[BugFix] Consolidate lazy stacks of non-tensors

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Consolidate lazy stacks of non-tensors #1222
• [Feature] capture_non_tensor_stack #1221
• [Feature] Allow update to increment the number of tds in a lazy stack #1220

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}23$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.8140μs	21.0699μs	47.4610 KOps/s	49.0189 KOps/s	$\color{#d91a1a}-3.18\%$
test_plain_set_stack_nested	41.3780μs	20.9454μs	47.7431 KOps/s	48.2252 KOps/s	$\color{#d91a1a}-1.00\%$
test_plain_set_nested_inplace	52.3380μs	22.8619μs	43.7408 KOps/s	44.6930 KOps/s	$\color{#d91a1a}-2.13\%$
test_plain_set_stack_nested_inplace	93.8690μs	22.8275μs	43.8068 KOps/s	44.7768 KOps/s	$\color{#d91a1a}-2.17\%$
test_items	0.1028ms	4.2246μs	236.7081 KOps/s	240.6147 KOps/s	$\color{#d91a1a}-1.62\%$
test_items_nested	0.5880ms	0.4042ms	2.4739 KOps/s	2.4419 KOps/s	$\color{#35bf28}+1.31\%$
test_items_nested_locked	0.5196ms	0.4039ms	2.4761 KOps/s	2.4268 KOps/s	$\color{#35bf28}+2.03\%$
test_items_nested_leaf	0.1383ms	76.7135μs	13.0355 KOps/s	12.8818 KOps/s	$\color{#35bf28}+1.19\%$
test_items_stack_nested	0.5989ms	0.4121ms	2.4264 KOps/s	2.4322 KOps/s	$\color{#d91a1a}-0.24\%$
test_items_stack_nested_leaf	0.1327ms	77.6983μs	12.8703 KOps/s	12.4785 KOps/s	$\color{#35bf28}+3.14\%$
test_items_stack_nested_locked	0.5776ms	0.4070ms	2.4569 KOps/s	2.4194 KOps/s	$\color{#35bf28}+1.55\%$
test_keys	28.6130μs	3.4566μs	289.3011 KOps/s	289.4759 KOps/s	$\color{#d91a1a}-0.06\%$
test_keys_nested	0.2684ms	0.1649ms	6.0648 KOps/s	6.0884 KOps/s	$\color{#d91a1a}-0.39\%$
test_keys_nested_locked	0.7396ms	0.1713ms	5.8381 KOps/s	5.8541 KOps/s	$\color{#d91a1a}-0.27\%$
test_keys_nested_leaf	0.2294ms	0.1433ms	6.9794 KOps/s	6.9652 KOps/s	$\color{#35bf28}+0.20\%$
test_keys_stack_nested	0.3040ms	0.1659ms	6.0290 KOps/s	6.0627 KOps/s	$\color{#d91a1a}-0.56\%$
test_keys_stack_nested_leaf	0.2597ms	0.1444ms	6.9248 KOps/s	7.1811 KOps/s	$\color{#d91a1a}-3.57\%$
test_keys_stack_nested_locked	0.2700ms	0.1715ms	5.8304 KOps/s	5.9465 KOps/s	$\color{#d91a1a}-1.95\%$
test_values	5.5846μs	1.0325μs	968.5399 KOps/s	969.5777 KOps/s	$\color{#d91a1a}-0.11\%$
test_values_nested	0.1054ms	63.5790μs	15.7285 KOps/s	16.1620 KOps/s	$\color{#d91a1a}-2.68\%$
test_values_nested_locked	0.1111ms	63.9656μs	15.6334 KOps/s	16.2223 KOps/s	$\color{#d91a1a}-3.63\%$
test_values_nested_leaf	0.1217ms	71.6702μs	13.9528 KOps/s	13.8582 KOps/s	$\color{#35bf28}+0.68\%$
test_values_stack_nested	0.1114ms	64.7863μs	15.4354 KOps/s	14.8523 KOps/s	$\color{#35bf28}+3.93\%$
test_values_stack_nested_leaf	0.1258ms	72.2738μs	13.8363 KOps/s	14.2342 KOps/s	$\color{#d91a1a}-2.80\%$
test_values_stack_nested_locked	0.1103ms	64.6482μs	15.4683 KOps/s	15.7605 KOps/s	$\color{#d91a1a}-1.85\%$
test_membership	3.5251μs	0.7165μs	1.3957 MOps/s	1.1459 MOps/s	$\textbf{\color{#35bf28}+21.80\%}$
test_membership_nested	43.0900μs	2.9470μs	339.3234 KOps/s	340.5480 KOps/s	$\color{#d91a1a}-0.36\%$
test_membership_nested_leaf	30.6680μs	2.9153μs	343.0187 KOps/s	344.6672 KOps/s	$\color{#d91a1a}-0.48\%$
test_membership_stacked_nested	31.6190μs	2.8985μs	345.0027 KOps/s	344.5362 KOps/s	$\color{#35bf28}+0.14\%$
test_membership_stacked_nested_leaf	39.6550μs	2.9411μs	340.0124 KOps/s	340.1864 KOps/s	$\color{#d91a1a}-0.05\%$
test_membership_nested_last	29.5960μs	4.3264μs	231.1388 KOps/s	228.2099 KOps/s	$\color{#35bf28}+1.28\%$
test_membership_nested_leaf_last	35.0460μs	4.3336μs	230.7552 KOps/s	219.9783 KOps/s	$\color{#35bf28}+4.90\%$
test_membership_stacked_nested_last	0.1276ms	4.4446μs	224.9913 KOps/s	74.3158 KOps/s	$\textbf{\color{#35bf28}+202.75\%}$
test_membership_stacked_nested_leaf_last	89.7780μs	4.3622μs	229.2423 KOps/s	74.1329 KOps/s	$\textbf{\color{#35bf28}+209.23\%}$
test_nested_getleaf	38.0210μs	10.8786μs	91.9232 KOps/s	92.7874 KOps/s	$\color{#d91a1a}-0.93\%$
test_nested_get	33.6240μs	10.4930μs	95.3020 KOps/s	98.5251 KOps/s	$\color{#d91a1a}-3.27\%$
test_stacked_getleaf	26.7600μs	10.6496μs	93.9002 KOps/s	94.2900 KOps/s	$\color{#d91a1a}-0.41\%$
test_stacked_get	37.1590μs	10.3617μs	96.5094 KOps/s	98.1857 KOps/s	$\color{#d91a1a}-1.71\%$
test_nested_getitemleaf	38.2420μs	11.5409μs	86.6482 KOps/s	87.4805 KOps/s	$\color{#d91a1a}-0.95\%$
test_nested_getitem	41.1270μs	10.9231μs	91.5487 KOps/s	94.8482 KOps/s	$\color{#d91a1a}-3.48\%$
test_stacked_getitemleaf	43.0810μs	11.3269μs	88.2857 KOps/s	89.8069 KOps/s	$\color{#d91a1a}-1.69\%$
test_stacked_getitem	41.8680μs	11.0153μs	90.7828 KOps/s	93.6309 KOps/s	$\color{#d91a1a}-3.04\%$
test_lock_nested	7.2654ms	0.4138ms	2.4164 KOps/s	2.4496 KOps/s	$\color{#d91a1a}-1.36\%$
test_lock_stack_nested	0.8251ms	0.4180ms	2.3921 KOps/s	2.4366 KOps/s	$\color{#d91a1a}-1.82\%$
test_unlock_nested	0.4130ms	0.3300ms	3.0305 KOps/s	2.9932 KOps/s	$\color{#35bf28}+1.24\%$
test_unlock_stack_nested	0.4138ms	0.3361ms	2.9756 KOps/s	3.0138 KOps/s	$\color{#d91a1a}-1.27\%$
test_flatten_speed	0.1794ms	99.6279μs	10.0374 KOps/s	9.4253 KOps/s	$\textbf{\color{#35bf28}+6.49\%}$
test_unflatten_speed	1.0881ms	0.5308ms	1.8840 KOps/s	1.8697 KOps/s	$\color{#35bf28}+0.76\%$
test_common_ops	6.9947ms	0.8422ms	1.1874 KOps/s	1.2661 KOps/s	$\textbf{\color{#d91a1a}-6.22\%}$
test_creation	41.4780μs	2.5136μs	397.8409 KOps/s	397.1191 KOps/s	$\color{#35bf28}+0.18\%$
test_creation_empty	37.1800μs	12.5786μs	79.5001 KOps/s	87.8107 KOps/s	$\textbf{\color{#d91a1a}-9.46\%}$
test_creation_nested_1	52.4580μs	15.6082μs	64.0690 KOps/s	70.1971 KOps/s	$\textbf{\color{#d91a1a}-8.73\%}$
test_creation_nested_2	53.3390μs	20.3457μs	49.1505 KOps/s	53.8173 KOps/s	$\textbf{\color{#d91a1a}-8.67\%}$
test_clone	63.8290μs	13.4902μs	74.1279 KOps/s	73.4214 KOps/s	$\color{#35bf28}+0.96\%$
test_getitem[int]	0.8942ms	12.3347μs	81.0723 KOps/s	78.7365 KOps/s	$\color{#35bf28}+2.97\%$
