[BugFix] Fix h5 auto batch size

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 144. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	39.8840μs	16.9042μs	59.1569 KOps/s	54.9812 KOps/s	$\textbf{\color{#35bf28}+7.59\%}$
test_plain_set_stack_nested	37.3200μs	17.0638μs	58.6036 KOps/s	53.8870 KOps/s	$\textbf{\color{#35bf28}+8.75\%}$
test_plain_set_nested_inplace	57.1960μs	19.2225μs	52.0224 KOps/s	48.0917 KOps/s	$\textbf{\color{#35bf28}+8.17\%}$
test_plain_set_stack_nested_inplace	60.5830μs	19.2570μs	51.9292 KOps/s	48.8966 KOps/s	$\textbf{\color{#35bf28}+6.20\%}$
test_items	17.2420μs	2.4914μs	401.3818 KOps/s	366.4193 KOps/s	$\textbf{\color{#35bf28}+9.54\%}$
test_items_nested	0.4358ms	0.2742ms	3.6471 KOps/s	3.8047 KOps/s	$\color{#d91a1a}-4.14\%$
test_items_nested_locked	1.1363ms	0.2762ms	3.6210 KOps/s	3.7464 KOps/s	$\color{#d91a1a}-3.35\%$
test_items_nested_leaf	0.1629ms	78.7390μs	12.7002 KOps/s	12.6152 KOps/s	$\color{#35bf28}+0.67\%$
test_items_stack_nested	1.7757ms	0.2764ms	3.6174 KOps/s	3.7383 KOps/s	$\color{#d91a1a}-3.24\%$
test_items_stack_nested_leaf	0.1476ms	79.0930μs	12.6433 KOps/s	12.3574 KOps/s	$\color{#35bf28}+2.31\%$
test_items_stack_nested_locked	0.4160ms	0.2737ms	3.6531 KOps/s	3.7767 KOps/s	$\color{#d91a1a}-3.27\%$
test_keys	16.8310μs	3.9968μs	250.1998 KOps/s	259.4504 KOps/s	$\color{#d91a1a}-3.57\%$
test_keys_nested	0.2410ms	0.1398ms	7.1518 KOps/s	7.0534 KOps/s	$\color{#35bf28}+1.39\%$
test_keys_nested_locked	0.6922ms	0.1436ms	6.9629 KOps/s	6.7488 KOps/s	$\color{#35bf28}+3.17\%$
test_keys_nested_leaf	0.2124ms	0.1185ms	8.4391 KOps/s	8.4939 KOps/s	$\color{#d91a1a}-0.65\%$
test_keys_stack_nested	0.2448ms	0.1405ms	7.1198 KOps/s	6.9813 KOps/s	$\color{#35bf28}+1.98\%$
test_keys_stack_nested_leaf	0.2358ms	0.1181ms	8.4661 KOps/s	8.1772 KOps/s	$\color{#35bf28}+3.53\%$
test_keys_stack_nested_locked	0.2114ms	0.1438ms	6.9560 KOps/s	6.7030 KOps/s	$\color{#35bf28}+3.77\%$
test_values	4.6446μs	1.1389μs	878.0284 KOps/s	850.7945 KOps/s	$\color{#35bf28}+3.20\%$
test_values_nested	89.2760μs	50.7111μs	19.7195 KOps/s	18.8829 KOps/s	$\color{#35bf28}+4.43\%$
test_values_nested_locked	88.0740μs	50.9424μs	19.6300 KOps/s	18.9382 KOps/s	$\color{#35bf28}+3.65\%$
test_values_nested_leaf	76.2230μs	45.9874μs	21.7451 KOps/s	20.7698 KOps/s	$\color{#35bf28}+4.70\%$
test_values_stack_nested	87.3230μs	50.9253μs	19.6366 KOps/s	18.7600 KOps/s	$\color{#35bf28}+4.67\%$
test_values_stack_nested_leaf	92.8130μs	45.6432μs	21.9091 KOps/s	21.0965 KOps/s	$\color{#35bf28}+3.85\%$
test_values_stack_nested_locked	0.1053ms	50.8282μs	19.6741 KOps/s	18.4395 KOps/s	$\textbf{\color{#35bf28}+6.70\%}$
test_membership	11.4310μs	1.3400μs	746.2479 KOps/s	742.7522 KOps/s	$\color{#35bf28}+0.47\%$
test_membership_nested	19.2860μs	3.5288μs	283.3828 KOps/s	288.1074 KOps/s	$\color{#d91a1a}-1.64\%$
test_membership_nested_leaf	28.0330μs	3.5635μs	280.6266 KOps/s	286.8593 KOps/s	$\color{#d91a1a}-2.17\%$
test_membership_stacked_nested	20.6180μs	3.4917μs	286.3929 KOps/s	286.6680 KOps/s	$\color{#d91a1a}-0.10\%$
test_membership_stacked_nested_leaf	29.3840μs	3.5519μs	281.5380 KOps/s	288.7939 KOps/s	$\color{#d91a1a}-2.51\%$
test_membership_nested_last	25.9390μs	4.3427μs	230.2691 KOps/s	234.9774 KOps/s	$\color{#d91a1a}-2.00\%$
test_membership_nested_leaf_last	24.8870μs	4.3930μs	227.6366 KOps/s	234.1534 KOps/s	$\color{#d91a1a}-2.78\%$
test_membership_stacked_nested_last	18.1940μs	4.3385μs	230.4954 KOps/s	239.9229 KOps/s	$\color{#d91a1a}-3.93\%$
test_membership_stacked_nested_leaf_last	19.6060μs	4.3733μs	228.6582 KOps/s	235.5802 KOps/s	$\color{#d91a1a}-2.94\%$
test_nested_getleaf	36.1180μs	10.5870μs	94.4557 KOps/s	93.8552 KOps/s	$\color{#35bf28}+0.64\%$
