[BugFix] module hook fixes

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 126. Improved: $\large\color{#35bf28}24$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	42.0190μs	15.7652μs	63.4307 KOps/s	58.9446 KOps/s	$\textbf{\color{#35bf28}+7.61\%}$
test_plain_set_stack_nested	0.1768ms	0.1435ms	6.9678 KOps/s	6.9408 KOps/s	$\color{#35bf28}+0.39\%$
test_plain_set_nested_inplace	55.7220μs	18.0598μs	55.3715 KOps/s	51.5138 KOps/s	$\textbf{\color{#35bf28}+7.49\%}$
test_plain_set_stack_nested_inplace	0.3027ms	0.1776ms	5.6308 KOps/s	5.5897 KOps/s	$\color{#35bf28}+0.74\%$
test_items	20.3480μs	2.5208μs	396.7058 KOps/s	397.2808 KOps/s	$\color{#d91a1a}-0.14\%$
test_items_nested	0.4087ms	0.2728ms	3.6663 KOps/s	3.7352 KOps/s	$\color{#d91a1a}-1.84\%$
test_items_nested_locked	0.4528ms	0.2714ms	3.6841 KOps/s	3.6942 KOps/s	$\color{#d91a1a}-0.28\%$
test_items_nested_leaf	0.5617ms	0.1690ms	5.9171 KOps/s	5.9955 KOps/s	$\color{#d91a1a}-1.31\%$
test_items_stack_nested	1.5757ms	1.3224ms	756.2284 Ops/s	760.4450 Ops/s	$\color{#d91a1a}-0.55\%$
test_items_stack_nested_leaf	1.5288ms	1.1968ms	835.5718 Ops/s	846.1525 Ops/s	$\color{#d91a1a}-1.25\%$
test_items_stack_nested_locked	1.1745ms	0.8944ms	1.1181 KOps/s	1.1429 KOps/s	$\color{#d91a1a}-2.17\%$
test_keys	25.1870μs	3.8670μs	258.6002 KOps/s	260.5898 KOps/s	$\color{#d91a1a}-0.76\%$
test_keys_nested	1.8425ms	0.1457ms	6.8645 KOps/s	6.7186 KOps/s	$\color{#35bf28}+2.17\%$
test_keys_nested_locked	0.2913ms	0.1508ms	6.6309 KOps/s	6.6043 KOps/s	$\color{#35bf28}+0.40\%$
test_keys_nested_leaf	0.2494ms	0.1289ms	7.7566 KOps/s	7.6776 KOps/s	$\color{#35bf28}+1.03\%$
test_keys_stack_nested	1.8100ms	1.2967ms	771.1661 Ops/s	778.8856 Ops/s	$\color{#d91a1a}-0.99\%$
test_keys_stack_nested_leaf	1.5317ms	1.2727ms	785.7141 Ops/s	792.0405 Ops/s	$\color{#d91a1a}-0.80\%$
test_keys_stack_nested_locked	1.3023ms	0.8206ms	1.2186 KOps/s	1.2494 KOps/s	$\color{#d91a1a}-2.46\%$
test_values	10.1290μs	1.1732μs	852.3939 KOps/s	857.6014 KOps/s	$\color{#d91a1a}-0.61\%$
test_values_nested	0.1281ms	51.9470μs	19.2504 KOps/s	19.3097 KOps/s	$\color{#d91a1a}-0.31\%$
test_values_nested_locked	0.1225ms	52.3612μs	19.0981 KOps/s	19.3199 KOps/s	$\color{#d91a1a}-1.15\%$
test_values_nested_leaf	86.1710μs	46.4675μs	21.5204 KOps/s	21.4120 KOps/s	$\color{#35bf28}+0.51\%$
test_values_stack_nested	1.6372ms	1.0288ms	972.0153 Ops/s	959.8000 Ops/s	$\color{#35bf28}+1.27\%$
test_values_stack_nested_leaf	1.3031ms	1.0152ms	984.9862 Ops/s	969.7380 Ops/s	$\color{#35bf28}+1.57\%$
test_values_stack_nested_locked	0.8109ms	0.6026ms	1.6596 KOps/s	1.6748 KOps/s	$\color{#d91a1a}-0.91\%$
test_membership	13.6260μs	1.3430μs	744.6095 KOps/s	748.1500 KOps/s	$\color{#d91a1a}-0.47\%$
test_membership_nested	52.9390μs	3.4537μs	289.5412 KOps/s	286.7943 KOps/s	$\color{#35bf28}+0.96\%$
test_membership_nested_leaf	25.9690μs	3.4912μs	286.4384 KOps/s	288.8277 KOps/s	$\color{#d91a1a}-0.83\%$
test_membership_stacked_nested	49.2320μs	11.8620μs	84.3031 KOps/s	83.7436 KOps/s	$\color{#35bf28}+0.67\%$
test_membership_stacked_nested_leaf	40.3260μs	11.7948μs	84.7831 KOps/s	80.3773 KOps/s	$\textbf{\color{#35bf28}+5.48\%}$
test_membership_nested_last	32.0300μs	6.6263μs	150.9145 KOps/s	151.4305 KOps/s	$\color{#d91a1a}-0.34\%$
