[Feature] Better logs of key errors in assert_close

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] Better logs of key errors in assert_close #1082

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}21$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	46.1960μs	17.4445μs	57.3247 KOps/s	58.2340 KOps/s	$\color{#d91a1a}-1.56\%$
test_plain_set_stack_nested	67.1160μs	17.6156μs	56.7680 KOps/s	57.7851 KOps/s	$\color{#d91a1a}-1.76\%$
test_plain_set_nested_inplace	54.9530μs	19.5523μs	51.1449 KOps/s	52.7678 KOps/s	$\color{#d91a1a}-3.08\%$
test_plain_set_stack_nested_inplace	57.0070μs	19.5264μs	51.2128 KOps/s	53.1820 KOps/s	$\color{#d91a1a}-3.70\%$
test_items	39.2940μs	4.1856μs	238.9118 KOps/s	242.9429 KOps/s	$\color{#d91a1a}-1.66\%$
test_items_nested	0.6305ms	0.3444ms	2.9039 KOps/s	2.9749 KOps/s	$\color{#d91a1a}-2.39\%$
test_items_nested_locked	0.5389ms	0.3465ms	2.8856 KOps/s	2.9342 KOps/s	$\color{#d91a1a}-1.66\%$
test_items_nested_leaf	0.1323ms	71.5351μs	13.9791 KOps/s	14.1328 KOps/s	$\color{#d91a1a}-1.09\%$
test_items_stack_nested	0.4944ms	0.3501ms	2.8565 KOps/s	2.9182 KOps/s	$\color{#d91a1a}-2.12\%$
test_items_stack_nested_leaf	0.1301ms	74.8469μs	13.3606 KOps/s	13.7868 KOps/s	$\color{#d91a1a}-3.09\%$
test_items_stack_nested_locked	1.2763ms	0.3485ms	2.8697 KOps/s	2.9457 KOps/s	$\color{#d91a1a}-2.58\%$
test_keys	0.1357ms	3.8046μs	262.8409 KOps/s	283.6990 KOps/s	$\textbf{\color{#d91a1a}-7.35\%}$
test_keys_nested	0.1891ms	0.1379ms	7.2535 KOps/s	7.4564 KOps/s	$\color{#d91a1a}-2.72\%$
test_keys_nested_locked	1.8663ms	0.1421ms	7.0351 KOps/s	7.1291 KOps/s	$\color{#d91a1a}-1.32\%$
test_keys_nested_leaf	0.1830ms	0.1179ms	8.4812 KOps/s	8.7131 KOps/s	$\color{#d91a1a}-2.66\%$
test_keys_stack_nested	0.2259ms	0.1370ms	7.3017 KOps/s	7.4577 KOps/s	$\color{#d91a1a}-2.09\%$
test_keys_stack_nested_leaf	0.1979ms	0.1174ms	8.5153 KOps/s	8.7587 KOps/s	$\color{#d91a1a}-2.78\%$
test_keys_stack_nested_locked	0.2396ms	0.1425ms	7.0160 KOps/s	7.1706 KOps/s	$\color{#d91a1a}-2.16\%$
test_values	54.1452μs	1.0591μs	944.2286 KOps/s	926.1740 KOps/s	$\color{#35bf28}+1.95\%$
test_values_nested	0.1135ms	56.5050μs	17.6976 KOps/s	18.2271 KOps/s	$\color{#d91a1a}-2.91\%$
test_values_nested_locked	0.1109ms	56.7439μs	17.6230 KOps/s	18.4050 KOps/s	$\color{#d91a1a}-4.25\%$
test_values_nested_leaf	0.1240ms	61.1707μs	16.3477 KOps/s	16.8083 KOps/s	$\color{#d91a1a}-2.74\%$
test_values_stack_nested	0.1132ms	57.9086μs	17.2686 KOps/s	16.0228 KOps/s	$\textbf{\color{#35bf28}+7.78\%}$
test_values_stack_nested_leaf	0.1075ms	61.3448μs	16.3013 KOps/s	16.7756 KOps/s	$\color{#d91a1a}-2.83\%$
test_values_stack_nested_locked	0.1158ms	58.3781μs	17.1297 KOps/s	18.2655 KOps/s	$\textbf{\color{#d91a1a}-6.22\%}$
test_membership	15.9900μs	0.9117μs	1.0969 MOps/s	1.1305 MOps/s	$\color{#d91a1a}-2.98\%$
test_membership_nested	43.3610μs	2.7229μs	367.2559 KOps/s	366.7777 KOps/s	$\color{#35bf28}+0.13\%$
test_membership_nested_leaf	43.2610μs	2.7483μs	363.8604 KOps/s	364.7415 KOps/s	$\color{#d91a1a}-0.24\%$
test_membership_stacked_nested	23.6140μs	2.7271μs	366.6852 KOps/s	364.7062 KOps/s	$\color{#35bf28}+0.54\%$
test_membership_stacked_nested_leaf	15.2190μs	2.7206μs	367.5597 KOps/s	370.0019 KOps/s	$\color{#d91a1a}-0.66\%$
test_membership_nested_last	46.8680μs	4.1026μs	243.7455 KOps/s	249.4830 KOps/s	$\color{#d91a1a}-2.30\%$
test_membership_nested_leaf_last	27.6920μs	4.0500μs	246.9163 KOps/s	247.6267 KOps/s	$\color{#d91a1a}-0.29\%$
test_membership_stacked_nested_last	52.5380μs	5.6801μs	176.0530 KOps/s	251.4704 KOps/s	$\textbf{\color{#d91a1a}-29.99\%}$
test_membership_stacked_nested_leaf_last	26.2190μs	5.6697μs	176.3773 KOps/s	245.2387 KOps/s	$\textbf{\color{#d91a1a}-28.08\%}$
test_nested_getleaf	51.1650μs	10.7157μs	93.3212 KOps/s	93.6924 KOps/s	$\color{#d91a1a}-0.40\%$
test_nested_get	56.0540μs	10.3333μs	96.7745 KOps/s	98.6755 KOps/s	$\color{#d91a1a}-1.93\%$
test_stacked_getleaf	54.6730μs	11.1659μs	89.5582 KOps/s	92.7016 KOps/s	$\color{#d91a1a}-3.39\%$
test_stacked_get	51.9180μs	10.2619μs	97.4479 KOps/s	98.5594 KOps/s	$\color{#d91a1a}-1.13\%$
test_nested_getitemleaf	0.2726ms	11.6665μs	85.7152 KOps/s	88.8269 KOps/s	$\color{#d91a1a}-3.50\%$
test_nested_getitem	38.6530μs	10.4532μs	95.6649 KOps/s	95.0731 KOps/s	$\color{#35bf28}+0.62\%$
test_stacked_getitemleaf	57.4680μs	11.0913μs	90.1611 KOps/s	89.2378 KOps/s	$\color{#35bf28}+1.03\%$
test_stacked_getitem	59.2800μs	10.3418μs	96.6951 KOps/s	95.0596 KOps/s	$\color{#35bf28}+1.72\%$
test_lock_nested	3.2045ms	0.4564ms	2.1912 KOps/s	1.7718 KOps/s	$\textbf{\color{#35bf28}+23.68\%}$
test_lock_stack_nested	0.7549ms	0.4170ms	2.3982 KOps/s	2.3909 KOps/s	$\color{#35bf28}+0.31\%$
test_unlock_nested	1.4654ms	0.3740ms	2.6740 KOps/s	2.7114 KOps/s	$\color{#d91a1a}-1.38\%$
test_unlock_stack_nested	0.6316ms	0.3326ms	3.0063 KOps/s	3.0121 KOps/s	$\color{#d91a1a}-0.19\%$
test_flatten_speed	0.1840ms	91.8334μs	10.8893 KOps/s	11.0543 KOps/s	$\color{#d91a1a}-1.49\%$
test_unflatten_speed	1.1432ms	0.4843ms	2.0649 KOps/s	2.1089 KOps/s	$\color{#d91a1a}-2.09\%$
test_common_ops	5.5204ms	0.7723ms	1.2948 KOps/s	1.3107 KOps/s	$\color{#d91a1a}-1.21\%$
test_creation	0.1276ms	2.0959μs	477.1116 KOps/s	488.0688 KOps/s	$\color{#d91a1a}-2.25\%$
test_creation_empty	0.2581ms	11.0168μs	90.7705 KOps/s	101.3007 KOps/s	$\textbf{\color{#d91a1a}-10.39\%}$
test_creation_nested_1	40.4260μs	12.8663μs	77.7223 KOps/s	79.2346 KOps/s	$\color{#d91a1a}-1.91\%$
test_creation_nested_2	49.6830μs	17.2822μs	57.8629 KOps/s	60.3288 KOps/s	$\color{#d91a1a}-4.09\%$
test_clone	56.6770μs	13.1599μs	75.9887 KOps/s	75.9491 KOps/s	$\color{#35bf28}+0.05\%$
test_getitem[int]	1.1884ms	12.7240μs	78.5915 KOps/s	80.0554 KOps/s	$\color{#d91a1a}-1.83\%$
test_getitem[slice_int]	0.1400ms	24.3681μs	41.0373 KOps/s	41.7188 KOps/s	$\color{#d91a1a}-1.63\%$
test_getitem[range]	0.1664ms	48.9436μs	20.4317 KOps/s	21.0688 KOps/s	$\color{#d91a1a}-3.02\%$
