[Refactor] __eq__ to identity check in non-tensor stacking

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Refactor] __eq__ to identity check in non-tensor stacking #1083

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	37.1300μs	18.1266μs	55.1677 KOps/s	56.7243 KOps/s	$\color{#d91a1a}-2.74\%$
test_plain_set_stack_nested	65.8430μs	18.3472μs	54.5041 KOps/s	55.9594 KOps/s	$\color{#d91a1a}-2.60\%$
test_plain_set_nested_inplace	42.1380μs	19.7925μs	50.5241 KOps/s	51.7223 KOps/s	$\color{#d91a1a}-2.32\%$
test_plain_set_stack_nested_inplace	76.7440μs	19.8986μs	50.2549 KOps/s	52.2524 KOps/s	$\color{#d91a1a}-3.82\%$
test_items	44.1920μs	4.2323μs	236.2764 KOps/s	242.0192 KOps/s	$\color{#d91a1a}-2.37\%$
test_items_nested	0.7288ms	0.3453ms	2.8963 KOps/s	2.9854 KOps/s	$\color{#d91a1a}-2.99\%$
test_items_nested_locked	0.5577ms	0.3428ms	2.9171 KOps/s	2.9560 KOps/s	$\color{#d91a1a}-1.32\%$
test_items_nested_leaf	0.1504ms	70.9741μs	14.0896 KOps/s	13.8349 KOps/s	$\color{#35bf28}+1.84\%$
test_items_stack_nested	0.5009ms	0.3428ms	2.9170 KOps/s	2.9084 KOps/s	$\color{#35bf28}+0.29\%$
test_items_stack_nested_leaf	0.1397ms	73.0440μs	13.6904 KOps/s	13.7914 KOps/s	$\color{#d91a1a}-0.73\%$
test_items_stack_nested_locked	0.4959ms	0.3465ms	2.8859 KOps/s	2.8951 KOps/s	$\color{#d91a1a}-0.32\%$
test_keys	41.6070μs	3.5036μs	285.4241 KOps/s	285.0020 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_nested	0.2341ms	0.1339ms	7.4704 KOps/s	7.2913 KOps/s	$\color{#35bf28}+2.46\%$
test_keys_nested_locked	1.8843ms	0.1409ms	7.0952 KOps/s	7.0604 KOps/s	$\color{#35bf28}+0.49\%$
test_keys_nested_leaf	0.1956ms	0.1153ms	8.6695 KOps/s	8.6942 KOps/s	$\color{#d91a1a}-0.28\%$
test_keys_stack_nested	0.2456ms	0.1357ms	7.3681 KOps/s	7.2925 KOps/s	$\color{#35bf28}+1.04\%$
test_keys_stack_nested_leaf	0.2662ms	0.1149ms	8.7017 KOps/s	8.5990 KOps/s	$\color{#35bf28}+1.19\%$
test_keys_stack_nested_locked	0.2379ms	0.1397ms	7.1607 KOps/s	7.0485 KOps/s	$\color{#35bf28}+1.59\%$
test_values	7.0412μs	1.0395μs	962.0246 KOps/s	944.3459 KOps/s	$\color{#35bf28}+1.87\%$
test_values_nested	0.1024ms	55.8055μs	17.9194 KOps/s	17.9041 KOps/s	$\color{#35bf28}+0.09\%$
test_values_nested_locked	0.1313ms	55.7239μs	17.9456 KOps/s	18.1515 KOps/s	$\color{#d91a1a}-1.13\%$
test_values_nested_leaf	0.1114ms	61.5466μs	16.2478 KOps/s	16.0624 KOps/s	$\color{#35bf28}+1.15\%$
test_values_stack_nested	0.1138ms	56.4047μs	17.7290 KOps/s	17.7679 KOps/s	$\color{#d91a1a}-0.22\%$
test_values_stack_nested_leaf	0.1073ms	61.2673μs	16.3219 KOps/s	16.3552 KOps/s	$\color{#d91a1a}-0.20\%$
test_values_stack_nested_locked	0.1169ms	56.1061μs	17.8234 KOps/s	17.5748 KOps/s	$\color{#35bf28}+1.41\%$
test_membership	25.2970μs	0.8683μs	1.1517 MOps/s	1.3367 MOps/s	$\textbf{\color{#d91a1a}-13.84\%}$
test_membership_nested	22.1510μs	2.7131μs	368.5869 KOps/s	357.9969 KOps/s	$\color{#35bf28}+2.96\%$
test_membership_nested_leaf	38.5440μs	2.7055μs	369.6219 KOps/s	359.0427 KOps/s	$\color{#35bf28}+2.95\%$
test_membership_stacked_nested	23.7540μs	2.7159μs	368.1984 KOps/s	361.3772 KOps/s	$\color{#35bf28}+1.89\%$
test_membership_stacked_nested_leaf	24.2560μs	2.6912μs	371.5800 KOps/s	363.0677 KOps/s	$\color{#35bf28}+2.34\%$
test_membership_nested_last	24.5160μs	4.0227μs	248.5904 KOps/s	242.8690 KOps/s	$\color{#35bf28}+2.36\%$
test_membership_nested_leaf_last	55.3310μs	4.0240μs	248.5102 KOps/s	244.0609 KOps/s	$\color{#35bf28}+1.82\%$
test_membership_stacked_nested_last	29.3050μs	4.1089μs	243.3740 KOps/s	246.3114 KOps/s	$\color{#d91a1a}-1.19\%$
test_membership_stacked_nested_leaf_last	27.7920μs	4.0883μs	244.5978 KOps/s	244.9571 KOps/s	$\color{#d91a1a}-0.15\%$
test_nested_getleaf	46.9300μs	10.6226μs	94.1391 KOps/s	92.6765 KOps/s	$\color{#35bf28}+1.58\%$
test_nested_get	35.6570μs	10.3175μs	96.9227 KOps/s	100.6888 KOps/s	$\color{#d91a1a}-3.74\%$
test_stacked_getleaf	40.3350μs	10.6649μs	93.7656 KOps/s	95.9154 KOps/s	$\color{#d91a1a}-2.24\%$
test_stacked_get	55.3320μs	10.0604μs	99.3996 KOps/s	100.9412 KOps/s	$\color{#d91a1a}-1.53\%$
test_nested_getitemleaf	40.3160μs	11.1278μs	89.8651 KOps/s	91.5568 KOps/s	$\color{#d91a1a}-1.85\%$
test_nested_getitem	40.5850μs	10.3927μs	96.2211 KOps/s	98.6796 KOps/s	$\color{#d91a1a}-2.49\%$
test_stacked_getitemleaf	0.1273ms	11.5850μs	86.3182 KOps/s	91.8423 KOps/s	$\textbf{\color{#d91a1a}-6.01\%}$
test_stacked_getitem	46.7970μs	10.5582μs	94.7128 KOps/s	98.1600 KOps/s	$\color{#d91a1a}-3.51\%$
test_lock_nested	2.9089ms	0.4384ms	2.2808 KOps/s	1.8268 KOps/s	$\textbf{\color{#35bf28}+24.85\%}$
test_lock_stack_nested	0.7151ms	0.4087ms	2.4467 KOps/s	2.4160 KOps/s	$\color{#35bf28}+1.27\%$
test_unlock_nested	1.0300ms	0.3564ms	2.8060 KOps/s	2.7564 KOps/s	$\color{#35bf28}+1.80\%$
test_unlock_stack_nested	0.9814ms	0.3316ms	3.0158 KOps/s	3.0076 KOps/s	$\color{#35bf28}+0.27\%$
test_flatten_speed	0.1910ms	92.7468μs	10.7820 KOps/s	10.9644 KOps/s	$\color{#d91a1a}-1.66\%$
test_unflatten_speed	0.7938ms	0.4695ms	2.1300 KOps/s	2.0995 KOps/s	$\color{#35bf28}+1.45\%$
test_common_ops	4.4194ms	0.7561ms	1.3226 KOps/s	1.3196 KOps/s	$\color{#35bf28}+0.22\%$
test_creation	0.1184ms	2.1263μs	470.2916 KOps/s	485.3495 KOps/s	$\color{#d91a1a}-3.10\%$
test_creation_empty	36.4290μs	10.4196μs	95.9733 KOps/s	92.7883 KOps/s	$\color{#35bf28}+3.43\%$
test_creation_nested_1	43.3110μs	13.0409μs	76.6818 KOps/s	73.9802 KOps/s	$\color{#35bf28}+3.65\%$
test_creation_nested_2	47.2690μs	17.2653μs	57.9198 KOps/s	57.9802 KOps/s	$\color{#d91a1a}-0.10\%$
test_clone	76.1230μs	12.8063μs	78.0868 KOps/s	76.4615 KOps/s	$\color{#35bf28}+2.13\%$
test_getitem[int]	0.9854ms	12.4198μs	80.5165 KOps/s	80.7850 KOps/s	$\color{#d91a1a}-0.33\%$
test_getitem[slice_int]	0.1733ms	23.3138μs	42.8931 KOps/s	43.6302 KOps/s	$\color{#d91a1a}-1.69\%$
test_getitem[range]	0.1696ms	47.2778μs	21.1516 KOps/s	20.6639 KOps/s	$\color{#35bf28}+2.36\%$
