[Feature] ProbabilisticTensorDictSequential.default_interaction_type

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] ProbabilisticTensorDictSequential.default_interaction_type #1123

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}40$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	41.1770μs	16.7187μs	59.8132 KOps/s	53.1047 KOps/s	$\textbf{\color{#35bf28}+12.63\%}$
test_plain_set_stack_nested	46.6880μs	17.0168μs	58.7653 KOps/s	53.3213 KOps/s	$\textbf{\color{#35bf28}+10.21\%}$
test_plain_set_nested_inplace	57.6080μs	18.6290μs	53.6797 KOps/s	48.9709 KOps/s	$\textbf{\color{#35bf28}+9.62\%}$
test_plain_set_stack_nested_inplace	56.2050μs	18.8841μs	52.9545 KOps/s	48.9320 KOps/s	$\textbf{\color{#35bf28}+8.22\%}$
test_items	39.5140μs	4.1183μs	242.8166 KOps/s	237.5566 KOps/s	$\color{#35bf28}+2.21\%$
test_items_nested	0.4956ms	0.3989ms	2.5067 KOps/s	2.5260 KOps/s	$\color{#d91a1a}-0.76\%$
test_items_nested_locked	0.5460ms	0.3974ms	2.5161 KOps/s	2.5226 KOps/s	$\color{#d91a1a}-0.26\%$
test_items_nested_leaf	0.1446ms	71.7985μs	13.9279 KOps/s	14.1811 KOps/s	$\color{#d91a1a}-1.79\%$
test_items_stack_nested	0.4504ms	0.3979ms	2.5132 KOps/s	2.4864 KOps/s	$\color{#35bf28}+1.08\%$
test_items_stack_nested_leaf	0.1408ms	72.1359μs	13.8627 KOps/s	13.5444 KOps/s	$\color{#35bf28}+2.35\%$
test_items_stack_nested_locked	0.6687ms	0.4000ms	2.4999 KOps/s	2.4891 KOps/s	$\color{#35bf28}+0.43\%$
test_keys	37.2000μs	3.5725μs	279.9121 KOps/s	287.1015 KOps/s	$\color{#d91a1a}-2.50\%$
test_keys_nested	0.2224ms	0.1367ms	7.3172 KOps/s	7.2617 KOps/s	$\color{#35bf28}+0.76\%$
test_keys_nested_locked	1.9192ms	0.1431ms	6.9902 KOps/s	7.0524 KOps/s	$\color{#d91a1a}-0.88\%$
test_keys_nested_leaf	0.2324ms	0.1178ms	8.4856 KOps/s	8.6778 KOps/s	$\color{#d91a1a}-2.21\%$
test_keys_stack_nested	0.3248ms	0.1376ms	7.2650 KOps/s	7.2476 KOps/s	$\color{#35bf28}+0.24\%$
test_keys_stack_nested_leaf	0.2304ms	0.1178ms	8.4895 KOps/s	8.5987 KOps/s	$\color{#d91a1a}-1.27\%$
test_keys_stack_nested_locked	0.2081ms	0.1419ms	7.0469 KOps/s	6.9925 KOps/s	$\color{#35bf28}+0.78\%$
test_values	8.5178μs	1.0347μs	966.4209 KOps/s	968.7885 KOps/s	$\color{#d91a1a}-0.24\%$
test_values_nested	98.7650μs	54.6649μs	18.2933 KOps/s	18.2279 KOps/s	$\color{#35bf28}+0.36\%$
test_values_nested_locked	0.1103ms	55.0326μs	18.1710 KOps/s	17.3944 KOps/s	$\color{#35bf28}+4.47\%$
test_values_nested_leaf	0.1167ms	59.1706μs	16.9003 KOps/s	16.7482 KOps/s	$\color{#35bf28}+0.91\%$
test_values_stack_nested	0.1092ms	54.8410μs	18.2345 KOps/s	17.8686 KOps/s	$\color{#35bf28}+2.05\%$
test_values_stack_nested_leaf	0.1199ms	59.8239μs	16.7157 KOps/s	16.6248 KOps/s	$\color{#35bf28}+0.55\%$
test_values_stack_nested_locked	0.1166ms	54.9245μs	18.2068 KOps/s	17.9456 KOps/s	$\color{#35bf28}+1.46\%$
test_membership	24.6160μs	0.8986μs	1.1128 MOps/s	1.3826 MOps/s	$\textbf{\color{#d91a1a}-19.51\%}$
test_membership_nested	21.5310μs	3.0007μs	333.2520 KOps/s	339.9992 KOps/s	$\color{#d91a1a}-1.98\%$
test_membership_nested_leaf	34.5450μs	3.0228μs	330.8138 KOps/s	335.9451 KOps/s	$\color{#d91a1a}-1.53\%$
test_membership_stacked_nested	23.5740μs	2.9924μs	334.1761 KOps/s	337.3948 KOps/s	$\color{#d91a1a}-0.95\%$
test_membership_stacked_nested_leaf	22.7220μs	2.9936μs	334.0479 KOps/s	341.6974 KOps/s	$\color{#d91a1a}-2.24\%$
test_membership_nested_last	25.8280μs	4.2710μs	234.1380 KOps/s	237.8577 KOps/s	$\color{#d91a1a}-1.56\%$
test_membership_nested_leaf_last	36.3970μs	4.3071μs	232.1748 KOps/s	236.7947 KOps/s	$\color{#d91a1a}-1.95\%$
test_membership_stacked_nested_last	28.6540μs	4.2639μs	234.5254 KOps/s	150.6598 KOps/s	$\textbf{\color{#35bf28}+55.67\%}$
test_membership_stacked_nested_leaf_last	28.0320μs	4.2785μs	233.7284 KOps/s	147.4782 KOps/s	$\textbf{\color{#35bf28}+58.48\%}$
test_nested_getleaf	32.5810μs	10.9375μs	91.4289 KOps/s	90.5232 KOps/s	$\color{#35bf28}+1.00\%$
test_nested_get	44.2430μs	10.3905μs	96.2417 KOps/s	96.0901 KOps/s	$\color{#35bf28}+0.16\%$
test_stacked_getleaf	45.9160μs	10.7869μs	92.7054 KOps/s	93.4741 KOps/s	$\color{#d91a1a}-0.82\%$
test_stacked_get	28.8340μs	10.4217μs	95.9535 KOps/s	95.8040 KOps/s	$\color{#35bf28}+0.16\%$
test_nested_getitemleaf	48.1700μs	11.3046μs	88.4598 KOps/s	88.2442 KOps/s	$\color{#35bf28}+0.24\%$
test_nested_getitem	51.1660μs	10.7305μs	93.1925 KOps/s	94.3421 KOps/s	$\color{#d91a1a}-1.22\%$
test_stacked_getitemleaf	34.3340μs	11.3613μs	88.0185 KOps/s	87.3167 KOps/s	$\color{#35bf28}+0.80\%$
test_stacked_getitem	51.9470μs	10.4347μs	95.8345 KOps/s	95.5720 KOps/s	$\color{#35bf28}+0.27\%$
test_lock_nested	2.8392ms	0.4383ms	2.2817 KOps/s	2.2584 KOps/s	$\color{#35bf28}+1.03\%$
test_lock_stack_nested	0.6772ms	0.4127ms	2.4230 KOps/s	2.4186 KOps/s	$\color{#35bf28}+0.18\%$
test_unlock_nested	0.8338ms	0.3553ms	2.8144 KOps/s	2.7705 KOps/s	$\color{#35bf28}+1.58\%$
test_unlock_stack_nested	0.6145ms	0.3288ms	3.0414 KOps/s	3.0175 KOps/s	$\color{#35bf28}+0.79\%$
test_flatten_speed	0.1739ms	94.9887μs	10.5276 KOps/s	10.5635 KOps/s	$\color{#d91a1a}-0.34\%$
test_unflatten_speed	0.7099ms	0.4995ms	2.0021 KOps/s	2.0017 KOps/s	$\color{#35bf28}+0.02\%$
test_common_ops	1.3715ms	0.7238ms	1.3815 KOps/s	1.2262 KOps/s	$\textbf{\color{#35bf28}+12.67\%}$
test_creation	22.1520μs	2.0685μs	483.4404 KOps/s	492.3215 KOps/s	$\color{#d91a1a}-1.80\%$
test_creation_empty	41.3570μs	9.3580μs	106.8609 KOps/s	76.6531 KOps/s	$\textbf{\color{#35bf28}+39.41\%}$
test_creation_nested_1	54.1310μs	11.9613μs	83.6032 KOps/s	62.4528 KOps/s	$\textbf{\color{#35bf28}+33.87\%}$
test_creation_nested_2	1.4653ms	16.5849μs	60.2957 KOps/s	49.3353 KOps/s	$\textbf{\color{#35bf28}+22.22\%}$
test_clone	57.6880μs	12.9465μs	77.2408 KOps/s	76.2568 KOps/s	$\color{#35bf28}+1.29\%$
test_getitem[int]	1.0398ms	12.4879μs	80.0778 KOps/s	82.6541 KOps/s	$\color{#d91a1a}-3.12\%$
test_getitem[slice_int]	0.1337ms	24.2521μs	41.2335 KOps/s	41.1359 KOps/s	$\color{#35bf28}+0.24\%$