test_getitem[tuple]	0.1359ms	20.1329μs	49.6700 KOps/s	49.8353 KOps/s	$\color{#d91a1a}-0.33\%$
test_getitem[list]	0.2751ms	45.0336μs	22.2056 KOps/s	23.1481 KOps/s	$\color{#d91a1a}-4.07\%$
test_setitem_dim[int]	54.8630μs	25.4917μs	39.2284 KOps/s	39.3813 KOps/s	$\color{#d91a1a}-0.39\%$
test_setitem_dim[slice_int]	91.5520μs	51.4818μs	19.4243 KOps/s	18.4430 KOps/s	$\textbf{\color{#35bf28}+5.32\%}$
test_setitem_dim[range]	0.1233ms	74.0639μs	13.5018 KOps/s	13.4334 KOps/s	$\color{#35bf28}+0.51\%$
test_setitem_dim[tuple]	75.1410μs	40.5389μs	24.6677 KOps/s	24.0548 KOps/s	$\color{#35bf28}+2.55\%$
test_setitem	60.6830μs	19.9728μs	50.0682 KOps/s	49.5005 KOps/s	$\color{#35bf28}+1.15\%$
test_set	67.7370μs	19.5108μs	51.2536 KOps/s	52.0046 KOps/s	$\color{#d91a1a}-1.44\%$
test_set_shared	3.7239ms	0.1730ms	5.7819 KOps/s	5.6643 KOps/s	$\color{#35bf28}+2.08\%$
test_update	0.2581ms	21.5862μs	46.3259 KOps/s	45.4785 KOps/s	$\color{#35bf28}+1.86\%$
test_update_nested	0.1832ms	30.8422μs	32.4231 KOps/s	30.7095 KOps/s	$\textbf{\color{#35bf28}+5.58\%}$
test_update__nested	0.3860ms	32.1992μs	31.0567 KOps/s	30.0893 KOps/s	$\color{#35bf28}+3.21\%$
test_set_nested	0.1533ms	21.5799μs	46.3395 KOps/s	45.9943 KOps/s	$\color{#35bf28}+0.75\%$
test_set_nested_new	0.1177ms	25.9201μs	38.5801 KOps/s	38.3289 KOps/s	$\color{#35bf28}+0.66\%$
test_select	0.2609ms	42.3001μs	23.6406 KOps/s	23.1496 KOps/s	$\color{#35bf28}+2.12\%$
test_select_nested	0.1584ms	60.2019μs	16.6108 KOps/s	16.8805 KOps/s	$\color{#d91a1a}-1.60\%$
test_exclude_nested	0.1447ms	75.7543μs	13.2006 KOps/s	13.3414 KOps/s	$\color{#d91a1a}-1.06\%$
test_empty[True]	0.7004ms	0.3478ms	2.8749 KOps/s	2.6750 KOps/s	$\textbf{\color{#35bf28}+7.47\%}$
test_empty[False]	9.8185μs	1.2208μs	819.1395 KOps/s	804.6226 KOps/s	$\color{#35bf28}+1.80\%$
test_unbind_speed	0.3688ms	0.2660ms	3.7592 KOps/s	3.8875 KOps/s	$\color{#d91a1a}-3.30\%$
test_unbind_speed_stack0	0.5353ms	0.2593ms	3.8572 KOps/s	3.9194 KOps/s	$\color{#d91a1a}-1.59\%$
test_unbind_speed_stack1	0.1188s	0.7773ms	1.2865 KOps/s	1.4057 KOps/s	$\textbf{\color{#d91a1a}-8.49\%}$
test_split	0.1136s	1.7730ms	564.0043 Ops/s	564.7251 Ops/s	$\color{#d91a1a}-0.13\%$
test_chunk	0.1172s	1.7910ms	558.3333 Ops/s	569.3965 Ops/s	$\color{#d91a1a}-1.94\%$
test_consolidate_njt[False-None]	10.4609ms	8.2099ms	121.8038 Ops/s	120.6497 Ops/s	$\color{#35bf28}+0.96\%$
test_creation[device0]	4.4034ms	94.7248μs	10.5569 KOps/s	10.7435 KOps/s	$\color{#d91a1a}-1.74\%$
test_creation_from_tensor	0.2769ms	94.5529μs	10.5761 KOps/s	10.1404 KOps/s	$\color{#35bf28}+4.30\%$
test_add_one[memmap_tensor0]	0.1553ms	4.9887μs	200.4535 KOps/s	196.9655 KOps/s	$\color{#35bf28}+1.77\%$
test_contiguous[memmap_tensor0]	22.6420μs	0.5402μs	1.8510 MOps/s	1.9493 MOps/s	$\textbf{\color{#d91a1a}-5.04\%}$
test_stack[memmap_tensor0]	31.8300μs	3.4719μs	288.0262 KOps/s	276.8262 KOps/s	$\color{#35bf28}+4.05\%$
test_memmaptd_index	1.1533ms	0.2369ms	4.2209 KOps/s	4.2060 KOps/s	$\color{#35bf28}+0.35\%$
test_memmaptd_index_astensor	2.0748ms	0.3311ms	3.0203 KOps/s	3.1902 KOps/s	$\textbf{\color{#d91a1a}-5.33\%}$
test_memmaptd_index_op	1.0937ms	0.5832ms	1.7147 KOps/s	1.7331 KOps/s	$\color{#d91a1a}-1.06\%$
test_serialize_model	0.1322s	0.1181s	8.4662 Ops/s	8.2499 Ops/s	$\color{#35bf28}+2.62\%$
test_serialize_model_pickle	0.5105s	0.4048s	2.4701 Ops/s	2.5617 Ops/s	$\color{#d91a1a}-3.58\%$
test_serialize_weights	0.2234s	0.1314s	7.6097 Ops/s	8.5047 Ops/s	$\textbf{\color{#d91a1a}-10.52\%}$
test_serialize_weights_returnearly	0.1755s	0.1633s	6.1252 Ops/s	6.3734 Ops/s	$\color{#d91a1a}-3.89\%$
test_serialize_weights_pickle	1.0371s	0.7154s	1.3979 Ops/s	1.1067 Ops/s	$\textbf{\color{#35bf28}+26.31\%}$
test_serialize_weights_filesystem	0.1557s	0.1427s	7.0063 Ops/s	7.0406 Ops/s	$\color{#d91a1a}-0.49\%$
test_serialize_model_filesystem	0.2486s	0.1558s	6.4175 Ops/s	6.8399 Ops/s	$\textbf{\color{#d91a1a}-6.18\%}$
test_reshape_pytree	59.0910μs	27.1816μs	36.7897 KOps/s	37.9952 KOps/s	$\color{#d91a1a}-3.17\%$
test_reshape_td	72.6570μs	32.3281μs	30.9328 KOps/s	31.6845 KOps/s	$\color{#d91a1a}-2.37\%$
test_view_pytree	76.0520μs	27.3956μs	36.5022 KOps/s	37.4023 KOps/s	$\color{#d91a1a}-2.41\%$
test_view_td	0.1120ms	39.5188μs	25.3044 KOps/s	27.4391 KOps/s	$\textbf{\color{#d91a1a}-7.78\%}$
test_unbind_pytree	0.1538ms	30.2670μs	33.0392 KOps/s	34.0389 KOps/s	$\color{#d91a1a}-2.94\%$
test_unbind_td	0.3555ms	39.3744μs	25.3972 KOps/s	26.2399 KOps/s	$\color{#d91a1a}-3.21\%$
test_split_pytree	89.2080μs	29.6642μs	33.7106 KOps/s	34.3827 KOps/s	$\color{#d91a1a}-1.95\%$
test_split_td	0.5269ms	44.2927μs	22.5771 KOps/s	22.7601 KOps/s	$\color{#d91a1a}-0.80\%$
test_add_pytree	0.1051ms	36.3278μs	27.5271 KOps/s	27.3070 KOps/s	$\color{#35bf28}+0.81\%$
test_add_td	0.1353ms	52.4058μs	19.0819 KOps/s	19.3425 KOps/s	$\color{#d91a1a}-1.35\%$
test_compile_add_one_nested[tensordict-compile]	0.1235ms	62.5252μs	15.9936 KOps/s	15.9417 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_add_one_nested[tensordict-eager]	0.3994ms	0.1587ms	6.3026 KOps/s	6.2963 KOps/s	$\color{#35bf28}+0.10\%$
test_compile_add_one_nested[pytree-compile]	0.1223ms	46.0768μs	21.7029 KOps/s	21.5701 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_add_one_nested[pytree-eager]	0.2495ms	0.1193ms	8.3800 KOps/s	8.3584 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_copy_nested[tensordict-compile]	74.2890μs	26.1652μs	38.2188 KOps/s	37.6090 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_copy_nested[tensordict-eager]	0.1092ms	53.2261μs	18.7878 KOps/s	18.5675 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_copy_nested[pytree-compile]	0.1738ms	79.2570μs	12.6172 KOps/s	12.8229 KOps/s	$\color{#d91a1a}-1.60\%$
test_compile_copy_nested[pytree-eager]	0.1334ms	69.0046μs	14.4918 KOps/s	14.8919 KOps/s	$\color{#d91a1a}-2.69\%$
test_compile_add_one_flat[tensordict-compile]	0.1869ms	0.1055ms	9.4815 KOps/s	9.4667 KOps/s	$\color{#35bf28}+0.16\%$
