[BugFix] Use correct default cuda device

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Use correct default cuda device #1161

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}38$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	51.1560μs	22.5351μs	44.3752 KOps/s	47.5697 KOps/s	$\textbf{\color{#d91a1a}-6.72\%}$
test_plain_set_stack_nested	66.8650μs	22.4793μs	44.4854 KOps/s	46.7984 KOps/s	$\color{#d91a1a}-4.94\%$
test_plain_set_nested_inplace	56.7560μs	24.0769μs	41.5336 KOps/s	41.0582 KOps/s	$\color{#35bf28}+1.16\%$
test_plain_set_stack_nested_inplace	79.6300μs	24.1452μs	41.4161 KOps/s	42.8944 KOps/s	$\color{#d91a1a}-3.45\%$
test_items	25.1470μs	4.1751μs	239.5155 KOps/s	237.6470 KOps/s	$\color{#35bf28}+0.79\%$
test_items_nested	0.6204ms	0.4024ms	2.4849 KOps/s	2.4560 KOps/s	$\color{#35bf28}+1.18\%$
test_items_nested_locked	0.8290ms	0.4041ms	2.4748 KOps/s	2.4364 KOps/s	$\color{#35bf28}+1.58\%$
test_items_nested_leaf	0.1474ms	77.3365μs	12.9305 KOps/s	12.8379 KOps/s	$\color{#35bf28}+0.72\%$
test_items_stack_nested	0.8339ms	0.4048ms	2.4702 KOps/s	2.4538 KOps/s	$\color{#35bf28}+0.67\%$
test_items_stack_nested_leaf	0.1500ms	78.8436μs	12.6833 KOps/s	12.5081 KOps/s	$\color{#35bf28}+1.40\%$
test_items_stack_nested_locked	0.5260ms	0.4059ms	2.4636 KOps/s	2.4313 KOps/s	$\color{#35bf28}+1.33\%$
test_keys	23.7340μs	3.4975μs	285.9178 KOps/s	279.5141 KOps/s	$\color{#35bf28}+2.29\%$
test_keys_nested	0.2266ms	0.1636ms	6.1133 KOps/s	6.0532 KOps/s	$\color{#35bf28}+0.99\%$
test_keys_nested_locked	0.8332ms	0.1710ms	5.8479 KOps/s	5.7794 KOps/s	$\color{#35bf28}+1.18\%$
test_keys_nested_leaf	0.2238ms	0.1431ms	6.9858 KOps/s	6.8072 KOps/s	$\color{#35bf28}+2.62\%$
test_keys_stack_nested	0.3428ms	0.1650ms	6.0609 KOps/s	5.9797 KOps/s	$\color{#35bf28}+1.36\%$
test_keys_stack_nested_leaf	0.2068ms	0.1425ms	7.0188 KOps/s	6.9653 KOps/s	$\color{#35bf28}+0.77\%$
test_keys_stack_nested_locked	0.3122ms	0.1708ms	5.8545 KOps/s	5.7856 KOps/s	$\color{#35bf28}+1.19\%$
test_values	9.8466μs	1.0625μs	941.1791 KOps/s	914.7289 KOps/s	$\color{#35bf28}+2.89\%$
test_values_nested	0.1354ms	62.3268μs	16.0445 KOps/s	15.8808 KOps/s	$\color{#35bf28}+1.03\%$
test_values_nested_locked	0.1198ms	62.1124μs	16.0999 KOps/s	15.8111 KOps/s	$\color{#35bf28}+1.83\%$
test_values_nested_leaf	0.1583ms	71.7873μs	13.9300 KOps/s	13.3383 KOps/s	$\color{#35bf28}+4.44\%$
test_values_stack_nested	0.1215ms	62.5333μs	15.9915 KOps/s	15.7583 KOps/s	$\color{#35bf28}+1.48\%$
test_values_stack_nested_leaf	0.1268ms	71.3636μs	14.0128 KOps/s	13.5710 KOps/s	$\color{#35bf28}+3.26\%$
test_values_stack_nested_locked	0.1249ms	63.4397μs	15.7630 KOps/s	15.6838 KOps/s	$\color{#35bf28}+0.51\%$
test_membership	2.5688μs	0.6980μs	1.4327 MOps/s	1.0864 MOps/s	$\textbf{\color{#35bf28}+31.88\%}$
test_membership_nested	41.9290μs	2.9709μs	336.6016 KOps/s	335.6417 KOps/s	$\color{#35bf28}+0.29\%$
test_membership_nested_leaf	28.4840μs	2.9429μs	339.8024 KOps/s	337.0828 KOps/s	$\color{#35bf28}+0.81\%$
test_membership_stacked_nested	44.7940μs	2.9201μs	342.4536 KOps/s	331.6355 KOps/s	$\color{#35bf28}+3.26\%$
test_membership_stacked_nested_leaf	27.2510μs	2.9032μs	344.4430 KOps/s	334.0766 KOps/s	$\color{#35bf28}+3.10\%$
test_membership_nested_last	35.9270μs	4.3820μs	228.2047 KOps/s	224.2816 KOps/s	$\color{#35bf28}+1.75\%$
test_membership_nested_leaf_last	29.4050μs	4.4280μs	225.8373 KOps/s	222.7880 KOps/s	$\color{#35bf28}+1.37\%$
test_membership_stacked_nested_last	33.1320μs	4.4470μs	224.8683 KOps/s	224.7291 KOps/s	$\color{#35bf28}+0.06\%$
test_membership_stacked_nested_leaf_last	47.1780μs	4.4900μs	222.7186 KOps/s	226.3257 KOps/s	$\color{#d91a1a}-1.59\%$
test_nested_getleaf	53.2200μs	10.8259μs	92.3707 KOps/s	92.6302 KOps/s	$\color{#d91a1a}-0.28\%$
test_nested_get	36.9690μs	10.2930μs	97.1532 KOps/s	99.4984 KOps/s	$\color{#d91a1a}-2.36\%$
test_stacked_getleaf	35.4460μs	10.6700μs	93.7207 KOps/s	93.4567 KOps/s	$\color{#35bf28}+0.28\%$
test_stacked_get	55.4740μs	10.2958μs	97.1273 KOps/s	97.8267 KOps/s	$\color{#d91a1a}-0.71\%$
test_nested_getitemleaf	61.7690μs	11.1832μs	89.4201 KOps/s	89.7020 KOps/s	$\color{#d91a1a}-0.31\%$
test_nested_getitem	31.8500μs	10.4766μs	95.4510 KOps/s	96.2504 KOps/s	$\color{#d91a1a}-0.83\%$
test_stacked_getitemleaf	61.7850μs	10.7860μs	92.7124 KOps/s	88.8494 KOps/s	$\color{#35bf28}+4.35\%$
test_stacked_getitem	35.3260μs	10.4255μs	95.9189 KOps/s	96.1860 KOps/s	$\color{#d91a1a}-0.28\%$
test_lock_nested	6.8719ms	0.4735ms	2.1121 KOps/s	2.1847 KOps/s	$\color{#d91a1a}-3.32\%$
test_lock_stack_nested	0.7278ms	0.4364ms	2.2916 KOps/s	2.3239 KOps/s	$\color{#d91a1a}-1.39\%$
test_unlock_nested	0.9233ms	0.3875ms	2.5806 KOps/s	2.6492 KOps/s	$\color{#d91a1a}-2.59\%$
test_unlock_stack_nested	0.6580ms	0.3554ms	2.8135 KOps/s	2.8729 KOps/s	$\color{#d91a1a}-2.07\%$
test_flatten_speed	0.1757ms	0.1005ms	9.9536 KOps/s	9.9474 KOps/s	$\color{#35bf28}+0.06\%$
test_unflatten_speed	0.6815ms	0.5323ms	1.8787 KOps/s	1.8383 KOps/s	$\color{#35bf28}+2.20\%$
test_common_ops	1.7740ms	0.8290ms	1.2063 KOps/s	1.2919 KOps/s	$\textbf{\color{#d91a1a}-6.63\%}$
test_creation	19.7370μs	2.4873μs	402.0424 KOps/s	394.1277 KOps/s	$\color{#35bf28}+2.01\%$
test_creation_empty	35.8970μs	13.2192μs	75.6478 KOps/s	91.8668 KOps/s	$\textbf{\color{#d91a1a}-17.65\%}$
test_creation_nested_1	44.7940μs	16.1109μs	62.0697 KOps/s	72.3997 KOps/s	$\textbf{\color{#d91a1a}-14.27\%}$
test_creation_nested_2	74.1560μs	20.5064μs	48.7654 KOps/s	54.8130 KOps/s	$\textbf{\color{#d91a1a}-11.03\%}$
test_clone	57.2670μs	13.4182μs	74.5254 KOps/s	72.6537 KOps/s	$\color{#35bf28}+2.58\%$
test_getitem[int]	1.2903ms	13.2139μs	75.6780 KOps/s	77.5645 KOps/s	$\color{#d91a1a}-2.43\%$
test_getitem[slice_int]	0.1468ms	25.1435μs	39.7718 KOps/s	41.4749 KOps/s	$\color{#d91a1a}-4.11\%$
test_getitem[range]	0.1781ms	48.9485μs	20.4296 KOps/s	20.6459 KOps/s	$\color{#d91a1a}-1.05\%$
test_getitem[tuple]	0.1391ms	20.8199μs	48.0310 KOps/s	49.1941 KOps/s	$\color{#d91a1a}-2.36\%$
