[Feature] map_names for composite dists

[vmoens]

No description provided.

[vmoens]

See pytorch/rl#2167

[github-actions]

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

Total Benchmarks: 144. Improved: $\large\color{#35bf28}27$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	31.8790μs	16.8143μs	59.4730 KOps/s	54.5038 KOps/s	$\textbf{\color{#35bf28}+9.12\%}$
test_plain_set_stack_nested	68.7680μs	17.1376μs	58.3511 KOps/s	54.3385 KOps/s	$\textbf{\color{#35bf28}+7.38\%}$
test_plain_set_nested_inplace	55.6440μs	19.2848μs	51.8544 KOps/s	48.5122 KOps/s	$\textbf{\color{#35bf28}+6.89\%}$
test_plain_set_stack_nested_inplace	77.4050μs	19.2770μs	51.8753 KOps/s	49.0593 KOps/s	$\textbf{\color{#35bf28}+5.74\%}$
test_items	27.0510μs	2.5107μs	398.2896 KOps/s	390.5026 KOps/s	$\color{#35bf28}+1.99\%$
test_items_nested	1.3040ms	0.2771ms	3.6087 KOps/s	3.6974 KOps/s	$\color{#d91a1a}-2.40\%$
test_items_nested_locked	0.4595ms	0.2849ms	3.5105 KOps/s	3.6689 KOps/s	$\color{#d91a1a}-4.32\%$
test_items_nested_leaf	0.1374ms	77.2897μs	12.9383 KOps/s	12.6904 KOps/s	$\color{#35bf28}+1.95\%$
test_items_stack_nested	0.3980ms	0.2783ms	3.5932 KOps/s	3.7058 KOps/s	$\color{#d91a1a}-3.04\%$
test_items_stack_nested_leaf	0.1383ms	78.0929μs	12.8053 KOps/s	12.3716 KOps/s	$\color{#35bf28}+3.51\%$
test_items_stack_nested_locked	0.4654ms	0.2767ms	3.6136 KOps/s	3.7140 KOps/s	$\color{#d91a1a}-2.70\%$
test_keys	30.6870μs	3.7674μs	265.4377 KOps/s	254.8918 KOps/s	$\color{#35bf28}+4.14\%$
test_keys_nested	0.2292ms	0.1369ms	7.3034 KOps/s	6.9295 KOps/s	$\textbf{\color{#35bf28}+5.40\%}$
test_keys_nested_locked	0.7355ms	0.1421ms	7.0395 KOps/s	6.7785 KOps/s	$\color{#35bf28}+3.85\%$
test_keys_nested_leaf	0.2119ms	0.1170ms	8.5492 KOps/s	8.5187 KOps/s	$\color{#35bf28}+0.36\%$
test_keys_stack_nested	0.2300ms	0.1373ms	7.2840 KOps/s	7.0484 KOps/s	$\color{#35bf28}+3.34\%$
test_keys_stack_nested_leaf	0.1990ms	0.1160ms	8.6201 KOps/s	8.3646 KOps/s	$\color{#35bf28}+3.05\%$
test_keys_stack_nested_locked	0.2528ms	0.1420ms	7.0432 KOps/s	6.8531 KOps/s	$\color{#35bf28}+2.77\%$
test_values	11.3788μs	1.1789μs	848.2213 KOps/s	816.7362 KOps/s	$\color{#35bf28}+3.85\%$
test_values_nested	0.1149ms	50.9994μs	19.6081 KOps/s	18.8143 KOps/s	$\color{#35bf28}+4.22\%$
test_values_nested_locked	0.1069ms	50.6815μs	19.7311 KOps/s	19.2342 KOps/s	$\color{#35bf28}+2.58\%$
test_values_nested_leaf	0.1021ms	46.2830μs	21.6062 KOps/s	20.8158 KOps/s	$\color{#35bf28}+3.80\%$
test_values_stack_nested	0.1040ms	51.6593μs	19.3576 KOps/s	18.5309 KOps/s	$\color{#35bf28}+4.46\%$
test_values_stack_nested_leaf	98.5240μs	45.4683μs	21.9933 KOps/s	20.9040 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_values_stack_nested_locked	98.6450μs	51.4971μs	19.4186 KOps/s	18.7215 KOps/s	$\color{#35bf28}+3.72\%$
test_membership	22.1920μs	1.3513μs	740.0370 KOps/s	736.8167 KOps/s	$\color{#35bf28}+0.44\%$
test_membership_nested	45.3650μs	3.4738μs	287.8708 KOps/s	293.5836 KOps/s	$\color{#d91a1a}-1.95\%$
test_membership_nested_leaf	38.9130μs	3.4506μs	289.8044 KOps/s	292.7973 KOps/s	$\color{#d91a1a}-1.02\%$
test_membership_stacked_nested	18.1940μs	3.4686μs	288.3022 KOps/s	294.8410 KOps/s	$\color{#d91a1a}-2.22\%$
test_membership_stacked_nested_leaf	38.4120μs	3.4627μs	288.7948 KOps/s	291.9745 KOps/s	$\color{#d91a1a}-1.09\%$
test_membership_nested_last	33.6830μs	4.1863μs	238.8724 KOps/s	238.0334 KOps/s	$\color{#35bf28}+0.35\%$
test_membership_nested_leaf_last	40.3260μs	4.2349μs	236.1308 KOps/s	240.6240 KOps/s	$\color{#d91a1a}-1.87\%$
test_membership_stacked_nested_last	45.7850μs	5.2262μs	191.3454 KOps/s	236.2173 KOps/s	$\textbf{\color{#d91a1a}-19.00\%}$
test_membership_stacked_nested_leaf_last	48.9210μs	5.3059μs	188.4704 KOps/s	236.9515 KOps/s	$\textbf{\color{#d91a1a}-20.46\%}$
