[BugFix] remove inplace updates when using td as a decorator

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 144. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.0770μs	16.6264μs	60.1454 KOps/s	61.3415 KOps/s	$\color{#d91a1a}-1.95\%$
test_plain_set_stack_nested	51.3050μs	16.7642μs	59.6510 KOps/s	60.9553 KOps/s	$\color{#d91a1a}-2.14\%$
test_plain_set_nested_inplace	49.5220μs	19.1098μs	52.3291 KOps/s	52.3965 KOps/s	$\color{#d91a1a}-0.13\%$
test_plain_set_stack_nested_inplace	51.4560μs	19.0593μs	52.4678 KOps/s	53.9400 KOps/s	$\color{#d91a1a}-2.73\%$
test_items	15.5590μs	2.5255μs	395.9598 KOps/s	393.8502 KOps/s	$\color{#35bf28}+0.54\%$
test_items_nested	0.4215ms	0.2590ms	3.8613 KOps/s	3.7580 KOps/s	$\color{#35bf28}+2.75\%$
test_items_nested_locked	1.1214ms	0.2637ms	3.7924 KOps/s	3.6869 KOps/s	$\color{#35bf28}+2.86\%$
test_items_nested_leaf	0.1511ms	76.7021μs	13.0375 KOps/s	12.9575 KOps/s	$\color{#35bf28}+0.62\%$
test_items_stack_nested	0.4011ms	0.2629ms	3.8031 KOps/s	3.6974 KOps/s	$\color{#35bf28}+2.86\%$
test_items_stack_nested_leaf	0.1531ms	78.5717μs	12.7272 KOps/s	12.6484 KOps/s	$\color{#35bf28}+0.62\%$
test_items_stack_nested_locked	0.4241ms	0.2663ms	3.7546 KOps/s	3.6911 KOps/s	$\color{#35bf28}+1.72\%$
test_keys	19.1660μs	3.8554μs	259.3760 KOps/s	262.5280 KOps/s	$\color{#d91a1a}-1.20\%$
test_keys_nested	0.2579ms	0.1358ms	7.3663 KOps/s	7.2424 KOps/s	$\color{#35bf28}+1.71\%$
test_keys_nested_locked	0.6679ms	0.1405ms	7.1163 KOps/s	7.0466 KOps/s	$\color{#35bf28}+0.99\%$
test_keys_nested_leaf	0.1763ms	0.1136ms	8.8001 KOps/s	8.6124 KOps/s	$\color{#35bf28}+2.18\%$
test_keys_stack_nested	0.2504ms	0.1353ms	7.3889 KOps/s	7.3031 KOps/s	$\color{#35bf28}+1.18\%$
test_keys_stack_nested_leaf	0.2006ms	0.1140ms	8.7753 KOps/s	8.5889 KOps/s	$\color{#35bf28}+2.17\%$
test_keys_stack_nested_locked	0.3109ms	0.1412ms	7.0805 KOps/s	7.0513 KOps/s	$\color{#35bf28}+0.41\%$
test_values	8.7613μs	1.1592μs	862.6760 KOps/s	863.5318 KOps/s	$\color{#d91a1a}-0.10\%$
test_values_nested	0.1022ms	50.2421μs	19.9036 KOps/s	19.7791 KOps/s	$\color{#35bf28}+0.63\%$
test_values_nested_locked	0.1081ms	50.0313μs	19.9875 KOps/s	19.8979 KOps/s	$\color{#35bf28}+0.45\%$
test_values_nested_leaf	0.1031ms	45.1077μs	22.1691 KOps/s	21.8273 KOps/s	$\color{#35bf28}+1.57\%$
test_values_stack_nested	0.1048ms	51.1514μs	19.5498 KOps/s	19.3183 KOps/s	$\color{#35bf28}+1.20\%$
test_values_stack_nested_leaf	96.8600μs	44.7717μs	22.3355 KOps/s	22.0130 KOps/s	$\color{#35bf28}+1.47\%$
test_values_stack_nested_locked	93.5540μs	50.7467μs	19.7057 KOps/s	19.4757 KOps/s	$\color{#35bf28}+1.18\%$
test_membership	15.4990μs	1.3201μs	757.5261 KOps/s	742.5469 KOps/s	$\color{#35bf28}+2.02\%$
test_membership_nested	37.0690μs	3.4351μs	291.1107 KOps/s	288.6640 KOps/s	$\color{#35bf28}+0.85\%$
test_membership_nested_leaf	31.7590μs	3.4443μs	290.3306 KOps/s	285.6046 KOps/s	$\color{#35bf28}+1.65\%$
test_membership_stacked_nested	35.3860μs	3.4527μs	289.6287 KOps/s	286.4016 KOps/s	$\color{#35bf28}+1.13\%$
test_membership_stacked_nested_leaf	17.7420μs	3.4544μs	289.4859 KOps/s	286.9823 KOps/s	$\color{#35bf28}+0.87\%$
test_membership_nested_last	34.1340μs	4.2214μs	236.8906 KOps/s	213.0825 KOps/s	$\textbf{\color{#35bf28}+11.17\%}$
test_membership_nested_leaf_last	25.8880μs	4.1894μs	238.6962 KOps/s	236.8380 KOps/s	$\color{#35bf28}+0.78\%$
test_membership_stacked_nested_last	37.0190μs	5.7662μs	173.4240 KOps/s	190.0648 KOps/s	$\textbf{\color{#d91a1a}-8.76\%}$
test_membership_stacked_nested_leaf_last	40.5250μs	5.7609μs	173.5836 KOps/s	185.8135 KOps/s	$\textbf{\color{#d91a1a}-6.58\%}$
test_nested_getleaf	41.1160μs	10.5584μs	94.7113 KOps/s	90.0767 KOps/s	$\textbf{\color{#35bf28}+5.15\%}$