test_getitem[list]	0.3046ms	44.4635μs	22.4904 KOps/s	23.5438 KOps/s	$\color{#d91a1a}-4.47\%$
test_setitem_dim[int]	64.4510μs	25.0852μs	39.8641 KOps/s	39.9649 KOps/s	$\color{#d91a1a}-0.25\%$
test_setitem_dim[slice_int]	93.0350μs	51.6956μs	19.3440 KOps/s	19.9817 KOps/s	$\color{#d91a1a}-3.19\%$
test_setitem_dim[range]	0.1264ms	74.5407μs	13.4155 KOps/s	13.7392 KOps/s	$\color{#d91a1a}-2.36\%$
test_setitem_dim[tuple]	88.9870μs	40.6694μs	24.5885 KOps/s	25.4883 KOps/s	$\color{#d91a1a}-3.53\%$
test_setitem	0.1426ms	21.7150μs	46.0511 KOps/s	49.3700 KOps/s	$\textbf{\color{#d91a1a}-6.72\%}$
test_set	0.1177ms	21.3153μs	46.9146 KOps/s	50.3841 KOps/s	$\textbf{\color{#d91a1a}-6.89\%}$
test_set_shared	3.7082ms	0.1734ms	5.7678 KOps/s	5.7862 KOps/s	$\color{#d91a1a}-0.32\%$
test_update	0.1763ms	25.5489μs	39.1406 KOps/s	45.7166 KOps/s	$\textbf{\color{#d91a1a}-14.38\%}$
test_update_nested	0.1518ms	35.6299μs	28.0663 KOps/s	30.9256 KOps/s	$\textbf{\color{#d91a1a}-9.25\%}$
test_update__nested	0.5394ms	34.6944μs	28.8231 KOps/s	29.3627 KOps/s	$\color{#d91a1a}-1.84\%$
test_set_nested	81.9420μs	23.6226μs	42.3323 KOps/s	45.6352 KOps/s	$\textbf{\color{#d91a1a}-7.24\%}$
test_set_nested_new	0.1195ms	28.4904μs	35.0995 KOps/s	37.1322 KOps/s	$\textbf{\color{#d91a1a}-5.47\%}$
test_select	0.1443ms	45.0277μs	22.2086 KOps/s	22.8965 KOps/s	$\color{#d91a1a}-3.00\%$
test_select_nested	0.1253ms	62.6160μs	15.9704 KOps/s	15.3775 KOps/s	$\color{#35bf28}+3.86\%$
test_exclude_nested	0.1485ms	81.2071μs	12.3142 KOps/s	12.1456 KOps/s	$\color{#35bf28}+1.39\%$
test_empty[True]	0.6916ms	0.4083ms	2.4491 KOps/s	2.3996 KOps/s	$\color{#35bf28}+2.06\%$
test_empty[False]	13.7358μs	1.4261μs	701.1886 KOps/s	713.3399 KOps/s	$\color{#d91a1a}-1.70\%$
test_unbind_speed	0.4804ms	0.2787ms	3.5878 KOps/s	3.6946 KOps/s	$\color{#d91a1a}-2.89\%$
test_unbind_speed_stack0	0.4461ms	0.2775ms	3.6039 KOps/s	3.7021 KOps/s	$\color{#d91a1a}-2.65\%$
test_unbind_speed_stack1	0.1129s	0.8364ms	1.1957 KOps/s	1.4921 KOps/s	$\textbf{\color{#d91a1a}-19.87\%}$
test_split	2.5217ms	1.6137ms	619.6866 Ops/s	560.4651 Ops/s	$\textbf{\color{#35bf28}+10.57\%}$
test_chunk	0.1126s	1.9692ms	507.8129 Ops/s	561.5805 Ops/s	$\textbf{\color{#d91a1a}-9.57\%}$
test_consolidate_njt[False-None]	9.5909ms	8.3621ms	119.5873 Ops/s	120.8374 Ops/s	$\color{#d91a1a}-1.03\%$
test_creation[device0]	0.2866ms	91.4746μs	10.9320 KOps/s	10.7942 KOps/s	$\color{#35bf28}+1.28\%$
test_creation_from_tensor	4.7156ms	95.8346μs	10.4346 KOps/s	10.0323 KOps/s	$\color{#35bf28}+4.01\%$
test_add_one[memmap_tensor0]	0.1503ms	5.2123μs	191.8542 KOps/s	202.2188 KOps/s	$\textbf{\color{#d91a1a}-5.13\%}$
test_contiguous[memmap_tensor0]	12.4330μs	0.5309μs	1.8836 MOps/s	1.8836 MOps/s	$-0.00\%$
test_stack[memmap_tensor0]	60.0620μs	3.5750μs	279.7236 KOps/s	281.2384 KOps/s	$\color{#d91a1a}-0.54\%$
test_memmaptd_index	1.1016ms	0.2450ms	4.0824 KOps/s	4.1706 KOps/s	$\color{#d91a1a}-2.12\%$
test_memmaptd_index_astensor	0.6109ms	0.3337ms	2.9964 KOps/s	3.0596 KOps/s	$\color{#d91a1a}-2.07\%$
test_memmaptd_index_op	1.0577ms	0.6416ms	1.5587 KOps/s	1.6816 KOps/s	$\textbf{\color{#d91a1a}-7.31\%}$
test_serialize_model	0.1267s	0.1190s	8.4027 Ops/s	7.3436 Ops/s	$\textbf{\color{#35bf28}+14.42\%}$
test_serialize_model_pickle	0.4438s	0.3859s	2.5915 Ops/s	2.5160 Ops/s	$\color{#35bf28}+3.00\%$
test_serialize_weights	0.1276s	0.1173s	8.5260 Ops/s	8.5367 Ops/s	$\color{#d91a1a}-0.12\%$
test_serialize_weights_returnearly	0.1821s	0.1630s	6.1334 Ops/s	6.3767 Ops/s	$\color{#d91a1a}-3.82\%$
test_serialize_weights_pickle	0.5383s	0.4430s	2.2575 Ops/s	2.4767 Ops/s	$\textbf{\color{#d91a1a}-8.85\%}$
test_serialize_weights_filesystem	0.1524s	0.1460s	6.8482 Ops/s	6.9318 Ops/s	$\color{#d91a1a}-1.21\%$
test_serialize_model_filesystem	0.1627s	0.1521s	6.5742 Ops/s	6.3939 Ops/s	$\color{#35bf28}+2.82\%$
test_reshape_pytree	57.9790μs	26.7320μs	37.4083 KOps/s	37.3594 KOps/s	$\color{#35bf28}+0.13\%$
test_reshape_td	83.9470μs	33.7528μs	29.6271 KOps/s	30.3985 KOps/s	$\color{#d91a1a}-2.54\%$
test_view_pytree	70.3820μs	26.9967μs	37.0415 KOps/s	37.3066 KOps/s	$\color{#d91a1a}-0.71\%$
test_view_td	95.9700μs	38.0381μs	26.2895 KOps/s	25.6340 KOps/s	$\color{#35bf28}+2.56\%$
test_unbind_pytree	92.3330μs	29.7503μs	33.6132 KOps/s	33.3326 KOps/s	$\color{#35bf28}+0.84\%$
test_unbind_td	0.3678ms	41.2075μs	24.2674 KOps/s	25.5747 KOps/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_split_pytree	61.7150μs	29.7283μs	33.6379 KOps/s	34.1518 KOps/s	$\color{#d91a1a}-1.50\%$
test_split_td	0.6003ms	45.4562μs	21.9992 KOps/s	22.3536 KOps/s	$\color{#d91a1a}-1.59\%$
test_add_pytree	88.6460μs	35.6102μs	28.0819 KOps/s	28.0746 KOps/s	$\color{#35bf28}+0.03\%$
test_add_td	0.1467ms	60.9259μs	16.4134 KOps/s	18.0976 KOps/s	$\textbf{\color{#d91a1a}-9.31\%}$
test_compile_add_one_nested[tensordict-compile]	0.1265ms	63.2435μs	15.8119 KOps/s	15.7418 KOps/s	$\color{#35bf28}+0.45\%$
test_compile_add_one_nested[tensordict-eager]	0.4521ms	0.1751ms	5.7105 KOps/s	5.9110 KOps/s	$\color{#d91a1a}-3.39\%$
test_compile_add_one_nested[pytree-compile]	0.1598ms	46.1125μs	21.6861 KOps/s	22.0663 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_add_one_nested[pytree-eager]	0.2556ms	0.1200ms	8.3317 KOps/s	8.4309 KOps/s	$\color{#d91a1a}-1.18\%$
test_compile_copy_nested[tensordict-compile]	86.8830μs	25.7203μs	38.8799 KOps/s	37.9865 KOps/s	$\color{#35bf28}+2.35\%$
test_compile_copy_nested[tensordict-eager]	0.1292ms	59.1181μs	16.9153 KOps/s	16.9534 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_copy_nested[pytree-compile]	0.1658ms	77.7741μs	12.8577 KOps/s	12.5977 KOps/s	$\color{#35bf28}+2.06\%$
test_compile_copy_nested[pytree-eager]	0.1422ms	67.4701μs	14.8214 KOps/s	14.6598 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_add_one_flat[tensordict-compile]	0.1900ms	0.1064ms	9.3986 KOps/s	9.4933 KOps/s	$\color{#d91a1a}-1.00\%$
test_compile_add_one_flat[tensordict-eager]	0.4609ms	0.2220ms	4.5047 KOps/s	4.7070 KOps/s	$\color{#d91a1a}-4.30\%$