test_getitem[range]	0.1667ms	47.4663μs	21.0676 KOps/s	21.0941 KOps/s	$\color{#d91a1a}-0.13\%$
test_getitem[tuple]	0.1282ms	20.0742μs	49.8153 KOps/s	48.9811 KOps/s	$\color{#35bf28}+1.70\%$
test_getitem[list]	0.1635ms	42.6308μs	23.4572 KOps/s	23.0182 KOps/s	$\color{#35bf28}+1.91\%$
test_setitem_dim[int]	44.8740μs	24.3053μs	41.1433 KOps/s	36.4696 KOps/s	$\textbf{\color{#35bf28}+12.82\%}$
test_setitem_dim[slice_int]	0.1067ms	49.8956μs	20.0419 KOps/s	18.5475 KOps/s	$\textbf{\color{#35bf28}+8.06\%}$
test_setitem_dim[range]	96.9510μs	70.9718μs	14.0901 KOps/s	13.6671 KOps/s	$\color{#35bf28}+3.10\%$
test_setitem_dim[tuple]	62.4860μs	39.7297μs	25.1701 KOps/s	22.8902 KOps/s	$\textbf{\color{#35bf28}+9.96\%}$
test_setitem	68.8790μs	19.1285μs	52.2781 KOps/s	47.2394 KOps/s	$\textbf{\color{#35bf28}+10.67\%}$
test_set	65.8130μs	18.7017μs	53.4711 KOps/s	48.7394 KOps/s	$\textbf{\color{#35bf28}+9.71\%}$
test_set_shared	2.7369ms	0.1676ms	5.9670 KOps/s	6.0224 KOps/s	$\color{#d91a1a}-0.92\%$
test_update	0.1310ms	21.4877μs	46.5382 KOps/s	41.4337 KOps/s	$\textbf{\color{#35bf28}+12.32\%}$
test_update_nested	0.1290ms	30.5643μs	32.7179 KOps/s	28.9172 KOps/s	$\textbf{\color{#35bf28}+13.14\%}$
test_update__nested	0.1460ms	32.3612μs	30.9012 KOps/s	30.3611 KOps/s	$\color{#35bf28}+1.78\%$
test_set_nested	64.0400μs	20.7385μs	48.2194 KOps/s	44.3547 KOps/s	$\textbf{\color{#35bf28}+8.71\%}$
test_set_nested_new	77.3850μs	24.7150μs	40.4613 KOps/s	35.6441 KOps/s	$\textbf{\color{#35bf28}+13.51\%}$
test_select	0.2056ms	40.5068μs	24.6872 KOps/s	22.7711 KOps/s	$\textbf{\color{#35bf28}+8.41\%}$
test_select_nested	0.1221ms	59.4621μs	16.8174 KOps/s	17.1576 KOps/s	$\color{#d91a1a}-1.98\%$
test_exclude_nested	0.1553ms	78.0829μs	12.8069 KOps/s	12.9250 KOps/s	$\color{#d91a1a}-0.91\%$
test_empty[True]	0.5226ms	0.3808ms	2.6261 KOps/s	2.6394 KOps/s	$\color{#d91a1a}-0.50\%$
test_empty[False]	6.2645μs	1.2188μs	820.4525 KOps/s	824.3291 KOps/s	$\color{#d91a1a}-0.47\%$
test_unbind_speed	0.3382ms	0.2627ms	3.8071 KOps/s	3.7633 KOps/s	$\color{#35bf28}+1.16\%$
test_unbind_speed_stack0	0.3313ms	0.2613ms	3.8276 KOps/s	3.8262 KOps/s	$\color{#35bf28}+0.04\%$
test_unbind_speed_stack1	93.9145ms	0.7536ms	1.3269 KOps/s	1.4291 KOps/s	$\textbf{\color{#d91a1a}-7.15\%}$
test_split	85.1642ms	1.6563ms	603.7689 Ops/s	585.8142 Ops/s	$\color{#35bf28}+3.06\%$
test_chunk	93.1928ms	1.6765ms	596.4933 Ops/s	589.0529 Ops/s	$\color{#35bf28}+1.26\%$
test_consolidate_njt[False-None]	8.5098ms	8.0410ms	124.3632 Ops/s	122.5849 Ops/s	$\color{#35bf28}+1.45\%$
test_creation[device0]	0.2227ms	89.3511μs	11.1918 KOps/s	10.8362 KOps/s	$\color{#35bf28}+3.28\%$
test_creation_from_tensor	2.8395ms	93.6423μs	10.6789 KOps/s	10.5453 KOps/s	$\color{#35bf28}+1.27\%$
test_add_one[memmap_tensor0]	0.2003ms	4.8477μs	206.2835 KOps/s	207.9880 KOps/s	$\color{#d91a1a}-0.82\%$
test_contiguous[memmap_tensor0]	31.0780μs	0.5056μs	1.9779 MOps/s	1.9387 MOps/s	$\color{#35bf28}+2.02\%$
test_stack[memmap_tensor0]	45.4450μs	3.2644μs	306.3364 KOps/s	291.3357 KOps/s	$\textbf{\color{#35bf28}+5.15\%}$
test_memmaptd_index	1.1111ms	0.2352ms	4.2518 KOps/s	4.0689 KOps/s	$\color{#35bf28}+4.49\%$
test_memmaptd_index_astensor	0.7433ms	0.3099ms	3.2271 KOps/s	3.0962 KOps/s	$\color{#35bf28}+4.23\%$
test_memmaptd_index_op	0.8958ms	0.5418ms	1.8458 KOps/s	1.6273 KOps/s	$\textbf{\color{#35bf28}+13.43\%}$
test_serialize_model	0.1235s	0.1115s	8.9671 Ops/s	7.8193 Ops/s	$\textbf{\color{#35bf28}+14.68\%}$
test_serialize_model_pickle	0.4865s	0.3941s	2.5375 Ops/s	2.5569 Ops/s	$\color{#d91a1a}-0.76\%$
test_serialize_weights	0.2118s	0.1243s	8.0418 Ops/s	9.0615 Ops/s	$\textbf{\color{#d91a1a}-11.25\%}$
test_serialize_weights_returnearly	0.1738s	0.1589s	6.2941 Ops/s	6.3041 Ops/s	$\color{#d91a1a}-0.16\%$
test_serialize_weights_pickle	0.5638s	0.4304s	2.3232 Ops/s	2.4394 Ops/s	$\color{#d91a1a}-4.76\%$
test_serialize_weights_filesystem	0.1439s	0.1399s	7.1502 Ops/s	6.5235 Ops/s	$\textbf{\color{#35bf28}+9.61\%}$
test_serialize_model_filesystem	0.2400s	0.1581s	6.3252 Ops/s	6.8018 Ops/s	$\textbf{\color{#d91a1a}-7.01\%}$
test_reshape_pytree	68.2370μs	27.2431μs	36.7065 KOps/s	36.9135 KOps/s	$\color{#d91a1a}-0.56\%$
test_reshape_td	62.5570μs	32.5005μs	30.7688 KOps/s	30.3191 KOps/s	$\color{#35bf28}+1.48\%$
test_view_pytree	92.8940μs	27.0365μs	36.9870 KOps/s	37.0520 KOps/s	$\color{#d91a1a}-0.18\%$
test_view_td	83.3460μs	37.3213μs	26.7944 KOps/s	26.9942 KOps/s	$\color{#d91a1a}-0.74\%$
test_unbind_pytree	88.1650μs	31.0185μs	32.2389 KOps/s	32.5067 KOps/s	$\color{#d91a1a}-0.82\%$
test_unbind_td	0.3296ms	38.7517μs	25.8053 KOps/s	25.3434 KOps/s	$\color{#35bf28}+1.82\%$
test_split_pytree	72.7260μs	30.2705μs	33.0355 KOps/s	32.7856 KOps/s	$\color{#35bf28}+0.76\%$
test_split_td	0.5476ms	42.9760μs	23.2688 KOps/s	22.5067 KOps/s	$\color{#35bf28}+3.39\%$
test_add_pytree	83.1350μs	36.0122μs	27.7684 KOps/s	27.6164 KOps/s	$\color{#35bf28}+0.55\%$
test_add_td	0.1281ms	50.0430μs	19.9828 KOps/s	17.9005 KOps/s	$\textbf{\color{#35bf28}+11.63\%}$
test_compile_add_one_nested[tensordict-compile]	0.1616ms	60.9400μs	16.4096 KOps/s	16.1851 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_add_one_nested[tensordict-eager]	0.3477ms	0.1609ms	6.2141 KOps/s	6.1378 KOps/s	$\color{#35bf28}+1.24\%$
test_compile_add_one_nested[pytree-compile]	0.1092ms	45.2832μs	22.0833 KOps/s	22.2696 KOps/s	$\color{#d91a1a}-0.84\%$
test_compile_add_one_nested[pytree-eager]	0.2091ms	0.1214ms	8.2366 KOps/s	8.3046 KOps/s	$\color{#d91a1a}-0.82\%$
test_compile_copy_nested[tensordict-compile]	72.3960μs	25.2305μs	39.6346 KOps/s	39.0923 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_copy_nested[tensordict-eager]	0.1143ms	53.5590μs	18.6710 KOps/s	19.0217 KOps/s	$\color{#d91a1a}-1.84\%$
test_compile_copy_nested[pytree-compile]	0.1565ms	79.1941μs	12.6272 KOps/s	12.6687 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_copy_nested[pytree-eager]	0.1363ms	68.9315μs	14.5072 KOps/s	14.6342 KOps/s	$\color{#d91a1a}-0.87\%$