test_getitem[tuple]	0.1320ms	19.2042μs	52.0718 KOps/s	51.4670 KOps/s	$\color{#35bf28}+1.18\%$
test_getitem[list]	0.1701ms	42.6897μs	23.4249 KOps/s	22.9244 KOps/s	$\color{#35bf28}+2.18\%$
test_setitem_dim[int]	44.7240μs	24.9217μs	40.1257 KOps/s	39.1597 KOps/s	$\color{#35bf28}+2.47\%$
test_setitem_dim[slice_int]	92.3030μs	50.1285μs	19.9487 KOps/s	19.9523 KOps/s	$\color{#d91a1a}-0.02\%$
test_setitem_dim[range]	0.1312ms	74.0709μs	13.5006 KOps/s	13.5367 KOps/s	$\color{#d91a1a}-0.27\%$
test_setitem_dim[tuple]	75.9720μs	39.5276μs	25.2988 KOps/s	25.3593 KOps/s	$\color{#d91a1a}-0.24\%$
test_setitem	0.1298ms	19.5369μs	51.1853 KOps/s	50.7508 KOps/s	$\color{#35bf28}+0.86\%$
test_set	0.1958ms	19.0775μs	52.4177 KOps/s	52.3500 KOps/s	$\color{#35bf28}+0.13\%$
test_set_shared	1.3505ms	0.1677ms	5.9623 KOps/s	5.9697 KOps/s	$\color{#d91a1a}-0.12\%$
test_update	0.1585ms	21.4727μs	46.5707 KOps/s	45.6917 KOps/s	$\color{#35bf28}+1.92\%$
test_update_nested	0.1800ms	31.4199μs	31.8270 KOps/s	31.8809 KOps/s	$\color{#d91a1a}-0.17\%$
test_update__nested	0.9491ms	32.6236μs	30.6526 KOps/s	30.9396 KOps/s	$\color{#d91a1a}-0.93\%$
test_set_nested	0.1476ms	21.5110μs	46.4879 KOps/s	46.7074 KOps/s	$\color{#d91a1a}-0.47\%$
test_set_nested_new	0.1005ms	25.8850μs	38.6324 KOps/s	37.8347 KOps/s	$\color{#35bf28}+2.11\%$
test_select	0.1033ms	41.6428μs	24.0138 KOps/s	24.0366 KOps/s	$\color{#d91a1a}-0.09\%$
test_select_nested	0.1290ms	59.0452μs	16.9362 KOps/s	16.6210 KOps/s	$\color{#35bf28}+1.90\%$
test_exclude_nested	0.1600ms	75.1073μs	13.3143 KOps/s	13.5464 KOps/s	$\color{#d91a1a}-1.71\%$
test_empty[True]	0.5291ms	0.3470ms	2.8819 KOps/s	2.8899 KOps/s	$\color{#d91a1a}-0.28\%$
test_empty[False]	9.4202μs	1.2193μs	820.1763 KOps/s	825.3668 KOps/s	$\color{#d91a1a}-0.63\%$
test_unbind_speed	0.3557ms	0.2572ms	3.8877 KOps/s	3.9037 KOps/s	$\color{#d91a1a}-0.41\%$
test_unbind_speed_stack0	0.4557ms	0.2552ms	3.9181 KOps/s	3.8368 KOps/s	$\color{#35bf28}+2.12\%$
test_unbind_speed_stack1	0.1013s	0.7636ms	1.3096 KOps/s	1.5867 KOps/s	$\textbf{\color{#d91a1a}-17.47\%}$
test_split	0.1007s	1.6917ms	591.1162 Ops/s	573.1255 Ops/s	$\color{#35bf28}+3.14\%$
test_chunk	0.1033s	1.7032ms	587.1268 Ops/s	556.6661 Ops/s	$\textbf{\color{#35bf28}+5.47\%}$
test_consolidate_njt[False-None]	8.8103ms	8.2193ms	121.6655 Ops/s	124.4331 Ops/s	$\color{#d91a1a}-2.22\%$
test_creation[device0]	4.2142ms	92.0170μs	10.8676 KOps/s	10.6019 KOps/s	$\color{#35bf28}+2.51\%$
test_creation_from_tensor	0.2843ms	93.6424μs	10.6789 KOps/s	10.5163 KOps/s	$\color{#35bf28}+1.55\%$
test_add_one[memmap_tensor0]	0.1826ms	4.9022μs	203.9908 KOps/s	195.5254 KOps/s	$\color{#35bf28}+4.33\%$
test_contiguous[memmap_tensor0]	19.3360μs	0.5183μs	1.9295 MOps/s	1.9376 MOps/s	$\color{#d91a1a}-0.42\%$
test_stack[memmap_tensor0]	36.4780μs	3.3855μs	295.3791 KOps/s	280.2083 KOps/s	$\textbf{\color{#35bf28}+5.41\%}$
test_memmaptd_index	1.0530ms	0.2374ms	4.2131 KOps/s	4.2343 KOps/s	$\color{#d91a1a}-0.50\%$
test_memmaptd_index_astensor	0.5773ms	0.3147ms	3.1778 KOps/s	3.1856 KOps/s	$\color{#d91a1a}-0.24\%$
test_memmaptd_index_op	0.9740ms	0.5824ms	1.7169 KOps/s	1.7356 KOps/s	$\color{#d91a1a}-1.08\%$
test_serialize_model	0.1290s	0.1212s	8.2509 Ops/s	7.1504 Ops/s	$\textbf{\color{#35bf28}+15.39\%}$
test_serialize_model_pickle	0.4424s	0.3944s	2.5355 Ops/s	2.5918 Ops/s	$\color{#d91a1a}-2.17\%$
test_serialize_weights	0.1188s	0.1135s	8.8116 Ops/s	8.7548 Ops/s	$\color{#35bf28}+0.65\%$
test_serialize_weights_returnearly	0.1848s	0.1591s	6.2859 Ops/s	6.2748 Ops/s	$\color{#35bf28}+0.18\%$
test_serialize_weights_pickle	0.4568s	0.4102s	2.4380 Ops/s	2.5577 Ops/s	$\color{#d91a1a}-4.68\%$
test_serialize_weights_filesystem	0.1513s	0.1421s	7.0357 Ops/s	6.9657 Ops/s	$\color{#35bf28}+1.01\%$
test_serialize_model_filesystem	0.2606s	0.1654s	6.0468 Ops/s	6.7212 Ops/s	$\textbf{\color{#d91a1a}-10.03\%}$
test_reshape_pytree	0.1191ms	26.3517μs	37.9482 KOps/s	37.4094 KOps/s	$\color{#35bf28}+1.44\%$
test_reshape_td	72.3260μs	32.0905μs	31.1619 KOps/s	30.4443 KOps/s	$\color{#35bf28}+2.36\%$
test_view_pytree	69.3700μs	26.3928μs	37.8892 KOps/s	37.4912 KOps/s	$\color{#35bf28}+1.06\%$
test_view_td	83.2860μs	37.5643μs	26.6210 KOps/s	25.7954 KOps/s	$\color{#35bf28}+3.20\%$
test_unbind_pytree	0.1096ms	30.1022μs	33.2202 KOps/s	33.3135 KOps/s	$\color{#d91a1a}-0.28\%$
test_unbind_td	0.3464ms	37.9110μs	26.3775 KOps/s	25.8487 KOps/s	$\color{#35bf28}+2.05\%$
test_split_pytree	62.2270μs	29.4853μs	33.9152 KOps/s	33.9290 KOps/s	$\color{#d91a1a}-0.04\%$
test_split_td	0.2040ms	43.6623μs	22.9030 KOps/s	22.5529 KOps/s	$\color{#35bf28}+1.55\%$
test_add_pytree	85.9710μs	35.4832μs	28.1823 KOps/s	27.5246 KOps/s	$\color{#35bf28}+2.39\%$
test_add_td	0.1144ms	55.6459μs	17.9708 KOps/s	17.5167 KOps/s	$\color{#35bf28}+2.59\%$
test_compile_add_one_nested[tensordict-compile]	0.1422ms	61.3319μs	16.3047 KOps/s	15.9840 KOps/s	$\color{#35bf28}+2.01\%$
test_compile_add_one_nested[tensordict-eager]	0.4050ms	0.1673ms	5.9787 KOps/s	6.1858 KOps/s	$\color{#d91a1a}-3.35\%$
test_compile_add_one_nested[pytree-compile]	0.1476ms	45.9555μs	21.7602 KOps/s	22.1211 KOps/s	$\color{#d91a1a}-1.63\%$
test_compile_add_one_nested[pytree-eager]	0.2628ms	0.1188ms	8.4166 KOps/s	8.4135 KOps/s	$\color{#35bf28}+0.04\%$
test_compile_copy_nested[tensordict-compile]	75.3110μs	25.8805μs	38.6392 KOps/s	37.8773 KOps/s	$\color{#35bf28}+2.01\%$
test_compile_copy_nested[tensordict-eager]	0.1120ms	53.6428μs	18.6418 KOps/s	18.5884 KOps/s	$\color{#35bf28}+0.29\%$
test_compile_copy_nested[pytree-compile]	0.1796ms	78.2829μs	12.7742 KOps/s	12.7077 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_copy_nested[pytree-eager]	0.1364ms	67.0311μs	14.9184 KOps/s	14.7171 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_add_one_flat[tensordict-compile]	0.1872ms	0.1043ms	9.5866 KOps/s	9.4563 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_add_one_flat[tensordict-eager]	0.3660ms	0.1995ms	5.0114 KOps/s	4.9765 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_add_one_flat[tensorclass-compile]	0.1029ms	44.6443μs	22.3993 KOps/s	21.9030 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_add_one_flat[tensorclass-eager]	0.5123ms	61.8032μs	16.1804 KOps/s	16.1781 KOps/s	$\color{#35bf28}+0.01\%$
