[Feature] TensorDictModule method and kwargs specification

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] TensorDictModule method and kwargs specification #1228
• [BE] Better errors for TensorDictSequential #1227
• [Doc] Better doc for TensorDictModuleBase #1226

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}16$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	44.8740μs	21.4773μs	46.5607 KOps/s	48.2540 KOps/s	$\color{#d91a1a}-3.51\%$
test_plain_set_stack_nested	42.9300μs	21.4954μs	46.5216 KOps/s	47.0196 KOps/s	$\color{#d91a1a}-1.06\%$
test_plain_set_nested_inplace	61.3840μs	22.6978μs	44.0570 KOps/s	43.0603 KOps/s	$\color{#35bf28}+2.31\%$
test_plain_set_stack_nested_inplace	70.4510μs	23.2250μs	43.0571 KOps/s	43.1141 KOps/s	$\color{#d91a1a}-0.13\%$
test_items	27.6920μs	4.1483μs	241.0635 KOps/s	241.5692 KOps/s	$\color{#d91a1a}-0.21\%$
test_items_nested	0.4716ms	0.4019ms	2.4882 KOps/s	2.4702 KOps/s	$\color{#35bf28}+0.73\%$
test_items_nested_locked	0.5411ms	0.4037ms	2.4770 KOps/s	2.4687 KOps/s	$\color{#35bf28}+0.34\%$
test_items_nested_leaf	0.1381ms	76.6575μs	13.0450 KOps/s	12.8204 KOps/s	$\color{#35bf28}+1.75\%$
test_items_stack_nested	0.5591ms	0.4057ms	2.4650 KOps/s	2.4486 KOps/s	$\color{#35bf28}+0.67\%$
test_items_stack_nested_leaf	0.1478ms	79.8885μs	12.5174 KOps/s	12.5663 KOps/s	$\color{#d91a1a}-0.39\%$
test_items_stack_nested_locked	0.6020ms	0.4051ms	2.4685 KOps/s	2.4605 KOps/s	$\color{#35bf28}+0.32\%$
test_keys	28.2430μs	3.4471μs	290.0994 KOps/s	289.1844 KOps/s	$\color{#35bf28}+0.32\%$
test_keys_nested	0.2342ms	0.1620ms	6.1710 KOps/s	6.1346 KOps/s	$\color{#35bf28}+0.59\%$
test_keys_nested_locked	1.5687ms	0.1705ms	5.8664 KOps/s	5.9131 KOps/s	$\color{#d91a1a}-0.79\%$
test_keys_nested_leaf	0.2664ms	0.1421ms	7.0397 KOps/s	6.9897 KOps/s	$\color{#35bf28}+0.71\%$
test_keys_stack_nested	0.2553ms	0.1620ms	6.1745 KOps/s	6.1087 KOps/s	$\color{#35bf28}+1.08\%$
test_keys_stack_nested_leaf	0.2245ms	0.1424ms	7.0201 KOps/s	7.1010 KOps/s	$\color{#d91a1a}-1.14\%$
test_keys_stack_nested_locked	0.2622ms	0.1674ms	5.9742 KOps/s	5.9476 KOps/s	$\color{#35bf28}+0.45\%$
test_values	5.3640μs	1.0948μs	913.4378 KOps/s	977.0980 KOps/s	$\textbf{\color{#d91a1a}-6.52\%}$
test_values_nested	0.1124ms	62.9643μs	15.8820 KOps/s	16.0617 KOps/s	$\color{#d91a1a}-1.12\%$
test_values_nested_locked	0.1368ms	63.5834μs	15.7274 KOps/s	16.0588 KOps/s	$\color{#d91a1a}-2.06\%$
test_values_nested_leaf	0.1263ms	72.0711μs	13.8752 KOps/s	13.4757 KOps/s	$\color{#35bf28}+2.96\%$
test_values_stack_nested	0.1142ms	64.0097μs	15.6226 KOps/s	15.9508 KOps/s	$\color{#d91a1a}-2.06\%$
test_values_stack_nested_leaf	0.1257ms	72.0165μs	13.8857 KOps/s	14.1706 KOps/s	$\color{#d91a1a}-2.01\%$
test_values_stack_nested_locked	0.1404ms	63.5705μs	15.7306 KOps/s	15.8710 KOps/s	$\color{#d91a1a}-0.89\%$
test_membership	3.8657μs	0.7082μs	1.4121 MOps/s	1.4384 MOps/s	$\color{#d91a1a}-1.83\%$
test_membership_nested	26.5690μs	2.8780μs	347.4599 KOps/s	350.5388 KOps/s	$\color{#d91a1a}-0.88\%$
test_membership_nested_leaf	21.8510μs	2.8892μs	346.1213 KOps/s	340.9366 KOps/s	$\color{#35bf28}+1.52\%$
test_membership_stacked_nested	25.9290μs	2.9076μs	343.9294 KOps/s	348.5097 KOps/s	$\color{#d91a1a}-1.31\%$
test_membership_stacked_nested_leaf	17.4730μs	2.8417μs	351.9027 KOps/s	345.6168 KOps/s	$\color{#35bf28}+1.82\%$
test_membership_nested_last	32.2090μs	4.3696μs	228.8523 KOps/s	228.9545 KOps/s	$\color{#d91a1a}-0.04\%$
test_membership_nested_leaf_last	38.8020μs	4.3510μs	229.8342 KOps/s	227.8905 KOps/s	$\color{#35bf28}+0.85\%$
test_membership_stacked_nested_last	25.8080μs	5.6635μs	176.5681 KOps/s	230.5547 KOps/s	$\textbf{\color{#d91a1a}-23.42\%}$
test_membership_stacked_nested_leaf_last	27.3000μs	5.6255μs	177.7622 KOps/s	228.7071 KOps/s	$\textbf{\color{#d91a1a}-22.28\%}$
test_nested_getleaf	31.7290μs	10.6857μs	93.5834 KOps/s	93.7028 KOps/s	$\color{#d91a1a}-0.13\%$
test_nested_get	30.4770μs	10.1937μs	98.1000 KOps/s	97.2232 KOps/s	$\color{#35bf28}+0.90\%$
test_stacked_getleaf	30.5770μs	10.5163μs	95.0902 KOps/s	93.1855 KOps/s	$\color{#35bf28}+2.04\%$
test_stacked_get	32.5100μs	10.1530μs	98.4927 KOps/s	97.0804 KOps/s	$\color{#35bf28}+1.45\%$
test_nested_getitemleaf	37.4300μs	11.4046μs	87.6839 KOps/s	87.4610 KOps/s	$\color{#35bf28}+0.25\%$
test_nested_getitem	39.7740μs	10.7593μs	92.9427 KOps/s	90.9469 KOps/s	$\color{#35bf28}+2.19\%$
test_stacked_getitemleaf	33.5020μs	11.2615μs	88.7978 KOps/s	88.3016 KOps/s	$\color{#35bf28}+0.56\%$
test_stacked_getitem	36.6880μs	10.7989μs	92.6020 KOps/s	91.5597 KOps/s	$\color{#35bf28}+1.14\%$
