[Feature] Non-blocking for consolidated TD

[vmoens]

Stack from ghstack (oldest at bottom):

• [Refactor] Put values, lengths and offsets of NJTs together in storage #1023
• [Feature] TD+NJT to(device) support #1022
• [Feature] NJT with lengths #1021
• -> [Feature] Non-blocking for consolidated TD #1020

[github-actions]

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

Total Benchmarks: 222. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}29$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	52.6380μs	26.0287μs	38.4192 KOps/s	40.6917 KOps/s	$\textbf{\color{#d91a1a}-5.58\%}$
test_plain_set_stack_nested	57.8880μs	26.2237μs	38.1335 KOps/s	40.4269 KOps/s	$\textbf{\color{#d91a1a}-5.67\%}$
test_plain_set_nested_inplace	62.8380μs	28.5825μs	34.9865 KOps/s	37.0205 KOps/s	$\textbf{\color{#d91a1a}-5.49\%}$
test_plain_set_stack_nested_inplace	67.0860μs	28.2036μs	35.4564 KOps/s	36.5039 KOps/s	$\color{#d91a1a}-2.87\%$
test_items	28.7540μs	4.2781μs	233.7487 KOps/s	229.8629 KOps/s	$\color{#35bf28}+1.69\%$
test_items_nested	0.5258ms	0.3806ms	2.6274 KOps/s	2.5650 KOps/s	$\color{#35bf28}+2.43\%$
test_items_nested_locked	0.7185ms	0.3853ms	2.5951 KOps/s	2.5656 KOps/s	$\color{#35bf28}+1.15\%$
test_items_nested_leaf	0.1471ms	80.9546μs	12.3526 KOps/s	12.1977 KOps/s	$\color{#35bf28}+1.27\%$
test_items_stack_nested	0.8131ms	0.3890ms	2.5709 KOps/s	2.5545 KOps/s	$\color{#35bf28}+0.64\%$
test_items_stack_nested_leaf	0.1494ms	84.8068μs	11.7915 KOps/s	11.7265 KOps/s	$\color{#35bf28}+0.55\%$
test_items_stack_nested_locked	0.7096ms	0.3906ms	2.5604 KOps/s	2.5619 KOps/s	$\color{#d91a1a}-0.06\%$
test_keys	28.6330μs	3.5929μs	278.3267 KOps/s	281.1507 KOps/s	$\color{#d91a1a}-1.00\%$
test_keys_nested	0.2293ms	0.1368ms	7.3077 KOps/s	7.2484 KOps/s	$\color{#35bf28}+0.82\%$
test_keys_nested_locked	0.7032ms	0.1429ms	6.9983 KOps/s	6.9209 KOps/s	$\color{#35bf28}+1.12\%$
test_keys_nested_leaf	0.1957ms	0.1197ms	8.3519 KOps/s	8.2171 KOps/s	$\color{#35bf28}+1.64\%$
test_keys_stack_nested	0.2760ms	0.1373ms	7.2838 KOps/s	7.2415 KOps/s	$\color{#35bf28}+0.58\%$
test_keys_stack_nested_leaf	0.2079ms	0.1200ms	8.3351 KOps/s	8.3472 KOps/s	$\color{#d91a1a}-0.14\%$
test_keys_stack_nested_locked	0.3405ms	0.1439ms	6.9472 KOps/s	6.9497 KOps/s	$\color{#d91a1a}-0.03\%$
test_values	5.9972μs	1.0577μs	945.4583 KOps/s	919.1975 KOps/s	$\color{#35bf28}+2.86\%$
test_values_nested	0.1773ms	94.0094μs	10.6372 KOps/s	10.5050 KOps/s	$\color{#35bf28}+1.26\%$
test_values_nested_locked	0.1580ms	92.7870μs	10.7774 KOps/s	10.4981 KOps/s	$\color{#35bf28}+2.66\%$
test_values_nested_leaf	0.1757ms	79.2858μs	12.6126 KOps/s	12.3476 KOps/s	$\color{#35bf28}+2.15\%$
test_values_stack_nested	0.1449ms	94.0777μs	10.6295 KOps/s	10.4799 KOps/s	$\color{#35bf28}+1.43\%$
test_values_stack_nested_leaf	0.1554ms	79.5996μs	12.5629 KOps/s	12.5748 KOps/s	$\color{#d91a1a}-0.10\%$
test_values_stack_nested_locked	0.1576ms	92.7963μs	10.7763 KOps/s	10.5382 KOps/s	$\color{#35bf28}+2.26\%$
test_membership	6.1111μs	0.7381μs	1.3548 MOps/s	1.3632 MOps/s	$\color{#d91a1a}-0.62\%$
test_membership_nested	18.9050μs	2.7518μs	363.4006 KOps/s	363.8772 KOps/s	$\color{#d91a1a}-0.13\%$
test_membership_nested_leaf	30.6170μs	2.7850μs	359.0662 KOps/s	360.5264 KOps/s	$\color{#d91a1a}-0.41\%$
test_membership_stacked_nested	26.4690μs	2.7764μs	360.1734 KOps/s	360.4654 KOps/s	$\color{#d91a1a}-0.08\%$
test_membership_stacked_nested_leaf	23.4940μs	2.8086μs	356.0519 KOps/s	345.5350 KOps/s	$\color{#35bf28}+3.04\%$
test_membership_nested_last	45.8680μs	4.2825μs	233.5079 KOps/s	237.9729 KOps/s	$\color{#d91a1a}-1.88\%$
test_membership_nested_leaf_last	98.3230μs	4.3631μs	229.1928 KOps/s	238.5868 KOps/s	$\color{#d91a1a}-3.94\%$
test_membership_stacked_nested_last	33.5720μs	4.2633μs	234.5589 KOps/s	169.5345 KOps/s	$\textbf{\color{#35bf28}+38.35\%}$
test_membership_stacked_nested_leaf_last	32.9610μs	4.2871μs	233.2566 KOps/s	169.9251 KOps/s	$\textbf{\color{#35bf28}+37.27\%}$
test_nested_getleaf	38.1020μs	10.8945μs	91.7892 KOps/s	92.8809 KOps/s	$\color{#d91a1a}-1.18\%$
test_nested_get	36.4580μs	10.5377μs	94.8972 KOps/s	98.2193 KOps/s	$\color{#d91a1a}-3.38\%$
test_stacked_getleaf	45.4150μs	10.9887μs	91.0025 KOps/s	94.3366 KOps/s	$\color{#d91a1a}-3.53\%$
test_stacked_get	36.9490μs	10.4774μs	95.4433 KOps/s	98.3367 KOps/s	$\color{#d91a1a}-2.94\%$
test_nested_getitemleaf	37.5610μs	11.4004μs	87.7163 KOps/s	89.3315 KOps/s	$\color{#d91a1a}-1.81\%$
test_nested_getitem	40.7460μs	10.3505μs	96.6136 KOps/s	96.0038 KOps/s	$\color{#35bf28}+0.64\%$
test_stacked_getitemleaf	35.8670μs	11.4276μs	87.5075 KOps/s	88.8264 KOps/s	$\color{#d91a1a}-1.48\%$
test_stacked_getitem	36.0570μs	10.2864μs	97.2157 KOps/s	95.8853 KOps/s	$\color{#35bf28}+1.39\%$
test_lock_nested	85.5422ms	0.5979ms	1.6725 KOps/s	1.9676 KOps/s	$\textbf{\color{#d91a1a}-15.00\%}$
test_lock_stack_nested	0.7483ms	0.4849ms	2.0622 KOps/s	2.1296 KOps/s	$\color{#d91a1a}-3.16\%$
test_unlock_nested	85.8721ms	0.5128ms	1.9501 KOps/s	2.3770 KOps/s	$\textbf{\color{#d91a1a}-17.96\%}$
test_unlock_stack_nested	0.7134ms	0.3960ms	2.5250 KOps/s	2.6108 KOps/s	$\color{#d91a1a}-3.28\%$
test_flatten_speed	0.2063ms	0.1018ms	9.8190 KOps/s	9.8980 KOps/s	$\color{#d91a1a}-0.80\%$
test_unflatten_speed	0.7012ms	0.5306ms	1.8846 KOps/s	1.9174 KOps/s	$\color{#d91a1a}-1.71\%$
test_common_ops	6.3955ms	1.2181ms	820.9513 Ops/s	862.7404 Ops/s	$\color{#d91a1a}-4.84\%$
test_creation	36.0880μs	2.0935μs	477.6626 KOps/s	469.3700 KOps/s	$\color{#35bf28}+1.77\%$
test_creation_empty	51.3560μs	20.3174μs	49.2190 KOps/s	54.8212 KOps/s	$\textbf{\color{#d91a1a}-10.22\%}$
test_creation_nested_1	74.6900μs	24.1624μs	41.3867 KOps/s	45.5840 KOps/s	$\textbf{\color{#d91a1a}-9.21\%}$
test_creation_nested_2	71.1840μs	28.0966μs	35.5915 KOps/s	37.1035 KOps/s	$\color{#d91a1a}-4.08\%$
test_clone	0.1461ms	17.5250μs	57.0614 KOps/s	58.0671 KOps/s	$\color{#d91a1a}-1.73\%$
test_getitem[int]	1.1730ms	17.5475μs	56.9881 KOps/s	59.7277 KOps/s	$\color{#d91a1a}-4.59\%$