test_compile_add_one_flat[pytree-compile]	0.1971ms	0.1029ms	9.7207 KOps/s	9.4894 KOps/s	$\color{#35bf28}+2.44\%$
test_compile_add_one_flat[pytree-eager]	0.3762ms	0.2002ms	4.9944 KOps/s	4.9569 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_add_self_flat[tensordict-eager]	0.3366ms	0.2106ms	4.7480 KOps/s	4.7100 KOps/s	$\color{#35bf28}+0.81\%$
test_compile_add_self_flat[tensordict-compile]	0.1889ms	0.1066ms	9.3778 KOps/s	9.2089 KOps/s	$\color{#35bf28}+1.83\%$
test_compile_add_self_flat[tensorclass-eager]	0.2236ms	58.0562μs	17.2247 KOps/s	18.3009 KOps/s	$\textbf{\color{#d91a1a}-5.88\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1127ms	44.8077μs	22.3176 KOps/s	20.7523 KOps/s	$\textbf{\color{#35bf28}+7.54\%}$
test_compile_add_self_flat[pytree-eager]	0.6101ms	0.1581ms	6.3263 KOps/s	6.2052 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_add_self_flat[pytree-compile]	0.1700ms	0.1027ms	9.7334 KOps/s	9.6687 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_copy_flat[tensordict-compile]	55.0940μs	20.8965μs	47.8548 KOps/s	47.3306 KOps/s	$\color{#35bf28}+1.11\%$
test_compile_copy_flat[tensordict-eager]	0.1352ms	58.8791μs	16.9840 KOps/s	16.8269 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_copy_flat[pytree-compile]	0.1764ms	81.8096μs	12.2235 KOps/s	12.1961 KOps/s	$\color{#35bf28}+0.22\%$
test_compile_copy_flat[pytree-eager]	0.1453ms	69.6705μs	14.3533 KOps/s	14.2501 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_assign_and_add[tensordict-compile]	0.3921ms	0.2081ms	4.8056 KOps/s	4.6444 KOps/s	$\color{#35bf28}+3.47\%$
test_compile_assign_and_add[tensordict-eager]	1.4909ms	1.2494ms	800.4108 Ops/s	789.8082 Ops/s	$\color{#35bf28}+1.34\%$
test_compile_assign_and_add[pytree-compile]	0.3093ms	0.2025ms	4.9383 KOps/s	4.9243 KOps/s	$\color{#35bf28}+0.29\%$
test_compile_assign_and_add[pytree-eager]	1.6025ms	0.7801ms	1.2819 KOps/s	1.2712 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_assign_and_add_stack[compile]	0.5714ms	0.4539ms	2.2034 KOps/s	2.1759 KOps/s	$\color{#35bf28}+1.26\%$
test_compile_assign_and_add_stack[eager]	2.7637ms	2.5198ms	396.8607 Ops/s	384.7153 Ops/s	$\color{#35bf28}+3.16\%$
test_compile_indexing[tensor-tensordict-compile]	87.5540μs	35.2455μs	28.3725 KOps/s	26.9600 KOps/s	$\textbf{\color{#35bf28}+5.24\%}$
test_compile_indexing[tensor-tensordict-eager]	0.4105ms	31.6198μs	31.6258 KOps/s	29.9443 KOps/s	$\textbf{\color{#35bf28}+5.62\%}$
test_compile_indexing[tensor-tensorclass-compile]	83.5160μs	28.9805μs	34.5060 KOps/s	33.7462 KOps/s	$\color{#35bf28}+2.25\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1865ms	22.9849μs	43.5069 KOps/s	41.2601 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_compile_indexing[tensor-pytree-compile]	68.3380μs	30.0759μs	33.2492 KOps/s	32.4235 KOps/s	$\color{#35bf28}+2.55\%$
test_compile_indexing[tensor-pytree-eager]	73.6080μs	22.9085μs	43.6520 KOps/s	41.1097 KOps/s	$\textbf{\color{#35bf28}+6.18\%}$
test_compile_indexing[slice-tensordict-compile]	0.1185ms	49.6483μs	20.1417 KOps/s	19.1225 KOps/s	$\textbf{\color{#35bf28}+5.33\%}$
test_compile_indexing[slice-tensordict-eager]	0.5203ms	18.8873μs	52.9457 KOps/s	50.5193 KOps/s	$\color{#35bf28}+4.80\%$
test_compile_indexing[slice-tensorclass-compile]	0.1333ms	43.5093μs	22.9836 KOps/s	21.8799 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_compile_indexing[slice-tensorclass-eager]	64.2100μs	18.3684μs	54.4412 KOps/s	52.4758 KOps/s	$\color{#35bf28}+3.75\%$
test_compile_indexing[slice-pytree-compile]	99.4360μs	43.7949μs	22.8337 KOps/s	21.4955 KOps/s	$\textbf{\color{#35bf28}+6.23\%}$
test_compile_indexing[slice-pytree-eager]	59.1510μs	18.4303μs	54.2586 KOps/s	52.3100 KOps/s	$\color{#35bf28}+3.73\%$
test_compile_indexing[int-tensordict-compile]	0.1265ms	50.9734μs	19.6181 KOps/s	18.7274 KOps/s	$\color{#35bf28}+4.76\%$
test_compile_indexing[int-tensordict-eager]	0.8708ms	18.5203μs	53.9948 KOps/s	50.4822 KOps/s	$\textbf{\color{#35bf28}+6.96\%}$
test_compile_indexing[int-tensorclass-compile]	0.1154ms	43.8121μs	22.8247 KOps/s	21.7252 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_compile_indexing[int-tensorclass-eager]	58.6600μs	18.4551μs	54.1855 KOps/s	52.6002 KOps/s	$\color{#35bf28}+3.01\%$
test_compile_indexing[int-pytree-compile]	89.3580μs	43.2749μs	23.1081 KOps/s	21.9584 KOps/s	$\textbf{\color{#35bf28}+5.24\%}$
test_compile_indexing[int-pytree-eager]	62.7970μs	18.3769μs	54.4163 KOps/s	53.0553 KOps/s	$\color{#35bf28}+2.57\%$
test_mod_add[eager]	0.1078ms	26.9439μs	37.1141 KOps/s	37.6852 KOps/s	$\color{#d91a1a}-1.52\%$
test_mod_add[compile]	0.1018ms	44.3142μs	22.5661 KOps/s	21.9738 KOps/s	$\color{#35bf28}+2.70\%$
test_mod_add[compile-overhead]	0.1041ms	45.0285μs	22.2081 KOps/s	21.7722 KOps/s	$\color{#35bf28}+2.00\%$
test_mod_wrap[eager]	0.3499ms	0.2096ms	4.7721 KOps/s	4.6076 KOps/s	$\color{#35bf28}+3.57\%$
test_mod_wrap[compile]	1.8212ms	0.2016ms	4.9606 KOps/s	4.7968 KOps/s	$\color{#35bf28}+3.42\%$
test_mod_wrap[compile-overhead]	1.9572ms	0.2014ms	4.9658 KOps/s	4.8348 KOps/s	$\color{#35bf28}+2.71\%$
test_mod_wrap_and_backward[eager]	13.5530ms	11.1780ms	89.4611 Ops/s	76.4549 Ops/s	$\textbf{\color{#35bf28}+17.01\%}$
test_mod_wrap_and_backward[compile]	17.3539ms	11.0537ms	90.4671 Ops/s	78.3695 Ops/s	$\textbf{\color{#35bf28}+15.44\%}$
test_mod_wrap_and_backward[compile-overhead]	13.1500ms	11.2558ms	88.8434 Ops/s	75.7940 Ops/s	$\textbf{\color{#35bf28}+17.22\%}$
test_seq_add[eager]	0.1714ms	92.2369μs	10.8416 KOps/s	10.7955 KOps/s	$\color{#35bf28}+0.43\%$