test_lock_nested	0.5611ms	0.4141ms	2.4147 KOps/s	2.4315 KOps/s	$\color{#d91a1a}-0.69\%$
test_lock_stack_nested	0.4931ms	0.4217ms	2.3715 KOps/s	2.3481 KOps/s	$\color{#35bf28}+0.99\%$
test_unlock_nested	0.4446ms	0.3365ms	2.9716 KOps/s	2.9739 KOps/s	$\color{#d91a1a}-0.08\%$
test_unlock_stack_nested	0.5589ms	0.3398ms	2.9432 KOps/s	2.9315 KOps/s	$\color{#35bf28}+0.40\%$
test_flatten_speed	0.1877ms	99.9256μs	10.0074 KOps/s	9.9045 KOps/s	$\color{#35bf28}+1.04\%$
test_unflatten_speed	0.9267ms	0.5294ms	1.8889 KOps/s	1.9252 KOps/s	$\color{#d91a1a}-1.88\%$
test_common_ops	5.3217ms	0.8377ms	1.1937 KOps/s	1.2310 KOps/s	$\color{#d91a1a}-3.03\%$
test_creation	48.3600μs	2.4330μs	411.0131 KOps/s	405.6586 KOps/s	$\color{#35bf28}+1.32\%$
test_creation_empty	41.0260μs	13.4219μs	74.5053 KOps/s	76.8628 KOps/s	$\color{#d91a1a}-3.07\%$
test_creation_nested_1	51.2550μs	16.1945μs	61.7493 KOps/s	63.1763 KOps/s	$\color{#d91a1a}-2.26\%$
test_creation_nested_2	59.3810μs	21.0653μs	47.4713 KOps/s	48.0116 KOps/s	$\color{#d91a1a}-1.13\%$
test_clone	74.6990μs	13.2298μs	75.5871 KOps/s	75.2544 KOps/s	$\color{#35bf28}+0.44\%$
test_getitem[int]	0.8433ms	12.9073μs	77.4756 KOps/s	79.2577 KOps/s	$\color{#d91a1a}-2.25\%$
test_getitem[slice_int]	0.1184ms	25.0667μs	39.8935 KOps/s	40.9629 KOps/s	$\color{#d91a1a}-2.61\%$
test_getitem[range]	0.2210ms	51.4717μs	19.4281 KOps/s	20.5270 KOps/s	$\textbf{\color{#d91a1a}-5.35\%}$
test_getitem[tuple]	0.1261ms	21.0251μs	47.5622 KOps/s	50.0117 KOps/s	$\color{#d91a1a}-4.90\%$
test_getitem[list]	0.2018ms	46.1135μs	21.6856 KOps/s	22.9115 KOps/s	$\textbf{\color{#d91a1a}-5.35\%}$
test_setitem_dim[int]	57.6480μs	25.6695μs	38.9567 KOps/s	39.9329 KOps/s	$\color{#d91a1a}-2.44\%$
test_setitem_dim[slice_int]	0.1154ms	51.6944μs	19.3445 KOps/s	19.5666 KOps/s	$\color{#d91a1a}-1.14\%$
test_setitem_dim[range]	0.1293ms	79.5886μs	12.5646 KOps/s	13.3352 KOps/s	$\textbf{\color{#d91a1a}-5.78\%}$
test_setitem_dim[tuple]	85.9300μs	41.0873μs	24.3384 KOps/s	24.8261 KOps/s	$\color{#d91a1a}-1.96\%$
test_setitem	67.4960μs	20.7728μs	48.1398 KOps/s	49.0949 KOps/s	$\color{#d91a1a}-1.95\%$
test_set	78.2960μs	20.1255μs	49.6881 KOps/s	49.0351 KOps/s	$\color{#35bf28}+1.33\%$
test_set_shared	5.2416ms	0.1801ms	5.5523 KOps/s	5.6835 KOps/s	$\color{#d91a1a}-2.31\%$
test_update	0.1194ms	23.7371μs	42.1281 KOps/s	41.7162 KOps/s	$\color{#35bf28}+0.99\%$
test_update_nested	92.5130μs	34.7189μs	28.8027 KOps/s	29.1285 KOps/s	$\color{#d91a1a}-1.12\%$
test_update__nested	0.5201ms	32.7095μs	30.5721 KOps/s	30.2686 KOps/s	$\color{#35bf28}+1.00\%$
test_set_nested	68.3770μs	22.1391μs	45.1689 KOps/s	44.6827 KOps/s	$\color{#35bf28}+1.09\%$
test_set_nested_new	74.4890μs	26.0159μs	38.4380 KOps/s	37.0872 KOps/s	$\color{#35bf28}+3.64\%$
test_select	97.8320μs	43.9849μs	22.7351 KOps/s	23.8489 KOps/s	$\color{#d91a1a}-4.67\%$
test_select_nested	0.1416ms	63.2403μs	15.8127 KOps/s	16.0436 KOps/s	$\color{#d91a1a}-1.44\%$
test_exclude_nested	0.4024ms	81.9277μs	12.2059 KOps/s	12.4788 KOps/s	$\color{#d91a1a}-2.19\%$
test_empty[True]	0.5968ms	0.4119ms	2.4277 KOps/s	2.4493 KOps/s	$\color{#d91a1a}-0.88\%$
test_empty[False]	7.3235μs	1.3522μs	739.5258 KOps/s	751.1995 KOps/s	$\color{#d91a1a}-1.55\%$
test_unbind_speed	0.3444ms	0.2699ms	3.7045 KOps/s	3.7402 KOps/s	$\color{#d91a1a}-0.96\%$
test_unbind_speed_stack0	0.4130ms	0.2684ms	3.7252 KOps/s	3.7960 KOps/s	$\color{#d91a1a}-1.86\%$
test_unbind_speed_stack1	99.2940ms	0.7232ms	1.3828 KOps/s	1.2781 KOps/s	$\textbf{\color{#35bf28}+8.19\%}$
test_split	98.0984ms	1.7572ms	569.0986 Ops/s	555.5052 Ops/s	$\color{#35bf28}+2.45\%$
test_chunk	0.1019s	1.7519ms	570.8032 Ops/s	626.4371 Ops/s	$\textbf{\color{#d91a1a}-8.88\%}$
test_consolidate_njt[False-None]	8.6989ms	8.1842ms	122.1860 Ops/s	109.2208 Ops/s	$\textbf{\color{#35bf28}+11.87\%}$
test_creation[device0]	4.9892ms	91.9734μs	10.8727 KOps/s	11.1692 KOps/s	$\color{#d91a1a}-2.65\%$
test_creation_from_tensor	0.2168ms	92.5494μs	10.8050 KOps/s	10.6821 KOps/s	$\color{#35bf28}+1.15\%$
test_add_one[memmap_tensor0]	0.1231ms	5.1299μs	194.9362 KOps/s	202.0463 KOps/s	$\color{#d91a1a}-3.52\%$
test_contiguous[memmap_tensor0]	22.9020μs	0.5109μs	1.9575 MOps/s	1.9688 MOps/s	$\color{#d91a1a}-0.57\%$
test_stack[memmap_tensor0]	30.3070μs	3.3907μs	294.9232 KOps/s	304.4990 KOps/s	$\color{#d91a1a}-3.14\%$
test_memmaptd_index	0.3131ms	0.2377ms	4.2071 KOps/s	4.3933 KOps/s	$\color{#d91a1a}-4.24\%$
test_memmaptd_index_astensor	1.0328ms	0.3219ms	3.1069 KOps/s	3.1694 KOps/s	$\color{#d91a1a}-1.97\%$