test_membership_nested_leaf_last	45.8660μs	6.6796μs	149.7086 KOps/s	150.8024 KOps/s	$\color{#d91a1a}-0.73\%$
test_membership_stacked_nested_last	0.3971ms	0.1781ms	5.6140 KOps/s	5.6004 KOps/s	$\color{#35bf28}+0.24\%$
test_membership_stacked_nested_leaf_last	43.9220μs	13.9652μs	71.6068 KOps/s	71.7731 KOps/s	$\color{#d91a1a}-0.23\%$
test_nested_getleaf	47.4010μs	10.6819μs	93.6161 KOps/s	93.3619 KOps/s	$\color{#35bf28}+0.27\%$
test_nested_get	36.5390μs	10.0942μs	99.0668 KOps/s	97.7826 KOps/s	$\color{#35bf28}+1.31\%$
test_stacked_getleaf	0.4853ms	0.3984ms	2.5103 KOps/s	2.4899 KOps/s	$\color{#35bf28}+0.82\%$
test_stacked_get	0.4914ms	0.3677ms	2.7198 KOps/s	2.7259 KOps/s	$\color{#d91a1a}-0.23\%$
test_nested_getitemleaf	41.2970μs	12.1466μs	82.3274 KOps/s	81.5118 KOps/s	$\color{#35bf28}+1.00\%$
test_nested_getitem	44.2130μs	11.5429μs	86.6330 KOps/s	84.8724 KOps/s	$\color{#35bf28}+2.07\%$
test_stacked_getitemleaf	0.5726ms	0.4006ms	2.4965 KOps/s	2.4250 KOps/s	$\color{#35bf28}+2.95\%$
test_stacked_getitem	0.5747ms	0.3686ms	2.7133 KOps/s	2.6876 KOps/s	$\color{#35bf28}+0.96\%$
test_lock_nested	2.8698ms	0.3410ms	2.9326 KOps/s	2.9462 KOps/s	$\color{#d91a1a}-0.46\%$
test_lock_stack_nested	90.4692ms	6.1682ms	162.1222 Ops/s	164.1127 Ops/s	$\color{#d91a1a}-1.21\%$
test_unlock_nested	73.1753ms	0.4104ms	2.4369 KOps/s	2.9364 KOps/s	$\textbf{\color{#d91a1a}-17.01\%}$
test_unlock_stack_nested	95.6620ms	6.5056ms	153.7146 Ops/s	157.4749 Ops/s	$\color{#d91a1a}-2.39\%$
test_flatten_speed	0.7384ms	0.3648ms	2.7414 KOps/s	2.7004 KOps/s	$\color{#35bf28}+1.52\%$
test_unflatten_speed	0.6869ms	0.4676ms	2.1385 KOps/s	2.1342 KOps/s	$\color{#35bf28}+0.20\%$
test_common_ops	4.7164ms	0.6360ms	1.5723 KOps/s	1.4119 KOps/s	$\textbf{\color{#35bf28}+11.36\%}$
test_creation	57.0270μs	1.8459μs	541.7473 KOps/s	525.4360 KOps/s	$\color{#35bf28}+3.10\%$
test_creation_empty	53.3130μs	7.4728μs	133.8180 KOps/s	102.2521 KOps/s	$\textbf{\color{#35bf28}+30.87\%}$
test_creation_nested_1	43.8920μs	10.1481μs	98.5406 KOps/s	80.0932 KOps/s	$\textbf{\color{#35bf28}+23.03\%}$
test_creation_nested_2	52.9000μs	13.1972μs	75.7739 KOps/s	63.2706 KOps/s	$\textbf{\color{#35bf28}+19.76\%}$
test_clone	67.2460μs	13.0992μs	76.3408 KOps/s	77.0425 KOps/s	$\color{#d91a1a}-0.91\%$
test_getitem[int]	37.2300μs	10.9493μs	91.3303 KOps/s	90.5900 KOps/s	$\color{#35bf28}+0.82\%$
test_getitem[slice_int]	90.6000μs	22.1782μs	45.0893 KOps/s	44.1969 KOps/s	$\color{#35bf28}+2.02\%$
test_getitem[range]	0.1497ms	43.2827μs	23.1039 KOps/s	23.7130 KOps/s	$\color{#d91a1a}-2.57\%$
test_getitem[tuple]	0.1519ms	18.5274μs	53.9741 KOps/s	54.5847 KOps/s	$\color{#d91a1a}-1.12\%$
test_getitem[list]	0.2663ms	39.0709μs	25.5945 KOps/s	26.9076 KOps/s	$\color{#d91a1a}-4.88\%$
test_setitem_dim[int]	55.5640μs	27.2390μs	36.7121 KOps/s	33.2036 KOps/s	$\textbf{\color{#35bf28}+10.57\%}$
test_setitem_dim[slice_int]	98.4950μs	52.0555μs	19.2103 KOps/s	17.7964 KOps/s	$\textbf{\color{#35bf28}+7.94\%}$
test_setitem_dim[range]	0.1216ms	74.4592μs	13.4302 KOps/s	12.9155 KOps/s	$\color{#35bf28}+3.99\%$
test_setitem_dim[tuple]	76.3730μs	41.4207μs	24.1425 KOps/s	22.5213 KOps/s	$\textbf{\color{#35bf28}+7.20\%}$
test_setitem	91.8620μs	18.3026μs	54.6372 KOps/s	50.4158 KOps/s	$\textbf{\color{#35bf28}+8.37\%}$
