094 KOps/s	$\color{#d91a1a}-4.16\%$
test_items_stack_nested	2.6182ms	1.5238ms	656.2486 Ops/s	684.7777 Ops/s	$\color{#d91a1a}-4.17\%$
test_items_stack_nested_leaf	2.5099ms	1.4151ms	706.6753 Ops/s	754.9729 Ops/s	$\textbf{\color{#d91a1a}-6.40\%}$
test_items_stack_nested_locked	1.1996ms	0.7626ms	1.3113 KOps/s	1.2988 KOps/s	$\color{#35bf28}+0.97\%$
test_keys	41.6180μs	4.8424μs	206.5080 KOps/s	255.3977 KOps/s	$\textbf{\color{#d91a1a}-19.14\%}$
test_keys_nested	1.5238ms	0.1509ms	6.6252 KOps/s	6.7517 KOps/s	$\color{#d91a1a}-1.87\%$
test_keys_nested_locked	0.2670ms	0.1411ms	7.0877 KOps/s	7.0900 KOps/s	$\color{#d91a1a}-0.03\%$
test_keys_nested_leaf	0.3867ms	0.1407ms	7.1048 KOps/s	7.2393 KOps/s	$\color{#d91a1a}-1.86\%$
test_keys_stack_nested	1.8532ms	1.4088ms	709.8094 Ops/s	727.3771 Ops/s	$\color{#d91a1a}-2.42\%$
test_keys_stack_nested_leaf	1.5072ms	1.4035ms	712.4796 Ops/s	723.6928 Ops/s	$\color{#d91a1a}-1.55\%$
test_keys_stack_nested_locked	1.3501ms	0.6610ms	1.5128 KOps/s	1.4748 KOps/s	$\color{#35bf28}+2.58\%$
test_values	21.9283μs	1.1628μs	859.9575 KOps/s	859.4670 KOps/s	$\color{#35bf28}+0.06\%$
test_values_nested	95.8180μs	50.2259μs	19.9100 KOps/s	20.2167 KOps/s	$\color{#d91a1a}-1.52\%$
test_values_nested_locked	97.2900μs	49.9500μs	20.0200 KOps/s	20.0410 KOps/s	$\color{#d91a1a}-0.10\%$
test_values_nested_leaf	75.4600μs	44.4162μs	22.5143 KOps/s	22.3275 KOps/s	$\color{#35bf28}+0.84\%$
test_values_stack_nested	1.3409ms	1.2076ms	828.0931 Ops/s	851.6675 Ops/s	$\color{#d91a1a}-2.77\%$
test_values_stack_nested_leaf	1.4246ms	1.1900ms	840.3581 Ops/s	859.6559 Ops/s	$\color{#d91a1a}-2.24\%$
test_values_stack_nested_locked	0.6945ms	0.5060ms	1.9765 KOps/s	1.9328 KOps/s	$\color{#35bf28}+2.26\%$
test_membership	22.2610μs	1.3507μs	740.3711 KOps/s	726.3895 KOps/s	$\color{#35bf28}+1.92\%$
test_membership_nested	42.8900μs	2.8028μs	356.7881 KOps/s	355.1947 KOps/s	$\color{#35bf28}+0.45\%$
test_membership_nested_leaf	20.4080μs	2.7785μs	359.9057 KOps/s	353.0000 KOps/s	$\color{#35bf28}+1.96\%$
test_membership_stacked_nested	56.5750μs	12.1121μs	82.5619 KOps/s	84.1272 KOps/s	$\color{#d91a1a}-1.86\%$
test_membership_stacked_nested_leaf	0.1187ms	12.9604μs	77.1581 KOps/s	84.8124 KOps/s	$\textbf{\color{#d91a1a}-9.02\%}$
test_membership_nested_last	49.1010μs	5.9223μs	168.8527 KOps/s	168.9552 KOps/s	$\color{#d91a1a}-0.06\%$
test_membership_nested_leaf_last	32.5100μs	5.8923μs	169.7128 KOps/s	167.6926 KOps/s	$\color{#35bf28}+1.20\%$
test_membership_stacked_nested_last	0.2853ms	0.1695ms	5.9002 KOps/s	5.8767 KOps/s	$\color{#35bf28}+0.40\%$
test_membership_stacked_nested_leaf_last	65.5690μs	14.1912μs	70.4660 KOps/s	71.9069 KOps/s	$\color{#d91a1a}-2.00\%$
test_nested_getleaf	30.8170μs	10.8293μs	92.3425 KOps/s	93.9048 KOps/s	$\color{#d91a1a}-1.66\%$
test_nested_get	34.1130μs	10.2198μs	97.8494 KOps/s	98.7061 KOps/s	$\color{#d91a1a}-0.87\%$