test_nested_get	38.8830μs	10.0031μs	99.9690 KOps/s	101.0254 KOps/s	$\color{#d91a1a}-1.05\%$
test_stacked_getleaf	29.9460μs	10.5405μs	94.8724 KOps/s	95.4387 KOps/s	$\color{#d91a1a}-0.59\%$
test_stacked_get	33.3320μs	10.1562μs	98.4623 KOps/s	101.9049 KOps/s	$\color{#d91a1a}-3.38\%$
test_nested_getitemleaf	32.7710μs	11.2108μs	89.1993 KOps/s	91.4450 KOps/s	$\color{#d91a1a}-2.46\%$
test_nested_getitem	29.2040μs	10.3851μs	96.2920 KOps/s	97.7728 KOps/s	$\color{#d91a1a}-1.51\%$
test_stacked_getitemleaf	33.0410μs	11.1298μs	89.8487 KOps/s	91.3947 KOps/s	$\color{#d91a1a}-1.69\%$
test_stacked_getitem	29.8960μs	10.2874μs	97.2065 KOps/s	98.8913 KOps/s	$\color{#d91a1a}-1.70\%$
test_lock_nested	52.2181ms	0.4099ms	2.4395 KOps/s	2.8178 KOps/s	$\textbf{\color{#d91a1a}-13.43\%}$
test_lock_stack_nested	0.4472ms	0.3164ms	3.1603 KOps/s	3.1435 KOps/s	$\color{#35bf28}+0.53\%$
test_unlock_nested	0.7355ms	0.3540ms	2.8251 KOps/s	2.4072 KOps/s	$\textbf{\color{#35bf28}+17.36\%}$
test_unlock_stack_nested	0.3964ms	0.3255ms	3.0719 KOps/s	3.0804 KOps/s	$\color{#d91a1a}-0.28\%$
test_flatten_speed	0.1805ms	0.1002ms	9.9781 KOps/s	10.2295 KOps/s	$\color{#d91a1a}-2.46\%$
test_unflatten_speed	0.5820ms	0.4149ms	2.4103 KOps/s	2.3663 KOps/s	$\color{#35bf28}+1.86\%$
test_common_ops	4.5301ms	0.7228ms	1.3836 KOps/s	1.3395 KOps/s	$\color{#35bf28}+3.29\%$
test_creation	19.1150μs	1.9481μs	513.3234 KOps/s	508.7629 KOps/s	$\color{#35bf28}+0.90\%$
test_creation_empty	23.2830μs	10.9334μs	91.4631 KOps/s	86.2039 KOps/s	$\textbf{\color{#35bf28}+6.10\%}$
test_creation_nested_1	78.8170μs	13.7863μs	72.5355 KOps/s	67.6625 KOps/s	$\textbf{\color{#35bf28}+7.20\%}$
test_creation_nested_2	41.0870μs	17.1754μs	58.2227 KOps/s	55.3384 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_clone	66.9650μs	13.9756μs	71.5532 KOps/s	71.6028 KOps/s	$\color{#d91a1a}-0.07\%$
test_getitem[int]	39.4040μs	11.8298μs	84.5321 KOps/s	86.2776 KOps/s	$\color{#d91a1a}-2.02\%$
test_getitem[slice_int]	55.7640μs	23.1956μs	43.1116 KOps/s	42.4458 KOps/s	$\color{#35bf28}+1.57\%$
test_getitem[range]	86.8010μs	61.1618μs	16.3501 KOps/s	16.6228 KOps/s	$\color{#d91a1a}-1.64\%$
test_getitem[tuple]	58.5590μs	19.5286μs	51.2070 KOps/s	52.0666 KOps/s	$\color{#d91a1a}-1.65\%$
test_getitem[list]	93.5440μs	42.0249μs	23.7954 KOps/s	24.1733 KOps/s	$\color{#d91a1a}-1.56\%$
test_setitem_dim[int]	67.4460μs	35.6789μs	28.0278 KOps/s	27.4384 KOps/s	$\color{#35bf28}+2.15\%$
test_setitem_dim[slice_int]	0.1123ms	62.3747μs	16.0321 KOps/s	15.6062 KOps/s	$\color{#35bf28}+2.73\%$
test_setitem_dim[range]	0.1303ms	84.5542μs	11.8267 KOps/s	11.7214 KOps/s	$\color{#35bf28}+0.90\%$
test_setitem_dim[tuple]	76.0420μs	51.7563μs	19.3213 KOps/s	19.1554 KOps/s	$\color{#35bf28}+0.87\%$
test_setitem	60.1620μs	21.3618μs	46.8125 KOps/s	47.5460 KOps/s	$\color{#d91a1a}-1.54\%$
test_set	57.3770μs	20.7092μs	48.2878 KOps/s	48.1711 KOps/s	$\color{#35bf28}+0.24\%$
test_set_shared	1.5910ms	0.1400ms	7.1411 KOps/s	7.0319 KOps/s	$\color{#35bf28}+1.55\%$
test_update	0.1251ms	22.8014μs	43.8570 KOps/s	43.0510 KOps/s	$\color{#35bf28}+1.87\%$
test_update_nested	71.8540μs	31.4308μs	31.8159 KOps/s	31.3856 KOps/s	$\color{#35bf28}+1.37\%$
test_update__nested	66.8250μs	26.4980μs	37.7387 KOps/s	38.5816 KOps/s	$\color{#d91a1a}-2.18\%$
test_set_nested	64.9810μs	22.6472μs	44.1556 KOps/s	44.2922 KOps/s	$\color{#d91a1a}-0.31\%$
test_set_nested_new	78.4270μs	27.6934μs	36.1096 KOps/s	37.4120 KOps/s	$\color{#d91a1a}-3.48\%$
test_select	0.1033ms	42.5794μs	23.4855 KOps/s	23.5324 KOps/s	$\color{#d91a1a}-0.20\%$
test_select_nested	0.1299ms	61.6022μs	16.2332 KOps/s	16.4699 KOps/s	$\color{#d91a1a}-1.44\%$