test_set	90.9710μs	17.7179μs	56.4400 KOps/s	52.6239 KOps/s	$\textbf{\color{#35bf28}+7.25\%}$
test_set_shared	1.8696ms	0.1399ms	7.1455 KOps/s	6.9445 KOps/s	$\color{#35bf28}+2.89\%$
test_update	0.1291ms	19.0160μs	52.5873 KOps/s	46.2622 KOps/s	$\textbf{\color{#35bf28}+13.67\%}$
test_update_nested	0.1077ms	26.6164μs	37.5709 KOps/s	33.9998 KOps/s	$\textbf{\color{#35bf28}+10.50\%}$
test_set_nested	0.1126ms	19.7206μs	50.7085 KOps/s	46.8162 KOps/s	$\textbf{\color{#35bf28}+8.31\%}$
test_set_nested_new	0.1036ms	23.4728μs	42.6025 KOps/s	39.7159 KOps/s	$\textbf{\color{#35bf28}+7.27\%}$
test_select	0.1488ms	37.0931μs	26.9592 KOps/s	26.2019 KOps/s	$\color{#35bf28}+2.89\%$
test_select_nested	0.1288ms	58.9968μs	16.9501 KOps/s	17.4206 KOps/s	$\color{#d91a1a}-2.70\%$
test_exclude_nested	0.2156ms	0.1178ms	8.4856 KOps/s	8.6290 KOps/s	$\color{#d91a1a}-1.66\%$
test_empty[True]	5.5056ms	0.4266ms	2.3443 KOps/s	2.4254 KOps/s	$\color{#d91a1a}-3.34\%$
test_empty[False]	7.9950μs	1.0650μs	938.9936 KOps/s	954.4091 KOps/s	$\color{#d91a1a}-1.62\%$
test_unbind_speed	0.3114ms	0.2459ms	4.0665 KOps/s	4.0522 KOps/s	$\color{#35bf28}+0.35\%$
test_unbind_speed_stack0	87.2141ms	3.5241ms	283.7575 Ops/s	306.3701 Ops/s	$\textbf{\color{#d91a1a}-7.38\%}$
test_unbind_speed_stack1	35.7660μs	1.9438μs	514.4438 KOps/s	520.2641 KOps/s	$\color{#d91a1a}-1.12\%$
test_split	1.5522ms	1.4618ms	684.0675 Ops/s	602.5273 Ops/s	$\textbf{\color{#35bf28}+13.53\%}$
test_chunk	79.9509ms	1.5775ms	633.9334 Ops/s	632.4532 Ops/s	$\color{#35bf28}+0.23\%$
test_creation[device0]	3.8061ms	0.1047ms	9.5470 KOps/s	9.8014 KOps/s	$\color{#d91a1a}-2.60\%$
test_creation_from_tensor	0.2611ms	81.9975μs	12.1955 KOps/s	11.8303 KOps/s	$\color{#35bf28}+3.09\%$
test_add_one[memmap_tensor0]	0.3257ms	5.4759μs	182.6197 KOps/s	185.4720 KOps/s	$\color{#d91a1a}-1.54\%$
test_contiguous[memmap_tensor0]	31.1390μs	0.6361μs	1.5720 MOps/s	1.5109 MOps/s	$\color{#35bf28}+4.05\%$
test_stack[memmap_tensor0]	53.7210μs	3.6971μs	270.4808 KOps/s	279.0552 KOps/s	$\color{#d91a1a}-3.07\%$
test_memmaptd_index	0.9950ms	0.2380ms	4.2010 KOps/s	4.1996 KOps/s	$\color{#35bf28}+0.03\%$
test_memmaptd_index_astensor	0.6976ms	0.3014ms	3.3180 KOps/s	3.3474 KOps/s	$\color{#d91a1a}-0.88\%$
test_memmaptd_index_op	1.0165ms	0.5632ms	1.7757 KOps/s	1.7117 KOps/s	$\color{#35bf28}+3.74\%$
test_serialize_model	0.1902s	0.1090s	9.1706 Ops/s	8.6425 Ops/s	$\textbf{\color{#35bf28}+6.11\%}$
test_serialize_model_pickle	0.4497s	0.3751s	2.6656 Ops/s	2.6084 Ops/s	$\color{#35bf28}+2.19\%$
test_serialize_weights	0.1045s	0.1009s	9.9120 Ops/s	9.7175 Ops/s	$\color{#35bf28}+2.00\%$
test_serialize_weights_returnearly	0.2066s	0.1328s	7.5285 Ops/s	7.8763 Ops/s	$\color{#d91a1a}-4.42\%$
test_serialize_weights_pickle	1.0403s	0.6088s	1.6426 Ops/s	2.3752 Ops/s	$\textbf{\color{#d91a1a}-30.84\%}$
test_serialize_weights_filesystem	99.7693ms	94.0694ms	10.6304 Ops/s	9.5055 Ops/s	$\textbf{\color{#35bf28}+11.84\%}$
test_serialize_model_filesystem	0.1813s	0.1055s	9.4751 Ops/s	10.3755 Ops/s	$\textbf{\color{#d91a1a}-8.68\%}$
test_reshape_pytree	60.3430μs	20.6303μs	48.4723 KOps/s	48.3882 KOps/s	$\color{#35bf28}+0.17\%$
test_reshape_td	66.9660μs	30.4078μs	32.8863 KOps/s	31.9888 KOps/s	$\color{#35bf28}+2.81\%$