test_nested_get	41.7680μs	10.1035μs	98.9754 KOps/s	96.7879 KOps/s	$\color{#35bf28}+2.26\%$
test_stacked_getleaf	31.1580μs	10.5290μs	94.9761 KOps/s	91.6780 KOps/s	$\color{#35bf28}+3.60\%$
test_stacked_get	48.7310μs	10.0061μs	99.9391 KOps/s	97.4725 KOps/s	$\color{#35bf28}+2.53\%$
test_nested_getitemleaf	40.5750μs	11.1149μs	89.9692 KOps/s	85.6543 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_nested_getitem	52.0270μs	10.1638μs	98.3887 KOps/s	92.6856 KOps/s	$\textbf{\color{#35bf28}+6.15\%}$
test_stacked_getitemleaf	45.5450μs	10.9513μs	91.3138 KOps/s	87.8238 KOps/s	$\color{#35bf28}+3.97\%$
test_stacked_getitem	33.4220μs	10.2581μs	97.4843 KOps/s	94.8254 KOps/s	$\color{#35bf28}+2.80\%$
test_lock_nested	45.3264ms	0.4020ms	2.4878 KOps/s	2.8712 KOps/s	$\textbf{\color{#d91a1a}-13.35\%}$
test_lock_stack_nested	0.4732ms	0.3079ms	3.2477 KOps/s	3.2439 KOps/s	$\color{#35bf28}+0.11\%$
test_unlock_nested	0.7559ms	0.3527ms	2.8351 KOps/s	2.4989 KOps/s	$\textbf{\color{#35bf28}+13.45\%}$
test_unlock_stack_nested	0.5101ms	0.3158ms	3.1665 KOps/s	3.1604 KOps/s	$\color{#35bf28}+0.19\%$
test_flatten_speed	0.1972ms	97.0291μs	10.3062 KOps/s	10.3264 KOps/s	$\color{#d91a1a}-0.20\%$
test_unflatten_speed	0.7185ms	0.4105ms	2.4361 KOps/s	2.3394 KOps/s	$\color{#35bf28}+4.13\%$
test_common_ops	1.4378ms	0.6918ms	1.4455 KOps/s	1.4764 KOps/s	$\color{#d91a1a}-2.09\%$
test_creation	27.9520μs	1.9220μs	520.2855 KOps/s	532.4833 KOps/s	$\color{#d91a1a}-2.29\%$
test_creation_empty	28.1720μs	9.9139μs	100.8683 KOps/s	112.9277 KOps/s	$\textbf{\color{#d91a1a}-10.68\%}$
test_creation_nested_1	40.2440μs	12.6404μs	79.1112 KOps/s	85.2265 KOps/s	$\textbf{\color{#d91a1a}-7.18\%}$
test_creation_nested_2	39.3230μs	16.0729μs	62.2167 KOps/s	66.9231 KOps/s	$\textbf{\color{#d91a1a}-7.03\%}$
test_clone	89.8170μs	13.2114μs	75.6924 KOps/s	74.4858 KOps/s	$\color{#35bf28}+1.62\%$
test_getitem[int]	30.8070μs	11.4969μs	86.9799 KOps/s	86.1730 KOps/s	$\color{#35bf28}+0.94\%$
test_getitem[slice_int]	59.9210μs	22.9777μs	43.5206 KOps/s	42.6225 KOps/s	$\color{#35bf28}+2.11\%$
test_getitem[range]	78.6860μs	58.2080μs	17.1798 KOps/s	16.5756 KOps/s	$\color{#35bf28}+3.65\%$
test_getitem[tuple]	51.0240μs	18.8451μs	53.0641 KOps/s	50.5683 KOps/s	$\color{#35bf28}+4.94\%$
test_getitem[list]	0.1056ms	40.6501μs	24.6002 KOps/s	24.1126 KOps/s	$\color{#35bf28}+2.02\%$
test_setitem_dim[int]	60.5230μs	33.5250μs	29.8285 KOps/s	29.8343 KOps/s	$\color{#d91a1a}-0.02\%$
test_setitem_dim[slice_int]	97.8810μs	59.9365μs	16.6843 KOps/s	15.4304 KOps/s	$\textbf{\color{#35bf28}+8.13\%}$
test_setitem_dim[range]	0.1341ms	83.3738μs	11.9942 KOps/s	11.1501 KOps/s	$\textbf{\color{#35bf28}+7.57\%}$
test_setitem_dim[tuple]	80.5400μs	48.2393μs	20.7300 KOps/s	19.8883 KOps/s	$\color{#35bf28}+4.23\%$
test_setitem	56.9160μs	19.9282μs	50.1800 KOps/s	52.9062 KOps/s	$\textbf{\color{#d91a1a}-5.15\%}$
test_set	52.8180μs	19.4589μs	51.3904 KOps/s	54.0419 KOps/s	$\color{#d91a1a}-4.91\%$
test_set_shared	3.0270ms	0.1414ms	7.0717 KOps/s	6.9235 KOps/s	$\color{#35bf28}+2.14\%$
test_update	0.1342ms	21.1766μs	47.2218 KOps/s	51.0061 KOps/s	$\textbf{\color{#d91a1a}-7.42\%}$
test_update_nested	75.7500μs	29.5114μs	33.8852 KOps/s	36.0717 KOps/s	$\textbf{\color{#d91a1a}-6.06\%}$
test_update__nested	60.5730μs	25.3453μs	39.4551 KOps/s	39.0802 KOps/s	$\color{#35bf28}+0.96\%$
test_set_nested	58.7790μs	21.5264μs	46.4545 KOps/s	48.3434 KOps/s	$\color{#d91a1a}-3.91\%$
test_set_nested_new	55.7840μs	25.1563μs	39.7514 KOps/s	39.9007 KOps/s	$\color{#d91a1a}-0.37\%$
test_select	1.0543ms	40.5702μs	24.6486 KOps/s	24.7371 KOps/s	$\color{#d91a1a}-0.36\%$