test_memmaptd_index_op	0.8997ms	0.6102ms	1.6389 KOps/s	1.6669 KOps/s	$\color{#d91a1a}-1.68\%$
test_serialize_model	0.1226s	0.1119s	8.9383 Ops/s	8.7836 Ops/s	$\color{#35bf28}+1.76\%$
test_serialize_model_pickle	0.4519s	0.3861s	2.5902 Ops/s	2.5508 Ops/s	$\color{#35bf28}+1.55\%$
test_serialize_weights	0.1227s	0.1146s	8.7255 Ops/s	8.7205 Ops/s	$\color{#35bf28}+0.06\%$
test_serialize_weights_returnearly	0.1763s	0.1590s	6.2912 Ops/s	5.6015 Ops/s	$\textbf{\color{#35bf28}+12.31\%}$
test_serialize_weights_pickle	0.6049s	0.4577s	2.1849 Ops/s	1.1664 Ops/s	$\textbf{\color{#35bf28}+87.32\%}$
test_serialize_weights_filesystem	0.1506s	0.1427s	7.0065 Ops/s	6.8195 Ops/s	$\color{#35bf28}+2.74\%$
test_serialize_model_filesystem	0.1539s	0.1469s	6.8065 Ops/s	7.0272 Ops/s	$\color{#d91a1a}-3.14\%$
test_reshape_pytree	77.1340μs	26.7765μs	37.3461 KOps/s	38.2778 KOps/s	$\color{#d91a1a}-2.43\%$
test_reshape_td	82.9350μs	33.4155μs	29.9262 KOps/s	30.0403 KOps/s	$\color{#d91a1a}-0.38\%$
test_view_pytree	92.3320μs	26.5696μs	37.6370 KOps/s	38.5728 KOps/s	$\color{#d91a1a}-2.43\%$
test_view_td	92.1820μs	42.0048μs	23.8068 KOps/s	24.4725 KOps/s	$\color{#d91a1a}-2.72\%$
test_unbind_pytree	77.1540μs	29.3126μs	34.1150 KOps/s	34.0330 KOps/s	$\color{#35bf28}+0.24\%$
test_unbind_td	0.3067ms	40.2806μs	24.8259 KOps/s	25.2571 KOps/s	$\color{#d91a1a}-1.71\%$
test_split_pytree	69.7500μs	29.6700μs	33.7041 KOps/s	34.4067 KOps/s	$\color{#d91a1a}-2.04\%$
test_split_td	0.5051ms	47.1216μs	21.2217 KOps/s	22.2146 KOps/s	$\color{#d91a1a}-4.47\%$
test_add_pytree	0.8040ms	36.3776μs	27.4895 KOps/s	29.1382 KOps/s	$\textbf{\color{#d91a1a}-5.66\%}$
test_add_td	0.1777ms	61.4798μs	16.2655 KOps/s	17.6284 KOps/s	$\textbf{\color{#d91a1a}-7.73\%}$
test_compile_add_one_nested[tensordict-compile]	0.1597ms	65.8988μs	15.1748 KOps/s	15.0365 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_add_one_nested[tensordict-eager]	0.3172ms	0.1721ms	5.8115 KOps/s	5.8030 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_add_one_nested[pytree-compile]	95.3580μs	44.9422μs	22.2508 KOps/s	22.4068 KOps/s	$\color{#d91a1a}-0.70\%$
test_compile_add_one_nested[pytree-eager]	0.2321ms	0.1197ms	8.3544 KOps/s	8.5027 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_copy_nested[tensordict-compile]	97.0700μs	28.3927μs	35.2203 KOps/s	36.1275 KOps/s	$\color{#d91a1a}-2.51\%$
test_compile_copy_nested[tensordict-eager]	0.1144ms	57.5701μs	17.3701 KOps/s	16.9276 KOps/s	$\color{#35bf28}+2.61\%$
test_compile_copy_nested[pytree-compile]	0.1498ms	78.9679μs	12.6634 KOps/s	12.6178 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_copy_nested[pytree-eager]	0.1201ms	67.2611μs	14.8674 KOps/s	15.0487 KOps/s	$\color{#d91a1a}-1.20\%$
test_compile_add_one_flat[tensordict-compile]	0.2418ms	0.1086ms	9.2069 KOps/s	9.1991 KOps/s	$\color{#35bf28}+0.09\%$
test_compile_add_one_flat[tensordict-eager]	0.3113ms	0.2264ms	4.4169 KOps/s	4.6325 KOps/s	$\color{#d91a1a}-4.65\%$
test_compile_add_one_flat[tensorclass-compile]	0.1157ms	47.5317μs	21.0386 KOps/s	21.5624 KOps/s	$\color{#d91a1a}-2.43\%$
test_compile_add_one_flat[tensorclass-eager]	0.1342ms	66.9851μs	14.9287 KOps/s	14.9814 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_add_one_flat[pytree-compile]	0.1678ms	97.9817μs	10.2060 KOps/s	9.8252 KOps/s	$\color{#35bf28}+3.88\%$
test_compile_add_one_flat[pytree-eager]	0.3962ms	0.2030ms	4.9271 KOps/s	4.9799 KOps/s	$\color{#d91a1a}-1.06\%$
test_compile_add_self_flat[tensordict-eager]	0.4832ms	0.2376ms	4.2092 KOps/s	4.2984 KOps/s	$\color{#d91a1a}-2.08\%$
test_compile_add_self_flat[tensordict-compile]	0.1997ms	0.1056ms	9.4702 KOps/s	9.2700 KOps/s	$\color{#35bf28}+2.16\%$
test_compile_add_self_flat[tensorclass-eager]	0.1554ms	64.8138μs	15.4288 KOps/s	16.1243 KOps/s	$\color{#d91a1a}-4.31\%$
test_compile_add_self_flat[tensorclass-compile]	0.1344ms	48.8553μs	20.4686 KOps/s	20.8646 KOps/s	$\color{#d91a1a}-1.90\%$
test_compile_add_self_flat[pytree-eager]	0.3374ms	0.1587ms	6.2993 KOps/s	6.2389 KOps/s	$\color{#35bf28}+0.97\%$
test_compile_add_self_flat[pytree-compile]	0.1971ms	0.1004ms	9.9624 KOps/s	10.0807 KOps/s	$\color{#d91a1a}-1.17\%$
test_compile_copy_flat[tensordict-compile]	73.3790μs	21.2497μs	47.0594 KOps/s	45.3256 KOps/s	$\color{#35bf28}+3.83\%$
test_compile_copy_flat[tensordict-eager]	0.1451ms	67.6475μs	14.7825 KOps/s	14.6284 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_copy_flat[pytree-compile]	0.1558ms	80.9396μs	12.3549 KOps/s	12.0223 KOps/s	$\color{#35bf28}+2.77\%$