test_view_pytree	75.4510μs	20.8062μs	48.0625 KOps/s	48.0016 KOps/s	$\color{#35bf28}+0.13\%$
test_view_td	91.1384ms	11.9999μs	83.3342 KOps/s	81.4289 KOps/s	$\color{#35bf28}+2.34\%$
test_unbind_pytree	58.4790μs	24.1396μs	41.4257 KOps/s	41.4894 KOps/s	$\color{#d91a1a}-0.15\%$
test_unbind_td	0.5272ms	36.2814μs	27.5623 KOps/s	28.1909 KOps/s	$\color{#d91a1a}-2.23\%$
test_split_pytree	60.2530μs	23.8558μs	41.9186 KOps/s	42.2923 KOps/s	$\color{#d91a1a}-0.88\%$
test_split_td	0.1259ms	39.3891μs	25.3877 KOps/s	24.9938 KOps/s	$\color{#35bf28}+1.58\%$
test_add_pytree	76.3630μs	30.0275μs	33.3028 KOps/s	34.2791 KOps/s	$\color{#d91a1a}-2.85\%$
test_add_td	0.1136ms	48.4701μs	20.6313 KOps/s	19.7707 KOps/s	$\color{#35bf28}+4.35\%$
test_distributed	0.2308ms	99.5828μs	10.0419 KOps/s	9.7831 KOps/s	$\color{#35bf28}+2.65\%$
test_tdmodule	0.1853ms	20.9845μs	47.6543 KOps/s	44.9547 KOps/s	$\textbf{\color{#35bf28}+6.01\%}$
test_tdmodule_dispatch	0.2280ms	41.2342μs	24.2517 KOps/s	23.0311 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_tdseq	0.1154ms	23.6662μs	42.2544 KOps/s	39.8439 KOps/s	$\textbf{\color{#35bf28}+6.05\%}$
test_tdseq_dispatch	0.1486ms	44.7098μs	22.3664 KOps/s	20.8719 KOps/s	$\textbf{\color{#35bf28}+7.16\%}$
test_instantiation_functorch	1.7639ms	1.3131ms	761.5525 Ops/s	772.1585 Ops/s	$\color{#d91a1a}-1.37\%$
test_instantiation_td	1.6631ms	1.0093ms	990.8121 Ops/s	993.3930 Ops/s	$\color{#d91a1a}-0.26\%$
test_exec_functorch	0.3474ms	0.1622ms	6.1668 KOps/s	6.3665 KOps/s	$\color{#d91a1a}-3.14\%$
test_exec_functional_call	0.2758ms	0.1472ms	6.7928 KOps/s	6.6334 KOps/s	$\color{#35bf28}+2.40\%$
test_exec_td	0.2733ms	0.1459ms	6.8558 KOps/s	6.8757 KOps/s	$\color{#d91a1a}-0.29\%$
test_exec_td_decorator	0.6677ms	0.1763ms	5.6720 KOps/s	5.7268 KOps/s	$\color{#d91a1a}-0.96\%$
test_vmap_mlp_speed[True-True]	1.2028ms	0.8747ms	1.1432 KOps/s	1.1124 KOps/s	$\color{#35bf28}+2.77\%$
test_vmap_mlp_speed[True-False]	0.7193ms	0.4673ms	2.1397 KOps/s	2.1372 KOps/s	$\color{#35bf28}+0.12\%$
test_vmap_mlp_speed[False-True]	1.1864ms	0.7684ms	1.3014 KOps/s	1.2818 KOps/s	$\color{#35bf28}+1.53\%$
test_vmap_mlp_speed[False-False]	0.5908ms	0.3907ms	2.5596 KOps/s	2.5788 KOps/s	$\color{#d91a1a}-0.74\%$
test_vmap_mlp_speed_decorator[True-True]	1.7724ms	1.5179ms	658.8224 Ops/s	621.8177 Ops/s	$\textbf{\color{#35bf28}+5.95\%}$
test_vmap_mlp_speed_decorator[True-False]	1.0903ms	0.5128ms	1.9502 KOps/s	1.9313 KOps/s	$\color{#35bf28}+0.98\%$
test_vmap_mlp_speed_decorator[False-True]	1.6268ms	1.2699ms	787.4415 Ops/s	759.7174 Ops/s	$\color{#35bf28}+3.65\%$
test_vmap_mlp_speed_decorator[False-False]	0.7088ms	0.4003ms	2.4981 KOps/s	2.5141 KOps/s	$\color{#d91a1a}-0.64\%$
test_to_module_speed[True]	1.3205ms	1.1276ms	886.8062 Ops/s	897.5916 Ops/s	$\color{#d91a1a}-1.20\%$
test_to_module_speed[False]	1.2194ms	1.1096ms	901.2592 Ops/s	906.9225 Ops/s	$\color{#d91a1a}-0.62\%$

[github-actions]

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

Total Benchmarks: 134. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	80.9020μs	13.9109μs	71.8859 KOps/s	75.7444 KOps/s	$\textbf{\color{#d91a1a}-5.09\%}$
test_plain_set_stack_nested	0.1681ms	0.1204ms	8.3069 KOps/s	8.3351 KOps/s	$\color{#d91a1a}-0.34\%$
test_plain_set_nested_inplace	40.2200μs	15.4664μs	64.6564 KOps/s	68.6088 KOps/s	$\textbf{\color{#d91a1a}-5.76\%}$