test_getitem[slice_int]	0.1571ms	33.0245μs	30.2805 KOps/s	32.1597 KOps/s	$\textbf{\color{#d91a1a}-5.84\%}$
test_getitem[range]	0.1877ms	59.0228μs	16.9426 KOps/s	16.8939 KOps/s	$\color{#35bf28}+0.29\%$
test_getitem[tuple]	0.1450ms	26.0494μs	38.3885 KOps/s	39.3414 KOps/s	$\color{#d91a1a}-2.42\%$
test_getitem[list]	0.1645ms	54.8637μs	18.2270 KOps/s	18.2725 KOps/s	$\color{#d91a1a}-0.25\%$
test_setitem_dim[int]	66.8450μs	34.0141μs	29.3995 KOps/s	29.0570 KOps/s	$\color{#35bf28}+1.18\%$
test_setitem_dim[slice_int]	0.1133ms	62.6627μs	15.9584 KOps/s	15.8745 KOps/s	$\color{#35bf28}+0.53\%$
test_setitem_dim[range]	0.1446ms	85.6468μs	11.6759 KOps/s	11.4693 KOps/s	$\color{#35bf28}+1.80\%$
test_setitem_dim[tuple]	87.4130μs	50.1390μs	19.9446 KOps/s	19.8189 KOps/s	$\color{#35bf28}+0.63\%$
test_setitem	92.3130μs	32.4241μs	30.8412 KOps/s	33.1869 KOps/s	$\textbf{\color{#d91a1a}-7.07\%}$
test_set	76.3020μs	31.2186μs	32.0321 KOps/s	33.9443 KOps/s	$\textbf{\color{#d91a1a}-5.63\%}$
test_set_shared	3.8809ms	0.2220ms	4.5042 KOps/s	4.5006 KOps/s	$\color{#35bf28}+0.08\%$
test_update	0.1403ms	40.8770μs	24.4636 KOps/s	26.5258 KOps/s	$\textbf{\color{#d91a1a}-7.77\%}$
test_update_nested	0.1943ms	51.9247μs	19.2587 KOps/s	20.3695 KOps/s	$\textbf{\color{#d91a1a}-5.45\%}$
test_update__nested	1.0118ms	38.1497μs	26.2125 KOps/s	26.6678 KOps/s	$\color{#d91a1a}-1.71\%$
test_set_nested	83.0040μs	35.2756μs	28.3482 KOps/s	30.3337 KOps/s	$\textbf{\color{#d91a1a}-6.55\%}$
test_set_nested_new	96.5300μs	40.1146μs	24.9286 KOps/s	26.2681 KOps/s	$\textbf{\color{#d91a1a}-5.10\%}$
test_select	0.2446ms	57.4027μs	17.4208 KOps/s	17.9137 KOps/s	$\color{#d91a1a}-2.75\%$
test_select_nested	0.1214ms	60.3048μs	16.5824 KOps/s	16.7976 KOps/s	$\color{#d91a1a}-1.28\%$
test_exclude_nested	0.1673ms	75.9072μs	13.1740 KOps/s	13.2284 KOps/s	$\color{#d91a1a}-0.41\%$
test_empty[True]	0.6688ms	0.3620ms	2.7622 KOps/s	2.7594 KOps/s	$\color{#35bf28}+0.10\%$
test_empty[False]	9.6102μs	1.2536μs	797.7187 KOps/s	723.7633 KOps/s	$\textbf{\color{#35bf28}+10.22\%}$
test_unbind_speed	0.5191ms	0.3080ms	3.2472 KOps/s	3.2847 KOps/s	$\color{#d91a1a}-1.14\%$
test_unbind_speed_stack0	0.9010ms	0.3087ms	3.2399 KOps/s	3.4468 KOps/s	$\textbf{\color{#d91a1a}-6.00\%}$
test_unbind_speed_stack1	90.1346ms	0.8208ms	1.2183 KOps/s	1.3794 KOps/s	$\textbf{\color{#d91a1a}-11.68\%}$
test_split	3.2966ms	2.0163ms	495.9555 Ops/s	455.6154 Ops/s	$\textbf{\color{#35bf28}+8.85\%}$
test_chunk	94.9570ms	2.2056ms	453.3907 Ops/s	451.2668 Ops/s	$\color{#35bf28}+0.47\%$
test_creation[device0]	0.2748ms	0.1192ms	8.3878 KOps/s	8.2238 KOps/s	$\color{#35bf28}+2.00\%$
test_creation_from_tensor	4.0891ms	0.1217ms	8.2147 KOps/s	8.4289 KOps/s	$\color{#d91a1a}-2.54\%$
test_add_one[memmap_tensor0]	0.6369ms	7.1150μs	140.5490 KOps/s	132.4672 KOps/s	$\textbf{\color{#35bf28}+6.10\%}$
test_contiguous[memmap_tensor0]	20.3380μs	1.8728μs	533.9637 KOps/s	517.3099 KOps/s	$\color{#35bf28}+3.22\%$
test_stack[memmap_tensor0]	29.3550μs	5.6265μs	177.7311 KOps/s	174.0301 KOps/s	$\color{#35bf28}+2.13\%$
test_memmaptd_index	1.2025ms	0.4118ms	2.4284 KOps/s	2.3429 KOps/s	$\color{#35bf28}+3.65\%$
test_memmaptd_index_astensor	1.0297ms	0.5124ms	1.9517 KOps/s	1.8891 KOps/s	$\color{#35bf28}+3.31\%$
test_memmaptd_index_op	2.2664ms	1.1027ms	906.8953 Ops/s	942.4719 Ops/s	$\color{#d91a1a}-3.77\%$
test_serialize_model	0.1233s	0.1179s	8.4826 Ops/s	8.5199 Ops/s	$\color{#d91a1a}-0.44\%$
test_serialize_model_pickle	0.4269s	0.4006s	2.4962 Ops/s	2.5256 Ops/s	$\color{#d91a1a}-1.16\%$
test_serialize_weights	0.1311s	0.1170s	8.5464 Ops/s	8.7400 Ops/s	$\color{#d91a1a}-2.22\%$
test_serialize_weights_returnearly	0.2814s	0.1722s	5.8061 Ops/s	5.5995 Ops/s	$\color{#35bf28}+3.69\%$
test_serialize_weights_pickle	1.0524s	0.6660s	1.5015 Ops/s	2.3733 Ops/s	$\textbf{\color{#d91a1a}-36.73\%}$
test_serialize_weights_filesystem	0.1500s	0.1407s	7.1092 Ops/s	7.1684 Ops/s	$\color{#d91a1a}-0.83\%$
test_serialize_model_filesystem	0.1513s	0.1411s	7.0847 Ops/s	6.7164 Ops/s	$\textbf{\color{#35bf28}+5.48\%}$
test_reshape_pytree	0.1168ms	40.8407μs	24.4854 KOps/s	25.4148 KOps/s	$\color{#d91a1a}-3.66\%$
test_reshape_td	0.1084ms	46.6729μs	21.4257 KOps/s	20.9636 KOps/s	$\color{#35bf28}+2.20\%$
test_view_pytree	0.1035ms	40.3886μs	24.7595 KOps/s	25.3585 KOps/s	$\color{#d91a1a}-2.36\%$
test_view_td	98.0240μs	52.3517μs	19.1016 KOps/s	18.7751 KOps/s	$\color{#35bf28}+1.74\%$
test_unbind_pytree	70.8730μs	37.4719μs	26.6867 KOps/s	27.6753 KOps/s	$\color{#d91a1a}-3.57\%$
test_unbind_td	0.3159ms	46.6561μs	21.4334 KOps/s	22.0447 KOps/s	$\color{#d91a1a}-2.77\%$
test_split_pytree	84.0770μs	39.7916μs	25.1310 KOps/s	26.1449 KOps/s	$\color{#d91a1a}-3.88\%$
test_split_td	84.2913ms	69.4311μs	14.4028 KOps/s	17.0024 KOps/s	$\textbf{\color{#d91a1a}-15.29\%}$
test_add_pytree	0.1085ms	45.4481μs	22.0031 KOps/s	21.4160 KOps/s	$\color{#35bf28}+2.74\%$
test_add_td	0.1844ms	93.5935μs	10.6845 KOps/s	11.4399 KOps/s	$\textbf{\color{#d91a1a}-6.60\%}$
test_compile_add_one_nested[tensordict-compile]	0.1046ms	57.1702μs	17.4916 KOps/s	16.9021 KOps/s	$\color{#35bf28}+3.49\%$
test_compile_add_one_nested[tensordict-eager]	0.3861ms	0.1999ms	5.0021 KOps/s	4.8893 KOps/s	$\color{#35bf28}+2.31\%$
test_compile_add_one_nested[pytree-compile]	0.1203ms	55.8066μs	17.9190 KOps/s	17.3490 KOps/s	$\color{#35bf28}+3.29\%$
test_compile_add_one_nested[pytree-eager]	0.3248ms	0.1420ms	7.0402 KOps/s	6.9650 KOps/s	$\color{#35bf28}+1.08\%$
test_compile_copy_nested[tensordict-compile]	76.4230μs	23.0956μs	43.2983 KOps/s	43.3169 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_copy_nested[tensordict-eager]	0.1467ms	74.3684μs	13.4466 KOps/s	12.9084 KOps/s	$\color{#35bf28}+4.17\%$
test_compile_copy_nested[pytree-compile]	0.1528ms	76.7800μs	13.0242 KOps/s	13.1338 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_copy_nested[pytree-eager]	0.1358ms	68.3473μs	14.6312 KOps/s	14.4388 KOps/s	$\color{#35bf28}+1.33\%$