test_compile_copy_flat[pytree-eager]	0.1342ms	68.2166μs	14.6592 KOps/s	14.7692 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_assign_and_add[tensordict-compile]	0.3113ms	0.2107ms	4.7466 KOps/s	4.6687 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_assign_and_add[tensordict-eager]	1.7408ms	1.4051ms	711.6972 Ops/s	731.7982 Ops/s	$\color{#d91a1a}-2.75\%$
test_compile_assign_and_add[pytree-compile]	0.2509ms	0.2035ms	4.9143 KOps/s	4.7620 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_assign_and_add[pytree-eager]	1.4656ms	0.8371ms	1.1945 KOps/s	1.2169 KOps/s	$\color{#d91a1a}-1.84\%$
test_compile_assign_and_add_stack[compile]	0.5647ms	0.4539ms	2.2030 KOps/s	2.2163 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_assign_and_add_stack[eager]	3.0892ms	2.7876ms	358.7311 Ops/s	366.8043 Ops/s	$\color{#d91a1a}-2.20\%$
test_compile_indexing[tensor-tensordict-compile]	0.1035ms	38.5910μs	25.9128 KOps/s	25.5724 KOps/s	$\color{#35bf28}+1.33\%$
test_compile_indexing[tensor-tensordict-eager]	0.6727ms	33.8071μs	29.5796 KOps/s	30.4930 KOps/s	$\color{#d91a1a}-3.00\%$
test_compile_indexing[tensor-tensorclass-compile]	85.1490μs	30.1397μs	33.1788 KOps/s	31.9122 KOps/s	$\color{#35bf28}+3.97\%$
test_compile_indexing[tensor-tensorclass-eager]	92.7410μs	23.1950μs	43.1127 KOps/s	44.5908 KOps/s	$\color{#d91a1a}-3.31\%$
test_compile_indexing[tensor-pytree-compile]	0.1331ms	31.8552μs	31.3920 KOps/s	32.2771 KOps/s	$\color{#d91a1a}-2.74\%$
test_compile_indexing[tensor-pytree-eager]	83.8260μs	23.3247μs	42.8730 KOps/s	45.0587 KOps/s	$\color{#d91a1a}-4.85\%$
test_compile_indexing[slice-tensordict-compile]	0.1114ms	54.0875μs	18.4886 KOps/s	18.7761 KOps/s	$\color{#d91a1a}-1.53\%$
test_compile_indexing[slice-tensordict-eager]	0.3741ms	21.1493μs	47.2828 KOps/s	48.6783 KOps/s	$\color{#d91a1a}-2.87\%$
test_compile_indexing[slice-tensorclass-compile]	0.1334ms	44.9344μs	22.2547 KOps/s	21.8875 KOps/s	$\color{#35bf28}+1.68\%$
test_compile_indexing[slice-tensorclass-eager]	78.5270μs	18.8265μs	53.1167 KOps/s	52.7399 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_indexing[slice-pytree-compile]	0.1083ms	45.5640μs	21.9472 KOps/s	21.7800 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_indexing[slice-pytree-eager]	88.7750μs	18.7646μs	53.2917 KOps/s	52.5839 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_indexing[int-tensordict-compile]	0.1227ms	54.1292μs	18.4743 KOps/s	18.3837 KOps/s	$\color{#35bf28}+0.49\%$
test_compile_indexing[int-tensordict-eager]	1.0730ms	20.9125μs	47.8182 KOps/s	50.0651 KOps/s	$\color{#d91a1a}-4.49\%$
test_compile_indexing[int-tensorclass-compile]	0.1074ms	45.3781μs	22.0371 KOps/s	21.3541 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_indexing[int-tensorclass-eager]	0.2866ms	18.7953μs	53.2047 KOps/s	53.7391 KOps/s	$\color{#d91a1a}-0.99\%$
test_compile_indexing[int-pytree-compile]	0.1127ms	46.0603μs	21.7107 KOps/s	21.3703 KOps/s	$\color{#35bf28}+1.59\%$
test_compile_indexing[int-pytree-eager]	56.8960μs	18.7989μs	53.1945 KOps/s	52.7917 KOps/s	$\color{#35bf28}+0.76\%$
test_mod_add[eager]	97.9720μs	36.1900μs	27.6319 KOps/s	27.7214 KOps/s	$\color{#d91a1a}-0.32\%$
test_mod_add[compile]	0.1736ms	64.5679μs	15.4876 KOps/s	15.1458 KOps/s	$\color{#35bf28}+2.26\%$
test_mod_add[compile-overhead]	0.1262ms	62.0238μs	16.1228 KOps/s	15.3554 KOps/s	$\color{#35bf28}+5.00\%$
test_mod_wrap[eager]	0.4196ms	0.2230ms	4.4845 KOps/s	4.4034 KOps/s	$\color{#35bf28}+1.84\%$
test_mod_wrap[compile]	1.6735ms	0.2311ms	4.3269 KOps/s	4.2827 KOps/s	$\color{#35bf28}+1.03\%$
test_mod_wrap[compile-overhead]	0.3520ms	0.2254ms	4.4368 KOps/s	4.3433 KOps/s	$\color{#35bf28}+2.15\%$
test_mod_wrap_and_backward[eager]	16.9554ms	12.3928ms	80.6918 Ops/s	89.8120 Ops/s	$\textbf{\color{#d91a1a}-10.15\%}$
test_mod_wrap_and_backward[compile]	19.3570ms	12.1629ms	82.2173 Ops/s	91.3112 Ops/s	$\textbf{\color{#d91a1a}-9.96\%}$
test_mod_wrap_and_backward[compile-overhead]	12.5501ms	10.8217ms	92.4069 Ops/s	90.4423 Ops/s	$\color{#35bf28}+2.17\%$
test_seq_add[eager]	0.2127ms	0.1181ms	8.4708 KOps/s	8.0801 KOps/s	$\color{#35bf28}+4.84\%$
test_seq_add[compile]	0.1336ms	74.8817μs	13.3544 KOps/s	12.6465 KOps/s	$\textbf{\color{#35bf28}+5.60\%}$
test_seq_add[compile-overhead]	0.1569ms	75.2181μs	13.2947 KOps/s	13.2587 KOps/s	$\color{#35bf28}+0.27\%$
test_seq_wrap[eager]	0.6604ms	0.4554ms	2.1958 KOps/s	2.2381 KOps/s	$\color{#d91a1a}-1.89\%$
test_seq_wrap[compile]	0.4590ms	0.2416ms	4.1395 KOps/s	3.9094 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_seq_wrap[compile-overhead]	0.4478ms	0.2409ms	4.1504 KOps/s	4.0754 KOps/s	$\color{#35bf28}+1.84\%$
