[BugFix] Allow fake-tensor detection pass through in torch 2.0

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 144. Improved: $\large\color{#35bf28}5$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	39.0320μs	16.1588μs	61.8858 KOps/s	61.3000 KOps/s	$\color{#35bf28}+0.96\%$
test_plain_set_stack_nested	72.8160μs	16.2233μs	61.6398 KOps/s	60.5802 KOps/s	$\color{#35bf28}+1.75\%$
test_plain_set_nested_inplace	66.5140μs	18.3289μs	54.5586 KOps/s	53.9782 KOps/s	$\color{#35bf28}+1.08\%$
test_plain_set_stack_nested_inplace	61.7550μs	18.5398μs	53.9379 KOps/s	53.6752 KOps/s	$\color{#35bf28}+0.49\%$
test_items	37.7410μs	2.5505μs	392.0843 KOps/s	396.6388 KOps/s	$\color{#d91a1a}-1.15\%$
test_items_nested	0.4382ms	0.2616ms	3.8223 KOps/s	3.7641 KOps/s	$\color{#35bf28}+1.55\%$
test_items_nested_locked	0.4516ms	0.2642ms	3.7851 KOps/s	3.7176 KOps/s	$\color{#35bf28}+1.81\%$
test_items_nested_leaf	0.1460ms	76.0457μs	13.1500 KOps/s	12.9220 KOps/s	$\color{#35bf28}+1.76\%$
test_items_stack_nested	0.4371ms	0.2678ms	3.7339 KOps/s	3.6854 KOps/s	$\color{#35bf28}+1.31\%$
test_items_stack_nested_leaf	0.1557ms	79.1637μs	12.6320 KOps/s	12.4015 KOps/s	$\color{#35bf28}+1.86\%$
test_items_stack_nested_locked	0.4758ms	0.2682ms	3.7282 KOps/s	3.6675 KOps/s	$\color{#35bf28}+1.66\%$
test_keys	35.1360μs	4.0383μs	247.6293 KOps/s	256.0010 KOps/s	$\color{#d91a1a}-3.27\%$
test_keys_nested	0.2349ms	0.1378ms	7.2552 KOps/s	7.0548 KOps/s	$\color{#35bf28}+2.84\%$
test_keys_nested_locked	0.8019ms	0.1427ms	7.0096 KOps/s	6.8315 KOps/s	$\color{#35bf28}+2.61\%$
test_keys_nested_leaf	0.2313ms	0.1162ms	8.6094 KOps/s	8.4010 KOps/s	$\color{#35bf28}+2.48\%$
test_keys_stack_nested	0.2867ms	0.1380ms	7.2489 KOps/s	7.1169 KOps/s	$\color{#35bf28}+1.85\%$
test_keys_stack_nested_leaf	0.2008ms	0.1163ms	8.5990 KOps/s	8.4059 KOps/s	$\color{#35bf28}+2.30\%$
test_keys_stack_nested_locked	0.2005ms	0.1426ms	7.0107 KOps/s	6.9339 KOps/s	$\color{#35bf28}+1.11\%$
test_values	10.9127μs	1.1763μs	850.1248 KOps/s	844.3399 KOps/s	$\color{#35bf28}+0.69\%$
test_values_nested	0.1128ms	50.4426μs	19.8245 KOps/s	19.5240 KOps/s	$\color{#35bf28}+1.54\%$
test_values_nested_locked	0.1026ms	50.2507μs	19.9002 KOps/s	19.5781 KOps/s	$\color{#35bf28}+1.65\%$
test_values_nested_leaf	0.1079ms	45.3544μs	22.0486 KOps/s	21.6229 KOps/s	$\color{#35bf28}+1.97\%$
test_values_stack_nested	0.1162ms	51.2375μs	19.5169 KOps/s	19.2146 KOps/s	$\color{#35bf28}+1.57\%$
test_values_stack_nested_leaf	91.9010μs	45.4129μs	22.0202 KOps/s	21.4565 KOps/s	$\color{#35bf28}+2.63\%$
test_values_stack_nested_locked	0.1023ms	51.1352μs	19.5560 KOps/s	19.1736 KOps/s	$\color{#35bf28}+1.99\%$
test_membership	16.7220μs	1.3790μs	725.1664 KOps/s	746.8887 KOps/s	$\color{#d91a1a}-2.91\%$
test_membership_nested	47.4690μs	3.4365μs	290.9976 KOps/s	296.3589 KOps/s	$\color{#d91a1a}-1.81\%$
test_membership_nested_leaf	39.2230μs	3.4786μs	287.4729 KOps/s	292.4235 KOps/s	$\color{#d91a1a}-1.69\%$
test_membership_stacked_nested	31.9790μs	3.4248μs	291.9916 KOps/s	284.7108 KOps/s	$\color{#35bf28}+2.56\%$
test_membership_stacked_nested_leaf	22.7230μs	3.4746μs	287.8039 KOps/s	291.2321 KOps/s	$\color{#d91a1a}-1.18\%$
test_membership_nested_last	28.4140μs	4.2586μs	234.8199 KOps/s	240.2103 KOps/s	$\color{#d91a1a}-2.24\%$
test_membership_nested_leaf_last	42.0180μs	4.2868μs	233.2734 KOps/s	240.1258 KOps/s	$\color{#d91a1a}-2.85\%$
test_membership_stacked_nested_last	26.8100μs	4.3546μs	229.6438 KOps/s	174.7706 KOps/s	$\textbf{\color{#35bf28}+31.40\%}$
test_membership_stacked_nested_leaf_last	25.1070μs	4.3175μs	231.6152 KOps/s	174.1648 KOps/s	$\textbf{\color{#35bf28}+32.99\%}$