test_compile_add_one_flat[tensordict-compile]	0.3989ms	0.1806ms	5.5370 KOps/s	5.4905 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_add_one_flat[tensordict-eager]	0.5059ms	0.2391ms	4.1831 KOps/s	4.0344 KOps/s	$\color{#35bf28}+3.68\%$
test_compile_add_one_flat[tensorclass-compile]	0.1143ms	47.8461μs	20.9003 KOps/s	20.2525 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_add_one_flat[tensorclass-eager]	0.1777ms	77.9574μs	12.8275 KOps/s	12.7105 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_add_one_flat[pytree-compile]	0.3649ms	0.1753ms	5.7048 KOps/s	5.6905 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_add_one_flat[pytree-eager]	0.6371ms	0.2887ms	3.4633 KOps/s	3.3414 KOps/s	$\color{#35bf28}+3.65\%$
test_compile_add_self_flat[tensordict-eager]	0.5340ms	0.2729ms	3.6642 KOps/s	3.5408 KOps/s	$\color{#35bf28}+3.49\%$
test_compile_add_self_flat[tensordict-compile]	0.5386ms	0.1831ms	5.4624 KOps/s	5.4063 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_add_self_flat[tensorclass-eager]	0.1594ms	74.7201μs	13.3833 KOps/s	13.2014 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_add_self_flat[tensorclass-compile]	0.1161ms	48.8297μs	20.4793 KOps/s	20.0060 KOps/s	$\color{#35bf28}+2.37\%$
test_compile_add_self_flat[pytree-eager]	0.4472ms	0.2330ms	4.2918 KOps/s	4.1718 KOps/s	$\color{#35bf28}+2.88\%$
test_compile_add_self_flat[pytree-compile]	0.2868ms	0.1760ms	5.6830 KOps/s	5.7491 KOps/s	$\color{#d91a1a}-1.15\%$
test_compile_copy_flat[tensordict-compile]	0.2427ms	0.1112ms	8.9962 KOps/s	9.0368 KOps/s	$\color{#d91a1a}-0.45\%$
test_compile_copy_flat[tensordict-eager]	0.1567ms	77.2072μs	12.9522 KOps/s	12.5735 KOps/s	$\color{#35bf28}+3.01\%$
test_compile_copy_flat[pytree-compile]	0.1440ms	77.3859μs	12.9222 KOps/s	12.7017 KOps/s	$\color{#35bf28}+1.74\%$
test_compile_copy_flat[pytree-eager]	0.1502ms	69.5137μs	14.3856 KOps/s	14.1095 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_assign_and_add[tensordict-compile]	0.3360ms	0.1931ms	5.1780 KOps/s	5.1823 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_assign_and_add[tensordict-eager]	2.9584ms	1.7360ms	576.0494 Ops/s	564.3420 Ops/s	$\color{#35bf28}+2.07\%$
test_compile_assign_and_add[pytree-compile]	0.3808ms	0.1925ms	5.1954 KOps/s	5.1951 KOps/s	$+0.01\%$
test_compile_assign_and_add[pytree-eager]	1.3292ms	1.0904ms	917.0850 Ops/s	878.7278 Ops/s	$\color{#35bf28}+4.37\%$
test_compile_assign_and_add_stack[compile]	0.6738ms	0.4134ms	2.4190 KOps/s	2.3897 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_assign_and_add_stack[eager]	4.7154ms	4.2808ms	233.5998 Ops/s	240.0512 Ops/s	$\color{#d91a1a}-2.69\%$
test_compile_indexing[tensor-tensordict-compile]	81.9030μs	33.7374μs	29.6407 KOps/s	28.3522 KOps/s	$\color{#35bf28}+4.54\%$
test_compile_indexing[tensor-tensordict-eager]	0.9220ms	50.5278μs	19.7911 KOps/s	20.2785 KOps/s	$\color{#d91a1a}-2.40\%$
test_compile_indexing[tensor-tensorclass-compile]	73.6570μs	30.4504μs	32.8403 KOps/s	33.2482 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_indexing[tensor-tensorclass-eager]	68.4170μs	29.5584μs	33.8313 KOps/s	34.0388 KOps/s	$\color{#d91a1a}-0.61\%$
test_compile_indexing[tensor-pytree-compile]	76.5030μs	30.1312μs	33.1882 KOps/s	33.0106 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_indexing[tensor-pytree-eager]	95.6990μs	29.5206μs	33.8747 KOps/s	34.1011 KOps/s	$\color{#d91a1a}-0.66\%$
test_compile_indexing[slice-tensordict-compile]	0.1811ms	73.7724μs	13.5552 KOps/s	13.3520 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_indexing[slice-tensordict-eager]	0.6452ms	28.7052μs	34.8369 KOps/s	34.2855 KOps/s	$\color{#35bf28}+1.61\%$
test_compile_indexing[slice-tensorclass-compile]	0.1214ms	68.9938μs	14.4941 KOps/s	14.6381 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_indexing[slice-tensorclass-eager]	63.2780μs	23.7056μs	42.1842 KOps/s	41.9194 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_indexing[slice-pytree-compile]	0.1253ms	68.4042μs	14.6190 KOps/s	14.5806 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_indexing[slice-pytree-eager]	63.9790μs	23.6004μs	42.3722 KOps/s	42.8127 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_indexing[int-tensordict-compile]	0.1376ms	73.4189μs	13.6205 KOps/s	13.3620 KOps/s	$\color{#35bf28}+1.93\%$
test_compile_indexing[int-tensordict-eager]	1.2070ms	28.7648μs	34.7647 KOps/s	34.1354 KOps/s	$\color{#35bf28}+1.84\%$
test_compile_indexing[int-tensorclass-compile]	0.1607ms	69.0793μs	14.4761 KOps/s	14.5963 KOps/s	$\color{#d91a1a}-0.82\%$
test_compile_indexing[int-tensorclass-eager]	64.7000μs	23.3906μs	42.7523 KOps/s	41.7736 KOps/s	$\color{#35bf28}+2.34\%$
test_compile_indexing[int-pytree-compile]	0.1330ms	67.3923μs	14.8385 KOps/s	14.7102 KOps/s	$\color{#35bf28}+0.87\%$
test_compile_indexing[int-pytree-eager]	69.2900μs	23.3515μs	42.8237 KOps/s	42.6733 KOps/s	$\color{#35bf28}+0.35\%$
test_mod_add[eager]	98.2540μs	28.0360μs	35.6684 KOps/s	38.6727 KOps/s	$\textbf{\color{#d91a1a}-7.77\%}$
test_mod_add[compile]	0.1109ms	38.3592μs	26.0694 KOps/s	26.0929 KOps/s	$\color{#d91a1a}-0.09\%$
test_mod_add[compile-overhead]	87.9840μs	38.5922μs	25.9120 KOps/s	26.0052 KOps/s	$\color{#d91a1a}-0.36\%$
test_mod_wrap[eager]	0.4041ms	0.2089ms	4.7859 KOps/s	4.5880 KOps/s	$\color{#35bf28}+4.31\%$
test_mod_wrap[compile]	0.3089ms	0.2288ms	4.3710 KOps/s	4.3391 KOps/s	$\color{#35bf28}+0.74\%$
test_mod_wrap[compile-overhead]	0.3091ms	0.2260ms	4.4241 KOps/s	4.4013 KOps/s	$\color{#35bf28}+0.52\%$
test_mod_wrap_and_backward[eager]	12.3277ms	10.7536ms	92.9917 Ops/s	93.5066 Ops/s	$\color{#d91a1a}-0.55\%$
test_mod_wrap_and_backward[compile]	12.6167ms	11.0992ms	90.0964 Ops/s	82.4422 Ops/s	$\textbf{\color{#35bf28}+9.28\%}$
test_mod_wrap_and_backward[compile-overhead]	13.7725ms	11.3706ms	87.9462 Ops/s	82.6547 Ops/s	$\textbf{\color{#35bf28}+6.40\%}$
test_seq_add[eager]	0.1644ms	94.3135μs	10.6029 KOps/s	10.3084 KOps/s	$\color{#35bf28}+2.86\%$
test_seq_add[compile]	0.1248ms	64.1098μs	15.5982 KOps/s	14.8683 KOps/s	$\color{#35bf28}+4.91\%$