test_nested_getleaf	0.1985ms	10.6628μs	93.7841 KOps/s	93.2969 KOps/s	$\color{#35bf28}+0.52\%$
test_nested_get	44.7040μs	10.0860μs	99.1476 KOps/s	100.8561 KOps/s	$\color{#d91a1a}-1.69\%$
test_stacked_getleaf	54.5820μs	10.4087μs	96.0737 KOps/s	96.1903 KOps/s	$\color{#d91a1a}-0.12\%$
test_stacked_get	49.8130μs	9.8976μs	101.0342 KOps/s	100.4825 KOps/s	$\color{#35bf28}+0.55\%$
test_nested_getitemleaf	39.9540μs	11.0661μs	90.3659 KOps/s	90.7825 KOps/s	$\color{#d91a1a}-0.46\%$
test_nested_getitem	56.1050μs	10.2950μs	97.1345 KOps/s	96.2044 KOps/s	$\color{#35bf28}+0.97\%$
test_stacked_getitemleaf	0.1277ms	11.0152μs	90.7836 KOps/s	90.9248 KOps/s	$\color{#d91a1a}-0.16\%$
test_stacked_getitem	98.3040μs	10.2170μs	97.8756 KOps/s	98.9146 KOps/s	$\color{#d91a1a}-1.05\%$
test_lock_nested	4.3913ms	0.3580ms	2.7935 KOps/s	2.7848 KOps/s	$\color{#35bf28}+0.31\%$
test_lock_stack_nested	0.5439ms	0.3137ms	3.1877 KOps/s	3.1417 KOps/s	$\color{#35bf28}+1.46\%$
test_unlock_nested	0.8870ms	0.3626ms	2.7579 KOps/s	2.4854 KOps/s	$\textbf{\color{#35bf28}+10.96\%}$
test_unlock_stack_nested	0.6547ms	0.3202ms	3.1233 KOps/s	3.0940 KOps/s	$\color{#35bf28}+0.95\%$
test_flatten_speed	0.5332ms	95.8344μs	10.4347 KOps/s	10.3504 KOps/s	$\color{#35bf28}+0.81\%$
test_unflatten_speed	0.8861ms	0.4113ms	2.4316 KOps/s	2.4065 KOps/s	$\color{#35bf28}+1.04\%$
test_common_ops	3.4339ms	0.7129ms	1.4027 KOps/s	1.3414 KOps/s	$\color{#35bf28}+4.57\%$
test_creation	54.7420μs	1.8878μs	529.7113 KOps/s	520.4356 KOps/s	$\color{#35bf28}+1.78\%$
test_creation_empty	45.9560μs	10.7946μs	92.6387 KOps/s	81.2010 KOps/s	$\textbf{\color{#35bf28}+14.09\%}$
test_creation_nested_1	37.2790μs	13.6453μs	73.2852 KOps/s	67.1438 KOps/s	$\textbf{\color{#35bf28}+9.15\%}$
test_creation_nested_2	75.2010μs	16.7778μs	59.6027 KOps/s	54.9433 KOps/s	$\textbf{\color{#35bf28}+8.48\%}$
test_clone	89.0670μs	13.6967μs	73.0105 KOps/s	73.9217 KOps/s	$\color{#d91a1a}-1.23\%$
test_getitem[int]	39.2940μs	11.5677μs	86.4476 KOps/s	87.7734 KOps/s	$\color{#d91a1a}-1.51\%$
test_getitem[slice_int]	4.4507ms	22.7964μs	43.8666 KOps/s	43.8757 KOps/s	$\color{#d91a1a}-0.02\%$
test_getitem[range]	81.3620μs	59.6294μs	16.7703 KOps/s	17.1147 KOps/s	$\color{#d91a1a}-2.01\%$
test_getitem[tuple]	80.3200μs	19.0195μs	52.5778 KOps/s	52.4934 KOps/s	$\color{#35bf28}+0.16\%$
test_getitem[list]	0.1244ms	40.5087μs	24.6860 KOps/s	23.8801 KOps/s	$\color{#35bf28}+3.37\%$
test_setitem_dim[int]	63.3790μs	34.3079μs	29.1478 KOps/s	26.8431 KOps/s	$\textbf{\color{#35bf28}+8.59\%}$
test_setitem_dim[slice_int]	0.1052ms	59.6781μs	16.7566 KOps/s	15.7056 KOps/s	$\textbf{\color{#35bf28}+6.69\%}$
test_setitem_dim[range]	0.1177ms	82.4164μs	12.1335 KOps/s	11.2161 KOps/s	$\textbf{\color{#35bf28}+8.18\%}$
test_setitem_dim[tuple]	86.7830μs	48.4529μs	20.6386 KOps/s	18.8844 KOps/s	$\textbf{\color{#35bf28}+9.29\%}$
test_setitem	61.2550μs	20.4973μs	48.7868 KOps/s	47.0906 KOps/s	$\color{#35bf28}+3.60\%$
test_set	80.1210μs	20.1872μs	49.5363 KOps/s	47.3562 KOps/s	$\color{#35bf28}+4.60\%$
test_set_shared	1.6975ms	0.1376ms	7.2671 KOps/s	6.9242 KOps/s	$\color{#35bf28}+4.95\%$
test_update	0.1074ms	22.6861μs	44.0800 KOps/s	41.7420 KOps/s	$\textbf{\color{#35bf28}+5.60\%}$
test_update_nested	99.3890μs	30.7050μs	32.5680 KOps/s	31.1777 KOps/s	$\color{#35bf28}+4.46\%$
test_update__nested	98.0130μs	25.7736μs	38.7993 KOps/s	37.6062 KOps/s	$\color{#35bf28}+3.17\%$
test_set_nested	97.0320μs	22.1365μs	45.1742 KOps/s	43.9610 KOps/s	$\color{#35bf28}+2.76\%$
test_set_nested_new	79.9900μs	26.5534μs	37.6600 KOps/s	37.1647 KOps/s	$\color{#35bf28}+1.33\%$
test_select	0.1062ms	41.6610μs	24.0033 KOps/s	23.2255 KOps/s	$\color{#35bf28}+3.35\%$
