[BugFix] Fix key ordering in pointwise ops

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 144. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	48.0400μs	17.4676μs	57.2487 KOps/s	59.3692 KOps/s	$\color{#d91a1a}-3.57\%$
test_plain_set_stack_nested	44.9830μs	17.6326μs	56.7133 KOps/s	59.0901 KOps/s	$\color{#d91a1a}-4.02\%$
test_plain_set_nested_inplace	52.9980μs	19.7295μs	50.6856 KOps/s	51.5276 KOps/s	$\color{#d91a1a}-1.63\%$
test_plain_set_stack_nested_inplace	58.3480μs	19.2812μs	51.8640 KOps/s	51.7272 KOps/s	$\color{#35bf28}+0.26\%$
test_items	33.1020μs	2.5346μs	394.5341 KOps/s	388.2307 KOps/s	$\color{#35bf28}+1.62\%$
test_items_nested	0.8977ms	0.2870ms	3.4838 KOps/s	3.6669 KOps/s	$\color{#d91a1a}-4.99\%$
test_items_nested_locked	0.4075ms	0.2873ms	3.4801 KOps/s	3.6545 KOps/s	$\color{#d91a1a}-4.77\%$
test_items_nested_leaf	0.1500ms	79.5179μs	12.5758 KOps/s	12.7497 KOps/s	$\color{#d91a1a}-1.36\%$
test_items_stack_nested	0.5658ms	0.2888ms	3.4628 KOps/s	3.6509 KOps/s	$\textbf{\color{#d91a1a}-5.15\%}$
test_items_stack_nested_leaf	0.1464ms	79.4202μs	12.5913 KOps/s	12.6915 KOps/s	$\color{#d91a1a}-0.79\%$
test_items_stack_nested_locked	0.7013ms	0.2901ms	3.4470 KOps/s	3.5722 KOps/s	$\color{#d91a1a}-3.50\%$
test_keys	32.8320μs	3.9897μs	250.6479 KOps/s	245.5290 KOps/s	$\color{#35bf28}+2.08\%$
test_keys_nested	0.2269ms	0.1372ms	7.2860 KOps/s	7.2812 KOps/s	$\color{#35bf28}+0.07\%$
test_keys_nested_locked	0.7233ms	0.1410ms	7.0918 KOps/s	6.8766 KOps/s	$\color{#35bf28}+3.13\%$
test_keys_nested_leaf	0.2068ms	0.1159ms	8.6302 KOps/s	8.4744 KOps/s	$\color{#35bf28}+1.84\%$
test_keys_stack_nested	0.2325ms	0.1371ms	7.2936 KOps/s	7.0867 KOps/s	$\color{#35bf28}+2.92\%$
test_keys_stack_nested_leaf	0.2517ms	0.1166ms	8.5729 KOps/s	8.5373 KOps/s	$\color{#35bf28}+0.42\%$
test_keys_stack_nested_locked	0.2305ms	0.1410ms	7.0936 KOps/s	6.9584 KOps/s	$\color{#35bf28}+1.94\%$
test_values	5.1634μs	1.1584μs	863.2640 KOps/s	866.1952 KOps/s	$\color{#d91a1a}-0.34\%$
test_values_nested	96.3490μs	50.7123μs	19.7191 KOps/s	20.0204 KOps/s	$\color{#d91a1a}-1.51\%$
test_values_nested_locked	0.1062ms	51.0613μs	19.5843 KOps/s	19.4873 KOps/s	$\color{#35bf28}+0.50\%$
test_values_nested_leaf	90.6490μs	46.1707μs	21.6587 KOps/s	22.1367 KOps/s	$\color{#d91a1a}-2.16\%$
test_values_stack_nested	0.1029ms	51.2310μs	19.5194 KOps/s	19.6480 KOps/s	$\color{#d91a1a}-0.65\%$
test_values_stack_nested_leaf	96.8600μs	45.8616μs	21.8047 KOps/s	21.9991 KOps/s	$\color{#d91a1a}-0.88\%$
test_values_stack_nested_locked	0.1022ms	51.1773μs	19.5399 KOps/s	19.8016 KOps/s	$\color{#d91a1a}-1.32\%$
test_membership	33.4420μs	1.3493μs	741.1241 KOps/s	742.3769 KOps/s	$\color{#d91a1a}-0.17\%$
test_membership_nested	39.9440μs	3.4440μs	290.3642 KOps/s	292.1907 KOps/s	$\color{#d91a1a}-0.63\%$
test_membership_nested_leaf	32.2610μs	3.4141μs	292.9023 KOps/s	286.1330 KOps/s	$\color{#35bf28}+2.37\%$
test_membership_stacked_nested	32.0790μs	3.3910μs	294.8950 KOps/s	292.6537 KOps/s	$\color{#35bf28}+0.77\%$
test_membership_stacked_nested_leaf	19.3460μs	3.4253μs	291.9477 KOps/s	284.5214 KOps/s	$\color{#35bf28}+2.61\%$
test_membership_nested_last	23.7940μs	4.2204μs	236.9462 KOps/s	237.5760 KOps/s	$\color{#d91a1a}-0.27\%$
test_membership_nested_leaf_last	21.7700μs	4.2264μs	236.6062 KOps/s	238.7599 KOps/s	$\color{#d91a1a}-0.90\%$
test_membership_stacked_nested_last	45.0740μs	4.1381μs	241.6560 KOps/s	239.7896 KOps/s	$\color{#35bf28}+0.78\%$
test_membership_stacked_nested_leaf_last	26.8100μs	4.1600μs	240.3844 KOps/s	236.6099 KOps/s	$\color{#35bf28}+1.60\%$
test_nested_getleaf	43.9410μs	10.7662μs	92.8836 KOps/s	90.3187 KOps/s	$\color{#35bf28}+2.84\%$