test_compile_add_one_flat[tensorclass-compile]	99.0150μs	45.3961μs	22.0283 KOps/s	22.1877 KOps/s	$\color{#d91a1a}-0.72\%$
test_compile_add_one_flat[tensorclass-eager]	0.5749ms	68.2118μs	14.6602 KOps/s	15.7102 KOps/s	$\textbf{\color{#d91a1a}-6.68\%}$
test_compile_add_one_flat[pytree-compile]	0.1949ms	0.1053ms	9.4965 KOps/s	9.8233 KOps/s	$\color{#d91a1a}-3.33\%$
test_compile_add_one_flat[pytree-eager]	0.4097ms	0.2046ms	4.8880 KOps/s	4.9453 KOps/s	$\color{#d91a1a}-1.16\%$
test_compile_add_self_flat[tensordict-eager]	0.4853ms	0.2332ms	4.2880 KOps/s	4.3105 KOps/s	$\color{#d91a1a}-0.52\%$
test_compile_add_self_flat[tensordict-compile]	0.2004ms	0.1058ms	9.4494 KOps/s	9.4713 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_add_self_flat[tensorclass-eager]	0.1555ms	60.7217μs	16.4686 KOps/s	17.1479 KOps/s	$\color{#d91a1a}-3.96\%$
test_compile_add_self_flat[tensorclass-compile]	0.1033ms	49.0355μs	20.3934 KOps/s	21.0376 KOps/s	$\color{#d91a1a}-3.06\%$
test_compile_add_self_flat[pytree-eager]	0.2514ms	0.1611ms	6.2068 KOps/s	6.3457 KOps/s	$\color{#d91a1a}-2.19\%$
test_compile_add_self_flat[pytree-compile]	0.1943ms	0.1076ms	9.2958 KOps/s	9.5102 KOps/s	$\color{#d91a1a}-2.25\%$
test_compile_copy_flat[tensordict-compile]	66.7760μs	22.6185μs	44.2116 KOps/s	46.4033 KOps/s	$\color{#d91a1a}-4.72\%$
test_compile_copy_flat[tensordict-eager]	0.1462ms	66.1071μs	15.1270 KOps/s	15.3560 KOps/s	$\color{#d91a1a}-1.49\%$
test_compile_copy_flat[pytree-compile]	0.1748ms	80.7990μs	12.3764 KOps/s	12.3707 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_copy_flat[pytree-eager]	0.1566ms	68.8963μs	14.5146 KOps/s	14.5927 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_assign_and_add[tensordict-compile]	0.4221ms	0.2158ms	4.6341 KOps/s	4.7858 KOps/s	$\color{#d91a1a}-3.17\%$
test_compile_assign_and_add[tensordict-eager]	1.4794ms	1.3418ms	745.2697 Ops/s	762.9671 Ops/s	$\color{#d91a1a}-2.32\%$
test_compile_assign_and_add[pytree-compile]	0.3074ms	0.2070ms	4.8311 KOps/s	4.9290 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_assign_and_add[pytree-eager]	1.0370ms	0.7967ms	1.2552 KOps/s	1.2854 KOps/s	$\color{#d91a1a}-2.35\%$
test_compile_assign_and_add_stack[compile]	1.0400ms	0.4693ms	2.1307 KOps/s	2.1852 KOps/s	$\color{#d91a1a}-2.49\%$
test_compile_assign_and_add_stack[eager]	3.9153ms	2.9981ms	333.5419 Ops/s	379.7694 Ops/s	$\textbf{\color{#d91a1a}-12.17\%}$
test_compile_indexing[tensor-tensordict-compile]	90.1890μs	37.3387μs	26.7819 KOps/s	27.9863 KOps/s	$\color{#d91a1a}-4.30\%$
test_compile_indexing[tensor-tensordict-eager]	0.5922ms	35.0066μs	28.5661 KOps/s	30.5827 KOps/s	$\textbf{\color{#d91a1a}-6.59\%}$
test_compile_indexing[tensor-tensorclass-compile]	98.7750μs	30.3378μs	32.9621 KOps/s	34.7493 KOps/s	$\textbf{\color{#d91a1a}-5.14\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.1106ms	23.4013μs	42.7326 KOps/s	40.2818 KOps/s	$\textbf{\color{#35bf28}+6.08\%}$
test_compile_indexing[tensor-pytree-compile]	81.7230μs	31.3010μs	31.9479 KOps/s	34.1422 KOps/s	$\textbf{\color{#d91a1a}-6.43\%}$
test_compile_indexing[tensor-pytree-eager]	62.9980μs	23.4776μs	42.5938 KOps/s	43.3736 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_indexing[slice-tensordict-compile]	0.1472ms	53.8731μs	18.5621 KOps/s	19.9375 KOps/s	$\textbf{\color{#d91a1a}-6.90\%}$
test_compile_indexing[slice-tensordict-eager]	0.5607ms	20.0367μs	49.9083 KOps/s	50.2563 KOps/s	$\color{#d91a1a}-0.69\%$
test_compile_indexing[slice-tensorclass-compile]	0.1355ms	45.5639μs	21.9472 KOps/s	23.2529 KOps/s	$\textbf{\color{#d91a1a}-5.62\%}$
test_compile_indexing[slice-tensorclass-eager]	57.5480μs	19.0411μs	52.5178 KOps/s	53.0936 KOps/s	$\color{#d91a1a}-1.08\%$
test_compile_indexing[slice-pytree-compile]	0.1036ms	47.1113μs	21.2263 KOps/s	22.9501 KOps/s	$\textbf{\color{#d91a1a}-7.51\%}$
test_compile_indexing[slice-pytree-eager]	67.9180μs	18.8082μs	53.1684 KOps/s	53.1754 KOps/s	$\color{#d91a1a}-0.01\%$
test_compile_indexing[int-tensordict-compile]	0.1354ms	55.0499μs	18.1653 KOps/s	19.6731 KOps/s	$\textbf{\color{#d91a1a}-7.66\%}$
test_compile_indexing[int-tensordict-eager]	1.0639ms	20.1245μs	49.6907 KOps/s	50.3605 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_indexing[int-tensorclass-compile]	0.1570ms	46.4096μs	21.5473 KOps/s	23.0145 KOps/s	$\textbf{\color{#d91a1a}-6.38\%}$
test_compile_indexing[int-tensorclass-eager]	65.9240μs	18.6845μs	53.5202 KOps/s	53.2914 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_indexing[int-pytree-compile]	0.1522ms	46.2260μs	21.6328 KOps/s	22.9859 KOps/s	$\textbf{\color{#d91a1a}-5.89\%}$
test_compile_indexing[int-pytree-eager]	0.5755ms	18.8678μs	53.0003 KOps/s	53.0434 KOps/s	$\color{#d91a1a}-0.08\%$
test_mod_add[eager]	94.4570μs	36.8455μs	27.1404 KOps/s	29.8416 KOps/s	$\textbf{\color{#d91a1a}-9.05\%}$
test_mod_add[compile]	99.8670μs	49.8559μs	20.0578 KOps/s	20.8412 KOps/s	$\color{#d91a1a}-3.76\%$
test_mod_add[compile-overhead]	0.1357ms	50.2008μs	19.9200 KOps/s	20.7435 KOps/s	$\color{#d91a1a}-3.97\%$
test_mod_wrap[eager]	0.4473ms	0.2334ms	4.2845 KOps/s	4.4548 KOps/s	$\color{#d91a1a}-3.82\%$
test_mod_wrap[compile]	0.3487ms	0.2114ms	4.7304 KOps/s	4.7903 KOps/s	$\color{#d91a1a}-1.25\%$
test_mod_wrap[compile-overhead]	0.3884ms	0.2098ms	4.7665 KOps/s	4.8886 KOps/s	$\color{#d91a1a}-2.50\%$
test_mod_wrap_and_backward[eager]	22.4251ms	13.5377ms	73.8677 Ops/s	83.0659 Ops/s	$\textbf{\color{#d91a1a}-11.07\%}$
test_mod_wrap_and_backward[compile]	22.5372ms	14.2634ms	70.1097 Ops/s	82.3573 Ops/s	$\textbf{\color{#d91a1a}-14.87\%}$
test_mod_wrap_and_backward[compile-overhead]	15.3021ms	12.6371ms	79.1319 Ops/s	72.1689 Ops/s	$\textbf{\color{#35bf28}+9.65\%}$
test_seq_add[eager]	0.2914ms	0.1201ms	8.3268 KOps/s	8.7889 KOps/s	$\textbf{\color{#d91a1a}-5.26\%}$
test_seq_add[compile]	0.1310ms	64.9783μs	15.3898 KOps/s	16.0996 KOps/s	$\color{#d91a1a}-4.41\%$