test_compile_add_one_flat[tensordict-compile]	0.2120ms	0.1023ms	9.7774 KOps/s	9.5384 KOps/s	$\color{#35bf28}+2.51\%$
test_compile_add_one_flat[tensordict-eager]	0.3230ms	0.1995ms	5.0136 KOps/s	5.1408 KOps/s	$\color{#d91a1a}-2.47\%$
test_compile_add_one_flat[tensorclass-compile]	0.1032ms	43.6213μs	22.9246 KOps/s	22.5364 KOps/s	$\color{#35bf28}+1.72\%$
test_compile_add_one_flat[tensorclass-eager]	0.3391ms	61.3995μs	16.2868 KOps/s	15.9640 KOps/s	$\color{#35bf28}+2.02\%$
test_compile_add_one_flat[pytree-compile]	0.1767ms	0.1018ms	9.8263 KOps/s	9.9607 KOps/s	$\color{#d91a1a}-1.35\%$
test_compile_add_one_flat[pytree-eager]	0.3109ms	0.2025ms	4.9377 KOps/s	4.9036 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_add_self_flat[tensordict-eager]	0.2742ms	0.2094ms	4.7748 KOps/s	4.8250 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_add_self_flat[tensordict-compile]	0.2475ms	0.1085ms	9.2193 KOps/s	9.5484 KOps/s	$\color{#d91a1a}-3.45\%$
test_compile_add_self_flat[tensorclass-eager]	0.2360ms	54.6690μs	18.2919 KOps/s	17.5536 KOps/s	$\color{#35bf28}+4.21\%$
test_compile_add_self_flat[tensorclass-compile]	0.1104ms	45.0214μs	22.2117 KOps/s	21.8331 KOps/s	$\color{#35bf28}+1.73\%$
test_compile_add_self_flat[pytree-eager]	0.6074ms	0.1603ms	6.2399 KOps/s	6.1758 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_add_self_flat[pytree-compile]	0.2164ms	0.1035ms	9.6649 KOps/s	9.8672 KOps/s	$\color{#d91a1a}-2.05\%$
test_compile_copy_flat[tensordict-compile]	74.3700μs	20.3062μs	49.2461 KOps/s	48.1794 KOps/s	$\color{#35bf28}+2.21\%$
test_compile_copy_flat[tensordict-eager]	0.1220ms	58.6720μs	17.0439 KOps/s	17.0046 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_copy_flat[pytree-compile]	0.1738ms	81.3909μs	12.2864 KOps/s	12.2361 KOps/s	$\color{#35bf28}+0.41\%$
test_compile_copy_flat[pytree-eager]	0.1497ms	71.4358μs	13.9986 KOps/s	14.1038 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_assign_and_add[tensordict-compile]	0.3168ms	0.2106ms	4.7476 KOps/s	4.7230 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_assign_and_add[tensordict-eager]	2.0910ms	1.2970ms	771.0135 Ops/s	793.4011 Ops/s	$\color{#d91a1a}-2.82\%$
test_compile_assign_and_add[pytree-compile]	0.2945ms	0.2034ms	4.9166 KOps/s	5.1015 KOps/s	$\color{#d91a1a}-3.62\%$
test_compile_assign_and_add[pytree-eager]	0.9767ms	0.7808ms	1.2807 KOps/s	1.2994 KOps/s	$\color{#d91a1a}-1.44\%$
test_compile_assign_and_add_stack[compile]	0.8857ms	0.4642ms	2.1541 KOps/s	2.2414 KOps/s	$\color{#d91a1a}-3.90\%$
test_compile_assign_and_add_stack[eager]	2.7875ms	2.5020ms	399.6870 Ops/s	362.7539 Ops/s	$\textbf{\color{#35bf28}+10.18\%}$
test_compile_indexing[tensor-tensordict-compile]	96.3100μs	35.5267μs	28.1478 KOps/s	27.3070 KOps/s	$\color{#35bf28}+3.08\%$
test_compile_indexing[tensor-tensordict-eager]	0.3684ms	31.3312μs	31.9171 KOps/s	30.4063 KOps/s	$\color{#35bf28}+4.97\%$
test_compile_indexing[tensor-tensorclass-compile]	81.0220μs	29.3289μs	34.0961 KOps/s	33.9665 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_indexing[tensor-tensorclass-eager]	74.5000μs	23.2503μs	43.0101 KOps/s	41.0317 KOps/s	$\color{#35bf28}+4.82\%$
test_compile_indexing[tensor-pytree-compile]	82.1330μs	29.5983μs	33.7857 KOps/s	33.7787 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_indexing[tensor-pytree-eager]	99.2760μs	23.1937μs	43.1152 KOps/s	41.3881 KOps/s	$\color{#35bf28}+4.17\%$
test_compile_indexing[slice-tensordict-compile]	0.1084ms	50.9107μs	19.6422 KOps/s	19.5317 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_indexing[slice-tensordict-eager]	0.4306ms	18.7999μs	53.1917 KOps/s	49.1108 KOps/s	$\textbf{\color{#35bf28}+8.31\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1049ms	43.5976μs	22.9370 KOps/s	22.8669 KOps/s	$\color{#35bf28}+0.31\%$
test_compile_indexing[slice-tensorclass-eager]	52.2070μs	19.0621μs	52.4602 KOps/s	52.1986 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_indexing[slice-pytree-compile]	98.5730μs	44.4183μs	22.5133 KOps/s	22.2883 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_indexing[slice-pytree-eager]	66.7460μs	19.0210μs	52.5735 KOps/s	52.0423 KOps/s	$\color{#35bf28}+1.02\%$
test_compile_indexing[int-tensordict-compile]	0.1569ms	52.8332μs	18.9275 KOps/s	19.1895 KOps/s	$\color{#d91a1a}-1.37\%$
test_compile_indexing[int-tensordict-eager]	0.8099ms	18.6863μs	53.5153 KOps/s	48.9794 KOps/s	$\textbf{\color{#35bf28}+9.26\%}$
test_compile_indexing[int-tensorclass-compile]	0.1344ms	45.1423μs	22.1522 KOps/s	22.3100 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_indexing[int-tensorclass-eager]	67.5970μs	18.9023μs	52.9035 KOps/s	52.9478 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_indexing[int-pytree-compile]	0.1434ms	45.6990μs	21.8823 KOps/s	22.3312 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_indexing[int-pytree-eager]	64.3610μs	18.8442μs	53.0667 KOps/s	52.5909 KOps/s	$\color{#35bf28}+0.90\%$
test_mod_add[eager]	92.3420μs	31.5143μs	31.7316 KOps/s	28.1754 KOps/s	$\textbf{\color{#35bf28}+12.62\%}$
test_mod_add[compile]	96.8110μs	46.0516μs	21.7148 KOps/s	21.3647 KOps/s	$\color{#35bf28}+1.64\%$
test_mod_add[compile-overhead]	0.1218ms	47.2113μs	21.1814 KOps/s	21.5652 KOps/s	$\color{#d91a1a}-1.78\%$
test_mod_wrap[eager]	0.3751ms	0.2162ms	4.6257 KOps/s	4.4667 KOps/s	$\color{#35bf28}+3.56\%$
test_mod_wrap[compile]	0.3754ms	0.2035ms	4.9141 KOps/s	4.8685 KOps/s	$\color{#35bf28}+0.94\%$
test_mod_wrap[compile-overhead]	0.3828ms	0.2009ms	4.9781 KOps/s	4.8949 KOps/s	$\color{#35bf28}+1.70\%$
test_mod_wrap_and_backward[eager]	12.1367ms	10.5337ms	94.9337 Ops/s	86.3798 Ops/s	$\textbf{\color{#35bf28}+9.90\%}$
test_mod_wrap_and_backward[compile]	11.6384ms	10.3106ms	96.9871 Ops/s	74.4883 Ops/s	$\textbf{\color{#35bf28}+30.20\%}$
test_mod_wrap_and_backward[compile-overhead]	11.9211ms	10.3168ms	96.9291 Ops/s	73.5329 Ops/s	$\textbf{\color{#35bf28}+31.82\%}$