test_nested_get	43.4710μs	10.1914μs	98.1222 KOps/s	97.0166 KOps/s	$\color{#35bf28}+1.14\%$
test_stacked_getleaf	90.4380μs	10.8017μs	92.5777 KOps/s	92.5464 KOps/s	$\color{#35bf28}+0.03\%$
test_stacked_get	27.4110μs	10.2003μs	98.0364 KOps/s	97.5946 KOps/s	$\color{#35bf28}+0.45\%$
test_nested_getitemleaf	50.9850μs	11.3136μs	88.3891 KOps/s	87.1703 KOps/s	$\color{#35bf28}+1.40\%$
test_nested_getitem	47.3980μs	10.4270μs	95.9046 KOps/s	96.5153 KOps/s	$\color{#d91a1a}-0.63\%$
test_stacked_getitemleaf	65.9920μs	11.1999μs	89.2864 KOps/s	88.2859 KOps/s	$\color{#35bf28}+1.13\%$
test_stacked_getitem	49.5820μs	10.4183μs	95.9848 KOps/s	94.4609 KOps/s	$\color{#35bf28}+1.61\%$
test_lock_nested	1.9713ms	0.3321ms	3.0114 KOps/s	3.0383 KOps/s	$\color{#d91a1a}-0.88\%$
test_lock_stack_nested	0.3647ms	0.3019ms	3.3125 KOps/s	3.3095 KOps/s	$\color{#35bf28}+0.09\%$
test_unlock_nested	0.8446ms	0.3336ms	2.9976 KOps/s	3.0015 KOps/s	$\color{#d91a1a}-0.13\%$
test_unlock_stack_nested	0.4362ms	0.3106ms	3.2197 KOps/s	3.2256 KOps/s	$\color{#d91a1a}-0.18\%$
test_flatten_speed	0.3599ms	97.8982μs	10.2147 KOps/s	10.3779 KOps/s	$\color{#d91a1a}-1.57\%$
test_unflatten_speed	0.5928ms	0.4081ms	2.4502 KOps/s	2.4476 KOps/s	$\color{#35bf28}+0.10\%$
test_common_ops	3.9255ms	0.7355ms	1.3595 KOps/s	1.3751 KOps/s	$\color{#d91a1a}-1.13\%$
test_creation	15.6990μs	1.8890μs	529.3852 KOps/s	530.2646 KOps/s	$\color{#d91a1a}-0.17\%$
test_creation_empty	40.6960μs	11.1645μs	89.5696 KOps/s	91.8493 KOps/s	$\color{#d91a1a}-2.48\%$
test_creation_nested_1	58.1180μs	13.5524μs	73.7878 KOps/s	73.9783 KOps/s	$\color{#d91a1a}-0.26\%$
test_creation_nested_2	41.0770μs	16.8276μs	59.4261 KOps/s	59.1302 KOps/s	$\color{#35bf28}+0.50\%$
test_clone	86.9710μs	12.7595μs	78.3729 KOps/s	75.6635 KOps/s	$\color{#35bf28}+3.58\%$
test_getitem[int]	86.6910μs	10.9375μs	91.4282 KOps/s	90.4363 KOps/s	$\color{#35bf28}+1.10\%$
test_getitem[slice_int]	79.5280μs	22.0188μs	45.4157 KOps/s	43.5293 KOps/s	$\color{#35bf28}+4.33\%$
test_getitem[range]	78.2050μs	59.0516μs	16.9343 KOps/s	17.1898 KOps/s	$\color{#d91a1a}-1.49\%$
test_getitem[tuple]	50.4640μs	18.4174μs	54.2965 KOps/s	53.4881 KOps/s	$\color{#35bf28}+1.51\%$
test_getitem[list]	77.9650μs	39.6869μs	25.1972 KOps/s	24.7631 KOps/s	$\color{#35bf28}+1.75\%$
test_setitem_dim[int]	0.1581ms	35.5473μs	28.1315 KOps/s	28.4799 KOps/s	$\color{#d91a1a}-1.22\%$
test_setitem_dim[slice_int]	89.6670μs	60.8280μs	16.4398 KOps/s	16.0174 KOps/s	$\color{#35bf28}+2.64\%$
test_setitem_dim[range]	0.1454ms	82.3422μs	12.1444 KOps/s	12.0598 KOps/s	$\color{#35bf28}+0.70\%$
test_setitem_dim[tuple]	91.3500μs	48.4054μs	20.6588 KOps/s	19.6323 KOps/s	$\textbf{\color{#35bf28}+5.23\%}$
test_setitem	51.9960μs	19.3406μs	51.7046 KOps/s	48.1569 KOps/s	$\textbf{\color{#35bf28}+7.37\%}$
test_set	73.2740μs	18.8158μs	53.1468 KOps/s	50.7435 KOps/s	$\color{#35bf28}+4.74\%$
test_set_shared	3.9772ms	0.1423ms	7.0269 KOps/s	6.9668 KOps/s	$\color{#35bf28}+0.86\%$
test_update	0.1298ms	22.2611μs	44.9215 KOps/s	43.2631 KOps/s	$\color{#35bf28}+3.83\%$
test_update_nested	74.4080μs	31.0185μs	32.2388 KOps/s	30.9600 KOps/s	$\color{#35bf28}+4.13\%$
test_update__nested	0.1183ms	24.8029μs	40.3179 KOps/s	39.3909 KOps/s	$\color{#35bf28}+2.35\%$
test_set_nested	67.1250μs	21.0451μs	47.5170 KOps/s	45.2224 KOps/s	$\textbf{\color{#35bf28}+5.07\%}$
test_set_nested_new	83.0850μs	25.1771μs	39.7186 KOps/s	38.0336 KOps/s	$\color{#35bf28}+4.43\%$
test_select	0.1014ms	40.3256μs	24.7982 KOps/s	24.0472 KOps/s	$\color{#35bf28}+3.12\%$
test_select_nested	0.1173ms	57.2151μs	17.4779 KOps/s	17.5885 KOps/s	$\color{#d91a1a}-0.63\%$
test_exclude_nested	0.2675ms	0.1176ms	8.5069 KOps/s	8.3296 KOps/s	$\color{#35bf28}+2.13\%$