test_select_nested	0.9942ms	61.6164μs	16.2294 KOps/s	16.3196 KOps/s	$\color{#d91a1a}-0.55\%$
test_exclude_nested	0.2409ms	0.1213ms	8.2443 KOps/s	8.1900 KOps/s	$\color{#35bf28}+0.66\%$
test_empty[True]	0.6337ms	0.4019ms	2.4883 KOps/s	2.4593 KOps/s	$\color{#35bf28}+1.18\%$
test_empty[False]	12.1428μs	1.1887μs	841.2842 KOps/s	854.3987 KOps/s	$\color{#d91a1a}-1.53\%$
test_unbind_speed	4.3537ms	0.2798ms	3.5745 KOps/s	3.7726 KOps/s	$\textbf{\color{#d91a1a}-5.25\%}$
test_unbind_speed_stack0	0.4148ms	0.2518ms	3.9719 KOps/s	3.8854 KOps/s	$\color{#35bf28}+2.23\%$
test_unbind_speed_stack1	68.1481ms	0.7247ms	1.3798 KOps/s	1.2599 KOps/s	$\textbf{\color{#35bf28}+9.51\%}$
test_split	67.1730ms	1.6300ms	613.4966 Ops/s	618.9558 Ops/s	$\color{#d91a1a}-0.88\%$
test_chunk	68.8898ms	1.6262ms	614.9385 Ops/s	662.2592 Ops/s	$\textbf{\color{#d91a1a}-7.15\%}$
test_creation[device0]	0.2046ms	83.5820μs	11.9643 KOps/s	11.5201 KOps/s	$\color{#35bf28}+3.86\%$
test_creation_from_tensor	0.2383ms	84.5034μs	11.8338 KOps/s	11.4785 KOps/s	$\color{#35bf28}+3.10\%$
test_add_one[memmap_tensor0]	66.7050μs	5.5010μs	181.7851 KOps/s	179.6485 KOps/s	$\color{#35bf28}+1.19\%$
test_contiguous[memmap_tensor0]	11.7120μs	0.6296μs	1.5883 MOps/s	1.5185 MOps/s	$\color{#35bf28}+4.60\%$
test_stack[memmap_tensor0]	14.4580μs	3.6599μs	273.2344 KOps/s	279.2174 KOps/s	$\color{#d91a1a}-2.14\%$
test_memmaptd_index	1.0975ms	0.2706ms	3.6955 KOps/s	3.9130 KOps/s	$\textbf{\color{#d91a1a}-5.56\%}$
test_memmaptd_index_astensor	1.1360ms	0.3491ms	2.8645 KOps/s	2.9853 KOps/s	$\color{#d91a1a}-4.04\%$
test_memmaptd_index_op	0.9943ms	0.6297ms	1.5880 KOps/s	1.4105 KOps/s	$\textbf{\color{#35bf28}+12.59\%}$
test_serialize_model	0.1838s	0.1136s	8.7998 Ops/s	8.2573 Ops/s	$\textbf{\color{#35bf28}+6.57\%}$
test_serialize_model_pickle	0.4516s	0.3804s	2.6291 Ops/s	2.5736 Ops/s	$\color{#35bf28}+2.16\%$
test_serialize_weights	0.1118s	0.1050s	9.5236 Ops/s	8.5132 Ops/s	$\textbf{\color{#35bf28}+11.87\%}$
test_serialize_weights_returnearly	0.2002s	0.1401s	7.1369 Ops/s	7.7649 Ops/s	$\textbf{\color{#d91a1a}-8.09\%}$
test_serialize_weights_pickle	1.0436s	0.6031s	1.6580 Ops/s	1.3937 Ops/s	$\textbf{\color{#35bf28}+18.96\%}$
test_serialize_weights_filesystem	0.1031s	94.3751ms	10.5960 Ops/s	9.3781 Ops/s	$\textbf{\color{#35bf28}+12.99\%}$
test_serialize_model_filesystem	0.1087s	97.9012ms	10.2144 Ops/s	9.4168 Ops/s	$\textbf{\color{#35bf28}+8.47\%}$
test_reshape_pytree	65.2220μs	25.3078μs	39.5135 KOps/s	38.1383 KOps/s	$\color{#35bf28}+3.61\%$
test_reshape_td	86.0410μs	34.6651μs	28.8475 KOps/s	28.9214 KOps/s	$\color{#d91a1a}-0.26\%$
test_view_pytree	68.7890μs	25.4230μs	39.3345 KOps/s	38.2858 KOps/s	$\color{#35bf28}+2.74\%$
test_view_td	82.5750μs	38.3329μs	26.0873 KOps/s	25.7733 KOps/s	$\color{#35bf28}+1.22\%$
test_unbind_pytree	62.9580μs	29.4388μs	33.9688 KOps/s	33.0548 KOps/s	$\color{#35bf28}+2.76\%$
test_unbind_td	0.4295ms	38.1801μs	26.1916 KOps/s	26.2425 KOps/s	$\color{#d91a1a}-0.19\%$
test_split_pytree	64.7210μs	29.1272μs	34.3321 KOps/s	33.3794 KOps/s	$\color{#35bf28}+2.85\%$
test_split_td	0.1240ms	41.5013μs	24.0957 KOps/s	24.3786 KOps/s	$\color{#d91a1a}-1.16\%$
test_add_pytree	0.1179ms	35.4673μs	28.1950 KOps/s	28.3758 KOps/s	$\color{#d91a1a}-0.64\%$
test_add_td	0.1577ms	56.1692μs	17.8033 KOps/s	17.2186 KOps/s	$\color{#35bf28}+3.40\%$
test_distributed	0.2523ms	99.7369μs	10.0264 KOps/s	9.7087 KOps/s	$\color{#35bf28}+3.27\%$
test_tdmodule	71.9050μs	17.3345μs	57.6883 KOps/s	55.4900 KOps/s	$\color{#35bf28}+3.96\%$
test_tdmodule_dispatch	62.7780μs	34.7805μs	28.7517 KOps/s	27.6584 KOps/s	$\color{#35bf28}+3.95\%$
test_tdseq	36.3980μs	20.1896μs	49.5304 KOps/s	47.5855 KOps/s	$\color{#35bf28}+4.09\%$