test_select_nested	0.1126ms	59.3928μs	16.8371 KOps/s	16.3309 KOps/s	$\color{#35bf28}+3.10\%$
test_exclude_nested	0.2018ms	0.1196ms	8.3612 KOps/s	8.0589 KOps/s	$\color{#35bf28}+3.75\%$
test_empty[True]	0.6177ms	0.3878ms	2.5789 KOps/s	2.5015 KOps/s	$\color{#35bf28}+3.09\%$
test_empty[False]	9.3416μs	1.1444μs	873.8016 KOps/s	861.4894 KOps/s	$\color{#35bf28}+1.43\%$
test_unbind_speed	0.4844ms	0.2613ms	3.8267 KOps/s	3.8765 KOps/s	$\color{#d91a1a}-1.28\%$
test_unbind_speed_stack0	0.3930ms	0.2557ms	3.9110 KOps/s	3.9671 KOps/s	$\color{#d91a1a}-1.41\%$
test_unbind_speed_stack1	64.6638ms	0.7585ms	1.3183 KOps/s	1.3097 KOps/s	$\color{#35bf28}+0.66\%$
test_split	66.1914ms	1.5844ms	631.1513 Ops/s	618.3391 Ops/s	$\color{#35bf28}+2.07\%$
test_chunk	60.9171ms	1.5793ms	633.1722 Ops/s	618.5913 Ops/s	$\color{#35bf28}+2.36\%$
test_creation[device0]	3.5171ms	86.2278μs	11.5972 KOps/s	11.6813 KOps/s	$\color{#d91a1a}-0.72\%$
test_creation_from_tensor	0.1680ms	85.6406μs	11.6767 KOps/s	11.5115 KOps/s	$\color{#35bf28}+1.44\%$
test_add_one[memmap_tensor0]	60.9830μs	5.2965μs	188.8053 KOps/s	181.1228 KOps/s	$\color{#35bf28}+4.24\%$
test_contiguous[memmap_tensor0]	6.1510μs	0.6364μs	1.5714 MOps/s	1.5520 MOps/s	$\color{#35bf28}+1.25\%$
test_stack[memmap_tensor0]	21.7900μs	3.5177μs	284.2772 KOps/s	276.9000 KOps/s	$\color{#35bf28}+2.66\%$
test_memmaptd_index	1.0506ms	0.2508ms	3.9869 KOps/s	3.8931 KOps/s	$\color{#35bf28}+2.41\%$
test_memmaptd_index_astensor	0.5309ms	0.3243ms	3.0832 KOps/s	3.0569 KOps/s	$\color{#35bf28}+0.86\%$
test_memmaptd_index_op	0.9303ms	0.5912ms	1.6914 KOps/s	1.7049 KOps/s	$\color{#d91a1a}-0.79\%$
test_serialize_model	0.1698s	0.1155s	8.6578 Ops/s	8.3599 Ops/s	$\color{#35bf28}+3.56\%$
test_serialize_model_pickle	0.4510s	0.3786s	2.6411 Ops/s	2.6549 Ops/s	$\color{#d91a1a}-0.52\%$
test_serialize_weights	0.1662s	0.1116s	8.9569 Ops/s	8.8279 Ops/s	$\color{#35bf28}+1.46\%$
test_serialize_weights_returnearly	0.1914s	0.1346s	7.4275 Ops/s	7.0918 Ops/s	$\color{#35bf28}+4.73\%$
test_serialize_weights_pickle	1.0556s	0.6101s	1.6391 Ops/s	2.3188 Ops/s	$\textbf{\color{#d91a1a}-29.31\%}$
test_serialize_weights_filesystem	97.0289ms	92.0905ms	10.8589 Ops/s	10.4861 Ops/s	$\color{#35bf28}+3.56\%$
test_serialize_model_filesystem	0.1017s	92.5041ms	10.8103 Ops/s	10.0789 Ops/s	$\textbf{\color{#35bf28}+7.26\%}$
test_reshape_pytree	61.9250μs	25.3895μs	39.3864 KOps/s	38.8305 KOps/s	$\color{#35bf28}+1.43\%$
test_reshape_td	0.1052ms	34.3397μs	29.1208 KOps/s	28.5553 KOps/s	$\color{#35bf28}+1.98\%$
test_view_pytree	77.2430μs	24.9930μs	40.0112 KOps/s	39.1539 KOps/s	$\color{#35bf28}+2.19\%$
test_view_td	88.8850μs	38.7885μs	25.7808 KOps/s	25.8699 KOps/s	$\color{#d91a1a}-0.34\%$
test_unbind_pytree	70.7510μs	29.5640μs	33.8250 KOps/s	34.3121 KOps/s	$\color{#d91a1a}-1.42\%$
test_unbind_td	0.4213ms	38.0848μs	26.2572 KOps/s	26.1034 KOps/s	$\color{#35bf28}+0.59\%$
test_split_pytree	61.3240μs	29.4742μs	33.9280 KOps/s	34.4032 KOps/s	$\color{#d91a1a}-1.38\%$
test_split_td	0.1262ms	40.9813μs	24.4014 KOps/s	24.7333 KOps/s	$\color{#d91a1a}-1.34\%$
test_add_pytree	0.1066ms	33.5743μs	29.7847 KOps/s	28.8130 KOps/s	$\color{#35bf28}+3.37\%$
test_add_td	0.1195ms	52.7081μs	18.9724 KOps/s	18.4598 KOps/s	$\color{#35bf28}+2.78\%$
test_distributed	0.2137ms	99.3720μs	10.0632 KOps/s	9.7127 KOps/s	$\color{#35bf28}+3.61\%$
test_tdmodule	31.2480μs	16.5974μs	60.2505 KOps/s	59.1349 KOps/s	$\color{#35bf28}+1.89\%$
test_tdmodule_dispatch	61.8450μs	33.0285μs	30.2769 KOps/s	30.3330 KOps/s	$\color{#d91a1a}-0.19\%$
test_tdseq	45.9650μs	19.3894μs	51.5745 KOps/s	50.0548 KOps/s	$\color{#35bf28}+3.04\%$
test_tdseq_dispatch	55.1620μs	38.3679μs	26.0634 KOps/s	25.8408 KOps/s	$\color{#35bf28}+0.86\%$