test_compile_add_one_flat[tensordict-eager]	0.4312ms	0.1968ms	5.0825 KOps/s	5.0408 KOps/s	$\color{#35bf28}+0.83\%$
test_compile_add_one_flat[tensorclass-compile]	91.6220μs	45.8227μs	21.8232 KOps/s	22.4032 KOps/s	$\color{#d91a1a}-2.59\%$
test_compile_add_one_flat[tensorclass-eager]	0.4959ms	60.5082μs	16.5267 KOps/s	16.1810 KOps/s	$\color{#35bf28}+2.14\%$
test_compile_add_one_flat[pytree-compile]	0.1887ms	0.1034ms	9.6752 KOps/s	9.7304 KOps/s	$\color{#d91a1a}-0.57\%$
test_compile_add_one_flat[pytree-eager]	0.3656ms	0.2007ms	4.9828 KOps/s	4.9774 KOps/s	$\color{#35bf28}+0.11\%$
test_compile_add_self_flat[tensordict-eager]	0.3758ms	0.2066ms	4.8393 KOps/s	4.6030 KOps/s	$\textbf{\color{#35bf28}+5.13\%}$
test_compile_add_self_flat[tensordict-compile]	0.2142ms	0.1058ms	9.4523 KOps/s	9.5585 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_add_self_flat[tensorclass-eager]	0.1968ms	56.3739μs	17.7387 KOps/s	18.4424 KOps/s	$\color{#d91a1a}-3.82\%$
test_compile_add_self_flat[tensorclass-compile]	0.1022ms	48.7485μs	20.5135 KOps/s	21.2804 KOps/s	$\color{#d91a1a}-3.60\%$
test_compile_add_self_flat[pytree-eager]	0.9621ms	0.1682ms	5.9455 KOps/s	6.2826 KOps/s	$\textbf{\color{#d91a1a}-5.37\%}$
test_compile_add_self_flat[pytree-compile]	0.2007ms	0.1035ms	9.6582 KOps/s	9.7434 KOps/s	$\color{#d91a1a}-0.88\%$
test_compile_copy_flat[tensordict-compile]	59.7220μs	20.8702μs	47.9151 KOps/s	47.7974 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_copy_flat[tensordict-eager]	0.1291ms	60.7323μs	16.4657 KOps/s	17.0361 KOps/s	$\color{#d91a1a}-3.35\%$
test_compile_copy_flat[pytree-compile]	0.1751ms	81.2849μs	12.3024 KOps/s	12.5019 KOps/s	$\color{#d91a1a}-1.60\%$
test_compile_copy_flat[pytree-eager]	0.1261ms	68.7439μs	14.5468 KOps/s	14.8913 KOps/s	$\color{#d91a1a}-2.31\%$
test_compile_assign_and_add[tensordict-compile]	0.3727ms	0.2079ms	4.8099 KOps/s	4.8302 KOps/s	$\color{#d91a1a}-0.42\%$
test_compile_assign_and_add[tensordict-eager]	1.4970ms	1.2417ms	805.3331 Ops/s	778.6268 Ops/s	$\color{#35bf28}+3.43\%$
test_compile_assign_and_add[pytree-compile]	0.4163ms	0.2081ms	4.8046 KOps/s	4.9104 KOps/s	$\color{#d91a1a}-2.15\%$
test_compile_assign_and_add[pytree-eager]	0.8843ms	0.7773ms	1.2864 KOps/s	1.2923 KOps/s	$\color{#d91a1a}-0.45\%$
test_compile_assign_and_add_stack[compile]	0.5666ms	0.4578ms	2.1845 KOps/s	2.1757 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_assign_and_add_stack[eager]	3.5204ms	2.5737ms	388.5406 Ops/s	387.8447 Ops/s	$\color{#35bf28}+0.18\%$
test_compile_indexing[tensor-tensordict-compile]	84.7890μs	36.1887μs	27.6330 KOps/s	27.2171 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_indexing[tensor-tensordict-eager]	0.5440ms	33.1809μs	30.1378 KOps/s	29.3428 KOps/s	$\color{#35bf28}+2.71\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1357ms	29.9018μs	33.4428 KOps/s	33.5563 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_indexing[tensor-tensorclass-eager]	78.3660μs	23.4195μs	42.6995 KOps/s	41.2465 KOps/s	$\color{#35bf28}+3.52\%$
test_compile_indexing[tensor-pytree-compile]	74.9710μs	30.3083μs	32.9942 KOps/s	32.8414 KOps/s	$\color{#35bf28}+0.47\%$
test_compile_indexing[tensor-pytree-eager]	68.2080μs	23.5974μs	42.3776 KOps/s	41.5877 KOps/s	$\color{#35bf28}+1.90\%$
test_compile_indexing[slice-tensordict-compile]	0.1115ms	53.5194μs	18.6848 KOps/s	19.4625 KOps/s	$\color{#d91a1a}-4.00\%$
test_compile_indexing[slice-tensordict-eager]	0.6058ms	20.1159μs	49.7120 KOps/s	49.7727 KOps/s	$\color{#d91a1a}-0.12\%$
test_compile_indexing[slice-tensorclass-compile]	0.1065ms	44.9498μs	22.2471 KOps/s	22.2867 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_indexing[slice-tensorclass-eager]	0.2952ms	20.4132μs	48.9879 KOps/s	52.8506 KOps/s	$\textbf{\color{#d91a1a}-7.31\%}$
test_compile_indexing[slice-pytree-compile]	0.1134ms	46.0199μs	21.7297 KOps/s	22.0662 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_indexing[slice-pytree-eager]	66.5750μs	19.1807μs	52.1358 KOps/s	53.1964 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_indexing[int-tensordict-compile]	0.1208ms	53.9460μs	18.5371 KOps/s	18.9083 KOps/s	$\color{#d91a1a}-1.96\%$
test_compile_indexing[int-tensordict-eager]	1.0765ms	19.6770μs	50.8207 KOps/s	49.9653 KOps/s	$\color{#35bf28}+1.71\%$
test_compile_indexing[int-tensorclass-compile]	0.1147ms	46.0006μs	21.7388 KOps/s	22.1676 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_indexing[int-tensorclass-eager]	0.1664ms	19.3728μs	51.6186 KOps/s	53.2825 KOps/s	$\color{#d91a1a}-3.12\%$
test_compile_indexing[int-pytree-compile]	97.7730μs	45.8783μs	21.7968 KOps/s	22.0680 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_indexing[int-pytree-eager]	94.7980μs	19.3952μs	51.5591 KOps/s	53.3593 KOps/s	$\color{#d91a1a}-3.37\%$
test_mod_add[eager]	92.2230μs	26.8337μs	37.2666 KOps/s	37.0690 KOps/s	$\color{#35bf28}+0.53\%$
test_mod_add[compile]	0.1106ms	45.0849μs	22.1804 KOps/s	21.3503 KOps/s	$\color{#35bf28}+3.89\%$
test_mod_add[compile-overhead]	0.1242ms	46.4027μs	21.5504 KOps/s	21.4860 KOps/s	$\color{#35bf28}+0.30\%$
test_mod_wrap[eager]	0.4377ms	0.2184ms	4.5779 KOps/s	4.4675 KOps/s	$\color{#35bf28}+2.47\%$
test_mod_wrap[compile]	2.2318ms	0.2046ms	4.8882 KOps/s	4.7781 KOps/s	$\color{#35bf28}+2.30\%$
test_mod_wrap[compile-overhead]	2.5146ms	0.2112ms	4.7343 KOps/s	4.8253 KOps/s	$\color{#d91a1a}-1.89\%$
test_mod_wrap_and_backward[eager]	15.1834ms	12.2663ms	81.5245 Ops/s	88.9472 Ops/s	$\textbf{\color{#d91a1a}-8.35\%}$
test_mod_wrap_and_backward[compile]	19.0657ms	13.4298ms	74.4615 Ops/s	88.8286 Ops/s	$\textbf{\color{#d91a1a}-16.17\%}$
test_mod_wrap_and_backward[compile-overhead]	16.8954ms	13.6259ms	73.3899 Ops/s	84.1636 Ops/s	$\textbf{\color{#d91a1a}-12.80\%}$
test_seq_add[eager]	0.2189ms	91.9988μs	10.8697 KOps/s	10.6361 KOps/s	$\color{#35bf28}+2.20\%$