test_exclude_nested	0.2603ms	0.1246ms	8.0282 KOps/s	8.0862 KOps/s	$\color{#d91a1a}-0.72\%$
test_empty[True]	0.7271ms	0.4005ms	2.4969 KOps/s	2.4522 KOps/s	$\color{#35bf28}+1.82\%$
test_empty[False]	7.7445μs	1.2132μs	824.2481 KOps/s	854.1842 KOps/s	$\color{#d91a1a}-3.50\%$
test_unbind_speed	0.3339ms	0.2667ms	3.7497 KOps/s	3.7327 KOps/s	$\color{#35bf28}+0.46\%$
test_unbind_speed_stack0	4.3208ms	0.2649ms	3.7755 KOps/s	3.8542 KOps/s	$\color{#d91a1a}-2.04\%$
test_unbind_speed_stack1	67.3243ms	0.7400ms	1.3514 KOps/s	1.2967 KOps/s	$\color{#35bf28}+4.22\%$
test_split	66.9079ms	1.6449ms	607.9273 Ops/s	612.7234 Ops/s	$\color{#d91a1a}-0.78\%$
test_chunk	67.8922ms	1.6336ms	612.1582 Ops/s	616.2573 Ops/s	$\color{#d91a1a}-0.67\%$
test_creation[device0]	0.1636ms	83.4964μs	11.9766 KOps/s	11.6825 KOps/s	$\color{#35bf28}+2.52\%$
test_creation_from_tensor	3.2344ms	85.4653μs	11.7007 KOps/s	11.5131 KOps/s	$\color{#35bf28}+1.63\%$
test_add_one[memmap_tensor0]	70.5220μs	5.3572μs	186.6630 KOps/s	182.1874 KOps/s	$\color{#35bf28}+2.46\%$
test_contiguous[memmap_tensor0]	10.0380μs	0.6372μs	1.5694 MOps/s	1.5277 MOps/s	$\color{#35bf28}+2.73\%$
test_stack[memmap_tensor0]	17.2720μs	3.6345μs	275.1396 KOps/s	279.1328 KOps/s	$\color{#d91a1a}-1.43\%$
test_memmaptd_index	0.9440ms	0.2565ms	3.8989 KOps/s	3.8838 KOps/s	$\color{#35bf28}+0.39\%$
test_memmaptd_index_astensor	0.7734ms	0.3302ms	3.0287 KOps/s	2.9856 KOps/s	$\color{#35bf28}+1.44\%$
test_memmaptd_index_op	0.8723ms	0.6221ms	1.6073 KOps/s	1.5496 KOps/s	$\color{#35bf28}+3.72\%$
test_serialize_model	0.1861s	0.1166s	8.5750 Ops/s	8.3742 Ops/s	$\color{#35bf28}+2.40\%$
test_serialize_model_pickle	0.4694s	0.3768s	2.6543 Ops/s	2.6380 Ops/s	$\color{#35bf28}+0.62\%$
test_serialize_weights	0.1072s	0.1015s	9.8562 Ops/s	8.6474 Ops/s	$\textbf{\color{#35bf28}+13.98\%}$
test_serialize_weights_returnearly	0.1979s	0.1386s	7.2150 Ops/s	7.7724 Ops/s	$\textbf{\color{#d91a1a}-7.17\%}$
test_serialize_weights_pickle	0.7530s	0.5111s	1.9567 Ops/s	1.5691 Ops/s	$\textbf{\color{#35bf28}+24.70\%}$
test_serialize_weights_filesystem	98.1820ms	92.5309ms	10.8072 Ops/s	10.7730 Ops/s	$\color{#35bf28}+0.32\%$
test_serialize_model_filesystem	0.1581s	0.1019s	9.8123 Ops/s	9.8609 Ops/s	$\color{#d91a1a}-0.49\%$
test_reshape_pytree	76.3630μs	25.9817μs	38.4886 KOps/s	37.7756 KOps/s	$\color{#35bf28}+1.89\%$
test_reshape_td	89.6270μs	35.7596μs	27.9645 KOps/s	29.4512 KOps/s	$\textbf{\color{#d91a1a}-5.05\%}$
test_view_pytree	72.6150μs	25.9395μs	38.5513 KOps/s	37.7384 KOps/s	$\color{#35bf28}+2.15\%$
test_view_td	82.9240μs	40.2937μs	24.8178 KOps/s	26.1042 KOps/s	$\color{#d91a1a}-4.93\%$
test_unbind_pytree	77.3140μs	29.4605μs	33.9438 KOps/s	33.5687 KOps/s	$\color{#35bf28}+1.12\%$
test_unbind_td	0.3982ms	39.1518μs	25.5416 KOps/s	25.9114 KOps/s	$\color{#d91a1a}-1.43\%$
test_split_pytree	94.9070μs	30.1545μs	33.1625 KOps/s	33.1057 KOps/s	$\color{#35bf28}+0.17\%$
test_split_td	0.1312ms	41.8029μs	23.9218 KOps/s	24.3492 KOps/s	$\color{#d91a1a}-1.76\%$
test_add_pytree	0.1316ms	35.6944μs	28.0156 KOps/s	27.6750 KOps/s	$\color{#35bf28}+1.23\%$
test_add_td	0.1298ms	56.3849μs	17.7353 KOps/s	17.7132 KOps/s	$\color{#35bf28}+0.12\%$
test_distributed	0.2193ms	0.1043ms	9.5909 KOps/s	9.7540 KOps/s	$\color{#d91a1a}-1.67\%$
test_tdmodule	50.4140μs	17.4178μs	57.4125 KOps/s	53.0395 KOps/s	$\textbf{\color{#35bf28}+8.24\%}$
test_tdmodule_dispatch	56.5860μs	35.1021μs	28.4884 KOps/s	26.6778 KOps/s	$\textbf{\color{#35bf28}+6.79\%}$
test_tdseq	38.2910μs	20.7489μs	48.1954 KOps/s	45.7738 KOps/s	$\textbf{\color{#35bf28}+5.29\%}$
test_tdseq_dispatch	79.4080μs	40.8601μs	24.4737 KOps/s	23.1413 KOps/s	$\textbf{\color{#35bf28}+5.76\%}$