test_seq_add[eager]	0.2113ms	0.1050ms	9.5217 KOps/s	8.4889 KOps/s	$\textbf{\color{#35bf28}+12.17\%}$
test_seq_add[compile]	0.1284ms	59.8518μs	16.7079 KOps/s	16.2981 KOps/s	$\color{#35bf28}+2.51\%$
test_seq_add[compile-overhead]	0.1152ms	58.0739μs	17.2194 KOps/s	17.3837 KOps/s	$\color{#d91a1a}-0.94\%$
test_seq_wrap[eager]	0.4984ms	0.4166ms	2.4002 KOps/s	2.1739 KOps/s	$\textbf{\color{#35bf28}+10.41\%}$
test_seq_wrap[compile]	0.4038ms	0.2277ms	4.3921 KOps/s	4.3468 KOps/s	$\color{#35bf28}+1.04\%$
test_seq_wrap[compile-overhead]	0.4141ms	0.2249ms	4.4466 KOps/s	4.3607 KOps/s	$\color{#35bf28}+1.97\%$
test_func_call_runtime[False-eager]	0.6556ms	0.5196ms	1.9244 KOps/s	1.8183 KOps/s	$\textbf{\color{#35bf28}+5.83\%}$
test_func_call_runtime[False-compile]	0.6087ms	0.4236ms	2.3607 KOps/s	2.3441 KOps/s	$\color{#35bf28}+0.71\%$
test_func_call_runtime[False-compile-overhead]	0.8750ms	0.4300ms	2.3255 KOps/s	2.3450 KOps/s	$\color{#d91a1a}-0.83\%$
test_func_call_runtime[True-eager]	1.5547ms	0.7363ms	1.3582 KOps/s	1.3127 KOps/s	$\color{#35bf28}+3.46\%$
test_func_call_runtime[True-compile]	0.5608ms	0.4590ms	2.1786 KOps/s	2.1526 KOps/s	$\color{#35bf28}+1.21\%$
test_func_call_runtime[True-compile-overhead]	0.5538ms	0.4624ms	2.1625 KOps/s	2.1683 KOps/s	$\color{#d91a1a}-0.27\%$
test_func_call_cm_runtime[False-eager]	0.7114ms	0.5183ms	1.9294 KOps/s	1.8336 KOps/s	$\textbf{\color{#35bf28}+5.22\%}$
test_func_call_cm_runtime[False-compile]	0.7628ms	0.4238ms	2.3596 KOps/s	2.3615 KOps/s	$\color{#d91a1a}-0.08\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5161ms	0.4233ms	2.3624 KOps/s	2.3555 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[True-eager]	1.0022ms	0.8673ms	1.1530 KOps/s	1.1086 KOps/s	$\color{#35bf28}+4.00\%$
test_func_call_cm_runtime[True-compile]	0.6278ms	0.4871ms	2.0528 KOps/s	2.0393 KOps/s	$\color{#35bf28}+0.66\%$
test_func_call_cm_runtime[True-compile-overhead]	0.8226ms	0.4858ms	2.0583 KOps/s	2.0658 KOps/s	$\color{#d91a1a}-0.36\%$
test_vmap_func_call_cm_runtime[eager]	2.2831ms	1.8558ms	538.8429 Ops/s	531.8743 Ops/s	$\color{#35bf28}+1.31\%$
test_vmap_func_call_cm_runtime[compile]	1.0987ms	0.5188ms	1.9276 KOps/s	1.9475 KOps/s	$\color{#d91a1a}-1.02\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9169ms	0.5181ms	1.9302 KOps/s	1.9571 KOps/s	$\color{#d91a1a}-1.38\%$
test_distributed	0.2507ms	0.1250ms	8.0028 KOps/s	7.9573 KOps/s	$\color{#35bf28}+0.57\%$
test_tdmodule	78.3760μs	24.0022μs	41.6629 KOps/s	35.3088 KOps/s	$\textbf{\color{#35bf28}+18.00\%}$
test_tdmodule_dispatch	76.3830μs	43.8194μs	22.8210 KOps/s	19.2283 KOps/s	$\textbf{\color{#35bf28}+18.68\%}$
test_tdseq	42.8500μs	23.7469μs	42.1108 KOps/s	36.1481 KOps/s	$\textbf{\color{#35bf28}+16.50\%}$
test_tdseq_dispatch	66.9960μs	46.8906μs	21.3262 KOps/s	18.5131 KOps/s	$\textbf{\color{#35bf28}+15.20\%}$
test_instantiation_functorch	1.7887ms	1.5178ms	658.8635 Ops/s	644.0961 Ops/s	$\color{#35bf28}+2.29\%$
test_exec_functorch	0.3868ms	0.1780ms	5.6193 KOps/s	5.5083 KOps/s	$\color{#35bf28}+2.02\%$
test_exec_functional_call	0.3185ms	0.1715ms	5.8296 KOps/s	5.7053 KOps/s	$\color{#35bf28}+2.18\%$
test_exec_td_decorator	0.5133ms	0.2294ms	4.3596 KOps/s	4.2541 KOps/s	$\color{#35bf28}+2.48\%$
test_vmap_mlp_speed_decorator[True-True]	1.1076ms	0.6419ms	1.5580 KOps/s	1.5122 KOps/s	$\color{#35bf28}+3.03\%$
test_vmap_mlp_speed_decorator[True-False]	0.8521ms	0.6386ms	1.5658 KOps/s	1.5185 KOps/s	$\color{#35bf28}+3.12\%$
test_vmap_mlp_speed_decorator[False-True]	0.7587ms	0.5204ms	1.9216 KOps/s	1.8935 KOps/s	$\color{#35bf28}+1.49\%$
test_vmap_mlp_speed_decorator[False-False]	0.8307ms	0.5223ms	1.9148 KOps/s	1.8952 KOps/s	$\color{#35bf28}+1.04\%$
test_to_module_speed[True]	1.6128ms	1.2797ms	781.4052 Ops/s	786.4339 Ops/s	$\color{#d91a1a}-0.64\%$
test_to_module_speed[False]	1.7842ms	1.2546ms	797.0451 Ops/s	806.4184 Ops/s	$\color{#d91a1a}-1.16\%$
test_tc_init	75.5810μs	45.8704μs	21.8005 KOps/s	20.8801 KOps/s	$\color{#35bf28}+4.41\%$
test_tc_init_nested	0.1566ms	91.6221μs	10.9144 KOps/s	10.6768 KOps/s	$\color{#35bf28}+2.23\%$
test_tc_first_layer_tensor	38.2610μs	1.5020μs	665.7629 KOps/s	644.3992 KOps/s	$\color{#35bf28}+3.32\%$
test_tc_first_layer_nontensor	25.7380μs	4.7576μs	210.1897 KOps/s	210.3187 KOps/s	$\color{#d91a1a}-0.06\%$
test_tc_second_layer_tensor	30.6860μs	2.8013μs	356.9749 KOps/s	341.9154 KOps/s	$\color{#35bf28}+4.40\%$
test_tc_second_layer_nontensor	37.4000μs	6.2809μs	159.2119 KOps/s	161.4883 KOps/s	$\color{#d91a1a}-1.41\%$
test_unbind	0.2025s	12.2547ms	81.6012 Ops/s	82.2283 Ops/s	$\color{#d91a1a}-0.76\%$
test_full_like	7.1053ms	6.5812ms	151.9483 Ops/s	150.6340 Ops/s	$\color{#35bf28}+0.87\%$
test_zeros_like	2.8304ms	2.5175ms	397.2116 Ops/s	390.6446 Ops/s	$\color{#35bf28}+1.68\%$
test_ones_like	3.3410ms	2.9890ms	334.5621 Ops/s	332.7533 Ops/s	$\color{#35bf28}+0.54\%$
test_clone	4.8979ms	4.5872ms	217.9970 Ops/s	216.9350 Ops/s	$\color{#35bf28}+0.49\%$
test_squeeze	53.1100μs	12.0488μs	82.9958 KOps/s	82.6191 KOps/s	$\color{#35bf28}+0.46\%$
test_unsqueeze	0.2671ms	90.6447μs	11.0321 KOps/s	11.2770 KOps/s	$\color{#d91a1a}-2.17\%$
test_split	0.3650ms	0.1913ms	5.2260 KOps/s	5.0681 KOps/s	$\color{#35bf28}+3.12\%$
test_permute	0.3440ms	0.2190ms	4.5656 KOps/s	4.4971 KOps/s	$\color{#35bf28}+1.52\%$
test_stack	30.0701ms	23.8993ms	41.8422 Ops/s	42.6195 Ops/s	$\color{#d91a1a}-1.82\%$
test_cat	31.6223ms	23.6961ms	42.2010 Ops/s	43.1356 Ops/s	$\color{#d91a1a}-2.17\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}27$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	30.9300μs	11.4266μs	87.5148 KOps/s	94.2472 KOps/s	$\textbf{\color{#d91a1a}-7.14\%}$
test_plain_set_stack_nested	28.7210μs	11.4154μs	87.6012 KOps/s	94.1052 KOps/s	$\textbf{\color{#d91a1a}-6.91\%}$
test_plain_set_nested_inplace	48.6300μs	12.2992μs	81.3058 KOps/s	85.8603 KOps/s	$\textbf{\color{#d91a1a}-5.30\%}$
test_plain_set_stack_nested_inplace	39.2810μs	12.2930μs	81.3471 KOps/s	86.8401 KOps/s	$\textbf{\color{#d91a1a}-6.33\%}$