test_seq_add[compile-overhead]	0.1306ms	63.2830μs	15.8020 KOps/s	15.6746 KOps/s	$\color{#35bf28}+0.81\%$
test_seq_wrap[eager]	0.5971ms	0.3950ms	2.5317 KOps/s	2.5327 KOps/s	$\color{#d91a1a}-0.04\%$
test_seq_wrap[compile]	1.1867ms	0.2676ms	3.7374 KOps/s	3.7205 KOps/s	$\color{#35bf28}+0.46\%$
test_seq_wrap[compile-overhead]	1.2183ms	0.2665ms	3.7528 KOps/s	3.5977 KOps/s	$\color{#35bf28}+4.31\%$
test_func_call_runtime[False-eager]	0.9956ms	0.5370ms	1.8624 KOps/s	1.8859 KOps/s	$\color{#d91a1a}-1.25\%$
test_func_call_runtime[False-compile]	0.8780ms	0.5005ms	1.9979 KOps/s	1.9830 KOps/s	$\color{#35bf28}+0.75\%$
test_func_call_runtime[False-compile-overhead]	0.6374ms	0.5022ms	1.9914 KOps/s	1.9977 KOps/s	$\color{#d91a1a}-0.31\%$
test_func_call_runtime[True-eager]	1.2901ms	0.7575ms	1.3201 KOps/s	1.3320 KOps/s	$\color{#d91a1a}-0.89\%$
test_func_call_runtime[True-compile]	0.6240ms	0.5138ms	1.9462 KOps/s	1.9292 KOps/s	$\color{#35bf28}+0.88\%$
test_func_call_runtime[True-compile-overhead]	0.6030ms	0.5109ms	1.9575 KOps/s	1.9486 KOps/s	$\color{#35bf28}+0.46\%$
test_func_call_cm_runtime[False-eager]	0.9037ms	0.5271ms	1.8972 KOps/s	1.8843 KOps/s	$\color{#35bf28}+0.68\%$
test_func_call_cm_runtime[False-compile]	0.6731ms	0.4951ms	2.0198 KOps/s	1.9724 KOps/s	$\color{#35bf28}+2.41\%$
test_func_call_cm_runtime[False-compile-overhead]	0.9270ms	0.4967ms	2.0134 KOps/s	1.9774 KOps/s	$\color{#35bf28}+1.82\%$
test_func_call_cm_runtime[True-eager]	1.0609ms	0.8973ms	1.1144 KOps/s	1.1056 KOps/s	$\color{#35bf28}+0.79\%$
test_func_call_cm_runtime[True-compile]	1.2185ms	0.7468ms	1.3390 KOps/s	1.3517 KOps/s	$\color{#d91a1a}-0.93\%$
test_func_call_cm_runtime[True-compile-overhead]	0.9294ms	0.7451ms	1.3421 KOps/s	1.3416 KOps/s	$\color{#35bf28}+0.04\%$
test_vmap_func_call_cm_runtime[eager]	2.4289ms	1.9167ms	521.7280 Ops/s	524.8040 Ops/s	$\color{#d91a1a}-0.59\%$
test_vmap_func_call_cm_runtime[compile]	2.6320ms	1.9762ms	506.0166 Ops/s	512.2920 Ops/s	$\color{#d91a1a}-1.22\%$
test_vmap_func_call_cm_runtime[compile-overhead]	2.6886ms	1.9797ms	505.1160 Ops/s	503.3427 Ops/s	$\color{#35bf28}+0.35\%$
test_distributed	0.2300ms	0.1259ms	7.9451 KOps/s	7.7393 KOps/s	$\color{#35bf28}+2.66\%$
test_tdmodule	42.8700μs	19.7457μs	50.6440 KOps/s	55.3814 KOps/s	$\textbf{\color{#d91a1a}-8.55\%}$
test_tdmodule_dispatch	70.3420μs	39.2082μs	25.5049 KOps/s	27.7483 KOps/s	$\textbf{\color{#d91a1a}-8.08\%}$
test_tdseq	67.8270μs	22.2603μs	44.9230 KOps/s	47.3509 KOps/s	$\textbf{\color{#d91a1a}-5.13\%}$
test_tdseq_dispatch	79.0170μs	44.8318μs	22.3056 KOps/s	22.2833 KOps/s	$\color{#35bf28}+0.10\%$
test_instantiation_functorch	2.5040ms	1.6129ms	620.0087 Ops/s	614.5818 Ops/s	$\color{#35bf28}+0.88\%$
test_instantiation_td	2.0066ms	1.1951ms	836.7588 Ops/s	820.6131 Ops/s	$\color{#35bf28}+1.97\%$
test_exec_functorch	0.4155ms	0.1899ms	5.2648 KOps/s	5.3141 KOps/s	$\color{#d91a1a}-0.93\%$
test_exec_functional_call	0.3088ms	0.1714ms	5.8359 KOps/s	5.5830 KOps/s	$\color{#35bf28}+4.53\%$
test_exec_td	0.3644ms	0.2003ms	4.9917 KOps/s	4.9298 KOps/s	$\color{#35bf28}+1.26\%$
test_exec_td_decorator	0.7814ms	0.2337ms	4.2785 KOps/s	4.1876 KOps/s	$\color{#35bf28}+2.17\%$
test_vmap_mlp_speed[True-True]	1.0157ms	0.6921ms	1.4449 KOps/s	1.4585 KOps/s	$\color{#d91a1a}-0.93\%$
test_vmap_mlp_speed[True-False]	1.0081ms	0.6879ms	1.4536 KOps/s	1.4693 KOps/s	$\color{#d91a1a}-1.07\%$
test_vmap_mlp_speed[False-True]	0.6686ms	0.5379ms	1.8590 KOps/s	1.8668 KOps/s	$\color{#d91a1a}-0.42\%$
test_vmap_mlp_speed[False-False]	0.8124ms	0.5417ms	1.8461 KOps/s	1.8598 KOps/s	$\color{#d91a1a}-0.74\%$
test_vmap_mlp_speed_decorator[True-True]	0.8833ms	0.6512ms	1.5357 KOps/s	1.5536 KOps/s	$\color{#d91a1a}-1.15\%$
test_vmap_mlp_speed_decorator[True-False]	1.0798ms	0.6546ms	1.5277 KOps/s	1.5521 KOps/s	$\color{#d91a1a}-1.57\%$
test_vmap_mlp_speed_decorator[False-True]	0.9341ms	0.5363ms	1.8647 KOps/s	1.8832 KOps/s	$\color{#d91a1a}-0.98\%$
test_vmap_mlp_speed_decorator[False-False]	0.8477ms	0.5367ms	1.8634 KOps/s	1.8930 KOps/s	$\color{#d91a1a}-1.56\%$
test_to_module_speed[True]	1.6080ms	1.3888ms	720.0479 Ops/s	688.4168 Ops/s	$\color{#35bf28}+4.59\%$
test_to_module_speed[False]	2.1808ms	1.3767ms	726.3889 Ops/s	723.1174 Ops/s	$\color{#35bf28}+0.45\%$
test_tc_init	94.7570μs	50.1545μs	19.9384 KOps/s	21.0269 KOps/s	$\textbf{\color{#d91a1a}-5.18\%}$
test_tc_init_nested	0.1749ms	99.6678μs	10.0333 KOps/s	10.4540 KOps/s	$\color{#d91a1a}-4.02\%$
test_tc_first_layer_tensor	22.9830μs	1.5531μs	643.8852 KOps/s	632.4159 KOps/s	$\color{#35bf28}+1.81\%$
test_tc_first_layer_nontensor	26.9800μs	4.7718μs	209.5652 KOps/s	204.8666 KOps/s	$\color{#35bf28}+2.29\%$
test_tc_second_layer_tensor	24.2150μs	2.8075μs	356.1893 KOps/s	345.1799 KOps/s	$\color{#35bf28}+3.19\%$
test_tc_second_layer_nontensor	22.3520μs	6.0015μs	166.6257 KOps/s	159.6663 KOps/s	$\color{#35bf28}+4.36\%$
test_unbind	0.4604s	13.0939ms	76.3716 Ops/s	78.2471 Ops/s	$\color{#d91a1a}-2.40\%$
test_full_like	15.0701ms	11.5648ms	86.4694 Ops/s	133.5770 Ops/s	$\textbf{\color{#d91a1a}-35.27\%}$
test_zeros_like	12.6066ms	6.8810ms	145.3284 Ops/s	350.8841 Ops/s	$\textbf{\color{#d91a1a}-58.58\%}$
test_ones_like	14.7679ms	7.4271ms	134.6426 Ops/s	308.0310 Ops/s	$\textbf{\color{#d91a1a}-56.29\%}$
test_clone	14.7695ms	9.1160ms	109.6972 Ops/s	196.5174 Ops/s	$\textbf{\color{#d91a1a}-44.18\%}$
test_squeeze	70.2210μs	13.7338μs	72.8130 KOps/s	78.5917 KOps/s	$\textbf{\color{#d91a1a}-7.35\%}$
test_unsqueeze	0.1919ms	97.1448μs	10.2939 KOps/s	10.8015 KOps/s	$\color{#d91a1a}-4.70\%$
test_split	0.4971ms	0.1975ms	5.0642 KOps/s	5.0553 KOps/s	$\color{#35bf28}+0.18\%$
test_permute	0.3020ms	0.2207ms	4.5314 KOps/s	4.4768 KOps/s	$\color{#35bf28}+1.22\%$
test_stack	30.1211ms	24.1658ms	41.3808 Ops/s	38.2408 Ops/s	$\textbf{\color{#35bf28}+8.21\%}$
test_cat	25.8665ms	23.9055ms	41.8314 Ops/s	38.0515 Ops/s	$\textbf{\color{#35bf28}+9.93\%}$

[github-actions]

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