test_empty[True]	0.5769ms	0.3937ms	2.5397 KOps/s	2.5363 KOps/s	$\color{#35bf28}+0.13\%$
test_empty[False]	8.2874μs	1.0231μs	977.4286 KOps/s	989.4440 KOps/s	$\color{#d91a1a}-1.21\%$
test_unbind_speed	0.4523ms	0.2435ms	4.1065 KOps/s	4.0806 KOps/s	$\color{#35bf28}+0.64\%$
test_unbind_speed_stack0	0.4401ms	0.2438ms	4.1018 KOps/s	4.1893 KOps/s	$\color{#d91a1a}-2.09\%$
test_unbind_speed_stack1	67.6612ms	0.7046ms	1.4192 KOps/s	1.4664 KOps/s	$\color{#d91a1a}-3.22\%$
test_split	72.7578ms	1.5814ms	632.3617 Ops/s	636.4504 Ops/s	$\color{#d91a1a}-0.64\%$
test_chunk	65.0146ms	1.5658ms	638.6522 Ops/s	633.8638 Ops/s	$\color{#35bf28}+0.76\%$
test_creation[device0]	0.1627ms	82.4405μs	12.1300 KOps/s	11.8951 KOps/s	$\color{#35bf28}+1.97\%$
test_creation_from_tensor	4.0303ms	84.8635μs	11.7836 KOps/s	11.4287 KOps/s	$\color{#35bf28}+3.11\%$
test_add_one[memmap_tensor0]	41.2670μs	5.7853μs	172.8504 KOps/s	182.0908 KOps/s	$\textbf{\color{#d91a1a}-5.07\%}$
test_contiguous[memmap_tensor0]	20.0270μs	0.6532μs	1.5310 MOps/s	1.5558 MOps/s	$\color{#d91a1a}-1.60\%$
test_stack[memmap_tensor0]	22.8430μs	3.6951μs	270.6283 KOps/s	264.5751 KOps/s	$\color{#35bf28}+2.29\%$
test_memmaptd_index	1.0457ms	0.2629ms	3.8040 KOps/s	3.8622 KOps/s	$\color{#d91a1a}-1.51\%$
test_memmaptd_index_astensor	0.6812ms	0.3355ms	2.9808 KOps/s	2.9821 KOps/s	$\color{#d91a1a}-0.05\%$
test_memmaptd_index_op	0.9768ms	0.6388ms	1.5655 KOps/s	1.5752 KOps/s	$\color{#d91a1a}-0.62\%$
test_serialize_model	0.1703s	0.1051s	9.5115 Ops/s	10.3705 Ops/s	$\textbf{\color{#d91a1a}-8.28\%}$
test_serialize_model_pickle	0.4564s	0.3785s	2.6422 Ops/s	2.6248 Ops/s	$\color{#35bf28}+0.66\%$
test_serialize_weights	97.3648ms	93.5610ms	10.6882 Ops/s	9.5075 Ops/s	$\textbf{\color{#35bf28}+12.42\%}$
test_serialize_weights_returnearly	0.1829s	0.1273s	7.8555 Ops/s	8.6030 Ops/s	$\textbf{\color{#d91a1a}-8.69\%}$
test_serialize_weights_pickle	0.7621s	0.4896s	2.0423 Ops/s	1.5805 Ops/s	$\textbf{\color{#35bf28}+29.22\%}$
test_serialize_weights_filesystem	99.1071ms	92.2741ms	10.8373 Ops/s	10.3740 Ops/s	$\color{#35bf28}+4.47\%$
test_serialize_model_filesystem	0.1545s	98.2059ms	10.1827 Ops/s	10.7650 Ops/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_reshape_pytree	73.3960μs	25.4517μs	39.2902 KOps/s	39.1798 KOps/s	$\color{#35bf28}+0.28\%$
test_reshape_td	79.5080μs	33.1326μs	30.1817 KOps/s	29.7268 KOps/s	$\color{#35bf28}+1.53\%$
test_view_pytree	71.2320μs	25.3080μs	39.5131 KOps/s	39.2397 KOps/s	$\color{#35bf28}+0.70\%$
test_view_td	83.7760μs	37.6721μs	26.5448 KOps/s	25.5600 KOps/s	$\color{#35bf28}+3.85\%$
test_unbind_pytree	90.9590μs	29.4025μs	34.0108 KOps/s	34.0353 KOps/s	$\color{#d91a1a}-0.07\%$
test_unbind_td	0.4424ms	36.2495μs	27.5866 KOps/s	27.5263 KOps/s	$\color{#35bf28}+0.22\%$
test_split_pytree	80.7210μs	28.9392μs	34.5553 KOps/s	33.8029 KOps/s	$\color{#35bf28}+2.23\%$
test_split_td	0.1193ms	38.1275μs	26.2278 KOps/s	25.0536 KOps/s	$\color{#35bf28}+4.69\%$
test_add_pytree	0.1209ms	35.0464μs	28.5336 KOps/s	27.6699 KOps/s	$\color{#35bf28}+3.12\%$
test_add_td	0.1332ms	57.7992μs	17.3013 KOps/s	17.4066 KOps/s	$\color{#d91a1a}-0.61\%$
test_distributed	0.1780ms	0.1033ms	9.6787 KOps/s	9.6895 KOps/s	$\color{#d91a1a}-0.11\%$
test_tdmodule	45.4650μs	18.5516μs	53.9036 KOps/s	56.8430 KOps/s	$\textbf{\color{#d91a1a}-5.17\%}$
test_tdmodule_dispatch	53.7400μs	36.3330μs	27.5232 KOps/s	27.8732 KOps/s	$\color{#d91a1a}-1.26\%$
test_tdseq	42.0080μs	21.2272μs	47.1093 KOps/s	49.3227 KOps/s	$\color{#d91a1a}-4.49\%$
test_tdseq_dispatch	77.7740μs	41.6460μs	24.0119 KOps/s	24.8745 KOps/s	$\color{#d91a1a}-3.47\%$
test_instantiation_functorch	1.9451ms	1.3217ms	756.5899 Ops/s	742.2690 Ops/s	$\color{#35bf28}+1.93\%$