test_plain_set_stack_nested_inplace	0.1897ms	0.1492ms	6.7023 KOps/s	6.6858 KOps/s	$\color{#35bf28}+0.25\%$
test_items	22.8000μs	4.6953μs	212.9783 KOps/s	211.1333 KOps/s	$\color{#35bf28}+0.87\%$
test_items_nested	0.4623ms	0.3366ms	2.9709 KOps/s	2.9321 KOps/s	$\color{#35bf28}+1.32\%$
test_items_nested_locked	0.3608ms	0.3401ms	2.9404 KOps/s	2.8936 KOps/s	$\color{#35bf28}+1.62\%$
test_items_nested_leaf	0.2189ms	0.1988ms	5.0293 KOps/s	4.9419 KOps/s	$\color{#35bf28}+1.77\%$
test_items_stack_nested	1.3609ms	1.3118ms	762.3059 Ops/s	754.1162 Ops/s	$\color{#35bf28}+1.09\%$
test_items_stack_nested_leaf	1.2209ms	1.1597ms	862.2663 Ops/s	861.1480 Ops/s	$\color{#35bf28}+0.13\%$
test_items_stack_nested_locked	0.9674ms	0.9124ms	1.0960 KOps/s	1.0925 KOps/s	$\color{#35bf28}+0.32\%$
test_keys	22.6400μs	4.5422μs	220.1581 KOps/s	219.3631 KOps/s	$\color{#35bf28}+0.36\%$
test_keys_nested	0.9047ms	93.5829μs	10.6857 KOps/s	10.6455 KOps/s	$\color{#35bf28}+0.38\%$
test_keys_nested_locked	0.1275ms	96.4318μs	10.3700 KOps/s	10.2576 KOps/s	$\color{#35bf28}+1.10\%$
test_keys_nested_leaf	0.1852ms	77.1421μs	12.9631 KOps/s	12.8127 KOps/s	$\color{#35bf28}+1.17\%$
test_keys_stack_nested	1.2075ms	1.1547ms	866.0417 Ops/s	858.6120 Ops/s	$\color{#35bf28}+0.87\%$
test_keys_stack_nested_leaf	1.6161ms	1.1429ms	874.9491 Ops/s	871.7281 Ops/s	$\color{#35bf28}+0.37\%$
test_keys_stack_nested_locked	0.7824ms	0.7293ms	1.3712 KOps/s	1.3633 KOps/s	$\color{#35bf28}+0.58\%$
test_values	8.8733μs	1.8757μs	533.1316 KOps/s	527.6168 KOps/s	$\color{#35bf28}+1.05\%$
test_values_nested	66.8310μs	44.8424μs	22.3003 KOps/s	21.7436 KOps/s	$\color{#35bf28}+2.56\%$
test_values_nested_locked	67.5110μs	47.1274μs	21.2191 KOps/s	20.8665 KOps/s	$\color{#35bf28}+1.69\%$
test_values_nested_leaf	59.7810μs	39.3502μs	25.4128 KOps/s	24.9822 KOps/s	$\color{#35bf28}+1.72\%$
test_values_stack_nested	1.0136ms	0.9548ms	1.0473 KOps/s	1.0450 KOps/s	$\color{#35bf28}+0.22\%$
test_values_stack_nested_leaf	1.0241ms	0.9520ms	1.0504 KOps/s	1.0415 KOps/s	$\color{#35bf28}+0.86\%$
test_values_stack_nested_locked	0.6339ms	0.5713ms	1.7503 KOps/s	1.6906 KOps/s	$\color{#35bf28}+3.53\%$
test_membership	5.0680μs	0.9434μs	1.0600 MOps/s	949.1530 KOps/s	$\textbf{\color{#35bf28}+11.68\%}$
test_membership_nested	32.3000μs	2.8448μs	351.5230 KOps/s	340.7262 KOps/s	$\color{#35bf28}+3.17\%$
test_membership_nested_leaf	21.8200μs	2.8678μs	348.6937 KOps/s	339.3515 KOps/s	$\color{#35bf28}+2.75\%$
test_membership_stacked_nested	46.1800μs	11.2208μs	89.1205 KOps/s	86.8177 KOps/s	$\color{#35bf28}+2.65\%$
test_membership_stacked_nested_leaf	29.5300μs	11.3708μs	87.9448 KOps/s	86.7293 KOps/s	$\color{#35bf28}+1.40\%$
test_membership_nested_last	23.3400μs	5.3075μs	188.4144 KOps/s	187.1172 KOps/s	$\color{#35bf28}+0.69\%$
test_membership_nested_leaf_last	30.2300μs	5.2905μs	189.0176 KOps/s	188.9033 KOps/s	$\color{#35bf28}+0.06\%$
test_membership_stacked_nested_last	0.1797ms	0.1555ms	6.4308 KOps/s	6.3986 KOps/s	$\color{#35bf28}+0.50\%$
test_membership_stacked_nested_leaf_last	48.9510μs	13.1683μs	75.9400 KOps/s	74.9850 KOps/s	$\color{#35bf28}+1.27\%$
test_nested_getleaf	35.0510μs	8.3641μs	119.5587 KOps/s	117.9652 KOps/s	$\color{#35bf28}+1.35\%$