test_nested_getleaf	49.2920μs	10.6009μs	94.3317 KOps/s	93.6616 KOps/s	$\color{#35bf28}+0.72\%$
test_nested_get	40.6350μs	9.9897μs	100.1026 KOps/s	98.5651 KOps/s	$\color{#35bf28}+1.56\%$
test_stacked_getleaf	68.0270μs	10.4585μs	95.6160 KOps/s	95.3767 KOps/s	$\color{#35bf28}+0.25\%$
test_stacked_get	42.3690μs	9.8978μs	101.0326 KOps/s	100.5283 KOps/s	$\color{#35bf28}+0.50\%$
test_nested_getitemleaf	40.5760μs	11.0842μs	90.2185 KOps/s	89.1588 KOps/s	$\color{#35bf28}+1.19\%$
test_nested_getitem	32.2400μs	10.1642μs	98.3841 KOps/s	97.0938 KOps/s	$\color{#35bf28}+1.33\%$
test_stacked_getitemleaf	52.7590μs	10.8426μs	92.2288 KOps/s	91.4629 KOps/s	$\color{#35bf28}+0.84\%$
test_stacked_getitem	32.9210μs	9.9882μs	100.1184 KOps/s	98.5618 KOps/s	$\color{#35bf28}+1.58\%$
test_lock_nested	0.7804ms	0.3531ms	2.8322 KOps/s	2.4223 KOps/s	$\textbf{\color{#35bf28}+16.92\%}$
test_lock_stack_nested	0.5496ms	0.3158ms	3.1670 KOps/s	3.2116 KOps/s	$\color{#d91a1a}-1.39\%$
test_unlock_nested	0.7652ms	0.3597ms	2.7801 KOps/s	2.3859 KOps/s	$\textbf{\color{#35bf28}+16.52\%}$
test_unlock_stack_nested	0.3844ms	0.3232ms	3.0944 KOps/s	3.1300 KOps/s	$\color{#d91a1a}-1.14\%$
test_flatten_speed	0.5987ms	96.9153μs	10.3183 KOps/s	10.2819 KOps/s	$\color{#35bf28}+0.35\%$
test_unflatten_speed	0.5829ms	0.4098ms	2.4401 KOps/s	2.4037 KOps/s	$\color{#35bf28}+1.52\%$
test_common_ops	1.2262ms	0.7010ms	1.4265 KOps/s	1.4347 KOps/s	$\color{#d91a1a}-0.57\%$
test_creation	18.6550μs	1.9216μs	520.4012 KOps/s	530.6737 KOps/s	$\color{#d91a1a}-1.94\%$
test_creation_empty	21.7000μs	9.4038μs	106.3401 KOps/s	106.4745 KOps/s	$\color{#d91a1a}-0.13\%$
test_creation_nested_1	40.3250μs	12.1691μs	82.1756 KOps/s	83.0445 KOps/s	$\color{#d91a1a}-1.05\%$
test_creation_nested_2	40.0350μs	15.4847μs	64.5801 KOps/s	64.6259 KOps/s	$\color{#d91a1a}-0.07\%$
test_clone	0.2212ms	14.0571μs	71.1382 KOps/s	74.0055 KOps/s	$\color{#d91a1a}-3.87\%$
test_getitem[int]	35.3050μs	11.9258μs	83.8518 KOps/s	86.8195 KOps/s	$\color{#d91a1a}-3.42\%$
test_getitem[slice_int]	68.6480μs	23.1653μs	43.1680 KOps/s	41.2798 KOps/s	$\color{#35bf28}+4.57\%$
test_getitem[range]	83.5460μs	60.3228μs	16.5775 KOps/s	16.6638 KOps/s	$\color{#d91a1a}-0.52\%$
test_getitem[tuple]	47.3480μs	19.3536μs	51.6700 KOps/s	50.8548 KOps/s	$\color{#35bf28}+1.60\%$
test_getitem[list]	0.1386ms	42.2365μs	23.6762 KOps/s	23.0906 KOps/s	$\color{#35bf28}+2.54\%$
test_setitem_dim[int]	54.6920μs	36.3203μs	27.5328 KOps/s	27.8146 KOps/s	$\color{#d91a1a}-1.01\%$
test_setitem_dim[slice_int]	99.8370μs	64.0072μs	15.6232 KOps/s	15.4460 KOps/s	$\color{#35bf28}+1.15\%$
test_setitem_dim[range]	0.1594ms	85.5726μs	11.6860 KOps/s	11.8942 KOps/s	$\color{#d91a1a}-1.75\%$
test_setitem_dim[tuple]	93.1340μs	52.0960μs	19.1953 KOps/s	19.5656 KOps/s	$\color{#d91a1a}-1.89\%$
test_setitem	80.6100μs	20.2472μs	49.3896 KOps/s	50.3635 KOps/s	$\color{#d91a1a}-1.93\%$
test_set	59.0700μs	19.6366μs	50.9253 KOps/s	51.9080 KOps/s	$\color{#d91a1a}-1.89\%$
test_set_shared	3.0161ms	0.1442ms	6.9359 KOps/s	7.0243 KOps/s	$\color{#d91a1a}-1.26\%$
test_update	0.1473ms	21.1994μs	47.1711 KOps/s	48.0305 KOps/s	$\color{#d91a1a}-1.79\%$
test_update_nested	0.1123ms	29.9873μs	33.3475 KOps/s	34.5010 KOps/s	$\color{#d91a1a}-3.34\%$
test_update__nested	72.5550μs	26.0521μs	38.3847 KOps/s	37.5422 KOps/s	$\color{#35bf28}+2.24\%$
test_set_nested	80.5970μs	21.7263μs	46.0272 KOps/s	46.9344 KOps/s	$\color{#d91a1a}-1.93\%$
test_set_nested_new	79.7090μs	26.0790μs	38.3450 KOps/s	38.8650 KOps/s	$\color{#d91a1a}-1.34\%$
test_select	0.1304ms	41.8725μs	23.8820 KOps/s	24.7004 KOps/s	$\color{#d91a1a}-3.31\%$
test_select_nested	0.1398ms	60.9235μs	16.4140 KOps/s	16.4145 KOps/s	$-0.00\%$