test_seq_add[compile]	0.1449ms	60.2183μs	16.6062 KOps/s	16.3625 KOps/s	$\color{#35bf28}+1.49\%$
test_seq_add[compile-overhead]	0.1276ms	58.4189μs	17.1178 KOps/s	16.8531 KOps/s	$\color{#35bf28}+1.57\%$
test_seq_wrap[eager]	0.6683ms	0.3837ms	2.6060 KOps/s	2.4920 KOps/s	$\color{#35bf28}+4.58\%$
test_seq_wrap[compile]	0.4340ms	0.2259ms	4.4266 KOps/s	4.4099 KOps/s	$\color{#35bf28}+0.38\%$
test_seq_wrap[compile-overhead]	0.4233ms	0.2239ms	4.4660 KOps/s	4.4358 KOps/s	$\color{#35bf28}+0.68\%$
test_func_call_runtime[False-eager]	0.8721ms	0.5283ms	1.8929 KOps/s	1.8300 KOps/s	$\color{#35bf28}+3.44\%$
test_func_call_runtime[False-compile]	0.5765ms	0.4262ms	2.3465 KOps/s	2.3210 KOps/s	$\color{#35bf28}+1.10\%$
test_func_call_runtime[False-compile-overhead]	0.5648ms	0.4251ms	2.3525 KOps/s	2.3149 KOps/s	$\color{#35bf28}+1.63\%$
test_func_call_runtime[True-eager]	1.2102ms	0.7398ms	1.3517 KOps/s	1.3157 KOps/s	$\color{#35bf28}+2.73\%$
test_func_call_runtime[True-compile]	0.7899ms	0.4738ms	2.1107 KOps/s	2.1192 KOps/s	$\color{#d91a1a}-0.40\%$
test_func_call_runtime[True-compile-overhead]	0.8990ms	0.4692ms	2.1314 KOps/s	2.1264 KOps/s	$\color{#35bf28}+0.23\%$
test_func_call_cm_runtime[False-eager]	0.9439ms	0.5388ms	1.8561 KOps/s	1.8325 KOps/s	$\color{#35bf28}+1.29\%$
test_func_call_cm_runtime[False-compile]	0.5460ms	0.4250ms	2.3532 KOps/s	2.3123 KOps/s	$\color{#35bf28}+1.77\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5754ms	0.4257ms	2.3491 KOps/s	2.2882 KOps/s	$\color{#35bf28}+2.66\%$
test_func_call_cm_runtime[True-eager]	1.4736ms	0.8810ms	1.1351 KOps/s	1.1122 KOps/s	$\color{#35bf28}+2.06\%$
test_func_call_cm_runtime[True-compile]	0.5978ms	0.4914ms	2.0352 KOps/s	2.0008 KOps/s	$\color{#35bf28}+1.72\%$
test_func_call_cm_runtime[True-compile-overhead]	0.8970ms	0.4918ms	2.0335 KOps/s	2.0332 KOps/s	$\color{#35bf28}+0.02\%$
test_vmap_func_call_cm_runtime[eager]	3.5001ms	1.8879ms	529.6839 Ops/s	521.6601 Ops/s	$\color{#35bf28}+1.54\%$
test_vmap_func_call_cm_runtime[compile]	0.8958ms	0.5143ms	1.9444 KOps/s	1.8695 KOps/s	$\color{#35bf28}+4.01\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.0564ms	0.5255ms	1.9030 KOps/s	1.9091 KOps/s	$\color{#d91a1a}-0.32\%$
test_distributed	0.2686ms	0.1258ms	7.9474 KOps/s	7.6982 KOps/s	$\color{#35bf28}+3.24\%$
test_tdmodule	32.4410μs	18.4770μs	54.1213 KOps/s	51.2943 KOps/s	$\textbf{\color{#35bf28}+5.51\%}$
test_tdmodule_dispatch	62.3260μs	36.2402μs	27.5936 KOps/s	26.7589 KOps/s	$\color{#35bf28}+3.12\%$
test_tdseq	55.8050μs	21.6148μs	46.2646 KOps/s	46.3607 KOps/s	$\color{#d91a1a}-0.21\%$
test_tdseq_dispatch	79.9200μs	43.5553μs	22.9593 KOps/s	23.1576 KOps/s	$\color{#d91a1a}-0.86\%$
test_instantiation_functorch	1.6736ms	1.5584ms	641.6989 Ops/s	633.2166 Ops/s	$\color{#35bf28}+1.34\%$
test_exec_functorch	0.2668ms	0.1761ms	5.6792 KOps/s	5.5001 KOps/s	$\color{#35bf28}+3.26\%$
test_exec_functional_call	0.4144ms	0.1733ms	5.7707 KOps/s	5.7071 KOps/s	$\color{#35bf28}+1.12\%$
test_exec_td_decorator	0.4823ms	0.2276ms	4.3931 KOps/s	4.3695 KOps/s	$\color{#35bf28}+0.54\%$
test_vmap_mlp_speed_decorator[True-True]	0.8711ms	0.6327ms	1.5805 KOps/s	1.5522 KOps/s	$\color{#35bf28}+1.83\%$
test_vmap_mlp_speed_decorator[True-False]	0.9801ms	0.6333ms	1.5789 KOps/s	1.5402 KOps/s	$\color{#35bf28}+2.52\%$
test_vmap_mlp_speed_decorator[False-True]	0.7029ms	0.5190ms	1.9266 KOps/s	1.8899 KOps/s	$\color{#35bf28}+1.94\%$
test_vmap_mlp_speed_decorator[False-False]	1.4086ms	0.5315ms	1.8813 KOps/s	1.8784 KOps/s	$\color{#35bf28}+0.16\%$
test_to_module_speed[True]	2.0945ms	1.2998ms	769.3778 Ops/s	781.5742 Ops/s	$\color{#d91a1a}-1.56\%$
test_to_module_speed[False]	1.7644ms	1.2620ms	792.4167 Ops/s	794.8850 Ops/s	$\color{#d91a1a}-0.31\%$
test_tc_init	0.1037ms	45.9499μs	21.7628 KOps/s	23.0025 KOps/s	$\textbf{\color{#d91a1a}-5.39\%}$
test_tc_init_nested	0.2333ms	92.5546μs	10.8044 KOps/s	11.6525 KOps/s	$\textbf{\color{#d91a1a}-7.28\%}$
test_tc_first_layer_tensor	21.2900μs	1.5038μs	665.0039 KOps/s	647.2513 KOps/s	$\color{#35bf28}+2.74\%$
test_tc_first_layer_nontensor	44.7290μs	4.6920μs	213.1301 KOps/s	215.4580 KOps/s	$\color{#d91a1a}-1.08\%$
test_tc_second_layer_tensor	42.4500μs	2.8402μs	352.0846 KOps/s	364.3454 KOps/s	$\color{#d91a1a}-3.37\%$
test_tc_second_layer_nontensor	46.1960μs	6.0868μs	164.2887 KOps/s	168.4869 KOps/s	$\color{#d91a1a}-2.49\%$
test_unbind	0.2216s	12.4748ms	80.1614 Ops/s	80.8818 Ops/s	$\color{#d91a1a}-0.89\%$
test_full_like	8.1148ms	7.3663ms	135.7539 Ops/s	125.2250 Ops/s	$\textbf{\color{#35bf28}+8.41\%}$
test_zeros_like	13.4827ms	7.2898ms	137.1778 Ops/s	332.2244 Ops/s	$\textbf{\color{#d91a1a}-58.71\%}$
test_ones_like	12.0366ms	7.8086ms	128.0634 Ops/s	278.2304 Ops/s	$\textbf{\color{#d91a1a}-53.97\%}$
test_clone	11.9199ms	9.2061ms	108.6238 Ops/s	171.3595 Ops/s	$\textbf{\color{#d91a1a}-36.61\%}$
test_squeeze	70.2810μs	11.8871μs	84.1246 KOps/s	84.5594 KOps/s	$\color{#d91a1a}-0.51\%$
test_unsqueeze	0.1520ms	87.2083μs	11.4668 KOps/s	11.4730 KOps/s	$\color{#d91a1a}-0.05\%$
test_split	0.5123ms	0.1888ms	5.2953 KOps/s	5.4318 KOps/s	$\color{#d91a1a}-2.51\%$
test_permute	0.3001ms	0.2172ms	4.6037 KOps/s	4.6293 KOps/s	$\color{#d91a1a}-0.55\%$
test_stack	33.5248ms	26.5465ms	37.6697 Ops/s	37.4051 Ops/s	$\color{#35bf28}+0.71\%$
test_cat	28.1857ms	25.8688ms	38.6566 Ops/s	37.9417 Ops/s	$\color{#35bf28}+1.88\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	32.0200μs	10.8341μs	92.3013 KOps/s	91.0731 KOps/s	$\color{#35bf28}+1.35\%$
test_plain_set_stack_nested	38.8610μs	10.9370μs	91.4328 KOps/s	90.9145 KOps/s	$\color{#35bf28}+0.57\%$
test_plain_set_nested_inplace	47.9900μs	11.6749μs	85.6542 KOps/s	84.2380 KOps/s	$\color{#35bf28}+1.68\%$
test_plain_set_stack_nested_inplace	47.0410μs	11.6561μs	85.7923 KOps/s	84.5530 KOps/s	$\color{#35bf28}+1.47\%$
test_items	36.5310μs	2.8931μs	345.6503 KOps/s	334.7520 KOps/s	$\color{#35bf28}+3.26\%$