test_getitem[slice_int]	0.1541ms	24.6210μs	40.6158 KOps/s	41.2602 KOps/s	$\color{#d91a1a}-1.56\%$
test_getitem[range]	0.1625ms	49.7312μs	20.1081 KOps/s	19.9930 KOps/s	$\color{#35bf28}+0.58\%$
test_getitem[tuple]	0.1371ms	19.8102μs	50.4790 KOps/s	49.9747 KOps/s	$\color{#35bf28}+1.01\%$
test_getitem[list]	0.2965ms	45.6687μs	21.8968 KOps/s	21.8896 KOps/s	$\color{#35bf28}+0.03\%$
test_setitem_dim[int]	56.7370μs	25.9076μs	38.5987 KOps/s	38.9021 KOps/s	$\color{#d91a1a}-0.78\%$
test_setitem_dim[slice_int]	88.4760μs	52.3291μs	19.1098 KOps/s	19.2256 KOps/s	$\color{#d91a1a}-0.60\%$
test_setitem_dim[range]	0.1312ms	75.5177μs	13.2419 KOps/s	12.8777 KOps/s	$\color{#35bf28}+2.83\%$
test_setitem_dim[tuple]	95.7290μs	42.4575μs	23.5530 KOps/s	24.8900 KOps/s	$\textbf{\color{#d91a1a}-5.37\%}$
test_setitem	86.7630μs	21.1030μs	47.3865 KOps/s	48.7144 KOps/s	$\color{#d91a1a}-2.73\%$
test_set	0.2593ms	20.6274μs	48.4793 KOps/s	50.3757 KOps/s	$\color{#d91a1a}-3.76\%$
test_set_shared	6.3092ms	0.1851ms	5.4037 KOps/s	5.4576 KOps/s	$\color{#d91a1a}-0.99\%$
test_update	0.1446ms	23.9832μs	41.6959 KOps/s	44.4226 KOps/s	$\textbf{\color{#d91a1a}-6.14\%}$
test_update_nested	0.2109ms	36.2747μs	27.5674 KOps/s	29.5031 KOps/s	$\textbf{\color{#d91a1a}-6.56\%}$
test_update__nested	0.2885ms	33.8627μs	29.5310 KOps/s	29.1522 KOps/s	$\color{#35bf28}+1.30\%$
test_set_nested	0.1289ms	22.8217μs	43.8180 KOps/s	44.2889 KOps/s	$\color{#d91a1a}-1.06\%$
test_set_nested_new	83.3370μs	28.1083μs	35.5767 KOps/s	36.4146 KOps/s	$\color{#d91a1a}-2.30\%$
test_select	0.1017ms	44.3833μs	22.5310 KOps/s	23.3325 KOps/s	$\color{#d91a1a}-3.43\%$
test_select_nested	0.1543ms	63.1741μs	15.8293 KOps/s	15.7809 KOps/s	$\color{#35bf28}+0.31\%$
test_exclude_nested	0.1787ms	81.8260μs	12.2210 KOps/s	12.2071 KOps/s	$\color{#35bf28}+0.11\%$
test_empty[True]	0.7378ms	0.4076ms	2.4531 KOps/s	2.3884 KOps/s	$\color{#35bf28}+2.71\%$
test_empty[False]	6.7450μs	1.3348μs	749.2023 KOps/s	726.9274 KOps/s	$\color{#35bf28}+3.06\%$
test_unbind_speed	0.5793ms	0.2674ms	3.7393 KOps/s	3.6991 KOps/s	$\color{#35bf28}+1.09\%$
test_unbind_speed_stack0	0.3505ms	0.2644ms	3.7827 KOps/s	3.7965 KOps/s	$\color{#d91a1a}-0.36\%$
test_unbind_speed_stack1	0.1059s	0.7294ms	1.3710 KOps/s	1.1116 KOps/s	$\textbf{\color{#35bf28}+23.34\%}$
test_split	0.1028s	1.7364ms	575.9178 Ops/s	564.5762 Ops/s	$\color{#35bf28}+2.01\%$
test_chunk	0.1030s	1.7602ms	568.1126 Ops/s	625.5824 Ops/s	$\textbf{\color{#d91a1a}-9.19\%}$
test_consolidate_njt[False-None]	9.4583ms	8.3331ms	120.0035 Ops/s	108.4618 Ops/s	$\textbf{\color{#35bf28}+10.64\%}$
test_creation[device0]	0.2257ms	91.2784μs	10.9555 KOps/s	10.8791 KOps/s	$\color{#35bf28}+0.70\%$
test_creation_from_tensor	0.2354ms	95.3409μs	10.4887 KOps/s	10.2197 KOps/s	$\color{#35bf28}+2.63\%$
test_add_one[memmap_tensor0]	0.1546ms	5.0884μs	196.5260 KOps/s	210.0658 KOps/s	$\textbf{\color{#d91a1a}-6.45\%}$
test_contiguous[memmap_tensor0]	10.6090μs	0.5192μs	1.9259 MOps/s	1.9824 MOps/s	$\color{#d91a1a}-2.85\%$
test_stack[memmap_tensor0]	19.3560μs	3.4829μs	287.1169 KOps/s	291.2094 KOps/s	$\color{#d91a1a}-1.41\%$
test_memmaptd_index	0.5425ms	0.2340ms	4.2739 KOps/s	4.3027 KOps/s	$\color{#d91a1a}-0.67\%$
test_memmaptd_index_astensor	0.6501ms	0.3181ms	3.1433 KOps/s	3.0880 KOps/s	$\color{#35bf28}+1.79\%$
test_memmaptd_index_op	0.7904ms	0.5980ms	1.6721 KOps/s	1.7530 KOps/s	$\color{#d91a1a}-4.61\%$
test_serialize_model	0.2162s	0.1302s	7.6834 Ops/s	8.3535 Ops/s	$\textbf{\color{#d91a1a}-8.02\%}$
test_serialize_model_pickle	0.4483s	0.3953s	2.5298 Ops/s	2.5202 Ops/s	$\color{#35bf28}+0.38\%$
test_serialize_weights	0.1226s	0.1164s	8.5938 Ops/s	8.7593 Ops/s	$\color{#d91a1a}-1.89\%$
test_serialize_weights_returnearly	0.1881s	0.1650s	6.0592 Ops/s	5.5275 Ops/s	$\textbf{\color{#35bf28}+9.62\%}$
test_serialize_weights_pickle	1.1443s	0.6964s	1.4359 Ops/s	2.5664 Ops/s	$\textbf{\color{#d91a1a}-44.05\%}$
test_serialize_weights_filesystem	0.2470s	0.1589s	6.2920 Ops/s	6.9201 Ops/s	$\textbf{\color{#d91a1a}-9.08\%}$
test_serialize_model_filesystem	0.1512s	0.1433s	6.9786 Ops/s	6.5968 Ops/s	$\textbf{\color{#35bf28}+5.79\%}$
test_reshape_pytree	73.2770μs	26.6907μs	37.4662 KOps/s	37.5043 KOps/s	$\color{#d91a1a}-0.10\%$
test_reshape_td	69.0090μs	32.9719μs	30.3288 KOps/s	29.6238 KOps/s	$\color{#35bf28}+2.38\%$
test_view_pytree	0.1866ms	26.7905μs	37.3266 KOps/s	37.1970 KOps/s	$\color{#35bf28}+0.35\%$
test_view_td	90.1690μs	40.3229μs	24.7998 KOps/s	24.5840 KOps/s	$\color{#35bf28}+0.88\%$
test_unbind_pytree	63.0080μs	29.3664μs	34.0525 KOps/s	33.3044 KOps/s	$\color{#35bf28}+2.25\%$
test_unbind_td	0.3642ms	40.7499μs	24.5400 KOps/s	25.1773 KOps/s	$\color{#d91a1a}-2.53\%$
test_split_pytree	63.1480μs	29.3032μs	34.1260 KOps/s	34.2084 KOps/s	$\color{#d91a1a}-0.24\%$
test_split_td	0.5243ms	44.8969μs	22.2733 KOps/s	21.9722 KOps/s	$\color{#35bf28}+1.37\%$
test_add_pytree	84.1270μs	35.5820μs	28.1041 KOps/s	27.9588 KOps/s	$\color{#35bf28}+0.52\%$
test_add_td	0.1292ms	62.1528μs	16.0894 KOps/s	17.2033 KOps/s	$\textbf{\color{#d91a1a}-6.47\%}$
test_compile_add_one_nested[tensordict-compile]	0.1420ms	67.3318μs	14.8518 KOps/s	14.6892 KOps/s	$\color{#35bf28}+1.11\%$
test_compile_add_one_nested[tensordict-eager]	0.7703ms	0.1731ms	5.7784 KOps/s	5.7561 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_add_one_nested[pytree-compile]	0.1100ms	44.9765μs	22.2338 KOps/s	21.7205 KOps/s	$\color{#35bf28}+2.36\%$
test_compile_add_one_nested[pytree-eager]	0.2591ms	0.1196ms	8.3585 KOps/s	8.3591 KOps/s	$-0.01\%$
test_compile_copy_nested[tensordict-compile]	78.4670μs	28.8443μs	34.6689 KOps/s	35.3731 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_copy_nested[tensordict-eager]	0.1217ms	59.0911μs	16.9230 KOps/s	17.0069 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_copy_nested[pytree-compile]	0.1492ms	80.0687μs	12.4893 KOps/s	12.5519 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_copy_nested[pytree-eager]	0.1208ms	66.8261μs	14.9642 KOps/s	14.7707 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_add_one_flat[tensordict-compile]	0.1992ms	0.1067ms	9.3685 KOps/s	9.3417 KOps/s	$\color{#35bf28}+0.29\%$