test_items	36.0810μs	2.9201μs	342.4557 KOps/s	343.2325 KOps/s	$\color{#d91a1a}-0.23\%$
test_items_nested	0.3849ms	0.3569ms	2.8023 KOps/s	2.7922 KOps/s	$\color{#35bf28}+0.36\%$
test_items_nested_locked	0.3838ms	0.3554ms	2.8136 KOps/s	2.7732 KOps/s	$\color{#35bf28}+1.45\%$
test_items_nested_leaf	80.7000μs	58.5563μs	17.0776 KOps/s	17.2191 KOps/s	$\color{#d91a1a}-0.82\%$
test_items_stack_nested	0.4172ms	0.3598ms	2.7790 KOps/s	2.7776 KOps/s	$\color{#35bf28}+0.05\%$
test_items_stack_nested_leaf	0.1059ms	59.4525μs	16.8202 KOps/s	17.2489 KOps/s	$\color{#d91a1a}-2.49\%$
test_items_stack_nested_locked	0.3879ms	0.3591ms	2.7845 KOps/s	2.7771 KOps/s	$\color{#35bf28}+0.27\%$
test_keys	45.8510μs	3.4612μs	288.9181 KOps/s	291.5148 KOps/s	$\color{#d91a1a}-0.89\%$
test_keys_nested	95.2310μs	70.1369μs	14.2578 KOps/s	14.3293 KOps/s	$\color{#d91a1a}-0.50\%$
test_keys_nested_locked	0.7874ms	75.5475μs	13.2367 KOps/s	13.1775 KOps/s	$\color{#35bf28}+0.45\%$
test_keys_nested_leaf	87.8610μs	61.0188μs	16.3884 KOps/s	16.3376 KOps/s	$\color{#35bf28}+0.31\%$
test_keys_stack_nested	0.1028ms	71.0910μs	14.0665 KOps/s	14.2357 KOps/s	$\color{#d91a1a}-1.19\%$
test_keys_stack_nested_leaf	94.0710μs	62.1071μs	16.1012 KOps/s	16.3274 KOps/s	$\color{#d91a1a}-1.39\%$
test_keys_stack_nested_locked	0.2202ms	76.3645μs	13.0951 KOps/s	13.2147 KOps/s	$\color{#d91a1a}-0.90\%$
test_values	5.8852μs	0.8488μs	1.1782 MOps/s	748.2818 KOps/s	$\textbf{\color{#35bf28}+57.45\%}$
test_values_nested	55.5510μs	31.3454μs	31.9026 KOps/s	32.2606 KOps/s	$\color{#d91a1a}-1.11\%$
test_values_nested_locked	71.4210μs	32.9299μs	30.3676 KOps/s	30.7636 KOps/s	$\color{#d91a1a}-1.29\%$
test_values_nested_leaf	0.1367ms	33.3137μs	30.0177 KOps/s	29.6362 KOps/s	$\color{#35bf28}+1.29\%$
test_values_stack_nested	74.5010μs	31.5119μs	31.7340 KOps/s	31.9628 KOps/s	$\color{#d91a1a}-0.72\%$
test_values_stack_nested_leaf	96.0410μs	34.2219μs	29.2210 KOps/s	29.7552 KOps/s	$\color{#d91a1a}-1.80\%$
test_values_stack_nested_locked	69.5310μs	33.2089μs	30.1124 KOps/s	30.5042 KOps/s	$\color{#d91a1a}-1.28\%$
test_membership	1.7951μs	0.5053μs	1.9792 MOps/s	1.9821 MOps/s	$\color{#d91a1a}-0.15\%$
test_membership_nested	14.8755μs	1.9810μs	504.7952 KOps/s	482.1039 KOps/s	$\color{#35bf28}+4.71\%$
test_membership_nested_leaf	23.9800μs	2.0044μs	498.8903 KOps/s	481.5464 KOps/s	$\color{#35bf28}+3.60\%$
test_membership_stacked_nested	34.2900μs	2.0838μs	479.8923 KOps/s	465.8756 KOps/s	$\color{#35bf28}+3.01\%$
test_membership_stacked_nested_leaf	24.2700μs	2.1314μs	469.1654 KOps/s	465.1602 KOps/s	$\color{#35bf28}+0.86\%$
test_membership_nested_last	39.7000μs	2.9166μs	342.8625 KOps/s	333.3073 KOps/s	$\color{#35bf28}+2.87\%$
test_membership_nested_leaf_last	38.5810μs	2.9328μs	340.9678 KOps/s	331.9479 KOps/s	$\color{#35bf28}+2.72\%$
test_membership_stacked_nested_last	39.9700μs	2.9883μs	334.6434 KOps/s	332.5420 KOps/s	$\color{#35bf28}+0.63\%$
test_membership_stacked_nested_leaf_last	44.9410μs	2.9784μs	335.7543 KOps/s	342.0624 KOps/s	$\color{#d91a1a}-1.84\%$
test_nested_getleaf	55.9810μs	6.1383μs	162.9107 KOps/s	161.9080 KOps/s	$\color{#35bf28}+0.62\%$
test_nested_get	29.4800μs	5.8341μs	171.4061 KOps/s	169.2357 KOps/s	$\color{#35bf28}+1.28\%$
test_stacked_getleaf	55.3710μs	6.1490μs	162.6270 KOps/s	163.5429 KOps/s	$\color{#d91a1a}-0.56\%$
test_stacked_get	0.1714ms	5.8416μs	171.1856 KOps/s	170.8354 KOps/s	$\color{#35bf28}+0.20\%$
test_nested_getitemleaf	0.1908ms	6.2294μs	160.5291 KOps/s	158.7915 KOps/s	$\color{#35bf28}+1.09\%$
test_nested_getitem	36.5800μs	5.9156μs	169.0440 KOps/s	168.2255 KOps/s	$\color{#35bf28}+0.49\%$
test_stacked_getitemleaf	0.1618ms	6.2408μs	160.2356 KOps/s	158.4189 KOps/s	$\color{#35bf28}+1.15\%$
test_stacked_getitem	0.1667ms	5.9292μs	168.6560 KOps/s	167.6767 KOps/s	$\color{#35bf28}+0.58\%$
test_lock_nested	9.8122ms	0.3755ms	2.6634 KOps/s	2.6895 KOps/s	$\color{#d91a1a}-0.97\%$
test_lock_stack_nested	0.3818ms	0.3361ms	2.9754 KOps/s	2.9750 KOps/s	$\color{#35bf28}+0.01\%$
test_unlock_nested	0.6235ms	0.3030ms	3.3005 KOps/s	3.2778 KOps/s	$\color{#35bf28}+0.69\%$
test_unlock_stack_nested	0.3005ms	0.2732ms	3.6598 KOps/s	3.6365 KOps/s	$\color{#35bf28}+0.64\%$
test_flatten_speed	0.1202ms	75.5333μs	13.2392 KOps/s	13.6046 KOps/s	$\color{#d91a1a}-2.69\%$
test_unflatten_speed	0.3508ms	0.3067ms	3.2602 KOps/s	3.2876 KOps/s	$\color{#d91a1a}-0.83\%$
test_common_ops	1.7238ms	0.6032ms	1.6580 KOps/s	1.7337 KOps/s	$\color{#d91a1a}-4.37\%$
test_creation	58.4200μs	1.4782μs	676.4775 KOps/s	681.3520 KOps/s	$\color{#d91a1a}-0.72\%$
test_creation_empty	38.5700μs	9.0062μs	111.0347 KOps/s	139.4786 KOps/s	$\textbf{\color{#d91a1a}-20.39\%}$
test_creation_nested_1	57.0310μs	10.6506μs	93.8915 KOps/s	114.9178 KOps/s	$\textbf{\color{#d91a1a}-18.30\%}$
test_creation_nested_2	38.2510μs	13.2357μs	75.5533 KOps/s	89.0860 KOps/s	$\textbf{\color{#d91a1a}-15.19\%}$
test_clone	55.3100μs	9.7929μs	102.1145 KOps/s	100.1842 KOps/s	$\color{#35bf28}+1.93\%$
test_getitem[int]	1.4292ms	10.8087μs	92.5177 KOps/s	93.3440 KOps/s	$\color{#d91a1a}-0.89\%$
test_getitem[slice_int]	91.8786ms	28.9997μs	34.4831 KOps/s	47.4960 KOps/s	$\textbf{\color{#d91a1a}-27.40\%}$
test_getitem[range]	0.1775ms	36.8770μs	27.1172 KOps/s	27.3355 KOps/s	$\color{#d91a1a}-0.80\%$
test_getitem[tuple]	0.1047ms	18.3206μs	54.5833 KOps/s	54.5282 KOps/s	$\color{#35bf28}+0.10\%$
test_getitem[list]	0.1936ms	32.0084μs	31.2418 KOps/s	30.9210 KOps/s	$\color{#35bf28}+1.04\%$
test_setitem_dim[int]	39.5010μs	18.3927μs	54.3693 KOps/s	55.7716 KOps/s	$\color{#d91a1a}-2.51\%$
test_setitem_dim[slice_int]	58.2810μs	37.2107μs	26.8740 KOps/s	26.8826 KOps/s	$\color{#d91a1a}-0.03\%$
test_setitem_dim[range]	87.1310μs	52.2427μs	19.1414 KOps/s	19.1620 KOps/s	$\color{#d91a1a}-0.11\%$
test_setitem_dim[tuple]	51.2110μs	30.8660μs	32.3981 KOps/s	31.2624 KOps/s	$\color{#35bf28}+3.63\%$
test_setitem	0.2004ms	14.7341μs	67.8698 KOps/s	70.9896 KOps/s	$\color{#d91a1a}-4.39\%$