test_instantiation_td	1.5508ms	1.0194ms	980.9335 Ops/s	967.4085 Ops/s	$\color{#35bf28}+1.40\%$
test_exec_functorch	0.2979ms	0.1750ms	5.7151 KOps/s	6.1332 KOps/s	$\textbf{\color{#d91a1a}-6.82\%}$
test_exec_functional_call	0.2985ms	0.1536ms	6.5119 KOps/s	6.5214 KOps/s	$\color{#d91a1a}-0.15\%$
test_exec_td	0.2819ms	0.1477ms	6.7690 KOps/s	6.7960 KOps/s	$\color{#d91a1a}-0.40\%$
test_exec_td_decorator	0.3603ms	0.2237ms	4.4701 KOps/s	4.4692 KOps/s	$\color{#35bf28}+0.02\%$
test_vmap_mlp_speed[True-True]	0.8252ms	0.4962ms	2.0154 KOps/s	2.0567 KOps/s	$\color{#d91a1a}-2.01\%$
test_vmap_mlp_speed[True-False]	0.9009ms	0.4912ms	2.0357 KOps/s	2.0670 KOps/s	$\color{#d91a1a}-1.51\%$
test_vmap_mlp_speed[False-True]	0.6609ms	0.4034ms	2.4789 KOps/s	2.5394 KOps/s	$\color{#d91a1a}-2.38\%$
test_vmap_mlp_speed[False-False]	0.6138ms	0.4011ms	2.4930 KOps/s	2.5261 KOps/s	$\color{#d91a1a}-1.31\%$
test_vmap_mlp_speed_decorator[True-True]	0.8938ms	0.5672ms	1.7632 KOps/s	1.7913 KOps/s	$\color{#d91a1a}-1.57\%$
test_vmap_mlp_speed_decorator[True-False]	0.9672ms	0.5669ms	1.7640 KOps/s	1.8062 KOps/s	$\color{#d91a1a}-2.33\%$
test_vmap_mlp_speed_decorator[False-True]	0.7698ms	0.4636ms	2.1570 KOps/s	2.2038 KOps/s	$\color{#d91a1a}-2.12\%$
test_vmap_mlp_speed_decorator[False-False]	0.8328ms	0.4636ms	2.1571 KOps/s	2.1972 KOps/s	$\color{#d91a1a}-1.83\%$
test_to_module_speed[True]	2.7421ms	1.6719ms	598.1370 Ops/s	593.3370 Ops/s	$\color{#35bf28}+0.81\%$
test_to_module_speed[False]	1.7697ms	1.6353ms	611.4960 Ops/s	613.9396 Ops/s	$\color{#d91a1a}-0.40\%$
test_tc_init	0.1193ms	57.8828μs	17.2763 KOps/s	16.4827 KOps/s	$\color{#35bf28}+4.81\%$
test_tc_init_nested	0.1974ms	0.1187ms	8.4245 KOps/s	8.4609 KOps/s	$\color{#d91a1a}-0.43\%$
test_tc_first_layer_tensor	28.3130μs	8.2675μs	120.9561 KOps/s	114.8011 KOps/s	$\textbf{\color{#35bf28}+5.36\%}$
test_tc_first_layer_nontensor	48.9110μs	8.1887μs	122.1193 KOps/s	113.8747 KOps/s	$\textbf{\color{#35bf28}+7.24\%}$
test_tc_second_layer_tensor	25.7980μs	2.5271μs	395.7068 KOps/s	379.3959 KOps/s	$\color{#35bf28}+4.30\%$
test_tc_second_layer_nontensor	50.3230μs	9.2188μs	108.4736 KOps/s	101.4017 KOps/s	$\textbf{\color{#35bf28}+6.97\%}$
test_unbind	84.9729ms	14.5708ms	68.6304 Ops/s	71.1493 Ops/s	$\color{#d91a1a}-3.54\%$
test_full_like	8.8912ms	7.1534ms	139.7940 Ops/s	146.4651 Ops/s	$\color{#d91a1a}-4.55\%$
test_zeros_like	11.5723ms	5.3233ms	187.8549 Ops/s	170.8059 Ops/s	$\textbf{\color{#35bf28}+9.98\%}$
test_ones_like	12.3629ms	6.2157ms	160.8827 Ops/s	168.7361 Ops/s	$\color{#d91a1a}-4.65\%$
test_clone	15.3356ms	8.0565ms	124.1230 Ops/s	132.4394 Ops/s	$\textbf{\color{#d91a1a}-6.28\%}$
test_squeeze	66.8540μs	12.8595μs	77.7637 KOps/s	76.5473 KOps/s	$\color{#35bf28}+1.59\%$
test_unsqueeze	0.2603ms	96.7769μs	10.3330 KOps/s	9.8892 KOps/s	$\color{#35bf28}+4.49\%$
test_split	0.6016ms	0.2798ms	3.5740 KOps/s	3.5770 KOps/s	$\color{#d91a1a}-0.08\%$
test_permute	0.4804ms	0.2268ms	4.4085 KOps/s	4.3523 KOps/s	$\color{#35bf28}+1.29\%$
test_stack	23.8765ms	21.6894ms	46.1054 Ops/s	46.8003 Ops/s	$\color{#d91a1a}-1.48\%$
test_cat	24.0961ms	21.2849ms	46.9816 Ops/s	47.5332 Ops/s	$\color{#d91a1a}-1.16\%$

[github-actions]

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

Total Benchmarks: 152. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}22$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	26.1500μs	13.9320μs	71.7771 KOps/s	80.1548 KOps/s	$\textbf{\color{#d91a1a}-10.45\%}$
test_plain_set_stack_nested	32.1810μs	13.8584μs	72.1582 KOps/s	79.9748 KOps/s	$\textbf{\color{#d91a1a}-9.77\%}$
test_plain_set_nested_inplace	39.3300μs	15.1630μs	65.9499 KOps/s	72.6469 KOps/s	$\textbf{\color{#d91a1a}-9.22\%}$