test_instantiation_functorch	1.9339ms	1.2959ms	771.6425 Ops/s	748.3052 Ops/s	$\color{#35bf28}+3.12\%$
test_instantiation_td	1.4512ms	1.0105ms	989.5759 Ops/s	986.3358 Ops/s	$\color{#35bf28}+0.33\%$
test_exec_functorch	0.3158ms	0.1557ms	6.4233 KOps/s	6.1480 KOps/s	$\color{#35bf28}+4.48\%$
test_exec_functional_call	0.3035ms	0.1490ms	6.7103 KOps/s	6.6919 KOps/s	$\color{#35bf28}+0.27\%$
test_exec_td	0.2833ms	0.1447ms	6.9089 KOps/s	6.6817 KOps/s	$\color{#35bf28}+3.40\%$
test_exec_td_decorator	0.8267ms	0.2195ms	4.5560 KOps/s	4.5039 KOps/s	$\color{#35bf28}+1.16\%$
test_vmap_mlp_speed[True-True]	0.6533ms	0.4798ms	2.0841 KOps/s	2.0552 KOps/s	$\color{#35bf28}+1.40\%$
test_vmap_mlp_speed[True-False]	0.8525ms	0.4789ms	2.0881 KOps/s	2.0684 KOps/s	$\color{#35bf28}+0.95\%$
test_vmap_mlp_speed[False-True]	0.7735ms	0.3943ms	2.5364 KOps/s	2.4921 KOps/s	$\color{#35bf28}+1.78\%$
test_vmap_mlp_speed[False-False]	0.6591ms	0.3943ms	2.5364 KOps/s	2.5224 KOps/s	$\color{#35bf28}+0.55\%$
test_vmap_mlp_speed_decorator[True-True]	0.9493ms	0.5502ms	1.8174 KOps/s	1.8130 KOps/s	$\color{#35bf28}+0.24\%$
test_vmap_mlp_speed_decorator[True-False]	0.7585ms	0.5493ms	1.8205 KOps/s	1.8187 KOps/s	$\color{#35bf28}+0.09\%$
test_vmap_mlp_speed_decorator[False-True]	0.7350ms	0.4559ms	2.1936 KOps/s	2.1782 KOps/s	$\color{#35bf28}+0.71\%$
test_vmap_mlp_speed_decorator[False-False]	0.8506ms	0.4570ms	2.1884 KOps/s	2.1815 KOps/s	$\color{#35bf28}+0.31\%$
test_to_module_speed[True]	2.5338ms	1.6767ms	596.4135 Ops/s	583.2646 Ops/s	$\color{#35bf28}+2.25\%$
test_to_module_speed[False]	1.9765ms	1.6425ms	608.8304 Ops/s	596.7746 Ops/s	$\color{#35bf28}+2.02\%$
test_tc_init	56.9150μs	27.1345μs	36.8535 KOps/s	39.9541 KOps/s	$\textbf{\color{#d91a1a}-7.76\%}$
test_tc_init_nested	0.1111ms	55.2586μs	18.0967 KOps/s	20.2900 KOps/s	$\textbf{\color{#d91a1a}-10.81\%}$
test_tc_first_layer_tensor	5.8211μs	0.6804μs	1.4697 MOps/s	1.4202 MOps/s	$\color{#35bf28}+3.49\%$
test_tc_first_layer_nontensor	1.8845μs	0.6528μs	1.5318 MOps/s	1.4536 MOps/s	$\textbf{\color{#35bf28}+5.38\%}$
test_tc_second_layer_tensor	28.3930μs	1.8269μs	547.3611 KOps/s	531.8053 KOps/s	$\color{#35bf28}+2.93\%$
test_tc_second_layer_nontensor	13.3050μs	1.6076μs	622.0620 KOps/s	649.3007 KOps/s	$\color{#d91a1a}-4.20\%$
test_unbind	5.5329ms	5.1715ms	193.3659 Ops/s	142.4122 Ops/s	$\textbf{\color{#35bf28}+35.78\%}$
test_full_like	17.1019ms	11.5154ms	86.8406 Ops/s	130.2399 Ops/s	$\textbf{\color{#d91a1a}-33.32\%}$
test_zeros_like	11.4868ms	5.9739ms	167.3958 Ops/s	173.4469 Ops/s	$\color{#d91a1a}-3.49\%$
test_ones_like	13.9096ms	6.5227ms	153.3097 Ops/s	161.3338 Ops/s	$\color{#d91a1a}-4.97\%$
test_clone	14.2174ms	7.5964ms	131.6411 Ops/s	130.4626 Ops/s	$\color{#35bf28}+0.90\%$
test_squeeze	88.3140μs	13.7668μs	72.6387 KOps/s	69.9679 KOps/s	$\color{#35bf28}+3.82\%$
test_unsqueeze	0.1248ms	67.8432μs	14.7399 KOps/s	14.2068 KOps/s	$\color{#35bf28}+3.75\%$
test_split	0.4422ms	0.1089ms	9.1835 KOps/s	8.7282 KOps/s	$\textbf{\color{#35bf28}+5.22\%}$
test_permute	0.3123ms	0.1350ms	7.4093 KOps/s	7.2218 KOps/s	$\color{#35bf28}+2.60\%$
test_stack	28.9768ms	21.3566ms	46.8240 Ops/s	46.0072 Ops/s	$\color{#35bf28}+1.78\%$
test_cat	27.0671ms	21.6524ms	46.1842 Ops/s	45.6588 Ops/s	$\color{#35bf28}+1.15\%$

[github-actions]

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

Total Benchmarks: 152. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.5217ms	13.5560μs	73.7681 KOps/s	73.2226 KOps/s	$\color{#35bf28}+0.74\%$
test_plain_set_stack_nested	32.0400μs	13.5663μs	73.7119 KOps/s	72.7300 KOps/s	$\color{#35bf28}+1.35\%$
test_plain_set_nested_inplace	84.3510μs	14.7286μs	67.8950 KOps/s	66.9349 KOps/s	$\color{#35bf28}+1.43\%$