test_func_call_runtime[False-eager]	0.8488ms	0.5372ms	1.8613 KOps/s	1.8751 KOps/s	$\color{#d91a1a}-0.73\%$
test_func_call_runtime[False-compile]	0.6923ms	0.4397ms	2.2742 KOps/s	2.1825 KOps/s	$\color{#35bf28}+4.20\%$
test_func_call_runtime[False-compile-overhead]	0.6718ms	0.4437ms	2.2535 KOps/s	2.1963 KOps/s	$\color{#35bf28}+2.61\%$
test_func_call_runtime[True-eager]	0.9642ms	0.7509ms	1.3317 KOps/s	1.3208 KOps/s	$\color{#35bf28}+0.83\%$
test_func_call_runtime[True-compile]	0.6380ms	0.4561ms	2.1923 KOps/s	2.1240 KOps/s	$\color{#35bf28}+3.22\%$
test_func_call_runtime[True-compile-overhead]	0.5584ms	0.4625ms	2.1620 KOps/s	2.1003 KOps/s	$\color{#35bf28}+2.93\%$
test_func_call_cm_runtime[False-eager]	0.8179ms	0.5347ms	1.8701 KOps/s	1.8577 KOps/s	$\color{#35bf28}+0.67\%$
test_func_call_cm_runtime[False-compile]	0.5290ms	0.4409ms	2.2679 KOps/s	2.2115 KOps/s	$\color{#35bf28}+2.55\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5544ms	0.4347ms	2.3006 KOps/s	2.1867 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_func_call_cm_runtime[True-eager]	1.1939ms	0.8962ms	1.1159 KOps/s	1.1115 KOps/s	$\color{#35bf28}+0.39\%$
test_func_call_cm_runtime[True-compile]	1.0200ms	0.7975ms	1.2540 KOps/s	1.2730 KOps/s	$\color{#d91a1a}-1.49\%$
test_func_call_cm_runtime[True-compile-overhead]	1.2982ms	0.8118ms	1.2318 KOps/s	1.2320 KOps/s	$\color{#d91a1a}-0.02\%$
test_vmap_func_call_cm_runtime[eager]	2.4643ms	1.9208ms	520.6273 Ops/s	519.1906 Ops/s	$\color{#35bf28}+0.28\%$
test_vmap_func_call_cm_runtime[compile]	0.6908ms	0.5386ms	1.8567 KOps/s	1.8262 KOps/s	$\color{#35bf28}+1.67\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.8349ms	0.5430ms	1.8416 KOps/s	1.8144 KOps/s	$\color{#35bf28}+1.50\%$
test_distributed	0.2603ms	0.1253ms	7.9810 KOps/s	7.8827 KOps/s	$\color{#35bf28}+1.25\%$
test_tdmodule	45.0540μs	27.4614μs	36.4147 KOps/s	35.1041 KOps/s	$\color{#35bf28}+3.73\%$
test_tdmodule_dispatch	0.1093ms	61.2985μs	16.3136 KOps/s	20.1081 KOps/s	$\textbf{\color{#d91a1a}-18.87\%}$
test_tdseq	56.8660μs	29.9287μs	33.4128 KOps/s	34.7308 KOps/s	$\color{#d91a1a}-3.79\%$
test_tdseq_dispatch	77.7550μs	56.1878μs	17.7974 KOps/s	17.5223 KOps/s	$\color{#35bf28}+1.57\%$
test_instantiation_functorch	1.9754ms	1.4986ms	667.2765 Ops/s	631.4471 Ops/s	$\textbf{\color{#35bf28}+5.67\%}$
test_exec_functorch	0.2650ms	0.1755ms	5.6976 KOps/s	5.5444 KOps/s	$\color{#35bf28}+2.76\%$
test_exec_functional_call	0.2664ms	0.1712ms	5.8395 KOps/s	5.9109 KOps/s	$\color{#d91a1a}-1.21\%$
test_exec_td_decorator	0.4844ms	0.2292ms	4.3635 KOps/s	4.2588 KOps/s	$\color{#35bf28}+2.46\%$
test_vmap_mlp_speed_decorator[True-True]	0.8937ms	0.6669ms	1.4995 KOps/s	1.5072 KOps/s	$\color{#d91a1a}-0.51\%$
test_vmap_mlp_speed_decorator[True-False]	1.0677ms	0.6700ms	1.4925 KOps/s	1.5087 KOps/s	$\color{#d91a1a}-1.07\%$
test_vmap_mlp_speed_decorator[False-True]	0.7328ms	0.5396ms	1.8531 KOps/s	1.8875 KOps/s	$\color{#d91a1a}-1.82\%$
test_vmap_mlp_speed_decorator[False-False]	0.7048ms	0.5374ms	1.8609 KOps/s	1.8924 KOps/s	$\color{#d91a1a}-1.67\%$
test_to_module_speed[True]	2.3996ms	1.3472ms	742.2700 Ops/s	744.9221 Ops/s	$\color{#d91a1a}-0.36\%$
test_to_module_speed[False]	1.7524ms	1.2977ms	770.6098 Ops/s	765.9901 Ops/s	$\color{#35bf28}+0.60\%$
test_tc_init	98.0930μs	50.4392μs	19.8258 KOps/s	20.4589 KOps/s	$\color{#d91a1a}-3.09\%$
test_tc_init_nested	0.1771ms	0.1015ms	9.8518 KOps/s	10.5681 KOps/s	$\textbf{\color{#d91a1a}-6.78\%}$
test_tc_first_layer_tensor	30.7580μs	1.5071μs	663.5348 KOps/s	640.9113 KOps/s	$\color{#35bf28}+3.53\%$
test_tc_first_layer_nontensor	42.0190μs	4.7198μs	211.8733 KOps/s	212.7074 KOps/s	$\color{#d91a1a}-0.39\%$
test_tc_second_layer_tensor	23.9250μs	2.8338μs	352.8816 KOps/s	346.3398 KOps/s	$\color{#35bf28}+1.89\%$
test_tc_second_layer_nontensor	40.3150μs	5.9803μs	167.2154 KOps/s	164.9266 KOps/s	$\color{#35bf28}+1.39\%$
test_unbind	0.2375s	13.3092ms	75.1357 Ops/s	78.8417 Ops/s	$\color{#d91a1a}-4.70\%$
test_full_like	12.3951ms	7.9362ms	126.0047 Ops/s	118.1504 Ops/s	$\textbf{\color{#35bf28}+6.65\%}$
test_zeros_like	5.9020ms	4.6627ms	214.4668 Ops/s	351.8001 Ops/s	$\textbf{\color{#d91a1a}-39.04\%}$
test_ones_like	12.2307ms	5.8788ms	170.1021 Ops/s	286.9202 Ops/s	$\textbf{\color{#d91a1a}-40.71\%}$
test_clone	6.2389ms	5.0536ms	197.8802 Ops/s	147.7926 Ops/s	$\textbf{\color{#35bf28}+33.89\%}$
test_squeeze	66.7740μs	12.7972μs	78.1422 KOps/s	75.9890 KOps/s	$\color{#35bf28}+2.83\%$
test_unsqueeze	0.2796ms	96.2902μs	10.3853 KOps/s	10.9320 KOps/s	$\textbf{\color{#d91a1a}-5.00\%}$