test_seq_add[compile]	0.2103ms	62.7654μs	15.9323 KOps/s	16.5788 KOps/s	$\color{#d91a1a}-3.90\%$
test_seq_add[compile-overhead]	0.1462ms	59.9608μs	16.6776 KOps/s	16.5564 KOps/s	$\color{#35bf28}+0.73\%$
test_seq_wrap[eager]	0.5867ms	0.3983ms	2.5106 KOps/s	2.5223 KOps/s	$\color{#d91a1a}-0.47\%$
test_seq_wrap[compile]	0.4258ms	0.2283ms	4.3799 KOps/s	4.3578 KOps/s	$\color{#35bf28}+0.51\%$
test_seq_wrap[compile-overhead]	0.4376ms	0.2287ms	4.3728 KOps/s	4.3886 KOps/s	$\color{#d91a1a}-0.36\%$
test_func_call_runtime[False-eager]	0.8398ms	0.5602ms	1.7850 KOps/s	1.7865 KOps/s	$\color{#d91a1a}-0.09\%$
test_func_call_runtime[False-compile]	0.8853ms	0.4298ms	2.3266 KOps/s	2.3465 KOps/s	$\color{#d91a1a}-0.85\%$
test_func_call_runtime[False-compile-overhead]	0.7579ms	0.4300ms	2.3255 KOps/s	2.3474 KOps/s	$\color{#d91a1a}-0.93\%$
test_func_call_runtime[True-eager]	0.9392ms	0.7662ms	1.3052 KOps/s	1.2953 KOps/s	$\color{#35bf28}+0.77\%$
test_func_call_runtime[True-compile]	0.7438ms	0.4784ms	2.0904 KOps/s	2.1187 KOps/s	$\color{#d91a1a}-1.34\%$
test_func_call_runtime[True-compile-overhead]	0.7053ms	0.4712ms	2.1222 KOps/s	2.1613 KOps/s	$\color{#d91a1a}-1.81\%$
test_func_call_cm_runtime[False-eager]	2.2468ms	0.5796ms	1.7254 KOps/s	1.7830 KOps/s	$\color{#d91a1a}-3.23\%$
test_func_call_cm_runtime[False-compile]	0.6760ms	0.4266ms	2.3443 KOps/s	2.3326 KOps/s	$\color{#35bf28}+0.50\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5434ms	0.4268ms	2.3432 KOps/s	2.3369 KOps/s	$\color{#35bf28}+0.27\%$
test_func_call_cm_runtime[True-eager]	4.1665ms	0.9749ms	1.0258 KOps/s	1.0910 KOps/s	$\textbf{\color{#d91a1a}-5.98\%}$
test_func_call_cm_runtime[True-compile]	2.3812ms	0.5045ms	1.9820 KOps/s	2.0170 KOps/s	$\color{#d91a1a}-1.74\%$
test_func_call_cm_runtime[True-compile-overhead]	0.7031ms	0.4953ms	2.0191 KOps/s	2.0294 KOps/s	$\color{#d91a1a}-0.51\%$
test_vmap_func_call_cm_runtime[eager]	3.0737ms	1.9111ms	523.2688 Ops/s	503.3597 Ops/s	$\color{#35bf28}+3.96\%$
test_vmap_func_call_cm_runtime[compile]	0.7246ms	0.5151ms	1.9414 KOps/s	1.9460 KOps/s	$\color{#d91a1a}-0.24\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9996ms	0.5171ms	1.9338 KOps/s	1.9272 KOps/s	$\color{#35bf28}+0.35\%$
test_distributed	0.2523ms	0.1275ms	7.8415 KOps/s	7.6758 KOps/s	$\color{#35bf28}+2.16\%$
test_tdmodule	41.1670μs	17.8696μs	55.9611 KOps/s	53.3470 KOps/s	$\color{#35bf28}+4.90\%$
test_tdmodule_dispatch	70.8730μs	36.0201μs	27.7622 KOps/s	27.4648 KOps/s	$\color{#35bf28}+1.08\%$
test_tdseq	38.9530μs	20.5850μs	48.5790 KOps/s	45.8734 KOps/s	$\textbf{\color{#35bf28}+5.90\%}$
test_tdseq_dispatch	71.0430μs	40.5032μs	24.6894 KOps/s	23.7980 KOps/s	$\color{#35bf28}+3.75\%$
test_instantiation_functorch	1.9412ms	1.5699ms	636.9776 Ops/s	643.4491 Ops/s	$\color{#d91a1a}-1.01\%$
test_exec_functorch	0.2571ms	0.1807ms	5.5328 KOps/s	5.5152 KOps/s	$\color{#35bf28}+0.32\%$
test_exec_functional_call	0.2627ms	0.1740ms	5.7473 KOps/s	5.6946 KOps/s	$\color{#35bf28}+0.92\%$
test_exec_td_decorator	0.5043ms	0.2292ms	4.3634 KOps/s	4.4363 KOps/s	$\color{#d91a1a}-1.65\%$
test_vmap_mlp_speed_decorator[True-True]	0.7796ms	0.6426ms	1.5561 KOps/s	1.5392 KOps/s	$\color{#35bf28}+1.10\%$
test_vmap_mlp_speed_decorator[True-False]	1.1041ms	0.6418ms	1.5580 KOps/s	1.5515 KOps/s	$\color{#35bf28}+0.42\%$
test_vmap_mlp_speed_decorator[False-True]	0.9418ms	0.5286ms	1.8918 KOps/s	1.7943 KOps/s	$\textbf{\color{#35bf28}+5.44\%}$
test_vmap_mlp_speed_decorator[False-False]	0.8169ms	0.5280ms	1.8940 KOps/s	1.8713 KOps/s	$\color{#35bf28}+1.21\%$
test_to_module_speed[True]	1.9173ms	1.2772ms	782.9809 Ops/s	780.8937 Ops/s	$\color{#35bf28}+0.27\%$
test_to_module_speed[False]	1.3460ms	1.2464ms	802.2973 Ops/s	809.1749 Ops/s	$\color{#d91a1a}-0.85\%$
test_tc_init	88.9570μs	45.3909μs	22.0309 KOps/s	23.2900 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_tc_init_nested	0.1601ms	88.7482μs	11.2678 KOps/s	11.3410 KOps/s	$\color{#d91a1a}-0.65\%$
test_tc_first_layer_tensor	38.2020μs	1.5214μs	657.2876 KOps/s	660.5389 KOps/s	$\color{#d91a1a}-0.49\%$
test_tc_first_layer_nontensor	28.7140μs	4.7052μs	212.5328 KOps/s	214.8362 KOps/s	$\color{#d91a1a}-1.07\%$
test_tc_second_layer_tensor	37.7100μs	2.7638μs	361.8164 KOps/s	355.3124 KOps/s	$\color{#35bf28}+1.83\%$
test_tc_second_layer_nontensor	33.8330μs	5.9723μs	167.4409 KOps/s	168.2948 KOps/s	$\color{#d91a1a}-0.51\%$
test_unbind	0.2402s	13.6469ms	73.2767 Ops/s	83.3136 Ops/s	$\textbf{\color{#d91a1a}-12.05\%}$
test_full_like	11.1603ms	7.9563ms	125.6865 Ops/s	131.5668 Ops/s	$\color{#d91a1a}-4.47\%$
test_zeros_like	3.7980ms	3.0244ms	330.6411 Ops/s	346.0943 Ops/s	$\color{#d91a1a}-4.47\%$
test_ones_like	4.2009ms	3.6365ms	274.9893 Ops/s	291.5231 Ops/s	$\textbf{\color{#d91a1a}-5.67\%}$
test_clone	6.7203ms	5.8069ms	172.2096 Ops/s	184.4675 Ops/s	$\textbf{\color{#d91a1a}-6.65\%}$
test_squeeze	61.1050μs	11.8111μs	84.6661 KOps/s	87.6210 KOps/s	$\color{#d91a1a}-3.37\%$
test_unsqueeze	0.3692ms	89.4535μs	11.1790 KOps/s	11.5289 KOps/s	$\color{#d91a1a}-3.04\%$
test_split	0.3365ms	0.1896ms	5.2752 KOps/s	5.3143 KOps/s	$\color{#d91a1a}-0.73\%$
test_permute	0.4555ms	0.2195ms	4.5566 KOps/s	4.6302 KOps/s	$\color{#d91a1a}-1.59\%$
test_stack	29.7332ms	26.4384ms	37.8238 Ops/s	38.9738 Ops/s	$\color{#d91a1a}-2.95\%$
test_cat	29.5983ms	25.4351ms	39.3157 Ops/s	39.2163 Ops/s	$\color{#35bf28}+0.25\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}33$. Worsened: $\large\color{#d91a1a}16$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.7900μs	11.4789μs	87.1165 KOps/s	84.5021 KOps/s	$\color{#35bf28}+3.09\%$
test_plain_set_stack_nested	38.7710μs	11.5279μs	86.7464 KOps/s	83.9017 KOps/s	$\color{#35bf28}+3.39\%$
test_plain_set_nested_inplace	44.6210μs	12.4788μs	80.1361 KOps/s	77.8643 KOps/s	$\color{#35bf28}+2.92\%$
test_plain_set_stack_nested_inplace	38.5410μs	12.3960μs	80.6714 KOps/s	78.1957 KOps/s	$\color{#35bf28}+3.17\%$
test_items	24.2200μs	3.0599μs	326.8096 KOps/s	333.8965 KOps/s	$\color{#d91a1a}-2.12\%$