test_plain_set_stack_nested_inplace	55.2710μs	14.8757μs	67.2236 KOps/s	66.6207 KOps/s	$\color{#35bf28}+0.90\%$
test_items	0.1968ms	4.7042μs	212.5772 KOps/s	208.8366 KOps/s	$\color{#35bf28}+1.79\%$
test_items_nested	0.3864ms	0.3420ms	2.9239 KOps/s	2.9275 KOps/s	$\color{#d91a1a}-0.12\%$
test_items_nested_locked	0.3766ms	0.3435ms	2.9111 KOps/s	2.8583 KOps/s	$\color{#35bf28}+1.85\%$
test_items_nested_leaf	0.1118ms	83.5191μs	11.9733 KOps/s	11.9070 KOps/s	$\color{#35bf28}+0.56\%$
test_items_stack_nested	0.4191ms	0.3447ms	2.9011 KOps/s	2.8622 KOps/s	$\color{#35bf28}+1.36\%$
test_items_stack_nested_leaf	0.1085ms	84.9435μs	11.7725 KOps/s	11.7296 KOps/s	$\color{#35bf28}+0.37\%$
test_items_stack_nested_locked	0.3773ms	0.3484ms	2.8700 KOps/s	2.9102 KOps/s	$\color{#d91a1a}-1.38\%$
test_keys	21.8900μs	4.4279μs	225.8406 KOps/s	227.8274 KOps/s	$\color{#d91a1a}-0.87\%$
test_keys_nested	0.1402ms	69.9763μs	14.2905 KOps/s	14.3798 KOps/s	$\color{#d91a1a}-0.62\%$
test_keys_nested_locked	0.7650ms	74.6741μs	13.3915 KOps/s	13.4234 KOps/s	$\color{#d91a1a}-0.24\%$
test_keys_nested_leaf	87.3810μs	60.4702μs	16.5371 KOps/s	16.5495 KOps/s	$\color{#d91a1a}-0.08\%$
test_keys_stack_nested	92.8010μs	69.0111μs	14.4904 KOps/s	14.5483 KOps/s	$\color{#d91a1a}-0.40\%$
test_keys_stack_nested_leaf	98.1420μs	60.5929μs	16.5036 KOps/s	16.6989 KOps/s	$\color{#d91a1a}-1.17\%$
test_keys_stack_nested_locked	0.1021ms	73.9847μs	13.5163 KOps/s	13.5390 KOps/s	$\color{#d91a1a}-0.17\%$
test_values	29.2037μs	1.8609μs	537.3657 KOps/s	542.9509 KOps/s	$\color{#d91a1a}-1.03\%$
test_values_nested	66.8910μs	37.2472μs	26.8477 KOps/s	27.2059 KOps/s	$\color{#d91a1a}-1.32\%$
test_values_nested_locked	62.3210μs	39.1262μs	25.5583 KOps/s	25.9844 KOps/s	$\color{#d91a1a}-1.64\%$
test_values_nested_leaf	65.1010μs	33.0182μs	30.2864 KOps/s	30.5088 KOps/s	$\color{#d91a1a}-0.73\%$
test_values_stack_nested	57.1110μs	38.2060μs	26.1739 KOps/s	26.3962 KOps/s	$\color{#d91a1a}-0.84\%$
test_values_stack_nested_leaf	69.7100μs	33.9893μs	29.4211 KOps/s	29.7110 KOps/s	$\color{#d91a1a}-0.98\%$
test_values_stack_nested_locked	80.8010μs	39.8312μs	25.1060 KOps/s	25.4313 KOps/s	$\color{#d91a1a}-1.28\%$
test_membership	4.2257μs	0.7696μs	1.2993 MOps/s	1.3056 MOps/s	$\color{#d91a1a}-0.48\%$
test_membership_nested	33.9700μs	2.6899μs	371.7604 KOps/s	371.0268 KOps/s	$\color{#35bf28}+0.20\%$
test_membership_nested_leaf	17.4300μs	2.6907μs	371.6456 KOps/s	368.0398 KOps/s	$\color{#35bf28}+0.98\%$
test_membership_stacked_nested	47.7810μs	2.6749μs	373.8515 KOps/s	367.5661 KOps/s	$\color{#35bf28}+1.71\%$
test_membership_stacked_nested_leaf	19.5410μs	2.6917μs	371.5164 KOps/s	367.8240 KOps/s	$\color{#35bf28}+1.00\%$
test_membership_nested_last	44.3700μs	3.2735μs	305.4821 KOps/s	305.6602 KOps/s	$\color{#d91a1a}-0.06\%$
test_membership_nested_leaf_last	21.4700μs	3.2640μs	306.3759 KOps/s	305.7440 KOps/s	$\color{#35bf28}+0.21\%$
test_membership_stacked_nested_last	20.0100μs	6.4383μs	155.3207 KOps/s	209.0258 KOps/s	$\textbf{\color{#d91a1a}-25.69\%}$
test_membership_stacked_nested_leaf_last	83.9800μs	6.5234μs	153.2944 KOps/s	208.8175 KOps/s	$\textbf{\color{#d91a1a}-26.59\%}$
test_nested_getleaf	54.4410μs	8.5335μs	117.1855 KOps/s	116.3205 KOps/s	$\color{#35bf28}+0.74\%$
test_nested_get	23.7500μs	8.0057μs	124.9111 KOps/s	124.6785 KOps/s	$\color{#35bf28}+0.19\%$
test_stacked_getleaf	23.9600μs	8.5935μs	116.3669 KOps/s	116.3375 KOps/s	$\color{#35bf28}+0.03\%$
test_stacked_get	39.1700μs	8.0289μs	124.5499 KOps/s	124.4842 KOps/s	$\color{#35bf28}+0.05\%$
test_nested_getitemleaf	23.8490μs	8.6705μs	115.3329 KOps/s	114.2321 KOps/s	$\color{#35bf28}+0.96\%$