test_tdseq_dispatch	99.0260μs	45.6168μs	21.9217 KOps/s	23.5201 KOps/s	$\textbf{\color{#d91a1a}-6.80\%}$
test_instantiation_functorch	1.6018ms	1.2920ms	773.9830 Ops/s	739.4889 Ops/s	$\color{#35bf28}+4.66\%$
test_instantiation_td	73.6967ms	1.0792ms	926.5787 Ops/s	941.7299 Ops/s	$\color{#d91a1a}-1.61\%$
test_exec_functorch	0.2515ms	0.1598ms	6.2584 KOps/s	6.1696 KOps/s	$\color{#35bf28}+1.44\%$
test_exec_functional_call	0.2696ms	0.1468ms	6.8110 KOps/s	6.5915 KOps/s	$\color{#35bf28}+3.33\%$
test_exec_td	0.4173ms	0.1501ms	6.6633 KOps/s	6.6702 KOps/s	$\color{#d91a1a}-0.10\%$
test_exec_td_decorator	0.3292ms	0.2228ms	4.4883 KOps/s	4.4243 KOps/s	$\color{#35bf28}+1.45\%$
test_vmap_mlp_speed[True-True]	0.9465ms	0.4825ms	2.0725 KOps/s	2.0203 KOps/s	$\color{#35bf28}+2.59\%$
test_vmap_mlp_speed[True-False]	0.7988ms	0.4818ms	2.0757 KOps/s	2.0408 KOps/s	$\color{#35bf28}+1.71\%$
test_vmap_mlp_speed[False-True]	0.6355ms	0.3936ms	2.5409 KOps/s	2.4924 KOps/s	$\color{#35bf28}+1.95\%$
test_vmap_mlp_speed[False-False]	0.9035ms	0.3938ms	2.5391 KOps/s	2.4742 KOps/s	$\color{#35bf28}+2.62\%$
test_vmap_mlp_speed_decorator[True-True]	1.1999ms	0.5596ms	1.7870 KOps/s	1.7772 KOps/s	$\color{#35bf28}+0.55\%$
test_vmap_mlp_speed_decorator[True-False]	0.9312ms	0.5534ms	1.8071 KOps/s	1.7871 KOps/s	$\color{#35bf28}+1.12\%$
test_vmap_mlp_speed_decorator[False-True]	0.8706ms	0.4543ms	2.2010 KOps/s	2.1392 KOps/s	$\color{#35bf28}+2.89\%$
test_vmap_mlp_speed_decorator[False-False]	0.8832ms	0.4556ms	2.1950 KOps/s	2.1648 KOps/s	$\color{#35bf28}+1.39\%$
test_to_module_speed[True]	2.6282ms	1.7163ms	582.6468 Ops/s	578.1126 Ops/s	$\color{#35bf28}+0.78\%$
test_to_module_speed[False]	1.9282ms	1.6804ms	595.1134 Ops/s	589.7945 Ops/s	$\color{#35bf28}+0.90\%$
test_tc_init	64.4010μs	28.9753μs	34.5121 KOps/s	30.9916 KOps/s	$\textbf{\color{#35bf28}+11.36\%}$
test_tc_init_nested	0.1302ms	55.9244μs	17.8813 KOps/s	14.9857 KOps/s	$\textbf{\color{#35bf28}+19.32\%}$
test_tc_first_layer_tensor	3.9289μs	0.7053μs	1.4179 MOps/s	1.4104 MOps/s	$\color{#35bf28}+0.53\%$
test_tc_first_layer_nontensor	2.1010μs	0.6871μs	1.4554 MOps/s	1.4218 MOps/s	$\color{#35bf28}+2.37\%$
test_tc_second_layer_tensor	34.9460μs	1.8659μs	535.9487 KOps/s	531.0771 KOps/s	$\color{#35bf28}+0.92\%$
test_tc_second_layer_nontensor	26.3400μs	1.6498μs	606.1465 KOps/s	612.5426 KOps/s	$\color{#d91a1a}-1.04\%$
test_unbind	5.3361ms	5.1612ms	193.7517 Ops/s	121.2662 Ops/s	$\textbf{\color{#35bf28}+59.77\%}$
test_full_like	18.5613ms	11.3929ms	87.7738 Ops/s	84.0683 Ops/s	$\color{#35bf28}+4.41\%$
test_zeros_like	12.3630ms	5.3702ms	186.2143 Ops/s	168.6751 Ops/s	$\textbf{\color{#35bf28}+10.40\%}$
test_ones_like	11.5464ms	6.2960ms	158.8319 Ops/s	154.3663 Ops/s	$\color{#35bf28}+2.89\%$
test_clone	16.0743ms	7.8525ms	127.3473 Ops/s	122.2659 Ops/s	$\color{#35bf28}+4.16\%$
test_squeeze	56.2550μs	13.5702μs	73.6911 KOps/s	67.1102 KOps/s	$\textbf{\color{#35bf28}+9.81\%}$
test_unsqueeze	0.1100ms	59.4325μs	16.8258 KOps/s	16.4663 KOps/s	$\color{#35bf28}+2.18\%$
test_split	0.2297ms	0.1128ms	8.8651 KOps/s	8.9393 KOps/s	$\color{#d91a1a}-0.83\%$
test_permute	0.2755ms	0.1284ms	7.7910 KOps/s	7.7198 KOps/s	$\color{#35bf28}+0.92\%$
test_stack	29.8896ms	22.5711ms	44.3044 Ops/s	42.4738 Ops/s	$\color{#35bf28}+4.31\%$
test_cat	30.0099ms	22.5345ms	44.3763 Ops/s	42.5330 Ops/s	$\color{#35bf28}+4.33\%$

[github-actions]

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

Total Benchmarks: 152. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}21$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	34.0520μs	12.3184μs	81.1792 KOps/s	78.0187 KOps/s	$\color{#35bf28}+4.05\%$
test_plain_set_stack_nested	28.0520μs	12.4506μs	80.3173 KOps/s	77.0365 KOps/s	$\color{#35bf28}+4.26\%$