test_items_nested	0.4450ms	0.3191ms	3.1337 KOps/s	3.0895 KOps/s	$\color{#35bf28}+1.43\%$
test_items_nested_locked	0.4276ms	0.3237ms	3.0889 KOps/s	3.0764 KOps/s	$\color{#35bf28}+0.41\%$
test_items_nested_leaf	91.4320μs	58.2296μs	17.1734 KOps/s	17.0444 KOps/s	$\color{#35bf28}+0.76\%$
test_items_stack_nested	0.4011ms	0.3265ms	3.0624 KOps/s	3.0817 KOps/s	$\color{#d91a1a}-0.62\%$
test_items_stack_nested_leaf	94.6120μs	59.8025μs	16.7217 KOps/s	16.7817 KOps/s	$\color{#d91a1a}-0.36\%$
test_items_stack_nested_locked	0.3804ms	0.3252ms	3.0755 KOps/s	3.0680 KOps/s	$\color{#35bf28}+0.24\%$
test_keys	26.7210μs	3.4906μs	286.4805 KOps/s	287.5216 KOps/s	$\color{#d91a1a}-0.36\%$
test_keys_nested	0.1057ms	70.4266μs	14.1992 KOps/s	14.1832 KOps/s	$\color{#35bf28}+0.11\%$
test_keys_nested_locked	2.4500ms	75.7644μs	13.1988 KOps/s	13.1114 KOps/s	$\color{#35bf28}+0.67\%$
test_keys_nested_leaf	0.1095ms	61.5814μs	16.2387 KOps/s	16.1354 KOps/s	$\color{#35bf28}+0.64\%$
test_keys_stack_nested	0.1126ms	71.5564μs	13.9750 KOps/s	13.9389 KOps/s	$\color{#35bf28}+0.26\%$
test_keys_stack_nested_leaf	0.1289ms	61.4393μs	16.2762 KOps/s	15.9005 KOps/s	$\color{#35bf28}+2.36\%$
test_keys_stack_nested_locked	0.1087ms	76.9005μs	13.0038 KOps/s	13.0151 KOps/s	$\color{#d91a1a}-0.09\%$
test_values	5.2318μs	0.9588μs	1.0429 MOps/s	1.1710 MOps/s	$\textbf{\color{#d91a1a}-10.94\%}$
test_values_nested	60.0210μs	31.5625μs	31.6831 KOps/s	32.0174 KOps/s	$\color{#d91a1a}-1.04\%$
test_values_nested_locked	56.6010μs	33.0631μs	30.2452 KOps/s	30.3284 KOps/s	$\color{#d91a1a}-0.27\%$
test_values_nested_leaf	71.5010μs	33.8253μs	29.5637 KOps/s	29.6218 KOps/s	$\color{#d91a1a}-0.20\%$
test_values_stack_nested	63.0610μs	31.9664μs	31.2829 KOps/s	31.4511 KOps/s	$\color{#d91a1a}-0.53\%$
test_values_stack_nested_leaf	70.3110μs	34.2585μs	29.1898 KOps/s	29.1154 KOps/s	$\color{#35bf28}+0.26\%$
test_values_stack_nested_locked	64.4710μs	33.3500μs	29.9850 KOps/s	30.1007 KOps/s	$\color{#d91a1a}-0.38\%$
test_membership	1.8070μs	0.5158μs	1.9388 MOps/s	1.9600 MOps/s	$\color{#d91a1a}-1.08\%$
test_membership_nested	30.9310μs	1.9799μs	505.0883 KOps/s	511.6528 KOps/s	$\color{#d91a1a}-1.28\%$
test_membership_nested_leaf	21.1255μs	1.9109μs	523.3201 KOps/s	502.6116 KOps/s	$\color{#35bf28}+4.12\%$
test_membership_stacked_nested	49.5110μs	1.9877μs	503.1009 KOps/s	493.3918 KOps/s	$\color{#35bf28}+1.97\%$
test_membership_stacked_nested_leaf	36.4000μs	1.9959μs	501.0245 KOps/s	490.0518 KOps/s	$\color{#35bf28}+2.24\%$
test_membership_nested_last	28.1100μs	2.8270μs	353.7291 KOps/s	344.5751 KOps/s	$\color{#35bf28}+2.66\%$
test_membership_nested_leaf_last	29.6810μs	2.8163μs	355.0728 KOps/s	344.4415 KOps/s	$\color{#35bf28}+3.09\%$
test_membership_stacked_nested_last	29.4510μs	3.2877μs	304.1637 KOps/s	258.8179 KOps/s	$\textbf{\color{#35bf28}+17.52\%}$
test_membership_stacked_nested_leaf_last	26.7910μs	3.2691μs	305.8934 KOps/s	258.5847 KOps/s	$\textbf{\color{#35bf28}+18.30\%}$
test_nested_getleaf	37.6610μs	6.0009μs	166.6419 KOps/s	164.9067 KOps/s	$\color{#35bf28}+1.05\%$
test_nested_get	33.8510μs	5.7290μs	174.5498 KOps/s	173.5002 KOps/s	$\color{#35bf28}+0.60\%$
test_stacked_getleaf	30.3200μs	6.0295μs	165.8504 KOps/s	165.4444 KOps/s	$\color{#35bf28}+0.25\%$
test_stacked_get	56.8310μs	5.7281μs	174.5766 KOps/s	174.1822 KOps/s	$\color{#35bf28}+0.23\%$
test_nested_getitemleaf	28.5010μs	6.1214μs	163.3620 KOps/s	161.5854 KOps/s	$\color{#35bf28}+1.10\%$
test_nested_getitem	44.0710μs	5.8635μs	170.5457 KOps/s	172.4376 KOps/s	$\color{#d91a1a}-1.10\%$
test_stacked_getitemleaf	33.8300μs	6.1304μs	163.1209 KOps/s	162.3490 KOps/s	$\color{#35bf28}+0.48\%$
test_stacked_getitem	29.7410μs	5.8568μs	170.7418 KOps/s	171.3479 KOps/s	$\color{#d91a1a}-0.35\%$
test_lock_nested	4.5930ms	0.3688ms	2.7113 KOps/s	2.7208 KOps/s	$\color{#d91a1a}-0.35\%$
test_lock_stack_nested	0.3973ms	0.3357ms	2.9785 KOps/s	2.9655 KOps/s	$\color{#35bf28}+0.44\%$
test_unlock_nested	0.6857ms	0.3044ms	3.2850 KOps/s	3.2703 KOps/s	$\color{#35bf28}+0.45\%$
test_unlock_stack_nested	0.3545ms	0.2751ms	3.6356 KOps/s	3.6262 KOps/s	$\color{#35bf28}+0.26\%$
test_flatten_speed	0.1197ms	73.8689μs	13.5375 KOps/s	13.6676 KOps/s	$\color{#d91a1a}-0.95\%$
test_unflatten_speed	0.3493ms	0.2944ms	3.3964 KOps/s	3.3816 KOps/s	$\color{#35bf28}+0.44\%$
test_common_ops	1.6592ms	0.5780ms	1.7300 KOps/s	1.6925 KOps/s	$\color{#35bf28}+2.22\%$
test_creation	92.0420μs	1.5173μs	659.0740 KOps/s	661.5716 KOps/s	$\color{#d91a1a}-0.38\%$
test_creation_empty	30.0310μs	7.7693μs	128.7115 KOps/s	120.8950 KOps/s	$\textbf{\color{#35bf28}+6.47\%}$
test_creation_nested_1	37.2210μs	9.3304μs	107.1771 KOps/s	103.7416 KOps/s	$\color{#35bf28}+3.31\%$
test_creation_nested_2	35.6210μs	11.6439μs	85.8820 KOps/s	82.7235 KOps/s	$\color{#35bf28}+3.82\%$
test_clone	53.5910μs	9.7163μs	102.9195 KOps/s	100.1244 KOps/s	$\color{#35bf28}+2.79\%$
test_getitem[int]	1.3633ms	10.4913μs	95.3172 KOps/s	95.3759 KOps/s	$\color{#d91a1a}-0.06\%$
test_getitem[slice_int]	0.1259ms	20.8386μs	47.9880 KOps/s	48.0447 KOps/s	$\color{#d91a1a}-0.12\%$
test_getitem[range]	0.1440ms	36.0263μs	27.7575 KOps/s	26.6080 KOps/s	$\color{#35bf28}+4.32\%$
test_getitem[tuple]	0.1128ms	17.7204μs	56.4323 KOps/s	54.5372 KOps/s	$\color{#35bf28}+3.47\%$
test_getitem[list]	0.1591ms	32.5077μs	30.7620 KOps/s	30.5544 KOps/s	$\color{#35bf28}+0.68\%$
test_setitem_dim[int]	40.1610μs	18.1177μs	55.1946 KOps/s	55.2536 KOps/s	$\color{#d91a1a}-0.11\%$
test_setitem_dim[slice_int]	57.0510μs	35.8959μs	27.8583 KOps/s	27.6969 KOps/s	$\color{#35bf28}+0.58\%$
test_setitem_dim[range]	77.1520μs	51.5301μs	19.4061 KOps/s	19.6142 KOps/s	$\color{#d91a1a}-1.06\%$
test_setitem_dim[tuple]	52.6910μs	31.0188μs	32.2385 KOps/s	33.2283 KOps/s	$\color{#d91a1a}-2.98\%$
test_setitem	58.9610μs	13.9585μs	71.6409 KOps/s	68.5972 KOps/s	$\color{#35bf28}+4.44\%$