test_items_nested	0.4100ms	0.3153ms	3.1716 KOps/s	3.1194 KOps/s	$\color{#35bf28}+1.67\%$
test_items_nested_locked	0.3854ms	0.3224ms	3.1020 KOps/s	3.1020 KOps/s	$-0.00\%$
test_items_nested_leaf	84.8710μs	59.0495μs	16.9349 KOps/s	17.0222 KOps/s	$\color{#d91a1a}-0.51\%$
test_items_stack_nested	0.3996ms	0.3204ms	3.1211 KOps/s	3.0885 KOps/s	$\color{#35bf28}+1.06\%$
test_items_stack_nested_leaf	95.7420μs	59.8785μs	16.7005 KOps/s	16.3920 KOps/s	$\color{#35bf28}+1.88\%$
test_items_stack_nested_locked	0.3682ms	0.3228ms	3.0977 KOps/s	3.0975 KOps/s	$+0.01\%$
test_keys	28.6010μs	3.5323μs	283.0987 KOps/s	283.0188 KOps/s	$\color{#35bf28}+0.03\%$
test_keys_nested	0.1021ms	72.7924μs	13.7377 KOps/s	13.8288 KOps/s	$\color{#d91a1a}-0.66\%$
test_keys_nested_locked	2.5031ms	78.3014μs	12.7712 KOps/s	12.8627 KOps/s	$\color{#d91a1a}-0.71\%$
test_keys_nested_leaf	92.2020μs	64.0954μs	15.6017 KOps/s	15.7474 KOps/s	$\color{#d91a1a}-0.92\%$
test_keys_stack_nested	0.1126ms	72.3463μs	13.8224 KOps/s	13.7685 KOps/s	$\color{#35bf28}+0.39\%$
test_keys_stack_nested_leaf	88.5620μs	63.1531μs	15.8345 KOps/s	15.5917 KOps/s	$\color{#35bf28}+1.56\%$
test_keys_stack_nested_locked	0.1071ms	77.4838μs	12.9059 KOps/s	12.8777 KOps/s	$\color{#35bf28}+0.22\%$
test_values	5.5368μs	0.8793μs	1.1373 MOps/s	1.1300 MOps/s	$\color{#35bf28}+0.64\%$
test_values_nested	59.9010μs	32.9949μs	30.3077 KOps/s	30.4854 KOps/s	$\color{#d91a1a}-0.58\%$
test_values_nested_locked	63.5210μs	34.7924μs	28.7419 KOps/s	28.8046 KOps/s	$\color{#d91a1a}-0.22\%$
test_values_nested_leaf	68.4610μs	35.3012μs	28.3277 KOps/s	28.5505 KOps/s	$\color{#d91a1a}-0.78\%$
test_values_stack_nested	56.4310μs	33.3676μs	29.9692 KOps/s	30.0608 KOps/s	$\color{#d91a1a}-0.30\%$
test_values_stack_nested_leaf	62.7710μs	35.5446μs	28.1337 KOps/s	28.0894 KOps/s	$\color{#35bf28}+0.16\%$
test_values_stack_nested_locked	65.4410μs	35.0652μs	28.5183 KOps/s	28.4091 KOps/s	$\color{#35bf28}+0.38\%$
test_membership	1.8625μs	0.5583μs	1.7911 MOps/s	1.7962 MOps/s	$\color{#d91a1a}-0.29\%$
test_membership_nested	28.8610μs	1.9949μs	501.2899 KOps/s	492.9211 KOps/s	$\color{#35bf28}+1.70\%$
test_membership_nested_leaf	12.6750μs	1.9630μs	509.4294 KOps/s	505.7206 KOps/s	$\color{#35bf28}+0.73\%$
test_membership_stacked_nested	30.8200μs	2.0600μs	485.4433 KOps/s	485.5094 KOps/s	$\color{#d91a1a}-0.01\%$
test_membership_stacked_nested_leaf	24.5410μs	2.0385μs	490.5512 KOps/s	485.5604 KOps/s	$\color{#35bf28}+1.03\%$
test_membership_nested_last	34.6500μs	2.8594μs	349.7240 KOps/s	348.5511 KOps/s	$\color{#35bf28}+0.34\%$
test_membership_nested_leaf_last	27.5210μs	2.8998μs	344.8522 KOps/s	349.9991 KOps/s	$\color{#d91a1a}-1.47\%$
test_membership_stacked_nested_last	63.5620μs	7.8678μs	127.1011 KOps/s	345.6329 KOps/s	$\textbf{\color{#d91a1a}-63.23\%}$
test_membership_stacked_nested_leaf_last	36.8510μs	7.8509μs	127.3746 KOps/s	347.5539 KOps/s	$\textbf{\color{#d91a1a}-63.35\%}$
test_nested_getleaf	33.9800μs	6.0257μs	165.9564 KOps/s	166.8547 KOps/s	$\color{#d91a1a}-0.54\%$
test_nested_get	36.4700μs	5.6939μs	175.6275 KOps/s	176.2938 KOps/s	$\color{#d91a1a}-0.38\%$
test_stacked_getleaf	42.7110μs	5.9610μs	167.7564 KOps/s	166.6486 KOps/s	$\color{#35bf28}+0.66\%$
test_stacked_get	34.1500μs	5.6382μs	177.3621 KOps/s	174.3945 KOps/s	$\color{#35bf28}+1.70\%$
test_nested_getitemleaf	30.7210μs	6.0431μs	165.4782 KOps/s	163.8559 KOps/s	$\color{#35bf28}+0.99\%$
test_nested_getitem	25.6310μs	5.7246μs	174.6860 KOps/s	172.9716 KOps/s	$\color{#35bf28}+0.99\%$
test_stacked_getitemleaf	33.1110μs	6.0684μs	164.7874 KOps/s	164.9480 KOps/s	$\color{#d91a1a}-0.10\%$
test_stacked_getitem	26.4000μs	5.7191μs	174.8534 KOps/s	173.8217 KOps/s	$\color{#35bf28}+0.59\%$
test_lock_nested	4.2906ms	0.3707ms	2.6975 KOps/s	2.7003 KOps/s	$\color{#d91a1a}-0.10\%$
test_lock_stack_nested	0.3589ms	0.3296ms	3.0340 KOps/s	2.9662 KOps/s	$\color{#35bf28}+2.29\%$
test_unlock_nested	0.6754ms	0.3082ms	3.2452 KOps/s	3.2498 KOps/s	$\color{#d91a1a}-0.14\%$
test_unlock_stack_nested	0.2960ms	0.2681ms	3.7306 KOps/s	3.6182 KOps/s	$\color{#35bf28}+3.11\%$
test_flatten_speed	0.1008ms	74.2637μs	13.4655 KOps/s	13.7707 KOps/s	$\color{#d91a1a}-2.22\%$
test_unflatten_speed	0.3363ms	0.2963ms	3.3744 KOps/s	3.4101 KOps/s	$\color{#d91a1a}-1.05\%$
test_common_ops	1.8887ms	0.6404ms	1.5615 KOps/s	1.5448 KOps/s	$\color{#35bf28}+1.08\%$
test_creation	90.7510μs	1.5453μs	647.1313 KOps/s	636.4122 KOps/s	$\color{#35bf28}+1.68\%$
test_creation_empty	38.6810μs	9.2334μs	108.3023 KOps/s	100.2170 KOps/s	$\textbf{\color{#35bf28}+8.07\%}$
test_creation_nested_1	35.7100μs	10.7814μs	92.7524 KOps/s	88.1412 KOps/s	$\textbf{\color{#35bf28}+5.23\%}$
test_creation_nested_2	48.1510μs	13.2427μs	75.5134 KOps/s	71.1644 KOps/s	$\textbf{\color{#35bf28}+6.11\%}$
test_clone	49.5710μs	10.6257μs	94.1118 KOps/s	85.7590 KOps/s	$\textbf{\color{#35bf28}+9.74\%}$
test_getitem[int]	1.2206ms	13.2988μs	75.1948 KOps/s	90.8770 KOps/s	$\textbf{\color{#d91a1a}-17.26\%}$
test_getitem[slice_int]	0.1309ms	21.0261μs	47.5598 KOps/s	43.6835 KOps/s	$\textbf{\color{#35bf28}+8.87\%}$
test_getitem[range]	0.1680ms	38.3270μs	26.0912 KOps/s	24.1770 KOps/s	$\textbf{\color{#35bf28}+7.92\%}$
test_getitem[tuple]	0.1356ms	19.2266μs	52.0113 KOps/s	50.4320 KOps/s	$\color{#35bf28}+3.13\%$
test_getitem[list]	0.1653ms	36.0170μs	27.7647 KOps/s	27.3131 KOps/s	$\color{#35bf28}+1.65\%$
test_setitem_dim[int]	47.8410μs	20.4318μs	48.9434 KOps/s	49.1172 KOps/s	$\color{#d91a1a}-0.35\%$
test_setitem_dim[slice_int]	63.8210μs	38.6267μs	25.8888 KOps/s	25.0303 KOps/s	$\color{#35bf28}+3.43\%$
test_setitem_dim[range]	85.1210μs	55.4543μs	18.0329 KOps/s	18.1447 KOps/s	$\color{#d91a1a}-0.62\%$
test_setitem_dim[tuple]	54.3810μs	33.3105μs	30.0205 KOps/s	30.1212 KOps/s	$\color{#d91a1a}-0.33\%$
test_setitem	62.4010μs	17.0580μs	58.6234 KOps/s	55.9332 KOps/s	$\color{#35bf28}+4.81\%$