test_plain_set_nested_inplace	47.6530μs	13.6811μs	73.0937 KOps/s	71.1850 KOps/s	$\color{#35bf28}+2.68\%$
test_plain_set_stack_nested_inplace	38.6420μs	13.8223μs	72.3471 KOps/s	70.4414 KOps/s	$\color{#35bf28}+2.71\%$
test_items	15.8210μs	4.6759μs	213.8615 KOps/s	213.7449 KOps/s	$\color{#35bf28}+0.05\%$
test_items_nested	0.3952ms	0.3412ms	2.9305 KOps/s	2.9049 KOps/s	$\color{#35bf28}+0.88\%$
test_items_nested_locked	0.3994ms	0.3440ms	2.9067 KOps/s	2.9430 KOps/s	$\color{#d91a1a}-1.24\%$
test_items_nested_leaf	0.1241ms	83.5803μs	11.9645 KOps/s	12.1369 KOps/s	$\color{#d91a1a}-1.42\%$
test_items_stack_nested	0.4123ms	0.3432ms	2.9135 KOps/s	2.9398 KOps/s	$\color{#d91a1a}-0.89\%$
test_items_stack_nested_leaf	0.1204ms	84.2533μs	11.8690 KOps/s	11.8775 KOps/s	$\color{#d91a1a}-0.07\%$
test_items_stack_nested_locked	0.4118ms	0.3452ms	2.8970 KOps/s	2.9148 KOps/s	$\color{#d91a1a}-0.61\%$
test_keys	25.6610μs	4.3573μs	229.5012 KOps/s	228.0654 KOps/s	$\color{#35bf28}+0.63\%$
test_keys_nested	0.1088ms	66.6790μs	14.9972 KOps/s	14.7689 KOps/s	$\color{#35bf28}+1.55\%$
test_keys_nested_locked	0.8042ms	71.3981μs	14.0060 KOps/s	13.6880 KOps/s	$\color{#35bf28}+2.32\%$
test_keys_nested_leaf	0.1025ms	57.3176μs	17.4467 KOps/s	17.1631 KOps/s	$\color{#35bf28}+1.65\%$
test_keys_stack_nested	95.6750μs	66.4909μs	15.0397 KOps/s	14.7547 KOps/s	$\color{#35bf28}+1.93\%$
test_keys_stack_nested_leaf	89.8960μs	57.5910μs	17.3638 KOps/s	17.1471 KOps/s	$\color{#35bf28}+1.26\%$
test_keys_stack_nested_locked	0.1022ms	71.8033μs	13.9269 KOps/s	13.8448 KOps/s	$\color{#35bf28}+0.59\%$
test_values	10.6140μs	1.8095μs	552.6387 KOps/s	548.4384 KOps/s	$\color{#35bf28}+0.77\%$
test_values_nested	60.5740μs	34.8783μs	28.6711 KOps/s	28.4959 KOps/s	$\color{#35bf28}+0.62\%$
test_values_nested_locked	63.3840μs	36.4856μs	27.4080 KOps/s	27.0952 KOps/s	$\color{#35bf28}+1.15\%$
test_values_nested_leaf	54.3730μs	31.0269μs	32.2301 KOps/s	32.1136 KOps/s	$\color{#35bf28}+0.36\%$
test_values_stack_nested	56.3030μs	35.8214μs	27.9163 KOps/s	27.9782 KOps/s	$\color{#d91a1a}-0.22\%$
test_values_stack_nested_leaf	59.7240μs	31.9918μs	31.2580 KOps/s	31.3035 KOps/s	$\color{#d91a1a}-0.15\%$
test_values_stack_nested_locked	60.5040μs	37.4290μs	26.7172 KOps/s	26.7109 KOps/s	$\color{#35bf28}+0.02\%$
test_membership	4.4874μs	0.7273μs	1.3749 MOps/s	1.1384 MOps/s	$\textbf{\color{#35bf28}+20.77\%}$
test_membership_nested	17.1110μs	2.5825μs	387.2266 KOps/s	392.6248 KOps/s	$\color{#d91a1a}-1.37\%$
test_membership_nested_leaf	17.5610μs	2.5824μs	387.2325 KOps/s	394.3887 KOps/s	$\color{#d91a1a}-1.81\%$
test_membership_stacked_nested	20.8410μs	2.6048μs	383.9034 KOps/s	388.2634 KOps/s	$\color{#d91a1a}-1.12\%$
test_membership_stacked_nested_leaf	33.2020μs	2.5847μs	386.8993 KOps/s	387.8894 KOps/s	$\color{#d91a1a}-0.26\%$
test_membership_nested_last	23.7920μs	3.0900μs	323.6214 KOps/s	324.5826 KOps/s	$\color{#d91a1a}-0.30\%$
test_membership_nested_leaf_last	34.0720μs	3.0968μs	322.9163 KOps/s	325.3581 KOps/s	$\color{#d91a1a}-0.75\%$
test_membership_stacked_nested_last	20.5910μs	3.1362μs	318.8550 KOps/s	256.6773 KOps/s	$\textbf{\color{#35bf28}+24.22\%}$
test_membership_stacked_nested_leaf_last	18.1910μs	3.0731μs	325.4050 KOps/s	257.1147 KOps/s	$\textbf{\color{#35bf28}+26.56\%}$
test_nested_getleaf	40.3530μs	8.5848μs	116.4843 KOps/s	119.5785 KOps/s	$\color{#d91a1a}-2.59\%$
test_nested_get	34.9220μs	8.0458μs	124.2887 KOps/s	127.2979 KOps/s	$\color{#d91a1a}-2.36\%$
test_stacked_getleaf	22.5710μs	8.5830μs	116.5097 KOps/s	118.4745 KOps/s	$\color{#d91a1a}-1.66\%$
test_stacked_get	37.3120μs	8.0990μs	123.4720 KOps/s	126.6096 KOps/s	$\color{#d91a1a}-2.48\%$
test_nested_getitemleaf	35.1920μs	8.7095μs	114.8167 KOps/s	117.0385 KOps/s	$\color{#d91a1a}-1.90\%$