test_set	53.5710μs	13.4372μs	74.4202 KOps/s	71.7947 KOps/s	$\color{#35bf28}+3.66\%$
test_set_shared	95.1322ms	0.1679ms	5.9551 KOps/s	6.8437 KOps/s	$\textbf{\color{#d91a1a}-12.98\%}$
test_update	0.7608ms	16.1883μs	61.7732 KOps/s	59.1683 KOps/s	$\color{#35bf28}+4.40\%$
test_update_nested	96.0220μs	20.5472μs	48.6685 KOps/s	44.9873 KOps/s	$\textbf{\color{#35bf28}+8.18\%}$
test_update__nested	0.7312ms	23.0295μs	43.4226 KOps/s	41.6236 KOps/s	$\color{#35bf28}+4.32\%$
test_set_nested	89.3520μs	14.7731μs	67.6906 KOps/s	65.4407 KOps/s	$\color{#35bf28}+3.44\%$
test_set_nested_new	94.3620μs	16.6749μs	59.9704 KOps/s	56.5870 KOps/s	$\textbf{\color{#35bf28}+5.98\%}$
test_select	96.8120μs	28.9273μs	34.5695 KOps/s	33.8690 KOps/s	$\color{#35bf28}+2.07\%$
test_select_nested	81.7920μs	41.7221μs	23.9681 KOps/s	23.6297 KOps/s	$\color{#35bf28}+1.43\%$
test_exclude_nested	95.0720μs	60.1591μs	16.6226 KOps/s	16.7344 KOps/s	$\color{#d91a1a}-0.67\%$
test_empty[True]	0.3102ms	0.2559ms	3.9076 KOps/s	3.8844 KOps/s	$\color{#35bf28}+0.60\%$
test_empty[False]	3.8231μs	0.7625μs	1.3115 MOps/s	1.3344 MOps/s	$\color{#d91a1a}-1.71\%$
test_to	84.7210μs	55.2867μs	18.0875 KOps/s	17.9844 KOps/s	$\color{#35bf28}+0.57\%$
test_to_nonblocking	89.3210μs	46.3464μs	21.5767 KOps/s	21.4103 KOps/s	$\color{#35bf28}+0.78\%$
test_unbind_speed	0.2791ms	0.2276ms	4.3940 KOps/s	4.2935 KOps/s	$\color{#35bf28}+2.34\%$
test_unbind_speed_stack0	0.2822ms	0.2283ms	4.3794 KOps/s	4.3042 KOps/s	$\color{#35bf28}+1.75\%$
test_unbind_speed_stack1	92.9907ms	0.6379ms	1.5676 KOps/s	1.5532 KOps/s	$\color{#35bf28}+0.93\%$
test_split	92.9573ms	1.5672ms	638.0719 Ops/s	570.2894 Ops/s	$\textbf{\color{#35bf28}+11.89\%}$
test_chunk	95.1249ms	1.7190ms	581.7213 Ops/s	676.2862 Ops/s	$\textbf{\color{#d91a1a}-13.98\%}$
test_consolidate[False-None]	2.9052ms	2.6353ms	379.4679 Ops/s	348.8049 Ops/s	$\textbf{\color{#35bf28}+8.79\%}$
test_consolidate[default-None]	1.7921ms	1.6390ms	610.1321 Ops/s	605.9145 Ops/s	$\color{#35bf28}+0.70\%$
test_consolidate[reduce-overhead-None]	1.7973ms	1.6835ms	593.9988 Ops/s	592.8981 Ops/s	$\color{#35bf28}+0.19\%$
test_consolidate_njt[False-None]	7.2663ms	6.8083ms	146.8799 Ops/s	150.7639 Ops/s	$\color{#d91a1a}-2.58\%$
test_to[False-False-None]	1.7685ms	1.6664ms	600.1009 Ops/s	596.0614 Ops/s	$\color{#35bf28}+0.68\%$
test_to[True-False-None]	1.5144ms	1.2802ms	781.1162 Ops/s	776.3545 Ops/s	$\color{#35bf28}+0.61\%$
test_to[within-False-None]	4.1784ms	4.0098ms	249.3885 Ops/s	250.7492 Ops/s	$\color{#d91a1a}-0.54\%$
test_to[True-default-None]	5.6241ms	5.1395ms	194.5733 Ops/s	194.9543 Ops/s	$\color{#d91a1a}-0.20\%$
test_to_njt[False-False-None]	7.1933ms	7.0491ms	141.8623 Ops/s	141.7262 Ops/s	$\color{#35bf28}+0.10\%$
test_to_njt[True-False-None]	5.8269ms	5.6355ms	177.4463 Ops/s	178.2475 Ops/s	$\color{#d91a1a}-0.45\%$
test_to_njt[within-False-None]	13.0966ms	12.4985ms	80.0093 Ops/s	80.8149 Ops/s	$\color{#d91a1a}-1.00\%$
test_creation[device0]	0.4575ms	79.7422μs	12.5404 KOps/s	12.4096 KOps/s	$\color{#35bf28}+1.05\%$
test_creation_from_tensor	0.6217ms	82.6521μs	12.0989 KOps/s	12.0153 KOps/s	$\color{#35bf28}+0.70\%$
test_add_one[memmap_tensor0]	0.2467ms	6.1770μs	161.8902 KOps/s	156.5296 KOps/s	$\color{#35bf28}+3.42\%$
test_contiguous[memmap_tensor0]	1.8900μs	0.4264μs	2.3455 MOps/s	2.3363 MOps/s	$\color{#35bf28}+0.39\%$
test_stack[memmap_tensor0]	35.6010μs	4.3479μs	229.9954 KOps/s	218.0642 KOps/s	$\textbf{\color{#35bf28}+5.47\%}$
test_memmaptd_index	2.1969ms	0.2463ms	4.0599 KOps/s	4.0360 KOps/s	$\color{#35bf28}+0.59\%$
test_memmaptd_index_astensor	0.6062ms	0.3058ms	3.2697 KOps/s	3.2350 KOps/s	$\color{#35bf28}+1.07\%$
test_memmaptd_index_op	0.9951ms	0.5600ms	1.7857 KOps/s	1.7350 KOps/s	$\color{#35bf28}+2.92\%$
test_serialize_model	0.1327s	0.1310s	7.6307 Ops/s	7.6636 Ops/s	$\color{#d91a1a}-0.43\%$
test_serialize_model_pickle	1.4010s	1.2244s	0.8167 Ops/s	0.8206 Ops/s	$\color{#d91a1a}-0.47\%$
test_serialize_weights	0.1300s	0.1294s	7.7304 Ops/s	7.6806 Ops/s	$\color{#35bf28}+0.65\%$
test_serialize_weights_returnearly	0.3332s	53.8422ms	18.5728 Ops/s	11.0706 Ops/s	$\textbf{\color{#35bf28}+67.77\%}$
test_serialize_weights_pickle	1.3472s	1.2161s	0.8223 Ops/s	0.8205 Ops/s	$\color{#35bf28}+0.21\%$
test_reshape_pytree	64.6610μs	22.8348μs	43.7927 KOps/s	42.4807 KOps/s	$\color{#35bf28}+3.09\%$
test_reshape_td	61.9510μs	26.8656μs	37.2223 KOps/s	36.2811 KOps/s	$\color{#35bf28}+2.59\%$
test_view_pytree	55.6600μs	22.9872μs	43.5025 KOps/s	42.8697 KOps/s	$\color{#35bf28}+1.48\%$
test_view_td	71.5310μs	31.5769μs	31.6688 KOps/s	31.6882 KOps/s	$\color{#d91a1a}-0.06\%$
test_unbind_pytree	60.4410μs	28.5475μs	35.0294 KOps/s	34.5608 KOps/s	$\color{#35bf28}+1.36\%$
test_unbind_td	0.8401ms	35.6542μs	28.0472 KOps/s	27.7419 KOps/s	$\color{#35bf28}+1.10\%$
test_split_pytree	58.1210μs	30.9030μs	32.3593 KOps/s	31.6892 KOps/s	$\color{#35bf28}+2.11\%$
test_split_td	1.0061ms	39.0282μs	25.6225 KOps/s	24.8582 KOps/s	$\color{#35bf28}+3.07\%$
test_add_pytree	69.1010μs	34.4774μs	29.0045 KOps/s	29.9933 KOps/s	$\color{#d91a1a}-3.30\%$
test_add_td	0.1436ms	45.7595μs	21.8534 KOps/s	20.5532 KOps/s	$\textbf{\color{#35bf28}+6.33\%}$
test_compile_add_one_nested[tensordict-compile]	0.1705ms	0.1212ms	8.2485 KOps/s	7.9878 KOps/s	$\color{#35bf28}+3.26\%$
test_compile_add_one_nested[tensordict-eager]	0.2260ms	0.1262ms	7.9257 KOps/s	7.9605 KOps/s	$\color{#d91a1a}-0.44\%$
test_compile_add_one_nested[pytree-compile]	0.1423ms	97.2792μs	10.2797 KOps/s	10.1752 KOps/s	$\color{#35bf28}+1.03\%$
test_compile_add_one_nested[pytree-eager]	0.2286ms	0.1523ms	6.5662 KOps/s	6.4734 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_copy_nested[tensordict-compile]	56.5010μs	23.3248μs	42.8728 KOps/s	43.3247 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_copy_nested[tensordict-eager]	56.8910μs	27.1121μs	36.8840 KOps/s	36.6113 KOps/s	$\color{#35bf28}+0.74\%$