test_exclude_nested	0.2257ms	0.1238ms	8.0754 KOps/s	8.1286 KOps/s	$\color{#d91a1a}-0.65\%$
test_empty[True]	0.5955ms	0.3993ms	2.5046 KOps/s	2.4755 KOps/s	$\color{#35bf28}+1.18\%$
test_empty[False]	8.8264μs	1.1435μs	874.5098 KOps/s	865.7769 KOps/s	$\color{#35bf28}+1.01\%$
test_unbind_speed	1.6786ms	0.2754ms	3.6315 KOps/s	3.8019 KOps/s	$\color{#d91a1a}-4.48\%$
test_unbind_speed_stack0	0.6198ms	0.2582ms	3.8727 KOps/s	3.8867 KOps/s	$\color{#d91a1a}-0.36\%$
test_unbind_speed_stack1	82.6902ms	0.7502ms	1.3330 KOps/s	1.2842 KOps/s	$\color{#35bf28}+3.80\%$
test_split	77.6977ms	1.6480ms	606.8009 Ops/s	610.1766 Ops/s	$\color{#d91a1a}-0.55\%$
test_chunk	76.3349ms	1.6453ms	607.7773 Ops/s	611.7764 Ops/s	$\color{#d91a1a}-0.65\%$
test_creation[device0]	3.8271ms	87.5772μs	11.4185 KOps/s	11.6999 KOps/s	$\color{#d91a1a}-2.40\%$
test_creation_from_tensor	0.2522ms	84.8678μs	11.7830 KOps/s	11.9101 KOps/s	$\color{#d91a1a}-1.07\%$
test_add_one[memmap_tensor0]	0.1100ms	5.5033μs	181.7076 KOps/s	178.4003 KOps/s	$\color{#35bf28}+1.85\%$
test_contiguous[memmap_tensor0]	22.1210μs	0.6418μs	1.5580 MOps/s	1.5576 MOps/s	$\color{#35bf28}+0.02\%$
test_stack[memmap_tensor0]	25.4980μs	3.5576μs	281.0854 KOps/s	281.0261 KOps/s	$\color{#35bf28}+0.02\%$
test_memmaptd_index	1.2153ms	0.2550ms	3.9215 KOps/s	3.7785 KOps/s	$\color{#35bf28}+3.78\%$
test_memmaptd_index_astensor	0.7883ms	0.3298ms	3.0317 KOps/s	2.9386 KOps/s	$\color{#35bf28}+3.17\%$
test_memmaptd_index_op	0.9361ms	0.6124ms	1.6330 KOps/s	1.6223 KOps/s	$\color{#35bf28}+0.66\%$
test_serialize_model	0.2008s	0.1193s	8.3793 Ops/s	8.2886 Ops/s	$\color{#35bf28}+1.09\%$
test_serialize_model_pickle	0.4508s	0.3779s	2.6461 Ops/s	2.6021 Ops/s	$\color{#35bf28}+1.69\%$
test_serialize_weights	0.1917s	0.1183s	8.4555 Ops/s	8.5010 Ops/s	$\color{#d91a1a}-0.53\%$
test_serialize_weights_returnearly	0.2155s	0.1398s	7.1527 Ops/s	7.5554 Ops/s	$\textbf{\color{#d91a1a}-5.33\%}$
test_serialize_weights_pickle	1.0457s	0.5720s	1.7481 Ops/s	2.4341 Ops/s	$\textbf{\color{#d91a1a}-28.18\%}$
test_serialize_weights_filesystem	0.1055s	99.0740ms	10.0935 Ops/s	10.0190 Ops/s	$\color{#35bf28}+0.74\%$
test_serialize_model_filesystem	0.1811s	0.1090s	9.1778 Ops/s	9.6078 Ops/s	$\color{#d91a1a}-4.48\%$
test_reshape_pytree	51.4860μs	25.5161μs	39.1909 KOps/s	38.6761 KOps/s	$\color{#35bf28}+1.33\%$
test_reshape_td	88.0840μs	34.7638μs	28.7655 KOps/s	28.4107 KOps/s	$\color{#35bf28}+1.25\%$
test_view_pytree	62.0260μs	25.2491μs	39.6053 KOps/s	39.2269 KOps/s	$\color{#35bf28}+0.96\%$
test_view_td	97.9230μs	39.0496μs	25.6085 KOps/s	24.7689 KOps/s	$\color{#35bf28}+3.39\%$
test_unbind_pytree	91.8020μs	29.6393μs	33.7389 KOps/s	33.5292 KOps/s	$\color{#35bf28}+0.63\%$
test_unbind_td	0.4804ms	39.1626μs	25.5345 KOps/s	25.8894 KOps/s	$\color{#d91a1a}-1.37\%$
test_split_pytree	76.8130μs	30.1044μs	33.2178 KOps/s	34.3658 KOps/s	$\color{#d91a1a}-3.34\%$
test_split_td	0.5770ms	42.2799μs	23.6519 KOps/s	23.7001 KOps/s	$\color{#d91a1a}-0.20\%$
test_add_pytree	92.8640μs	35.9642μs	27.8054 KOps/s	28.3652 KOps/s	$\color{#d91a1a}-1.97\%$
test_add_td	0.1239ms	58.8353μs	16.9966 KOps/s	17.8589 KOps/s	$\color{#d91a1a}-4.83\%$
test_distributed	0.2752ms	0.1033ms	9.6848 KOps/s	9.5727 KOps/s	$\color{#35bf28}+1.17\%$
test_tdmodule	0.1097ms	17.3725μs	57.5624 KOps/s	59.4499 KOps/s	$\color{#d91a1a}-3.17\%$
test_tdmodule_dispatch	50.1630μs	33.5715μs	29.7872 KOps/s	30.0091 KOps/s	$\color{#d91a1a}-0.74\%$
test_tdseq	45.5450μs	20.3590μs	49.1183 KOps/s	49.7053 KOps/s	$\color{#d91a1a}-1.18\%$
test_tdseq_dispatch	82.2740μs	39.4541μs	25.3459 KOps/s	25.6098 KOps/s	$\color{#d91a1a}-1.03\%$