test_split	0.3411ms	0.1942ms	5.1495 KOps/s	5.0875 KOps/s	$\color{#35bf28}+1.22\%$
test_permute	0.3388ms	0.1994ms	5.0145 KOps/s	4.9466 KOps/s	$\color{#35bf28}+1.37\%$
test_stack	26.0298ms	24.2119ms	41.3019 Ops/s	38.5599 Ops/s	$\textbf{\color{#35bf28}+7.11\%}$
test_cat	30.5259ms	24.2552ms	41.2282 Ops/s	39.6547 Ops/s	$\color{#35bf28}+3.97\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}34$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.6510μs	13.4473μs	74.3644 KOps/s	78.0823 KOps/s	$\color{#d91a1a}-4.76\%$
test_plain_set_stack_nested	47.9710μs	13.4906μs	74.1256 KOps/s	77.8759 KOps/s	$\color{#d91a1a}-4.82\%$
test_plain_set_nested_inplace	41.6000μs	14.5870μs	68.5541 KOps/s	72.4108 KOps/s	$\textbf{\color{#d91a1a}-5.33\%}$
test_plain_set_stack_nested_inplace	43.3810μs	14.4176μs	69.3596 KOps/s	72.4064 KOps/s	$\color{#d91a1a}-4.21\%$
test_items	23.9610μs	2.8651μs	349.0308 KOps/s	345.9830 KOps/s	$\color{#35bf28}+0.88\%$
test_items_nested	0.4340ms	0.3635ms	2.7514 KOps/s	2.6958 KOps/s	$\color{#35bf28}+2.06\%$
test_items_nested_locked	0.4755ms	0.3703ms	2.7005 KOps/s	2.6737 KOps/s	$\color{#35bf28}+1.00\%$
test_items_nested_leaf	95.9310μs	65.2365μs	15.3288 KOps/s	15.4348 KOps/s	$\color{#d91a1a}-0.69\%$
test_items_stack_nested	0.4423ms	0.3634ms	2.7520 KOps/s	2.7364 KOps/s	$\color{#35bf28}+0.57\%$
test_items_stack_nested_leaf	96.2720μs	66.4943μs	15.0389 KOps/s	15.2711 KOps/s	$\color{#d91a1a}-1.52\%$
test_items_stack_nested_locked	0.4211ms	0.3656ms	2.7354 KOps/s	2.6937 KOps/s	$\color{#35bf28}+1.55\%$
test_keys	32.7200μs	3.9578μs	252.6636 KOps/s	291.1486 KOps/s	$\textbf{\color{#d91a1a}-13.22\%}$
test_keys_nested	0.1184ms	88.7149μs	11.2721 KOps/s	11.2158 KOps/s	$\color{#35bf28}+0.50\%$
test_keys_nested_locked	0.6999ms	94.6495μs	10.5653 KOps/s	10.5985 KOps/s	$\color{#d91a1a}-0.31\%$
test_keys_nested_leaf	0.1201ms	79.6797μs	12.5502 KOps/s	12.5944 KOps/s	$\color{#d91a1a}-0.35\%$
test_keys_stack_nested	0.1168ms	88.0091μs	11.3625 KOps/s	11.2578 KOps/s	$\color{#35bf28}+0.93\%$
test_keys_stack_nested_leaf	0.1158ms	79.9913μs	12.5014 KOps/s	12.5545 KOps/s	$\color{#d91a1a}-0.42\%$
test_keys_stack_nested_locked	0.1539ms	94.6835μs	10.5615 KOps/s	10.4779 KOps/s	$\color{#35bf28}+0.80\%$
test_values	4.9652μs	0.8543μs	1.1706 MOps/s	1.1667 MOps/s	$\color{#35bf28}+0.33\%$
test_values_nested	60.6010μs	37.6094μs	26.5891 KOps/s	26.9319 KOps/s	$\color{#d91a1a}-1.27\%$
test_values_nested_locked	78.0310μs	39.6329μs	25.2316 KOps/s	25.3745 KOps/s	$\color{#d91a1a}-0.56\%$
test_values_nested_leaf	64.7910μs	42.5212μs	23.5177 KOps/s	23.7470 KOps/s	$\color{#d91a1a}-0.97\%$
test_values_stack_nested	75.6910μs	37.8168μs	26.4433 KOps/s	26.5627 KOps/s	$\color{#d91a1a}-0.45\%$
test_values_stack_nested_leaf	75.1510μs	42.7640μs	23.3842 KOps/s	23.5912 KOps/s	$\color{#d91a1a}-0.88\%$
test_values_stack_nested_locked	68.1810μs	39.9630μs	25.0231 KOps/s	25.1724 KOps/s	$\color{#d91a1a}-0.59\%$
test_membership	1.9656μs	0.5004μs	1.9983 MOps/s	1.9978 MOps/s	$\color{#35bf28}+0.02\%$
test_membership_nested	12.4405μs	2.0118μs	497.0642 KOps/s	476.8613 KOps/s	$\color{#35bf28}+4.24\%$
test_membership_nested_leaf	18.0855μs	2.0217μs	494.6247 KOps/s	489.8913 KOps/s	$\color{#35bf28}+0.97\%$
test_membership_stacked_nested	29.2100μs	2.1025μs	475.6354 KOps/s	477.8115 KOps/s	$\color{#d91a1a}-0.46\%$
test_membership_stacked_nested_leaf	25.2800μs	2.0584μs	485.8144 KOps/s	480.3514 KOps/s	$\color{#35bf28}+1.14\%$
test_membership_nested_last	34.1100μs	3.0683μs	325.9082 KOps/s	322.8330 KOps/s	$\color{#35bf28}+0.95\%$
test_membership_nested_leaf_last	30.1610μs	3.0496μs	327.9065 KOps/s	325.8686 KOps/s	$\color{#35bf28}+0.63\%$
test_membership_stacked_nested_last	42.9010μs	8.2995μs	120.4897 KOps/s	326.1169 KOps/s	$\textbf{\color{#d91a1a}-63.05\%}$
test_membership_stacked_nested_leaf_last	32.7900μs	8.3221μs	120.1626 KOps/s	328.6027 KOps/s	$\textbf{\color{#d91a1a}-63.43\%}$
test_nested_getleaf	36.9100μs	6.2863μs	159.0750 KOps/s	162.6130 KOps/s	$\color{#d91a1a}-2.18\%$
test_nested_get	33.7800μs	5.9282μs	168.6855 KOps/s	167.9850 KOps/s	$\color{#35bf28}+0.42\%$
test_stacked_getleaf	29.7310μs	6.2188μs	160.8017 KOps/s	163.9850 KOps/s	$\color{#d91a1a}-1.94\%$
test_stacked_get	0.6833ms	5.7911μs	172.6794 KOps/s	174.7290 KOps/s	$\color{#d91a1a}-1.17\%$
test_nested_getitemleaf	44.1910μs	6.4216μs	155.7234 KOps/s	155.6270 KOps/s	$\color{#35bf28}+0.06\%$
test_nested_getitem	31.2500μs	6.1050μs	163.8007 KOps/s	163.5099 KOps/s	$\color{#35bf28}+0.18\%$
test_stacked_getitemleaf	34.4110μs	6.4308μs	155.5028 KOps/s	156.9946 KOps/s	$\color{#d91a1a}-0.95\%$