Total Benchmarks: 228. Improved: $\large\color{#35bf28}17$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1257ms	16.6185μs	60.1738 KOps/s	56.8502 KOps/s	$\textbf{\color{#35bf28}+5.85\%}$
test_plain_set_stack_nested	87.8350μs	16.7516μs	59.6957 KOps/s	57.1452 KOps/s	$\color{#35bf28}+4.46\%$
test_plain_set_nested_inplace	67.3140μs	18.0026μs	55.5477 KOps/s	53.3854 KOps/s	$\color{#35bf28}+4.05\%$
test_plain_set_stack_nested_inplace	57.9440μs	17.9768μs	55.6272 KOps/s	52.9722 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_items	28.8820μs	2.8403μs	352.0770 KOps/s	348.8082 KOps/s	$\color{#35bf28}+0.94\%$
test_items_nested	0.4479ms	0.3443ms	2.9045 KOps/s	2.8985 KOps/s	$\color{#35bf28}+0.21\%$
test_items_nested_locked	0.4985ms	0.3480ms	2.8736 KOps/s	2.8970 KOps/s	$\color{#d91a1a}-0.81\%$
test_items_nested_leaf	91.5660μs	62.8693μs	15.9060 KOps/s	16.0177 KOps/s	$\color{#d91a1a}-0.70\%$
test_items_stack_nested	0.4860ms	0.3529ms	2.8335 KOps/s	2.9021 KOps/s	$\color{#d91a1a}-2.36\%$
test_items_stack_nested_leaf	0.1056ms	63.6398μs	15.7134 KOps/s	15.9581 KOps/s	$\color{#d91a1a}-1.53\%$
test_items_stack_nested_locked	0.5219ms	0.3487ms	2.8674 KOps/s	2.8821 KOps/s	$\color{#d91a1a}-0.51\%$
test_keys	38.9230μs	3.4306μs	291.4944 KOps/s	293.8702 KOps/s	$\color{#d91a1a}-0.81\%$
test_keys_nested	0.1109ms	71.5381μs	13.9786 KOps/s	14.0279 KOps/s	$\color{#d91a1a}-0.35\%$
test_keys_nested_locked	2.8001ms	77.2597μs	12.9434 KOps/s	13.0082 KOps/s	$\color{#d91a1a}-0.50\%$
test_keys_nested_leaf	99.7060μs	62.0571μs	16.1142 KOps/s	16.2676 KOps/s	$\color{#d91a1a}-0.94\%$
test_keys_stack_nested	0.1118ms	71.7270μs	13.9417 KOps/s	14.1843 KOps/s	$\color{#d91a1a}-1.71\%$
test_keys_stack_nested_leaf	0.1239ms	62.7191μs	15.9441 KOps/s	16.2861 KOps/s	$\color{#d91a1a}-2.10\%$
test_keys_stack_nested_locked	0.1915ms	76.9617μs	12.9935 KOps/s	12.8707 KOps/s	$\color{#35bf28}+0.95\%$
test_values	4.4403μs	0.8602μs	1.1625 MOps/s	1.1399 MOps/s	$\color{#35bf28}+1.98\%$
test_values_nested	0.4248ms	48.8463μs	20.4724 KOps/s	20.5658 KOps/s	$\color{#d91a1a}-0.45\%$
test_values_nested_locked	0.4402ms	50.3905μs	19.8450 KOps/s	19.8867 KOps/s	$\color{#d91a1a}-0.21\%$
test_values_nested_leaf	69.6840μs	43.0022μs	23.2546 KOps/s	23.5702 KOps/s	$\color{#d91a1a}-1.34\%$
test_values_stack_nested	0.4433ms	49.7886μs	20.0849 KOps/s	20.2743 KOps/s	$\color{#d91a1a}-0.93\%$
test_values_stack_nested_leaf	0.4384ms	43.6414μs	22.9140 KOps/s	23.3047 KOps/s	$\color{#d91a1a}-1.68\%$
test_values_stack_nested_locked	0.4298ms	51.2322μs	19.5190 KOps/s	19.8732 KOps/s	$\color{#d91a1a}-1.78\%$
test_membership	19.7723μs	0.5017μs	1.9933 MOps/s	1.9875 MOps/s	$\color{#35bf28}+0.29\%$
test_membership_nested	0.1908ms	1.8720μs	534.2009 KOps/s	530.6618 KOps/s	$\color{#35bf28}+0.67\%$
test_membership_nested_leaf	0.1285ms	1.8365μs	544.5264 KOps/s	528.9284 KOps/s	$\color{#35bf28}+2.95\%$
test_membership_stacked_nested	29.1320μs	1.9189μs	521.1316 KOps/s	519.0438 KOps/s	$\color{#35bf28}+0.40\%$
test_membership_stacked_nested_leaf	47.1330μs	1.9474μs	513.5049 KOps/s	521.6067 KOps/s	$\color{#d91a1a}-1.55\%$
test_membership_nested_last	0.3869ms	2.9561μs	338.2851 KOps/s	334.2080 KOps/s	$\color{#35bf28}+1.22\%$
test_membership_nested_leaf_last	27.0420μs	2.9594μs	337.9106 KOps/s	335.8815 KOps/s	$\color{#35bf28}+0.60\%$
test_membership_stacked_nested_last	20.6110μs	2.9436μs	339.7179 KOps/s	331.2112 KOps/s	$\color{#35bf28}+2.57\%$
test_membership_stacked_nested_leaf_last	27.1220μs	2.9429μs	339.7972 KOps/s	332.8099 KOps/s	$\color{#35bf28}+2.10\%$
test_nested_getleaf	35.7020μs	6.1079μs	163.7212 KOps/s	164.3559 KOps/s	$\color{#d91a1a}-0.39\%$
test_nested_get	0.3849ms	5.8009μs	172.3882 KOps/s	172.9740 KOps/s	$\color{#d91a1a}-0.34\%$
test_stacked_getleaf	0.3895ms	6.0551μs	165.1506 KOps/s	166.1092 KOps/s	$\color{#d91a1a}-0.58\%$
test_stacked_get	36.7420μs	5.7518μs	173.8599 KOps/s	175.8366 KOps/s	$\color{#d91a1a}-1.12\%$
test_nested_getitemleaf	0.3872ms	6.2130μs	160.9522 KOps/s	162.6437 KOps/s	$\color{#d91a1a}-1.04\%$
test_nested_getitem	31.6020μs	5.8720μs	170.2991 KOps/s	172.8287 KOps/s	$\color{#d91a1a}-1.46\%$
test_stacked_getitemleaf	0.3924ms	6.1441μs	162.7578 KOps/s	164.7630 KOps/s	$\color{#d91a1a}-1.22\%$
test_stacked_getitem	29.1220μs	5.7504μs	173.9009 KOps/s	173.1550 KOps/s	$\color{#35bf28}+0.43\%$
test_lock_nested	6.8677ms	0.4365ms	2.2910 KOps/s	2.3210 KOps/s	$\color{#d91a1a}-1.29\%$
test_lock_stack_nested	0.4371ms	0.3968ms	2.5202 KOps/s	2.5599 KOps/s	$\color{#d91a1a}-1.55\%$
test_unlock_nested	0.7721ms	0.3691ms	2.7097 KOps/s	2.7148 KOps/s	$\color{#d91a1a}-0.19\%$
test_unlock_stack_nested	0.3665ms	0.3338ms	2.9962 KOps/s	3.0241 KOps/s	$\color{#d91a1a}-0.92\%$
test_flatten_speed	0.4628ms	76.2808μs	13.1095 KOps/s	13.1791 KOps/s	$\color{#d91a1a}-0.53\%$
test_unflatten_speed	0.3927ms	0.3238ms	3.0887 KOps/s	3.0488 KOps/s	$\color{#35bf28}+1.31\%$
test_common_ops	1.6582ms	1.2866ms	777.2187 Ops/s	774.0121 Ops/s	$\color{#35bf28}+0.41\%$
test_creation	0.3763ms	1.4758μs	677.5786 KOps/s	671.0766 KOps/s	$\color{#35bf28}+0.97\%$
test_creation_empty	45.9130μs	15.4134μs	64.8788 KOps/s	58.2219 KOps/s	$\textbf{\color{#35bf28}+11.43\%}$
test_creation_nested_1	0.4146ms	17.0596μs	58.6181 KOps/s	52.1152 KOps/s	$\textbf{\color{#35bf28}+12.48\%}$
test_creation_nested_2	61.4940μs	19.6898μs	50.7878 KOps/s	47.1154 KOps/s	$\textbf{\color{#35bf28}+7.79\%}$
test_clone	71.5850μs	29.3244μs	34.1013 KOps/s	35.1792 KOps/s	$\color{#d91a1a}-3.06\%$
test_getitem[int]	1.2098ms	16.5152μs	60.5504 KOps/s	63.2557 KOps/s	$\color{#d91a1a}-4.28\%$
test_getitem[slice_int]	0.1192ms	28.2324μs	35.4203 KOps/s	35.7426 KOps/s	$\color{#d91a1a}-0.90\%$
test_getitem[range]	0.1791ms	0.1092ms	9.1551 KOps/s	9.1721 KOps/s	$\color{#d91a1a}-0.19\%$
test_getitem[tuple]	0.1178ms	24.1052μs	41.4848 KOps/s	41.9750 KOps/s	$\color{#d91a1a}-1.17\%$
test_getitem[list]	0.5003ms	99.0685μs	10.0940 KOps/s	10.0334 KOps/s	$\color{#35bf28}+0.60\%$