test_seq_add[compile-overhead]	0.1404ms	62.8732μs	15.9050 KOps/s	16.6964 KOps/s	$\color{#d91a1a}-4.74\%$
test_seq_wrap[eager]	0.6339ms	0.4651ms	2.1502 KOps/s	2.2655 KOps/s	$\textbf{\color{#d91a1a}-5.09\%}$
test_seq_wrap[compile]	0.4797ms	0.2369ms	4.2215 KOps/s	4.3921 KOps/s	$\color{#d91a1a}-3.88\%$
test_seq_wrap[compile-overhead]	0.4058ms	0.2362ms	4.2345 KOps/s	4.4192 KOps/s	$\color{#d91a1a}-4.18\%$
test_func_call_runtime[False-eager]	0.8034ms	0.5599ms	1.7859 KOps/s	1.8637 KOps/s	$\color{#d91a1a}-4.18\%$
test_func_call_runtime[False-compile]	0.8874ms	0.4474ms	2.2350 KOps/s	2.3437 KOps/s	$\color{#d91a1a}-4.64\%$
test_func_call_runtime[False-compile-overhead]	0.5714ms	0.4363ms	2.2920 KOps/s	2.3395 KOps/s	$\color{#d91a1a}-2.03\%$
test_func_call_runtime[True-eager]	1.5906ms	0.8049ms	1.2424 KOps/s	1.3138 KOps/s	$\textbf{\color{#d91a1a}-5.44\%}$
test_func_call_runtime[True-compile]	0.6362ms	0.4814ms	2.0775 KOps/s	2.1385 KOps/s	$\color{#d91a1a}-2.85\%$
test_func_call_runtime[True-compile-overhead]	0.6165ms	0.4825ms	2.0725 KOps/s	2.1404 KOps/s	$\color{#d91a1a}-3.17\%$
test_func_call_cm_runtime[False-eager]	0.9684ms	0.5686ms	1.7587 KOps/s	1.8389 KOps/s	$\color{#d91a1a}-4.36\%$
test_func_call_cm_runtime[False-compile]	0.5519ms	0.4386ms	2.2799 KOps/s	2.3421 KOps/s	$\color{#d91a1a}-2.65\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6015ms	0.4352ms	2.2977 KOps/s	2.3479 KOps/s	$\color{#d91a1a}-2.14\%$
test_func_call_cm_runtime[True-eager]	1.1695ms	0.9301ms	1.0752 KOps/s	1.0866 KOps/s	$\color{#d91a1a}-1.05\%$
test_func_call_cm_runtime[True-compile]	0.8407ms	0.5093ms	1.9635 KOps/s	2.0281 KOps/s	$\color{#d91a1a}-3.18\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6378ms	0.5050ms	1.9802 KOps/s	2.0373 KOps/s	$\color{#d91a1a}-2.80\%$
test_vmap_func_call_cm_runtime[eager]	3.3414ms	2.0166ms	495.8930 Ops/s	516.2709 Ops/s	$\color{#d91a1a}-3.95\%$
test_vmap_func_call_cm_runtime[compile]	0.9404ms	0.5368ms	1.8630 KOps/s	1.8899 KOps/s	$\color{#d91a1a}-1.43\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6429ms	0.5307ms	1.8844 KOps/s	1.9149 KOps/s	$\color{#d91a1a}-1.59\%$
test_distributed	0.3771ms	0.1260ms	7.9393 KOps/s	7.8254 KOps/s	$\color{#35bf28}+1.46\%$
test_tdmodule	44.9340μs	26.9087μs	37.1628 KOps/s	38.3178 KOps/s	$\color{#d91a1a}-3.01\%$
test_tdmodule_dispatch	82.9250μs	50.0307μs	19.9877 KOps/s	19.2552 KOps/s	$\color{#35bf28}+3.80\%$
test_tdseq	51.7570μs	30.3095μs	32.9930 KOps/s	35.3183 KOps/s	$\textbf{\color{#d91a1a}-6.58\%}$
test_tdseq_dispatch	81.3530μs	56.2087μs	17.7909 KOps/s	19.0982 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_instantiation_functorch	3.2451ms	1.5565ms	642.4553 Ops/s	657.5907 Ops/s	$\color{#d91a1a}-2.30\%$
test_exec_functorch	0.3565ms	0.1835ms	5.4509 KOps/s	5.5725 KOps/s	$\color{#d91a1a}-2.18\%$
test_exec_functional_call	0.2902ms	0.1746ms	5.7277 KOps/s	5.9685 KOps/s	$\color{#d91a1a}-4.03\%$
test_exec_td_decorator	0.5495ms	0.2381ms	4.2002 KOps/s	4.3441 KOps/s	$\color{#d91a1a}-3.31\%$
test_vmap_mlp_speed_decorator[True-True]	0.8649ms	0.6620ms	1.5106 KOps/s	1.5120 KOps/s	$\color{#d91a1a}-0.09\%$
test_vmap_mlp_speed_decorator[True-False]	5.8787ms	0.6685ms	1.4959 KOps/s	1.4905 KOps/s	$\color{#35bf28}+0.36\%$
test_vmap_mlp_speed_decorator[False-True]	1.0137ms	0.5410ms	1.8483 KOps/s	1.8684 KOps/s	$\color{#d91a1a}-1.07\%$
test_vmap_mlp_speed_decorator[False-False]	0.8861ms	0.5354ms	1.8678 KOps/s	1.8738 KOps/s	$\color{#d91a1a}-0.32\%$
test_to_module_speed[True]	1.9430ms	1.3529ms	739.1685 Ops/s	738.1826 Ops/s	$\color{#35bf28}+0.13\%$
test_to_module_speed[False]	1.7953ms	1.3340ms	749.6432 Ops/s	762.2430 Ops/s	$\color{#d91a1a}-1.65\%$
test_tc_init	87.7650μs	47.8508μs	20.8983 KOps/s	20.2492 KOps/s	$\color{#35bf28}+3.21\%$
test_tc_init_nested	0.1683ms	94.2051μs	10.6151 KOps/s	10.4695 KOps/s	$\color{#35bf28}+1.39\%$
test_tc_first_layer_tensor	13.1650μs	1.5503μs	645.0163 KOps/s	620.3138 KOps/s	$\color{#35bf28}+3.98\%$
test_tc_first_layer_nontensor	34.0840μs	4.7412μs	210.9162 KOps/s	209.9560 KOps/s	$\color{#35bf28}+0.46\%$
test_tc_second_layer_tensor	32.2310μs	2.8773μs	347.5485 KOps/s	336.3533 KOps/s	$\color{#35bf28}+3.33\%$
test_tc_second_layer_nontensor	37.9710μs	6.0931μs	164.1213 KOps/s	163.2234 KOps/s	$\color{#35bf28}+0.55\%$
test_unbind	0.2322s	14.7087ms	67.9871 Ops/s	74.5443 Ops/s	$\textbf{\color{#d91a1a}-8.80\%}$
test_full_like	11.1449ms	9.3367ms	107.1043 Ops/s	82.8222 Ops/s	$\textbf{\color{#35bf28}+29.32\%}$
test_zeros_like	4.0458ms	3.4781ms	287.5105 Ops/s	134.4528 Ops/s	$\textbf{\color{#35bf28}+113.84\%}$
test_ones_like	4.3997ms	3.8858ms	257.3488 Ops/s	119.6149 Ops/s	$\textbf{\color{#35bf28}+115.15\%}$
test_clone	9.1585ms	5.7859ms	172.8333 Ops/s	97.7328 Ops/s	$\textbf{\color{#35bf28}+76.84\%}$
test_squeeze	70.8930μs	12.3729μs	80.8219 KOps/s	80.6264 KOps/s	$\color{#35bf28}+0.24\%$
test_unsqueeze	0.1967ms	94.9608μs	10.5307 KOps/s	10.6672 KOps/s	$\color{#d91a1a}-1.28\%$
test_split	0.5240ms	0.1979ms	5.0521 KOps/s	5.0933 KOps/s	$\color{#d91a1a}-0.81\%$
test_permute	0.4082ms	0.2131ms	4.6916 KOps/s	4.7196 KOps/s	$\color{#d91a1a}-0.59\%$
test_stack	28.3458ms	24.0717ms	41.5426 Ops/s	36.9314 Ops/s	$\textbf{\color{#35bf28}+12.49\%}$
test_cat	29.1371ms	23.6444ms	42.2934 Ops/s	38.0193 Ops/s	$\textbf{\color{#35bf28}+11.24\%}$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}57$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.4500μs	11.3558μs	88.0608 KOps/s	76.9000 KOps/s	$\textbf{\color{#35bf28}+14.51\%}$
test_plain_set_stack_nested	34.7110μs	11.4686μs	87.1944 KOps/s	74.7964 KOps/s	$\textbf{\color{#35bf28}+16.58\%}$
test_plain_set_nested_inplace	44.2710μs	12.5857μs	79.4551 KOps/s	69.4498 KOps/s	$\textbf{\color{#35bf28}+14.41\%}$
test_plain_set_stack_nested_inplace	43.8610μs	12.5094μs	79.9396 KOps/s	69.2494 KOps/s	$\textbf{\color{#35bf28}+15.44\%}$
test_items	37.1300μs	2.8507μs	350.7877 KOps/s	346.2051 KOps/s	$\color{#35bf28}+1.32\%$
test_items_nested	0.4137ms	0.3539ms	2.8253 KOps/s	2.7584 KOps/s	$\color{#35bf28}+2.43\%$