test_instantiation_functorch	1.6106ms	1.3301ms	751.8091 Ops/s	745.5155 Ops/s	$\color{#35bf28}+0.84\%$
test_instantiation_td	1.4828ms	1.0445ms	957.3548 Ops/s	968.3705 Ops/s	$\color{#d91a1a}-1.14\%$
test_exec_functorch	0.3546ms	0.1611ms	6.2060 KOps/s	6.1904 KOps/s	$\color{#35bf28}+0.25\%$
test_exec_functional_call	0.2322ms	0.1493ms	6.6985 KOps/s	6.6469 KOps/s	$\color{#35bf28}+0.78\%$
test_exec_td	0.2310ms	0.1474ms	6.7863 KOps/s	6.9352 KOps/s	$\color{#d91a1a}-2.15\%$
test_exec_td_decorator	0.3037ms	0.2201ms	4.5441 KOps/s	4.4947 KOps/s	$\color{#35bf28}+1.10\%$
test_vmap_mlp_speed[True-True]	0.7693ms	0.4927ms	2.0297 KOps/s	2.0176 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_mlp_speed[True-False]	0.7294ms	0.4896ms	2.0423 KOps/s	2.0268 KOps/s	$\color{#35bf28}+0.77\%$
test_vmap_mlp_speed[False-True]	0.8265ms	0.3989ms	2.5066 KOps/s	2.5226 KOps/s	$\color{#d91a1a}-0.64\%$
test_vmap_mlp_speed[False-False]	0.6342ms	0.3976ms	2.5151 KOps/s	2.5193 KOps/s	$\color{#d91a1a}-0.17\%$
test_vmap_mlp_speed_decorator[True-True]	0.6645ms	0.5596ms	1.7870 KOps/s	1.7759 KOps/s	$\color{#35bf28}+0.62\%$
test_vmap_mlp_speed_decorator[True-False]	0.8672ms	0.5622ms	1.7786 KOps/s	1.7735 KOps/s	$\color{#35bf28}+0.29\%$
test_vmap_mlp_speed_decorator[False-True]	0.6817ms	0.4602ms	2.1730 KOps/s	2.1513 KOps/s	$\color{#35bf28}+1.01\%$
test_vmap_mlp_speed_decorator[False-False]	0.5785ms	0.4594ms	2.1769 KOps/s	2.1664 KOps/s	$\color{#35bf28}+0.49\%$
test_to_module_speed[True]	1.7909ms	1.7029ms	587.2507 Ops/s	576.9294 Ops/s	$\color{#35bf28}+1.79\%$
test_to_module_speed[False]	1.7391ms	1.6659ms	600.2683 Ops/s	585.6720 Ops/s	$\color{#35bf28}+2.49\%$
test_tc_init	56.0450μs	28.7631μs	34.7667 KOps/s	32.0511 KOps/s	$\textbf{\color{#35bf28}+8.47\%}$
test_tc_init_nested	96.7510μs	61.6728μs	16.2146 KOps/s	15.7168 KOps/s	$\color{#35bf28}+3.17\%$
test_tc_first_layer_tensor	1.6776μs	0.6832μs	1.4638 MOps/s	1.4257 MOps/s	$\color{#35bf28}+2.67\%$
test_tc_first_layer_nontensor	1.9636μs	0.6771μs	1.4768 MOps/s	1.4425 MOps/s	$\color{#35bf28}+2.38\%$
test_tc_second_layer_tensor	14.3670μs	1.8437μs	542.3916 KOps/s	538.7432 KOps/s	$\color{#35bf28}+0.68\%$
test_tc_second_layer_nontensor	8.1953μs	1.5342μs	651.8146 KOps/s	674.5839 KOps/s	$\color{#d91a1a}-3.38\%$
test_unbind	79.1462ms	6.8547ms	145.8848 Ops/s	154.3698 Ops/s	$\textbf{\color{#d91a1a}-5.50\%}$
test_full_like	18.0895ms	10.9475ms	91.3452 Ops/s	83.6799 Ops/s	$\textbf{\color{#35bf28}+9.16\%}$
test_zeros_like	10.3735ms	5.9990ms	166.6958 Ops/s	158.0867 Ops/s	$\textbf{\color{#35bf28}+5.45\%}$
test_ones_like	14.6328ms	6.6459ms	150.4690 Ops/s	158.2326 Ops/s	$\color{#d91a1a}-4.91\%$
test_clone	14.8330ms	8.3505ms	119.7538 Ops/s	122.2998 Ops/s	$\color{#d91a1a}-2.08\%$
test_squeeze	79.6690μs	14.6946μs	68.0520 KOps/s	64.5233 KOps/s	$\textbf{\color{#35bf28}+5.47\%}$
test_unsqueeze	73.3677ms	88.7009μs	11.2738 KOps/s	14.4112 KOps/s	$\textbf{\color{#d91a1a}-21.77\%}$
test_split	0.2185ms	0.1141ms	8.7648 KOps/s	8.6529 KOps/s	$\color{#35bf28}+1.29\%$
test_permute	0.2458ms	0.1407ms	7.1091 KOps/s	7.2022 KOps/s	$\color{#d91a1a}-1.29\%$
test_stack	24.9314ms	24.2079ms	41.3088 Ops/s	41.8391 Ops/s	$\color{#d91a1a}-1.27\%$
test_cat	28.4918ms	24.5079ms	40.8031 Ops/s	41.7958 Ops/s	$\color{#d91a1a}-2.38\%$

[github-actions]

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

Total Benchmarks: 152. 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	0.5087ms	12.9486μs	77.2287 KOps/s	79.0829 KOps/s	$\color{#d91a1a}-2.34\%$
test_plain_set_stack_nested	40.9510μs	13.0513μs	76.6207 KOps/s	78.7070 KOps/s	$\color{#d91a1a}-2.65\%$
test_plain_set_nested_inplace	49.0510μs	14.0674μs	71.0865 KOps/s	71.9462 KOps/s	$\color{#d91a1a}-1.19\%$
