[CI] Fix benchmark on gpu

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 105. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	32.8110μs	15.0174μs	66.5896 KOps/s	67.6372 KOps/s	$\color{#d91a1a}-1.55\%$
test_plain_set_stack_nested	0.2666ms	0.1389ms	7.2015 KOps/s	7.1892 KOps/s	$\color{#35bf28}+0.17\%$
test_plain_set_nested_inplace	44.0530μs	18.1322μs	55.1506 KOps/s	56.4284 KOps/s	$\color{#d91a1a}-2.26\%$
test_plain_set_stack_nested_inplace	0.2174ms	0.1715ms	5.8316 KOps/s	5.9171 KOps/s	$\color{#d91a1a}-1.44\%$
test_items	24.5160μs	2.5501μs	392.1389 KOps/s	407.9049 KOps/s	$\color{#d91a1a}-3.87\%$
test_items_nested	0.8962ms	0.2737ms	3.6538 KOps/s	3.7161 KOps/s	$\color{#d91a1a}-1.68\%$
test_items_nested_locked	1.3444ms	0.2868ms	3.4873 KOps/s	3.7543 KOps/s	$\textbf{\color{#d91a1a}-7.11\%}$
test_items_nested_leaf	0.5263ms	0.1685ms	5.9350 KOps/s	6.0886 KOps/s	$\color{#d91a1a}-2.52\%$
test_items_stack_nested	1.4919ms	1.3734ms	728.1098 Ops/s	729.5118 Ops/s	$\color{#d91a1a}-0.19\%$
test_items_stack_nested_leaf	1.3394ms	1.2428ms	804.6587 Ops/s	799.6878 Ops/s	$\color{#35bf28}+0.62\%$
test_items_stack_nested_locked	1.8183ms	0.7459ms	1.3407 KOps/s	1.3335 KOps/s	$\color{#35bf28}+0.54\%$
test_keys	28.0220μs	3.8352μs	260.7409 KOps/s	259.6660 KOps/s	$\color{#35bf28}+0.41\%$
test_keys_nested	0.5208ms	0.1368ms	7.3076 KOps/s	6.8070 KOps/s	$\textbf{\color{#35bf28}+7.35\%}$
test_keys_nested_locked	0.2361ms	0.1370ms	7.3018 KOps/s	7.2956 KOps/s	$\color{#35bf28}+0.09\%$
test_keys_nested_leaf	0.5872ms	0.1348ms	7.4209 KOps/s	7.3736 KOps/s	$\color{#35bf28}+0.64\%$
test_keys_stack_nested	3.0926ms	1.2718ms	786.2568 Ops/s	790.5633 Ops/s	$\color{#d91a1a}-0.54\%$
test_keys_stack_nested_leaf	1.5245ms	1.2673ms	789.0876 Ops/s	788.9020 Ops/s	$\color{#35bf28}+0.02\%$
test_keys_stack_nested_locked	0.7326ms	0.6194ms	1.6145 KOps/s	1.5757 KOps/s	$\color{#35bf28}+2.46\%$
test_values	6.7125μs	1.1593μs	862.5927 KOps/s	867.8786 KOps/s	$\color{#d91a1a}-0.61\%$
test_values_nested	0.1361ms	47.9432μs	20.8580 KOps/s	21.5702 KOps/s	$\color{#d91a1a}-3.30\%$
test_values_nested_locked	91.0600μs	48.4163μs	20.6542 KOps/s	20.3825 KOps/s	$\color{#35bf28}+1.33\%$
test_values_nested_leaf	2.4463ms	43.0675μs	23.2194 KOps/s	24.0572 KOps/s	$\color{#d91a1a}-3.48\%$
test_values_stack_nested	1.2635ms	1.0961ms	912.3259 Ops/s	915.8886 Ops/s	$\color{#d91a1a}-0.39\%$
test_values_stack_nested_leaf	1.3428ms	1.0947ms	913.4774 Ops/s	920.9305 Ops/s	$\color{#d91a1a}-0.81\%$
test_values_stack_nested_locked	0.8541ms	0.4901ms	2.0405 KOps/s	2.0166 KOps/s	$\color{#35bf28}+1.18\%$
test_membership	13.4450μs	1.3328μs	750.3026 KOps/s	739.9266 KOps/s	$\color{#35bf28}+1.40\%$
test_membership_nested	19.6170μs	2.8144μs	355.3147 KOps/s	362.0207 KOps/s	$\color{#d91a1a}-1.85\%$