test_setitem_dim[int]	65.8650μs	44.6869μs	22.3779 KOps/s	22.1536 KOps/s	$\color{#35bf28}+1.01\%$
test_setitem_dim[slice_int]	99.9370μs	67.5130μs	14.8120 KOps/s	14.7270 KOps/s	$\color{#35bf28}+0.58\%$
test_setitem_dim[range]	0.1553ms	0.1290ms	7.7532 KOps/s	7.8445 KOps/s	$\color{#d91a1a}-1.16\%$
test_setitem_dim[tuple]	0.4667ms	61.4001μs	16.2866 KOps/s	16.3782 KOps/s	$\color{#d91a1a}-0.56\%$
test_setitem	88.6450μs	42.7115μs	23.4129 KOps/s	23.6923 KOps/s	$\color{#d91a1a}-1.18\%$
test_set	0.4457ms	41.7099μs	23.9751 KOps/s	23.9910 KOps/s	$\color{#d91a1a}-0.07\%$
test_set_shared	0.3776ms	54.6473μs	18.2992 KOps/s	18.4706 KOps/s	$\color{#d91a1a}-0.93\%$
test_update	0.4325ms	51.4649μs	19.4307 KOps/s	18.5020 KOps/s	$\textbf{\color{#35bf28}+5.02\%}$
test_update_nested	0.4586ms	59.5362μs	16.7965 KOps/s	16.0491 KOps/s	$\color{#35bf28}+4.66\%$
test_update__nested	0.1086ms	62.6071μs	15.9726 KOps/s	15.7888 KOps/s	$\color{#35bf28}+1.16\%$
test_set_nested	0.4355ms	44.8759μs	22.2837 KOps/s	21.2395 KOps/s	$\color{#35bf28}+4.92\%$
test_set_nested_new	98.2060μs	48.6411μs	20.5587 KOps/s	19.8255 KOps/s	$\color{#35bf28}+3.70\%$
test_select	0.1116ms	61.7999μs	16.1813 KOps/s	15.5360 KOps/s	$\color{#35bf28}+4.15\%$
test_select_nested	0.4283ms	41.7290μs	23.9641 KOps/s	21.9884 KOps/s	$\textbf{\color{#35bf28}+8.99\%}$
test_exclude_nested	0.4432ms	58.4259μs	17.1157 KOps/s	16.3774 KOps/s	$\color{#35bf28}+4.51\%$
test_empty[True]	0.6552ms	0.2608ms	3.8342 KOps/s	3.8159 KOps/s	$\color{#35bf28}+0.48\%$
test_empty[False]	4.7683μs	0.7365μs	1.3578 MOps/s	1.3538 MOps/s	$\color{#35bf28}+0.29\%$
test_to	0.4078ms	26.3262μs	37.9850 KOps/s	38.6573 KOps/s	$\color{#d91a1a}-1.74\%$
test_to_nonblocking	59.8740μs	25.5647μs	39.1164 KOps/s	39.8495 KOps/s	$\color{#d91a1a}-1.84\%$
test_unbind_speed	1.5111ms	0.2848ms	3.5108 KOps/s	3.4946 KOps/s	$\color{#35bf28}+0.46\%$
test_unbind_speed_stack0	0.6862ms	0.2802ms	3.5686 KOps/s	3.5685 KOps/s	$+0.00\%$
test_unbind_speed_stack1	91.1781ms	0.7122ms	1.4041 KOps/s	1.3959 KOps/s	$\color{#35bf28}+0.58\%$
test_split	93.1798ms	2.2470ms	445.0385 Ops/s	456.1572 Ops/s	$\color{#d91a1a}-2.44\%$
test_chunk	93.2378ms	2.2527ms	443.9182 Ops/s	458.1654 Ops/s	$\color{#d91a1a}-3.11\%$
test_creation[device0]	0.5048ms	0.1290ms	7.7529 KOps/s	7.8983 KOps/s	$\color{#d91a1a}-1.84\%$
test_creation_from_tensor	0.3783ms	0.1305ms	7.6649 KOps/s	7.6484 KOps/s	$\color{#35bf28}+0.22\%$
test_add_one[memmap_tensor0]	0.2663ms	8.5477μs	116.9909 KOps/s	113.6537 KOps/s	$\color{#35bf28}+2.94\%$
test_contiguous[memmap_tensor0]	17.7010μs	2.1938μs	455.8384 KOps/s	458.0518 KOps/s	$\color{#d91a1a}-0.48\%$
test_stack[memmap_tensor0]	36.4330μs	6.8708μs	145.5439 KOps/s	151.2467 KOps/s	$\color{#d91a1a}-3.77\%$
test_memmaptd_index	1.3094ms	0.4455ms	2.2448 KOps/s	2.3501 KOps/s	$\color{#d91a1a}-4.48\%$
test_memmaptd_index_astensor	1.0112ms	0.5137ms	1.9466 KOps/s	2.0175 KOps/s	$\color{#d91a1a}-3.52\%$
test_memmaptd_index_op	1.4210ms	1.0393ms	962.1907 Ops/s	931.3130 Ops/s	$\color{#35bf28}+3.32\%$
test_serialize_model	0.1309s	0.1294s	7.7268 Ops/s	7.6712 Ops/s	$\color{#35bf28}+0.73\%$
test_serialize_model_pickle	1.3473s	1.2125s	0.8247 Ops/s	0.8246 Ops/s	$\color{#35bf28}+0.02\%$
test_serialize_weights	0.2132s	0.1414s	7.0706 Ops/s	6.9608 Ops/s	$\color{#35bf28}+1.58\%$
test_serialize_weights_returnearly	0.2148s	55.9685ms	17.8672 Ops/s	18.0293 Ops/s	$\color{#d91a1a}-0.90\%$
test_serialize_weights_pickle	1.3825s	1.2183s	0.8208 Ops/s	0.8217 Ops/s	$\color{#d91a1a}-0.11\%$
test_reshape_pytree	66.7340μs	35.8876μs	27.8647 KOps/s	27.8510 KOps/s	$\color{#35bf28}+0.05\%$
test_reshape_td	71.5240μs	41.6169μs	24.0287 KOps/s	23.6934 KOps/s	$\color{#35bf28}+1.42\%$
test_view_pytree	75.1350μs	36.4172μs	27.4596 KOps/s	28.6591 KOps/s	$\color{#d91a1a}-4.19\%$
test_view_td	78.3750μs	46.5388μs	21.4874 KOps/s	22.1342 KOps/s	$\color{#d91a1a}-2.92\%$
test_unbind_pytree	68.1140μs	34.8244μs	28.7155 KOps/s	29.4502 KOps/s	$\color{#d91a1a}-2.49\%$
test_unbind_td	0.3878ms	43.6703μs	22.8989 KOps/s	23.4293 KOps/s	$\color{#d91a1a}-2.26\%$
test_split_pytree	86.4950μs	47.0056μs	21.2741 KOps/s	21.4403 KOps/s	$\color{#d91a1a}-0.78\%$
test_split_td	0.7066ms	58.3102μs	17.1497 KOps/s	17.7862 KOps/s	$\color{#d91a1a}-3.58\%$
test_add_pytree	97.5660μs	58.6546μs	17.0489 KOps/s	17.8383 KOps/s	$\color{#d91a1a}-4.42\%$
test_add_td	0.1365ms	98.3858μs	10.1641 KOps/s	10.3049 KOps/s	$\color{#d91a1a}-1.37\%$
test_compile_add_one_nested[tensordict-compile]	0.2287ms	0.1618ms	6.1795 KOps/s	5.9889 KOps/s	$\color{#35bf28}+3.18\%$
test_compile_add_one_nested[tensordict-eager]	0.2521ms	0.1667ms	5.9986 KOps/s	6.2638 KOps/s	$\color{#d91a1a}-4.23\%$
test_compile_add_one_nested[pytree-compile]	0.1778ms	0.1450ms	6.8963 KOps/s	6.9080 KOps/s	$\color{#d91a1a}-0.17\%$
test_compile_add_one_nested[pytree-eager]	0.2618ms	0.1885ms	5.3060 KOps/s	5.4912 KOps/s	$\color{#d91a1a}-3.37\%$
test_compile_copy_nested[tensordict-compile]	64.5140μs	22.2656μs	44.9124 KOps/s	47.3740 KOps/s	$\textbf{\color{#d91a1a}-5.20\%}$
test_compile_copy_nested[tensordict-eager]	83.1750μs	49.1678μs	20.3385 KOps/s	20.1494 KOps/s	$\color{#35bf28}+0.94\%$
test_compile_copy_nested[pytree-compile]	0.2310ms	64.9618μs	15.3937 KOps/s	15.4547 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_copy_nested[pytree-eager]	84.8550μs	50.2655μs	19.8944 KOps/s	19.9871 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_add_one_flat[tensordict-compile]	0.3568ms	0.3215ms	3.1104 KOps/s	3.1223 KOps/s	$\color{#d91a1a}-0.38\%$
test_compile_add_one_flat[tensordict-eager]	0.3431ms	0.2418ms	4.1356 KOps/s	4.2935 KOps/s	$\color{#d91a1a}-3.68\%$
test_compile_add_one_flat[tensorclass-compile]	0.1721ms	0.1274ms	7.8470 KOps/s	7.8338 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_add_one_flat[tensorclass-eager]	0.1756ms	67.4633μs	14.8229 KOps/s	15.3934 KOps/s	$\color{#d91a1a}-3.71\%$
test_compile_add_one_flat[pytree-compile]	0.3629ms	0.3200ms	3.1255 KOps/s	3.1534 KOps/s	$\color{#d91a1a}-0.89\%$