test_items_nested_locked	0.3947ms	0.3575ms	2.7972 KOps/s	2.7601 KOps/s	$\color{#35bf28}+1.34\%$
test_items_nested_leaf	79.8820μs	58.0874μs	17.2155 KOps/s	16.9634 KOps/s	$\color{#35bf28}+1.49\%$
test_items_stack_nested	0.4154ms	0.3580ms	2.7929 KOps/s	2.7742 KOps/s	$\color{#35bf28}+0.67\%$
test_items_stack_nested_leaf	86.4620μs	59.7525μs	16.7357 KOps/s	16.7515 KOps/s	$\color{#d91a1a}-0.09\%$
test_items_stack_nested_locked	0.4082ms	0.3593ms	2.7828 KOps/s	2.7632 KOps/s	$\color{#35bf28}+0.71\%$
test_keys	36.9210μs	3.4440μs	290.3596 KOps/s	289.5189 KOps/s	$\color{#35bf28}+0.29\%$
test_keys_nested	0.1148ms	82.5243μs	12.1176 KOps/s	11.9565 KOps/s	$\color{#35bf28}+1.35\%$
test_keys_nested_locked	0.7744ms	88.0926μs	11.3517 KOps/s	11.1141 KOps/s	$\color{#35bf28}+2.14\%$
test_keys_nested_leaf	97.7920μs	72.8485μs	13.7271 KOps/s	13.5217 KOps/s	$\color{#35bf28}+1.52\%$
test_keys_stack_nested	0.1165ms	84.0121μs	11.9031 KOps/s	11.7067 KOps/s	$\color{#35bf28}+1.68\%$
test_keys_stack_nested_leaf	0.1112ms	75.1910μs	13.2995 KOps/s	13.2321 KOps/s	$\color{#35bf28}+0.51\%$
test_keys_stack_nested_locked	0.1390ms	90.3628μs	11.0665 KOps/s	10.9628 KOps/s	$\color{#35bf28}+0.95\%$
test_values	5.5868μs	0.8479μs	1.1794 MOps/s	1.1742 MOps/s	$\color{#35bf28}+0.44\%$
test_values_nested	57.1220μs	34.1095μs	29.3174 KOps/s	28.8130 KOps/s	$\color{#35bf28}+1.75\%$
test_values_nested_locked	58.0110μs	35.5759μs	28.1089 KOps/s	27.5937 KOps/s	$\color{#35bf28}+1.87\%$
test_values_nested_leaf	64.2410μs	38.5986μs	25.9076 KOps/s	25.6502 KOps/s	$\color{#35bf28}+1.00\%$
test_values_stack_nested	65.0810μs	34.7936μs	28.7409 KOps/s	28.2639 KOps/s	$\color{#35bf28}+1.69\%$
test_values_stack_nested_leaf	71.8910μs	39.3061μs	25.4414 KOps/s	25.1332 KOps/s	$\color{#35bf28}+1.23\%$
test_values_stack_nested_locked	66.9110μs	36.2481μs	27.5876 KOps/s	27.0235 KOps/s	$\color{#35bf28}+2.09\%$
test_membership	1.8090μs	0.5066μs	1.9738 MOps/s	1.9865 MOps/s	$\color{#d91a1a}-0.64\%$
test_membership_nested	22.0755μs	2.0019μs	499.5249 KOps/s	496.2401 KOps/s	$\color{#35bf28}+0.66\%$
test_membership_nested_leaf	17.2205μs	2.0111μs	497.2368 KOps/s	501.4104 KOps/s	$\color{#d91a1a}-0.83\%$
test_membership_stacked_nested	29.7710μs	2.1176μs	472.2414 KOps/s	473.1745 KOps/s	$\color{#d91a1a}-0.20\%$
test_membership_stacked_nested_leaf	33.7810μs	2.0983μs	476.5803 KOps/s	470.5908 KOps/s	$\color{#35bf28}+1.27\%$
test_membership_nested_last	27.7410μs	3.1257μs	319.9289 KOps/s	319.1881 KOps/s	$\color{#35bf28}+0.23\%$
test_membership_nested_leaf_last	44.3710μs	3.0674μs	326.0037 KOps/s	315.3605 KOps/s	$\color{#35bf28}+3.37\%$
test_membership_stacked_nested_last	42.9410μs	4.2987μs	232.6282 KOps/s	264.2891 KOps/s	$\textbf{\color{#d91a1a}-11.98\%}$
test_membership_stacked_nested_leaf_last	38.1700μs	4.2660μs	234.4135 KOps/s	269.6416 KOps/s	$\textbf{\color{#d91a1a}-13.06\%}$
test_nested_getleaf	25.0010μs	6.2616μs	159.7034 KOps/s	160.0593 KOps/s	$\color{#d91a1a}-0.22\%$
test_nested_get	32.7510μs	5.8615μs	170.6048 KOps/s	170.8117 KOps/s	$\color{#d91a1a}-0.12\%$
test_stacked_getleaf	28.1610μs	6.1612μs	162.3054 KOps/s	161.4050 KOps/s	$\color{#35bf28}+0.56\%$
test_stacked_get	39.1310μs	5.8676μs	170.4286 KOps/s	171.2666 KOps/s	$\color{#d91a1a}-0.49\%$
test_nested_getitemleaf	26.8400μs	6.2606μs	159.7280 KOps/s	157.6464 KOps/s	$\color{#35bf28}+1.32\%$
test_nested_getitem	29.4810μs	5.9110μs	169.1756 KOps/s	165.8801 KOps/s	$\color{#35bf28}+1.99\%$
test_stacked_getitemleaf	31.3200μs	6.2431μs	160.1773 KOps/s	159.8281 KOps/s	$\color{#35bf28}+0.22\%$
test_stacked_getitem	28.6000μs	5.9203μs	168.9106 KOps/s	167.2234 KOps/s	$\color{#35bf28}+1.01\%$
test_lock_nested	9.1786ms	0.3887ms	2.5724 KOps/s	2.6062 KOps/s	$\color{#d91a1a}-1.29\%$
test_lock_stack_nested	0.4024ms	0.3483ms	2.8712 KOps/s	2.7934 KOps/s	$\color{#35bf28}+2.79\%$
test_unlock_nested	0.7284ms	0.3209ms	3.1159 KOps/s	3.0679 KOps/s	$\color{#35bf28}+1.56\%$
test_unlock_stack_nested	0.3453ms	0.2864ms	3.4912 KOps/s	3.4089 KOps/s	$\color{#35bf28}+2.41\%$
test_flatten_speed	0.1321ms	75.8709μs	13.1803 KOps/s	13.1941 KOps/s	$\color{#d91a1a}-0.10\%$
test_unflatten_speed	0.3707ms	0.3224ms	3.1021 KOps/s	3.0476 KOps/s	$\color{#35bf28}+1.79\%$
test_common_ops	92.5933ms	0.6343ms	1.5765 KOps/s	1.5312 KOps/s	$\color{#35bf28}+2.96\%$
test_creation	21.5210μs	1.7240μs	580.0603 KOps/s	574.1353 KOps/s	$\color{#35bf28}+1.03\%$
test_creation_empty	35.5700μs	6.4745μs	154.4520 KOps/s	102.5737 KOps/s	$\textbf{\color{#35bf28}+50.58\%}$
test_creation_nested_1	19.8300μs	8.3774μs	119.3695 KOps/s	87.1599 KOps/s	$\textbf{\color{#35bf28}+36.95\%}$
test_creation_nested_2	0.1175ms	10.8855μs	91.8649 KOps/s	70.1057 KOps/s	$\textbf{\color{#35bf28}+31.04\%}$
test_clone	88.6610μs	10.3506μs	96.6132 KOps/s	91.2443 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_getitem[int]	0.9843ms	10.7888μs	92.6885 KOps/s	85.7663 KOps/s	$\textbf{\color{#35bf28}+8.07\%}$
test_getitem[slice_int]	0.1028ms	20.9990μs	47.6214 KOps/s	45.0288 KOps/s	$\textbf{\color{#35bf28}+5.76\%}$
test_getitem[range]	0.1229ms	36.4124μs	27.4632 KOps/s	26.7345 KOps/s	$\color{#35bf28}+2.73\%$
test_getitem[tuple]	0.1029ms	18.3883μs	54.3824 KOps/s	50.5092 KOps/s	$\textbf{\color{#35bf28}+7.67\%}$
test_getitem[list]	0.2052ms	32.2017μs	31.0543 KOps/s	30.1793 KOps/s	$\color{#35bf28}+2.90\%$
test_setitem_dim[int]	39.7110μs	18.3593μs	54.4683 KOps/s	53.2331 KOps/s	$\color{#35bf28}+2.32\%$
test_setitem_dim[slice_int]	70.7710μs	37.8523μs	26.4185 KOps/s	26.3013 KOps/s	$\color{#35bf28}+0.45\%$
test_setitem_dim[range]	83.5020μs	50.4730μs	19.8126 KOps/s	19.1274 KOps/s	$\color{#35bf28}+3.58\%$
test_setitem_dim[tuple]	63.3310μs	30.5977μs	32.6822 KOps/s	30.7832 KOps/s	$\textbf{\color{#35bf28}+6.17\%}$
test_setitem	70.1720μs	13.4603μs	74.2928 KOps/s	62.1920 KOps/s	$\textbf{\color{#35bf28}+19.46\%}$
test_set	96.9020μs	13.0624μs	76.5555 KOps/s	63.7627 KOps/s	$\textbf{\color{#35bf28}+20.06\%}$