test_nested_getitem	81.5910μs	8.1310μs	122.9863 KOps/s	122.7134 KOps/s	$\color{#35bf28}+0.22\%$
test_stacked_getitemleaf	37.3610μs	8.7320μs	114.5219 KOps/s	113.7947 KOps/s	$\color{#35bf28}+0.64\%$
test_stacked_getitem	41.9010μs	8.1391μs	122.8644 KOps/s	122.4116 KOps/s	$\color{#35bf28}+0.37\%$
test_lock_nested	60.7644ms	0.4330ms	2.3092 KOps/s	2.3175 KOps/s	$\color{#d91a1a}-0.36\%$
test_lock_stack_nested	0.3734ms	0.3194ms	3.1307 KOps/s	3.0789 KOps/s	$\color{#35bf28}+1.68\%$
test_unlock_nested	0.8795ms	0.3704ms	2.6996 KOps/s	2.6815 KOps/s	$\color{#35bf28}+0.67\%$
test_unlock_stack_nested	0.3970ms	0.3251ms	3.0762 KOps/s	3.0180 KOps/s	$\color{#35bf28}+1.93\%$
test_flatten_speed	0.2583ms	0.1028ms	9.7264 KOps/s	9.7098 KOps/s	$\color{#35bf28}+0.17\%$
test_unflatten_speed	0.3449ms	0.3048ms	3.2812 KOps/s	3.2941 KOps/s	$\color{#d91a1a}-0.39\%$
test_common_ops	1.2318ms	0.6198ms	1.6134 KOps/s	1.6043 KOps/s	$\color{#35bf28}+0.57\%$
test_creation	34.2800μs	1.7450μs	573.0515 KOps/s	585.1280 KOps/s	$\color{#d91a1a}-2.06\%$
test_creation_empty	23.4600μs	9.2622μs	107.9660 KOps/s	105.5154 KOps/s	$\color{#35bf28}+2.32\%$
test_creation_nested_1	0.1956ms	11.1398μs	89.7681 KOps/s	87.8252 KOps/s	$\color{#35bf28}+2.21\%$
test_creation_nested_2	45.1210μs	13.3537μs	74.8859 KOps/s	73.4515 KOps/s	$\color{#35bf28}+1.95\%$
test_clone	57.1000μs	12.0563μs	82.9440 KOps/s	78.5781 KOps/s	$\textbf{\color{#35bf28}+5.56\%}$
test_getitem[int]	31.0200μs	11.6920μs	85.5287 KOps/s	86.1695 KOps/s	$\color{#d91a1a}-0.74\%$
test_getitem[slice_int]	40.3700μs	21.9867μs	45.4821 KOps/s	45.1785 KOps/s	$\color{#35bf28}+0.67\%$
test_getitem[range]	68.1500μs	49.9384μs	20.0247 KOps/s	19.8971 KOps/s	$\color{#35bf28}+0.64\%$
test_getitem[tuple]	49.2000μs	19.6476μs	50.8967 KOps/s	50.3051 KOps/s	$\color{#35bf28}+1.18\%$
test_getitem[list]	0.1682ms	36.0000μs	27.7778 KOps/s	28.3027 KOps/s	$\color{#d91a1a}-1.85\%$
test_setitem_dim[int]	48.0100μs	31.0527μs	32.2033 KOps/s	30.3619 KOps/s	$\textbf{\color{#35bf28}+6.06\%}$
test_setitem_dim[slice_int]	0.1007ms	54.1339μs	18.4727 KOps/s	18.6854 KOps/s	$\color{#d91a1a}-1.14\%$
test_setitem_dim[range]	95.3810μs	74.3668μs	13.4469 KOps/s	14.0912 KOps/s	$\color{#d91a1a}-4.57\%$
test_setitem_dim[tuple]	66.2910μs	47.9266μs	20.8652 KOps/s	20.9598 KOps/s	$\color{#d91a1a}-0.45\%$
test_setitem	59.4110μs	17.5051μs	57.1264 KOps/s	55.1142 KOps/s	$\color{#35bf28}+3.65\%$
test_set	52.0210μs	16.8381μs	59.3893 KOps/s	56.8277 KOps/s	$\color{#35bf28}+4.51\%$
test_set_shared	1.1827ms	0.1009ms	9.9126 KOps/s	9.8573 KOps/s	$\color{#35bf28}+0.56\%$
test_update	0.1022ms	19.2456μs	51.9598 KOps/s	49.2714 KOps/s	$\textbf{\color{#35bf28}+5.46\%}$
test_update_nested	0.1064ms	24.9637μs	40.0581 KOps/s	39.0831 KOps/s	$\color{#35bf28}+2.49\%$
test_update__nested	62.6300μs	23.4131μs	42.7112 KOps/s	38.6984 KOps/s	$\textbf{\color{#35bf28}+10.37\%}$
test_set_nested	0.1926ms	18.1778μs	55.0121 KOps/s	49.7178 KOps/s	$\textbf{\color{#35bf28}+10.65\%}$
test_set_nested_new	0.1903ms	21.0817μs	47.4345 KOps/s	42.4422 KOps/s	$\textbf{\color{#35bf28}+11.76\%}$
test_select	0.2396ms	35.4557μs	28.2042 KOps/s	27.4638 KOps/s	$\color{#35bf28}+2.70\%$
test_select_nested	0.8065ms	56.5990μs	17.6682 KOps/s	17.5070 KOps/s	$\color{#35bf28}+0.92\%$
test_exclude_nested	0.1334ms	0.1130ms	8.8522 KOps/s	8.7727 KOps/s	$\color{#35bf28}+0.91\%$
test_empty[True]	0.5593ms	0.3593ms	2.7832 KOps/s	2.8093 KOps/s	$\color{#d91a1a}-0.93\%$
test_empty[False]	6.7780μs	0.9857μs	1.0146 MOps/s	1.0284 MOps/s	$\color{#d91a1a}-1.34\%$
test_to	0.1048ms	78.3895μs	12.7568 KOps/s	12.5502 KOps/s	$\color{#35bf28}+1.65\%$
test_to_nonblocking	0.2673ms	68.3490μs	14.6308 KOps/s	15.7962 KOps/s	$\textbf{\color{#d91a1a}-7.38\%}$