test_membership_nested_leaf	17.5730μs	2.8470μs	351.2500 KOps/s	358.2648 KOps/s	$\color{#d91a1a}-1.96\%$
test_membership_stacked_nested	35.6060μs	11.5916μs	86.2693 KOps/s	88.6321 KOps/s	$\color{#d91a1a}-2.67\%$
test_membership_stacked_nested_leaf	36.2870μs	11.4610μs	87.2528 KOps/s	87.9574 KOps/s	$\color{#d91a1a}-0.80\%$
test_membership_nested_last	22.3620μs	5.8782μs	170.1201 KOps/s	150.8290 KOps/s	$\textbf{\color{#35bf28}+12.79\%}$
test_membership_nested_leaf_last	21.0290μs	5.8709μs	170.3310 KOps/s	173.2386 KOps/s	$\color{#d91a1a}-1.68\%$
test_membership_stacked_nested_last	0.2519ms	0.1782ms	5.6110 KOps/s	5.6937 KOps/s	$\color{#d91a1a}-1.45\%$
test_membership_stacked_nested_leaf_last	44.2820μs	13.3764μs	74.7588 KOps/s	75.3404 KOps/s	$\color{#d91a1a}-0.77\%$
test_nested_getleaf	29.8860μs	11.9594μs	83.6162 KOps/s	84.1485 KOps/s	$\color{#d91a1a}-0.63\%$
test_nested_get	31.0880μs	11.2533μs	88.8625 KOps/s	89.5569 KOps/s	$\color{#d91a1a}-0.78\%$
test_stacked_getleaf	3.4390ms	0.5668ms	1.7644 KOps/s	1.7771 KOps/s	$\color{#d91a1a}-0.71\%$
test_stacked_get	0.6164ms	0.5376ms	1.8602 KOps/s	1.8660 KOps/s	$\color{#d91a1a}-0.31\%$
test_nested_getitemleaf	38.5510μs	11.7643μs	85.0029 KOps/s	81.7771 KOps/s	$\color{#35bf28}+3.94\%$
test_nested_getitem	32.8710μs	11.2684μs	88.7438 KOps/s	87.0752 KOps/s	$\color{#35bf28}+1.92\%$
test_stacked_getitemleaf	0.7182ms	0.5639ms	1.7735 KOps/s	1.7817 KOps/s	$\color{#d91a1a}-0.46\%$
test_stacked_getitem	0.6847ms	0.5354ms	1.8677 KOps/s	1.8639 KOps/s	$\color{#35bf28}+0.20\%$
test_lock_nested	50.2652ms	0.9497ms	1.0529 KOps/s	1.1293 KOps/s	$\textbf{\color{#d91a1a}-6.76\%}$
test_lock_stack_nested	68.1261ms	12.1231ms	82.4871 Ops/s	76.0472 Ops/s	$\textbf{\color{#35bf28}+8.47\%}$
test_unlock_nested	46.9563ms	0.9394ms	1.0645 KOps/s	1.0689 KOps/s	$\color{#d91a1a}-0.40\%$
test_unlock_stack_nested	58.9615ms	12.4471ms	80.3397 Ops/s	73.3793 Ops/s	$\textbf{\color{#35bf28}+9.49\%}$
test_flatten_speed	0.7517ms	0.6620ms	1.5106 KOps/s	1.5325 KOps/s	$\color{#d91a1a}-1.43\%$
test_unflatten_speed	1.7602ms	1.1526ms	867.6162 Ops/s	879.6400 Ops/s	$\color{#d91a1a}-1.37\%$
test_common_ops	0.7766ms	0.6301ms	1.5871 KOps/s	1.5271 KOps/s	$\color{#35bf28}+3.93\%$
test_creation	25.7080μs	2.0901μs	478.4375 KOps/s	451.6151 KOps/s	$\textbf{\color{#35bf28}+5.94\%}$
test_creation_empty	23.9950μs	7.0561μs	141.7208 KOps/s	138.3305 KOps/s	$\color{#35bf28}+2.45\%$
test_creation_nested_1	34.4940μs	11.0209μs	90.7365 KOps/s	89.1912 KOps/s	$\color{#35bf28}+1.73\%$
test_creation_nested_2	55.2430μs	13.4992μs	74.0787 KOps/s	74.6328 KOps/s	$\color{#d91a1a}-0.74\%$
test_clone	64.3600μs	10.6614μs	93.7959 KOps/s	93.8949 KOps/s	$\color{#d91a1a}-0.11\%$