test_stacked_getleaf	1.1213ms	0.6495ms	1.5396 KOps/s	1.6099 KOps/s	$\color{#d91a1a}-4.37\%$
test_stacked_get	1.2159ms	0.6131ms	1.6310 KOps/s	1.6942 KOps/s	$\color{#d91a1a}-3.73\%$
test_nested_getitemleaf	0.1270ms	11.2172μs	89.1491 KOps/s	93.1218 KOps/s	$\color{#d91a1a}-4.27\%$
test_nested_getitem	42.8290μs	10.0364μs	99.6374 KOps/s	98.1196 KOps/s	$\color{#35bf28}+1.55\%$
test_stacked_getitemleaf	1.2130ms	0.6500ms	1.5385 KOps/s	1.5472 KOps/s	$\color{#d91a1a}-0.56\%$
test_stacked_getitem	1.0322ms	0.6122ms	1.6333 KOps/s	1.6941 KOps/s	$\color{#d91a1a}-3.59\%$
test_lock_nested	61.6082ms	0.5543ms	1.8041 KOps/s	2.0121 KOps/s	$\textbf{\color{#d91a1a}-10.34\%}$
test_lock_stack_nested	91.3765ms	8.9422ms	111.8289 Ops/s	117.1341 Ops/s	$\color{#d91a1a}-4.53\%$
test_unlock_nested	71.0665ms	0.5176ms	1.9321 KOps/s	1.9421 KOps/s	$\color{#d91a1a}-0.51\%$
test_unlock_stack_nested	79.0098ms	8.4626ms	118.1669 Ops/s	202.9720 Ops/s	$\textbf{\color{#d91a1a}-41.78\%}$
test_flatten_speed	0.7222ms	0.2705ms	3.6973 KOps/s	3.6769 KOps/s	$\color{#35bf28}+0.56\%$
test_unflatten_speed	1.2667ms	0.4728ms	2.1149 KOps/s	2.1077 KOps/s	$\color{#35bf28}+0.34\%$
test_common_ops	1.4755ms	0.6937ms	1.4416 KOps/s	1.4068 KOps/s	$\color{#35bf28}+2.47\%$
test_creation	26.2480μs	2.4419μs	409.5184 KOps/s	405.5477 KOps/s	$\color{#35bf28}+0.98\%$
test_creation_empty	38.2920μs	8.5579μs	116.8509 KOps/s	112.0649 KOps/s	$\color{#35bf28}+4.27\%$
test_creation_nested_1	25.9380μs	12.1243μs	82.4792 KOps/s	77.3122 KOps/s	$\textbf{\color{#35bf28}+6.68\%}$
test_creation_nested_2	65.8220μs	15.3840μs	65.0024 KOps/s	61.4068 KOps/s	$\textbf{\color{#35bf28}+5.86\%}$
test_clone	0.2143ms	13.4427μs	74.3899 KOps/s	76.0741 KOps/s	$\color{#d91a1a}-2.21\%$
test_getitem[int]	58.8100μs	12.6932μs	78.7822 KOps/s	77.9121 KOps/s	$\color{#35bf28}+1.12\%$
test_getitem[slice_int]	69.6500μs	24.6366μs	40.5901 KOps/s	40.6617 KOps/s	$\color{#d91a1a}-0.18\%$
test_getitem[range]	96.6090μs	43.1414μs	23.1796 KOps/s	22.5144 KOps/s	$\color{#35bf28}+2.95\%$
test_getitem[tuple]	53.2680μs	20.4100μs	48.9957 KOps/s	49.8519 KOps/s	$\color{#d91a1a}-1.72\%$
test_getitem[list]	80.3980μs	38.7185μs	25.8275 KOps/s	24.9051 KOps/s	$\color{#35bf28}+3.70\%$
test_setitem_dim[int]	47.3470μs	27.4519μs	36.4274 KOps/s	36.0508 KOps/s	$\color{#35bf28}+1.04\%$
test_setitem_dim[slice_int]	84.3760μs	51.8672μs	19.2800 KOps/s	19.3474 KOps/s	$\color{#d91a1a}-0.35\%$
test_setitem_dim[range]	0.1212ms	72.0116μs	13.8867 KOps/s	13.7889 KOps/s	$\color{#35bf28}+0.71\%$
test_setitem_dim[tuple]	0.1135ms	40.9736μs	24.4060 KOps/s	24.3050 KOps/s	$\color{#35bf28}+0.42\%$
test_setitem	0.2502ms	18.7232μs	53.4097 KOps/s	53.4992 KOps/s	$\color{#d91a1a}-0.17\%$
test_set	0.2319ms	18.1232μs	55.1778 KOps/s	55.1898 KOps/s	$\color{#d91a1a}-0.02\%$
test_set_shared	3.3573ms	0.1408ms	7.1033 KOps/s	7.2263 KOps/s	$\color{#d91a1a}-1.70\%$
test_update	0.2095ms	24.4040μs	40.9769 KOps/s	41.4602 KOps/s	$\color{#d91a1a}-1.17\%$