test_compile_add_one_flat[tensordict-eager]	0.4276ms	0.2160ms	4.6299 KOps/s	4.5322 KOps/s	$\color{#35bf28}+2.16\%$
test_compile_add_one_flat[tensorclass-compile]	0.2038ms	46.8527μs	21.3435 KOps/s	21.0383 KOps/s	$\color{#35bf28}+1.45\%$
test_compile_add_one_flat[tensorclass-eager]	0.2009ms	66.9186μs	14.9435 KOps/s	14.7576 KOps/s	$\color{#35bf28}+1.26\%$
test_compile_add_one_flat[pytree-compile]	0.1949ms	0.1014ms	9.8657 KOps/s	9.9463 KOps/s	$\color{#d91a1a}-0.81\%$
test_compile_add_one_flat[pytree-eager]	0.2875ms	0.2016ms	4.9592 KOps/s	4.9117 KOps/s	$\color{#35bf28}+0.97\%$
test_compile_add_self_flat[tensordict-eager]	0.4715ms	0.2349ms	4.2569 KOps/s	4.2971 KOps/s	$\color{#d91a1a}-0.94\%$
test_compile_add_self_flat[tensordict-compile]	0.2250ms	0.1107ms	9.0320 KOps/s	9.1829 KOps/s	$\color{#d91a1a}-1.64\%$
test_compile_add_self_flat[tensorclass-eager]	0.3214ms	64.0077μs	15.6231 KOps/s	15.6076 KOps/s	$\color{#35bf28}+0.10\%$
test_compile_add_self_flat[tensorclass-compile]	0.2054ms	48.9510μs	20.4286 KOps/s	20.1912 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_add_self_flat[pytree-eager]	0.3321ms	0.1581ms	6.3256 KOps/s	6.3079 KOps/s	$\color{#35bf28}+0.28\%$
test_compile_add_self_flat[pytree-compile]	0.2018ms	0.1003ms	9.9691 KOps/s	9.8557 KOps/s	$\color{#35bf28}+1.15\%$
test_compile_copy_flat[tensordict-compile]	52.1480μs	21.4025μs	46.7235 KOps/s	46.3006 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_copy_flat[tensordict-eager]	0.1459ms	69.2793μs	14.4343 KOps/s	14.7805 KOps/s	$\color{#d91a1a}-2.34\%$
test_compile_copy_flat[pytree-compile]	0.1936ms	81.7199μs	12.2369 KOps/s	11.9306 KOps/s	$\color{#35bf28}+2.57\%$
test_compile_copy_flat[pytree-eager]	0.1422ms	68.2067μs	14.6613 KOps/s	14.5296 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_assign_and_add[tensordict-compile]	0.3154ms	0.2161ms	4.6285 KOps/s	4.5294 KOps/s	$\color{#35bf28}+2.19\%$
test_compile_assign_and_add[tensordict-eager]	2.7708ms	1.3654ms	732.3805 Ops/s	716.4002 Ops/s	$\color{#35bf28}+2.23\%$
test_compile_assign_and_add[pytree-compile]	0.3084ms	0.2092ms	4.7793 KOps/s	4.6667 KOps/s	$\color{#35bf28}+2.41\%$
test_compile_assign_and_add[pytree-eager]	1.4820ms	0.8329ms	1.2006 KOps/s	1.1885 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_assign_and_add_stack[compile]	0.5804ms	0.4618ms	2.1655 KOps/s	2.1614 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_assign_and_add_stack[eager]	3.5238ms	2.7934ms	357.9804 Ops/s	375.2794 Ops/s	$\color{#d91a1a}-4.61\%$
test_compile_indexing[tensor-tensordict-compile]	0.1878ms	39.9192μs	25.0506 KOps/s	25.2424 KOps/s	$\color{#d91a1a}-0.76\%$
test_compile_indexing[tensor-tensordict-eager]	0.6247ms	32.8196μs	30.4696 KOps/s	28.9821 KOps/s	$\textbf{\color{#35bf28}+5.13\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.1766ms	30.8135μs	32.4534 KOps/s	31.5156 KOps/s	$\color{#35bf28}+2.98\%$
test_compile_indexing[tensor-tensorclass-eager]	89.8480μs	22.9968μs	43.4844 KOps/s	43.1157 KOps/s	$\color{#35bf28}+0.86\%$
test_compile_indexing[tensor-pytree-compile]	87.9240μs	31.5412μs	31.7046 KOps/s	30.3306 KOps/s	$\color{#35bf28}+4.53\%$
test_compile_indexing[tensor-pytree-eager]	79.7510μs	23.0607μs	43.3639 KOps/s	43.4152 KOps/s	$\color{#d91a1a}-0.12\%$
test_compile_indexing[slice-tensordict-compile]	0.1289ms	52.9340μs	18.8914 KOps/s	18.4015 KOps/s	$\color{#35bf28}+2.66\%$
test_compile_indexing[slice-tensordict-eager]	0.3172ms	20.2229μs	49.4490 KOps/s	49.1425 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_indexing[slice-tensorclass-compile]	0.2588ms	45.2577μs	22.0957 KOps/s	20.6437 KOps/s	$\textbf{\color{#35bf28}+7.03\%}$
test_compile_indexing[slice-tensorclass-eager]	80.3000μs	18.8654μs	53.0071 KOps/s	53.4603 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_indexing[slice-pytree-compile]	93.8360μs	46.0953μs	21.6942 KOps/s	20.2595 KOps/s	$\textbf{\color{#35bf28}+7.08\%}$
test_compile_indexing[slice-pytree-eager]	76.3030μs	19.2307μs	52.0003 KOps/s	53.0337 KOps/s	$\color{#d91a1a}-1.95\%$
test_compile_indexing[int-tensordict-compile]	0.1350ms	54.1959μs	18.4516 KOps/s	17.7923 KOps/s	$\color{#35bf28}+3.71\%$
test_compile_indexing[int-tensordict-eager]	0.9307ms	19.9554μs	50.1117 KOps/s	49.3230 KOps/s	$\color{#35bf28}+1.60\%$
test_compile_indexing[int-tensorclass-compile]	0.1328ms	45.7021μs	21.8808 KOps/s	20.0593 KOps/s	$\textbf{\color{#35bf28}+9.08\%}$
test_compile_indexing[int-tensorclass-eager]	81.7430μs	19.0197μs	52.5770 KOps/s	52.9553 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_indexing[int-pytree-compile]	97.1120μs	46.4538μs	21.5268 KOps/s	20.1124 KOps/s	$\textbf{\color{#35bf28}+7.03\%}$
test_compile_indexing[int-pytree-eager]	65.0820μs	19.0369μs	52.5296 KOps/s	53.6067 KOps/s	$\color{#d91a1a}-2.01\%$
test_mod_add[eager]	84.4380μs	36.7230μs	27.2309 KOps/s	28.2854 KOps/s	$\color{#d91a1a}-3.73\%$
test_mod_add[compile]	0.1465ms	66.5918μs	15.0169 KOps/s	14.4393 KOps/s	$\color{#35bf28}+4.00\%$
test_mod_add[compile-overhead]	0.1278ms	66.1579μs	15.1154 KOps/s	14.7728 KOps/s	$\color{#35bf28}+2.32\%$
test_mod_wrap[eager]	0.4487ms	0.2270ms	4.4049 KOps/s	4.4530 KOps/s	$\color{#d91a1a}-1.08\%$
test_mod_wrap[compile]	1.9457ms	0.2350ms	4.2552 KOps/s	4.3002 KOps/s	$\color{#d91a1a}-1.05\%$
test_mod_wrap[compile-overhead]	0.3401ms	0.2347ms	4.2604 KOps/s	4.3384 KOps/s	$\color{#d91a1a}-1.80\%$
test_mod_wrap_and_backward[eager]	16.6994ms	13.5685ms	73.7002 Ops/s	77.7724 Ops/s	$\textbf{\color{#d91a1a}-5.24\%}$
test_mod_wrap_and_backward[compile]	15.7046ms	12.0448ms	83.0235 Ops/s	87.8823 Ops/s	$\textbf{\color{#d91a1a}-5.53\%}$
test_mod_wrap_and_backward[compile-overhead]	15.0462ms	11.8179ms	84.6176 Ops/s	88.6191 Ops/s	$\color{#d91a1a}-4.52\%$