test_getitem[int]	31.3990μs	13.0654μs	76.5381 KOps/s	76.0302 KOps/s	$\color{#35bf28}+0.67\%$
test_getitem[slice_int]	0.3794ms	32.9490μs	30.3500 KOps/s	32.1224 KOps/s	$\textbf{\color{#d91a1a}-5.52\%}$
test_getitem[range]	0.1709ms	54.1174μs	18.4783 KOps/s	18.7081 KOps/s	$\color{#d91a1a}-1.23\%$
test_getitem[tuple]	65.8720μs	23.6103μs	42.3545 KOps/s	41.0725 KOps/s	$\color{#35bf28}+3.12\%$
test_getitem[list]	0.1769ms	48.6429μs	20.5580 KOps/s	18.7149 KOps/s	$\textbf{\color{#35bf28}+9.85\%}$
test_setitem_dim[int]	52.4380μs	25.6440μs	38.9954 KOps/s	38.0116 KOps/s	$\color{#35bf28}+2.59\%$
test_setitem_dim[slice_int]	90.6290μs	49.6643μs	20.1352 KOps/s	19.9082 KOps/s	$\color{#35bf28}+1.14\%$
test_setitem_dim[range]	0.1027ms	70.2550μs	14.2339 KOps/s	14.1430 KOps/s	$\color{#35bf28}+0.64\%$
test_setitem_dim[tuple]	80.9810μs	39.0954μs	25.5785 KOps/s	25.5216 KOps/s	$\color{#35bf28}+0.22\%$
test_setitem	65.5920μs	14.9509μs	66.8856 KOps/s	67.1441 KOps/s	$\color{#d91a1a}-0.38\%$
test_set	62.9570μs	14.2338μs	70.2553 KOps/s	70.4042 KOps/s	$\color{#d91a1a}-0.21\%$
test_set_shared	3.3062ms	0.1576ms	6.3470 KOps/s	6.3694 KOps/s	$\color{#d91a1a}-0.35\%$
test_update	0.1205ms	18.8311μs	53.1037 KOps/s	52.8918 KOps/s	$\color{#35bf28}+0.40\%$
test_update_nested	73.2260μs	27.1920μs	36.7755 KOps/s	36.2440 KOps/s	$\color{#35bf28}+1.47\%$
test_set_nested	85.6700μs	16.2125μs	61.6809 KOps/s	62.6507 KOps/s	$\color{#d91a1a}-1.55\%$
test_set_nested_new	87.1520μs	22.3787μs	44.6854 KOps/s	43.2272 KOps/s	$\color{#35bf28}+3.37\%$
test_select	94.5460μs	45.6572μs	21.9023 KOps/s	21.2024 KOps/s	$\color{#35bf28}+3.30\%$
test_unbind_speed	0.7873ms	0.2888ms	3.4622 KOps/s	3.5481 KOps/s	$\color{#d91a1a}-2.42\%$
test_unbind_speed_stack0	54.8914ms	4.2757ms	233.8787 Ops/s	221.7600 Ops/s	$\textbf{\color{#35bf28}+5.46\%}$
test_unbind_speed_stack1	1.9947μs	0.5964μs	1.6768 MOps/s	1.6065 MOps/s	$\color{#35bf28}+4.38\%$
test_creation[device0]	3.4215ms	0.3029ms	3.3012 KOps/s	3.3388 KOps/s	$\color{#d91a1a}-1.13\%$
test_creation_from_tensor	0.7272ms	0.3324ms	3.0080 KOps/s	2.9701 KOps/s	$\color{#35bf28}+1.27\%$
test_add_one[memmap_tensor0]	0.1274ms	26.5851μs	37.6151 KOps/s	38.7605 KOps/s	$\color{#d91a1a}-2.96\%$
test_contiguous[memmap_tensor0]	23.1530μs	5.7710μs	173.2792 KOps/s	173.4526 KOps/s	$\color{#d91a1a}-0.10\%$
test_stack[memmap_tensor0]	51.4560μs	19.6720μs	50.8337 KOps/s	51.7013 KOps/s	$\color{#d91a1a}-1.68\%$
test_memmaptd_index	0.2390ms	0.1815ms	5.5102 KOps/s	5.3761 KOps/s	$\color{#35bf28}+2.49\%$
test_memmaptd_index_astensor	0.2947ms	0.2425ms	4.1242 KOps/s	4.0411 KOps/s	$\color{#35bf28}+2.06\%$
test_memmaptd_index_op	0.8274ms	0.4968ms	2.0129 KOps/s	2.1646 KOps/s	$\textbf{\color{#d91a1a}-7.01\%}$