test_plain_set_stack_nested_inplace	31.8510μs	15.1347μs	66.0733 KOps/s	72.6988 KOps/s	$\textbf{\color{#d91a1a}-9.11\%}$
test_items	28.5200μs	4.6457μs	215.2550 KOps/s	212.9098 KOps/s	$\color{#35bf28}+1.10\%$
test_items_nested	0.3786ms	0.3407ms	2.9350 KOps/s	2.9311 KOps/s	$\color{#35bf28}+0.13\%$
test_items_nested_locked	0.3591ms	0.3381ms	2.9581 KOps/s	2.8855 KOps/s	$\color{#35bf28}+2.51\%$
test_items_nested_leaf	99.1610μs	82.6127μs	12.1047 KOps/s	12.1198 KOps/s	$\color{#d91a1a}-0.12\%$
test_items_stack_nested	0.3712ms	0.3452ms	2.8970 KOps/s	2.9347 KOps/s	$\color{#d91a1a}-1.29\%$
test_items_stack_nested_leaf	0.1058ms	83.2768μs	12.0081 KOps/s	11.7012 KOps/s	$\color{#35bf28}+2.62\%$
test_items_stack_nested_locked	0.3755ms	0.3436ms	2.9107 KOps/s	2.9287 KOps/s	$\color{#d91a1a}-0.61\%$
test_keys	28.2210μs	4.3562μs	229.5588 KOps/s	228.1846 KOps/s	$\color{#35bf28}+0.60\%$
test_keys_nested	93.6020μs	68.6672μs	14.5630 KOps/s	14.4130 KOps/s	$\color{#35bf28}+1.04\%$
test_keys_nested_locked	0.7580ms	74.2838μs	13.4619 KOps/s	13.4048 KOps/s	$\color{#35bf28}+0.43\%$
test_keys_nested_leaf	84.8210μs	59.5444μs	16.7942 KOps/s	16.8691 KOps/s	$\color{#d91a1a}-0.44\%$
test_keys_stack_nested	98.7410μs	68.1472μs	14.6741 KOps/s	14.3885 KOps/s	$\color{#35bf28}+1.99\%$
test_keys_stack_nested_leaf	82.8410μs	57.4708μs	17.4002 KOps/s	16.8073 KOps/s	$\color{#35bf28}+3.53\%$
test_keys_stack_nested_locked	97.4320μs	72.9814μs	13.7021 KOps/s	13.3041 KOps/s	$\color{#35bf28}+2.99\%$
test_values	9.7870μs	1.8054μs	553.8880 KOps/s	555.2515 KOps/s	$\color{#d91a1a}-0.25\%$
test_values_nested	65.5410μs	35.1973μs	28.4113 KOps/s	28.3311 KOps/s	$\color{#35bf28}+0.28\%$
test_values_nested_locked	66.8010μs	37.1937μs	26.8863 KOps/s	26.8611 KOps/s	$\color{#35bf28}+0.09\%$
test_values_nested_leaf	48.9510μs	31.4843μs	31.7619 KOps/s	31.5795 KOps/s	$\color{#35bf28}+0.58\%$
test_values_stack_nested	66.9310μs	36.2381μs	27.5953 KOps/s	27.7524 KOps/s	$\color{#d91a1a}-0.57\%$
test_values_stack_nested_leaf	59.2810μs	32.3150μs	30.9453 KOps/s	31.2352 KOps/s	$\color{#d91a1a}-0.93\%$
test_values_stack_nested_locked	61.1010μs	38.1045μs	26.2436 KOps/s	26.6896 KOps/s	$\color{#d91a1a}-1.67\%$
test_membership	1.7140μs	0.6932μs	1.4425 MOps/s	1.4463 MOps/s	$\color{#d91a1a}-0.27\%$
test_membership_nested	17.6800μs	2.6391μs	378.9142 KOps/s	388.1110 KOps/s	$\color{#d91a1a}-2.37\%$
test_membership_nested_leaf	30.5610μs	2.6160μs	382.2583 KOps/s	388.8326 KOps/s	$\color{#d91a1a}-1.69\%$
test_membership_stacked_nested	20.2100μs	2.5756μs	388.2560 KOps/s	392.9615 KOps/s	$\color{#d91a1a}-1.20\%$
test_membership_stacked_nested_leaf	17.8100μs	2.5794μs	387.6842 KOps/s	388.9431 KOps/s	$\color{#d91a1a}-0.32\%$
test_membership_nested_last	32.8610μs	3.0640μs	326.3743 KOps/s	326.7200 KOps/s	$\color{#d91a1a}-0.11\%$
test_membership_nested_leaf_last	21.3110μs	3.1207μs	320.4453 KOps/s	325.7590 KOps/s	$\color{#d91a1a}-1.63\%$
test_membership_stacked_nested_last	33.9400μs	3.1235μs	320.1540 KOps/s	323.4086 KOps/s	$\color{#d91a1a}-1.01\%$
test_membership_stacked_nested_leaf_last	21.6310μs	3.1310μs	319.3901 KOps/s	325.6722 KOps/s	$\color{#d91a1a}-1.93\%$
test_nested_getleaf	34.1800μs	8.2974μs	120.5197 KOps/s	119.0074 KOps/s	$\color{#35bf28}+1.27\%$
test_nested_get	31.8900μs	7.8031μs	128.1541 KOps/s	127.4612 KOps/s	$\color{#35bf28}+0.54\%$
test_stacked_getleaf	27.7010μs	8.3832μs	119.2858 KOps/s	118.8295 KOps/s	$\color{#35bf28}+0.38\%$
test_stacked_get	30.8310μs	7.7933μs	128.3147 KOps/s	127.0020 KOps/s	$\color{#35bf28}+1.03\%$
test_nested_getitemleaf	30.0700μs	8.4632μs	118.1589 KOps/s	116.4495 KOps/s	$\color{#35bf28}+1.47\%$
test_nested_getitem	33.9410μs	7.9988μs	125.0180 KOps/s	124.5599 KOps/s	$\color{#35bf28}+0.37\%$