test_compile_copy_nested[pytree-compile]	0.2324ms	68.3649μs	14.6274 KOps/s	14.5229 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_copy_nested[pytree-eager]	0.1057ms	51.8312μs	19.2934 KOps/s	18.6304 KOps/s	$\color{#35bf28}+3.56\%$
test_compile_add_one_flat[tensordict-compile]	0.2247ms	0.1448ms	6.9081 KOps/s	6.9205 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_add_one_flat[tensordict-eager]	0.2967ms	0.2101ms	4.7601 KOps/s	4.7929 KOps/s	$\color{#d91a1a}-0.68\%$
test_compile_add_one_flat[tensorclass-compile]	0.1555ms	99.0842μs	10.0924 KOps/s	10.2649 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_add_one_flat[tensorclass-eager]	0.1479ms	52.1338μs	19.1814 KOps/s	19.0068 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_add_one_flat[pytree-compile]	0.1955ms	0.1401ms	7.1358 KOps/s	7.2419 KOps/s	$\color{#d91a1a}-1.47\%$
test_compile_add_one_flat[pytree-eager]	0.5733ms	0.5005ms	1.9980 KOps/s	2.0747 KOps/s	$\color{#d91a1a}-3.70\%$
test_compile_add_self_flat[tensordict-eager]	0.3674ms	0.2507ms	3.9894 KOps/s	4.0312 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_add_self_flat[tensordict-compile]	0.2025ms	0.1498ms	6.6758 KOps/s	6.7777 KOps/s	$\color{#d91a1a}-1.50\%$
test_compile_add_self_flat[tensorclass-eager]	0.1626ms	61.9714μs	16.1365 KOps/s	15.7429 KOps/s	$\color{#35bf28}+2.50\%$
test_compile_add_self_flat[tensorclass-compile]	0.1475ms	0.1023ms	9.7719 KOps/s	9.8720 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_add_self_flat[pytree-eager]	0.4674ms	0.4051ms	2.4684 KOps/s	2.4379 KOps/s	$\color{#35bf28}+1.25\%$
test_compile_add_self_flat[pytree-compile]	0.1882ms	0.1395ms	7.1706 KOps/s	7.4134 KOps/s	$\color{#d91a1a}-3.28\%$
test_compile_copy_flat[tensordict-compile]	0.1621ms	22.1915μs	45.0622 KOps/s	51.4662 KOps/s	$\textbf{\color{#d91a1a}-12.44\%}$
test_compile_copy_flat[tensordict-eager]	59.5310μs	27.3335μs	36.5851 KOps/s	37.4351 KOps/s	$\color{#d91a1a}-2.27\%$
test_compile_copy_flat[pytree-compile]	0.1191ms	71.1449μs	14.0558 KOps/s	14.2521 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_copy_flat[pytree-eager]	79.8810μs	52.2479μs	19.1395 KOps/s	19.1693 KOps/s	$\color{#d91a1a}-0.16\%$
test_compile_assign_and_add[tensordict-compile]	1.6909ms	0.4030ms	2.4813 KOps/s	2.2117 KOps/s	$\textbf{\color{#35bf28}+12.19\%}$
test_compile_assign_and_add[tensordict-eager]	2.8403ms	2.6748ms	373.8579 Ops/s	382.9761 Ops/s	$\color{#d91a1a}-2.38\%$
test_compile_assign_and_add[pytree-compile]	1.6043ms	0.4337ms	2.3059 KOps/s	2.2560 KOps/s	$\color{#35bf28}+2.21\%$
test_compile_assign_and_add[pytree-eager]	2.7588ms	2.6460ms	377.9349 Ops/s	379.9320 Ops/s	$\color{#d91a1a}-0.53\%$
test_compile_indexing[tensor-tensordict-compile]	0.4770ms	0.1133ms	8.8253 KOps/s	8.9525 KOps/s	$\color{#d91a1a}-1.42\%$
test_compile_indexing[tensor-tensordict-eager]	0.5642ms	80.9370μs	12.3553 KOps/s	12.4935 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_indexing[tensor-tensorclass-compile]	0.3679ms	0.1090ms	9.1749 KOps/s	9.6674 KOps/s	$\textbf{\color{#d91a1a}-5.09\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.1675ms	68.9379μs	14.5058 KOps/s	14.5760 KOps/s	$\color{#d91a1a}-0.48\%$
test_compile_indexing[tensor-pytree-compile]	0.1708ms	0.1120ms	8.9270 KOps/s	9.6054 KOps/s	$\textbf{\color{#d91a1a}-7.06\%}$
test_compile_indexing[tensor-pytree-eager]	0.1162ms	70.6787μs	14.1485 KOps/s	14.4922 KOps/s	$\color{#d91a1a}-2.37\%$
test_compile_indexing[slice-tensordict-compile]	0.1571ms	0.1023ms	9.7755 KOps/s	9.8899 KOps/s	$\color{#d91a1a}-1.16\%$
test_compile_indexing[slice-tensordict-eager]	0.1534ms	17.1903μs	58.1725 KOps/s	55.8614 KOps/s	$\color{#35bf28}+4.14\%$
test_compile_indexing[slice-tensorclass-compile]	0.2115ms	96.6212μs	10.3497 KOps/s	10.3228 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_indexing[slice-tensorclass-eager]	54.0910μs	15.8727μs	63.0011 KOps/s	59.8500 KOps/s	$\textbf{\color{#35bf28}+5.27\%}$
test_compile_indexing[slice-pytree-compile]	0.1433ms	97.3518μs	10.2720 KOps/s	10.3105 KOps/s	$\color{#d91a1a}-0.37\%$
test_compile_indexing[slice-pytree-eager]	55.8510μs	16.0096μs	62.4626 KOps/s	61.3140 KOps/s	$\color{#35bf28}+1.87\%$
test_compile_indexing[int-tensordict-compile]	0.1553ms	0.1064ms	9.3976 KOps/s	9.8246 KOps/s	$\color{#d91a1a}-4.35\%$
test_compile_indexing[int-tensordict-eager]	0.5918ms	16.9776μs	58.9011 KOps/s	58.0107 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_indexing[int-tensorclass-compile]	0.1397ms	96.9517μs	10.3144 KOps/s	10.3201 KOps/s	$\color{#d91a1a}-0.06\%$
test_compile_indexing[int-tensorclass-eager]	83.8810μs	18.0744μs	55.3268 KOps/s	61.2786 KOps/s	$\textbf{\color{#d91a1a}-9.71\%}$
test_compile_indexing[int-pytree-compile]	0.2075ms	97.2759μs	10.2800 KOps/s	10.3580 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_indexing[int-pytree-eager]	45.2000μs	15.9839μs	62.5630 KOps/s	60.8524 KOps/s	$\color{#35bf28}+2.81\%$
test_mod_add[eager]	75.7310μs	31.3891μs	31.8582 KOps/s	30.8756 KOps/s	$\color{#35bf28}+3.18\%$
test_mod_add[compile]	0.2496ms	80.1710μs	12.4733 KOps/s	12.4697 KOps/s	$\color{#35bf28}+0.03\%$
test_mod_add[compile-overhead]	0.3187ms	0.1659ms	6.0271 KOps/s	5.8160 KOps/s	$\color{#35bf28}+3.63\%$
test_mod_wrap[eager]	0.3310ms	0.2415ms	4.1413 KOps/s	4.1111 KOps/s	$\color{#35bf28}+0.74\%$
test_mod_wrap[compile]	1.6259ms	0.2999ms	3.3343 KOps/s	3.4634 KOps/s	$\color{#d91a1a}-3.73\%$
test_mod_wrap[compile-overhead]	7.8851ms	4.1495ms	240.9947 Ops/s	242.9222 Ops/s	$\color{#d91a1a}-0.79\%$
test_mod_wrap_and_backward[eager]	1.5378ms	1.3208ms	757.1195 Ops/s	694.5154 Ops/s	$\textbf{\color{#35bf28}+9.01\%}$
test_mod_wrap_and_backward[compile]	1.4679ms	1.2922ms	773.8743 Ops/s	713.6550 Ops/s	$\textbf{\color{#35bf28}+8.44\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3630ms	0.9247ms	1.0815 KOps/s	947.2069 Ops/s	$\textbf{\color{#35bf28}+14.17\%}$