test_reshape_pytree	68.9180μs	22.7993μs	43.8611 KOps/s	43.5543 KOps/s	$\color{#35bf28}+0.70\%$
test_reshape_td	55.6240μs	20.7115μs	48.2823 KOps/s	48.4528 KOps/s	$\color{#d91a1a}-0.35\%$
test_view_pytree	56.4550μs	22.6816μs	44.0885 KOps/s	43.4113 KOps/s	$\color{#35bf28}+1.56\%$
test_view_td	16.5510μs	4.0878μs	244.6302 KOps/s	246.1416 KOps/s	$\color{#d91a1a}-0.61\%$
test_unbind_pytree	68.4680μs	26.2128μs	38.1493 KOps/s	37.6576 KOps/s	$\color{#35bf28}+1.31\%$
test_unbind_td	89.2370μs	39.4235μs	25.3656 KOps/s	25.0111 KOps/s	$\color{#35bf28}+1.42\%$
test_split_pytree	56.8960μs	25.9174μs	38.5842 KOps/s	37.7629 KOps/s	$\color{#35bf28}+2.17\%$
test_split_td	0.1585ms	74.4740μs	13.4275 KOps/s	13.2572 KOps/s	$\color{#35bf28}+1.28\%$
test_add_pytree	71.1320μs	32.0725μs	31.1793 KOps/s	31.0985 KOps/s	$\color{#35bf28}+0.26\%$
test_add_td	0.1240ms	42.2238μs	23.6833 KOps/s	23.6382 KOps/s	$\color{#35bf28}+0.19\%$
test_distributed	17.6330μs	5.8927μs	169.7007 KOps/s	167.8953 KOps/s	$\color{#35bf28}+1.08\%$
test_tdmodule	1.0494ms	21.9455μs	45.5674 KOps/s	49.2205 KOps/s	$\textbf{\color{#d91a1a}-7.42\%}$
test_tdmodule_dispatch	0.1923ms	37.8644μs	26.4100 KOps/s	27.3217 KOps/s	$\color{#d91a1a}-3.34\%$
test_tdseq	0.1180ms	23.1662μs	43.1663 KOps/s	43.7531 KOps/s	$\color{#d91a1a}-1.34\%$
test_tdseq_dispatch	0.1322ms	40.4695μs	24.7100 KOps/s	24.5219 KOps/s	$\color{#35bf28}+0.77\%$
test_instantiation_functorch	2.2132ms	1.3241ms	755.2473 Ops/s	777.3539 Ops/s	$\color{#d91a1a}-2.84\%$
test_instantiation_td	71.4649ms	1.1474ms	871.5006 Ops/s	974.1470 Ops/s	$\textbf{\color{#d91a1a}-10.54\%}$
test_exec_functorch	0.3823ms	0.1502ms	6.6570 KOps/s	6.7436 KOps/s	$\color{#d91a1a}-1.28\%$
test_exec_td	0.2310ms	0.1493ms	6.6977 KOps/s	7.1063 KOps/s	$\textbf{\color{#d91a1a}-5.75\%}$
test_vmap_mlp_speed[True-True]	1.1895ms	0.8553ms	1.1692 KOps/s	1.2024 KOps/s	$\color{#d91a1a}-2.77\%$
test_vmap_mlp_speed[True-False]	0.6074ms	0.4690ms	2.1322 KOps/s	2.1745 KOps/s	$\color{#d91a1a}-1.95\%$
test_vmap_mlp_speed[False-True]	1.0908ms	0.7452ms	1.3419 KOps/s	1.3735 KOps/s	$\color{#d91a1a}-2.30\%$
test_vmap_mlp_speed[False-False]	0.5439ms	0.3900ms	2.5638 KOps/s	2.5208 KOps/s	$\color{#35bf28}+1.71\%$

[github-actions]

$\color{#35bf28}\textsf{\Large✔\kern{0.2cm}\normalsize OK}$ Result of GPU Benchmark Tests

Total Benchmarks: 115. Improved: $\large\color{#35bf28}8$. Worsened: $\large\color{#d91a1a}0$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	92.4110μs	12.1926μs	82.0170 KOps/s	81.5906 KOps/s	$\color{#35bf28}+0.52\%$
test_plain_set_stack_nested	0.1279ms	0.1150ms	8.6971 KOps/s	8.6189 KOps/s	$\color{#35bf28}+0.91\%$
test_plain_set_nested_inplace	31.3700μs	14.4738μs	69.0903 KOps/s	68.6399 KOps/s	$\color{#35bf28}+0.66\%$