test_nested_get	31.1310μs	7.9238μs	126.2021 KOps/s	125.3103 KOps/s	$\color{#35bf28}+0.71\%$
test_stacked_getleaf	0.3933ms	0.3254ms	3.0730 KOps/s	3.0782 KOps/s	$\color{#d91a1a}-0.17\%$
test_stacked_get	0.3402ms	0.2926ms	3.4180 KOps/s	3.4102 KOps/s	$\color{#35bf28}+0.23\%$
test_nested_getitemleaf	29.2000μs	9.7762μs	102.2894 KOps/s	101.6367 KOps/s	$\color{#35bf28}+0.64\%$
test_nested_getitem	30.2100μs	9.2968μs	107.5637 KOps/s	106.1167 KOps/s	$\color{#35bf28}+1.36\%$
test_stacked_getitemleaf	0.3562ms	0.3307ms	3.0240 KOps/s	3.0291 KOps/s	$\color{#d91a1a}-0.17\%$
test_stacked_getitem	0.3449ms	0.2962ms	3.3760 KOps/s	3.3710 KOps/s	$\color{#35bf28}+0.15\%$
test_lock_nested	1.2536ms	0.3507ms	2.8517 KOps/s	2.2245 KOps/s	$\textbf{\color{#35bf28}+28.19\%}$
test_lock_stack_nested	0.1050s	6.7482ms	148.1868 Ops/s	148.1417 Ops/s	$\color{#35bf28}+0.03\%$
test_unlock_nested	92.1366ms	0.4437ms	2.2537 KOps/s	2.2483 KOps/s	$\color{#35bf28}+0.24\%$
test_unlock_stack_nested	0.1100s	6.8357ms	146.2914 Ops/s	147.3188 Ops/s	$\color{#d91a1a}-0.70\%$
test_flatten_speed	0.3467ms	0.2571ms	3.8893 KOps/s	3.8346 KOps/s	$\color{#35bf28}+1.43\%$
test_unflatten_speed	0.3896ms	0.3589ms	2.7864 KOps/s	2.7807 KOps/s	$\color{#35bf28}+0.21\%$
test_common_ops	1.0647ms	0.6040ms	1.6558 KOps/s	1.6602 KOps/s	$\color{#d91a1a}-0.27\%$
test_creation	29.4700μs	1.5391μs	649.7378 KOps/s	636.7428 KOps/s	$\color{#35bf28}+2.04\%$
test_creation_empty	27.1110μs	8.7009μs	114.9312 KOps/s	133.4783 KOps/s	$\textbf{\color{#d91a1a}-13.90\%}$
test_creation_nested_1	35.8000μs	10.4479μs	95.7128 KOps/s	109.0621 KOps/s	$\textbf{\color{#d91a1a}-12.24\%}$
test_creation_nested_2	38.3510μs	12.7927μs	78.1698 KOps/s	86.9446 KOps/s	$\textbf{\color{#d91a1a}-10.09\%}$
test_clone	54.5910μs	13.5937μs	73.5633 KOps/s	68.8378 KOps/s	$\textbf{\color{#35bf28}+6.86\%}$
test_getitem[int]	36.4700μs	11.0848μs	90.2137 KOps/s	92.6769 KOps/s	$\color{#d91a1a}-2.66\%$
test_getitem[slice_int]	79.4410μs	21.0022μs	47.6141 KOps/s	46.4012 KOps/s	$\color{#35bf28}+2.61\%$
test_getitem[range]	60.1910μs	39.0270μs	25.6233 KOps/s	25.2264 KOps/s	$\color{#35bf28}+1.57\%$
test_getitem[tuple]	55.8310μs	18.4498μs	54.2012 KOps/s	52.9683 KOps/s	$\color{#35bf28}+2.33\%$
test_getitem[list]	0.2022ms	36.2306μs	27.6010 KOps/s	27.2107 KOps/s	$\color{#35bf28}+1.43\%$
test_setitem_dim[int]	43.4410μs	27.2458μs	36.7029 KOps/s	37.6165 KOps/s	$\color{#d91a1a}-2.43\%$
test_setitem_dim[slice_int]	73.1910μs	48.7419μs	20.5162 KOps/s	20.6236 KOps/s	$\color{#d91a1a}-0.52\%$
test_setitem_dim[range]	0.1027ms	67.1082μs	14.9013 KOps/s	15.2181 KOps/s	$\color{#d91a1a}-2.08\%$
test_setitem_dim[tuple]	60.5710μs	42.8722μs	23.3252 KOps/s	24.1658 KOps/s	$\color{#d91a1a}-3.48\%$
test_setitem	77.8020μs	18.5147μs	54.0110 KOps/s	53.2603 KOps/s	$\color{#35bf28}+1.41\%$
test_set	62.4210μs	18.1356μs	55.1402 KOps/s	54.6214 KOps/s	$\color{#35bf28}+0.95\%$
test_set_shared	2.9593ms	0.1027ms	9.7415 KOps/s	9.5014 KOps/s	$\color{#35bf28}+2.53\%$
test_update	87.5710μs	20.6628μs	48.3961 KOps/s	49.1553 KOps/s	$\color{#d91a1a}-1.54\%$
test_update_nested	81.7810μs	27.0426μs	36.9786 KOps/s	37.1258 KOps/s	$\color{#d91a1a}-0.40\%$
test_set_nested	75.1120μs	19.2660μs	51.9050 KOps/s	50.5507 KOps/s	$\color{#35bf28}+2.68\%$