test_instantiation_functorch	2.1217ms	1.3387ms	746.9915 Ops/s	760.1598 Ops/s	$\color{#d91a1a}-1.73\%$
test_instantiation_td	1.6560ms	1.0452ms	956.7279 Ops/s	975.2500 Ops/s	$\color{#d91a1a}-1.90\%$
test_exec_functorch	0.2974ms	0.1607ms	6.2215 KOps/s	6.1036 KOps/s	$\color{#35bf28}+1.93\%$
test_exec_functional_call	0.3550ms	0.1499ms	6.6722 KOps/s	6.4478 KOps/s	$\color{#35bf28}+3.48\%$
test_exec_td	0.2243ms	0.1451ms	6.8907 KOps/s	6.6124 KOps/s	$\color{#35bf28}+4.21\%$
test_exec_td_decorator	0.9652ms	0.2227ms	4.4894 KOps/s	4.4628 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_mlp_speed[True-True]	0.6837ms	0.4820ms	2.0746 KOps/s	2.0649 KOps/s	$\color{#35bf28}+0.47\%$
test_vmap_mlp_speed[True-False]	0.8482ms	0.4784ms	2.0904 KOps/s	2.0721 KOps/s	$\color{#35bf28}+0.89\%$
test_vmap_mlp_speed[False-True]	0.6510ms	0.3871ms	2.5833 KOps/s	2.5206 KOps/s	$\color{#35bf28}+2.49\%$
test_vmap_mlp_speed[False-False]	0.6244ms	0.3867ms	2.5860 KOps/s	2.5167 KOps/s	$\color{#35bf28}+2.75\%$
test_vmap_mlp_speed_decorator[True-True]	1.3220ms	0.5531ms	1.8079 KOps/s	1.8102 KOps/s	$\color{#d91a1a}-0.13\%$
test_vmap_mlp_speed_decorator[True-False]	1.0037ms	0.5519ms	1.8118 KOps/s	1.8156 KOps/s	$\color{#d91a1a}-0.21\%$
test_vmap_mlp_speed_decorator[False-True]	0.9073ms	0.4527ms	2.2092 KOps/s	2.1402 KOps/s	$\color{#35bf28}+3.22\%$
test_vmap_mlp_speed_decorator[False-False]	0.6836ms	0.4495ms	2.2247 KOps/s	2.0456 KOps/s	$\textbf{\color{#35bf28}+8.76\%}$
test_to_module_speed[True]	2.0025ms	1.7044ms	586.7073 Ops/s	584.7119 Ops/s	$\color{#35bf28}+0.34\%$
test_to_module_speed[False]	1.9457ms	1.6649ms	600.6396 Ops/s	598.9103 Ops/s	$\color{#35bf28}+0.29\%$
test_tc_init	72.6360μs	26.1016μs	38.3119 KOps/s	38.9323 KOps/s	$\color{#d91a1a}-1.59\%$
test_tc_init_nested	0.1452ms	50.2632μs	19.8953 KOps/s	19.0047 KOps/s	$\color{#35bf28}+4.69\%$
test_tc_first_layer_tensor	2.7065μs	0.6713μs	1.4897 MOps/s	1.4449 MOps/s	$\color{#35bf28}+3.10\%$
test_tc_first_layer_nontensor	2.5798μs	0.6827μs	1.4647 MOps/s	1.4646 MOps/s	$+0.00\%$
test_tc_second_layer_tensor	26.8500μs	1.8265μs	547.5062 KOps/s	540.3751 KOps/s	$\color{#35bf28}+1.32\%$
test_tc_second_layer_nontensor	15.1783μs	1.5426μs	648.2757 KOps/s	650.9552 KOps/s	$\color{#d91a1a}-0.41\%$
test_unbind	0.1037s	8.4270ms	118.6656 Ops/s	128.9333 Ops/s	$\textbf{\color{#d91a1a}-7.96\%}$
test_full_like	21.8453ms	12.7081ms	78.6898 Ops/s	83.9345 Ops/s	$\textbf{\color{#d91a1a}-6.25\%}$
test_zeros_like	14.5463ms	6.4320ms	155.4722 Ops/s	153.2131 Ops/s	$\color{#35bf28}+1.47\%$
test_ones_like	13.8814ms	7.0478ms	141.8880 Ops/s	149.6360 Ops/s	$\textbf{\color{#d91a1a}-5.18\%}$
test_clone	17.5469ms	9.0347ms	110.6850 Ops/s	112.8175 Ops/s	$\color{#d91a1a}-1.89\%$
test_squeeze	72.5450μs	14.5614μs	68.6746 KOps/s	68.0629 KOps/s	$\color{#35bf28}+0.90\%$
test_unsqueeze	0.1275ms	62.6214μs	15.9690 KOps/s	16.0801 KOps/s	$\color{#d91a1a}-0.69\%$
test_split	0.1928ms	0.1165ms	8.5820 KOps/s	8.8439 KOps/s	$\color{#d91a1a}-2.96\%$
test_permute	0.2286ms	0.1330ms	7.5184 KOps/s	7.8225 KOps/s	$\color{#d91a1a}-3.89\%$
test_stack	35.1629ms	25.4338ms	39.3178 Ops/s	41.6849 Ops/s	$\textbf{\color{#d91a1a}-5.68\%}$
test_cat	31.4785ms	25.5851ms	39.0852 Ops/s	41.0581 Ops/s	$\color{#d91a1a}-4.81\%$

[github-actions]

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

Total Benchmarks: 152. Improved: $\large\color{#35bf28}3$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.4863ms	13.6204μs	73.4191 KOps/s	75.3939 KOps/s	$\color{#d91a1a}-2.62\%$
test_plain_set_stack_nested	0.2013ms	13.8484μs	72.2105 KOps/s	74.6287 KOps/s	$\color{#d91a1a}-3.24\%$
test_plain_set_nested_inplace	0.2047ms	15.0247μs	66.5573 KOps/s	68.7854 KOps/s	$\color{#d91a1a}-3.24\%$