test_plain_set_stack_nested_inplace	0.2091ms	0.1409ms	7.0986 KOps/s	7.0937 KOps/s	$\color{#35bf28}+0.07\%$
test_items	20.3410μs	4.7345μs	211.2146 KOps/s	210.1243 KOps/s	$\color{#35bf28}+0.52\%$
test_items_nested	0.3898ms	0.3367ms	2.9703 KOps/s	2.9746 KOps/s	$\color{#d91a1a}-0.14\%$
test_items_nested_locked	0.3786ms	0.3370ms	2.9677 KOps/s	2.9677 KOps/s	$-0.00\%$
test_items_nested_leaf	0.2469ms	0.1992ms	5.0201 KOps/s	5.0349 KOps/s	$\color{#d91a1a}-0.29\%$
test_items_stack_nested	1.4571ms	1.4167ms	705.8879 Ops/s	705.3260 Ops/s	$\color{#35bf28}+0.08\%$
test_items_stack_nested_leaf	1.3258ms	1.2488ms	800.7984 Ops/s	797.0642 Ops/s	$\color{#35bf28}+0.47\%$
test_items_stack_nested_locked	0.8452ms	0.7923ms	1.2621 KOps/s	1.2604 KOps/s	$\color{#35bf28}+0.14\%$
test_keys	26.3600μs	4.6179μs	216.5477 KOps/s	202.2872 KOps/s	$\textbf{\color{#35bf28}+7.05\%}$
test_keys_nested	0.5408ms	90.1231μs	11.0959 KOps/s	10.4007 KOps/s	$\textbf{\color{#35bf28}+6.68\%}$
test_keys_nested_locked	0.1053ms	89.7879μs	11.1374 KOps/s	10.9854 KOps/s	$\color{#35bf28}+1.38\%$
test_keys_nested_leaf	0.1603ms	81.5775μs	12.2583 KOps/s	12.0583 KOps/s	$\color{#35bf28}+1.66\%$
test_keys_stack_nested	1.2893ms	1.2085ms	827.4975 Ops/s	830.9497 Ops/s	$\color{#d91a1a}-0.42\%$
test_keys_stack_nested_leaf	1.3297ms	1.2116ms	825.3555 Ops/s	839.2417 Ops/s	$\color{#d91a1a}-1.65\%$
test_keys_stack_nested_locked	0.6331ms	0.5852ms	1.7088 KOps/s	1.7294 KOps/s	$\color{#d91a1a}-1.19\%$
test_values	9.9303μs	1.8838μs	530.8503 KOps/s	521.7376 KOps/s	$\color{#35bf28}+1.75\%$
test_values_nested	65.0010μs	43.4554μs	23.0121 KOps/s	23.0069 KOps/s	$\color{#35bf28}+0.02\%$
test_values_nested_locked	58.2410μs	43.5325μs	22.9713 KOps/s	22.9603 KOps/s	$\color{#35bf28}+0.05\%$
test_values_nested_leaf	52.0410μs	37.7098μs	26.5183 KOps/s	26.5824 KOps/s	$\color{#d91a1a}-0.24\%$
test_values_stack_nested	1.1235ms	1.0724ms	932.5194 Ops/s	930.0000 Ops/s	$\color{#35bf28}+0.27\%$
test_values_stack_nested_leaf	1.2352ms	1.0669ms	937.2617 Ops/s	938.8818 Ops/s	$\color{#d91a1a}-0.17\%$
test_values_stack_nested_locked	0.5761ms	0.4784ms	2.0905 KOps/s	2.1031 KOps/s	$\color{#d91a1a}-0.60\%$
test_membership	3.5540μs	0.9315μs	1.0736 MOps/s	1.0686 MOps/s	$\color{#35bf28}+0.47\%$
test_membership_nested	13.9600μs	2.1943μs	455.7247 KOps/s	448.8045 KOps/s	$\color{#35bf28}+1.54\%$
test_membership_nested_leaf	17.1600μs	2.1905μs	456.5150 KOps/s	449.2791 KOps/s	$\color{#35bf28}+1.61\%$
test_membership_stacked_nested	33.8800μs	10.8392μs	92.2577 KOps/s	92.1503 KOps/s	$\color{#35bf28}+0.12\%$
test_membership_stacked_nested_leaf	29.1020μs	10.7255μs	93.2357 KOps/s	93.0755 KOps/s	$\color{#35bf28}+0.17\%$
test_membership_nested_last	28.0600μs	4.5002μs	222.2124 KOps/s	218.4311 KOps/s	$\color{#35bf28}+1.73\%$