test_update_nested	0.3807ms	36.6963μs	27.2507 KOps/s	27.8001 KOps/s	$\color{#d91a1a}-1.98\%$
test_set_nested	0.2146ms	19.8791μs	50.3041 KOps/s	50.0880 KOps/s	$\color{#35bf28}+0.43\%$
test_set_nested_new	0.2211ms	25.3880μs	39.3887 KOps/s	37.9763 KOps/s	$\color{#35bf28}+3.72\%$
test_select	98.5830μs	49.7152μs	20.1146 KOps/s	19.2303 KOps/s	$\color{#35bf28}+4.60\%$
test_unbind_speed	0.4633ms	0.3685ms	2.7138 KOps/s	2.6979 KOps/s	$\color{#35bf28}+0.59\%$
test_unbind_speed_stack0	69.7567ms	5.6275ms	177.6997 Ops/s	175.6044 Ops/s	$\color{#35bf28}+1.19\%$
test_unbind_speed_stack1	1.6030μs	0.6445μs	1.5517 MOps/s	1.5998 MOps/s	$\color{#d91a1a}-3.01\%$
test_split	1.7025ms	1.6314ms	612.9627 Ops/s	611.4281 Ops/s	$\color{#35bf28}+0.25\%$
test_chunk	62.3735ms	1.7439ms	573.4190 Ops/s	569.1813 Ops/s	$\color{#35bf28}+0.74\%$
test_creation[device0]	3.6752ms	0.3027ms	3.3034 KOps/s	3.1244 KOps/s	$\textbf{\color{#35bf28}+5.73\%}$
test_creation_from_tensor	60.0502ms	0.3677ms	2.7198 KOps/s	2.6764 KOps/s	$\color{#35bf28}+1.62\%$
test_add_one[memmap_tensor0]	70.4610μs	24.4884μs	40.8357 KOps/s	39.8896 KOps/s	$\color{#35bf28}+2.37\%$
test_contiguous[memmap_tensor0]	33.1810μs	5.6485μs	177.0374 KOps/s	175.4126 KOps/s	$\color{#35bf28}+0.93\%$
test_stack[memmap_tensor0]	67.1740μs	18.6293μs	53.6789 KOps/s	51.6743 KOps/s	$\color{#35bf28}+3.88\%$
test_memmaptd_index	0.4305ms	0.1879ms	5.3210 KOps/s	5.2488 KOps/s	$\color{#35bf28}+1.38\%$
test_memmaptd_index_astensor	0.4081ms	0.2499ms	4.0015 KOps/s	3.9370 KOps/s	$\color{#35bf28}+1.64\%$
test_memmaptd_index_op	0.9349ms	0.4864ms	2.0560 KOps/s	2.0120 KOps/s	$\color{#35bf28}+2.18\%$
test_reshape_pytree	52.0570μs	23.2417μs	43.0260 KOps/s	42.9671 KOps/s	$\color{#35bf28}+0.14\%$
test_reshape_td	73.2360μs	31.2186μs	32.0322 KOps/s	30.9588 KOps/s	$\color{#35bf28}+3.47\%$
test_view_pytree	52.5280μs	23.3549μs	42.8176 KOps/s	43.4072 KOps/s	$\color{#d91a1a}-1.36\%$
test_view_td	22.6920μs	4.8195μs	207.4895 KOps/s	206.8551 KOps/s	$\color{#35bf28}+0.31\%$
test_unbind_pytree	0.2067ms	28.6962μs	34.8478 KOps/s	38.0687 KOps/s	$\textbf{\color{#d91a1a}-8.46\%}$
test_unbind_td	0.1092ms	58.5145μs	17.0898 KOps/s	16.8906 KOps/s	$\color{#35bf28}+1.18\%$
test_split_pytree	56.4650μs	26.3504μs	37.9501 KOps/s	37.6899 KOps/s	$\color{#35bf28}+0.69\%$
test_split_td	0.1323ms	45.5219μs	21.9674 KOps/s	21.5815 KOps/s	$\color{#35bf28}+1.79\%$
test_add_pytree	79.1470μs	31.7549μs	31.4912 KOps/s	31.4766 KOps/s	$\color{#35bf28}+0.05\%$
test_add_td	0.1422ms	44.1417μs	22.6543 KOps/s	21.8671 KOps/s	$\color{#35bf28}+3.60\%$
test_distributed	17.8530μs	6.3633μs	157.1500 KOps/s	168.0954 KOps/s	$\textbf{\color{#d91a1a}-6.51\%}$
test_tdmodule	0.1066ms	21.1744μs	47.2267 KOps/s	46.4830 KOps/s	$\color{#35bf28}+1.60\%$
test_tdmodule_dispatch	0.1800ms	39.0708μs	25.5946 KOps/s	25.2814 KOps/s	$\color{#35bf28}+1.24\%$