test_stacked_getitem	37.2210μs	6.0076μs	166.4568 KOps/s	167.1671 KOps/s	$\color{#d91a1a}-0.42\%$
test_lock_nested	8.7838ms	0.3598ms	2.7796 KOps/s	2.9792 KOps/s	$\textbf{\color{#d91a1a}-6.70\%}$
test_lock_stack_nested	0.4280ms	0.3436ms	2.9107 KOps/s	2.9272 KOps/s	$\color{#d91a1a}-0.57\%$
test_unlock_nested	0.3535ms	0.2964ms	3.3737 KOps/s	3.5425 KOps/s	$\color{#d91a1a}-4.76\%$
test_unlock_stack_nested	0.3512ms	0.2821ms	3.5449 KOps/s	3.5222 KOps/s	$\color{#35bf28}+0.64\%$
test_flatten_speed	0.1190ms	82.6794μs	12.0949 KOps/s	12.0327 KOps/s	$\color{#35bf28}+0.52\%$
test_unflatten_speed	0.3932ms	0.3237ms	3.0897 KOps/s	3.1305 KOps/s	$\color{#d91a1a}-1.30\%$
test_common_ops	0.7881ms	0.6502ms	1.5380 KOps/s	1.5990 KOps/s	$\color{#d91a1a}-3.81\%$
test_creation	0.1161ms	1.7660μs	566.2660 KOps/s	569.4428 KOps/s	$\color{#d91a1a}-0.56\%$
test_creation_empty	39.8900μs	10.3172μs	96.9255 KOps/s	113.2117 KOps/s	$\textbf{\color{#d91a1a}-14.39\%}$
test_creation_nested_1	46.3300μs	11.8907μs	84.0991 KOps/s	95.7736 KOps/s	$\textbf{\color{#d91a1a}-12.19\%}$
test_creation_nested_2	42.2800μs	14.6918μs	68.0650 KOps/s	76.2079 KOps/s	$\textbf{\color{#d91a1a}-10.69\%}$
test_clone	50.8710μs	11.4622μs	87.2432 KOps/s	92.9191 KOps/s	$\textbf{\color{#d91a1a}-6.11\%}$
test_getitem[int]	1.2033ms	11.0770μs	90.2771 KOps/s	95.5614 KOps/s	$\textbf{\color{#d91a1a}-5.53\%}$
test_getitem[slice_int]	0.1111ms	22.0196μs	45.4141 KOps/s	49.1371 KOps/s	$\textbf{\color{#d91a1a}-7.58\%}$
test_getitem[range]	0.1307ms	39.9345μs	25.0410 KOps/s	26.1514 KOps/s	$\color{#d91a1a}-4.25\%$
test_getitem[tuple]	0.1065ms	18.7069μs	53.4563 KOps/s	56.2931 KOps/s	$\textbf{\color{#d91a1a}-5.04\%}$
test_getitem[list]	0.1484ms	34.2581μs	29.1902 KOps/s	29.3228 KOps/s	$\color{#d91a1a}-0.45\%$
test_setitem_dim[int]	43.5700μs	20.9384μs	47.7592 KOps/s	51.3093 KOps/s	$\textbf{\color{#d91a1a}-6.92\%}$
test_setitem_dim[slice_int]	74.8510μs	40.1078μs	24.9328 KOps/s	25.9871 KOps/s	$\color{#d91a1a}-4.06\%$
test_setitem_dim[range]	88.8410μs	54.7752μs	18.2565 KOps/s	18.7835 KOps/s	$\color{#d91a1a}-2.81\%$
test_setitem_dim[tuple]	64.4710μs	33.4999μs	29.8508 KOps/s	30.5041 KOps/s	$\color{#d91a1a}-2.14\%$
test_setitem	57.8110μs	16.8791μs	59.2448 KOps/s	63.7153 KOps/s	$\textbf{\color{#d91a1a}-7.02\%}$
test_set	49.5210μs	16.3657μs	61.1034 KOps/s	65.0037 KOps/s	$\textbf{\color{#d91a1a}-6.00\%}$
test_set_shared	0.5235ms	0.1638ms	6.1034 KOps/s	6.2829 KOps/s	$\color{#d91a1a}-2.86\%$
test_update	0.3831ms	20.5146μs	48.7457 KOps/s	54.3051 KOps/s	$\textbf{\color{#d91a1a}-10.24\%}$
test_update_nested	63.3410μs	26.1979μs	38.1709 KOps/s	41.5305 KOps/s	$\textbf{\color{#d91a1a}-8.09\%}$
test_update__nested	0.4572ms	26.1360μs	38.2614 KOps/s	38.6012 KOps/s	$\color{#d91a1a}-0.88\%$
test_set_nested	78.5610μs	18.2760μs	54.7166 KOps/s	60.7253 KOps/s	$\textbf{\color{#d91a1a}-9.89\%}$
test_set_nested_new	70.3810μs	20.4239μs	48.9622 KOps/s	52.2564 KOps/s	$\textbf{\color{#d91a1a}-6.30\%}$
test_select	79.2310μs	33.3884μs	29.9505 KOps/s	31.8435 KOps/s	$\textbf{\color{#d91a1a}-5.94\%}$
test_select_nested	88.7220μs	44.8335μs	22.3047 KOps/s	22.4386 KOps/s	$\color{#d91a1a}-0.60\%$
test_exclude_nested	0.1341ms	63.8986μs	15.6498 KOps/s	15.5622 KOps/s	$\color{#35bf28}+0.56\%$
test_empty[True]	0.3661ms	0.2932ms	3.4105 KOps/s	3.4123 KOps/s	$\color{#d91a1a}-0.05\%$
test_empty[False]	3.4811μs	0.8418μs	1.1879 MOps/s	1.1871 MOps/s	$\color{#35bf28}+0.07\%$
test_to	88.2410μs	57.2564μs	17.4653 KOps/s	17.3868 KOps/s	$\color{#35bf28}+0.45\%$
test_to_nonblocking	0.1052ms	48.3762μs	20.6713 KOps/s	20.7499 KOps/s	$\color{#d91a1a}-0.38\%$
test_unbind_speed	0.2939ms	0.2499ms	4.0016 KOps/s	4.1979 KOps/s	$\color{#d91a1a}-4.68\%$
test_unbind_speed_stack0	0.2944ms	0.2407ms	4.1541 KOps/s	4.1658 KOps/s	$\color{#d91a1a}-0.28\%$
test_unbind_speed_stack1	92.6498ms	0.7324ms	1.3653 KOps/s	1.3393 KOps/s	$\color{#35bf28}+1.94\%$
test_split	94.1062ms	1.6423ms	608.9123 Ops/s	622.1267 Ops/s	$\color{#d91a1a}-2.12\%$
test_chunk	95.4914ms	1.6498ms	606.1363 Ops/s	615.8397 Ops/s	$\color{#d91a1a}-1.58\%$
test_consolidate[False-None]	2.8620ms	2.7796ms	359.7633 Ops/s	360.4996 Ops/s	$\color{#d91a1a}-0.20\%$
test_consolidate[default-None]	1.8392ms	1.7485ms	571.9293 Ops/s	585.5509 Ops/s	$\color{#d91a1a}-2.33\%$
test_consolidate[reduce-overhead-None]	1.8906ms	1.7941ms	557.3856 Ops/s	573.3956 Ops/s	$\color{#d91a1a}-2.79\%$
test_consolidate_njt[False-None]	6.9964ms	6.7890ms	147.2965 Ops/s	152.4783 Ops/s	$\color{#d91a1a}-3.40\%$