test_plain_set_stack_nested_inplace	0.2025ms	15.1571μs	65.9756 KOps/s	68.1298 KOps/s	$\color{#d91a1a}-3.16\%$
test_items	0.1893ms	4.6895μs	213.2431 KOps/s	209.7541 KOps/s	$\color{#35bf28}+1.66\%$
test_items_nested	0.5230ms	0.3370ms	2.9669 KOps/s	2.9297 KOps/s	$\color{#35bf28}+1.27\%$
test_items_nested_locked	0.5247ms	0.3465ms	2.8864 KOps/s	2.9179 KOps/s	$\color{#d91a1a}-1.08\%$
test_items_nested_leaf	0.1316ms	83.6175μs	11.9592 KOps/s	11.9973 KOps/s	$\color{#d91a1a}-0.32\%$
test_items_stack_nested	0.5371ms	0.3481ms	2.8724 KOps/s	2.9169 KOps/s	$\color{#d91a1a}-1.53\%$
test_items_stack_nested_leaf	0.2681ms	85.3337μs	11.7187 KOps/s	11.9648 KOps/s	$\color{#d91a1a}-2.06\%$
test_items_stack_nested_locked	0.5266ms	0.3433ms	2.9132 KOps/s	2.9048 KOps/s	$\color{#35bf28}+0.29\%$
test_keys	0.1887ms	4.3454μs	230.1279 KOps/s	230.5786 KOps/s	$\color{#d91a1a}-0.20\%$
test_keys_nested	0.2489ms	67.1273μs	14.8971 KOps/s	14.8077 KOps/s	$\color{#35bf28}+0.60\%$
test_keys_nested_locked	0.8232ms	73.5168μs	13.6023 KOps/s	13.8980 KOps/s	$\color{#d91a1a}-2.13\%$
test_keys_nested_leaf	2.4231ms	58.5615μs	17.0761 KOps/s	17.2575 KOps/s	$\color{#d91a1a}-1.05\%$
test_keys_stack_nested	99.6410μs	67.5642μs	14.8007 KOps/s	14.8025 KOps/s	$\color{#d91a1a}-0.01\%$
test_keys_stack_nested_leaf	0.2540ms	58.5367μs	17.0833 KOps/s	17.2340 KOps/s	$\color{#d91a1a}-0.87\%$
test_keys_stack_nested_locked	0.1183ms	72.2221μs	13.8462 KOps/s	13.7578 KOps/s	$\color{#35bf28}+0.64\%$
test_values	55.4373μs	1.8091μs	552.7742 KOps/s	553.4950 KOps/s	$\color{#d91a1a}-0.13\%$
test_values_nested	68.0810μs	36.0162μs	27.7653 KOps/s	28.5411 KOps/s	$\color{#d91a1a}-2.72\%$
test_values_nested_locked	0.2292ms	38.1388μs	26.2200 KOps/s	27.0126 KOps/s	$\color{#d91a1a}-2.93\%$
test_values_nested_leaf	63.7910μs	32.2912μs	30.9682 KOps/s	31.9641 KOps/s	$\color{#d91a1a}-3.12\%$
test_values_stack_nested	95.6810μs	36.5620μs	27.3508 KOps/s	27.8548 KOps/s	$\color{#d91a1a}-1.81\%$
test_values_stack_nested_leaf	56.7510μs	32.9404μs	30.3578 KOps/s	31.4608 KOps/s	$\color{#d91a1a}-3.51\%$
test_values_stack_nested_locked	68.7310μs	38.8350μs	25.7500 KOps/s	26.6140 KOps/s	$\color{#d91a1a}-3.25\%$
test_membership	1.8275μs	0.6984μs	1.4318 MOps/s	1.3882 MOps/s	$\color{#35bf28}+3.14\%$
test_membership_nested	26.7300μs	2.5415μs	393.4703 KOps/s	389.0147 KOps/s	$\color{#35bf28}+1.15\%$
test_membership_nested_leaf	26.8200μs	2.5410μs	393.5436 KOps/s	389.2237 KOps/s	$\color{#35bf28}+1.11\%$
test_membership_stacked_nested	23.3000μs	2.5890μs	386.2436 KOps/s	393.4256 KOps/s	$\color{#d91a1a}-1.83\%$
test_membership_stacked_nested_leaf	40.4910μs	2.5627μs	390.2127 KOps/s	391.7307 KOps/s	$\color{#d91a1a}-0.39\%$
test_membership_nested_last	32.4710μs	3.0746μs	325.2477 KOps/s	324.6305 KOps/s	$\color{#35bf28}+0.19\%$
test_membership_nested_leaf_last	27.5010μs	3.0950μs	323.1063 KOps/s	326.3676 KOps/s	$\color{#d91a1a}-1.00\%$
test_membership_stacked_nested_last	47.2510μs	3.8511μs	259.6651 KOps/s	281.7535 KOps/s	$\textbf{\color{#d91a1a}-7.84\%}$
test_membership_stacked_nested_leaf_last	93.5310μs	3.8611μs	258.9945 KOps/s	284.8311 KOps/s	$\textbf{\color{#d91a1a}-9.07\%}$
test_nested_getleaf	25.5210μs	8.4010μs	119.0329 KOps/s	119.2282 KOps/s	$\color{#d91a1a}-0.16\%$
test_nested_get	23.1900μs	7.8795μs	126.9111 KOps/s	127.1278 KOps/s	$\color{#d91a1a}-0.17\%$
test_stacked_getleaf	29.2710μs	8.3757μs	119.3930 KOps/s	118.7707 KOps/s	$\color{#35bf28}+0.52\%$
test_stacked_get	30.4200μs	7.8760μs	126.9676 KOps/s	126.3099 KOps/s	$\color{#35bf28}+0.52\%$
test_nested_getitemleaf	34.1310μs	8.5480μs	116.9870 KOps/s	117.0606 KOps/s	$\color{#d91a1a}-0.06\%$