test_compile_add_one_flat[pytree-eager]	0.7671ms	0.6348ms	1.5752 KOps/s	1.6386 KOps/s	$\color{#d91a1a}-3.87\%$
test_compile_add_self_flat[tensordict-eager]	0.3453ms	0.2886ms	3.4656 KOps/s	3.5167 KOps/s	$\color{#d91a1a}-1.45\%$
test_compile_add_self_flat[tensordict-compile]	0.3740ms	0.3210ms	3.1153 KOps/s	3.1116 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_add_self_flat[tensorclass-eager]	0.1696ms	78.8933μs	12.6754 KOps/s	13.1454 KOps/s	$\color{#d91a1a}-3.58\%$
test_compile_add_self_flat[tensorclass-compile]	0.1866ms	0.1306ms	7.6596 KOps/s	7.6020 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_add_self_flat[pytree-eager]	0.6238ms	0.5477ms	1.8258 KOps/s	1.9392 KOps/s	$\textbf{\color{#d91a1a}-5.85\%}$
test_compile_add_self_flat[pytree-compile]	0.3601ms	0.3193ms	3.1323 KOps/s	3.1431 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_copy_flat[tensordict-compile]	70.8240μs	19.5859μs	51.0572 KOps/s	51.2055 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_copy_flat[tensordict-eager]	69.1740μs	38.1989μs	26.1788 KOps/s	24.3866 KOps/s	$\textbf{\color{#35bf28}+7.35\%}$
test_compile_copy_flat[pytree-compile]	0.1059ms	69.4339μs	14.4022 KOps/s	14.3951 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_copy_flat[pytree-eager]	87.7660μs	51.2345μs	19.5181 KOps/s	19.3914 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_assign_and_add[tensordict-compile]	2.3699ms	0.8405ms	1.1898 KOps/s	1.1231 KOps/s	$\textbf{\color{#35bf28}+5.94\%}$
test_compile_assign_and_add[tensordict-eager]	3.4529ms	3.3019ms	302.8534 Ops/s	314.8395 Ops/s	$\color{#d91a1a}-3.81\%$
test_compile_assign_and_add[pytree-compile]	2.3501ms	0.8253ms	1.2117 KOps/s	1.1301 KOps/s	$\textbf{\color{#35bf28}+7.23\%}$
test_compile_assign_and_add[pytree-eager]	3.4166ms	3.2671ms	306.0800 Ops/s	318.5538 Ops/s	$\color{#d91a1a}-3.92\%$
test_compile_indexing[tensor-tensordict-compile]	0.1425ms	0.1086ms	9.2059 KOps/s	9.1195 KOps/s	$\color{#35bf28}+0.95\%$
test_compile_indexing[tensor-tensordict-eager]	0.1900ms	60.8971μs	16.4212 KOps/s	15.9640 KOps/s	$\color{#35bf28}+2.86\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1550ms	0.1031ms	9.7022 KOps/s	9.6877 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1467ms	44.2565μs	22.5955 KOps/s	23.2013 KOps/s	$\color{#d91a1a}-2.61\%$
test_compile_indexing[tensor-pytree-compile]	0.1815ms	0.1068ms	9.3620 KOps/s	9.6021 KOps/s	$\color{#d91a1a}-2.50\%$
test_compile_indexing[tensor-pytree-eager]	85.0960μs	47.5803μs	21.0171 KOps/s	23.1239 KOps/s	$\textbf{\color{#d91a1a}-9.11\%}$
test_compile_indexing[slice-tensordict-compile]	0.1742ms	0.1377ms	7.2606 KOps/s	7.2862 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_indexing[slice-tensordict-eager]	0.1619ms	25.9298μs	38.5657 KOps/s	39.8013 KOps/s	$\color{#d91a1a}-3.10\%$
test_compile_indexing[slice-tensorclass-compile]	0.1779ms	0.1312ms	7.6227 KOps/s	7.6073 KOps/s	$\color{#35bf28}+0.20\%$
test_compile_indexing[slice-tensorclass-eager]	59.5440μs	21.5170μs	46.4750 KOps/s	48.3435 KOps/s	$\color{#d91a1a}-3.87\%$
test_compile_indexing[slice-pytree-compile]	0.1912ms	0.1354ms	7.3874 KOps/s	7.5869 KOps/s	$\color{#d91a1a}-2.63\%$
test_compile_indexing[slice-pytree-eager]	89.0860μs	20.9572μs	47.7163 KOps/s	47.8067 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_indexing[int-tensordict-compile]	0.1880ms	0.1385ms	7.2182 KOps/s	7.2191 KOps/s	$\color{#d91a1a}-0.01\%$
test_compile_indexing[int-tensordict-eager]	0.4850ms	25.2703μs	39.5722 KOps/s	39.4738 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_indexing[int-tensorclass-compile]	0.5816ms	0.1330ms	7.5214 KOps/s	7.5698 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_indexing[int-tensorclass-eager]	51.7630μs	21.3027μs	46.9425 KOps/s	48.0774 KOps/s	$\color{#d91a1a}-2.36\%$
test_compile_indexing[int-pytree-compile]	0.1925ms	0.1326ms	7.5431 KOps/s	7.5610 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_indexing[int-pytree-eager]	50.1830μs	20.9972μs	47.6254 KOps/s	39.4005 KOps/s	$\textbf{\color{#35bf28}+20.88\%}$
test_mod_add[eager]	64.7240μs	33.0189μs	30.2857 KOps/s	29.9693 KOps/s	$\color{#35bf28}+1.06\%$
test_mod_add[compile]	0.1093ms	71.0353μs	14.0775 KOps/s	14.2314 KOps/s	$\color{#d91a1a}-1.08\%$
test_mod_add[compile-overhead]	0.2590ms	0.1350ms	7.4098 KOps/s	6.8049 KOps/s	$\textbf{\color{#35bf28}+8.89\%}$
test_mod_wrap[eager]	0.8540ms	0.7934ms	1.2604 KOps/s	1.2574 KOps/s	$\color{#35bf28}+0.24\%$
test_mod_wrap[compile]	2.0035ms	0.8477ms	1.1797 KOps/s	1.1806 KOps/s	$\color{#d91a1a}-0.08\%$
test_mod_wrap[compile-overhead]	4.8545ms	3.0389ms	329.0658 Ops/s	320.9817 Ops/s	$\color{#35bf28}+2.52\%$
test_mod_wrap_and_backward[eager]	4.5993ms	4.0799ms	245.1057 Ops/s	237.9322 Ops/s	$\color{#35bf28}+3.01\%$
test_mod_wrap_and_backward[compile]	5.0422ms	4.1094ms	243.3458 Ops/s	239.5729 Ops/s	$\color{#35bf28}+1.57\%$
test_mod_wrap_and_backward[compile-overhead]	1.3510ms	0.9051ms	1.1048 KOps/s	991.4427 Ops/s	$\textbf{\color{#35bf28}+11.44\%}$
test_seq_add[eager]	0.2000ms	0.1035ms	9.6583 KOps/s	9.8867 KOps/s	$\color{#d91a1a}-2.31\%$
test_seq_add[compile]	0.1750ms	81.5427μs	12.2635 KOps/s	12.3823 KOps/s	$\color{#d91a1a}-0.96\%$
test_seq_add[compile-overhead]	0.1508ms	0.1142ms	8.7531 KOps/s	8.7423 KOps/s	$\color{#35bf28}+0.12\%$
test_seq_wrap[eager]	1.0160ms	0.9418ms	1.0618 KOps/s	1.0435 KOps/s	$\color{#35bf28}+1.76\%$
test_seq_wrap[compile]	0.9551ms	0.8643ms	1.1570 KOps/s	1.1571 KOps/s	$\color{#d91a1a}-0.01\%$
test_seq_wrap[compile-overhead]	0.2970ms	0.2193ms	4.5591 KOps/s	4.5373 KOps/s	$\color{#35bf28}+0.48\%$
test_func_call_runtime[False-eager]	2.6259ms	2.4073ms	415.4015 Ops/s	414.3886 Ops/s	$\color{#35bf28}+0.24\%$
test_func_call_runtime[False-compile]	2.5334ms	2.4476ms	408.5715 Ops/s	411.6690 Ops/s	$\color{#d91a1a}-0.75\%$
test_func_call_runtime[False-compile-overhead]	0.4294ms	0.3609ms	2.7710 KOps/s	2.7852 KOps/s	$\color{#d91a1a}-0.51\%$
test_func_call_runtime[True-eager]	2.6733ms	2.5623ms	390.2690 Ops/s	388.9693 Ops/s	$\color{#35bf28}+0.33\%$