test_plain_set_stack_nested_inplace	56.4920μs	14.2086μs	70.3797 KOps/s	71.3893 KOps/s	$\color{#d91a1a}-1.41\%$
test_items	26.2000μs	4.6318μs	215.9006 KOps/s	210.2393 KOps/s	$\color{#35bf28}+2.69\%$
test_items_nested	0.3958ms	0.3433ms	2.9128 KOps/s	2.9061 KOps/s	$\color{#35bf28}+0.23\%$
test_items_nested_locked	0.4223ms	0.3525ms	2.8366 KOps/s	2.8576 KOps/s	$\color{#d91a1a}-0.73\%$
test_items_nested_leaf	0.1041ms	83.1412μs	12.0277 KOps/s	12.0233 KOps/s	$\color{#35bf28}+0.04\%$
test_items_stack_nested	0.4123ms	0.3526ms	2.8361 KOps/s	2.8582 KOps/s	$\color{#d91a1a}-0.77\%$
test_items_stack_nested_leaf	0.1242ms	85.5698μs	11.6864 KOps/s	11.9961 KOps/s	$\color{#d91a1a}-2.58\%$
test_items_stack_nested_locked	0.4142ms	0.3576ms	2.7962 KOps/s	2.8536 KOps/s	$\color{#d91a1a}-2.01\%$
test_keys	18.8000μs	4.4064μs	226.9438 KOps/s	229.6933 KOps/s	$\color{#d91a1a}-1.20\%$
test_keys_nested	97.2220μs	67.2066μs	14.8795 KOps/s	14.9336 KOps/s	$\color{#d91a1a}-0.36\%$
test_keys_nested_locked	0.7832ms	72.2174μs	13.8471 KOps/s	13.7619 KOps/s	$\color{#35bf28}+0.62\%$
test_keys_nested_leaf	91.3320μs	57.4265μs	17.4136 KOps/s	17.3117 KOps/s	$\color{#35bf28}+0.59\%$
test_keys_stack_nested	0.1025ms	67.6897μs	14.7733 KOps/s	15.0243 KOps/s	$\color{#d91a1a}-1.67\%$
test_keys_stack_nested_leaf	90.8620μs	58.0075μs	17.2391 KOps/s	17.2595 KOps/s	$\color{#d91a1a}-0.12\%$
test_keys_stack_nested_locked	97.6810μs	72.6458μs	13.7654 KOps/s	14.1516 KOps/s	$\color{#d91a1a}-2.73\%$
test_values	8.5300μs	1.8157μs	550.7565 KOps/s	554.4631 KOps/s	$\color{#d91a1a}-0.67\%$
test_values_nested	62.3610μs	34.9287μs	28.6298 KOps/s	28.6734 KOps/s	$\color{#d91a1a}-0.15\%$
test_values_nested_locked	60.1310μs	37.1891μs	26.8896 KOps/s	26.9573 KOps/s	$\color{#d91a1a}-0.25\%$
test_values_nested_leaf	51.3410μs	31.0512μs	32.2048 KOps/s	32.1957 KOps/s	$\color{#35bf28}+0.03\%$
test_values_stack_nested	60.2210μs	35.5422μs	28.1356 KOps/s	27.7653 KOps/s	$\color{#35bf28}+1.33\%$
test_values_stack_nested_leaf	60.9910μs	31.6568μs	31.5888 KOps/s	31.3352 KOps/s	$\color{#35bf28}+0.81\%$
test_values_stack_nested_locked	58.0010μs	37.6011μs	26.5950 KOps/s	26.4220 KOps/s	$\color{#35bf28}+0.65\%$
test_membership	2.0050μs	0.7172μs	1.3942 MOps/s	1.4006 MOps/s	$\color{#d91a1a}-0.45\%$
test_membership_nested	29.8700μs	2.5828μs	387.1773 KOps/s	392.3771 KOps/s	$\color{#d91a1a}-1.33\%$
test_membership_nested_leaf	20.6210μs	2.6126μs	382.7536 KOps/s	386.7910 KOps/s	$\color{#d91a1a}-1.04\%$
test_membership_stacked_nested	14.4210μs	2.5997μs	384.6649 KOps/s	389.4211 KOps/s	$\color{#d91a1a}-1.22\%$
test_membership_stacked_nested_leaf	33.6610μs	2.6368μs	379.2514 KOps/s	386.8834 KOps/s	$\color{#d91a1a}-1.97\%$
test_membership_nested_last	21.6890μs	3.1101μs	321.5331 KOps/s	325.3120 KOps/s	$\color{#d91a1a}-1.16\%$
test_membership_nested_leaf_last	34.3100μs	3.1265μs	319.8467 KOps/s	324.5298 KOps/s	$\color{#d91a1a}-1.44\%$
test_membership_stacked_nested_last	23.4010μs	3.5981μs	277.9229 KOps/s	160.0610 KOps/s	$\textbf{\color{#35bf28}+73.64\%}$
test_membership_stacked_nested_leaf_last	34.5200μs	3.5715μs	279.9907 KOps/s	159.4279 KOps/s	$\textbf{\color{#35bf28}+75.62\%}$
test_nested_getleaf	28.1800μs	8.3800μs	119.3320 KOps/s	119.2979 KOps/s	$\color{#35bf28}+0.03\%$
test_nested_get	35.4910μs	7.9052μs	126.4997 KOps/s	127.1482 KOps/s	$\color{#d91a1a}-0.51\%$
test_stacked_getleaf	36.3510μs	8.4285μs	118.6455 KOps/s	119.1471 KOps/s	$\color{#d91a1a}-0.42\%$
test_stacked_get	25.8310μs	7.9241μs	126.1976 KOps/s	126.4059 KOps/s	$\color{#d91a1a}-0.16\%$
test_nested_getitemleaf	33.5100μs	8.5570μs	116.8635 KOps/s	117.0851 KOps/s	$\color{#d91a1a}-0.19\%$