test_set	63.0310μs	16.4939μs	60.6284 KOps/s	59.7202 KOps/s	$\color{#35bf28}+1.52\%$
test_set_shared	95.3387ms	0.1722ms	5.8076 KOps/s	6.6919 KOps/s	$\textbf{\color{#d91a1a}-13.21\%}$
test_update	0.3553ms	18.8016μs	53.1869 KOps/s	48.5116 KOps/s	$\textbf{\color{#35bf28}+9.64\%}$
test_update_nested	94.7520μs	23.4149μs	42.7079 KOps/s	39.2003 KOps/s	$\textbf{\color{#35bf28}+8.95\%}$
test_update__nested	0.5643ms	24.8802μs	40.1926 KOps/s	39.0443 KOps/s	$\color{#35bf28}+2.94\%$
test_set_nested	0.1006ms	16.3886μs	61.0181 KOps/s	55.0307 KOps/s	$\textbf{\color{#35bf28}+10.88\%}$
test_set_nested_new	0.1056ms	18.3891μs	54.3801 KOps/s	47.0101 KOps/s	$\textbf{\color{#35bf28}+15.68\%}$
test_select	0.1092ms	31.2788μs	31.9706 KOps/s	29.1927 KOps/s	$\textbf{\color{#35bf28}+9.52\%}$
test_select_nested	69.9010μs	42.8325μs	23.3467 KOps/s	23.6750 KOps/s	$\color{#d91a1a}-1.39\%$
test_exclude_nested	95.9710μs	59.8933μs	16.6964 KOps/s	16.3913 KOps/s	$\color{#35bf28}+1.86\%$
test_empty[True]	0.2940ms	0.2562ms	3.9027 KOps/s	3.8515 KOps/s	$\color{#35bf28}+1.33\%$
test_empty[False]	6.0531μs	0.7432μs	1.3456 MOps/s	1.3419 MOps/s	$\color{#35bf28}+0.28\%$
test_to	83.4510μs	53.2735μs	18.7711 KOps/s	17.8464 KOps/s	$\textbf{\color{#35bf28}+5.18\%}$
test_to_nonblocking	80.3610μs	45.9551μs	21.7604 KOps/s	21.1883 KOps/s	$\color{#35bf28}+2.70\%$
test_unbind_speed	0.2629ms	0.2329ms	4.2936 KOps/s	4.2487 KOps/s	$\color{#35bf28}+1.06\%$
test_unbind_speed_stack0	0.2806ms	0.2261ms	4.4229 KOps/s	4.2494 KOps/s	$\color{#35bf28}+4.08\%$
test_unbind_speed_stack1	92.5792ms	0.6391ms	1.5647 KOps/s	1.5370 KOps/s	$\color{#35bf28}+1.81\%$
test_split	94.7276ms	1.5807ms	632.6463 Ops/s	608.5633 Ops/s	$\color{#35bf28}+3.96\%$
test_chunk	95.9768ms	1.5845ms	631.1280 Ops/s	612.6860 Ops/s	$\color{#35bf28}+3.01\%$
test_consolidate[False-None]	96.0870ms	2.8861ms	346.4891 Ops/s	345.4974 Ops/s	$\color{#35bf28}+0.29\%$
test_consolidate[default-None]	1.7293ms	1.6470ms	607.1709 Ops/s	594.9969 Ops/s	$\color{#35bf28}+2.05\%$
test_consolidate[reduce-overhead-None]	1.7503ms	1.6721ms	598.0645 Ops/s	580.9537 Ops/s	$\color{#35bf28}+2.95\%$
test_consolidate_njt[False-None]	7.1445ms	6.5249ms	153.2597 Ops/s	149.7064 Ops/s	$\color{#35bf28}+2.37\%$
test_to[False-False-None]	1.8486ms	1.6844ms	593.6893 Ops/s	583.8211 Ops/s	$\color{#35bf28}+1.69\%$
test_to[True-False-None]	1.5261ms	1.2970ms	771.0042 Ops/s	748.7999 Ops/s	$\color{#35bf28}+2.97\%$
test_to[within-False-None]	4.1445ms	4.0437ms	247.2993 Ops/s	248.3316 Ops/s	$\color{#d91a1a}-0.42\%$
test_to[True-default-None]	5.1630ms	5.0354ms	198.5948 Ops/s	192.2019 Ops/s	$\color{#35bf28}+3.33\%$
test_to_njt[False-False-None]	7.1123ms	6.9650ms	143.5753 Ops/s	141.2409 Ops/s	$\color{#35bf28}+1.65\%$
test_to_njt[True-False-None]	5.7020ms	5.4245ms	184.3487 Ops/s	179.0725 Ops/s	$\color{#35bf28}+2.95\%$
test_to_njt[within-False-None]	12.0659ms	11.9547ms	83.6494 Ops/s	82.5053 Ops/s	$\color{#35bf28}+1.39\%$
test_creation[device0]	0.5326ms	81.4691μs	12.2746 KOps/s	12.0127 KOps/s	$\color{#35bf28}+2.18\%$
test_creation_from_tensor	0.5507ms	85.6322μs	11.6779 KOps/s	11.5392 KOps/s	$\color{#35bf28}+1.20\%$
test_add_one[memmap_tensor0]	0.3873ms	7.0960μs	140.9244 KOps/s	131.3761 KOps/s	$\textbf{\color{#35bf28}+7.27\%}$
test_contiguous[memmap_tensor0]	4.9496μs	0.4249μs	2.3538 MOps/s	2.3760 MOps/s	$\color{#d91a1a}-0.94\%$
test_stack[memmap_tensor0]	37.1100μs	4.5522μs	219.6733 KOps/s	205.2211 KOps/s	$\textbf{\color{#35bf28}+7.04\%}$
test_memmaptd_index	1.7783ms	0.2529ms	3.9538 KOps/s	3.9023 KOps/s	$\color{#35bf28}+1.32\%$
test_memmaptd_index_astensor	1.0169ms	0.3107ms	3.2184 KOps/s	3.1334 KOps/s	$\color{#35bf28}+2.71\%$
test_memmaptd_index_op	1.0314ms	0.6117ms	1.6349 KOps/s	1.5234 KOps/s	$\textbf{\color{#35bf28}+7.32\%}$
test_serialize_model	0.1319s	0.1301s	7.6863 Ops/s	7.6650 Ops/s	$\color{#35bf28}+0.28\%$
test_serialize_model_pickle	1.3518s	1.1845s	0.8442 Ops/s	0.8248 Ops/s	$\color{#35bf28}+2.35\%$
test_serialize_weights	0.1310s	0.1295s	7.7238 Ops/s	7.6932 Ops/s	$\color{#35bf28}+0.40\%$
test_serialize_weights_returnearly	0.6704s	78.2183ms	12.7847 Ops/s	10.6848 Ops/s	$\textbf{\color{#35bf28}+19.65\%}$
test_serialize_weights_pickle	1.3770s	1.2300s	0.8130 Ops/s	0.8386 Ops/s	$\color{#d91a1a}-3.05\%$
test_reshape_pytree	54.6310μs	22.1152μs	45.2177 KOps/s	43.0544 KOps/s	$\textbf{\color{#35bf28}+5.02\%}$
test_reshape_td	50.7710μs	26.5533μs	37.6601 KOps/s	35.5291 KOps/s	$\textbf{\color{#35bf28}+6.00\%}$
test_view_pytree	50.1310μs	22.1444μs	45.1582 KOps/s	44.5141 KOps/s	$\color{#35bf28}+1.45\%$
test_view_td	55.4310μs	28.5902μs	34.9770 KOps/s	30.7899 KOps/s	$\textbf{\color{#35bf28}+13.60\%}$
test_unbind_pytree	55.7010μs	27.9312μs	35.8023 KOps/s	34.9078 KOps/s	$\color{#35bf28}+2.56\%$
test_unbind_td	0.7128ms	34.8128μs	28.7250 KOps/s	27.7005 KOps/s	$\color{#35bf28}+3.70\%$
test_split_pytree	56.3910μs	29.7212μs	33.6460 KOps/s	32.5425 KOps/s	$\color{#35bf28}+3.39\%$
test_split_td	0.8748ms	37.6192μs	26.5822 KOps/s	24.7648 KOps/s	$\textbf{\color{#35bf28}+7.34\%}$
test_add_pytree	66.6820μs	34.6559μs	28.8551 KOps/s	26.9537 KOps/s	$\textbf{\color{#35bf28}+7.05\%}$
test_add_td	88.9210μs	48.4546μs	20.6379 KOps/s	18.4875 KOps/s	$\textbf{\color{#35bf28}+11.63\%}$
test_compile_add_one_nested[tensordict-compile]	0.1789ms	0.1197ms	8.3560 KOps/s	7.7308 KOps/s	$\textbf{\color{#35bf28}+8.09\%}$
test_compile_add_one_nested[tensordict-eager]	0.2244ms	0.1234ms	8.1059 KOps/s	7.7382 KOps/s	$\color{#35bf28}+4.75\%$
test_compile_add_one_nested[pytree-compile]	0.3780ms	98.8970μs	10.1115 KOps/s	9.9652 KOps/s	$\color{#35bf28}+1.47\%$
test_compile_add_one_nested[pytree-eager]	1.6856ms	0.1521ms	6.5735 KOps/s	6.5971 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_copy_nested[tensordict-compile]	67.3310μs	21.3734μs	46.7872 KOps/s	41.3892 KOps/s	$\textbf{\color{#35bf28}+13.04\%}$