test_set_shared	1.5588ms	0.1510ms	6.6208 KOps/s	6.4339 KOps/s	$\color{#35bf28}+2.91\%$
test_update	0.5644ms	14.7856μs	67.6334 KOps/s	50.9841 KOps/s	$\textbf{\color{#35bf28}+32.66\%}$
test_update_nested	1.1266ms	20.2293μs	49.4333 KOps/s	39.4320 KOps/s	$\textbf{\color{#35bf28}+25.36\%}$
test_update__nested	72.7120μs	24.5746μs	40.6924 KOps/s	37.9223 KOps/s	$\textbf{\color{#35bf28}+7.30\%}$
test_set_nested	73.7410μs	14.4379μs	69.2623 KOps/s	59.7168 KOps/s	$\textbf{\color{#35bf28}+15.98\%}$
test_set_nested_new	84.0610μs	16.9422μs	59.0242 KOps/s	52.2162 KOps/s	$\textbf{\color{#35bf28}+13.04\%}$
test_select	84.4020μs	28.7850μs	34.7403 KOps/s	32.0038 KOps/s	$\textbf{\color{#35bf28}+8.55\%}$
test_select_nested	68.8820μs	45.0094μs	22.2176 KOps/s	22.1125 KOps/s	$\color{#35bf28}+0.48\%$
test_exclude_nested	90.5120μs	62.5920μs	15.9765 KOps/s	15.7258 KOps/s	$\color{#35bf28}+1.59\%$
test_empty[True]	0.3242ms	0.2915ms	3.4309 KOps/s	3.4327 KOps/s	$\color{#d91a1a}-0.05\%$
test_empty[False]	8.4431μs	0.8286μs	1.2069 MOps/s	1.1827 MOps/s	$\color{#35bf28}+2.04\%$
test_to	86.0020μs	56.6783μs	17.6434 KOps/s	17.4669 KOps/s	$\color{#35bf28}+1.01\%$
test_to_nonblocking	80.1520μs	47.3804μs	21.1058 KOps/s	19.3275 KOps/s	$\textbf{\color{#35bf28}+9.20\%}$
test_unbind_speed	1.3581ms	0.2410ms	4.1494 KOps/s	4.0301 KOps/s	$\color{#35bf28}+2.96\%$
test_unbind_speed_stack0	0.2739ms	0.2428ms	4.1178 KOps/s	4.0081 KOps/s	$\color{#35bf28}+2.74\%$
test_unbind_speed_stack1	92.3831ms	0.6651ms	1.5036 KOps/s	1.4641 KOps/s	$\color{#35bf28}+2.70\%$
test_split	93.2546ms	1.5927ms	627.8688 Ops/s	593.2662 Ops/s	$\textbf{\color{#35bf28}+5.83\%}$
test_chunk	95.5949ms	1.6186ms	617.8090 Ops/s	596.3680 Ops/s	$\color{#35bf28}+3.60\%$
test_consolidate[False-None]	95.8422ms	2.9972ms	333.6476 Ops/s	318.8718 Ops/s	$\color{#35bf28}+4.63\%$
test_consolidate[default-None]	1.9223ms	1.6891ms	592.0276 Ops/s	565.0876 Ops/s	$\color{#35bf28}+4.77\%$
test_consolidate[reduce-overhead-None]	1.8497ms	1.7271ms	579.0140 Ops/s	548.8652 Ops/s	$\textbf{\color{#35bf28}+5.49\%}$
test_consolidate_njt[False-None]	7.5035ms	6.8714ms	145.5303 Ops/s	145.9922 Ops/s	$\color{#d91a1a}-0.32\%$
test_to[False-False-None]	2.1966ms	1.7315ms	577.5344 Ops/s	554.3052 Ops/s	$\color{#35bf28}+4.19\%$
test_to[True-False-None]	1.6149ms	1.3701ms	729.8822 Ops/s	715.0407 Ops/s	$\color{#35bf28}+2.08\%$
test_to[within-False-None]	4.4118ms	4.1696ms	239.8299 Ops/s	227.2525 Ops/s	$\textbf{\color{#35bf28}+5.53\%}$
test_to[True-default-None]	5.9816ms	5.7196ms	174.8374 Ops/s	181.4840 Ops/s	$\color{#d91a1a}-3.66\%$
test_to_njt[False-False-None]	7.4593ms	7.0543ms	141.7576 Ops/s	141.3078 Ops/s	$\color{#35bf28}+0.32\%$
test_to_njt[True-False-None]	6.5128ms	5.8143ms	171.9911 Ops/s	174.7113 Ops/s	$\color{#d91a1a}-1.56\%$
test_to_njt[within-False-None]	13.3855ms	13.1587ms	75.9951 Ops/s	79.0094 Ops/s	$\color{#d91a1a}-3.82\%$
test_creation[device0]	0.5525ms	85.7165μs	11.6664 KOps/s	12.4050 KOps/s	$\textbf{\color{#d91a1a}-5.95\%}$
test_creation_from_tensor	0.5386ms	87.3424μs	11.4492 KOps/s	11.9036 KOps/s	$\color{#d91a1a}-3.82\%$
test_add_one[memmap_tensor0]	0.2387ms	6.3427μs	157.6623 KOps/s	152.7964 KOps/s	$\color{#35bf28}+3.18\%$
test_contiguous[memmap_tensor0]	2.1505μs	0.3982μs	2.5111 MOps/s	2.4950 MOps/s	$\color{#35bf28}+0.64\%$
test_stack[memmap_tensor0]	30.1710μs	4.5478μs	219.8881 KOps/s	200.7368 KOps/s	$\textbf{\color{#35bf28}+9.54\%}$
test_memmaptd_index	1.3982ms	0.2506ms	3.9912 KOps/s	3.6932 KOps/s	$\textbf{\color{#35bf28}+8.07\%}$
test_memmaptd_index_astensor	0.5691ms	0.3090ms	3.2368 KOps/s	2.9564 KOps/s	$\textbf{\color{#35bf28}+9.48\%}$
test_memmaptd_index_op	0.9965ms	0.5464ms	1.8300 KOps/s	1.6002 KOps/s	$\textbf{\color{#35bf28}+14.36\%}$
test_serialize_model	0.1318s	0.1311s	7.6279 Ops/s	7.5810 Ops/s	$\color{#35bf28}+0.62\%$
test_serialize_model_pickle	1.3462s	1.2129s	0.8245 Ops/s	0.8260 Ops/s	$\color{#d91a1a}-0.18\%$
test_serialize_weights	0.4228s	0.1725s	5.7965 Ops/s	7.6057 Ops/s	$\textbf{\color{#d91a1a}-23.79\%}$
test_serialize_weights_returnearly	0.3270s	56.0382ms	17.8450 Ops/s	22.9116 Ops/s	$\textbf{\color{#d91a1a}-22.11\%}$
test_serialize_weights_pickle	1.3789s	1.2266s	0.8152 Ops/s	0.8401 Ops/s	$\color{#d91a1a}-2.95\%$
test_reshape_pytree	68.4510μs	23.3998μs	42.7353 KOps/s	43.4002 KOps/s	$\color{#d91a1a}-1.53\%$
test_reshape_td	78.3120μs	31.4405μs	31.8061 KOps/s	35.4129 KOps/s	$\textbf{\color{#d91a1a}-10.18\%}$
test_view_pytree	67.3320μs	23.8026μs	42.0122 KOps/s	44.2706 KOps/s	$\textbf{\color{#d91a1a}-5.10\%}$
test_view_td	76.7310μs	34.8799μs	28.6698 KOps/s	31.3985 KOps/s	$\textbf{\color{#d91a1a}-8.69\%}$
test_unbind_pytree	58.1610μs	29.7103μs	33.6584 KOps/s	34.4897 KOps/s	$\color{#d91a1a}-2.41\%$
test_unbind_td	0.7799ms	38.9448μs	25.6774 KOps/s	26.4355 KOps/s	$\color{#d91a1a}-2.87\%$
test_split_pytree	62.7410μs	32.8287μs	30.4612 KOps/s	31.5781 KOps/s	$\color{#d91a1a}-3.54\%$
test_split_td	0.9033ms	42.0729μs	23.7683 KOps/s	23.6284 KOps/s	$\color{#35bf28}+0.59\%$
test_add_pytree	73.6510μs	34.5845μs	28.9147 KOps/s	29.1753 KOps/s	$\color{#d91a1a}-0.89\%$
test_add_td	85.5210μs	45.0435μs	22.2008 KOps/s	19.7494 KOps/s	$\textbf{\color{#35bf28}+12.41\%}$
test_compile_add_one_nested[tensordict-compile]	0.1749ms	0.1215ms	8.2326 KOps/s	7.4296 KOps/s	$\textbf{\color{#35bf28}+10.81\%}$
test_compile_add_one_nested[tensordict-eager]	0.2252ms	0.1312ms	7.6191 KOps/s	7.1241 KOps/s	$\textbf{\color{#35bf28}+6.95\%}$
test_compile_add_one_nested[pytree-compile]	0.1423ms	97.4993μs	10.2565 KOps/s	9.6458 KOps/s	$\textbf{\color{#35bf28}+6.33\%}$
test_compile_add_one_nested[pytree-eager]	1.4158ms	0.1493ms	6.7000 KOps/s	6.1575 KOps/s	$\textbf{\color{#35bf28}+8.81\%}$
test_compile_copy_nested[tensordict-compile]	52.5010μs	23.6029μs	42.3677 KOps/s	44.5295 KOps/s	$\color{#d91a1a}-4.85\%$