test_seq_add[eager]	0.1935ms	0.1188ms	8.4178 KOps/s	8.4009 KOps/s	$\color{#35bf28}+0.20\%$
test_seq_add[compile]	0.1534ms	78.5051μs	12.7380 KOps/s	12.8595 KOps/s	$\color{#d91a1a}-0.94\%$
test_seq_add[compile-overhead]	0.1568ms	75.2982μs	13.2805 KOps/s	12.9165 KOps/s	$\color{#35bf28}+2.82\%$
test_seq_wrap[eager]	0.6306ms	0.4546ms	2.1997 KOps/s	2.2418 KOps/s	$\color{#d91a1a}-1.88\%$
test_seq_wrap[compile]	0.4392ms	0.2477ms	4.0374 KOps/s	4.0999 KOps/s	$\color{#d91a1a}-1.52\%$
test_seq_wrap[compile-overhead]	0.3753ms	0.2479ms	4.0338 KOps/s	4.0903 KOps/s	$\color{#d91a1a}-1.38\%$
test_func_call_runtime[False-eager]	0.8365ms	0.5465ms	1.8297 KOps/s	1.8258 KOps/s	$\color{#35bf28}+0.21\%$
test_func_call_runtime[False-compile]	0.6137ms	0.4509ms	2.2177 KOps/s	2.2348 KOps/s	$\color{#d91a1a}-0.76\%$
test_func_call_runtime[False-compile-overhead]	0.6161ms	0.4496ms	2.2242 KOps/s	2.2536 KOps/s	$\color{#d91a1a}-1.31\%$
test_func_call_runtime[True-eager]	0.9667ms	0.7649ms	1.3073 KOps/s	1.3186 KOps/s	$\color{#d91a1a}-0.85\%$
test_func_call_runtime[True-compile]	0.7431ms	0.4702ms	2.1269 KOps/s	2.1499 KOps/s	$\color{#d91a1a}-1.07\%$
test_func_call_runtime[True-compile-overhead]	0.6723ms	0.4720ms	2.1185 KOps/s	2.1534 KOps/s	$\color{#d91a1a}-1.62\%$
test_func_call_cm_runtime[False-eager]	0.8785ms	0.5428ms	1.8423 KOps/s	1.8502 KOps/s	$\color{#d91a1a}-0.42\%$
test_func_call_cm_runtime[False-compile]	0.8419ms	0.4521ms	2.2121 KOps/s	2.2500 KOps/s	$\color{#d91a1a}-1.68\%$
test_func_call_cm_runtime[False-compile-overhead]	0.8387ms	0.4518ms	2.2135 KOps/s	2.2349 KOps/s	$\color{#d91a1a}-0.96\%$
test_func_call_cm_runtime[True-eager]	1.0786ms	0.9101ms	1.0988 KOps/s	1.1017 KOps/s	$\color{#d91a1a}-0.27\%$
test_func_call_cm_runtime[True-compile]	0.9687ms	0.8034ms	1.2446 KOps/s	1.2354 KOps/s	$\color{#35bf28}+0.75\%$
test_func_call_cm_runtime[True-compile-overhead]	1.4338ms	0.8207ms	1.2184 KOps/s	1.2180 KOps/s	$\color{#35bf28}+0.03\%$
test_vmap_func_call_cm_runtime[eager]	4.6775ms	2.1474ms	465.6829 Ops/s	523.6769 Ops/s	$\textbf{\color{#d91a1a}-11.07\%}$
test_vmap_func_call_cm_runtime[compile]	0.9376ms	0.5505ms	1.8166 KOps/s	1.8217 KOps/s	$\color{#d91a1a}-0.28\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6522ms	0.5422ms	1.8442 KOps/s	1.8479 KOps/s	$\color{#d91a1a}-0.20\%$
test_distributed	0.2462ms	0.1275ms	7.8453 KOps/s	7.9587 KOps/s	$\color{#d91a1a}-1.42\%$
test_tdmodule	73.4780μs	27.6771μs	36.1310 KOps/s	37.6581 KOps/s	$\color{#d91a1a}-4.06\%$
test_tdmodule_dispatch	0.3012ms	61.6748μs	16.2141 KOps/s	21.0470 KOps/s	$\textbf{\color{#d91a1a}-22.96\%}$
test_tdseq	48.2200μs	29.4612μs	33.9430 KOps/s	34.3135 KOps/s	$\color{#d91a1a}-1.08\%$
test_tdseq_dispatch	0.1183ms	56.2681μs	17.7721 KOps/s	18.8336 KOps/s	$\textbf{\color{#d91a1a}-5.64\%}$
test_instantiation_functorch	2.2217ms	1.5719ms	636.1615 Ops/s	639.3416 Ops/s	$\color{#d91a1a}-0.50\%$
test_exec_functorch	0.3447ms	0.1779ms	5.6219 KOps/s	5.4789 KOps/s	$\color{#35bf28}+2.61\%$
test_exec_functional_call	0.3268ms	0.1720ms	5.8142 KOps/s	5.7713 KOps/s	$\color{#35bf28}+0.74\%$
test_exec_td_decorator	0.5130ms	0.2376ms	4.2085 KOps/s	4.2162 KOps/s	$\color{#d91a1a}-0.18\%$
test_vmap_mlp_speed_decorator[True-True]	0.8623ms	0.6829ms	1.4644 KOps/s	1.5518 KOps/s	$\textbf{\color{#d91a1a}-5.63\%}$
test_vmap_mlp_speed_decorator[True-False]	1.0170ms	0.6847ms	1.4604 KOps/s	1.5360 KOps/s	$\color{#d91a1a}-4.92\%$
test_vmap_mlp_speed_decorator[False-True]	0.8854ms	0.5520ms	1.8116 KOps/s	1.9205 KOps/s	$\textbf{\color{#d91a1a}-5.67\%}$
test_vmap_mlp_speed_decorator[False-False]	0.9194ms	0.5527ms	1.8092 KOps/s	1.9114 KOps/s	$\textbf{\color{#d91a1a}-5.35\%}$
test_to_module_speed[True]	2.2078ms	1.3670ms	731.5037 Ops/s	734.4825 Ops/s	$\color{#d91a1a}-0.41\%$
test_to_module_speed[False]	2.1232ms	1.3211ms	756.9313 Ops/s	759.1190 Ops/s	$\color{#d91a1a}-0.29\%$
test_tc_init	0.1101ms	47.5078μs	21.0492 KOps/s	21.6205 KOps/s	$\color{#d91a1a}-2.64\%$
test_tc_init_nested	0.1721ms	94.1172μs	10.6250 KOps/s	10.8364 KOps/s	$\color{#d91a1a}-1.95\%$
test_tc_first_layer_tensor	39.1610μs	1.5523μs	644.2204 KOps/s	647.3888 KOps/s	$\color{#d91a1a}-0.49\%$
test_tc_first_layer_nontensor	34.7870μs	4.7672μs	209.7652 KOps/s	208.7714 KOps/s	$\color{#35bf28}+0.48\%$
test_tc_second_layer_tensor	32.3300μs	3.2059μs	311.9283 KOps/s	339.6703 KOps/s	$\textbf{\color{#d91a1a}-8.17\%}$
test_tc_second_layer_nontensor	52.1080μs	6.3732μs	156.9058 KOps/s	164.1090 KOps/s	$\color{#d91a1a}-4.39\%$
test_unbind	0.2395s	13.5139ms	73.9982 Ops/s	69.2619 Ops/s	$\textbf{\color{#35bf28}+6.84\%}$
test_full_like	9.5861ms	7.8229ms	127.8296 Ops/s	133.1770 Ops/s	$\color{#d91a1a}-4.02\%$
test_zeros_like	6.2521ms	2.8868ms	346.4076 Ops/s	350.9916 Ops/s	$\color{#d91a1a}-1.31\%$
test_ones_like	4.7427ms	3.2398ms	308.6596 Ops/s	308.4542 Ops/s	$\color{#35bf28}+0.07\%$
test_clone	12.1052ms	6.9749ms	143.3704 Ops/s	191.2663 Ops/s	$\textbf{\color{#d91a1a}-25.04\%}$
test_squeeze	68.9190μs	13.0284μs	76.7551 KOps/s	78.3659 KOps/s	$\color{#d91a1a}-2.06\%$
test_unsqueeze	0.1608ms	93.6259μs	10.6808 KOps/s	10.6300 KOps/s	$\color{#35bf28}+0.48\%$
test_split	0.4663ms	0.1979ms	5.0529 KOps/s	5.1414 KOps/s	$\color{#d91a1a}-1.72\%$
test_permute	0.3498ms	0.2006ms	4.9860 KOps/s	4.9406 KOps/s	$\color{#35bf28}+0.92\%$
test_stack	29.7444ms	25.2266ms	39.6407 Ops/s	38.8995 Ops/s	$\color{#35bf28}+1.91\%$
test_cat	27.0820ms	24.9148ms	40.1367 Ops/s	38.8643 Ops/s	$\color{#35bf28}+3.27\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	32.5810μs	13.1181μs	76.2307 KOps/s	79.4407 KOps/s	$\color{#d91a1a}-4.04\%$
test_plain_set_stack_nested	40.7810μs	13.1157μs	76.2446 KOps/s	79.7696 KOps/s	$\color{#d91a1a}-4.42\%$
test_plain_set_nested_inplace	44.0610μs	14.1709μs	70.5671 KOps/s	73.7085 KOps/s	$\color{#d91a1a}-4.26\%$