test_set	87.5410μs	14.5309μs	68.8188 KOps/s	73.6116 KOps/s	$\textbf{\color{#d91a1a}-6.51\%}$
test_set_shared	1.5743ms	0.1437ms	6.9568 KOps/s	6.9215 KOps/s	$\color{#35bf28}+0.51\%$
test_update	0.5786ms	17.8091μs	56.1511 KOps/s	62.8748 KOps/s	$\textbf{\color{#d91a1a}-10.69\%}$
test_update_nested	84.8010μs	22.8645μs	43.7359 KOps/s	46.8172 KOps/s	$\textbf{\color{#d91a1a}-6.58\%}$
test_update__nested	1.0405ms	23.5480μs	42.4665 KOps/s	42.5794 KOps/s	$\color{#d91a1a}-0.27\%$
test_set_nested	69.8500μs	15.8577μs	63.0607 KOps/s	67.7870 KOps/s	$\textbf{\color{#d91a1a}-6.97\%}$
test_set_nested_new	99.8510μs	18.0898μs	55.2796 KOps/s	58.7801 KOps/s	$\textbf{\color{#d91a1a}-5.96\%}$
test_select	0.1036ms	30.2171μs	33.0938 KOps/s	34.4782 KOps/s	$\color{#d91a1a}-4.02\%$
test_select_nested	0.1069ms	41.3041μs	24.2107 KOps/s	23.8198 KOps/s	$\color{#35bf28}+1.64\%$
test_exclude_nested	0.1007ms	60.7766μs	16.4537 KOps/s	16.1082 KOps/s	$\color{#35bf28}+2.14\%$
test_empty[True]	0.3134ms	0.2767ms	3.6135 KOps/s	3.5313 KOps/s	$\color{#35bf28}+2.33\%$
test_empty[False]	4.7481μs	0.7409μs	1.3497 MOps/s	1.3551 MOps/s	$\color{#d91a1a}-0.40\%$
test_to	85.0910μs	54.4977μs	18.3494 KOps/s	18.1373 KOps/s	$\color{#35bf28}+1.17\%$
test_to_nonblocking	0.1661ms	45.4255μs	22.0141 KOps/s	21.8733 KOps/s	$\color{#35bf28}+0.64\%$
test_unbind_speed	1.9527ms	0.2298ms	4.3510 KOps/s	4.3689 KOps/s	$\color{#d91a1a}-0.41\%$
test_unbind_speed_stack0	0.3289ms	0.2316ms	4.3170 KOps/s	4.3524 KOps/s	$\color{#d91a1a}-0.81\%$
test_unbind_speed_stack1	94.3523ms	0.6385ms	1.5661 KOps/s	1.5488 KOps/s	$\color{#35bf28}+1.12\%$
test_split	93.4102ms	1.6066ms	622.4310 Ops/s	584.0716 Ops/s	$\textbf{\color{#35bf28}+6.57\%}$
test_chunk	96.0006ms	1.7569ms	569.1717 Ops/s	632.5129 Ops/s	$\textbf{\color{#d91a1a}-10.01\%}$
test_consolidate[False-None]	3.0407ms	2.6118ms	382.8751 Ops/s	382.8363 Ops/s	$\color{#35bf28}+0.01\%$
test_consolidate[default-None]	1.7991ms	1.6772ms	596.2484 Ops/s	595.3768 Ops/s	$\color{#35bf28}+0.15\%$
test_consolidate[reduce-overhead-None]	1.8342ms	1.7019ms	587.5781 Ops/s	577.1710 Ops/s	$\color{#35bf28}+1.80\%$
test_consolidate_njt[False-None]	6.7684ms	6.5556ms	152.5422 Ops/s	112.0113 Ops/s	$\textbf{\color{#35bf28}+36.18\%}$
test_to[False-False-None]	1.8428ms	1.6660ms	600.2542 Ops/s	598.6226 Ops/s	$\color{#35bf28}+0.27\%$
test_to[True-False-None]	1.5326ms	1.2899ms	775.2554 Ops/s	762.9302 Ops/s	$\color{#35bf28}+1.62\%$
test_to[within-False-None]	0.2881s	5.1087ms	195.7457 Ops/s	244.3237 Ops/s	$\textbf{\color{#d91a1a}-19.88\%}$
test_to[True-default-None]	5.3310ms	5.0464ms	198.1595 Ops/s	192.1265 Ops/s	$\color{#35bf28}+3.14\%$
test_to_njt[False-False-None]	7.1391ms	6.9101ms	144.7164 Ops/s	135.0991 Ops/s	$\textbf{\color{#35bf28}+7.12\%}$
test_to_njt[True-False-None]	5.7716ms	5.4331ms	184.0585 Ops/s	172.1234 Ops/s	$\textbf{\color{#35bf28}+6.93\%}$
test_to_njt[within-False-None]	12.3783ms	12.1704ms	82.1665 Ops/s	79.7033 Ops/s	$\color{#35bf28}+3.09\%$
test_creation[device0]	0.4743ms	78.8622μs	12.6803 KOps/s	12.3747 KOps/s	$\color{#35bf28}+2.47\%$
test_creation_from_tensor	0.5026ms	82.0136μs	12.1931 KOps/s	12.1356 KOps/s	$\color{#35bf28}+0.47\%$
test_add_one[memmap_tensor0]	0.2661ms	6.2914μs	158.9464 KOps/s	160.1865 KOps/s	$\color{#d91a1a}-0.77\%$
test_contiguous[memmap_tensor0]	1.9120μs	0.4127μs	2.4232 MOps/s	2.4147 MOps/s	$\color{#35bf28}+0.35\%$
test_stack[memmap_tensor0]	0.1550ms	4.6555μs	214.8018 KOps/s	215.7048 KOps/s	$\color{#d91a1a}-0.42\%$
test_memmaptd_index	1.9308ms	0.2447ms	4.0866 KOps/s	3.8772 KOps/s	$\textbf{\color{#35bf28}+5.40\%}$
test_memmaptd_index_astensor	1.1084ms	0.3017ms	3.3150 KOps/s	3.1716 KOps/s	$\color{#35bf28}+4.52\%$
test_memmaptd_index_op	0.9756ms	0.5793ms	1.7261 KOps/s	1.7787 KOps/s	$\color{#d91a1a}-2.95\%$
test_serialize_model	0.1304s	0.1295s	7.7224 Ops/s	7.6815 Ops/s	$\color{#35bf28}+0.53\%$
test_serialize_model_pickle	1.3720s	1.1917s	0.8391 Ops/s	0.8236 Ops/s	$\color{#35bf28}+1.89\%$
test_serialize_weights	0.1310s	0.1296s	7.7173 Ops/s	7.6813 Ops/s	$\color{#35bf28}+0.47\%$
test_serialize_weights_returnearly	0.3376s	60.2201ms	16.6058 Ops/s	15.3539 Ops/s	$\textbf{\color{#35bf28}+8.15\%}$
test_serialize_weights_pickle	1.3806s	1.2177s	0.8213 Ops/s	0.8438 Ops/s	$\color{#d91a1a}-2.67\%$
test_reshape_pytree	51.6810μs	21.9838μs	45.4880 KOps/s	44.2887 KOps/s	$\color{#35bf28}+2.71\%$
test_reshape_td	0.1016ms	26.7068μs	37.4436 KOps/s	37.8715 KOps/s	$\color{#d91a1a}-1.13\%$
test_view_pytree	80.4900μs	21.9288μs	45.6022 KOps/s	44.6476 KOps/s	$\color{#35bf28}+2.14\%$
test_view_td	57.4610μs	28.4191μs	35.1876 KOps/s	33.9849 KOps/s	$\color{#35bf28}+3.54\%$
test_unbind_pytree	74.4010μs	27.9597μs	35.7657 KOps/s	35.3751 KOps/s	$\color{#35bf28}+1.10\%$
test_unbind_td	0.5659ms	34.9892μs	28.5803 KOps/s	27.7153 KOps/s	$\color{#35bf28}+3.12\%$
test_split_pytree	0.1239ms	29.4795μs	33.9219 KOps/s	32.5706 KOps/s	$\color{#35bf28}+4.15\%$
test_split_td	0.7346ms	37.6847μs	26.5360 KOps/s	25.8423 KOps/s	$\color{#35bf28}+2.68\%$
test_add_pytree	0.1524ms	32.2261μs	31.0307 KOps/s	30.4735 KOps/s	$\color{#35bf28}+1.83\%$
test_add_td	0.2257ms	46.8408μs	21.3489 KOps/s	22.5194 KOps/s	$\textbf{\color{#d91a1a}-5.20\%}$
test_compile_add_one_nested[tensordict-compile]	0.2667ms	0.1198ms	8.3459 KOps/s	7.8729 KOps/s	$\textbf{\color{#35bf28}+6.01\%}$
test_compile_add_one_nested[tensordict-eager]	0.2727ms	0.1235ms	8.0941 KOps/s	8.0150 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_add_one_nested[pytree-compile]	0.1382ms	95.8534μs	10.4326 KOps/s	10.1794 KOps/s	$\color{#35bf28}+2.49\%$
test_compile_add_one_nested[pytree-eager]	1.1265ms	0.1467ms	6.8146 KOps/s	6.8282 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_copy_nested[tensordict-compile]	0.4269ms	30.1911μs	33.1223 KOps/s	45.3466 KOps/s	$\textbf{\color{#d91a1a}-26.96\%}$