test_nested_getitem	21.9210μs	8.2084μs	121.8271 KOps/s	124.7481 KOps/s	$\color{#d91a1a}-2.34\%$
test_stacked_getitemleaf	48.6930μs	8.7127μs	114.7753 KOps/s	115.6728 KOps/s	$\color{#d91a1a}-0.78\%$
test_stacked_getitem	44.7330μs	8.2289μs	121.5235 KOps/s	123.4967 KOps/s	$\color{#d91a1a}-1.60\%$
test_lock_nested	57.1989ms	0.4181ms	2.3918 KOps/s	2.3952 KOps/s	$\color{#d91a1a}-0.14\%$
test_lock_stack_nested	0.3662ms	0.3136ms	3.1885 KOps/s	3.2564 KOps/s	$\color{#d91a1a}-2.09\%$
test_unlock_nested	0.7302ms	0.3621ms	2.7619 KOps/s	2.8021 KOps/s	$\color{#d91a1a}-1.43\%$
test_unlock_stack_nested	0.3927ms	0.3211ms	3.1146 KOps/s	3.1792 KOps/s	$\color{#d91a1a}-2.03\%$
test_flatten_speed	0.1883ms	0.1029ms	9.7165 KOps/s	9.7712 KOps/s	$\color{#d91a1a}-0.56\%$
test_unflatten_speed	0.3652ms	0.2933ms	3.4100 KOps/s	3.4766 KOps/s	$\color{#d91a1a}-1.92\%$
test_common_ops	1.1389ms	0.5732ms	1.7447 KOps/s	1.7628 KOps/s	$\color{#d91a1a}-1.02\%$
test_creation	38.5920μs	1.6642μs	600.9036 KOps/s	598.6802 KOps/s	$\color{#35bf28}+0.37\%$
test_creation_empty	28.6320μs	7.6100μs	131.4065 KOps/s	115.3144 KOps/s	$\textbf{\color{#35bf28}+13.95\%}$
test_creation_nested_1	29.0110μs	9.4238μs	106.1148 KOps/s	95.7513 KOps/s	$\textbf{\color{#35bf28}+10.82\%}$
test_creation_nested_2	46.7230μs	11.6532μs	85.8137 KOps/s	79.8337 KOps/s	$\textbf{\color{#35bf28}+7.49\%}$
test_clone	77.4350μs	12.1450μs	82.3386 KOps/s	88.3952 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_getitem[int]	33.3220μs	11.9358μs	83.7812 KOps/s	91.8029 KOps/s	$\textbf{\color{#d91a1a}-8.74\%}$
test_getitem[slice_int]	57.8930μs	21.7115μs	46.0584 KOps/s	49.0835 KOps/s	$\textbf{\color{#d91a1a}-6.16\%}$
test_getitem[range]	69.0040μs	49.9890μs	20.0044 KOps/s	19.1193 KOps/s	$\color{#35bf28}+4.63\%$
test_getitem[tuple]	48.5320μs	19.6746μs	50.8269 KOps/s	54.8287 KOps/s	$\textbf{\color{#d91a1a}-7.30\%}$
test_getitem[list]	0.1230ms	35.7794μs	27.9490 KOps/s	29.9067 KOps/s	$\textbf{\color{#d91a1a}-6.55\%}$
test_setitem_dim[int]	51.5730μs	28.6500μs	34.9040 KOps/s	33.8125 KOps/s	$\color{#35bf28}+3.23\%$
test_setitem_dim[slice_int]	72.4940μs	48.8933μs	20.4527 KOps/s	20.1544 KOps/s	$\color{#35bf28}+1.48\%$
test_setitem_dim[range]	0.1091ms	67.4865μs	14.8178 KOps/s	14.7248 KOps/s	$\color{#35bf28}+0.63\%$
test_setitem_dim[tuple]	78.1150μs	42.5519μs	23.5007 KOps/s	22.8846 KOps/s	$\color{#35bf28}+2.69\%$
test_setitem	45.9220μs	16.4785μs	60.6850 KOps/s	62.8143 KOps/s	$\color{#d91a1a}-3.39\%$
test_set	57.2030μs	15.8347μs	63.1524 KOps/s	65.6941 KOps/s	$\color{#d91a1a}-3.87\%$
test_set_shared	1.1689ms	98.5446μs	10.1477 KOps/s	10.3743 KOps/s	$\color{#d91a1a}-2.18\%$
test_update	89.3860μs	17.6776μs	56.5687 KOps/s	56.8203 KOps/s	$\color{#d91a1a}-0.44\%$
test_update_nested	85.0460μs	22.9192μs	43.6316 KOps/s	44.0651 KOps/s	$\color{#d91a1a}-0.98\%$
test_update__nested	61.1840μs	22.7622μs	43.9324 KOps/s	45.9820 KOps/s	$\color{#d91a1a}-4.46\%$
test_set_nested	74.5040μs	17.1537μs	58.2965 KOps/s	61.2690 KOps/s	$\color{#d91a1a}-4.85\%$
test_set_nested_new	71.4140μs	20.0572μs	49.8575 KOps/s	52.2910 KOps/s	$\color{#d91a1a}-4.65\%$
test_select	75.1240μs	33.2483μs	30.0767 KOps/s	30.9835 KOps/s	$\color{#d91a1a}-2.93\%$
test_select_nested	96.3760μs	56.1446μs	17.8112 KOps/s	18.5414 KOps/s	$\color{#d91a1a}-3.94\%$
test_exclude_nested	0.1565ms	0.1129ms	8.8598 KOps/s	9.2051 KOps/s	$\color{#d91a1a}-3.75\%$
test_empty[True]	0.4200ms	0.3499ms	2.8576 KOps/s	2.9063 KOps/s	$\color{#d91a1a}-1.68\%$
test_empty[False]	3.0972μs	0.9451μs	1.0581 MOps/s	1.0571 MOps/s	$\color{#35bf28}+0.10\%$
test_to	0.1039ms	79.9786μs	12.5033 KOps/s	13.3669 KOps/s	$\textbf{\color{#d91a1a}-6.46\%}$
test_to_nonblocking	0.1170ms	64.5584μs	15.4898 KOps/s	16.3365 KOps/s	$\textbf{\color{#d91a1a}-5.18\%}$