test_tdseq	50.1830μs	24.5409μs	40.7483 KOps/s	40.9599 KOps/s	$\color{#d91a1a}-0.52\%$
test_tdseq_dispatch	0.1347ms	43.4939μs	22.9917 KOps/s	22.7276 KOps/s	$\color{#35bf28}+1.16\%$
test_instantiation_functorch	1.3813ms	1.2806ms	780.9079 Ops/s	777.5176 Ops/s	$\color{#35bf28}+0.44\%$
test_instantiation_td	1.4325ms	0.9992ms	1.0008 KOps/s	995.5540 Ops/s	$\color{#35bf28}+0.53\%$
test_exec_functorch	0.3485ms	0.1617ms	6.1855 KOps/s	6.3716 KOps/s	$\color{#d91a1a}-2.92\%$
test_exec_functional_call	0.2255ms	0.1439ms	6.9490 KOps/s	6.8448 KOps/s	$\color{#35bf28}+1.52\%$
test_exec_td	0.4636ms	0.1435ms	6.9669 KOps/s	7.0820 KOps/s	$\color{#d91a1a}-1.63\%$
test_exec_td_decorator	1.0311ms	0.2171ms	4.6054 KOps/s	4.0619 KOps/s	$\textbf{\color{#35bf28}+13.38\%}$
test_vmap_mlp_speed[True-True]	1.2534ms	0.8935ms	1.1192 KOps/s	1.1242 KOps/s	$\color{#d91a1a}-0.45\%$
test_vmap_mlp_speed[True-False]	1.3172ms	0.4848ms	2.0628 KOps/s	2.1430 KOps/s	$\color{#d91a1a}-3.75\%$
test_vmap_mlp_speed[False-True]	1.2179ms	0.7777ms	1.2859 KOps/s	1.2859 KOps/s	$+0.00\%$
test_vmap_mlp_speed[False-False]	1.0457ms	0.4056ms	2.4656 KOps/s	2.5991 KOps/s	$\textbf{\color{#d91a1a}-5.14\%}$
test_vmap_mlp_speed_decorator[True-True]	2.2594ms	1.5668ms	638.2491 Ops/s	540.0427 Ops/s	$\textbf{\color{#35bf28}+18.18\%}$
test_vmap_mlp_speed_decorator[True-False]	0.9945ms	0.5492ms	1.8208 KOps/s	1.7247 KOps/s	$\textbf{\color{#35bf28}+5.57\%}$
test_vmap_mlp_speed_decorator[False-True]	1.8058ms	1.3554ms	737.7642 Ops/s	619.8881 Ops/s	$\textbf{\color{#35bf28}+19.02\%}$
test_vmap_mlp_speed_decorator[False-False]	0.9009ms	0.4251ms	2.3526 KOps/s	2.1978 KOps/s	$\textbf{\color{#35bf28}+7.05\%}$

[github-actions]

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

Total Benchmarks: 127. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.5600ms	12.4435μs	80.3634 KOps/s	78.8103 KOps/s	$\color{#35bf28}+1.97\%$
test_plain_set_stack_nested	0.1368ms	0.1140ms	8.7732 KOps/s	8.2374 KOps/s	$\textbf{\color{#35bf28}+6.51\%}$
test_plain_set_nested_inplace	31.9910μs	14.9392μs	66.9382 KOps/s	66.7472 KOps/s	$\color{#35bf28}+0.29\%$
test_plain_set_stack_nested_inplace	0.1737ms	0.1398ms	7.1525 KOps/s	7.1543 KOps/s	$\color{#d91a1a}-0.03\%$
test_items	27.1900μs	4.7468μs	210.6672 KOps/s	211.1098 KOps/s	$\color{#d91a1a}-0.21\%$
test_items_nested	0.3738ms	0.3383ms	2.9557 KOps/s	2.9742 KOps/s	$\color{#d91a1a}-0.62\%$
test_items_nested_locked	0.3587ms	0.3368ms	2.9691 KOps/s	2.9601 KOps/s	$\color{#35bf28}+0.31\%$
test_items_nested_leaf	0.2221ms	0.1988ms	5.0293 KOps/s	5.0367 KOps/s	$\color{#d91a1a}-0.15\%$
test_items_stack_nested	1.5322ms	1.4929ms	669.8484 Ops/s	687.0648 Ops/s	$\color{#d91a1a}-2.51\%$
test_items_stack_nested_leaf	1.3610ms	1.3096ms	763.5654 Ops/s	779.9794 Ops/s	$\color{#d91a1a}-2.10\%$
test_items_stack_nested_locked	0.8597ms	0.8185ms	1.2217 KOps/s	1.2188 KOps/s	$\color{#35bf28}+0.24\%$