test_stacked_getitemleaf	35.6310μs	8.5558μs	116.8797 KOps/s	116.1692 KOps/s	$\color{#35bf28}+0.61\%$
test_stacked_getitem	25.0200μs	7.9922μs	125.1227 KOps/s	124.2232 KOps/s	$\color{#35bf28}+0.72\%$
test_lock_nested	57.4349ms	0.3969ms	2.5194 KOps/s	2.4649 KOps/s	$\color{#35bf28}+2.21\%$
test_lock_stack_nested	0.3425ms	0.2919ms	3.4258 KOps/s	3.3111 KOps/s	$\color{#35bf28}+3.47\%$
test_unlock_nested	59.8476ms	0.3998ms	2.5015 KOps/s	2.4628 KOps/s	$\color{#35bf28}+1.57\%$
test_unlock_stack_nested	0.3212ms	0.3017ms	3.3150 KOps/s	3.2362 KOps/s	$\color{#35bf28}+2.43\%$
test_flatten_speed	0.4123ms	0.1017ms	9.8362 KOps/s	9.6626 KOps/s	$\color{#35bf28}+1.80\%$
test_unflatten_speed	0.3139ms	0.2898ms	3.4502 KOps/s	3.4180 KOps/s	$\color{#35bf28}+0.94\%$
test_common_ops	1.0930ms	0.6233ms	1.6043 KOps/s	1.7273 KOps/s	$\textbf{\color{#d91a1a}-7.12\%}$
test_creation	29.0000μs	1.6481μs	606.7445 KOps/s	616.8917 KOps/s	$\color{#d91a1a}-1.64\%$
test_creation_empty	26.4110μs	10.7088μs	93.3812 KOps/s	126.8563 KOps/s	$\textbf{\color{#d91a1a}-26.39\%}$
test_creation_nested_1	28.0310μs	12.5112μs	79.9284 KOps/s	103.4794 KOps/s	$\textbf{\color{#d91a1a}-22.76\%}$
test_creation_nested_2	51.3810μs	14.7616μs	67.7434 KOps/s	84.3790 KOps/s	$\textbf{\color{#d91a1a}-19.72\%}$
test_clone	56.9610μs	11.6527μs	85.8174 KOps/s	85.3292 KOps/s	$\color{#35bf28}+0.57\%$
test_getitem[int]	31.0600μs	10.6368μs	94.0133 KOps/s	94.0386 KOps/s	$\color{#d91a1a}-0.03\%$
test_getitem[slice_int]	46.9210μs	20.4875μs	48.8102 KOps/s	48.8220 KOps/s	$\color{#d91a1a}-0.02\%$
test_getitem[range]	64.4310μs	46.5046μs	21.5032 KOps/s	22.0678 KOps/s	$\color{#d91a1a}-2.56\%$
test_getitem[tuple]	42.0410μs	18.3026μs	54.6370 KOps/s	54.0202 KOps/s	$\color{#35bf28}+1.14\%$
test_getitem[list]	0.1519ms	31.9083μs	31.3398 KOps/s	30.5250 KOps/s	$\color{#35bf28}+2.67\%$
test_setitem_dim[int]	66.8410μs	28.4367μs	35.1658 KOps/s	37.2978 KOps/s	$\textbf{\color{#d91a1a}-5.72\%}$
test_setitem_dim[slice_int]	70.0310μs	48.5305μs	20.6056 KOps/s	20.8238 KOps/s	$\color{#d91a1a}-1.05\%$
test_setitem_dim[range]	88.6910μs	65.2632μs	15.3226 KOps/s	15.8482 KOps/s	$\color{#d91a1a}-3.32\%$
test_setitem_dim[tuple]	60.5010μs	42.1918μs	23.7013 KOps/s	23.9793 KOps/s	$\color{#d91a1a}-1.16\%$
test_setitem	47.4700μs	17.6002μs	56.8174 KOps/s	62.6276 KOps/s	$\textbf{\color{#d91a1a}-9.28\%}$
test_set	49.3110μs	17.1548μs	58.2927 KOps/s	63.7642 KOps/s	$\textbf{\color{#d91a1a}-8.58\%}$
test_set_shared	1.6643ms	98.4177μs	10.1608 KOps/s	10.1831 KOps/s	$\color{#d91a1a}-0.22\%$
test_update	90.7210μs	20.5981μs	48.5483 KOps/s	55.5269 KOps/s	$\textbf{\color{#d91a1a}-12.57\%}$
test_update_nested	69.1020μs	26.5024μs	37.7325 KOps/s	42.1088 KOps/s	$\textbf{\color{#d91a1a}-10.39\%}$
test_update__nested	67.7410μs	22.3876μs	44.6675 KOps/s	43.9879 KOps/s	$\color{#35bf28}+1.55\%$
test_set_nested	54.0410μs	17.9797μs	55.6183 KOps/s	59.3294 KOps/s	$\textbf{\color{#d91a1a}-6.26\%}$
test_set_nested_new	69.0810μs	20.8766μs	47.9006 KOps/s	51.3847 KOps/s	$\textbf{\color{#d91a1a}-6.78\%}$
test_select	66.6510μs	33.4357μs	29.9081 KOps/s	30.4024 KOps/s	$\color{#d91a1a}-1.63\%$
test_select_nested	0.6063ms	51.1052μs	19.5675 KOps/s	19.1088 KOps/s	$\color{#35bf28}+2.40\%$
test_exclude_nested	0.1409ms	0.1077ms	9.2833 KOps/s	9.0389 KOps/s	$\color{#35bf28}+2.70\%$
test_empty[True]	0.3779ms	0.3406ms	2.9360 KOps/s	2.8756 KOps/s	$\color{#35bf28}+2.10\%$
test_empty[False]	2.7930μs	0.7932μs	1.2608 MOps/s	1.2431 MOps/s	$\color{#35bf28}+1.42\%$
test_to	89.8920μs	58.9966μs	16.9501 KOps/s	17.1530 KOps/s	$\color{#d91a1a}-1.18\%$
test_to_nonblocking	75.2710μs	35.2139μs	28.3979 KOps/s	27.6741 KOps/s	$\color{#35bf28}+2.62\%$