test_unbind_speed	1.6060ms	0.2759ms	3.6247 KOps/s	3.7606 KOps/s	$\color{#d91a1a}-3.61\%$
test_unbind_speed_stack0	0.3425ms	0.2754ms	3.6306 KOps/s	3.7215 KOps/s	$\color{#d91a1a}-2.44\%$
test_unbind_speed_stack1	74.5231ms	0.8080ms	1.2377 KOps/s	1.2293 KOps/s	$\color{#35bf28}+0.68\%$
test_split	74.9811ms	1.7894ms	558.8612 Ops/s	633.3254 Ops/s	$\textbf{\color{#d91a1a}-11.76\%}$
test_chunk	74.9102ms	1.7908ms	558.4169 Ops/s	642.6691 Ops/s	$\textbf{\color{#d91a1a}-13.11\%}$
test_creation[device0]	0.1336ms	60.3812μs	16.5614 KOps/s	17.1957 KOps/s	$\color{#d91a1a}-3.69\%$
test_creation_from_tensor	0.1352ms	55.3706μs	18.0601 KOps/s	18.5590 KOps/s	$\color{#d91a1a}-2.69\%$
test_add_one[memmap_tensor0]	94.8060μs	7.1287μs	140.2774 KOps/s	152.9119 KOps/s	$\textbf{\color{#d91a1a}-8.26\%}$
test_contiguous[memmap_tensor0]	23.2710μs	0.7279μs	1.3739 MOps/s	1.4350 MOps/s	$\color{#d91a1a}-4.26\%$
test_stack[memmap_tensor0]	27.3820μs	5.2427μs	190.7403 KOps/s	217.4964 KOps/s	$\textbf{\color{#d91a1a}-12.30\%}$
test_memmaptd_index	1.3202ms	0.3158ms	3.1669 KOps/s	3.4103 KOps/s	$\textbf{\color{#d91a1a}-7.14\%}$
test_memmaptd_index_astensor	0.6608ms	0.3880ms	2.5776 KOps/s	2.7778 KOps/s	$\textbf{\color{#d91a1a}-7.21\%}$
test_memmaptd_index_op	1.1121ms	0.6771ms	1.4768 KOps/s	1.5345 KOps/s	$\color{#d91a1a}-3.76\%$
test_serialize_model	0.1803s	0.1104s	9.0572 Ops/s	9.3825 Ops/s	$\color{#d91a1a}-3.47\%$
test_serialize_model_pickle	1.3512s	1.2359s	0.8091 Ops/s	0.8086 Ops/s	$\color{#35bf28}+0.06\%$
test_serialize_weights	0.1809s	0.1094s	9.1374 Ops/s	8.8200 Ops/s	$\color{#35bf28}+3.60\%$
test_serialize_weights_returnearly	0.2556s	99.7480ms	10.0253 Ops/s	10.2959 Ops/s	$\color{#d91a1a}-2.63\%$
test_serialize_weights_pickle	1.3492s	1.2484s	0.8010 Ops/s	0.8012 Ops/s	$\color{#d91a1a}-0.03\%$
test_reshape_pytree	49.9230μs	28.1518μs	35.5217 KOps/s	38.5752 KOps/s	$\textbf{\color{#d91a1a}-7.92\%}$
test_reshape_td	66.7640μs	33.0578μs	30.2501 KOps/s	31.9068 KOps/s	$\textbf{\color{#d91a1a}-5.19\%}$
test_view_pytree	0.1875ms	28.5485μs	35.0281 KOps/s	39.2087 KOps/s	$\textbf{\color{#d91a1a}-10.66\%}$
test_view_td	0.1659ms	37.6636μs	26.5508 KOps/s	28.8335 KOps/s	$\textbf{\color{#d91a1a}-7.92\%}$
test_unbind_pytree	69.3740μs	32.6484μs	30.6293 KOps/s	31.5587 KOps/s	$\color{#d91a1a}-2.94\%$
test_unbind_td	0.4725ms	42.3295μs	23.6242 KOps/s	24.4217 KOps/s	$\color{#d91a1a}-3.27\%$
test_split_pytree	75.5340μs	36.6741μs	27.2672 KOps/s	28.6374 KOps/s	$\color{#d91a1a}-4.78\%$
test_split_td	0.1078ms	41.1522μs	24.3001 KOps/s	24.0548 KOps/s	$\color{#35bf28}+1.02\%$
test_add_pytree	71.4940μs	38.7936μs	25.7774 KOps/s	27.3155 KOps/s	$\textbf{\color{#d91a1a}-5.63\%}$
test_add_td	94.1150μs	47.8135μs	20.9146 KOps/s	19.0672 KOps/s	$\textbf{\color{#35bf28}+9.69\%}$
test_distributed	2.0137ms	70.4727μs	14.1899 KOps/s	13.6238 KOps/s	$\color{#35bf28}+4.16\%$
test_tdmodule	60.5540μs	14.3548μs	69.6631 KOps/s	66.0757 KOps/s	$\textbf{\color{#35bf28}+5.43\%}$
test_tdmodule_dispatch	52.4530μs	28.0403μs	35.6630 KOps/s	32.9037 KOps/s	$\textbf{\color{#35bf28}+8.39\%}$
test_tdseq	36.6320μs	16.2548μs	61.5203 KOps/s	60.1750 KOps/s	$\color{#35bf28}+2.24\%$
test_tdseq_dispatch	59.9530μs	31.4363μs	31.8104 KOps/s	31.4130 KOps/s	$\color{#35bf28}+1.26\%$
test_instantiation_functorch	1.6951ms	1.5840ms	631.3041 Ops/s	652.1915 Ops/s	$\color{#d91a1a}-3.20\%$
test_instantiation_td	1.5517ms	1.0656ms	938.4011 Ops/s	881.7645 Ops/s	$\textbf{\color{#35bf28}+6.42\%}$
test_exec_functorch	0.1901ms	0.1501ms	6.6612 KOps/s	6.9283 KOps/s	$\color{#d91a1a}-3.86\%$
test_exec_functional_call	0.1684ms	0.1354ms	7.3852 KOps/s	7.5526 KOps/s	$\color{#d91a1a}-2.22\%$
test_exec_td	0.1732ms	0.1363ms	7.3346 KOps/s	7.6291 KOps/s	$\color{#d91a1a}-3.86\%$