test_keys	25.4400μs	4.7912μs	208.7149 KOps/s	216.1124 KOps/s	$\color{#d91a1a}-3.42\%$
test_keys_nested	1.2051ms	89.8260μs	11.1326 KOps/s	11.0851 KOps/s	$\color{#35bf28}+0.43\%$
test_keys_nested_locked	0.1201ms	89.2809μs	11.2006 KOps/s	11.1176 KOps/s	$\color{#35bf28}+0.75\%$
test_keys_nested_leaf	42.3150ms	85.9087μs	11.6403 KOps/s	12.2449 KOps/s	$\color{#d91a1a}-4.94\%$
test_keys_stack_nested	1.3565ms	1.3065ms	765.4025 Ops/s	786.6514 Ops/s	$\color{#d91a1a}-2.70\%$
test_keys_stack_nested_leaf	1.3327ms	1.3013ms	768.4577 Ops/s	783.5534 Ops/s	$\color{#d91a1a}-1.93\%$
test_keys_stack_nested_locked	0.6776ms	0.6233ms	1.6043 KOps/s	1.6142 KOps/s	$\color{#d91a1a}-0.61\%$
test_values	9.2600μs	1.9002μs	526.2615 KOps/s	524.0092 KOps/s	$\color{#35bf28}+0.43\%$
test_values_nested	73.3410μs	43.0276μs	23.2409 KOps/s	23.1851 KOps/s	$\color{#35bf28}+0.24\%$
test_values_nested_locked	66.2310μs	43.1296μs	23.1859 KOps/s	22.9951 KOps/s	$\color{#35bf28}+0.83\%$
test_values_nested_leaf	64.2710μs	37.2400μs	26.8528 KOps/s	26.6395 KOps/s	$\color{#35bf28}+0.80\%$
test_values_stack_nested	1.1891ms	1.1482ms	870.9081 Ops/s	903.7586 Ops/s	$\color{#d91a1a}-3.63\%$
test_values_stack_nested_leaf	1.1615ms	1.1255ms	888.4834 Ops/s	906.3536 Ops/s	$\color{#d91a1a}-1.97\%$
test_values_stack_nested_locked	0.5307ms	0.4962ms	2.0153 KOps/s	2.0166 KOps/s	$\color{#d91a1a}-0.07\%$
test_membership	5.4380μs	0.9341μs	1.0705 MOps/s	1.0505 MOps/s	$\color{#35bf28}+1.90\%$
test_membership_nested	13.4455μs	2.1409μs	467.0866 KOps/s	472.6314 KOps/s	$\color{#d91a1a}-1.17\%$
test_membership_nested_leaf	16.8500μs	2.1365μs	468.0498 KOps/s	472.9976 KOps/s	$\color{#d91a1a}-1.05\%$
test_membership_stacked_nested	45.8510μs	10.8318μs	92.3208 KOps/s	92.2587 KOps/s	$\color{#35bf28}+0.07\%$
test_membership_stacked_nested_leaf	33.1900μs	10.9361μs	91.4400 KOps/s	92.1949 KOps/s	$\color{#d91a1a}-0.82\%$
test_membership_nested_last	23.4800μs	4.6464μs	215.2195 KOps/s	215.9088 KOps/s	$\color{#d91a1a}-0.32\%$
test_membership_nested_leaf_last	43.5610μs	4.6473μs	215.1765 KOps/s	216.8417 KOps/s	$\color{#d91a1a}-0.77\%$
test_membership_stacked_nested_last	0.1699ms	0.1348ms	7.4157 KOps/s	7.5515 KOps/s	$\color{#d91a1a}-1.80\%$
test_membership_stacked_nested_leaf_last	43.6600μs	12.6286μs	79.1851 KOps/s	78.0133 KOps/s	$\color{#35bf28}+1.50\%$
test_nested_getleaf	28.9900μs	8.4037μs	118.9951 KOps/s	119.2803 KOps/s	$\color{#d91a1a}-0.24\%$
test_nested_get	33.1700μs	7.9474μs	125.8273 KOps/s	125.7483 KOps/s	$\color{#35bf28}+0.06\%$
test_stacked_getleaf	0.6280ms	0.5681ms	1.7602 KOps/s	1.8373 KOps/s	$\color{#d91a1a}-4.20\%$
test_stacked_get	0.5863ms	0.5419ms	1.8453 KOps/s	1.9479 KOps/s	$\textbf{\color{#d91a1a}-5.27\%}$
test_nested_getitemleaf	30.0010μs	8.4310μs	118.6103 KOps/s	117.9235 KOps/s	$\color{#35bf28}+0.58\%$
test_nested_getitem	32.2210μs	7.9841μs	125.2487 KOps/s	125.0657 KOps/s	$\color{#35bf28}+0.15\%$