test_unbind_speed	0.2927ms	0.2540ms	3.9376 KOps/s	3.8755 KOps/s	$\color{#35bf28}+1.60\%$
test_unbind_speed_stack0	0.2927ms	0.2547ms	3.9258 KOps/s	3.8177 KOps/s	$\color{#35bf28}+2.83\%$
test_unbind_speed_stack1	75.3276ms	0.7656ms	1.3062 KOps/s	1.2705 KOps/s	$\color{#35bf28}+2.81\%$
test_split	75.0175ms	1.6498ms	606.1403 Ops/s	600.4106 Ops/s	$\color{#35bf28}+0.95\%$
test_chunk	76.0235ms	1.6517ms	605.4437 Ops/s	599.9829 Ops/s	$\color{#35bf28}+0.91\%$
test_creation[device0]	0.1280ms	59.3094μs	16.8607 KOps/s	17.5575 KOps/s	$\color{#d91a1a}-3.97\%$
test_creation_from_tensor	0.1334ms	56.0179μs	17.8514 KOps/s	17.6168 KOps/s	$\color{#35bf28}+1.33\%$
test_add_one[memmap_tensor0]	51.7810μs	6.8567μs	145.8435 KOps/s	146.0735 KOps/s	$\color{#d91a1a}-0.16\%$
test_contiguous[memmap_tensor0]	29.5800μs	0.6601μs	1.5150 MOps/s	1.5153 MOps/s	$\color{#d91a1a}-0.02\%$
test_stack[memmap_tensor0]	29.8500μs	4.6944μs	213.0203 KOps/s	210.0070 KOps/s	$\color{#35bf28}+1.43\%$
test_memmaptd_index	1.0432ms	0.2764ms	3.6177 KOps/s	3.4875 KOps/s	$\color{#35bf28}+3.73\%$
test_memmaptd_index_astensor	0.5899ms	0.3329ms	3.0043 KOps/s	2.8611 KOps/s	$\textbf{\color{#35bf28}+5.00\%}$
test_memmaptd_index_op	0.9466ms	0.6587ms	1.5182 KOps/s	1.5932 KOps/s	$\color{#d91a1a}-4.71\%$
test_serialize_model	96.0782ms	90.8790ms	11.0036 Ops/s	10.4370 Ops/s	$\textbf{\color{#35bf28}+5.43\%}$
test_serialize_model_pickle	1.3485s	1.2353s	0.8095 Ops/s	0.8083 Ops/s	$\color{#35bf28}+0.15\%$
test_serialize_weights	93.0226ms	89.2009ms	11.2106 Ops/s	9.6680 Ops/s	$\textbf{\color{#35bf28}+15.96\%}$
test_serialize_weights_returnearly	0.2211s	75.8420ms	13.1853 Ops/s	13.2590 Ops/s	$\color{#d91a1a}-0.56\%$
test_serialize_weights_pickle	1.4070s	1.2542s	0.7973 Ops/s	0.7975 Ops/s	$\color{#d91a1a}-0.02\%$
test_reshape_pytree	51.2110μs	26.0183μs	38.4344 KOps/s	38.2878 KOps/s	$\color{#35bf28}+0.38\%$
test_reshape_td	56.0110μs	31.6022μs	31.6433 KOps/s	31.5799 KOps/s	$\color{#35bf28}+0.20\%$
test_view_pytree	48.8010μs	25.6677μs	38.9594 KOps/s	38.7218 KOps/s	$\color{#35bf28}+0.61\%$
test_view_td	67.1210μs	36.4615μs	27.4262 KOps/s	26.1745 KOps/s	$\color{#35bf28}+4.78\%$
test_unbind_pytree	54.9910μs	31.8505μs	31.3967 KOps/s	30.5265 KOps/s	$\color{#35bf28}+2.85\%$
test_unbind_td	0.4760ms	39.5292μs	25.2978 KOps/s	24.7536 KOps/s	$\color{#35bf28}+2.20\%$
test_split_pytree	65.4110μs	34.5099μs	28.9772 KOps/s	28.2453 KOps/s	$\color{#35bf28}+2.59\%$
test_split_td	0.4740ms	38.9037μs	25.7045 KOps/s	25.6009 KOps/s	$\color{#35bf28}+0.40\%$
test_add_pytree	64.7610μs	37.7100μs	26.5181 KOps/s	26.4722 KOps/s	$\color{#35bf28}+0.17\%$
test_add_td	82.1310μs	52.8089μs	18.9362 KOps/s	19.6436 KOps/s	$\color{#d91a1a}-3.60\%$
test_distributed	1.7340ms	85.8737μs	11.6450 KOps/s	11.4122 KOps/s	$\color{#35bf28}+2.04\%$
test_tdmodule	37.3710μs	16.4573μs	60.7632 KOps/s	65.2114 KOps/s	$\textbf{\color{#d91a1a}-6.82\%}$
test_tdmodule_dispatch	47.7100μs	31.7356μs	31.5103 KOps/s	33.5855 KOps/s	$\textbf{\color{#d91a1a}-6.18\%}$
test_tdseq	42.2200μs	17.6800μs	56.5612 KOps/s	59.2304 KOps/s	$\color{#d91a1a}-4.51\%$
test_tdseq_dispatch	63.1210μs	36.1993μs	27.6249 KOps/s	30.6682 KOps/s	$\textbf{\color{#d91a1a}-9.92\%}$
test_instantiation_functorch	1.5290ms	1.4049ms	711.8021 Ops/s	706.5376 Ops/s	$\color{#35bf28}+0.75\%$
test_instantiation_td	77.9213ms	1.0753ms	929.9976 Ops/s	922.8870 Ops/s	$\color{#35bf28}+0.77\%$
test_exec_functorch	0.1887ms	0.1432ms	6.9826 KOps/s	6.9129 KOps/s	$\color{#35bf28}+1.01\%$
test_exec_functional_call	0.3081ms	0.1323ms	7.5589 KOps/s	7.5565 KOps/s	$\color{#35bf28}+0.03\%$
test_exec_td	0.1770ms	0.1278ms	7.8243 KOps/s	7.7741 KOps/s	$\color{#35bf28}+0.64\%$