test_func_call_runtime[True-compile]	2.5482ms	2.4720ms	404.5255 Ops/s	407.4661 Ops/s	$\color{#d91a1a}-0.72\%$
test_func_call_runtime[True-compile-overhead]	0.4324ms	0.3845ms	2.6007 KOps/s	2.6404 KOps/s	$\color{#d91a1a}-1.50\%$
test_func_call_cm_runtime[False-eager]	2.4797ms	2.4048ms	415.8415 Ops/s	414.7612 Ops/s	$\color{#35bf28}+0.26\%$
test_func_call_cm_runtime[False-compile]	2.5299ms	2.4457ms	408.8758 Ops/s	414.1822 Ops/s	$\color{#d91a1a}-1.28\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4173ms	0.3653ms	2.7376 KOps/s	2.7646 KOps/s	$\color{#d91a1a}-0.98\%$
test_func_call_cm_runtime[True-eager]	2.8352ms	2.6804ms	373.0727 Ops/s	373.0262 Ops/s	$\color{#35bf28}+0.01\%$
test_func_call_cm_runtime[True-compile]	2.6041ms	2.5063ms	398.9988 Ops/s	402.8327 Ops/s	$\color{#d91a1a}-0.95\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4973ms	0.4088ms	2.4465 KOps/s	2.4571 KOps/s	$\color{#d91a1a}-0.43\%$
test_vmap_func_call_cm_runtime[eager]	4.2696ms	3.8435ms	260.1819 Ops/s	263.5596 Ops/s	$\color{#d91a1a}-1.28\%$
test_vmap_func_call_cm_runtime[compile]	2.5813ms	2.5072ms	398.8504 Ops/s	403.4707 Ops/s	$\color{#d91a1a}-1.15\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4597ms	0.4118ms	2.4287 KOps/s	2.4517 KOps/s	$\color{#d91a1a}-0.94\%$
test_distributed	2.5392ms	0.1757ms	5.6927 KOps/s	8.8475 KOps/s	$\textbf{\color{#d91a1a}-35.66\%}$
test_tdmodule	62.3740μs	15.2659μs	65.5056 KOps/s	61.6484 KOps/s	$\textbf{\color{#35bf28}+6.26\%}$
test_tdmodule_dispatch	59.5040μs	29.5661μs	33.8225 KOps/s	30.6073 KOps/s	$\textbf{\color{#35bf28}+10.50\%}$
test_tdseq	36.1520μs	16.2337μs	61.6003 KOps/s	57.2296 KOps/s	$\textbf{\color{#35bf28}+7.64\%}$
test_tdseq_dispatch	53.4230μs	32.3248μs	30.9360 KOps/s	28.3367 KOps/s	$\textbf{\color{#35bf28}+9.17\%}$
test_instantiation_functorch	2.0974ms	1.8776ms	532.6061 Ops/s	545.2851 Ops/s	$\color{#d91a1a}-2.33\%$
test_instantiation_td	1.8255ms	1.2147ms	823.2339 Ops/s	831.8947 Ops/s	$\color{#d91a1a}-1.04\%$
test_exec_functorch	1.1319ms	1.0177ms	982.6443 Ops/s	987.9135 Ops/s	$\color{#d91a1a}-0.53\%$
test_exec_functional_call	1.0736ms	1.0275ms	973.2424 Ops/s	977.7719 Ops/s	$\color{#d91a1a}-0.46\%$
test_exec_td	1.2473ms	1.0608ms	942.7256 Ops/s	976.8026 Ops/s	$\color{#d91a1a}-3.49\%$
test_exec_td_decorator	1.9103ms	1.0819ms	924.3331 Ops/s	941.5361 Ops/s	$\color{#d91a1a}-1.83\%$
test_vmap_mlp_speed[True-True]	1.3521ms	1.2810ms	780.6133 Ops/s	781.0284 Ops/s	$\color{#d91a1a}-0.05\%$
test_vmap_mlp_speed[True-False]	1.3571ms	1.2770ms	783.0696 Ops/s	787.1952 Ops/s	$\color{#d91a1a}-0.52\%$
test_vmap_mlp_speed[False-True]	1.2654ms	1.1690ms	855.4472 Ops/s	863.3416 Ops/s	$\color{#d91a1a}-0.91\%$
test_vmap_mlp_speed[False-False]	1.2257ms	1.1745ms	851.4392 Ops/s	860.4453 Ops/s	$\color{#d91a1a}-1.05\%$
test_vmap_mlp_speed_decorator[True-True]	1.7547ms	1.2533ms	797.9184 Ops/s	799.8749 Ops/s	$\color{#d91a1a}-0.24\%$
test_vmap_mlp_speed_decorator[True-False]	1.3428ms	1.2526ms	798.3587 Ops/s	800.1403 Ops/s	$\color{#d91a1a}-0.22\%$
test_vmap_mlp_speed_decorator[False-True]	1.3211ms	1.1723ms	853.0006 Ops/s	858.0404 Ops/s	$\color{#d91a1a}-0.59\%$
test_vmap_mlp_speed_decorator[False-False]	1.2794ms	1.1705ms	854.3565 Ops/s	859.9525 Ops/s	$\color{#d91a1a}-0.65\%$
test_vmap_transformer_speed[True-True]	13.2957ms	13.1967ms	75.7764 Ops/s	75.6901 Ops/s	$\color{#35bf28}+0.11\%$
test_vmap_transformer_speed[True-False]	13.3208ms	13.1628ms	75.9717 Ops/s	75.8283 Ops/s	$\color{#35bf28}+0.19\%$
test_vmap_transformer_speed[False-True]	13.0310ms	12.9498ms	77.2212 Ops/s	76.8370 Ops/s	$\color{#35bf28}+0.50\%$
test_vmap_transformer_speed[False-False]	13.0104ms	12.9358ms	77.3049 Ops/s	76.9785 Ops/s	$\color{#35bf28}+0.42\%$
test_vmap_transformer_speed_decorator[True-True]	33.9718ms	33.8839ms	29.5126 Ops/s	29.6791 Ops/s	$\color{#d91a1a}-0.56\%$
test_vmap_transformer_speed_decorator[True-False]	34.1805ms	34.0526ms	29.3664 Ops/s	29.6879 Ops/s	$\color{#d91a1a}-1.08\%$
test_vmap_transformer_speed_decorator[False-True]	34.3522ms	34.0215ms	29.3931 Ops/s	29.5662 Ops/s	$\color{#d91a1a}-0.59\%$
test_vmap_transformer_speed_decorator[False-False]	34.1088ms	33.9989ms	29.4127 Ops/s	29.5955 Ops/s	$\color{#d91a1a}-0.62\%$
test_to_module_speed[True]	2.0741ms	0.9998ms	1.0002 KOps/s	1.0084 KOps/s	$\color{#d91a1a}-0.81\%$
test_to_module_speed[False]	1.0761ms	0.9806ms	1.0198 KOps/s	1.0366 KOps/s	$\color{#d91a1a}-1.62\%$
test_tc_init	63.5840μs	35.8911μs	27.8620 KOps/s	27.5524 KOps/s	$\color{#35bf28}+1.12\%$
test_tc_init_nested	0.1062ms	74.4452μs	13.4327 KOps/s	13.9670 KOps/s	$\color{#d91a1a}-3.83\%$
test_tc_first_layer_tensor	4.6120μs	0.6534μs	1.5305 MOps/s	1.4879 MOps/s	$\color{#35bf28}+2.86\%$
test_tc_first_layer_nontensor	23.2120μs	2.2066μs	453.1885 KOps/s	443.6885 KOps/s	$\color{#35bf28}+2.14\%$
test_tc_second_layer_tensor	10.6180μs	1.3601μs	735.2245 KOps/s	726.7870 KOps/s	$\color{#35bf28}+1.16\%$
test_tc_second_layer_nontensor	24.0220μs	2.9678μs	336.9510 KOps/s	336.9989 KOps/s	$\color{#d91a1a}-0.01\%$
test_unbind	0.1927s	12.2283ms	81.7777 Ops/s	93.8199 Ops/s	$\textbf{\color{#d91a1a}-12.84\%}$
test_full_like	0.6561ms	0.5740ms	1.7423 KOps/s	1.7408 KOps/s	$\color{#35bf28}+0.09\%$
test_zeros_like	0.2619ms	0.1980ms	5.0513 KOps/s	5.0501 KOps/s	$\color{#35bf28}+0.02\%$
test_ones_like	0.2342ms	0.1978ms	5.0549 KOps/s	5.0555 KOps/s	$\color{#d91a1a}-0.01\%$
test_clone	0.4612ms	0.4137ms	2.4170 KOps/s	2.4165 KOps/s	$\color{#35bf28}+0.02\%$
test_squeeze	43.1630μs	10.0592μs	99.4111 KOps/s	104.0571 KOps/s	$\color{#d91a1a}-4.46\%$
test_unsqueeze	0.2194ms	76.0830μs	13.1435 KOps/s	13.4978 KOps/s	$\color{#d91a1a}-2.62\%$
test_split	0.4295ms	0.1616ms	6.1869 KOps/s	6.4290 KOps/s	$\color{#d91a1a}-3.77\%$
test_permute	0.2696ms	0.1772ms	5.6433 KOps/s	5.7301 KOps/s	$\color{#d91a1a}-1.51\%$
test_stack	1.2454ms	0.8657ms	1.1551 KOps/s	1.1516 KOps/s	$\color{#35bf28}+0.30\%$
test_cat	1.2520ms	1.2315ms	812.0051 Ops/s	811.9832 Ops/s	$+0.00\%$