test_exec_td_decorator	0.6277ms	0.2063ms	4.8475 KOps/s	4.8805 KOps/s	$\color{#d91a1a}-0.68\%$
test_vmap_mlp_speed[True-True]	0.7049ms	0.6013ms	1.6632 KOps/s	1.6667 KOps/s	$\color{#d91a1a}-0.21\%$
test_vmap_mlp_speed[True-False]	0.7314ms	0.5995ms	1.6680 KOps/s	1.6815 KOps/s	$\color{#d91a1a}-0.80\%$
test_vmap_mlp_speed[False-True]	0.5945ms	0.5306ms	1.8848 KOps/s	1.9017 KOps/s	$\color{#d91a1a}-0.89\%$
test_vmap_mlp_speed[False-False]	0.6003ms	0.5321ms	1.8793 KOps/s	1.9007 KOps/s	$\color{#d91a1a}-1.13\%$
test_vmap_mlp_speed_decorator[True-True]	0.7702ms	0.6683ms	1.4964 KOps/s	1.5073 KOps/s	$\color{#d91a1a}-0.72\%$
test_vmap_mlp_speed_decorator[True-False]	0.8823ms	0.6735ms	1.4848 KOps/s	1.5133 KOps/s	$\color{#d91a1a}-1.89\%$
test_vmap_mlp_speed_decorator[False-True]	0.7369ms	0.5924ms	1.6880 KOps/s	1.7088 KOps/s	$\color{#d91a1a}-1.22\%$
test_vmap_mlp_speed_decorator[False-False]	0.7200ms	0.5893ms	1.6970 KOps/s	1.7101 KOps/s	$\color{#d91a1a}-0.77\%$
test_vmap_transformer_speed[True-True]	8.9793ms	8.5621ms	116.7941 Ops/s	126.4600 Ops/s	$\textbf{\color{#d91a1a}-7.64\%}$
test_vmap_transformer_speed[True-False]	8.9100ms	8.3268ms	120.0948 Ops/s	126.9000 Ops/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_vmap_transformer_speed[False-True]	8.6961ms	8.2315ms	121.4852 Ops/s	128.3681 Ops/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_vmap_transformer_speed[False-False]	8.6351ms	8.1730ms	122.3542 Ops/s	127.9720 Ops/s	$\color{#d91a1a}-4.39\%$
test_vmap_transformer_speed_decorator[True-True]	20.6357ms	19.9066ms	50.2345 Ops/s	52.2831 Ops/s	$\color{#d91a1a}-3.92\%$
test_vmap_transformer_speed_decorator[True-False]	20.6219ms	19.9576ms	50.1062 Ops/s	52.3111 Ops/s	$\color{#d91a1a}-4.21\%$
test_vmap_transformer_speed_decorator[False-True]	20.5003ms	19.8854ms	50.2881 Ops/s	52.4217 Ops/s	$\color{#d91a1a}-4.07\%$
test_vmap_transformer_speed_decorator[False-False]	20.3093ms	19.8123ms	50.4737 Ops/s	52.6818 Ops/s	$\color{#d91a1a}-4.19\%$
test_to_module_speed[True]	3.0952ms	1.5982ms	625.7029 Ops/s	650.6334 Ops/s	$\color{#d91a1a}-3.83\%$
test_to_module_speed[False]	2.0586ms	1.5798ms	632.9798 Ops/s	657.7509 Ops/s	$\color{#d91a1a}-3.77\%$
test_tc_init	0.1518ms	22.4810μs	44.4820 KOps/s	40.0170 KOps/s	$\textbf{\color{#35bf28}+11.16\%}$
test_tc_init_nested	0.1938ms	47.6141μs	21.0022 KOps/s	20.3099 KOps/s	$\color{#35bf28}+3.41\%$
test_tc_first_layer_tensor	3.4102μs	0.3637μs	2.7497 MOps/s	2.7998 MOps/s	$\color{#d91a1a}-1.79\%$
test_tc_first_layer_nontensor	7.3889μs	0.3874μs	2.5812 MOps/s	2.5822 MOps/s	$\color{#d91a1a}-0.04\%$
test_tc_second_layer_tensor	30.5300μs	0.9890μs	1.0112 MOps/s	1.0230 MOps/s	$\color{#d91a1a}-1.16\%$
test_tc_second_layer_nontensor	21.9465μs	0.8352μs	1.1973 MOps/s	1.2179 MOps/s	$\color{#d91a1a}-1.69\%$
test_unbind	98.2544ms	7.6940ms	129.9715 Ops/s	110.9132 Ops/s	$\textbf{\color{#35bf28}+17.18\%}$
test_full_like	13.7675ms	13.1476ms	76.0596 Ops/s	76.6439 Ops/s	$\color{#d91a1a}-0.76\%$
test_zeros_like	8.2214ms	7.8228ms	127.8310 Ops/s	126.0532 Ops/s	$\color{#35bf28}+1.41\%$
test_ones_like	8.6684ms	7.9025ms	126.5419 Ops/s	126.4027 Ops/s	$\color{#35bf28}+0.11\%$
test_clone	9.4636ms	9.1772ms	108.9655 Ops/s	108.3999 Ops/s	$\color{#35bf28}+0.52\%$
test_squeeze	60.4330μs	11.0721μs	90.3172 KOps/s	89.8805 KOps/s	$\color{#35bf28}+0.49\%$
test_unsqueeze	97.5260μs	55.1563μs	18.1303 KOps/s	18.8013 KOps/s	$\color{#d91a1a}-3.57\%$
test_split	0.1437ms	0.1018ms	9.8206 KOps/s	10.0585 KOps/s	$\color{#d91a1a}-2.37\%$
test_permute	0.1607ms	0.1172ms	8.5323 KOps/s	8.4481 KOps/s	$\color{#35bf28}+1.00\%$
test_stack	26.8676ms	26.6005ms	37.5932 Ops/s	37.6125 Ops/s	$\color{#d91a1a}-0.05\%$
test_cat	26.8871ms	26.5434ms	37.6741 Ops/s	37.6704 Ops/s	$+0.01\%$