test_nested_getitem	35.3000μs	8.0198μs	124.6916 KOps/s	124.6215 KOps/s	$\color{#35bf28}+0.06\%$
test_stacked_getitemleaf	37.5700μs	8.5664μs	116.7351 KOps/s	116.2512 KOps/s	$\color{#35bf28}+0.42\%$
test_stacked_getitem	27.1200μs	8.0798μs	123.7652 KOps/s	124.1249 KOps/s	$\color{#d91a1a}-0.29\%$
test_lock_nested	58.8036ms	0.4132ms	2.4200 KOps/s	2.4194 KOps/s	$\color{#35bf28}+0.02\%$
test_lock_stack_nested	0.3350ms	0.3067ms	3.2606 KOps/s	3.2528 KOps/s	$\color{#35bf28}+0.24\%$
test_unlock_nested	0.7089ms	0.3547ms	2.8192 KOps/s	2.8441 KOps/s	$\color{#d91a1a}-0.88\%$
test_unlock_stack_nested	0.4769ms	0.3160ms	3.1645 KOps/s	3.1771 KOps/s	$\color{#d91a1a}-0.40\%$
test_flatten_speed	0.1854ms	0.1017ms	9.8293 KOps/s	9.7438 KOps/s	$\color{#35bf28}+0.88\%$
test_unflatten_speed	0.4391ms	0.2914ms	3.4320 KOps/s	3.4185 KOps/s	$\color{#35bf28}+0.40\%$
test_common_ops	1.2413ms	0.6214ms	1.6092 KOps/s	1.6558 KOps/s	$\color{#d91a1a}-2.81\%$
test_creation	36.4410μs	1.6123μs	620.2436 KOps/s	614.7658 KOps/s	$\color{#35bf28}+0.89\%$
test_creation_empty	28.8410μs	10.4105μs	96.0569 KOps/s	104.3335 KOps/s	$\textbf{\color{#d91a1a}-7.93\%}$
test_creation_nested_1	48.9710μs	12.1868μs	82.0559 KOps/s	87.7691 KOps/s	$\textbf{\color{#d91a1a}-6.51\%}$
test_creation_nested_2	55.2110μs	14.4164μs	69.3657 KOps/s	73.3193 KOps/s	$\textbf{\color{#d91a1a}-5.39\%}$
test_clone	0.1557ms	12.0665μs	82.8742 KOps/s	83.8198 KOps/s	$\color{#d91a1a}-1.13\%$
test_getitem[int]	32.2200μs	10.6667μs	93.7493 KOps/s	94.0208 KOps/s	$\color{#d91a1a}-0.29\%$
test_getitem[slice_int]	72.1510μs	20.2076μs	49.4864 KOps/s	48.5753 KOps/s	$\color{#35bf28}+1.88\%$
test_getitem[range]	65.1410μs	46.5708μs	21.4727 KOps/s	21.3982 KOps/s	$\color{#35bf28}+0.35\%$
test_getitem[tuple]	64.1910μs	18.3839μs	54.3954 KOps/s	54.5319 KOps/s	$\color{#d91a1a}-0.25\%$
test_getitem[list]	0.1682ms	33.5157μs	29.8368 KOps/s	26.7552 KOps/s	$\textbf{\color{#35bf28}+11.52\%}$
test_setitem_dim[int]	53.7710μs	31.6383μs	31.6072 KOps/s	30.4085 KOps/s	$\color{#35bf28}+3.94\%$
test_setitem_dim[slice_int]	75.2710μs	50.9875μs	19.6126 KOps/s	18.5519 KOps/s	$\textbf{\color{#35bf28}+5.72\%}$
test_setitem_dim[range]	0.1907ms	67.9032μs	14.7269 KOps/s	14.5698 KOps/s	$\color{#35bf28}+1.08\%$
test_setitem_dim[tuple]	73.4210μs	44.7011μs	22.3708 KOps/s	22.4290 KOps/s	$\color{#d91a1a}-0.26\%$
test_setitem	0.1207ms	17.8129μs	56.1392 KOps/s	57.0409 KOps/s	$\color{#d91a1a}-1.58\%$
test_set	56.6810μs	17.5571μs	56.9572 KOps/s	56.7724 KOps/s	$\color{#35bf28}+0.33\%$
test_set_shared	1.2180ms	99.3159μs	10.0689 KOps/s	9.8520 KOps/s	$\color{#35bf28}+2.20\%$
test_update	87.7820μs	20.2830μs	49.3023 KOps/s	51.3277 KOps/s	$\color{#d91a1a}-3.95\%$
test_update_nested	67.7110μs	25.8202μs	38.7294 KOps/s	40.4575 KOps/s	$\color{#d91a1a}-4.27\%$
test_update__nested	0.1116ms	22.8737μs	43.7183 KOps/s	43.7117 KOps/s	$\color{#35bf28}+0.01\%$
test_set_nested	66.2110μs	18.5282μs	53.9719 KOps/s	55.7913 KOps/s	$\color{#d91a1a}-3.26\%$
test_set_nested_new	68.2410μs	21.3545μs	46.8285 KOps/s	47.9562 KOps/s	$\color{#d91a1a}-2.35\%$
test_select	0.1739ms	34.6949μs	28.8227 KOps/s	28.3728 KOps/s	$\color{#35bf28}+1.59\%$
test_select_nested	0.6453ms	53.9186μs	18.5465 KOps/s	18.3648 KOps/s	$\color{#35bf28}+0.99\%$
test_exclude_nested	0.1806ms	0.1101ms	9.0860 KOps/s	9.0905 KOps/s	$\color{#d91a1a}-0.05\%$
test_empty[True]	0.5031ms	0.3419ms	2.9245 KOps/s	2.8723 KOps/s	$\color{#35bf28}+1.82\%$
test_empty[False]	4.2821μs	0.9204μs	1.0865 MOps/s	1.0916 MOps/s	$\color{#d91a1a}-0.47\%$
test_to	99.4420μs	77.1868μs	12.9556 KOps/s	13.0586 KOps/s	$\color{#d91a1a}-0.79\%$
test_to_nonblocking	0.2659ms	61.4295μs	16.2788 KOps/s	15.9448 KOps/s	$\color{#35bf28}+2.09\%$