test_stacked_getitemleaf	0.6516ms	0.5762ms	1.7354 KOps/s	1.8302 KOps/s	$\textbf{\color{#d91a1a}-5.18\%}$
test_stacked_getitem	0.5665ms	0.5385ms	1.8569 KOps/s	1.9454 KOps/s	$\color{#d91a1a}-4.55\%$
test_lock_nested	4.3898ms	0.4591ms	2.1780 KOps/s	2.1683 KOps/s	$\color{#35bf28}+0.44\%$
test_lock_stack_nested	70.2399ms	6.6020ms	151.4682 Ops/s	149.0229 Ops/s	$\color{#35bf28}+1.64\%$
test_unlock_nested	1.2830ms	0.4360ms	2.2937 KOps/s	2.0005 KOps/s	$\textbf{\color{#35bf28}+14.66\%}$
test_unlock_stack_nested	67.9921ms	7.3462ms	136.1256 Ops/s	135.6508 Ops/s	$\color{#35bf28}+0.35\%$
test_flatten_speed	0.5160ms	0.1860ms	5.3753 KOps/s	5.2410 KOps/s	$\color{#35bf28}+2.56\%$
test_unflatten_speed	0.4136ms	0.3617ms	2.7649 KOps/s	2.6809 KOps/s	$\color{#35bf28}+3.13\%$
test_common_ops	1.0144ms	0.6037ms	1.6563 KOps/s	1.5813 KOps/s	$\color{#35bf28}+4.75\%$
test_creation	18.8900μs	1.9256μs	519.3206 KOps/s	526.2617 KOps/s	$\color{#d91a1a}-1.32\%$
test_creation_empty	21.3010μs	6.5755μs	152.0803 KOps/s	144.4321 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_creation_nested_1	24.0000μs	8.9291μs	111.9935 KOps/s	104.8845 KOps/s	$\textbf{\color{#35bf28}+6.78\%}$
test_creation_nested_2	30.4200μs	11.4306μs	87.4844 KOps/s	82.6993 KOps/s	$\textbf{\color{#35bf28}+5.79\%}$
test_clone	31.0900μs	14.3546μs	69.6643 KOps/s	70.2266 KOps/s	$\color{#d91a1a}-0.80\%$
test_getitem[int]	56.3310μs	12.1110μs	82.5697 KOps/s	82.0592 KOps/s	$\color{#35bf28}+0.62\%$
test_getitem[slice_int]	49.1710μs	23.7430μs	42.1177 KOps/s	43.2061 KOps/s	$\color{#d91a1a}-2.52\%$
test_getitem[range]	63.8910μs	40.2647μs	24.8356 KOps/s	25.3155 KOps/s	$\color{#d91a1a}-1.90\%$
test_getitem[tuple]	97.9810μs	20.1114μs	49.7231 KOps/s	49.8081 KOps/s	$\color{#d91a1a}-0.17\%$
test_getitem[list]	0.3321ms	37.2574μs	26.8403 KOps/s	27.0289 KOps/s	$\color{#d91a1a}-0.70\%$
test_setitem_dim[int]	51.1710μs	26.5841μs	37.6165 KOps/s	38.2038 KOps/s	$\color{#d91a1a}-1.54\%$
test_setitem_dim[slice_int]	64.4110μs	46.3562μs	21.5721 KOps/s	21.8220 KOps/s	$\color{#d91a1a}-1.15\%$
test_setitem_dim[range]	82.9820μs	63.9276μs	15.6427 KOps/s	15.9622 KOps/s	$\color{#d91a1a}-2.00\%$
test_setitem_dim[tuple]	57.0300μs	40.0798μs	24.9502 KOps/s	25.4376 KOps/s	$\color{#d91a1a}-1.92\%$
test_setitem	0.1348ms	18.2273μs	54.8627 KOps/s	55.9126 KOps/s	$\color{#d91a1a}-1.88\%$
test_set	0.1130ms	17.7122μs	56.4584 KOps/s	57.5634 KOps/s	$\color{#d91a1a}-1.92\%$
test_set_shared	3.2710ms	0.1024ms	9.7676 KOps/s	9.2968 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_update	79.4810μs	21.4295μs	46.6646 KOps/s	43.1235 KOps/s	$\textbf{\color{#35bf28}+8.21\%}$
test_update_nested	0.1344ms	31.1642μs	32.0881 KOps/s	31.8821 KOps/s	$\color{#35bf28}+0.65\%$
test_set_nested	0.1117ms	18.7856μs	53.2323 KOps/s	52.8417 KOps/s	$\color{#35bf28}+0.74\%$
test_set_nested_new	0.1297ms	23.0662μs	43.3536 KOps/s	41.9758 KOps/s	$\color{#35bf28}+3.28\%$