test_nested_getitem	30.0010μs	8.0682μs	123.9434 KOps/s	124.0931 KOps/s	$\color{#d91a1a}-0.12\%$
test_stacked_getitemleaf	32.1210μs	8.6246μs	115.9478 KOps/s	116.1020 KOps/s	$\color{#d91a1a}-0.13\%$
test_stacked_getitem	37.5210μs	8.0493μs	124.2340 KOps/s	123.5406 KOps/s	$\color{#35bf28}+0.56\%$
test_lock_nested	58.6341ms	0.4187ms	2.3885 KOps/s	2.3522 KOps/s	$\color{#35bf28}+1.54\%$
test_lock_stack_nested	0.3490ms	0.3146ms	3.1785 KOps/s	3.1814 KOps/s	$\color{#d91a1a}-0.09\%$
test_unlock_nested	0.7349ms	0.3596ms	2.7812 KOps/s	2.7429 KOps/s	$\color{#35bf28}+1.40\%$
test_unlock_stack_nested	0.3501ms	0.3212ms	3.1133 KOps/s	3.1142 KOps/s	$\color{#d91a1a}-0.03\%$
test_flatten_speed	0.1853ms	0.1021ms	9.7941 KOps/s	9.8931 KOps/s	$\color{#d91a1a}-1.00\%$
test_unflatten_speed	0.3282ms	0.2919ms	3.4261 KOps/s	3.4503 KOps/s	$\color{#d91a1a}-0.70\%$
test_common_ops	1.1758ms	0.5997ms	1.6674 KOps/s	1.7091 KOps/s	$\color{#d91a1a}-2.44\%$
test_creation	33.5500μs	1.6718μs	598.1561 KOps/s	604.8181 KOps/s	$\color{#d91a1a}-1.10\%$
test_creation_empty	39.8010μs	8.7883μs	113.7880 KOps/s	122.0020 KOps/s	$\textbf{\color{#d91a1a}-6.73\%}$
test_creation_nested_1	31.0510μs	10.4932μs	95.2994 KOps/s	98.6446 KOps/s	$\color{#d91a1a}-3.39\%$
test_creation_nested_2	30.9410μs	12.5667μs	79.5752 KOps/s	80.7912 KOps/s	$\color{#d91a1a}-1.51\%$
test_clone	85.5100μs	12.1442μs	82.3441 KOps/s	78.7927 KOps/s	$\color{#35bf28}+4.51\%$
test_getitem[int]	27.4810μs	11.4487μs	87.3460 KOps/s	86.7855 KOps/s	$\color{#35bf28}+0.65\%$
test_getitem[slice_int]	40.9200μs	21.7532μs	45.9703 KOps/s	45.8073 KOps/s	$\color{#35bf28}+0.36\%$
test_getitem[range]	68.1010μs	51.1966μs	19.5325 KOps/s	20.0541 KOps/s	$\color{#d91a1a}-2.60\%$
test_getitem[tuple]	71.9910μs	19.4603μs	51.3867 KOps/s	50.9324 KOps/s	$\color{#35bf28}+0.89\%$
test_getitem[list]	0.1308ms	35.7500μs	27.9720 KOps/s	27.1341 KOps/s	$\color{#35bf28}+3.09\%$
test_setitem_dim[int]	47.5010μs	30.7123μs	32.5602 KOps/s	30.8863 KOps/s	$\textbf{\color{#35bf28}+5.42\%}$
test_setitem_dim[slice_int]	74.0810μs	51.0646μs	19.5831 KOps/s	18.4935 KOps/s	$\textbf{\color{#35bf28}+5.89\%}$
test_setitem_dim[range]	87.5810μs	68.5389μs	14.5903 KOps/s	13.7388 KOps/s	$\textbf{\color{#35bf28}+6.20\%}$
test_setitem_dim[tuple]	87.3020μs	44.9667μs	22.2387 KOps/s	21.0890 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_setitem	46.1000μs	17.5346μs	57.0302 KOps/s	53.4372 KOps/s	$\textbf{\color{#35bf28}+6.72\%}$
test_set	49.7100μs	16.9124μs	59.1281 KOps/s	54.5767 KOps/s	$\textbf{\color{#35bf28}+8.34\%}$
test_set_shared	1.2967ms	0.1003ms	9.9741 KOps/s	10.0812 KOps/s	$\color{#d91a1a}-1.06\%$
test_update	87.1220μs	18.9685μs	52.7190 KOps/s	50.5892 KOps/s	$\color{#35bf28}+4.21\%$
test_update_nested	63.1010μs	24.0689μs	41.5474 KOps/s	40.1110 KOps/s	$\color{#35bf28}+3.58\%$
test_update__nested	57.1210μs	23.3213μs	42.8792 KOps/s	38.4539 KOps/s	$\textbf{\color{#35bf28}+11.51\%}$
test_set_nested	77.6210μs	18.1812μs	55.0020 KOps/s	52.3960 KOps/s	$\color{#35bf28}+4.97\%$
test_set_nested_new	53.0710μs	20.8883μs	47.8737 KOps/s	46.3749 KOps/s	$\color{#35bf28}+3.23\%$
test_select	67.9610μs	33.6312μs	29.7343 KOps/s	28.9451 KOps/s	$\color{#35bf28}+2.73\%$
test_select_nested	0.5384ms	54.6517μs	18.2977 KOps/s	18.5457 KOps/s	$\color{#d91a1a}-1.34\%$
test_exclude_nested	0.1532ms	0.1108ms	9.0272 KOps/s	8.8895 KOps/s	$\color{#35bf28}+1.55\%$
test_empty[True]	0.4183ms	0.3528ms	2.8343 KOps/s	2.8504 KOps/s	$\color{#d91a1a}-0.57\%$
test_empty[False]	3.1251μs	0.9191μs	1.0881 MOps/s	1.0897 MOps/s	$\color{#d91a1a}-0.15\%$
test_to	0.1043ms	76.9391μs	12.9973 KOps/s	12.8627 KOps/s	$\color{#35bf28}+1.05\%$
test_to_nonblocking	0.1008ms	62.3691μs	16.0336 KOps/s	15.5832 KOps/s	$\color{#35bf28}+2.89\%$