test_unbind_speed	0.4051ms	0.2676ms	3.7363 KOps/s	3.7345 KOps/s	$\color{#35bf28}+0.05\%$
test_unbind_speed_stack0	0.2954ms	0.2687ms	3.7209 KOps/s	3.6999 KOps/s	$\color{#35bf28}+0.57\%$
test_unbind_speed_stack1	76.2574ms	0.8125ms	1.2308 KOps/s	1.2226 KOps/s	$\color{#35bf28}+0.68\%$
test_split	76.6976ms	1.6596ms	602.5489 Ops/s	595.6292 Ops/s	$\color{#35bf28}+1.16\%$
test_chunk	76.1169ms	1.6554ms	604.0763 Ops/s	594.9098 Ops/s	$\color{#35bf28}+1.54\%$
test_creation[device0]	0.1847ms	58.3230μs	17.1459 KOps/s	17.2295 KOps/s	$\color{#d91a1a}-0.49\%$
test_creation_from_tensor	0.2093ms	54.8854μs	18.2198 KOps/s	18.4720 KOps/s	$\color{#d91a1a}-1.37\%$
test_add_one[memmap_tensor0]	81.4420μs	7.2150μs	138.6004 KOps/s	138.0541 KOps/s	$\color{#35bf28}+0.40\%$
test_contiguous[memmap_tensor0]	23.2200μs	0.6744μs	1.4827 MOps/s	1.4539 MOps/s	$\color{#35bf28}+1.98\%$
test_stack[memmap_tensor0]	34.4210μs	4.8348μs	206.8346 KOps/s	209.4806 KOps/s	$\color{#d91a1a}-1.26\%$
test_memmaptd_index	1.1778ms	0.2898ms	3.4510 KOps/s	3.4808 KOps/s	$\color{#d91a1a}-0.86\%$
test_memmaptd_index_astensor	0.6416ms	0.3637ms	2.7494 KOps/s	2.7777 KOps/s	$\color{#d91a1a}-1.02\%$
test_memmaptd_index_op	1.2503ms	0.7002ms	1.4282 KOps/s	1.4469 KOps/s	$\color{#d91a1a}-1.29\%$
test_serialize_model	0.1846s	0.1134s	8.8160 Ops/s	8.3816 Ops/s	$\textbf{\color{#35bf28}+5.18\%}$
test_serialize_model_pickle	1.3859s	1.2394s	0.8068 Ops/s	0.8054 Ops/s	$\color{#35bf28}+0.18\%$
test_serialize_weights	0.1822s	0.1103s	9.0652 Ops/s	8.7341 Ops/s	$\color{#35bf28}+3.79\%$
test_serialize_weights_returnearly	0.2612s	99.4605ms	10.0542 Ops/s	9.9602 Ops/s	$\color{#35bf28}+0.94\%$
test_serialize_weights_pickle	1.3588s	1.2482s	0.8012 Ops/s	0.8088 Ops/s	$\color{#d91a1a}-0.94\%$
test_reshape_pytree	0.2179ms	26.1621μs	38.2233 KOps/s	38.2078 KOps/s	$\color{#35bf28}+0.04\%$
test_reshape_td	61.0410μs	30.4621μs	32.8277 KOps/s	32.2042 KOps/s	$\color{#35bf28}+1.94\%$
test_view_pytree	0.2090ms	25.6755μs	38.9476 KOps/s	38.4973 KOps/s	$\color{#35bf28}+1.17\%$
test_view_td	0.1624ms	35.8910μs	27.8621 KOps/s	28.0209 KOps/s	$\color{#d91a1a}-0.57\%$
test_unbind_pytree	0.1412ms	32.3539μs	30.9082 KOps/s	31.0808 KOps/s	$\color{#d91a1a}-0.56\%$
test_unbind_td	0.4771ms	42.5262μs	23.5149 KOps/s	23.8881 KOps/s	$\color{#d91a1a}-1.56\%$
test_split_pytree	73.6220μs	34.6331μs	28.8741 KOps/s	28.0923 KOps/s	$\color{#35bf28}+2.78\%$
test_split_td	0.1275ms	39.0330μs	25.6194 KOps/s	25.7249 KOps/s	$\color{#d91a1a}-0.41\%$
test_add_pytree	0.1728ms	39.7252μs	25.1730 KOps/s	25.5647 KOps/s	$\color{#d91a1a}-1.53\%$
test_add_td	0.1567ms	53.5008μs	18.6913 KOps/s	18.1544 KOps/s	$\color{#35bf28}+2.96\%$
test_distributed	2.5881ms	91.9217μs	10.8788 KOps/s	14.0333 KOps/s	$\textbf{\color{#d91a1a}-22.48\%}$
test_tdmodule	0.1542ms	15.7650μs	63.4318 KOps/s	65.0990 KOps/s	$\color{#d91a1a}-2.56\%$
test_tdmodule_dispatch	46.5910μs	30.3400μs	32.9597 KOps/s	33.0037 KOps/s	$\color{#d91a1a}-0.13\%$
test_tdseq	33.9200μs	17.4314μs	57.3676 KOps/s	58.9104 KOps/s	$\color{#d91a1a}-2.62\%$
test_tdseq_dispatch	0.1276ms	34.0904μs	29.3338 KOps/s	30.0560 KOps/s	$\color{#d91a1a}-2.40\%$
test_instantiation_functorch	1.6417ms	1.5248ms	655.8318 Ops/s	654.3637 Ops/s	$\color{#35bf28}+0.22\%$
test_instantiation_td	1.5141ms	1.0504ms	952.0224 Ops/s	967.5564 Ops/s	$\color{#d91a1a}-1.61\%$
test_exec_functorch	0.2953ms	0.1549ms	6.4557 KOps/s	6.4470 KOps/s	$\color{#35bf28}+0.13\%$
test_exec_functional_call	0.1948ms	0.1442ms	6.9363 KOps/s	6.9839 KOps/s	$\color{#d91a1a}-0.68\%$
test_exec_td	0.2151ms	0.1424ms	7.0210 KOps/s	7.1132 KOps/s	$\color{#d91a1a}-1.30\%$