test_select	69.4810μs	45.4799μs	21.9877 KOps/s	20.8797 KOps/s	$\textbf{\color{#35bf28}+5.31\%}$
test_to	74.5910μs	53.2820μs	18.7681 KOps/s	18.7561 KOps/s	$\color{#35bf28}+0.06\%$
test_to_nonblocking	64.8010μs	34.8789μs	28.6706 KOps/s	28.5955 KOps/s	$\color{#35bf28}+0.26\%$
test_unbind_speed	0.3724ms	0.3568ms	2.8027 KOps/s	2.8332 KOps/s	$\color{#d91a1a}-1.08\%$
test_unbind_speed_stack0	66.4712ms	5.2004ms	192.2934 Ops/s	192.3873 Ops/s	$\color{#d91a1a}-0.05\%$
test_unbind_speed_stack1	3.4940μs	0.5216μs	1.9170 MOps/s	1.8938 MOps/s	$\color{#35bf28}+1.23\%$
test_split	53.9143ms	1.8135ms	551.4224 Ops/s	552.2271 Ops/s	$\color{#d91a1a}-0.15\%$
test_chunk	53.4272ms	1.7995ms	555.6988 Ops/s	556.8926 Ops/s	$\color{#d91a1a}-0.21\%$
test_creation[device0]	0.4964ms	0.3079ms	3.2480 KOps/s	3.2562 KOps/s	$\color{#d91a1a}-0.25\%$
test_creation[device1]	0.7881ms	0.3114ms	3.2111 KOps/s	3.2239 KOps/s	$\color{#d91a1a}-0.40\%$
test_creation_from_tensor	0.6377ms	0.3372ms	2.9655 KOps/s	2.7469 KOps/s	$\textbf{\color{#35bf28}+7.96\%}$
test_add_one[memmap_tensor0]	60.3610μs	24.7806μs	40.3541 KOps/s	40.8849 KOps/s	$\color{#d91a1a}-1.30\%$
test_add_one[memmap_tensor1]	0.2126ms	74.9794μs	13.3370 KOps/s	13.4492 KOps/s	$\color{#d91a1a}-0.83\%$
test_contiguous[memmap_tensor0]	22.7800μs	6.1070μs	163.7462 KOps/s	169.7714 KOps/s	$\color{#d91a1a}-3.55\%$
test_contiguous[memmap_tensor1]	47.8400μs	22.7212μs	44.0118 KOps/s	44.3417 KOps/s	$\color{#d91a1a}-0.74\%$
test_stack[memmap_tensor0]	50.0310μs	20.2392μs	49.4090 KOps/s	49.4225 KOps/s	$\color{#d91a1a}-0.03\%$
test_stack[memmap_tensor1]	0.1516ms	74.8658μs	13.3572 KOps/s	13.5224 KOps/s	$\color{#d91a1a}-1.22\%$
test_memmaptd_index	0.2687ms	0.2243ms	4.4578 KOps/s	4.5213 KOps/s	$\color{#d91a1a}-1.40\%$
test_memmaptd_index_astensor	0.3170ms	0.2781ms	3.5953 KOps/s	3.4951 KOps/s	$\color{#35bf28}+2.87\%$
test_memmaptd_index_op	0.5958ms	0.5401ms	1.8515 KOps/s	1.8346 KOps/s	$\color{#35bf28}+0.92\%$
test_reshape_pytree	37.5410μs	20.9622μs	47.7050 KOps/s	47.9799 KOps/s	$\color{#d91a1a}-0.57\%$
test_reshape_td	56.3910μs	30.3328μs	32.9676 KOps/s	32.8760 KOps/s	$\color{#35bf28}+0.28\%$
test_view_pytree	40.1910μs	20.6539μs	48.4170 KOps/s	48.6524 KOps/s	$\color{#d91a1a}-0.48\%$
test_view_td	21.2210μs	4.0757μs	245.3572 KOps/s	244.2347 KOps/s	$\color{#35bf28}+0.46\%$
test_unbind_pytree	44.9410μs	25.9160μs	38.5862 KOps/s	38.0345 KOps/s	$\color{#35bf28}+1.45\%$
test_unbind_td	75.3020μs	55.5101μs	18.0147 KOps/s	17.8492 KOps/s	$\color{#35bf28}+0.93\%$
test_split_pytree	51.6300μs	24.7577μs	40.3915 KOps/s	40.6124 KOps/s	$\color{#d91a1a}-0.54\%$
test_split_td	67.0910μs	45.2003μs	22.1238 KOps/s	22.4545 KOps/s	$\color{#d91a1a}-1.47\%$
test_add_pytree	60.2510μs	32.6394μs	30.6378 KOps/s	30.8746 KOps/s	$\color{#d91a1a}-0.77\%$
test_add_td	66.0310μs	43.1110μs	23.1960 KOps/s	22.7016 KOps/s	$\color{#35bf28}+2.18\%$