test_compile_copy_nested[tensordict-eager]	72.2710μs	28.1087μs	35.5762 KOps/s	34.3744 KOps/s	$\color{#35bf28}+3.50\%$
test_compile_copy_nested[pytree-compile]	0.2102ms	69.6557μs	14.3563 KOps/s	14.1860 KOps/s	$\color{#35bf28}+1.20\%$
test_compile_copy_nested[pytree-eager]	80.3020μs	49.7839μs	20.0868 KOps/s	19.8045 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_add_one_flat[tensordict-compile]	0.2147ms	0.1454ms	6.8773 KOps/s	6.9563 KOps/s	$\color{#d91a1a}-1.14\%$
test_compile_add_one_flat[tensordict-eager]	0.2868ms	0.2071ms	4.8293 KOps/s	4.8313 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_add_one_flat[tensorclass-compile]	0.1839ms	0.1022ms	9.7895 KOps/s	9.7045 KOps/s	$\color{#35bf28}+0.88\%$
test_compile_add_one_flat[tensorclass-eager]	0.1620ms	51.0702μs	19.5809 KOps/s	18.1373 KOps/s	$\textbf{\color{#35bf28}+7.96\%}$
test_compile_add_one_flat[pytree-compile]	0.1931ms	0.1397ms	7.1573 KOps/s	7.1447 KOps/s	$\color{#35bf28}+0.18\%$
test_compile_add_one_flat[pytree-eager]	0.5677ms	0.4867ms	2.0546 KOps/s	2.0697 KOps/s	$\color{#d91a1a}-0.73\%$
test_compile_add_self_flat[tensordict-eager]	0.3463ms	0.2475ms	4.0403 KOps/s	4.0505 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_self_flat[tensordict-compile]	0.2277ms	0.1493ms	6.6995 KOps/s	6.9559 KOps/s	$\color{#d91a1a}-3.69\%$
test_compile_add_self_flat[tensorclass-eager]	0.3001ms	61.3096μs	16.3107 KOps/s	15.5173 KOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1473ms	0.1008ms	9.9201 KOps/s	9.8379 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_add_self_flat[pytree-eager]	0.4477ms	0.4049ms	2.4699 KOps/s	2.4877 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_add_self_flat[pytree-compile]	0.2002ms	0.1395ms	7.1678 KOps/s	7.2594 KOps/s	$\color{#d91a1a}-1.26\%$
test_compile_copy_flat[tensordict-compile]	54.7210μs	17.9147μs	55.8201 KOps/s	53.7170 KOps/s	$\color{#35bf28}+3.92\%$
test_compile_copy_flat[tensordict-eager]	0.1380ms	28.6936μs	34.8509 KOps/s	35.0530 KOps/s	$\color{#d91a1a}-0.58\%$
test_compile_copy_flat[pytree-compile]	0.1179ms	76.0345μs	13.1519 KOps/s	13.2574 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_copy_flat[pytree-eager]	0.1015ms	52.2173μs	19.1507 KOps/s	19.5336 KOps/s	$\color{#d91a1a}-1.96\%$
test_compile_assign_and_add[tensordict-compile]	1.7408ms	0.4138ms	2.4167 KOps/s	2.1803 KOps/s	$\textbf{\color{#35bf28}+10.84\%}$
test_compile_assign_and_add[tensordict-eager]	2.7426ms	2.5953ms	385.3076 Ops/s	386.9848 Ops/s	$\color{#d91a1a}-0.43\%$
test_compile_assign_and_add[pytree-compile]	1.6271ms	0.4408ms	2.2687 KOps/s	2.2148 KOps/s	$\color{#35bf28}+2.43\%$
test_compile_assign_and_add[pytree-eager]	2.7322ms	2.6717ms	374.2902 Ops/s	376.3387 Ops/s	$\color{#d91a1a}-0.54\%$
test_compile_indexing[tensor-tensordict-compile]	0.1717ms	0.1211ms	8.2564 KOps/s	8.8211 KOps/s	$\textbf{\color{#d91a1a}-6.40\%}$
test_compile_indexing[tensor-tensordict-eager]	0.5598ms	85.0404μs	11.7591 KOps/s	12.4538 KOps/s	$\textbf{\color{#d91a1a}-5.58\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.2885ms	0.1129ms	8.8536 KOps/s	9.4925 KOps/s	$\textbf{\color{#d91a1a}-6.73\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.1165ms	73.7880μs	13.5523 KOps/s	14.7090 KOps/s	$\textbf{\color{#d91a1a}-7.86\%}$
test_compile_indexing[tensor-pytree-compile]	0.1890ms	0.1145ms	8.7344 KOps/s	9.4330 KOps/s	$\textbf{\color{#d91a1a}-7.41\%}$
test_compile_indexing[tensor-pytree-eager]	0.1152ms	73.7644μs	13.5567 KOps/s	14.6821 KOps/s	$\textbf{\color{#d91a1a}-7.67\%}$
test_compile_indexing[slice-tensordict-compile]	0.1492ms	0.1072ms	9.3309 KOps/s	9.8069 KOps/s	$\color{#d91a1a}-4.85\%$
test_compile_indexing[slice-tensordict-eager]	0.1480ms	17.2229μs	58.0623 KOps/s	54.9784 KOps/s	$\textbf{\color{#35bf28}+5.61\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1413ms	97.2671μs	10.2810 KOps/s	10.3368 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_indexing[slice-tensorclass-eager]	52.2310μs	15.9881μs	62.5467 KOps/s	61.8027 KOps/s	$\color{#35bf28}+1.20\%$
test_compile_indexing[slice-pytree-compile]	0.1469ms	0.1021ms	9.7961 KOps/s	10.2530 KOps/s	$\color{#d91a1a}-4.46\%$
test_compile_indexing[slice-pytree-eager]	49.8510μs	15.8629μs	63.0400 KOps/s	62.4610 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_indexing[int-tensordict-compile]	0.1494ms	0.1069ms	9.3530 KOps/s	9.7182 KOps/s	$\color{#d91a1a}-3.76\%$
test_compile_indexing[int-tensordict-eager]	0.5567ms	16.9538μs	58.9837 KOps/s	57.6120 KOps/s	$\color{#35bf28}+2.38\%$
test_compile_indexing[int-tensorclass-compile]	0.1480ms	0.1016ms	9.8463 KOps/s	9.8398 KOps/s	$\color{#35bf28}+0.07\%$
test_compile_indexing[int-tensorclass-eager]	46.4010μs	15.7831μs	63.3590 KOps/s	62.4366 KOps/s	$\color{#35bf28}+1.48\%$
test_compile_indexing[int-pytree-compile]	0.2083ms	0.1023ms	9.7790 KOps/s	10.2562 KOps/s	$\color{#d91a1a}-4.65\%$
test_compile_indexing[int-pytree-eager]	0.3938ms	15.7643μs	63.4346 KOps/s	62.3605 KOps/s	$\color{#35bf28}+1.72\%$
test_mod_add[eager]	97.1620μs	33.7437μs	29.6352 KOps/s	31.4571 KOps/s	$\textbf{\color{#d91a1a}-5.79\%}$
test_mod_add[compile]	0.3895ms	76.2549μs	13.1139 KOps/s	13.1525 KOps/s	$\color{#d91a1a}-0.29\%$
test_mod_add[compile-overhead]	0.3176ms	0.1646ms	6.0753 KOps/s	5.6078 KOps/s	$\textbf{\color{#35bf28}+8.34\%}$
test_mod_wrap[eager]	0.3237ms	0.2427ms	4.1197 KOps/s	4.0637 KOps/s	$\color{#35bf28}+1.38\%$
test_mod_wrap[compile]	1.6611ms	0.2835ms	3.5274 KOps/s	3.5233 KOps/s	$\color{#35bf28}+0.12\%$
test_mod_wrap[compile-overhead]	7.6567ms	3.9756ms	251.5329 Ops/s	241.3766 Ops/s	$\color{#35bf28}+4.21\%$
test_mod_wrap_and_backward[eager]	1.4921ms	1.3768ms	726.3121 Ops/s	683.9355 Ops/s	$\textbf{\color{#35bf28}+6.20\%}$
test_mod_wrap_and_backward[compile]	1.4063ms	1.2681ms	788.5591 Ops/s	718.0035 Ops/s	$\textbf{\color{#35bf28}+9.83\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3945ms	0.9304ms	1.0748 KOps/s	957.3607 Ops/s	$\textbf{\color{#35bf28}+12.26\%}$