test_unbind_speed	0.3202ms	0.2773ms	3.6067 KOps/s	3.5513 KOps/s	$\color{#35bf28}+1.56\%$
test_unbind_speed_stack0	0.3139ms	0.2799ms	3.5730 KOps/s	3.5934 KOps/s	$\color{#d91a1a}-0.57\%$
test_unbind_speed_stack1	75.4955ms	0.8324ms	1.2013 KOps/s	1.2255 KOps/s	$\color{#d91a1a}-1.98\%$
test_split	75.7234ms	1.7017ms	587.6613 Ops/s	579.1127 Ops/s	$\color{#35bf28}+1.48\%$
test_chunk	75.5595ms	1.7068ms	585.8750 Ops/s	576.6418 Ops/s	$\color{#35bf28}+1.60\%$
test_creation[device0]	0.1333ms	62.1641μs	16.0864 KOps/s	15.9254 KOps/s	$\color{#35bf28}+1.01\%$
test_creation_from_tensor	0.1315ms	58.7775μs	17.0133 KOps/s	16.4219 KOps/s	$\color{#35bf28}+3.60\%$
test_add_one[memmap_tensor0]	68.9510μs	7.4886μs	133.5358 KOps/s	131.3234 KOps/s	$\color{#35bf28}+1.68\%$
test_contiguous[memmap_tensor0]	25.0810μs	0.6771μs	1.4768 MOps/s	1.4707 MOps/s	$\color{#35bf28}+0.41\%$
test_stack[memmap_tensor0]	36.3110μs	4.9785μs	200.8648 KOps/s	200.5286 KOps/s	$\color{#35bf28}+0.17\%$
test_memmaptd_index	1.1360ms	0.2984ms	3.3518 KOps/s	3.1444 KOps/s	$\textbf{\color{#35bf28}+6.60\%}$
test_memmaptd_index_astensor	0.7067ms	0.3696ms	2.7059 KOps/s	2.6634 KOps/s	$\color{#35bf28}+1.60\%$
test_memmaptd_index_op	1.1790ms	0.6933ms	1.4423 KOps/s	1.4307 KOps/s	$\color{#35bf28}+0.81\%$
test_serialize_model	0.1827s	0.1114s	8.9793 Ops/s	8.5946 Ops/s	$\color{#35bf28}+4.48\%$
test_serialize_model_pickle	1.3499s	1.2355s	0.8094 Ops/s	0.8084 Ops/s	$\color{#35bf28}+0.11\%$
test_serialize_weights	0.1808s	0.1095s	9.1331 Ops/s	8.7115 Ops/s	$\color{#35bf28}+4.84\%$
test_serialize_weights_returnearly	0.2485s	0.1007s	9.9276 Ops/s	10.2564 Ops/s	$\color{#d91a1a}-3.21\%$
test_serialize_weights_pickle	1.3743s	1.2542s	0.7973 Ops/s	0.7983 Ops/s	$\color{#d91a1a}-0.12\%$
test_reshape_pytree	55.9800μs	26.7315μs	37.4090 KOps/s	37.6085 KOps/s	$\color{#d91a1a}-0.53\%$
test_reshape_td	60.5710μs	32.5192μs	30.7511 KOps/s	30.8583 KOps/s	$\color{#d91a1a}-0.35\%$
test_view_pytree	0.2608ms	26.5862μs	37.6135 KOps/s	38.0519 KOps/s	$\color{#d91a1a}-1.15\%$
test_view_td	60.6910μs	37.4968μs	26.6690 KOps/s	27.1947 KOps/s	$\color{#d91a1a}-1.93\%$
test_unbind_pytree	0.2282ms	33.1357μs	30.1789 KOps/s	30.0098 KOps/s	$\color{#35bf28}+0.56\%$
test_unbind_td	0.4250ms	43.5101μs	22.9831 KOps/s	21.8757 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_split_pytree	70.5020μs	36.9178μs	27.0872 KOps/s	27.7575 KOps/s	$\color{#d91a1a}-2.41\%$
test_split_td	0.4965ms	42.1038μs	23.7508 KOps/s	23.1926 KOps/s	$\color{#35bf28}+2.41\%$
test_add_pytree	0.2393ms	40.0350μs	24.9781 KOps/s	24.9722 KOps/s	$\color{#35bf28}+0.02\%$
test_add_td	83.8910μs	52.6096μs	19.0079 KOps/s	18.4559 KOps/s	$\color{#35bf28}+2.99\%$
test_distributed	1.6277ms	68.7884μs	14.5373 KOps/s	11.4025 KOps/s	$\textbf{\color{#35bf28}+27.49\%}$
test_tdmodule	30.2110μs	14.5790μs	68.5916 KOps/s	67.7339 KOps/s	$\color{#35bf28}+1.27\%$
test_tdmodule_dispatch	44.4420μs	28.6593μs	34.8927 KOps/s	34.1641 KOps/s	$\color{#35bf28}+2.13\%$
test_tdseq	31.5410μs	16.8016μs	59.5182 KOps/s	58.3701 KOps/s	$\color{#35bf28}+1.97\%$
test_tdseq_dispatch	54.2800μs	32.3959μs	30.8681 KOps/s	30.6300 KOps/s	$\color{#35bf28}+0.78\%$
test_instantiation_functorch	1.7459ms	1.5641ms	639.3429 Ops/s	635.7864 Ops/s	$\color{#35bf28}+0.56\%$
test_instantiation_td	1.5808ms	1.0757ms	929.5855 Ops/s	927.7306 Ops/s	$\color{#35bf28}+0.20\%$
test_exec_functorch	0.1996ms	0.1554ms	6.4353 KOps/s	6.4280 KOps/s	$\color{#35bf28}+0.11\%$
test_exec_functional_call	0.2140ms	0.1443ms	6.9299 KOps/s	6.8789 KOps/s	$\color{#35bf28}+0.74\%$
test_exec_td	0.1983ms	0.1460ms	6.8515 KOps/s	6.9176 KOps/s	$\color{#d91a1a}-0.96\%$