test_seq_add[eager]	0.1804ms	96.9314μs	10.3166 KOps/s	9.9037 KOps/s	$\color{#35bf28}+4.17\%$
test_seq_add[compile]	0.1975ms	94.5308μs	10.5786 KOps/s	11.3546 KOps/s	$\textbf{\color{#d91a1a}-6.83\%}$
test_seq_add[compile-overhead]	0.1852ms	0.1346ms	7.4301 KOps/s	7.6539 KOps/s	$\color{#d91a1a}-2.92\%$
test_seq_wrap[eager]	0.4656ms	0.3957ms	2.5275 KOps/s	2.4735 KOps/s	$\color{#35bf28}+2.18\%$
test_seq_wrap[compile]	0.4257ms	0.3189ms	3.1357 KOps/s	3.1225 KOps/s	$\color{#35bf28}+0.42\%$
test_seq_wrap[compile-overhead]	0.3132ms	0.2366ms	4.2262 KOps/s	4.3928 KOps/s	$\color{#d91a1a}-3.79\%$
test_func_call_runtime[False-eager]	0.9672ms	0.7752ms	1.2899 KOps/s	1.3578 KOps/s	$\color{#d91a1a}-5.00\%$
test_func_call_runtime[False-compile]	0.9950ms	0.7907ms	1.2646 KOps/s	1.2959 KOps/s	$\color{#d91a1a}-2.42\%$
test_func_call_runtime[False-compile-overhead]	0.4532ms	0.3685ms	2.7138 KOps/s	2.6968 KOps/s	$\color{#35bf28}+0.63\%$
test_func_call_runtime[True-eager]	0.9977ms	0.8826ms	1.1330 KOps/s	1.1158 KOps/s	$\color{#35bf28}+1.54\%$
test_func_call_runtime[True-compile]	0.8988ms	0.8099ms	1.2347 KOps/s	1.2560 KOps/s	$\color{#d91a1a}-1.69\%$
test_func_call_runtime[True-compile-overhead]	0.5775ms	0.3924ms	2.5483 KOps/s	2.5445 KOps/s	$\color{#35bf28}+0.15\%$
test_func_call_cm_runtime[False-eager]	0.7859ms	0.7149ms	1.3989 KOps/s	1.3648 KOps/s	$\color{#35bf28}+2.50\%$
test_func_call_cm_runtime[False-compile]	0.8407ms	0.7727ms	1.2942 KOps/s	1.2638 KOps/s	$\color{#35bf28}+2.40\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4391ms	0.3731ms	2.6801 KOps/s	2.6877 KOps/s	$\color{#d91a1a}-0.28\%$
test_func_call_cm_runtime[True-eager]	1.0632ms	0.9836ms	1.0167 KOps/s	1.0010 KOps/s	$\color{#35bf28}+1.57\%$
test_func_call_cm_runtime[True-compile]	0.9053ms	0.8246ms	1.2127 KOps/s	1.2131 KOps/s	$\color{#d91a1a}-0.03\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4858ms	0.4163ms	2.4022 KOps/s	2.3720 KOps/s	$\color{#35bf28}+1.27\%$
test_vmap_func_call_cm_runtime[eager]	2.5007ms	2.0248ms	493.8741 Ops/s	489.3318 Ops/s	$\color{#35bf28}+0.93\%$
test_vmap_func_call_cm_runtime[compile]	0.9719ms	0.8495ms	1.1771 KOps/s	1.2044 KOps/s	$\color{#d91a1a}-2.26\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4902ms	0.4219ms	2.3705 KOps/s	2.3777 KOps/s	$\color{#d91a1a}-0.30\%$
test_distributed	3.5277ms	0.1265ms	7.9032 KOps/s	8.1979 KOps/s	$\color{#d91a1a}-3.59\%$
test_tdmodule	20.7000μs	13.5754μs	73.6627 KOps/s	66.9443 KOps/s	$\textbf{\color{#35bf28}+10.04\%}$
test_tdmodule_dispatch	55.3810μs	27.0690μs	36.9426 KOps/s	35.4806 KOps/s	$\color{#35bf28}+4.12\%$
test_tdseq	35.5000μs	15.3558μs	65.1219 KOps/s	63.3119 KOps/s	$\color{#35bf28}+2.86\%$
test_tdseq_dispatch	51.8510μs	31.3475μs	31.9004 KOps/s	32.1698 KOps/s	$\color{#d91a1a}-0.84\%$
test_instantiation_functorch	1.7887ms	1.5478ms	646.0906 Ops/s	639.2202 Ops/s	$\color{#35bf28}+1.07\%$
test_exec_functorch	0.1959ms	0.1409ms	7.0984 KOps/s	7.0567 KOps/s	$\color{#35bf28}+0.59\%$
test_exec_functional_call	0.2354ms	0.1416ms	7.0610 KOps/s	7.4234 KOps/s	$\color{#d91a1a}-4.88\%$
test_exec_td_decorator	0.3792ms	0.1800ms	5.5545 KOps/s	5.5059 KOps/s	$\color{#35bf28}+0.88\%$
test_vmap_mlp_speed_decorator[True-True]	0.8200ms	0.6708ms	1.4908 KOps/s	1.5047 KOps/s	$\color{#d91a1a}-0.92\%$
test_vmap_mlp_speed_decorator[True-False]	0.7806ms	0.6772ms	1.4767 KOps/s	1.4696 KOps/s	$\color{#35bf28}+0.48\%$
test_vmap_mlp_speed_decorator[False-True]	0.7233ms	0.6053ms	1.6520 KOps/s	1.6260 KOps/s	$\color{#35bf28}+1.60\%$
test_vmap_mlp_speed_decorator[False-False]	0.7182ms	0.6012ms	1.6634 KOps/s	1.6264 KOps/s	$\color{#35bf28}+2.27\%$
test_vmap_transformer_speed_decorator[True-True]	19.8475ms	18.8412ms	53.0751 Ops/s	53.2587 Ops/s	$\color{#d91a1a}-0.34\%$
test_vmap_transformer_speed_decorator[True-False]	19.8014ms	18.8422ms	53.0723 Ops/s	53.3055 Ops/s	$\color{#d91a1a}-0.44\%$
test_vmap_transformer_speed_decorator[False-True]	19.7874ms	18.9262ms	52.8368 Ops/s	53.7272 Ops/s	$\color{#d91a1a}-1.66\%$
test_vmap_transformer_speed_decorator[False-False]	19.6961ms	18.6016ms	53.7587 Ops/s	53.3424 Ops/s	$\color{#35bf28}+0.78\%$
test_to_module_speed[True]	1.0719ms	0.9412ms	1.0625 KOps/s	1.0450 KOps/s	$\color{#35bf28}+1.67\%$
test_to_module_speed[False]	1.3518ms	0.9299ms	1.0754 KOps/s	1.0751 KOps/s	$\color{#35bf28}+0.02\%$
test_tc_init	76.8210μs	36.5230μs	27.3800 KOps/s	27.0630 KOps/s	$\color{#35bf28}+1.17\%$
test_tc_init_nested	0.1140ms	72.5895μs	13.7761 KOps/s	13.1017 KOps/s	$\textbf{\color{#35bf28}+5.15\%}$
test_tc_first_layer_tensor	5.1144μs	0.7195μs	1.3898 MOps/s	1.2974 MOps/s	$\textbf{\color{#35bf28}+7.12\%}$
test_tc_first_layer_nontensor	35.4900μs	2.4266μs	412.1001 KOps/s	419.3406 KOps/s	$\color{#d91a1a}-1.73\%$
test_tc_second_layer_tensor	9.0850μs	1.4499μs	689.6926 KOps/s	668.5726 KOps/s	$\color{#35bf28}+3.16\%$
test_tc_second_layer_nontensor	27.8100μs	3.1778μs	314.6879 KOps/s	319.8217 KOps/s	$\color{#d91a1a}-1.61\%$
test_unbind	0.2276s	9.8035ms	102.0049 Ops/s	151.2060 Ops/s	$\textbf{\color{#d91a1a}-32.54\%}$
test_full_like	10.5602ms	9.1191ms	109.6594 Ops/s	107.9929 Ops/s	$\color{#35bf28}+1.54\%$
test_zeros_like	9.1915ms	7.1361ms	140.1334 Ops/s	114.4083 Ops/s	$\textbf{\color{#35bf28}+22.49\%}$
test_ones_like	4.9928ms	4.2517ms	235.2023 Ops/s	231.8687 Ops/s	$\color{#35bf28}+1.44\%$
test_clone	7.5079ms	6.3254ms	158.0922 Ops/s	159.1434 Ops/s	$\color{#d91a1a}-0.66\%$
test_squeeze	62.6310μs	9.6152μs	104.0021 KOps/s	103.7040 KOps/s	$\color{#35bf28}+0.29\%$
test_unsqueeze	0.1322ms	73.4465μs	13.6153 KOps/s	13.7378 KOps/s	$\color{#d91a1a}-0.89\%$
test_split	0.4026ms	0.1593ms	6.2777 KOps/s	5.9933 KOps/s	$\color{#35bf28}+4.74\%$
test_permute	0.2534ms	0.1888ms	5.2960 KOps/s	5.1798 KOps/s	$\color{#35bf28}+2.24\%$
test_stack	51.0700ms	50.6656ms	19.7373 Ops/s	19.6920 Ops/s	$\color{#35bf28}+0.23\%$
test_cat	50.9556ms	50.4981ms	19.8027 Ops/s	19.6861 Ops/s	$\color{#35bf28}+0.59\%$