test_membership_nested_leaf_last	23.0800μs	4.5662μs	218.9991 KOps/s	218.9421 KOps/s	$\color{#35bf28}+0.03\%$
test_membership_stacked_nested_last	0.1845ms	0.1439ms	6.9470 KOps/s	7.0447 KOps/s	$\color{#d91a1a}-1.39\%$
test_membership_stacked_nested_leaf_last	27.8100μs	12.5917μs	79.4172 KOps/s	79.2715 KOps/s	$\color{#35bf28}+0.18\%$
test_nested_getleaf	22.4110μs	9.3837μs	106.5683 KOps/s	106.4895 KOps/s	$\color{#35bf28}+0.07\%$
test_nested_get	26.1700μs	8.8224μs	113.3477 KOps/s	112.6540 KOps/s	$\color{#35bf28}+0.62\%$
test_stacked_getleaf	0.5808ms	0.5299ms	1.8871 KOps/s	1.9144 KOps/s	$\color{#d91a1a}-1.43\%$
test_stacked_get	0.5328ms	0.5024ms	1.9903 KOps/s	2.0218 KOps/s	$\color{#d91a1a}-1.56\%$
test_nested_getitemleaf	23.2900μs	9.4181μs	106.1787 KOps/s	105.9320 KOps/s	$\color{#35bf28}+0.23\%$
test_nested_getitem	21.4100μs	8.8945μs	112.4287 KOps/s	111.6064 KOps/s	$\color{#35bf28}+0.74\%$
test_stacked_getitemleaf	0.5832ms	0.5189ms	1.9272 KOps/s	1.9254 KOps/s	$\color{#35bf28}+0.09\%$
test_stacked_getitem	0.5453ms	0.4912ms	2.0357 KOps/s	2.0378 KOps/s	$\color{#d91a1a}-0.11\%$
test_lock_nested	46.7588ms	0.9297ms	1.0757 KOps/s	1.1272 KOps/s	$\color{#d91a1a}-4.58\%$
test_lock_stack_nested	57.0526ms	11.9637ms	83.5862 Ops/s	82.2695 Ops/s	$\color{#35bf28}+1.60\%$
test_unlock_nested	45.6521ms	0.9540ms	1.0483 KOps/s	1.0332 KOps/s	$\color{#35bf28}+1.46\%$
test_unlock_stack_nested	57.8770ms	12.7480ms	78.4439 Ops/s	76.7368 Ops/s	$\color{#35bf28}+2.22\%$
test_flatten_speed	0.6116ms	0.5464ms	1.8301 KOps/s	1.8347 KOps/s	$\color{#d91a1a}-0.25\%$
test_unflatten_speed	1.0393ms	0.9794ms	1.0210 KOps/s	1.0141 KOps/s	$\color{#35bf28}+0.69\%$
test_common_ops	0.6340ms	0.5600ms	1.7856 KOps/s	1.7882 KOps/s	$\color{#d91a1a}-0.15\%$
test_creation	27.1710μs	1.7602μs	568.1028 KOps/s	575.2659 KOps/s	$\color{#d91a1a}-1.25\%$
test_creation_empty	19.7600μs	5.9800μs	167.2237 KOps/s	168.7143 KOps/s	$\color{#d91a1a}-0.88\%$
test_creation_nested_1	22.7000μs	8.8453μs	113.0539 KOps/s	115.9023 KOps/s	$\color{#d91a1a}-2.46\%$
test_creation_nested_2	29.6200μs	10.6196μs	94.1654 KOps/s	95.8110 KOps/s	$\color{#d91a1a}-1.72\%$
test_clone	70.2010μs	11.5695μs	86.4341 KOps/s	87.7803 KOps/s	$\color{#d91a1a}-1.53\%$
test_getitem[int]	27.0000μs	12.3105μs	81.2313 KOps/s	80.3624 KOps/s	$\color{#35bf28}+1.08\%$
test_getitem[slice_int]	53.7300μs	27.7116μs	36.0860 KOps/s	35.9572 KOps/s	$\color{#35bf28}+0.36\%$
test_getitem[range]	78.0020μs	46.8662μs	21.3373 KOps/s	21.2215 KOps/s	$\color{#35bf28}+0.55\%$
test_getitem[tuple]	57.2810μs	23.5850μs	42.3998 KOps/s	42.7968 KOps/s	$\color{#d91a1a}-0.93\%$
test_getitem[list]	0.1768ms	44.0559μs	22.6984 KOps/s	22.4660 KOps/s	$\color{#35bf28}+1.03\%$