test_plain_set_stack_nested_inplace	42.7800μs	14.1118μs	70.8629 KOps/s	74.4135 KOps/s	$\color{#d91a1a}-4.77\%$
test_items	27.8500μs	2.9147μs	343.0860 KOps/s	334.5725 KOps/s	$\color{#35bf28}+2.54\%$
test_items_nested	0.4203ms	0.3657ms	2.7342 KOps/s	2.7334 KOps/s	$\color{#35bf28}+0.03\%$
test_items_nested_locked	0.5652ms	0.3677ms	2.7197 KOps/s	2.7174 KOps/s	$\color{#35bf28}+0.08\%$
test_items_nested_leaf	0.1220ms	60.7742μs	16.4543 KOps/s	16.3566 KOps/s	$\color{#35bf28}+0.60\%$
test_items_stack_nested	0.3966ms	0.3656ms	2.7354 KOps/s	2.7365 KOps/s	$\color{#d91a1a}-0.04\%$
test_items_stack_nested_leaf	92.5020μs	62.5244μs	15.9938 KOps/s	16.2240 KOps/s	$\color{#d91a1a}-1.42\%$
test_items_stack_nested_locked	0.4144ms	0.3639ms	2.7482 KOps/s	2.7413 KOps/s	$\color{#35bf28}+0.25\%$
test_keys	28.2200μs	3.4482μs	290.0028 KOps/s	285.9957 KOps/s	$\color{#35bf28}+1.40\%$
test_keys_nested	0.1409ms	88.6211μs	11.2840 KOps/s	11.3073 KOps/s	$\color{#d91a1a}-0.21\%$
test_keys_nested_locked	0.7621ms	94.8208μs	10.5462 KOps/s	10.6177 KOps/s	$\color{#d91a1a}-0.67\%$
test_keys_nested_leaf	0.1097ms	79.8425μs	12.5247 KOps/s	12.5995 KOps/s	$\color{#d91a1a}-0.59\%$
test_keys_stack_nested	0.1425ms	89.4136μs	11.1840 KOps/s	11.3583 KOps/s	$\color{#d91a1a}-1.53\%$
test_keys_stack_nested_leaf	0.1039ms	81.1643μs	12.3207 KOps/s	12.5292 KOps/s	$\color{#d91a1a}-1.66\%$
test_keys_stack_nested_locked	0.1660ms	95.2646μs	10.4971 KOps/s	10.6813 KOps/s	$\color{#d91a1a}-1.72\%$
test_values	5.9617μs	0.8652μs	1.1558 MOps/s	1.1666 MOps/s	$\color{#d91a1a}-0.92\%$
test_values_nested	69.0310μs	37.5837μs	26.6073 KOps/s	26.6530 KOps/s	$\color{#d91a1a}-0.17\%$
test_values_nested_locked	71.7610μs	39.3310μs	25.4252 KOps/s	25.2526 KOps/s	$\color{#35bf28}+0.68\%$
test_values_nested_leaf	0.1219ms	42.8121μs	23.3579 KOps/s	23.4566 KOps/s	$\color{#d91a1a}-0.42\%$
test_values_stack_nested	75.5410μs	38.2857μs	26.1194 KOps/s	26.3029 KOps/s	$\color{#d91a1a}-0.70\%$
test_values_stack_nested_leaf	74.2210μs	43.1559μs	23.1718 KOps/s	23.4219 KOps/s	$\color{#d91a1a}-1.07\%$
test_values_stack_nested_locked	84.9520μs	40.2523μs	24.8433 KOps/s	24.9702 KOps/s	$\color{#d91a1a}-0.51\%$
test_membership	1.5630μs	0.5050μs	1.9804 MOps/s	1.9928 MOps/s	$\color{#d91a1a}-0.62\%$
test_membership_nested	25.2400μs	2.0864μs	479.2981 KOps/s	474.3379 KOps/s	$\color{#35bf28}+1.05\%$
test_membership_nested_leaf	17.3605μs	2.0528μs	487.1292 KOps/s	492.2626 KOps/s	$\color{#d91a1a}-1.04\%$
test_membership_stacked_nested	30.3010μs	2.0916μs	478.1130 KOps/s	474.4662 KOps/s	$\color{#35bf28}+0.77\%$
test_membership_stacked_nested_leaf	27.3910μs	2.0822μs	480.2662 KOps/s	471.1602 KOps/s	$\color{#35bf28}+1.93\%$
test_membership_nested_last	35.1310μs	3.1195μs	320.5633 KOps/s	320.2239 KOps/s	$\color{#35bf28}+0.11\%$
test_membership_nested_leaf_last	31.0800μs	3.1185μs	320.6707 KOps/s	320.6960 KOps/s	$-0.01\%$
test_membership_stacked_nested_last	28.8700μs	3.5961μs	278.0808 KOps/s	118.1031 KOps/s	$\textbf{\color{#35bf28}+135.46\%}$
test_membership_stacked_nested_leaf_last	96.5320μs	3.5819μs	279.1776 KOps/s	118.1504 KOps/s	$\textbf{\color{#35bf28}+136.29\%}$
test_nested_getleaf	36.3110μs	6.2846μs	159.1179 KOps/s	159.3153 KOps/s	$\color{#d91a1a}-0.12\%$
test_nested_get	35.4510μs	6.0031μs	166.5812 KOps/s	167.5453 KOps/s	$\color{#d91a1a}-0.58\%$
test_stacked_getleaf	26.3800μs	6.1809μs	161.7886 KOps/s	161.2302 KOps/s	$\color{#35bf28}+0.35\%$
test_stacked_get	55.6210μs	5.8332μs	171.4336 KOps/s	170.6425 KOps/s	$\color{#35bf28}+0.46\%$
test_nested_getitemleaf	32.9310μs	6.4984μs	153.8838 KOps/s	155.2167 KOps/s	$\color{#d91a1a}-0.86\%$
test_nested_getitem	40.1900μs	6.1254μs	163.2555 KOps/s	162.7608 KOps/s	$\color{#35bf28}+0.30\%$
test_stacked_getitemleaf	26.2700μs	6.4281μs	155.5677 KOps/s	155.0606 KOps/s	$\color{#35bf28}+0.33\%$
test_stacked_getitem	38.9010μs	6.0358μs	165.6775 KOps/s	165.5376 KOps/s	$\color{#35bf28}+0.08\%$
test_lock_nested	0.3991ms	0.3457ms	2.8927 KOps/s	2.9136 KOps/s	$\color{#d91a1a}-0.72\%$
test_lock_stack_nested	0.4563ms	0.3491ms	2.8642 KOps/s	2.8905 KOps/s	$\color{#d91a1a}-0.91\%$
test_unlock_nested	0.3527ms	0.2839ms	3.5221 KOps/s	3.5390 KOps/s	$\color{#d91a1a}-0.48\%$
test_unlock_stack_nested	0.3283ms	0.2818ms	3.5486 KOps/s	3.5996 KOps/s	$\color{#d91a1a}-1.42\%$
test_flatten_speed	0.1159ms	77.9490μs	12.8289 KOps/s	12.6002 KOps/s	$\color{#35bf28}+1.81\%$
test_unflatten_speed	0.3990ms	0.3246ms	3.0809 KOps/s	3.0713 KOps/s	$\color{#35bf28}+0.31\%$
test_common_ops	0.7924ms	0.6155ms	1.6246 KOps/s	1.6594 KOps/s	$\color{#d91a1a}-2.10\%$
test_creation	28.0600μs	1.7796μs	561.9374 KOps/s	560.9758 KOps/s	$\color{#35bf28}+0.17\%$
test_creation_empty	35.1110μs	9.3784μs	106.6276 KOps/s	118.0526 KOps/s	$\textbf{\color{#d91a1a}-9.68\%}$
test_creation_nested_1	40.6810μs	11.1063μs	90.0390 KOps/s	98.0476 KOps/s	$\textbf{\color{#d91a1a}-8.17\%}$
test_creation_nested_2	40.9800μs	13.8350μs	72.2805 KOps/s	77.6777 KOps/s	$\textbf{\color{#d91a1a}-6.95\%}$
test_clone	29.7900μs	10.2025μs	98.0152 KOps/s	100.0601 KOps/s	$\color{#d91a1a}-2.04\%$
test_getitem[int]	1.1932ms	10.8984μs	91.7564 KOps/s	92.0662 KOps/s	$\color{#d91a1a}-0.34\%$
test_getitem[slice_int]	0.1082ms	20.7481μs	48.1971 KOps/s	47.0807 KOps/s	$\color{#35bf28}+2.37\%$
test_getitem[range]	0.1229ms	36.7622μs	27.2018 KOps/s	27.0402 KOps/s	$\color{#35bf28}+0.60\%$
test_getitem[tuple]	0.1036ms	18.2811μs	54.7012 KOps/s	54.5493 KOps/s	$\color{#35bf28}+0.28\%$
test_getitem[list]	0.1247ms	32.9567μs	30.3429 KOps/s	30.6559 KOps/s	$\color{#d91a1a}-1.02\%$
test_setitem_dim[int]	41.4210μs	18.6287μs	53.6806 KOps/s	53.0447 KOps/s	$\color{#35bf28}+1.20\%$
test_setitem_dim[slice_int]	60.7110μs	36.8039μs	27.1710 KOps/s	26.4793 KOps/s	$\color{#35bf28}+2.61\%$
test_setitem_dim[range]	80.8120μs	52.2978μs	19.1213 KOps/s	19.2374 KOps/s	$\color{#d91a1a}-0.60\%$
test_setitem_dim[tuple]	61.2910μs	31.8053μs	31.4413 KOps/s	31.4620 KOps/s	$\color{#d91a1a}-0.07\%$