test_set_nested_new	67.3810μs	21.7880μs	45.8969 KOps/s	44.9312 KOps/s	$\color{#35bf28}+2.15\%$
test_select	79.4410μs	34.3415μs	29.1193 KOps/s	27.4313 KOps/s	$\textbf{\color{#35bf28}+6.15\%}$
test_select_nested	69.3710μs	52.7503μs	18.9572 KOps/s	19.0091 KOps/s	$\color{#d91a1a}-0.27\%$
test_exclude_nested	0.1408ms	0.1135ms	8.8139 KOps/s	8.9363 KOps/s	$\color{#d91a1a}-1.37\%$
test_empty[True]	0.4210ms	0.3862ms	2.5894 KOps/s	2.6260 KOps/s	$\color{#d91a1a}-1.40\%$
test_empty[False]	2.2691μs	0.8857μs	1.1291 MOps/s	1.1734 MOps/s	$\color{#d91a1a}-3.78\%$
test_to	77.3020μs	55.2830μs	18.0887 KOps/s	17.8485 KOps/s	$\color{#35bf28}+1.35\%$
test_to_nonblocking	61.4210μs	33.9444μs	29.4599 KOps/s	28.3947 KOps/s	$\color{#35bf28}+3.75\%$
test_unbind_speed	0.3050ms	0.2678ms	3.7345 KOps/s	3.6426 KOps/s	$\color{#35bf28}+2.52\%$
test_unbind_speed_stack0	91.5055ms	3.4130ms	292.9946 Ops/s	229.5041 Ops/s	$\textbf{\color{#35bf28}+27.66\%}$
test_unbind_speed_stack1	18.2300μs	1.8197μs	549.5267 KOps/s	541.7269 KOps/s	$\color{#35bf28}+1.44\%$
test_split	88.7273ms	1.7593ms	568.3983 Ops/s	636.4226 Ops/s	$\textbf{\color{#d91a1a}-10.69\%}$
test_chunk	90.7286ms	1.6928ms	590.7310 Ops/s	586.7734 Ops/s	$\color{#35bf28}+0.67\%$
test_creation[device0]	0.1369ms	73.0671μs	13.6860 KOps/s	13.2431 KOps/s	$\color{#35bf28}+3.34\%$
test_creation_from_tensor	0.1292ms	56.8854μs	17.5792 KOps/s	17.5592 KOps/s	$\color{#35bf28}+0.11\%$
test_add_one[memmap_tensor0]	0.1230ms	7.1171μs	140.5073 KOps/s	135.4086 KOps/s	$\color{#35bf28}+3.77\%$
test_contiguous[memmap_tensor0]	24.6110μs	0.6638μs	1.5065 MOps/s	1.5412 MOps/s	$\color{#d91a1a}-2.25\%$
test_stack[memmap_tensor0]	29.4400μs	4.6450μs	215.2870 KOps/s	214.2582 KOps/s	$\color{#35bf28}+0.48\%$
test_memmaptd_index	1.1774ms	0.2620ms	3.8168 KOps/s	3.8279 KOps/s	$\color{#d91a1a}-0.29\%$
test_memmaptd_index_astensor	0.5720ms	0.3198ms	3.1266 KOps/s	3.1337 KOps/s	$\color{#d91a1a}-0.23\%$
test_memmaptd_index_op	0.9223ms	0.6360ms	1.5722 KOps/s	1.6170 KOps/s	$\color{#d91a1a}-2.77\%$
test_serialize_model	92.7156ms	89.2449ms	11.2051 Ops/s	9.4409 Ops/s	$\textbf{\color{#35bf28}+18.69\%}$
test_serialize_model_pickle	1.3491s	1.2360s	0.8091 Ops/s	0.8066 Ops/s	$\color{#35bf28}+0.30\%$
test_serialize_weights	89.5314ms	86.6980ms	11.5343 Ops/s	10.9571 Ops/s	$\textbf{\color{#35bf28}+5.27\%}$
test_serialize_weights_returnearly	63.0392ms	57.6493ms	17.3463 Ops/s	13.4533 Ops/s	$\textbf{\color{#35bf28}+28.94\%}$
test_serialize_weights_pickle	1.3464s	1.2488s	0.8007 Ops/s	0.8084 Ops/s	$\color{#d91a1a}-0.94\%$
test_reshape_pytree	60.1310μs	24.5485μs	40.7357 KOps/s	40.0021 KOps/s	$\color{#35bf28}+1.83\%$
test_reshape_td	0.2445ms	30.8698μs	32.3941 KOps/s	31.5649 KOps/s	$\color{#35bf28}+2.63\%$
test_view_pytree	0.1109ms	24.2022μs	41.3186 KOps/s	40.7737 KOps/s	$\color{#35bf28}+1.34\%$
test_view_td	90.8999ms	10.9241μs	91.5406 KOps/s	88.8892 KOps/s	$\color{#35bf28}+2.98\%$
test_unbind_pytree	73.0010μs	30.4970μs	32.7901 KOps/s	32.2662 KOps/s	$\color{#35bf28}+1.62\%$
test_unbind_td	0.2499ms	40.2070μs	24.8713 KOps/s	23.2382 KOps/s	$\textbf{\color{#35bf28}+7.03\%}$
test_split_pytree	54.4810μs	28.9178μs	34.5808 KOps/s	33.9613 KOps/s	$\color{#35bf28}+1.82\%$
test_split_td	0.1851ms	40.2710μs	24.8317 KOps/s	25.1467 KOps/s	$\color{#d91a1a}-1.25\%$