test_setitem_dim[int]	40.2010μs	24.7997μs	40.3231 KOps/s	37.0647 KOps/s	$\textbf{\color{#35bf28}+8.79\%}$
test_setitem_dim[slice_int]	60.7910μs	43.8856μs	22.7865 KOps/s	21.7280 KOps/s	$\color{#35bf28}+4.87\%$
test_setitem_dim[range]	79.3720μs	61.2701μs	16.3212 KOps/s	15.8088 KOps/s	$\color{#35bf28}+3.24\%$
test_setitem_dim[tuple]	59.4210μs	37.9338μs	26.3617 KOps/s	25.0789 KOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_setitem	77.4310μs	14.6363μs	68.3231 KOps/s	66.9711 KOps/s	$\color{#35bf28}+2.02\%$
test_set	58.6610μs	13.9550μs	71.6591 KOps/s	69.5657 KOps/s	$\color{#35bf28}+3.01\%$
test_set_shared	0.1862ms	0.1130ms	8.8505 KOps/s	8.9332 KOps/s	$\color{#d91a1a}-0.93\%$
test_update	77.4910μs	17.5000μs	57.1428 KOps/s	57.3233 KOps/s	$\color{#d91a1a}-0.31\%$
test_update_nested	83.3620μs	24.4276μs	40.9372 KOps/s	40.5192 KOps/s	$\color{#35bf28}+1.03\%$
test_set_nested	70.9910μs	15.4373μs	64.7783 KOps/s	64.7432 KOps/s	$\color{#35bf28}+0.05\%$
test_set_nested_new	79.1520μs	20.5873μs	48.5736 KOps/s	47.5503 KOps/s	$\color{#35bf28}+2.15\%$
test_select	93.5310μs	42.3815μs	23.5952 KOps/s	23.4565 KOps/s	$\color{#35bf28}+0.59\%$
test_to	75.1610μs	53.9321μs	18.5418 KOps/s	18.7033 KOps/s	$\color{#d91a1a}-0.86\%$
test_to_nonblocking	53.4010μs	34.3596μs	29.1040 KOps/s	27.7752 KOps/s	$\color{#35bf28}+4.78\%$
test_unbind_speed	0.3863ms	0.2832ms	3.5307 KOps/s	3.5080 KOps/s	$\color{#35bf28}+0.64\%$
test_unbind_speed_stack0	51.7947ms	3.7138ms	269.2655 Ops/s	254.0564 Ops/s	$\textbf{\color{#35bf28}+5.99\%}$
test_unbind_speed_stack1	1.3010μs	0.4933μs	2.0273 MOps/s	2.0390 MOps/s	$\color{#d91a1a}-0.57\%$
test_creation[device0]	0.7475ms	0.3088ms	3.2378 KOps/s	3.2278 KOps/s	$\color{#35bf28}+0.31\%$
test_creation[device1]	0.6625ms	0.3115ms	3.2104 KOps/s	3.1750 KOps/s	$\color{#35bf28}+1.11\%$
test_creation_from_tensor	0.6618ms	0.3373ms	2.9644 KOps/s	2.8993 KOps/s	$\color{#35bf28}+2.25\%$
test_add_one[memmap_tensor0]	88.1920μs	24.6183μs	40.6202 KOps/s	38.4322 KOps/s	$\textbf{\color{#35bf28}+5.69\%}$
test_add_one[memmap_tensor1]	0.2031ms	74.6339μs	13.3987 KOps/s	13.1869 KOps/s	$\color{#35bf28}+1.61\%$
test_contiguous[memmap_tensor0]	26.0700μs	5.8537μs	170.8336 KOps/s	167.2464 KOps/s	$\color{#35bf28}+2.14\%$
test_contiguous[memmap_tensor1]	46.7910μs	21.5969μs	46.3029 KOps/s	45.7398 KOps/s	$\color{#35bf28}+1.23\%$
test_stack[memmap_tensor0]	47.1310μs	19.3575μs	51.6597 KOps/s	48.6033 KOps/s	$\textbf{\color{#35bf28}+6.29\%}$
test_stack[memmap_tensor1]	0.1543ms	74.3075μs	13.4576 KOps/s	12.7174 KOps/s	$\textbf{\color{#35bf28}+5.82\%}$
test_memmaptd_index	0.2657ms	0.2155ms	4.6397 KOps/s	4.4889 KOps/s	$\color{#35bf28}+3.36\%$
test_memmaptd_index_astensor	0.3294ms	0.2764ms	3.6182 KOps/s	3.5861 KOps/s	$\color{#35bf28}+0.90\%$