test_compile_copy_nested[tensordict-eager]	0.4003ms	26.3816μs	37.9052 KOps/s	36.9966 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_copy_nested[pytree-compile]	0.4650ms	63.7646μs	15.6827 KOps/s	15.1018 KOps/s	$\color{#35bf28}+3.85\%$
test_compile_copy_nested[pytree-eager]	0.4268ms	48.7621μs	20.5077 KOps/s	20.0086 KOps/s	$\color{#35bf28}+2.49\%$
test_compile_add_one_flat[tensordict-compile]	0.2688ms	0.1414ms	7.0711 KOps/s	7.1103 KOps/s	$\color{#d91a1a}-0.55\%$
test_compile_add_one_flat[tensordict-eager]	0.3273ms	0.2084ms	4.7981 KOps/s	4.8637 KOps/s	$\color{#d91a1a}-1.35\%$
test_compile_add_one_flat[tensorclass-compile]	0.2478ms	97.0268μs	10.3064 KOps/s	10.2796 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_add_one_flat[tensorclass-eager]	0.2013ms	51.7110μs	19.3383 KOps/s	19.6576 KOps/s	$\color{#d91a1a}-1.62\%$
test_compile_add_one_flat[pytree-compile]	0.2819ms	0.1370ms	7.2988 KOps/s	7.3246 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_add_one_flat[pytree-eager]	0.6359ms	0.4712ms	2.1223 KOps/s	2.1132 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_add_self_flat[tensordict-eager]	0.3735ms	0.2457ms	4.0706 KOps/s	4.0550 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_add_self_flat[tensordict-compile]	0.2880ms	0.1422ms	7.0316 KOps/s	6.9146 KOps/s	$\color{#35bf28}+1.69\%$
test_compile_add_self_flat[tensorclass-eager]	0.4434ms	61.6403μs	16.2232 KOps/s	16.1346 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_add_self_flat[tensorclass-compile]	0.2019ms	97.8338μs	10.2214 KOps/s	10.2249 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_add_self_flat[pytree-eager]	0.5588ms	0.4000ms	2.5000 KOps/s	2.4910 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_add_self_flat[pytree-compile]	0.2780ms	0.1347ms	7.4223 KOps/s	7.2198 KOps/s	$\color{#35bf28}+2.80\%$
test_compile_copy_flat[tensordict-compile]	0.1632ms	18.8756μs	52.9785 KOps/s	55.4662 KOps/s	$\color{#d91a1a}-4.49\%$
test_compile_copy_flat[tensordict-eager]	57.1910μs	27.2724μs	36.6671 KOps/s	37.7902 KOps/s	$\color{#d91a1a}-2.97\%$
test_compile_copy_flat[pytree-compile]	0.1797ms	69.5788μs	14.3722 KOps/s	14.4049 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_copy_flat[pytree-eager]	0.1124ms	51.3446μs	19.4763 KOps/s	19.4090 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_assign_and_add[tensordict-compile]	1.7067ms	0.4116ms	2.4297 KOps/s	2.2185 KOps/s	$\textbf{\color{#35bf28}+9.52\%}$
test_compile_assign_and_add[tensordict-eager]	2.7526ms	2.5626ms	390.2261 Ops/s	392.8697 Ops/s	$\color{#d91a1a}-0.67\%$
test_compile_assign_and_add[pytree-compile]	1.6252ms	0.3847ms	2.5991 KOps/s	2.2529 KOps/s	$\textbf{\color{#35bf28}+15.37\%}$
test_compile_assign_and_add[pytree-eager]	2.7874ms	2.5486ms	392.3788 Ops/s	391.9231 Ops/s	$\color{#35bf28}+0.12\%$
test_compile_indexing[tensor-tensordict-compile]	0.2600ms	0.1138ms	8.7894 KOps/s	8.7186 KOps/s	$\color{#35bf28}+0.81\%$
test_compile_indexing[tensor-tensordict-eager]	0.5876ms	79.8098μs	12.5298 KOps/s	13.0076 KOps/s	$\color{#d91a1a}-3.67\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2589ms	0.1068ms	9.3630 KOps/s	9.7017 KOps/s	$\color{#d91a1a}-3.49\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2205ms	66.1382μs	15.1199 KOps/s	14.9191 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_indexing[tensor-pytree-compile]	0.2611ms	0.1118ms	8.9440 KOps/s	9.6194 KOps/s	$\textbf{\color{#d91a1a}-7.02\%}$
test_compile_indexing[tensor-pytree-eager]	0.2506ms	68.7868μs	14.5377 KOps/s	15.3889 KOps/s	$\textbf{\color{#d91a1a}-5.53\%}$
test_compile_indexing[slice-tensordict-compile]	0.2489ms	0.1001ms	9.9924 KOps/s	9.8810 KOps/s	$\color{#35bf28}+1.13\%$
test_compile_indexing[slice-tensordict-eager]	0.1465ms	17.0031μs	58.8127 KOps/s	57.1031 KOps/s	$\color{#35bf28}+2.99\%$
test_compile_indexing[slice-tensorclass-compile]	0.2400ms	97.2944μs	10.2781 KOps/s	10.3403 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_indexing[slice-tensorclass-eager]	0.1360ms	15.8037μs	63.2762 KOps/s	62.8023 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_indexing[slice-pytree-compile]	0.2476ms	99.8039μs	10.0196 KOps/s	10.0661 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_indexing[slice-pytree-eager]	0.1414ms	15.9987μs	62.5051 KOps/s	62.7928 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_indexing[int-tensordict-compile]	0.2474ms	0.1009ms	9.9098 KOps/s	9.8508 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_indexing[int-tensordict-eager]	0.6198ms	17.2078μs	58.1131 KOps/s	57.9939 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_indexing[int-tensorclass-compile]	0.2417ms	99.2737μs	10.0732 KOps/s	10.3355 KOps/s	$\color{#d91a1a}-2.54\%$
test_compile_indexing[int-tensorclass-eager]	0.1363ms	16.0509μs	62.3018 KOps/s	63.3861 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_indexing[int-pytree-compile]	0.2484ms	96.2030μs	10.3947 KOps/s	10.2890 KOps/s	$\color{#35bf28}+1.03\%$
test_compile_indexing[int-pytree-eager]	0.1085ms	15.9416μs	62.7288 KOps/s	63.3290 KOps/s	$\color{#d91a1a}-0.95\%$
test_mod_add[eager]	0.1857ms	39.2377μs	25.4857 KOps/s	27.3591 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_mod_add[compile]	0.3010ms	80.5735μs	12.4110 KOps/s	12.1937 KOps/s	$\color{#35bf28}+1.78\%$
test_mod_add[compile-overhead]	0.3343ms	0.1678ms	5.9607 KOps/s	5.6498 KOps/s	$\textbf{\color{#35bf28}+5.50\%}$
test_mod_wrap[eager]	0.4267ms	0.2506ms	3.9909 KOps/s	4.1218 KOps/s	$\color{#d91a1a}-3.18\%$
test_mod_wrap[compile]	0.4253ms	0.2800ms	3.5714 KOps/s	3.4932 KOps/s	$\color{#35bf28}+2.24\%$
test_mod_wrap[compile-overhead]	7.7375ms	3.8278ms	261.2499 Ops/s	256.7591 Ops/s	$\color{#35bf28}+1.75\%$
test_mod_wrap_and_backward[eager]	1.5262ms	1.3199ms	757.6063 Ops/s	704.2318 Ops/s	$\textbf{\color{#35bf28}+7.58\%}$
test_mod_wrap_and_backward[compile]	1.4279ms	1.2573ms	795.3747 Ops/s	732.0674 Ops/s	$\textbf{\color{#35bf28}+8.65\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3705ms	0.9178ms	1.0896 KOps/s	907.2782 Ops/s	$\textbf{\color{#35bf28}+20.09\%}$