test_unbind_speed	0.3433ms	0.2858ms	3.4987 KOps/s	3.4887 KOps/s	$\color{#35bf28}+0.29\%$
test_unbind_speed_stack0	0.4265ms	0.2846ms	3.5140 KOps/s	3.4631 KOps/s	$\color{#35bf28}+1.47\%$
test_unbind_speed_stack1	77.2376ms	0.8659ms	1.1549 KOps/s	1.1776 KOps/s	$\color{#d91a1a}-1.93\%$
test_split	78.5357ms	1.7487ms	571.8573 Ops/s	564.4338 Ops/s	$\color{#35bf28}+1.32\%$
test_chunk	77.7322ms	1.7398ms	574.7784 Ops/s	564.8296 Ops/s	$\color{#35bf28}+1.76\%$
test_creation[device0]	0.1859ms	60.1199μs	16.6334 KOps/s	16.6008 KOps/s	$\color{#35bf28}+0.20\%$
test_creation_from_tensor	0.1922ms	56.0172μs	17.8517 KOps/s	16.3600 KOps/s	$\textbf{\color{#35bf28}+9.12\%}$
test_add_one[memmap_tensor0]	64.5610μs	7.3190μs	136.6301 KOps/s	127.6353 KOps/s	$\textbf{\color{#35bf28}+7.05\%}$
test_contiguous[memmap_tensor0]	34.1900μs	0.6936μs	1.4417 MOps/s	1.4231 MOps/s	$\color{#35bf28}+1.30\%$
test_stack[memmap_tensor0]	36.7610μs	4.9107μs	203.6354 KOps/s	198.5167 KOps/s	$\color{#35bf28}+2.58\%$
test_memmaptd_index	1.0948ms	0.2994ms	3.3405 KOps/s	3.3251 KOps/s	$\color{#35bf28}+0.46\%$
test_memmaptd_index_astensor	0.7423ms	0.3740ms	2.6740 KOps/s	2.6709 KOps/s	$\color{#35bf28}+0.12\%$
test_memmaptd_index_op	1.2315ms	0.6889ms	1.4516 KOps/s	1.4063 KOps/s	$\color{#35bf28}+3.22\%$
test_serialize_model	0.1853s	0.1140s	8.7699 Ops/s	8.5538 Ops/s	$\color{#35bf28}+2.53\%$
test_serialize_model_pickle	1.3497s	1.2353s	0.8095 Ops/s	0.8086 Ops/s	$\color{#35bf28}+0.11\%$
test_serialize_weights	0.1827s	0.1106s	9.0424 Ops/s	8.5971 Ops/s	$\textbf{\color{#35bf28}+5.18\%}$
test_serialize_weights_returnearly	0.2847s	0.1088s	9.1941 Ops/s	11.9003 Ops/s	$\textbf{\color{#d91a1a}-22.74\%}$
test_serialize_weights_pickle	1.3500s	1.2484s	0.8010 Ops/s	0.8008 Ops/s	$\color{#35bf28}+0.03\%$
test_reshape_pytree	0.1285ms	27.5747μs	36.2651 KOps/s	35.9773 KOps/s	$\color{#35bf28}+0.80\%$
test_reshape_td	0.1070ms	32.4813μs	30.7870 KOps/s	29.9408 KOps/s	$\color{#35bf28}+2.83\%$
test_view_pytree	0.1255ms	27.3900μs	36.5097 KOps/s	36.4922 KOps/s	$\color{#35bf28}+0.05\%$
test_view_td	0.1596ms	36.9585μs	27.0574 KOps/s	26.2608 KOps/s	$\color{#35bf28}+3.03\%$
test_unbind_pytree	0.1296ms	33.6816μs	29.6898 KOps/s	29.7211 KOps/s	$\color{#d91a1a}-0.11\%$
test_unbind_td	0.4424ms	44.4304μs	22.5071 KOps/s	22.5704 KOps/s	$\color{#d91a1a}-0.28\%$
test_split_pytree	0.1338ms	37.1370μs	26.9273 KOps/s	27.5545 KOps/s	$\color{#d91a1a}-2.28\%$
test_split_td	0.4652ms	41.7717μs	23.9396 KOps/s	23.9644 KOps/s	$\color{#d91a1a}-0.10\%$
test_add_pytree	67.0210μs	40.3713μs	24.7701 KOps/s	24.6371 KOps/s	$\color{#35bf28}+0.54\%$
test_add_td	87.3110μs	51.0712μs	19.5805 KOps/s	17.9572 KOps/s	$\textbf{\color{#35bf28}+9.04\%}$
test_distributed	2.2232ms	88.6635μs	11.2786 KOps/s	14.8511 KOps/s	$\textbf{\color{#d91a1a}-24.06\%}$
test_tdmodule	35.8900μs	15.5720μs	64.2180 KOps/s	65.3039 KOps/s	$\color{#d91a1a}-1.66\%$
test_tdmodule_dispatch	0.1154ms	30.7888μs	32.4793 KOps/s	32.9742 KOps/s	$\color{#d91a1a}-1.50\%$
test_tdseq	42.9700μs	17.9109μs	55.8318 KOps/s	55.9178 KOps/s	$\color{#d91a1a}-0.15\%$
test_tdseq_dispatch	63.8490μs	34.4121μs	29.0596 KOps/s	28.9391 KOps/s	$\color{#35bf28}+0.42\%$
test_instantiation_functorch	1.7582ms	1.5754ms	634.7678 Ops/s	627.2571 Ops/s	$\color{#35bf28}+1.20\%$
test_instantiation_td	1.7757ms	1.0982ms	910.6199 Ops/s	910.6748 Ops/s	$-0.01\%$
test_exec_functorch	0.2297ms	0.1580ms	6.3285 KOps/s	6.2820 KOps/s	$\color{#35bf28}+0.74\%$
test_exec_functional_call	0.2762ms	0.1479ms	6.7631 KOps/s	6.7110 KOps/s	$\color{#35bf28}+0.78\%$
test_exec_td	0.2518ms	0.1460ms	6.8493 KOps/s	6.8392 KOps/s	$\color{#35bf28}+0.15\%$