test_setitem	63.5410μs	15.3794μs	65.0222 KOps/s	68.4324 KOps/s	$\color{#d91a1a}-4.98\%$
test_set	67.6410μs	14.8947μs	67.1379 KOps/s	71.0523 KOps/s	$\textbf{\color{#d91a1a}-5.51\%}$
test_set_shared	0.5080ms	0.1552ms	6.4436 KOps/s	6.3977 KOps/s	$\color{#35bf28}+0.72\%$
test_update	0.4187ms	18.7631μs	53.2962 KOps/s	58.3289 KOps/s	$\textbf{\color{#d91a1a}-8.63\%}$
test_update_nested	76.4310μs	24.2554μs	41.2279 KOps/s	43.2062 KOps/s	$\color{#d91a1a}-4.58\%$
test_update__nested	0.4917ms	24.5936μs	40.6610 KOps/s	40.2525 KOps/s	$\color{#35bf28}+1.01\%$
test_set_nested	65.8310μs	16.2997μs	61.3508 KOps/s	64.8809 KOps/s	$\textbf{\color{#d91a1a}-5.44\%}$
test_set_nested_new	69.3310μs	18.7962μs	53.2022 KOps/s	55.8034 KOps/s	$\color{#d91a1a}-4.66\%$
test_select	62.5110μs	29.8553μs	33.4949 KOps/s	33.8052 KOps/s	$\color{#d91a1a}-0.92\%$
test_select_nested	71.2410μs	43.8804μs	22.7892 KOps/s	22.5931 KOps/s	$\color{#35bf28}+0.87\%$
test_exclude_nested	95.6510μs	63.8302μs	15.6666 KOps/s	15.4582 KOps/s	$\color{#35bf28}+1.35\%$
test_empty[True]	0.3271ms	0.2974ms	3.3624 KOps/s	3.3753 KOps/s	$\color{#d91a1a}-0.38\%$
test_empty[False]	3.2681μs	0.8185μs	1.2217 MOps/s	1.2129 MOps/s	$\color{#35bf28}+0.73\%$
test_to	88.9820μs	54.8447μs	18.2333 KOps/s	17.9100 KOps/s	$\color{#35bf28}+1.80\%$
test_to_nonblocking	85.1810μs	46.7949μs	21.3698 KOps/s	21.2083 KOps/s	$\color{#35bf28}+0.76\%$
test_unbind_speed	0.2773ms	0.2423ms	4.1279 KOps/s	4.1331 KOps/s	$\color{#d91a1a}-0.13\%$
test_unbind_speed_stack0	0.3108ms	0.2372ms	4.2159 KOps/s	4.2058 KOps/s	$\color{#35bf28}+0.24\%$
test_unbind_speed_stack1	92.9696ms	0.7199ms	1.3891 KOps/s	1.3865 KOps/s	$\color{#35bf28}+0.18\%$
test_split	95.7998ms	1.6065ms	622.4583 Ops/s	617.1189 Ops/s	$\color{#35bf28}+0.87\%$
test_chunk	95.1355ms	1.6041ms	623.4199 Ops/s	614.5226 Ops/s	$\color{#35bf28}+1.45\%$
test_consolidate[False-None]	3.3749ms	2.6607ms	375.8454 Ops/s	372.9970 Ops/s	$\color{#35bf28}+0.76\%$
test_consolidate[default-None]	1.7722ms	1.6905ms	591.5393 Ops/s	589.9783 Ops/s	$\color{#35bf28}+0.26\%$
test_consolidate[reduce-overhead-None]	1.8119ms	1.7313ms	577.6106 Ops/s	579.1552 Ops/s	$\color{#d91a1a}-0.27\%$
test_consolidate_njt[False-None]	6.8317ms	6.5840ms	151.8839 Ops/s	113.2139 Ops/s	$\textbf{\color{#35bf28}+34.16\%}$
test_to[False-False-None]	1.8000ms	1.6876ms	592.5675 Ops/s	588.9173 Ops/s	$\color{#35bf28}+0.62\%$
test_to[True-False-None]	1.5798ms	1.3570ms	736.9057 Ops/s	721.9913 Ops/s	$\color{#35bf28}+2.07\%$
test_to[within-False-None]	4.4443ms	4.1686ms	239.8904 Ops/s	236.2559 Ops/s	$\color{#35bf28}+1.54\%$
test_to[True-default-None]	5.5471ms	5.3284ms	187.6748 Ops/s	184.1381 Ops/s	$\color{#35bf28}+1.92\%$
test_to_njt[False-False-None]	7.0621ms	6.9371ms	144.1529 Ops/s	141.8789 Ops/s	$\color{#35bf28}+1.60\%$
test_to_njt[True-False-None]	5.7030ms	5.5799ms	179.2147 Ops/s	176.7204 Ops/s	$\color{#35bf28}+1.41\%$
test_to_njt[within-False-None]	12.3841ms	12.2249ms	81.8002 Ops/s	80.5304 Ops/s	$\color{#35bf28}+1.58\%$
test_creation[device0]	0.4601ms	78.9770μs	12.6619 KOps/s	11.8791 KOps/s	$\textbf{\color{#35bf28}+6.59\%}$
test_creation_from_tensor	0.6149ms	83.7987μs	11.9334 KOps/s	11.5049 KOps/s	$\color{#35bf28}+3.72\%$
test_add_one[memmap_tensor0]	0.3524ms	6.3997μs	156.2571 KOps/s	157.2588 KOps/s	$\color{#d91a1a}-0.64\%$
test_contiguous[memmap_tensor0]	1.9380μs	0.4246μs	2.3554 MOps/s	2.3448 MOps/s	$\color{#35bf28}+0.45\%$
test_stack[memmap_tensor0]	47.3000μs	4.5940μs	217.6739 KOps/s	229.4514 KOps/s	$\textbf{\color{#d91a1a}-5.13\%}$
test_memmaptd_index	1.4722ms	0.2429ms	4.1166 KOps/s	4.0947 KOps/s	$\color{#35bf28}+0.53\%$
test_memmaptd_index_astensor	0.4381ms	0.3057ms	3.2711 KOps/s	3.2643 KOps/s	$\color{#35bf28}+0.21\%$
test_memmaptd_index_op	0.7271ms	0.5817ms	1.7192 KOps/s	1.7628 KOps/s	$\color{#d91a1a}-2.47\%$
test_serialize_model	0.1309s	0.1300s	7.6953 Ops/s	7.6725 Ops/s	$\color{#35bf28}+0.30\%$
test_serialize_model_pickle	1.3655s	1.2161s	0.8223 Ops/s	0.8240 Ops/s	$\color{#d91a1a}-0.21\%$
test_serialize_weights	0.1316s	0.1293s	7.7340 Ops/s	7.7115 Ops/s	$\color{#35bf28}+0.29\%$
test_serialize_weights_returnearly	0.4397s	69.6106ms	14.3656 Ops/s	14.6216 Ops/s	$\color{#d91a1a}-1.75\%$
test_serialize_weights_pickle	1.3503s	1.1874s	0.8422 Ops/s	0.8396 Ops/s	$\color{#35bf28}+0.31\%$
test_reshape_pytree	59.5210μs	22.9409μs	43.5903 KOps/s	43.2346 KOps/s	$\color{#35bf28}+0.82\%$
test_reshape_td	0.1647ms	26.7824μs	37.3379 KOps/s	36.3591 KOps/s	$\color{#35bf28}+2.69\%$
test_view_pytree	47.5910μs	22.5243μs	44.3965 KOps/s	45.0643 KOps/s	$\color{#d91a1a}-1.48\%$
test_view_td	66.1910μs	31.5590μs	31.6867 KOps/s	30.9095 KOps/s	$\color{#35bf28}+2.51\%$
test_unbind_pytree	57.7410μs	28.8184μs	34.7000 KOps/s	35.1478 KOps/s	$\color{#d91a1a}-1.27\%$
test_unbind_td	0.9564ms	37.6159μs	26.5845 KOps/s	26.8543 KOps/s	$\color{#d91a1a}-1.00\%$
test_split_pytree	59.6110μs	30.8136μs	32.4532 KOps/s	32.7437 KOps/s	$\color{#d91a1a}-0.89\%$
test_split_td	0.1770ms	38.8441μs	25.7439 KOps/s	24.8252 KOps/s	$\color{#35bf28}+3.70\%$
test_add_pytree	69.4510μs	33.6187μs	29.7453 KOps/s	29.8480 KOps/s	$\color{#d91a1a}-0.34\%$
test_add_td	86.4620μs	50.4692μs	19.8141 KOps/s	20.7791 KOps/s	$\color{#d91a1a}-4.64\%$
test_compile_add_one_nested[tensordict-compile]	0.1758ms	0.1233ms	8.1118 KOps/s	7.8032 KOps/s	$\color{#35bf28}+3.96\%$
test_compile_add_one_nested[tensordict-eager]	0.2233ms	0.1357ms	7.3699 KOps/s	7.2587 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_add_one_nested[pytree-compile]	0.1424ms	0.1002ms	9.9804 KOps/s	10.2128 KOps/s	$\color{#d91a1a}-2.28\%$
test_compile_add_one_nested[pytree-eager]	1.4031ms	0.1457ms	6.8632 KOps/s	6.6583 KOps/s	$\color{#35bf28}+3.08\%$
test_compile_copy_nested[tensordict-compile]	99.3020μs	25.3979μs	39.3733 KOps/s	38.5577 KOps/s	$\color{#35bf28}+2.12\%$