test_memmaptd_index_op	0.5821ms	0.5287ms	1.8915 KOps/s	1.8745 KOps/s	$\color{#35bf28}+0.91\%$
test_reshape_pytree	46.0210μs	21.1115μs	47.3676 KOps/s	46.0152 KOps/s	$\color{#35bf28}+2.94\%$
test_reshape_td	43.3010μs	21.1460μs	47.2903 KOps/s	47.7587 KOps/s	$\color{#d91a1a}-0.98\%$
test_view_pytree	46.0310μs	20.5103μs	48.7561 KOps/s	48.2243 KOps/s	$\color{#35bf28}+1.10\%$
test_view_td	15.3800μs	3.3233μs	300.9022 KOps/s	300.6594 KOps/s	$\color{#35bf28}+0.08\%$
test_unbind_pytree	42.6510μs	25.7700μs	38.8048 KOps/s	37.6719 KOps/s	$\color{#35bf28}+3.01\%$
test_unbind_td	58.6710μs	40.4776μs	24.7050 KOps/s	24.1791 KOps/s	$\color{#35bf28}+2.17\%$
test_split_pytree	44.4810μs	23.8097μs	41.9997 KOps/s	41.9980 KOps/s	$+0.00\%$
test_split_td	90.4620μs	71.3908μs	14.0074 KOps/s	13.9989 KOps/s	$\color{#35bf28}+0.06\%$
test_add_pytree	50.0010μs	32.9705μs	30.3301 KOps/s	30.6131 KOps/s	$\color{#d91a1a}-0.92\%$
test_add_td	67.3210μs	42.6716μs	23.4348 KOps/s	23.1827 KOps/s	$\color{#35bf28}+1.09\%$
test_distributed	18.9900μs	5.5043μs	181.6777 KOps/s	179.2840 KOps/s	$\color{#35bf28}+1.34\%$
test_tdmodule	30.9700μs	16.2118μs	61.6833 KOps/s	60.0380 KOps/s	$\color{#35bf28}+2.74\%$
test_tdmodule_dispatch	0.1227ms	30.4425μs	32.8488 KOps/s	32.3439 KOps/s	$\color{#35bf28}+1.56\%$
test_tdseq	34.7010μs	19.4107μs	51.5180 KOps/s	51.1183 KOps/s	$\color{#35bf28}+0.78\%$
test_tdseq_dispatch	49.9810μs	33.9604μs	29.4461 KOps/s	29.3242 KOps/s	$\color{#35bf28}+0.42\%$
test_instantiation_functorch	1.7282ms	1.6788ms	595.6772 Ops/s	587.4243 Ops/s	$\color{#35bf28}+1.40\%$
test_instantiation_td	1.7949ms	1.2052ms	829.7099 Ops/s	834.2530 Ops/s	$\color{#d91a1a}-0.54\%$
test_exec_functorch	0.2000ms	0.1589ms	6.2952 KOps/s	6.3283 KOps/s	$\color{#d91a1a}-0.52\%$
test_exec_td	0.1798ms	0.1498ms	6.6735 KOps/s	6.7070 KOps/s	$\color{#d91a1a}-0.50\%$
test_vmap_mlp_speed[True-True]	1.1075ms	1.0508ms	951.6783 Ops/s	953.1853 Ops/s	$\color{#d91a1a}-0.16\%$
test_vmap_mlp_speed[True-False]	0.6671ms	0.6215ms	1.6089 KOps/s	1.6038 KOps/s	$\color{#35bf28}+0.32\%$
test_vmap_mlp_speed[False-True]	1.0009ms	0.9636ms	1.0377 KOps/s	1.0374 KOps/s	$\color{#35bf28}+0.03\%$
test_vmap_mlp_speed[False-False]	0.5971ms	0.5577ms	1.7930 KOps/s	1.7193 KOps/s	$\color{#35bf28}+4.29\%$
test_vmap_transformer_speed[True-True]	12.5471ms	12.4505ms	80.3178 Ops/s	79.6500 Ops/s	$\color{#35bf28}+0.84\%$
test_vmap_transformer_speed[True-False]	8.4848ms	8.3449ms	119.8333 Ops/s	118.5951 Ops/s	$\color{#35bf28}+1.04\%$
test_vmap_transformer_speed[False-True]	12.4473ms	12.3517ms	80.9606 Ops/s	80.7120 Ops/s	$\color{#35bf28}+0.31\%$
test_vmap_transformer_speed[False-False]	8.4649ms	8.2917ms	120.6025 Ops/s	119.9105 Ops/s	$\color{#35bf28}+0.58\%$