test_distributed	23.8800μs	5.4491μs	183.5165 KOps/s	181.1962 KOps/s	$\color{#35bf28}+1.28\%$
test_tdmodule	31.8910μs	16.5140μs	60.5546 KOps/s	59.1035 KOps/s	$\color{#35bf28}+2.46\%$
test_tdmodule_dispatch	0.1308ms	32.0093μs	31.2409 KOps/s	30.6072 KOps/s	$\color{#35bf28}+2.07\%$
test_tdseq	36.3200μs	19.9486μs	50.1288 KOps/s	49.4903 KOps/s	$\color{#35bf28}+1.29\%$
test_tdseq_dispatch	0.1318ms	35.1524μs	28.4475 KOps/s	27.7594 KOps/s	$\color{#35bf28}+2.48\%$
test_instantiation_functorch	2.0295ms	1.6961ms	589.5974 Ops/s	597.7524 Ops/s	$\color{#d91a1a}-1.36\%$
test_instantiation_td	1.6707ms	1.1903ms	840.0944 Ops/s	852.0547 Ops/s	$\color{#d91a1a}-1.40\%$
test_exec_functorch	0.2032ms	0.1628ms	6.1434 KOps/s	6.2042 KOps/s	$\color{#d91a1a}-0.98\%$
test_exec_functional_call	0.2278ms	0.1639ms	6.0996 KOps/s	6.2171 KOps/s	$\color{#d91a1a}-1.89\%$
test_exec_td	0.1965ms	0.1545ms	6.4714 KOps/s	6.5071 KOps/s	$\color{#d91a1a}-0.55\%$
test_exec_td_decorator	0.7872ms	0.2264ms	4.4165 KOps/s	3.9988 KOps/s	$\textbf{\color{#35bf28}+10.44\%}$
test_vmap_mlp_speed[True-True]	1.1511ms	1.0792ms	926.6249 Ops/s	939.5392 Ops/s	$\color{#d91a1a}-1.37\%$
test_vmap_mlp_speed[True-False]	0.8070ms	0.6171ms	1.6204 KOps/s	1.6241 KOps/s	$\color{#d91a1a}-0.22\%$
test_vmap_mlp_speed[False-True]	1.1154ms	1.0402ms	961.3480 Ops/s	1.0270 KOps/s	$\textbf{\color{#d91a1a}-6.39\%}$
test_vmap_mlp_speed[False-False]	0.6150ms	0.5460ms	1.8316 KOps/s	1.8320 KOps/s	$\color{#d91a1a}-0.02\%$
test_vmap_mlp_speed_decorator[True-True]	2.6243ms	1.7956ms	556.9136 Ops/s	482.6189 Ops/s	$\textbf{\color{#35bf28}+15.39\%}$
test_vmap_mlp_speed_decorator[True-False]	1.1706ms	0.6905ms	1.4482 KOps/s	1.4070 KOps/s	$\color{#35bf28}+2.93\%$
test_vmap_mlp_speed_decorator[False-True]	2.1119ms	1.6157ms	618.9403 Ops/s	534.6432 Ops/s	$\textbf{\color{#35bf28}+15.77\%}$
test_vmap_mlp_speed_decorator[False-False]	1.0170ms	0.5879ms	1.7008 KOps/s	1.6464 KOps/s	$\color{#35bf28}+3.30\%$
test_vmap_transformer_speed[True-True]	12.7199ms	12.6224ms	79.2242 Ops/s	79.1637 Ops/s	$\color{#35bf28}+0.08\%$
test_vmap_transformer_speed[True-False]	8.7169ms	8.2975ms	120.5176 Ops/s	120.7391 Ops/s	$\color{#d91a1a}-0.18\%$
test_vmap_transformer_speed[False-True]	12.6758ms	12.5791ms	79.4968 Ops/s	80.1661 Ops/s	$\color{#d91a1a}-0.83\%$
test_vmap_transformer_speed[False-False]	8.3318ms	8.2351ms	121.4308 Ops/s	121.7175 Ops/s	$\color{#d91a1a}-0.24\%$
test_vmap_transformer_speed_decorator[True-True]	44.5421ms	43.0429ms	23.2326 Ops/s	19.0166 Ops/s	$\textbf{\color{#35bf28}+22.17\%}$
test_vmap_transformer_speed_decorator[True-False]	0.1010s	22.0845ms	45.2805 Ops/s	47.8972 Ops/s	$\textbf{\color{#d91a1a}-5.46\%}$
test_vmap_transformer_speed_decorator[False-True]	43.7874ms	42.6218ms	23.4622 Ops/s	19.0503 Ops/s	$\textbf{\color{#35bf28}+23.16\%}$
test_vmap_transformer_speed_decorator[False-False]	0.1002s	21.6044ms	46.2869 Ops/s	48.7028 Ops/s	$\color{#d91a1a}-4.96\%$