test_compile_copy_nested[tensordict-eager]	71.0520μs	29.8597μs	33.4900 KOps/s	34.0503 KOps/s	$\color{#d91a1a}-1.65\%$
test_compile_copy_nested[pytree-compile]	0.3758ms	65.0348μs	15.3764 KOps/s	15.0873 KOps/s	$\color{#35bf28}+1.92\%$
test_compile_copy_nested[pytree-eager]	84.4220μs	49.6870μs	20.1260 KOps/s	19.6321 KOps/s	$\color{#35bf28}+2.52\%$
test_compile_add_one_flat[tensordict-compile]	0.1861ms	0.1424ms	7.0216 KOps/s	7.0513 KOps/s	$\color{#d91a1a}-0.42\%$
test_compile_add_one_flat[tensordict-eager]	0.3207ms	0.2178ms	4.5914 KOps/s	4.6137 KOps/s	$\color{#d91a1a}-0.48\%$
test_compile_add_one_flat[tensorclass-compile]	0.1343ms	97.7556μs	10.2296 KOps/s	10.2508 KOps/s	$\color{#d91a1a}-0.21\%$
test_compile_add_one_flat[tensorclass-eager]	0.1137ms	56.6725μs	17.6452 KOps/s	17.2531 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_add_one_flat[pytree-compile]	0.1840ms	0.1366ms	7.3231 KOps/s	7.3623 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_add_one_flat[pytree-eager]	0.5998ms	0.4609ms	2.1698 KOps/s	2.1027 KOps/s	$\color{#35bf28}+3.19\%$
test_compile_add_self_flat[tensordict-eager]	0.3786ms	0.2616ms	3.8221 KOps/s	3.8055 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_self_flat[tensordict-compile]	0.2853ms	0.1457ms	6.8630 KOps/s	7.0370 KOps/s	$\color{#d91a1a}-2.47\%$
test_compile_add_self_flat[tensorclass-eager]	0.1556ms	70.6923μs	14.1458 KOps/s	14.5429 KOps/s	$\color{#d91a1a}-2.73\%$
test_compile_add_self_flat[tensorclass-compile]	0.2309ms	99.1181μs	10.0890 KOps/s	10.1544 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_add_self_flat[pytree-eager]	0.4403ms	0.3937ms	2.5401 KOps/s	2.5233 KOps/s	$\color{#35bf28}+0.66\%$
test_compile_add_self_flat[pytree-compile]	0.1799ms	0.1355ms	7.3797 KOps/s	7.4048 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_copy_flat[tensordict-compile]	53.0010μs	19.4865μs	51.3175 KOps/s	36.4500 KOps/s	$\textbf{\color{#35bf28}+40.79\%}$
test_compile_copy_flat[tensordict-eager]	67.7910μs	31.7742μs	31.4721 KOps/s	32.0749 KOps/s	$\color{#d91a1a}-1.88\%$
test_compile_copy_flat[pytree-compile]	0.1400ms	70.8526μs	14.1138 KOps/s	14.0032 KOps/s	$\color{#35bf28}+0.79\%$
test_compile_copy_flat[pytree-eager]	0.1537ms	53.2118μs	18.7928 KOps/s	18.8286 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_assign_and_add[tensordict-compile]	1.6570ms	0.3972ms	2.5174 KOps/s	2.1891 KOps/s	$\textbf{\color{#35bf28}+14.99\%}$
test_compile_assign_and_add[tensordict-eager]	3.0433ms	2.7467ms	364.0776 Ops/s	373.8378 Ops/s	$\color{#d91a1a}-2.61\%$
test_compile_assign_and_add[pytree-compile]	1.6074ms	0.4349ms	2.2992 KOps/s	2.2413 KOps/s	$\color{#35bf28}+2.59\%$
test_compile_assign_and_add[pytree-eager]	2.9113ms	2.6432ms	378.3338 Ops/s	386.0650 Ops/s	$\color{#d91a1a}-2.00\%$
test_compile_indexing[tensor-tensordict-compile]	0.4117ms	0.1200ms	8.3312 KOps/s	8.5653 KOps/s	$\color{#d91a1a}-2.73\%$
test_compile_indexing[tensor-tensordict-eager]	0.5757ms	81.2232μs	12.3118 KOps/s	12.2372 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_indexing[tensor-tensorclass-compile]	0.6596ms	0.1125ms	8.8857 KOps/s	8.8536 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1947ms	69.6714μs	14.3531 KOps/s	13.7713 KOps/s	$\color{#35bf28}+4.22\%$
test_compile_indexing[tensor-pytree-compile]	0.2932ms	0.1125ms	8.8857 KOps/s	8.7580 KOps/s	$\color{#35bf28}+1.46\%$
test_compile_indexing[tensor-pytree-eager]	0.2039ms	69.3709μs	14.4153 KOps/s	13.8401 KOps/s	$\color{#35bf28}+4.16\%$
test_compile_indexing[slice-tensordict-compile]	0.2479ms	0.1087ms	9.1996 KOps/s	9.4980 KOps/s	$\color{#d91a1a}-3.14\%$
test_compile_indexing[slice-tensordict-eager]	0.1690ms	17.4365μs	57.3509 KOps/s	55.3237 KOps/s	$\color{#35bf28}+3.66\%$
test_compile_indexing[slice-tensorclass-compile]	0.2165ms	0.1015ms	9.8480 KOps/s	9.9148 KOps/s	$\color{#d91a1a}-0.67\%$
test_compile_indexing[slice-tensorclass-eager]	88.0420μs	16.1546μs	61.9020 KOps/s	62.2776 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_indexing[slice-pytree-compile]	0.2298ms	0.1029ms	9.7160 KOps/s	9.8866 KOps/s	$\color{#d91a1a}-1.73\%$
test_compile_indexing[slice-pytree-eager]	0.1018ms	16.2286μs	61.6195 KOps/s	63.2141 KOps/s	$\color{#d91a1a}-2.52\%$
test_compile_indexing[int-tensordict-compile]	0.2278ms	0.1067ms	9.3715 KOps/s	9.8451 KOps/s	$\color{#d91a1a}-4.81\%$
test_compile_indexing[int-tensordict-eager]	0.5892ms	17.3233μs	57.7258 KOps/s	56.6625 KOps/s	$\color{#35bf28}+1.88\%$
test_compile_indexing[int-tensorclass-compile]	0.2322ms	0.1037ms	9.6451 KOps/s	10.3342 KOps/s	$\textbf{\color{#d91a1a}-6.67\%}$
test_compile_indexing[int-tensorclass-eager]	0.1112ms	16.0385μs	62.3500 KOps/s	63.1891 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_indexing[int-pytree-compile]	0.2342ms	0.1024ms	9.7627 KOps/s	10.3637 KOps/s	$\textbf{\color{#d91a1a}-5.80\%}$
test_compile_indexing[int-pytree-eager]	87.0010μs	15.9942μs	62.5226 KOps/s	60.4886 KOps/s	$\color{#35bf28}+3.36\%$
test_mod_add[eager]	0.1603ms	38.6734μs	25.8576 KOps/s	25.8636 KOps/s	$\color{#d91a1a}-0.02\%$
test_mod_add[compile]	0.2896ms	82.8233μs	12.0739 KOps/s	12.1533 KOps/s	$\color{#d91a1a}-0.65\%$
test_mod_add[compile-overhead]	0.3255ms	0.1685ms	5.9349 KOps/s	5.6184 KOps/s	$\textbf{\color{#35bf28}+5.63\%}$
test_mod_wrap[eager]	0.3296ms	0.2584ms	3.8701 KOps/s	3.9950 KOps/s	$\color{#d91a1a}-3.13\%$
test_mod_wrap[compile]	0.3770ms	0.2886ms	3.4650 KOps/s	3.4314 KOps/s	$\color{#35bf28}+0.98\%$
test_mod_wrap[compile-overhead]	7.2570ms	3.8890ms	257.1373 Ops/s	261.4183 Ops/s	$\color{#d91a1a}-1.64\%$
test_mod_wrap_and_backward[eager]	1.4677ms	1.3422ms	745.0196 Ops/s	700.6597 Ops/s	$\textbf{\color{#35bf28}+6.33\%}$
test_mod_wrap_and_backward[compile]	1.4074ms	1.2821ms	779.9593 Ops/s	722.3696 Ops/s	$\textbf{\color{#35bf28}+7.97\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3776ms	0.9261ms	1.0798 KOps/s	961.6483 Ops/s	$\textbf{\color{#35bf28}+12.28\%}$