test_exec_td_decorator	0.7028ms	0.2178ms	4.5914 KOps/s	4.6914 KOps/s	$\color{#d91a1a}-2.13\%$
test_vmap_mlp_speed[True-True]	0.6956ms	0.6097ms	1.6403 KOps/s	1.6500 KOps/s	$\color{#d91a1a}-0.59\%$
test_vmap_mlp_speed[True-False]	0.6787ms	0.6079ms	1.6451 KOps/s	1.6544 KOps/s	$\color{#d91a1a}-0.56\%$
test_vmap_mlp_speed[False-True]	0.6213ms	0.5615ms	1.7811 KOps/s	1.8373 KOps/s	$\color{#d91a1a}-3.06\%$
test_vmap_mlp_speed[False-False]	0.6296ms	0.5642ms	1.7725 KOps/s	1.8499 KOps/s	$\color{#d91a1a}-4.18\%$
test_vmap_mlp_speed_decorator[True-True]	1.5088ms	0.6746ms	1.4823 KOps/s	1.5043 KOps/s	$\color{#d91a1a}-1.46\%$
test_vmap_mlp_speed_decorator[True-False]	0.7818ms	0.6668ms	1.4998 KOps/s	1.5020 KOps/s	$\color{#d91a1a}-0.15\%$
test_vmap_mlp_speed_decorator[False-True]	0.7152ms	0.5929ms	1.6866 KOps/s	1.6811 KOps/s	$\color{#35bf28}+0.33\%$
test_vmap_mlp_speed_decorator[False-False]	0.7031ms	0.6110ms	1.6366 KOps/s	1.6665 KOps/s	$\color{#d91a1a}-1.79\%$
test_vmap_transformer_speed[True-True]	8.2149ms	8.1024ms	123.4197 Ops/s	122.9363 Ops/s	$\color{#35bf28}+0.39\%$
test_vmap_transformer_speed[True-False]	8.1591ms	8.0841ms	123.7000 Ops/s	122.2583 Ops/s	$\color{#35bf28}+1.18\%$
test_vmap_transformer_speed[False-True]	8.1480ms	8.0387ms	124.3977 Ops/s	124.1601 Ops/s	$\color{#35bf28}+0.19\%$
test_vmap_transformer_speed[False-False]	8.3889ms	8.0407ms	124.3666 Ops/s	124.5204 Ops/s	$\color{#d91a1a}-0.12\%$
test_vmap_transformer_speed_decorator[True-True]	20.5266ms	19.7107ms	50.7339 Ops/s	50.7390 Ops/s	$\color{#d91a1a}-0.01\%$
test_vmap_transformer_speed_decorator[True-False]	19.7376ms	19.6267ms	50.9511 Ops/s	50.9657 Ops/s	$\color{#d91a1a}-0.03\%$
test_vmap_transformer_speed_decorator[False-True]	20.2623ms	19.5328ms	51.1959 Ops/s	51.1612 Ops/s	$\color{#35bf28}+0.07\%$
test_vmap_transformer_speed_decorator[False-False]	20.3291ms	19.6007ms	51.0185 Ops/s	51.3245 Ops/s	$\color{#d91a1a}-0.60\%$
test_to_module_speed[True]	1.6399ms	1.5141ms	660.4553 Ops/s	645.9906 Ops/s	$\color{#35bf28}+2.24\%$
test_to_module_speed[False]	1.6464ms	1.4934ms	669.6290 Ops/s	663.9585 Ops/s	$\color{#35bf28}+0.85\%$
test_tc_init	50.6310μs	24.2625μs	41.2159 KOps/s	41.9176 KOps/s	$\color{#d91a1a}-1.67\%$
test_tc_init_nested	91.6320μs	51.9301μs	19.2566 KOps/s	20.3631 KOps/s	$\textbf{\color{#d91a1a}-5.43\%}$
test_tc_first_layer_tensor	0.8145μs	0.3561μs	2.8084 MOps/s	2.8093 MOps/s	$\color{#d91a1a}-0.03\%$
test_tc_first_layer_nontensor	1.5985μs	0.3859μs	2.5912 MOps/s	2.5761 MOps/s	$\color{#35bf28}+0.59\%$
test_tc_second_layer_tensor	14.9500μs	1.0739μs	931.2265 KOps/s	939.2171 KOps/s	$\color{#d91a1a}-0.85\%$
test_tc_second_layer_nontensor	6.1942μs	0.8260μs	1.2106 MOps/s	1.2322 MOps/s	$\color{#d91a1a}-1.76\%$
test_unbind	0.1017s	8.1962ms	122.0078 Ops/s	123.7158 Ops/s	$\color{#d91a1a}-1.38\%$
test_full_like	13.7560ms	13.2656ms	75.3829 Ops/s	87.6672 Ops/s	$\textbf{\color{#d91a1a}-14.01\%}$
test_zeros_like	96.5304ms	8.2778ms	120.8056 Ops/s	142.1129 Ops/s	$\textbf{\color{#d91a1a}-14.99\%}$
test_ones_like	7.9737ms	7.8280ms	127.7462 Ops/s	126.3143 Ops/s	$\color{#35bf28}+1.13\%$
test_clone	9.7805ms	9.5722ms	104.4687 Ops/s	105.1390 Ops/s	$\color{#d91a1a}-0.64\%$
test_squeeze	63.6610μs	10.9105μs	91.6546 KOps/s	83.6619 KOps/s	$\textbf{\color{#35bf28}+9.55\%}$
test_unsqueeze	0.1143ms	61.5723μs	16.2411 KOps/s	15.8346 KOps/s	$\color{#35bf28}+2.57\%$
test_split	0.1586ms	0.1010ms	9.9031 KOps/s	9.7364 KOps/s	$\color{#35bf28}+1.71\%$
test_permute	0.2050ms	0.1278ms	7.8234 KOps/s	7.9535 KOps/s	$\color{#d91a1a}-1.64\%$
test_stack	27.4730ms	27.2812ms	36.6553 Ops/s	36.1249 Ops/s	$\color{#35bf28}+1.47\%$
test_cat	27.6724ms	27.1845ms	36.7857 Ops/s	36.3662 Ops/s	$\color{#35bf28}+1.15\%$