test_exec_td_decorator	0.4124ms	0.2170ms	4.6088 KOps/s	4.6431 KOps/s	$\color{#d91a1a}-0.74\%$
test_vmap_mlp_speed[True-True]	0.7598ms	0.5893ms	1.6970 KOps/s	1.7060 KOps/s	$\color{#d91a1a}-0.53\%$
test_vmap_mlp_speed[True-False]	0.7411ms	0.5856ms	1.7077 KOps/s	1.7035 KOps/s	$\color{#35bf28}+0.25\%$
test_vmap_mlp_speed[False-True]	0.6701ms	0.5139ms	1.9460 KOps/s	1.9236 KOps/s	$\color{#35bf28}+1.16\%$
test_vmap_mlp_speed[False-False]	0.6822ms	0.5153ms	1.9406 KOps/s	1.9496 KOps/s	$\color{#d91a1a}-0.46\%$
test_vmap_mlp_speed_decorator[True-True]	1.1394ms	0.6532ms	1.5310 KOps/s	1.5385 KOps/s	$\color{#d91a1a}-0.49\%$
test_vmap_mlp_speed_decorator[True-False]	0.8044ms	0.6495ms	1.5395 KOps/s	1.5381 KOps/s	$\color{#35bf28}+0.09\%$
test_vmap_mlp_speed_decorator[False-True]	0.7363ms	0.5732ms	1.7444 KOps/s	1.7443 KOps/s	$+0.01\%$
test_vmap_mlp_speed_decorator[False-False]	0.7310ms	0.5740ms	1.7423 KOps/s	1.7490 KOps/s	$\color{#d91a1a}-0.38\%$
test_vmap_transformer_speed[True-True]	7.8831ms	7.7146ms	129.6246 Ops/s	130.7554 Ops/s	$\color{#d91a1a}-0.86\%$
test_vmap_transformer_speed[True-False]	8.1360ms	7.8055ms	128.1151 Ops/s	129.8291 Ops/s	$\color{#d91a1a}-1.32\%$
test_vmap_transformer_speed[False-True]	8.0755ms	7.7213ms	129.5117 Ops/s	131.4826 Ops/s	$\color{#d91a1a}-1.50\%$
test_vmap_transformer_speed[False-False]	8.0660ms	7.6984ms	129.8975 Ops/s	130.6666 Ops/s	$\color{#d91a1a}-0.59\%$
test_vmap_transformer_speed_decorator[True-True]	19.6687ms	18.8459ms	53.0620 Ops/s	52.9373 Ops/s	$\color{#35bf28}+0.24\%$
test_vmap_transformer_speed_decorator[True-False]	19.8342ms	18.8950ms	52.9242 Ops/s	53.2641 Ops/s	$\color{#d91a1a}-0.64\%$
test_vmap_transformer_speed_decorator[False-True]	19.3570ms	18.6896ms	53.5056 Ops/s	53.7910 Ops/s	$\color{#d91a1a}-0.53\%$
test_vmap_transformer_speed_decorator[False-False]	19.1860ms	18.6155ms	53.7188 Ops/s	53.6210 Ops/s	$\color{#35bf28}+0.18\%$
test_to_module_speed[True]	1.7390ms	1.5143ms	660.3522 Ops/s	646.5715 Ops/s	$\color{#35bf28}+2.13\%$
test_to_module_speed[False]	1.6207ms	1.5021ms	665.7400 Ops/s	656.2163 Ops/s	$\color{#35bf28}+1.45\%$
test_tc_init	58.7110μs	27.9355μs	35.7967 KOps/s	37.1084 KOps/s	$\color{#d91a1a}-3.53\%$
test_tc_init_nested	83.0110μs	58.3372μs	17.1417 KOps/s	19.1756 KOps/s	$\textbf{\color{#d91a1a}-10.61\%}$
test_tc_first_layer_tensor	0.7555μs	0.3660μs	2.7322 MOps/s	2.7409 MOps/s	$\color{#d91a1a}-0.32\%$
test_tc_first_layer_nontensor	1.3916μs	0.3945μs	2.5351 MOps/s	2.5604 MOps/s	$\color{#d91a1a}-0.99\%$
test_tc_second_layer_tensor	4.2582μs	0.9806μs	1.0198 MOps/s	1.0223 MOps/s	$\color{#d91a1a}-0.25\%$
test_tc_second_layer_nontensor	4.3118μs	0.8343μs	1.1986 MOps/s	1.2046 MOps/s	$\color{#d91a1a}-0.49\%$
test_unbind	0.1039s	8.0845ms	123.6929 Ops/s	147.4537 Ops/s	$\textbf{\color{#d91a1a}-16.11\%}$
test_full_like	13.8535ms	13.2983ms	75.1975 Ops/s	72.4146 Ops/s	$\color{#35bf28}+3.84\%$
test_zeros_like	8.0761ms	7.8372ms	127.5960 Ops/s	125.9241 Ops/s	$\color{#35bf28}+1.33\%$
test_ones_like	8.1415ms	7.8352ms	127.6299 Ops/s	126.3662 Ops/s	$\color{#35bf28}+1.00\%$
test_clone	10.8754ms	9.5871ms	104.3063 Ops/s	102.3626 Ops/s	$\color{#35bf28}+1.90\%$
test_squeeze	55.5400μs	11.0222μs	90.7260 KOps/s	91.5703 KOps/s	$\color{#d91a1a}-0.92\%$
test_unsqueeze	0.1826ms	50.9967μs	19.6091 KOps/s	19.8140 KOps/s	$\color{#d91a1a}-1.03\%$
test_split	0.2367ms	95.1297μs	10.5120 KOps/s	10.5573 KOps/s	$\color{#d91a1a}-0.43\%$
test_permute	0.2486ms	0.1085ms	9.2133 KOps/s	9.2312 KOps/s	$\color{#d91a1a}-0.19\%$
test_stack	28.7399ms	27.9310ms	35.8025 Ops/s	35.4222 Ops/s	$\color{#35bf28}+1.07\%$
test_cat	28.6212ms	27.8993ms	35.8432 Ops/s	35.3417 Ops/s	$\color{#35bf28}+1.42\%$