test_seq_add[eager]	0.1790ms	0.1231ms	8.1239 KOps/s	8.2844 KOps/s	$\color{#d91a1a}-1.94\%$
test_seq_add[compile]	0.2026ms	91.9513μs	10.8753 KOps/s	10.9719 KOps/s	$\color{#d91a1a}-0.88\%$
test_seq_add[compile-overhead]	0.1723ms	0.1312ms	7.6215 KOps/s	7.6354 KOps/s	$\color{#d91a1a}-0.18\%$
test_seq_wrap[eager]	0.4914ms	0.4277ms	2.3383 KOps/s	2.2148 KOps/s	$\textbf{\color{#35bf28}+5.58\%}$
test_seq_wrap[compile]	0.3461ms	0.3046ms	3.2828 KOps/s	3.2607 KOps/s	$\color{#35bf28}+0.68\%$
test_seq_wrap[compile-overhead]	0.2765ms	0.2283ms	4.3796 KOps/s	4.4041 KOps/s	$\color{#d91a1a}-0.55\%$
test_func_call_runtime[False-eager]	0.8353ms	0.7535ms	1.3271 KOps/s	1.3705 KOps/s	$\color{#d91a1a}-3.17\%$
test_func_call_runtime[False-compile]	0.8966ms	0.7856ms	1.2729 KOps/s	1.3046 KOps/s	$\color{#d91a1a}-2.43\%$
test_func_call_runtime[False-compile-overhead]	0.4788ms	0.3715ms	2.6917 KOps/s	2.7218 KOps/s	$\color{#d91a1a}-1.11\%$
test_func_call_runtime[True-eager]	0.9468ms	0.8855ms	1.1294 KOps/s	1.1135 KOps/s	$\color{#35bf28}+1.43\%$
test_func_call_runtime[True-compile]	0.8493ms	0.7788ms	1.2840 KOps/s	1.2736 KOps/s	$\color{#35bf28}+0.82\%$
test_func_call_runtime[True-compile-overhead]	0.4557ms	0.3887ms	2.5728 KOps/s	2.5740 KOps/s	$\color{#d91a1a}-0.04\%$
test_func_call_cm_runtime[False-eager]	0.8903ms	0.7285ms	1.3727 KOps/s	1.3831 KOps/s	$\color{#d91a1a}-0.76\%$
test_func_call_cm_runtime[False-compile]	1.2968ms	0.7588ms	1.3179 KOps/s	1.3066 KOps/s	$\color{#35bf28}+0.87\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4251ms	0.3699ms	2.7035 KOps/s	2.6934 KOps/s	$\color{#35bf28}+0.38\%$
test_func_call_cm_runtime[True-eager]	1.0703ms	0.9908ms	1.0093 KOps/s	986.5777 Ops/s	$\color{#35bf28}+2.31\%$
test_func_call_cm_runtime[True-compile]	1.1321ms	1.0228ms	977.7369 Ops/s	1.0115 KOps/s	$\color{#d91a1a}-3.33\%$
test_func_call_cm_runtime[True-compile-overhead]	1.0618ms	0.9842ms	1.0161 KOps/s	1.0034 KOps/s	$\color{#35bf28}+1.26\%$
test_vmap_func_call_cm_runtime[eager]	2.4638ms	2.0318ms	492.1627 Ops/s	487.2872 Ops/s	$\color{#35bf28}+1.00\%$
test_vmap_func_call_cm_runtime[compile]	0.9270ms	0.8231ms	1.2149 KOps/s	1.1912 KOps/s	$\color{#35bf28}+1.99\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4716ms	0.4211ms	2.3745 KOps/s	2.3695 KOps/s	$\color{#35bf28}+0.21\%$
test_distributed	2.6674ms	0.3375ms	2.9627 KOps/s	8.3495 KOps/s	$\textbf{\color{#d91a1a}-64.52\%}$
test_tdmodule	53.7110μs	20.8133μs	48.0463 KOps/s	47.9519 KOps/s	$\color{#35bf28}+0.20\%$
test_tdmodule_dispatch	72.0310μs	37.2834μs	26.8216 KOps/s	26.7225 KOps/s	$\color{#35bf28}+0.37\%$
test_tdseq	45.2800μs	21.5419μs	46.4212 KOps/s	47.1392 KOps/s	$\color{#d91a1a}-1.52\%$
test_tdseq_dispatch	60.1110μs	40.2384μs	24.8519 KOps/s	25.7853 KOps/s	$\color{#d91a1a}-3.62\%$
test_instantiation_functorch	1.6636ms	1.5560ms	642.6555 Ops/s	641.0899 Ops/s	$\color{#35bf28}+0.24\%$
test_exec_functorch	0.2020ms	0.1393ms	7.1794 KOps/s	7.0100 KOps/s	$\color{#35bf28}+2.42\%$
test_exec_functional_call	0.1855ms	0.1320ms	7.5732 KOps/s	7.5718 KOps/s	$\color{#35bf28}+0.02\%$
test_exec_td_decorator	0.3611ms	0.1816ms	5.5066 KOps/s	5.4263 KOps/s	$\color{#35bf28}+1.48\%$
test_vmap_mlp_speed_decorator[True-True]	0.7883ms	0.6714ms	1.4893 KOps/s	1.4285 KOps/s	$\color{#35bf28}+4.25\%$
test_vmap_mlp_speed_decorator[True-False]	0.7907ms	0.6707ms	1.4910 KOps/s	1.4724 KOps/s	$\color{#35bf28}+1.27\%$
test_vmap_mlp_speed_decorator[False-True]	0.7310ms	0.5766ms	1.7343 KOps/s	1.7293 KOps/s	$\color{#35bf28}+0.29\%$
test_vmap_mlp_speed_decorator[False-False]	0.7114ms	0.5759ms	1.7364 KOps/s	1.7270 KOps/s	$\color{#35bf28}+0.54\%$
test_vmap_transformer_speed_decorator[True-True]	18.7757ms	18.6390ms	53.6509 Ops/s	53.6844 Ops/s	$\color{#d91a1a}-0.06\%$
test_vmap_transformer_speed_decorator[True-False]	18.8548ms	18.6619ms	53.5852 Ops/s	53.1643 Ops/s	$\color{#35bf28}+0.79\%$
test_vmap_transformer_speed_decorator[False-True]	18.7087ms	18.4886ms	54.0874 Ops/s	54.0018 Ops/s	$\color{#35bf28}+0.16\%$
test_vmap_transformer_speed_decorator[False-False]	18.6755ms	18.5331ms	53.9575 Ops/s	53.9473 Ops/s	$\color{#35bf28}+0.02\%$
test_to_module_speed[True]	1.1839ms	0.9783ms	1.0222 KOps/s	1.0210 KOps/s	$\color{#35bf28}+0.11\%$
test_to_module_speed[False]	1.3594ms	0.9536ms	1.0487 KOps/s	1.0460 KOps/s	$\color{#35bf28}+0.25\%$
test_tc_init	70.7910μs	36.2284μs	27.6027 KOps/s	28.1831 KOps/s	$\color{#d91a1a}-2.06\%$
test_tc_init_nested	0.1168ms	73.7204μs	13.5648 KOps/s	13.5546 KOps/s	$\color{#35bf28}+0.07\%$
test_tc_first_layer_tensor	23.4300μs	0.8275μs	1.2084 MOps/s	1.4235 MOps/s	$\textbf{\color{#d91a1a}-15.11\%}$
test_tc_first_layer_nontensor	28.5600μs	2.2735μs	439.8592 KOps/s	425.1754 KOps/s	$\color{#35bf28}+3.45\%$
test_tc_second_layer_tensor	12.2437μs	1.4407μs	694.1296 KOps/s	694.2729 KOps/s	$\color{#d91a1a}-0.02\%$
test_tc_second_layer_nontensor	23.7700μs	3.0150μs	331.6732 KOps/s	322.1516 KOps/s	$\color{#35bf28}+2.96\%$
test_unbind	0.2238s	10.1404ms	98.6159 Ops/s	142.2247 Ops/s	$\textbf{\color{#d91a1a}-30.66\%}$
test_full_like	11.4011ms	9.3175ms	107.3249 Ops/s	107.8702 Ops/s	$\color{#d91a1a}-0.51\%$
test_zeros_like	4.9887ms	4.3329ms	230.7912 Ops/s	231.6129 Ops/s	$\color{#d91a1a}-0.35\%$
test_ones_like	9.4247ms	7.1552ms	139.7586 Ops/s	230.4833 Ops/s	$\textbf{\color{#d91a1a}-39.36\%}$
test_clone	7.3189ms	6.4523ms	154.9824 Ops/s	154.3629 Ops/s	$\color{#35bf28}+0.40\%$
test_squeeze	58.2010μs	10.0548μs	99.4547 KOps/s	100.9527 KOps/s	$\color{#d91a1a}-1.48\%$
test_unsqueeze	0.1337ms	74.1263μs	13.4905 KOps/s	13.4667 KOps/s	$\color{#35bf28}+0.18\%$
test_split	0.3691ms	0.1590ms	6.2901 KOps/s	6.0952 KOps/s	$\color{#35bf28}+3.20\%$
test_permute	0.2273ms	0.1823ms	5.4849 KOps/s	5.3682 KOps/s	$\color{#35bf28}+2.18\%$
test_stack	53.6053ms	51.7006ms	19.3421 Ops/s	19.6376 Ops/s	$\color{#d91a1a}-1.50\%$
test_cat	54.4925ms	51.1718ms	19.5420 Ops/s	19.5776 Ops/s	$\color{#d91a1a}-0.18\%$