test_compile_copy_nested[tensordict-eager]	60.6610μs	29.2167μs	34.2270 KOps/s	34.0333 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_copy_nested[pytree-compile]	0.3153ms	64.5449μs	15.4931 KOps/s	15.3326 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_copy_nested[pytree-eager]	0.1192ms	48.5559μs	20.5948 KOps/s	20.0716 KOps/s	$\color{#35bf28}+2.61\%$
test_compile_add_one_flat[tensordict-compile]	0.2150ms	0.1468ms	6.8142 KOps/s	6.8384 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_add_one_flat[tensordict-eager]	0.3096ms	0.2146ms	4.6600 KOps/s	4.6512 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_add_one_flat[tensorclass-compile]	0.1614ms	98.3378μs	10.1690 KOps/s	10.0243 KOps/s	$\color{#35bf28}+1.44\%$
test_compile_add_one_flat[tensorclass-eager]	0.1095ms	52.1866μs	19.1620 KOps/s	18.8257 KOps/s	$\color{#35bf28}+1.79\%$
test_compile_add_one_flat[pytree-compile]	0.2019ms	0.1365ms	7.3253 KOps/s	7.1256 KOps/s	$\color{#35bf28}+2.80\%$
test_compile_add_one_flat[pytree-eager]	0.5292ms	0.4750ms	2.1052 KOps/s	2.0027 KOps/s	$\textbf{\color{#35bf28}+5.12\%}$
test_compile_add_self_flat[tensordict-eager]	0.3649ms	0.2588ms	3.8635 KOps/s	3.8750 KOps/s	$\color{#d91a1a}-0.30\%$
test_compile_add_self_flat[tensordict-compile]	0.1843ms	0.1438ms	6.9520 KOps/s	6.8862 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_add_self_flat[tensorclass-eager]	0.1451ms	64.1698μs	15.5836 KOps/s	15.3764 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_add_self_flat[tensorclass-compile]	0.1376ms	99.2934μs	10.0712 KOps/s	10.0285 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_add_self_flat[pytree-eager]	0.4551ms	0.4072ms	2.4555 KOps/s	2.3480 KOps/s	$\color{#35bf28}+4.58\%$
test_compile_add_self_flat[pytree-compile]	0.1861ms	0.1394ms	7.1743 KOps/s	7.2401 KOps/s	$\color{#d91a1a}-0.91\%$
test_compile_copy_flat[tensordict-compile]	57.0210μs	19.2837μs	51.8572 KOps/s	55.0735 KOps/s	$\textbf{\color{#d91a1a}-5.84\%}$
test_compile_copy_flat[tensordict-eager]	63.3810μs	32.0126μs	31.2377 KOps/s	31.8030 KOps/s	$\color{#d91a1a}-1.78\%$
test_compile_copy_flat[pytree-compile]	0.1449ms	71.0364μs	14.0773 KOps/s	13.9710 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_copy_flat[pytree-eager]	0.1134ms	52.5160μs	19.0418 KOps/s	18.8360 KOps/s	$\color{#35bf28}+1.09\%$
test_compile_assign_and_add[tensordict-compile]	1.6182ms	0.3894ms	2.5681 KOps/s	2.1823 KOps/s	$\textbf{\color{#35bf28}+17.68\%}$
test_compile_assign_and_add[tensordict-eager]	2.6378ms	2.5687ms	389.3041 Ops/s	350.2480 Ops/s	$\textbf{\color{#35bf28}+11.15\%}$
test_compile_assign_and_add[pytree-compile]	1.5921ms	0.3808ms	2.6262 KOps/s	2.2414 KOps/s	$\textbf{\color{#35bf28}+17.17\%}$
test_compile_assign_and_add[pytree-eager]	2.8579ms	2.6273ms	380.6175 Ops/s	366.1222 Ops/s	$\color{#35bf28}+3.96\%$
test_compile_indexing[tensor-tensordict-compile]	0.1652ms	0.1128ms	8.8655 KOps/s	8.3925 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_compile_indexing[tensor-tensordict-eager]	0.5588ms	82.6665μs	12.0968 KOps/s	12.3067 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1669ms	0.1143ms	8.7500 KOps/s	9.3946 KOps/s	$\textbf{\color{#d91a1a}-6.86\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.1182ms	71.9232μs	13.9037 KOps/s	14.3003 KOps/s	$\color{#d91a1a}-2.77\%$
test_compile_indexing[tensor-pytree-compile]	0.1675ms	0.1143ms	8.7515 KOps/s	8.9058 KOps/s	$\color{#d91a1a}-1.73\%$
test_compile_indexing[tensor-pytree-eager]	0.1553ms	68.3241μs	14.6361 KOps/s	14.5792 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_indexing[slice-tensordict-compile]	0.1575ms	0.1016ms	9.8460 KOps/s	9.7309 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_indexing[slice-tensordict-eager]	0.1390ms	17.6077μs	56.7932 KOps/s	49.3673 KOps/s	$\textbf{\color{#35bf28}+15.04\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1440ms	97.2085μs	10.2872 KOps/s	10.1311 KOps/s	$\color{#35bf28}+1.54\%$
test_compile_indexing[slice-tensorclass-eager]	58.3810μs	15.8815μs	62.9665 KOps/s	60.6739 KOps/s	$\color{#35bf28}+3.78\%$
test_compile_indexing[slice-pytree-compile]	0.1541ms	0.1008ms	9.9246 KOps/s	10.0258 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_indexing[slice-pytree-eager]	49.1410μs	16.0135μs	62.4474 KOps/s	59.3583 KOps/s	$\textbf{\color{#35bf28}+5.20\%}$
test_compile_indexing[int-tensordict-compile]	0.1650ms	0.1088ms	9.1880 KOps/s	9.6119 KOps/s	$\color{#d91a1a}-4.41\%$
test_compile_indexing[int-tensordict-eager]	0.5992ms	18.5970μs	53.7720 KOps/s	53.9282 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_indexing[int-tensorclass-compile]	0.1546ms	0.1063ms	9.4116 KOps/s	9.9945 KOps/s	$\textbf{\color{#d91a1a}-5.83\%}$
test_compile_indexing[int-tensorclass-eager]	38.9810μs	17.1412μs	58.3390 KOps/s	60.0756 KOps/s	$\color{#d91a1a}-2.89\%$
test_compile_indexing[int-pytree-compile]	0.1512ms	0.1042ms	9.6005 KOps/s	9.9726 KOps/s	$\color{#d91a1a}-3.73\%$
test_compile_indexing[int-pytree-eager]	57.3010μs	17.5552μs	56.9631 KOps/s	60.1543 KOps/s	$\textbf{\color{#d91a1a}-5.31\%}$
test_mod_add[eager]	84.6320μs	36.5466μs	27.3624 KOps/s	25.8516 KOps/s	$\textbf{\color{#35bf28}+5.84\%}$
test_mod_add[compile]	0.1157ms	81.3666μs	12.2901 KOps/s	12.0922 KOps/s	$\color{#35bf28}+1.64\%$
test_mod_add[compile-overhead]	0.3328ms	0.1690ms	5.9174 KOps/s	5.5289 KOps/s	$\textbf{\color{#35bf28}+7.03\%}$
test_mod_wrap[eager]	0.3335ms	0.2422ms	4.1289 KOps/s	3.9947 KOps/s	$\color{#35bf28}+3.36\%$
test_mod_wrap[compile]	0.3519ms	0.2832ms	3.5308 KOps/s	3.4635 KOps/s	$\color{#35bf28}+1.94\%$
test_mod_wrap[compile-overhead]	7.0198ms	3.7440ms	267.0957 Ops/s	271.4366 Ops/s	$\color{#d91a1a}-1.60\%$
test_mod_wrap_and_backward[eager]	1.4322ms	1.3278ms	753.1200 Ops/s	693.8119 Ops/s	$\textbf{\color{#35bf28}+8.55\%}$
test_mod_wrap_and_backward[compile]	1.6717ms	1.2611ms	792.9655 Ops/s	714.8099 Ops/s	$\textbf{\color{#35bf28}+10.93\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3588ms	0.9300ms	1.0753 KOps/s	960.8276 Ops/s	$\textbf{\color{#35bf28}+11.91\%}$
test_seq_add[eager]	0.1804ms	0.1174ms	8.5182 KOps/s	8.3718 KOps/s	$\color{#35bf28}+1.75\%$