test_exec_td_decorator	0.5043ms	0.2011ms	4.9725 KOps/s	4.9529 KOps/s	$\color{#35bf28}+0.39\%$
test_vmap_mlp_speed[True-True]	0.5914ms	0.5585ms	1.7904 KOps/s	1.7964 KOps/s	$\color{#d91a1a}-0.34\%$
test_vmap_mlp_speed[True-False]	1.2862ms	0.5592ms	1.7883 KOps/s	1.7968 KOps/s	$\color{#d91a1a}-0.47\%$
test_vmap_mlp_speed[False-True]	0.5506ms	0.4856ms	2.0594 KOps/s	2.0485 KOps/s	$\color{#35bf28}+0.53\%$
test_vmap_mlp_speed[False-False]	0.6313ms	0.5115ms	1.9549 KOps/s	2.0096 KOps/s	$\color{#d91a1a}-2.72\%$
test_vmap_mlp_speed_decorator[True-True]	0.9343ms	0.6594ms	1.5165 KOps/s	1.4251 KOps/s	$\textbf{\color{#35bf28}+6.41\%}$
test_vmap_mlp_speed_decorator[True-False]	0.7961ms	0.6583ms	1.5190 KOps/s	1.6195 KOps/s	$\textbf{\color{#d91a1a}-6.21\%}$
test_vmap_mlp_speed_decorator[False-True]	0.7185ms	0.5780ms	1.7302 KOps/s	1.8290 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_vmap_mlp_speed_decorator[False-False]	0.7028ms	0.5761ms	1.7358 KOps/s	1.8171 KOps/s	$\color{#d91a1a}-4.47\%$
test_vmap_transformer_speed[True-True]	7.7595ms	7.3822ms	135.4607 Ops/s	135.9745 Ops/s	$\color{#d91a1a}-0.38\%$
test_vmap_transformer_speed[True-False]	7.3799ms	7.3087ms	136.8223 Ops/s	134.7067 Ops/s	$\color{#35bf28}+1.57\%$
test_vmap_transformer_speed[False-True]	7.3217ms	7.2553ms	137.8304 Ops/s	136.5672 Ops/s	$\color{#35bf28}+0.92\%$
test_vmap_transformer_speed[False-False]	7.7074ms	7.3509ms	136.0377 Ops/s	136.6042 Ops/s	$\color{#d91a1a}-0.41\%$
test_vmap_transformer_speed_decorator[True-True]	18.6781ms	17.7377ms	56.3771 Ops/s	56.0912 Ops/s	$\color{#35bf28}+0.51\%$
test_vmap_transformer_speed_decorator[True-False]	17.8692ms	17.7769ms	56.2529 Ops/s	56.1483 Ops/s	$\color{#35bf28}+0.19\%$
test_vmap_transformer_speed_decorator[False-True]	18.8808ms	17.6272ms	56.7305 Ops/s	56.5991 Ops/s	$\color{#35bf28}+0.23\%$
test_vmap_transformer_speed_decorator[False-False]	18.2325ms	17.6550ms	56.6411 Ops/s	56.3115 Ops/s	$\color{#35bf28}+0.59\%$
test_to_module_speed[True]	1.5995ms	1.4811ms	675.1798 Ops/s	669.7717 Ops/s	$\color{#35bf28}+0.81\%$
test_to_module_speed[False]	1.5838ms	1.4791ms	676.1063 Ops/s	677.8007 Ops/s	$\color{#d91a1a}-0.25\%$
test_tc_init	83.4710μs	59.5664μs	16.7880 KOps/s	19.4609 KOps/s	$\textbf{\color{#d91a1a}-13.73\%}$
test_tc_init_nested	0.1507ms	0.1174ms	8.5191 KOps/s	9.7781 KOps/s	$\textbf{\color{#d91a1a}-12.88\%}$
test_tc_first_layer_tensor	23.8410μs	3.6843μs	271.4221 KOps/s	268.8321 KOps/s	$\color{#35bf28}+0.96\%$
test_tc_first_layer_nontensor	26.1200μs	3.7084μs	269.6562 KOps/s	267.0216 KOps/s	$\color{#35bf28}+0.99\%$
test_tc_second_layer_tensor	6.5002μs	1.2089μs	827.1746 KOps/s	789.0325 KOps/s	$\color{#35bf28}+4.83\%$
test_tc_second_layer_nontensor	24.3000μs	4.2164μs	237.1689 KOps/s	234.1734 KOps/s	$\color{#35bf28}+1.28\%$
test_unbind	0.1115s	14.7457ms	67.8162 Ops/s	69.4159 Ops/s	$\color{#d91a1a}-2.30\%$
test_full_like	14.3288ms	13.5820ms	73.6267 Ops/s	72.2418 Ops/s	$\color{#35bf28}+1.92\%$
test_zeros_like	8.3559ms	7.9687ms	125.4908 Ops/s	124.7093 Ops/s	$\color{#35bf28}+0.63\%$
test_ones_like	8.5419ms	7.9642ms	125.5618 Ops/s	124.9674 Ops/s	$\color{#35bf28}+0.48\%$
test_clone	9.7521ms	9.4975ms	105.2905 Ops/s	100.2970 Ops/s	$\color{#35bf28}+4.98\%$
test_squeeze	72.1410μs	10.8217μs	92.4065 KOps/s	94.4496 KOps/s	$\color{#d91a1a}-2.16\%$
test_unsqueeze	0.1345ms	87.2276μs	11.4643 KOps/s	11.1088 KOps/s	$\color{#35bf28}+3.20\%$
test_split	3.4263ms	3.1011ms	322.4684 Ops/s	311.7268 Ops/s	$\color{#35bf28}+3.45\%$
test_permute	0.2630ms	0.2061ms	4.8509 KOps/s	4.9157 KOps/s	$\color{#d91a1a}-1.32\%$
test_stack	27.4510ms	27.2169ms	36.7419 Ops/s	36.3331 Ops/s	$\color{#35bf28}+1.13\%$
test_cat	27.2567ms	27.0044ms	37.0309 Ops/s	36.3955 Ops/s	$\color{#35bf28}+1.75\%$