test_add_pytree	0.2639ms	36.2707μs	27.5705 KOps/s	26.8307 KOps/s	$\color{#35bf28}+2.76\%$
test_add_td	0.1429ms	51.2575μs	19.5094 KOps/s	19.8191 KOps/s	$\color{#d91a1a}-1.56\%$
test_distributed	3.3148ms	92.0640μs	10.8620 KOps/s	14.3490 KOps/s	$\textbf{\color{#d91a1a}-24.30\%}$
test_tdmodule	41.5400μs	17.9207μs	55.8015 KOps/s	57.6995 KOps/s	$\color{#d91a1a}-3.29\%$
test_tdmodule_dispatch	0.1293ms	36.8499μs	27.1371 KOps/s	27.9376 KOps/s	$\color{#d91a1a}-2.87\%$
test_tdseq	39.9810μs	20.8310μs	48.0053 KOps/s	49.2878 KOps/s	$\color{#d91a1a}-2.60\%$
test_tdseq_dispatch	59.5310μs	39.4418μs	25.3538 KOps/s	26.3519 KOps/s	$\color{#d91a1a}-3.79\%$
test_instantiation_functorch	1.7508ms	1.6553ms	604.1257 Ops/s	600.4161 Ops/s	$\color{#35bf28}+0.62\%$
test_instantiation_td	1.6809ms	1.1481ms	871.0135 Ops/s	863.8144 Ops/s	$\color{#35bf28}+0.83\%$
test_exec_functorch	0.1829ms	0.1596ms	6.2650 KOps/s	6.1022 KOps/s	$\color{#35bf28}+2.67\%$
test_exec_functional_call	0.2122ms	0.1586ms	6.3036 KOps/s	6.0949 KOps/s	$\color{#35bf28}+3.42\%$
test_exec_td	0.1876ms	0.1504ms	6.6481 KOps/s	6.4619 KOps/s	$\color{#35bf28}+2.88\%$
test_exec_td_decorator	0.7176ms	0.1805ms	5.5416 KOps/s	5.4933 KOps/s	$\color{#35bf28}+0.88\%$
test_vmap_mlp_speed[True-True]	1.1001ms	1.0300ms	970.8623 Ops/s	965.8333 Ops/s	$\color{#35bf28}+0.52\%$
test_vmap_mlp_speed[True-False]	0.6382ms	0.5965ms	1.6764 KOps/s	1.6777 KOps/s	$\color{#d91a1a}-0.08\%$
test_vmap_mlp_speed[False-True]	1.0248ms	0.9485ms	1.0543 KOps/s	1.0499 KOps/s	$\color{#35bf28}+0.42\%$
test_vmap_mlp_speed[False-False]	0.5884ms	0.5243ms	1.9072 KOps/s	1.8960 KOps/s	$\color{#35bf28}+0.59\%$
test_vmap_mlp_speed_decorator[True-True]	2.2536ms	1.7965ms	556.6461 Ops/s	556.8608 Ops/s	$\color{#d91a1a}-0.04\%$
test_vmap_mlp_speed_decorator[True-False]	0.7468ms	0.6294ms	1.5889 KOps/s	1.5871 KOps/s	$\color{#35bf28}+0.11\%$
test_vmap_mlp_speed_decorator[False-True]	1.6710ms	1.5592ms	641.3376 Ops/s	635.5100 Ops/s	$\color{#35bf28}+0.92\%$
test_vmap_mlp_speed_decorator[False-False]	0.8851ms	0.5329ms	1.8765 KOps/s	1.8557 KOps/s	$\color{#35bf28}+1.12\%$
test_vmap_transformer_speed[True-True]	12.3059ms	12.1244ms	82.4783 Ops/s	81.6351 Ops/s	$\color{#35bf28}+1.03\%$
test_vmap_transformer_speed[True-False]	8.9001ms	8.0021ms	124.9670 Ops/s	118.8509 Ops/s	$\textbf{\color{#35bf28}+5.15\%}$
test_vmap_transformer_speed[False-True]	12.2586ms	12.0922ms	82.6978 Ops/s	80.1971 Ops/s	$\color{#35bf28}+3.12\%$
test_vmap_transformer_speed[False-False]	8.0967ms	7.9344ms	126.0334 Ops/s	122.0384 Ops/s	$\color{#35bf28}+3.27\%$
test_vmap_transformer_speed_decorator[True-True]	59.5661ms	58.6791ms	17.0418 Ops/s	16.4743 Ops/s	$\color{#35bf28}+3.44\%$
test_vmap_transformer_speed_decorator[True-False]	20.0422ms	19.6783ms	50.8174 Ops/s	51.2578 Ops/s	$\color{#d91a1a}-0.86\%$
test_vmap_transformer_speed_decorator[False-True]	55.0670ms	53.6974ms	18.6229 Ops/s	18.1505 Ops/s	$\color{#35bf28}+2.60\%$
test_vmap_transformer_speed_decorator[False-False]	19.5261ms	18.8790ms	52.9688 Ops/s	52.1727 Ops/s	$\color{#35bf28}+1.53\%$
test_to_module_speed[True]	0.1362s	1.1525ms	867.6957 Ops/s	968.0075 Ops/s	$\textbf{\color{#d91a1a}-10.36\%}$
test_to_module_speed[False]	1.6017ms	0.9757ms	1.0249 KOps/s	1.0152 KOps/s	$\color{#35bf28}+0.95\%$