test_seq_add[eager]	0.2927ms	0.1091ms	9.1632 KOps/s	9.0728 KOps/s	$\color{#35bf28}+1.00\%$
test_seq_add[compile]	0.2810ms	88.5130μs	11.2978 KOps/s	11.2303 KOps/s	$\color{#35bf28}+0.60\%$
test_seq_add[compile-overhead]	0.2669ms	0.1287ms	7.7694 KOps/s	7.7449 KOps/s	$\color{#35bf28}+0.32\%$
test_seq_wrap[eager]	0.5531ms	0.4079ms	2.4516 KOps/s	2.4521 KOps/s	$\color{#d91a1a}-0.02\%$
test_seq_wrap[compile]	0.4538ms	0.2953ms	3.3862 KOps/s	3.2947 KOps/s	$\color{#35bf28}+2.78\%$
test_seq_wrap[compile-overhead]	0.3585ms	0.2230ms	4.4847 KOps/s	4.4220 KOps/s	$\color{#35bf28}+1.42\%$
test_func_call_runtime[False-eager]	0.9232ms	0.7610ms	1.3140 KOps/s	1.3833 KOps/s	$\textbf{\color{#d91a1a}-5.01\%}$
test_func_call_runtime[False-compile]	0.8365ms	0.7458ms	1.3408 KOps/s	1.3422 KOps/s	$\color{#d91a1a}-0.10\%$
test_func_call_runtime[False-compile-overhead]	0.4964ms	0.3613ms	2.7674 KOps/s	2.7489 KOps/s	$\color{#35bf28}+0.67\%$
test_func_call_runtime[True-eager]	1.0608ms	0.8765ms	1.1410 KOps/s	1.1318 KOps/s	$\color{#35bf28}+0.81\%$
test_func_call_runtime[True-compile]	0.9902ms	0.7709ms	1.2972 KOps/s	1.3007 KOps/s	$\color{#d91a1a}-0.27\%$
test_func_call_runtime[True-compile-overhead]	0.4977ms	0.3823ms	2.6157 KOps/s	2.6001 KOps/s	$\color{#35bf28}+0.60\%$
test_func_call_cm_runtime[False-eager]	0.8922ms	0.7141ms	1.4003 KOps/s	1.3954 KOps/s	$\color{#35bf28}+0.35\%$
test_func_call_cm_runtime[False-compile]	0.9237ms	0.7425ms	1.3468 KOps/s	1.3327 KOps/s	$\color{#35bf28}+1.06\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5425ms	0.3629ms	2.7555 KOps/s	2.7487 KOps/s	$\color{#35bf28}+0.25\%$
test_func_call_cm_runtime[True-eager]	1.1426ms	0.9672ms	1.0339 KOps/s	1.0147 KOps/s	$\color{#35bf28}+1.89\%$
test_func_call_cm_runtime[True-compile]	0.9299ms	0.7877ms	1.2695 KOps/s	1.2584 KOps/s	$\color{#35bf28}+0.89\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5407ms	0.4073ms	2.4550 KOps/s	2.4379 KOps/s	$\color{#35bf28}+0.70\%$
test_vmap_func_call_cm_runtime[eager]	2.5712ms	2.0195ms	495.1616 Ops/s	494.9090 Ops/s	$\color{#35bf28}+0.05\%$
test_vmap_func_call_cm_runtime[compile]	1.0440ms	0.8273ms	1.2088 KOps/s	1.2237 KOps/s	$\color{#d91a1a}-1.22\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5142ms	0.4103ms	2.4373 KOps/s	2.4179 KOps/s	$\color{#35bf28}+0.80\%$
test_distributed	13.3669ms	0.3154ms	3.1704 KOps/s	7.9908 KOps/s	$\textbf{\color{#d91a1a}-60.32\%}$
test_tdmodule	0.3543ms	19.9651μs	50.0874 KOps/s	52.2414 KOps/s	$\color{#d91a1a}-4.12\%$
test_tdmodule_dispatch	0.1718ms	35.8875μs	27.8649 KOps/s	29.2294 KOps/s	$\color{#d91a1a}-4.67\%$
test_tdseq	38.5200μs	19.6444μs	50.9052 KOps/s	53.3358 KOps/s	$\color{#d91a1a}-4.56\%$
test_tdseq_dispatch	59.3700μs	37.5964μs	26.5983 KOps/s	28.2779 KOps/s	$\textbf{\color{#d91a1a}-5.94\%}$
test_instantiation_functorch	1.7179ms	1.5240ms	656.1804 Ops/s	659.8173 Ops/s	$\color{#d91a1a}-0.55\%$
test_exec_functorch	0.2524ms	0.1401ms	7.1372 KOps/s	7.1206 KOps/s	$\color{#35bf28}+0.23\%$
test_exec_functional_call	0.1731ms	0.1299ms	7.7008 KOps/s	7.6145 KOps/s	$\color{#35bf28}+1.13\%$
test_exec_td_decorator	0.3612ms	0.1761ms	5.6785 KOps/s	5.6957 KOps/s	$\color{#d91a1a}-0.30\%$
test_vmap_mlp_speed_decorator[True-True]	0.8294ms	0.6620ms	1.5106 KOps/s	1.5130 KOps/s	$\color{#d91a1a}-0.16\%$
test_vmap_mlp_speed_decorator[True-False]	0.8193ms	0.6646ms	1.5048 KOps/s	1.5122 KOps/s	$\color{#d91a1a}-0.49\%$
test_vmap_mlp_speed_decorator[False-True]	0.7665ms	0.5800ms	1.7240 KOps/s	1.7148 KOps/s	$\color{#35bf28}+0.54\%$
test_vmap_mlp_speed_decorator[False-False]	0.7316ms	0.5813ms	1.7203 KOps/s	1.6951 KOps/s	$\color{#35bf28}+1.49\%$
test_vmap_transformer_speed_decorator[True-True]	19.5587ms	18.5859ms	53.8043 Ops/s	54.0132 Ops/s	$\color{#d91a1a}-0.39\%$
test_vmap_transformer_speed_decorator[True-False]	19.5737ms	18.6596ms	53.5916 Ops/s	53.8463 Ops/s	$\color{#d91a1a}-0.47\%$
test_vmap_transformer_speed_decorator[False-True]	19.5311ms	18.5351ms	53.9516 Ops/s	54.4415 Ops/s	$\color{#d91a1a}-0.90\%$
test_vmap_transformer_speed_decorator[False-False]	19.0556ms	18.4689ms	54.1449 Ops/s	54.2484 Ops/s	$\color{#d91a1a}-0.19\%$
test_to_module_speed[True]	1.0327ms	0.9328ms	1.0720 KOps/s	1.0580 KOps/s	$\color{#35bf28}+1.32\%$
test_to_module_speed[False]	1.3742ms	0.9145ms	1.0935 KOps/s	1.0691 KOps/s	$\color{#35bf28}+2.28\%$
test_tc_init	69.7800μs	37.6344μs	26.5715 KOps/s	28.4537 KOps/s	$\textbf{\color{#d91a1a}-6.62\%}$
test_tc_init_nested	0.1149ms	75.5784μs	13.2313 KOps/s	14.2291 KOps/s	$\textbf{\color{#d91a1a}-7.01\%}$
test_tc_first_layer_tensor	4.9416μs	0.7048μs	1.4187 MOps/s	1.4269 MOps/s	$\color{#d91a1a}-0.57\%$
test_tc_first_layer_nontensor	23.6010μs	2.3241μs	430.2678 KOps/s	430.0649 KOps/s	$\color{#35bf28}+0.05\%$
test_tc_second_layer_tensor	8.8667μs	1.4480μs	690.5983 KOps/s	705.3242 KOps/s	$\color{#d91a1a}-2.09\%$
test_tc_second_layer_nontensor	25.0500μs	3.0806μs	324.6125 KOps/s	324.1851 KOps/s	$\color{#35bf28}+0.13\%$
test_unbind	0.2204s	9.7803ms	102.2468 Ops/s	151.7758 Ops/s	$\textbf{\color{#d91a1a}-32.63\%}$
test_full_like	9.8318ms	9.3423ms	107.0402 Ops/s	102.8481 Ops/s	$\color{#35bf28}+4.08\%$
test_zeros_like	4.8869ms	4.3378ms	230.5328 Ops/s	230.2125 Ops/s	$\color{#35bf28}+0.14\%$
test_ones_like	9.2663ms	7.2400ms	138.1213 Ops/s	229.4501 Ops/s	$\textbf{\color{#d91a1a}-39.80\%}$
test_clone	6.9183ms	6.5696ms	152.2172 Ops/s	151.8816 Ops/s	$\color{#35bf28}+0.22\%$
test_squeeze	62.0000μs	9.4527μs	105.7898 KOps/s	109.5806 KOps/s	$\color{#d91a1a}-3.46\%$
test_unsqueeze	0.2070ms	69.8785μs	14.3106 KOps/s	14.4039 KOps/s	$\color{#d91a1a}-0.65\%$
test_split	0.3847ms	0.1589ms	6.2951 KOps/s	6.2269 KOps/s	$\color{#35bf28}+1.10\%$
test_permute	0.2687ms	0.1732ms	5.7732 KOps/s	5.4855 KOps/s	$\textbf{\color{#35bf28}+5.25\%}$
test_stack	51.5032ms	51.0561ms	19.5863 Ops/s	19.5258 Ops/s	$\color{#35bf28}+0.31\%$
test_cat	51.8452ms	51.1474ms	19.5513 Ops/s	19.6034 Ops/s	$\color{#d91a1a}-0.27\%$