test_seq_add[compile]	0.2229ms	92.6358μs	10.7950 KOps/s	11.0663 KOps/s	$\color{#d91a1a}-2.45\%$
test_seq_add[compile-overhead]	0.1718ms	0.1306ms	7.6588 KOps/s	7.4920 KOps/s	$\color{#35bf28}+2.23\%$
test_seq_wrap[eager]	0.5385ms	0.4310ms	2.3199 KOps/s	2.3250 KOps/s	$\color{#d91a1a}-0.22\%$
test_seq_wrap[compile]	0.3975ms	0.2990ms	3.3448 KOps/s	3.1797 KOps/s	$\textbf{\color{#35bf28}+5.19\%}$
test_seq_wrap[compile-overhead]	0.2922ms	0.2255ms	4.4339 KOps/s	4.3925 KOps/s	$\color{#35bf28}+0.94\%$
test_func_call_runtime[False-eager]	0.7593ms	0.7145ms	1.3996 KOps/s	1.3106 KOps/s	$\textbf{\color{#35bf28}+6.80\%}$
test_func_call_runtime[False-compile]	0.9726ms	0.7413ms	1.3489 KOps/s	1.3161 KOps/s	$\color{#35bf28}+2.49\%$
test_func_call_runtime[False-compile-overhead]	0.4660ms	0.3674ms	2.7220 KOps/s	2.7340 KOps/s	$\color{#d91a1a}-0.44\%$
test_func_call_runtime[True-eager]	0.9273ms	0.8734ms	1.1450 KOps/s	1.1084 KOps/s	$\color{#35bf28}+3.30\%$
test_func_call_runtime[True-compile]	0.9921ms	0.7647ms	1.3077 KOps/s	1.2806 KOps/s	$\color{#35bf28}+2.11\%$
test_func_call_runtime[True-compile-overhead]	0.5036ms	0.3872ms	2.5827 KOps/s	2.6006 KOps/s	$\color{#d91a1a}-0.69\%$
test_func_call_cm_runtime[False-eager]	0.8479ms	0.7698ms	1.2990 KOps/s	1.3873 KOps/s	$\textbf{\color{#d91a1a}-6.37\%}$
test_func_call_cm_runtime[False-compile]	0.8064ms	0.7449ms	1.3424 KOps/s	1.3113 KOps/s	$\color{#35bf28}+2.37\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4137ms	0.3658ms	2.7339 KOps/s	2.7357 KOps/s	$\color{#d91a1a}-0.07\%$
test_func_call_cm_runtime[True-eager]	1.2142ms	1.0326ms	968.4758 Ops/s	997.3958 Ops/s	$\color{#d91a1a}-2.90\%$
test_func_call_cm_runtime[True-compile]	1.0435ms	0.7882ms	1.2688 KOps/s	1.2313 KOps/s	$\color{#35bf28}+3.04\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4866ms	0.4097ms	2.4409 KOps/s	2.4213 KOps/s	$\color{#35bf28}+0.81\%$
test_vmap_func_call_cm_runtime[eager]	2.5390ms	2.0431ms	489.4572 Ops/s	489.5119 Ops/s	$\color{#d91a1a}-0.01\%$
test_vmap_func_call_cm_runtime[compile]	0.8766ms	0.7993ms	1.2511 KOps/s	1.2070 KOps/s	$\color{#35bf28}+3.65\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4879ms	0.4134ms	2.4189 KOps/s	2.4067 KOps/s	$\color{#35bf28}+0.51\%$
test_distributed	4.2159ms	0.1779ms	5.6218 KOps/s	8.3170 KOps/s	$\textbf{\color{#d91a1a}-32.41\%}$
test_tdmodule	48.1110μs	18.6723μs	53.5552 KOps/s	47.5289 KOps/s	$\textbf{\color{#35bf28}+12.68\%}$
test_tdmodule_dispatch	66.2310μs	33.6452μs	29.7219 KOps/s	26.0775 KOps/s	$\textbf{\color{#35bf28}+13.98\%}$
test_tdseq	48.9110μs	19.8365μs	50.4122 KOps/s	44.3841 KOps/s	$\textbf{\color{#35bf28}+13.58\%}$
test_tdseq_dispatch	66.7920μs	36.8634μs	27.1272 KOps/s	23.9029 KOps/s	$\textbf{\color{#35bf28}+13.49\%}$
test_instantiation_functorch	1.6662ms	1.5607ms	640.7190 Ops/s	626.1298 Ops/s	$\color{#35bf28}+2.33\%$
test_exec_functorch	0.2026ms	0.1433ms	6.9784 KOps/s	7.0126 KOps/s	$\color{#d91a1a}-0.49\%$
test_exec_functional_call	0.1873ms	0.1350ms	7.4065 KOps/s	7.4140 KOps/s	$\color{#d91a1a}-0.10\%$
test_exec_td_decorator	0.3791ms	0.1842ms	5.4282 KOps/s	5.3907 KOps/s	$\color{#35bf28}+0.70\%$
test_vmap_mlp_speed_decorator[True-True]	0.7819ms	0.6724ms	1.4872 KOps/s	1.4184 KOps/s	$\color{#35bf28}+4.85\%$
test_vmap_mlp_speed_decorator[True-False]	0.8033ms	0.6761ms	1.4792 KOps/s	1.4087 KOps/s	$\textbf{\color{#35bf28}+5.00\%}$
test_vmap_mlp_speed_decorator[False-True]	0.7103ms	0.5999ms	1.6669 KOps/s	1.6449 KOps/s	$\color{#35bf28}+1.34\%$
test_vmap_mlp_speed_decorator[False-False]	0.7376ms	0.6014ms	1.6627 KOps/s	1.6289 KOps/s	$\color{#35bf28}+2.07\%$
test_vmap_transformer_speed_decorator[True-True]	20.1489ms	19.2757ms	51.8788 Ops/s	52.8742 Ops/s	$\color{#d91a1a}-1.88\%$
test_vmap_transformer_speed_decorator[True-False]	20.0146ms	19.2249ms	52.0158 Ops/s	52.8257 Ops/s	$\color{#d91a1a}-1.53\%$
test_vmap_transformer_speed_decorator[False-True]	19.8753ms	18.8729ms	52.9860 Ops/s	53.5567 Ops/s	$\color{#d91a1a}-1.07\%$
test_vmap_transformer_speed_decorator[False-False]	19.5490ms	18.7192ms	53.4212 Ops/s	53.2767 Ops/s	$\color{#35bf28}+0.27\%$
test_to_module_speed[True]	1.0713ms	0.9781ms	1.0223 KOps/s	989.3886 Ops/s	$\color{#35bf28}+3.33\%$
test_to_module_speed[False]	1.1040ms	0.9583ms	1.0435 KOps/s	1.0224 KOps/s	$\color{#35bf28}+2.06\%$
test_tc_init	54.6310μs	34.1426μs	29.2889 KOps/s	24.8283 KOps/s	$\textbf{\color{#35bf28}+17.97\%}$
test_tc_init_nested	0.1028ms	69.7310μs	14.3408 KOps/s	11.9598 KOps/s	$\textbf{\color{#35bf28}+19.91\%}$
test_tc_first_layer_tensor	32.3100μs	0.8328μs	1.2008 MOps/s	1.1817 MOps/s	$\color{#35bf28}+1.62\%$
test_tc_first_layer_nontensor	23.6700μs	2.2541μs	443.6354 KOps/s	437.5243 KOps/s	$\color{#35bf28}+1.40\%$
test_tc_second_layer_tensor	22.5855μs	1.4045μs	711.9792 KOps/s	644.6392 KOps/s	$\textbf{\color{#35bf28}+10.45\%}$
test_tc_second_layer_nontensor	38.8210μs	3.0115μs	332.0595 KOps/s	331.1266 KOps/s	$\color{#35bf28}+0.28\%$
test_unbind	0.2195s	11.6751ms	85.6522 Ops/s	139.7307 Ops/s	$\textbf{\color{#d91a1a}-38.70\%}$
test_full_like	9.4380ms	9.1173ms	109.6818 Ops/s	108.7795 Ops/s	$\color{#35bf28}+0.83\%$
test_zeros_like	5.3904ms	4.3279ms	231.0610 Ops/s	234.0943 Ops/s	$\color{#d91a1a}-1.30\%$
test_ones_like	4.9204ms	4.2151ms	237.2421 Ops/s	236.4597 Ops/s	$\color{#35bf28}+0.33\%$
test_clone	6.5047ms	6.3529ms	157.4075 Ops/s	109.5662 Ops/s	$\textbf{\color{#35bf28}+43.66\%}$
test_squeeze	59.2510μs	10.0455μs	99.5475 KOps/s	97.5498 KOps/s	$\color{#35bf28}+2.05\%$
test_unsqueeze	0.1236ms	71.7382μs	13.9396 KOps/s	12.8491 KOps/s	$\textbf{\color{#35bf28}+8.49\%}$
test_split	0.3952ms	0.1776ms	5.6293 KOps/s	5.7615 KOps/s	$\color{#d91a1a}-2.29\%$
test_permute	0.2421ms	0.1851ms	5.4023 KOps/s	5.3762 KOps/s	$\color{#35bf28}+0.49\%$
test_stack	50.6066ms	50.3276ms	19.8698 Ops/s	19.7511 Ops/s	$\color{#35bf28}+0.60\%$
test_cat	50.5404ms	50.3180ms	19.8736 Ops/s	19.7697 Ops/s	$\color{#35bf28}+0.53\%$