test_seq_add[eager]	0.1897ms	97.9045μs	10.2140 KOps/s	10.0615 KOps/s	$\color{#35bf28}+1.52\%$
test_seq_add[compile]	0.1644ms	87.8974μs	11.3769 KOps/s	11.5329 KOps/s	$\color{#d91a1a}-1.35\%$
test_seq_add[compile-overhead]	0.3348ms	0.1316ms	7.6006 KOps/s	7.7062 KOps/s	$\color{#d91a1a}-1.37\%$
test_seq_wrap[eager]	0.7072ms	0.3912ms	2.5565 KOps/s	2.5256 KOps/s	$\color{#35bf28}+1.22\%$
test_seq_wrap[compile]	0.5586ms	0.3013ms	3.3193 KOps/s	3.3095 KOps/s	$\color{#35bf28}+0.30\%$
test_seq_wrap[compile-overhead]	0.5417ms	0.2323ms	4.3050 KOps/s	4.4113 KOps/s	$\color{#d91a1a}-2.41\%$
test_func_call_runtime[False-eager]	1.0707ms	0.7797ms	1.2825 KOps/s	1.3527 KOps/s	$\textbf{\color{#d91a1a}-5.19\%}$
test_func_call_runtime[False-compile]	0.8630ms	0.7476ms	1.3377 KOps/s	1.3356 KOps/s	$\color{#35bf28}+0.15\%$
test_func_call_runtime[False-compile-overhead]	0.4120ms	0.3641ms	2.7464 KOps/s	2.7241 KOps/s	$\color{#35bf28}+0.82\%$
test_func_call_runtime[True-eager]	0.9514ms	0.8870ms	1.1274 KOps/s	1.1097 KOps/s	$\color{#35bf28}+1.59\%$
test_func_call_runtime[True-compile]	0.8933ms	0.7757ms	1.2892 KOps/s	1.3013 KOps/s	$\color{#d91a1a}-0.93\%$
test_func_call_runtime[True-compile-overhead]	0.4381ms	0.3879ms	2.5783 KOps/s	2.5632 KOps/s	$\color{#35bf28}+0.59\%$
test_func_call_cm_runtime[False-eager]	0.8891ms	0.7863ms	1.2718 KOps/s	1.3602 KOps/s	$\textbf{\color{#d91a1a}-6.49\%}$
test_func_call_cm_runtime[False-compile]	0.9606ms	0.7612ms	1.3137 KOps/s	1.3280 KOps/s	$\color{#d91a1a}-1.07\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4561ms	0.3761ms	2.6586 KOps/s	2.7110 KOps/s	$\color{#d91a1a}-1.93\%$
test_func_call_cm_runtime[True-eager]	1.3056ms	1.0466ms	955.4724 Ops/s	978.0637 Ops/s	$\color{#d91a1a}-2.31\%$
test_func_call_cm_runtime[True-compile]	0.8578ms	0.8003ms	1.2496 KOps/s	1.2502 KOps/s	$\color{#d91a1a}-0.05\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5403ms	0.4146ms	2.4119 KOps/s	2.3981 KOps/s	$\color{#35bf28}+0.58\%$
test_vmap_func_call_cm_runtime[eager]	2.5434ms	2.0839ms	479.8584 Ops/s	482.5549 Ops/s	$\color{#d91a1a}-0.56\%$
test_vmap_func_call_cm_runtime[compile]	0.9170ms	0.8332ms	1.2002 KOps/s	1.2381 KOps/s	$\color{#d91a1a}-3.06\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4859ms	0.4155ms	2.4065 KOps/s	2.3731 KOps/s	$\color{#35bf28}+1.41\%$
test_distributed	2.5490ms	0.1717ms	5.8226 KOps/s	8.3394 KOps/s	$\textbf{\color{#d91a1a}-30.18\%}$
test_tdmodule	0.2972ms	14.6821μs	68.1100 KOps/s	66.1430 KOps/s	$\color{#35bf28}+2.97\%$
test_tdmodule_dispatch	80.5620μs	28.7750μs	34.7524 KOps/s	34.2272 KOps/s	$\color{#35bf28}+1.53\%$
test_tdseq	36.7710μs	15.7565μs	63.4658 KOps/s	61.2715 KOps/s	$\color{#35bf28}+3.58\%$
test_tdseq_dispatch	56.2010μs	31.7960μs	31.4505 KOps/s	30.2025 KOps/s	$\color{#35bf28}+4.13\%$
test_instantiation_functorch	2.0608ms	1.5686ms	637.5246 Ops/s	634.3214 Ops/s	$\color{#35bf28}+0.50\%$
test_exec_functorch	0.1995ms	0.1497ms	6.6783 KOps/s	6.5446 KOps/s	$\color{#35bf28}+2.04\%$
test_exec_functional_call	0.2580ms	0.1442ms	6.9353 KOps/s	6.8871 KOps/s	$\color{#35bf28}+0.70\%$
test_exec_td_decorator	0.3774ms	0.1878ms	5.3241 KOps/s	5.1754 KOps/s	$\color{#35bf28}+2.87\%$
test_vmap_mlp_speed_decorator[True-True]	0.7558ms	0.6738ms	1.4841 KOps/s	1.4842 KOps/s	$-0.00\%$
test_vmap_mlp_speed_decorator[True-False]	0.8130ms	0.6700ms	1.4926 KOps/s	1.4827 KOps/s	$\color{#35bf28}+0.66\%$
test_vmap_mlp_speed_decorator[False-True]	0.7238ms	0.5905ms	1.6934 KOps/s	1.6892 KOps/s	$\color{#35bf28}+0.25\%$
test_vmap_mlp_speed_decorator[False-False]	0.7335ms	0.6120ms	1.6339 KOps/s	1.6926 KOps/s	$\color{#d91a1a}-3.47\%$
test_vmap_transformer_speed_decorator[True-True]	19.2241ms	19.0928ms	52.3759 Ops/s	52.6653 Ops/s	$\color{#d91a1a}-0.55\%$
test_vmap_transformer_speed_decorator[True-False]	19.8462ms	19.1041ms	52.3447 Ops/s	52.5021 Ops/s	$\color{#d91a1a}-0.30\%$
test_vmap_transformer_speed_decorator[False-True]	19.9397ms	19.2522ms	51.9422 Ops/s	53.0820 Ops/s	$\color{#d91a1a}-2.15\%$
test_vmap_transformer_speed_decorator[False-False]	19.6862ms	19.0514ms	52.4896 Ops/s	52.8457 Ops/s	$\color{#d91a1a}-0.67\%$
test_to_module_speed[True]	1.0977ms	0.9614ms	1.0401 KOps/s	1.0554 KOps/s	$\color{#d91a1a}-1.45\%$
test_to_module_speed[False]	1.3281ms	0.9311ms	1.0740 KOps/s	1.0682 KOps/s	$\color{#35bf28}+0.54\%$
test_tc_init	73.7420μs	37.5598μs	26.6242 KOps/s	28.0162 KOps/s	$\color{#d91a1a}-4.97\%$
test_tc_init_nested	0.1134ms	73.8422μs	13.5424 KOps/s	13.6544 KOps/s	$\color{#d91a1a}-0.82\%$
test_tc_first_layer_tensor	4.4257μs	0.7380μs	1.3550 MOps/s	1.3640 MOps/s	$\color{#d91a1a}-0.66\%$
test_tc_first_layer_nontensor	41.2310μs	2.4919μs	401.3023 KOps/s	394.5425 KOps/s	$\color{#35bf28}+1.71\%$
test_tc_second_layer_tensor	15.8437μs	1.4869μs	672.5432 KOps/s	667.1205 KOps/s	$\color{#35bf28}+0.81\%$
test_tc_second_layer_nontensor	27.4210μs	3.2436μs	308.2966 KOps/s	301.3653 KOps/s	$\color{#35bf28}+2.30\%$
test_unbind	0.2267s	10.0019ms	99.9805 Ops/s	146.3851 Ops/s	$\textbf{\color{#d91a1a}-31.70\%}$
test_full_like	9.6180ms	9.1185ms	109.6673 Ops/s	106.8217 Ops/s	$\color{#35bf28}+2.66\%$
test_zeros_like	5.4753ms	4.3390ms	230.4679 Ops/s	138.2449 Ops/s	$\textbf{\color{#35bf28}+66.71\%}$
test_ones_like	4.9430ms	4.2588ms	234.8080 Ops/s	230.9902 Ops/s	$\color{#35bf28}+1.65\%$
test_clone	6.7035ms	6.3579ms	157.2855 Ops/s	157.1502 Ops/s	$\color{#35bf28}+0.09\%$
test_squeeze	59.3010μs	9.5580μs	104.6245 KOps/s	109.8238 KOps/s	$\color{#d91a1a}-4.73\%$
test_unsqueeze	0.1955ms	70.4400μs	14.1965 KOps/s	14.2449 KOps/s	$\color{#d91a1a}-0.34\%$
test_split	0.4048ms	0.1592ms	6.2811 KOps/s	6.3721 KOps/s	$\color{#d91a1a}-1.43\%$
test_permute	0.2280ms	0.1787ms	5.5945 KOps/s	5.6207 KOps/s	$\color{#d91a1a}-0.47\%$
test_stack	51.0353ms	50.8585ms	19.6624 Ops/s	19.5779 Ops/s	$\color{#35bf28}+0.43\%$
test_cat	51.0622ms	50.7161ms	19.7176 Ops/s	23.5163 Ops/s	$\textbf{\color{#d91a1a}-16.15\%}$