test_exec_td_decorator	0.5153ms	0.2203ms	4.5386 KOps/s	4.5225 KOps/s	$\color{#35bf28}+0.36\%$
test_vmap_mlp_speed[True-True]	0.7750ms	0.6095ms	1.6406 KOps/s	1.6440 KOps/s	$\color{#d91a1a}-0.21\%$
test_vmap_mlp_speed[True-False]	0.7481ms	0.6109ms	1.6371 KOps/s	1.6408 KOps/s	$\color{#d91a1a}-0.23\%$
test_vmap_mlp_speed[False-True]	0.7208ms	0.5416ms	1.8464 KOps/s	1.8676 KOps/s	$\color{#d91a1a}-1.13\%$
test_vmap_mlp_speed[False-False]	0.7395ms	0.5555ms	1.8003 KOps/s	1.8676 KOps/s	$\color{#d91a1a}-3.60\%$
test_vmap_mlp_speed_decorator[True-True]	1.4679ms	0.6756ms	1.4802 KOps/s	1.4883 KOps/s	$\color{#d91a1a}-0.54\%$
test_vmap_mlp_speed_decorator[True-False]	0.8335ms	0.6736ms	1.4846 KOps/s	1.4888 KOps/s	$\color{#d91a1a}-0.28\%$
test_vmap_mlp_speed_decorator[False-True]	0.7829ms	0.6058ms	1.6507 KOps/s	1.6882 KOps/s	$\color{#d91a1a}-2.22\%$
test_vmap_mlp_speed_decorator[False-False]	0.7601ms	0.6022ms	1.6607 KOps/s	1.6872 KOps/s	$\color{#d91a1a}-1.57\%$
test_vmap_transformer_speed[True-True]	8.3279ms	8.0937ms	123.5526 Ops/s	123.5005 Ops/s	$\color{#35bf28}+0.04\%$
test_vmap_transformer_speed[True-False]	8.2804ms	8.0884ms	123.6343 Ops/s	123.4115 Ops/s	$\color{#35bf28}+0.18\%$
test_vmap_transformer_speed[False-True]	8.2538ms	8.0244ms	124.6204 Ops/s	124.3395 Ops/s	$\color{#35bf28}+0.23\%$
test_vmap_transformer_speed[False-False]	8.3520ms	8.0291ms	124.5476 Ops/s	124.5580 Ops/s	$-0.01\%$
test_vmap_transformer_speed_decorator[True-True]	20.4403ms	19.6568ms	50.8730 Ops/s	51.1538 Ops/s	$\color{#d91a1a}-0.55\%$
test_vmap_transformer_speed_decorator[True-False]	19.7691ms	19.5514ms	51.1473 Ops/s	51.2480 Ops/s	$\color{#d91a1a}-0.20\%$
test_vmap_transformer_speed_decorator[False-True]	20.5055ms	19.6029ms	51.0128 Ops/s	51.3648 Ops/s	$\color{#d91a1a}-0.69\%$
test_vmap_transformer_speed_decorator[False-False]	20.1571ms	19.5361ms	51.1872 Ops/s	51.3293 Ops/s	$\color{#d91a1a}-0.28\%$
test_to_module_speed[True]	2.1274ms	1.5760ms	634.5003 Ops/s	638.5452 Ops/s	$\color{#d91a1a}-0.63\%$
test_to_module_speed[False]	1.6585ms	1.5411ms	648.8972 Ops/s	648.6792 Ops/s	$\color{#35bf28}+0.03\%$
test_tc_init	49.9310μs	25.6962μs	38.9162 KOps/s	38.0081 KOps/s	$\color{#35bf28}+2.39\%$
test_tc_init_nested	0.1810ms	51.8082μs	19.3020 KOps/s	18.1248 KOps/s	$\textbf{\color{#35bf28}+6.49\%}$
test_tc_first_layer_tensor	1.8025μs	0.3638μs	2.7489 MOps/s	2.7396 MOps/s	$\color{#35bf28}+0.34\%$
test_tc_first_layer_nontensor	1.7225μs	0.3917μs	2.5527 MOps/s	2.5212 MOps/s	$\color{#35bf28}+1.25\%$
test_tc_second_layer_tensor	4.2182μs	0.9917μs	1.0084 MOps/s	988.0687 KOps/s	$\color{#35bf28}+2.06\%$
test_tc_second_layer_nontensor	3.9018μs	0.8376μs	1.1939 MOps/s	1.1933 MOps/s	$\color{#35bf28}+0.05\%$
test_unbind	5.4616ms	5.3136ms	188.1954 Ops/s	183.9971 Ops/s	$\color{#35bf28}+2.28\%$
test_full_like	15.7717ms	13.9141ms	71.8693 Ops/s	74.1829 Ops/s	$\color{#d91a1a}-3.12\%$
test_zeros_like	7.2870ms	7.0646ms	141.5518 Ops/s	126.9878 Ops/s	$\textbf{\color{#35bf28}+11.47\%}$
test_ones_like	7.3901ms	7.0809ms	141.2256 Ops/s	126.2048 Ops/s	$\textbf{\color{#35bf28}+11.90\%}$
test_clone	9.5250ms	8.9342ms	111.9297 Ops/s	103.8672 Ops/s	$\textbf{\color{#35bf28}+7.76\%}$
test_squeeze	0.1467ms	11.1905μs	89.3615 KOps/s	87.9753 KOps/s	$\color{#35bf28}+1.58\%$
test_unsqueeze	0.1915ms	66.8507μs	14.9587 KOps/s	15.5588 KOps/s	$\color{#d91a1a}-3.86\%$
test_split	0.1680ms	0.1043ms	9.5898 KOps/s	9.3596 KOps/s	$\color{#35bf28}+2.46\%$
test_permute	0.2014ms	0.1294ms	7.7292 KOps/s	7.5504 KOps/s	$\color{#35bf28}+2.37\%$
test_stack	28.4294ms	27.6159ms	36.2110 Ops/s	36.1031 Ops/s	$\color{#35bf28}+0.30\%$
test_cat	28.4572ms	27.6580ms	36.1559 Ops/s	35.6138 Ops/s	$\color{#35bf28}+1.52\%$