test_to[False-False-None]	1.8977ms	1.7912ms	558.2730 Ops/s	562.5944 Ops/s	$\color{#d91a1a}-0.77\%$
test_to[True-False-None]	1.5360ms	1.4077ms	710.3713 Ops/s	724.5008 Ops/s	$\color{#d91a1a}-1.95\%$
test_to[within-False-None]	4.5276ms	4.3024ms	232.4294 Ops/s	235.4776 Ops/s	$\color{#d91a1a}-1.29\%$
test_to[True-default-None]	5.6655ms	5.4343ms	184.0178 Ops/s	179.6929 Ops/s	$\color{#35bf28}+2.41\%$
test_to_njt[False-False-None]	7.2678ms	7.0690ms	141.4620 Ops/s	140.0957 Ops/s	$\color{#35bf28}+0.98\%$
test_to_njt[True-False-None]	5.8223ms	5.6694ms	176.3866 Ops/s	171.6584 Ops/s	$\color{#35bf28}+2.75\%$
test_to_njt[within-False-None]	12.7320ms	12.5099ms	79.9367 Ops/s	78.6539 Ops/s	$\color{#35bf28}+1.63\%$
test_creation[device0]	0.4624ms	80.5633μs	12.4126 KOps/s	11.8269 KOps/s	$\color{#35bf28}+4.95\%$
test_creation_from_tensor	0.5384ms	84.2266μs	11.8727 KOps/s	11.2536 KOps/s	$\textbf{\color{#35bf28}+5.50\%}$
test_add_one[memmap_tensor0]	0.4201ms	7.2279μs	138.3528 KOps/s	142.9096 KOps/s	$\color{#d91a1a}-3.19\%$
test_contiguous[memmap_tensor0]	2.7115μs	0.4331μs	2.3088 MOps/s	2.3811 MOps/s	$\color{#d91a1a}-3.03\%$
test_stack[memmap_tensor0]	39.7010μs	4.8449μs	206.4035 KOps/s	224.5089 KOps/s	$\textbf{\color{#d91a1a}-8.06\%}$
test_memmaptd_index	1.4131ms	0.2562ms	3.9031 KOps/s	4.0541 KOps/s	$\color{#d91a1a}-3.73\%$
test_memmaptd_index_astensor	0.4665ms	0.3161ms	3.1639 KOps/s	3.2520 KOps/s	$\color{#d91a1a}-2.71\%$
test_memmaptd_index_op	0.7749ms	0.6434ms	1.5543 KOps/s	1.6796 KOps/s	$\textbf{\color{#d91a1a}-7.46\%}$
test_serialize_model	0.4166s	0.1720s	5.8125 Ops/s	7.6538 Ops/s	$\textbf{\color{#d91a1a}-24.06\%}$
test_serialize_model_pickle	1.3487s	1.2163s	0.8222 Ops/s	0.8255 Ops/s	$\color{#d91a1a}-0.40\%$
test_serialize_weights	0.1308s	0.1298s	7.7062 Ops/s	7.6896 Ops/s	$\color{#35bf28}+0.22\%$
test_serialize_weights_returnearly	0.3197s	54.2634ms	18.4286 Ops/s	15.1709 Ops/s	$\textbf{\color{#35bf28}+21.47\%}$
test_serialize_weights_pickle	1.3777s	1.2164s	0.8221 Ops/s	0.8147 Ops/s	$\color{#35bf28}+0.91\%$
test_reshape_pytree	55.4210μs	22.5984μs	44.2508 KOps/s	43.5750 KOps/s	$\color{#35bf28}+1.55\%$
test_reshape_td	60.2010μs	26.4049μs	37.8717 KOps/s	36.2061 KOps/s	$\color{#35bf28}+4.60\%$
test_view_pytree	59.3710μs	22.2792μs	44.8850 KOps/s	43.9970 KOps/s	$\color{#35bf28}+2.02\%$
test_view_td	0.1269ms	31.5632μs	31.6824 KOps/s	30.1488 KOps/s	$\textbf{\color{#35bf28}+5.09\%}$
test_unbind_pytree	67.1010μs	29.5440μs	33.8478 KOps/s	34.9504 KOps/s	$\color{#d91a1a}-3.15\%$
test_unbind_td	0.8216ms	38.9250μs	25.6904 KOps/s	27.2359 KOps/s	$\textbf{\color{#d91a1a}-5.67\%}$
test_split_pytree	73.6810μs	30.6819μs	32.5925 KOps/s	32.9535 KOps/s	$\color{#d91a1a}-1.10\%$
test_split_td	0.9422ms	39.7283μs	25.1710 KOps/s	25.7634 KOps/s	$\color{#d91a1a}-2.30\%$
test_add_pytree	83.2110μs	36.7127μs	27.2385 KOps/s	27.6736 KOps/s	$\color{#d91a1a}-1.57\%$
test_add_td	0.2050ms	52.6307μs	19.0003 KOps/s	20.2465 KOps/s	$\textbf{\color{#d91a1a}-6.16\%}$
test_compile_add_one_nested[tensordict-compile]	0.1796ms	0.1240ms	8.0616 KOps/s	7.8313 KOps/s	$\color{#35bf28}+2.94\%$
test_compile_add_one_nested[tensordict-eager]	0.2368ms	0.1342ms	7.4501 KOps/s	7.3367 KOps/s	$\color{#35bf28}+1.55\%$
test_compile_add_one_nested[pytree-compile]	0.1533ms	98.0680μs	10.1970 KOps/s	10.2892 KOps/s	$\color{#d91a1a}-0.90\%$
test_compile_add_one_nested[pytree-eager]	1.2982ms	0.1574ms	6.3534 KOps/s	6.5310 KOps/s	$\color{#d91a1a}-2.72\%$
test_compile_copy_nested[tensordict-compile]	68.9310μs	26.5830μs	37.6180 KOps/s	39.7160 KOps/s	$\textbf{\color{#d91a1a}-5.28\%}$
test_compile_copy_nested[tensordict-eager]	62.7410μs	29.1541μs	34.3005 KOps/s	33.4794 KOps/s	$\color{#35bf28}+2.45\%$
test_compile_copy_nested[pytree-compile]	0.4534ms	65.0229μs	15.3792 KOps/s	15.1347 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_copy_nested[pytree-eager]	87.0610μs	50.0542μs	19.9783 KOps/s	19.4809 KOps/s	$\color{#35bf28}+2.55\%$
test_compile_add_one_flat[tensordict-compile]	0.1828ms	0.1435ms	6.9700 KOps/s	7.0453 KOps/s	$\color{#d91a1a}-1.07\%$
test_compile_add_one_flat[tensordict-eager]	0.3136ms	0.2195ms	4.5563 KOps/s	4.5971 KOps/s	$\color{#d91a1a}-0.89\%$
test_compile_add_one_flat[tensorclass-compile]	0.1473ms	99.2467μs	10.0759 KOps/s	10.2831 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_add_one_flat[tensorclass-eager]	0.1500ms	54.9678μs	18.1925 KOps/s	17.7048 KOps/s	$\color{#35bf28}+2.75\%$
test_compile_add_one_flat[pytree-compile]	0.1850ms	0.1375ms	7.2729 KOps/s	7.3603 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_add_one_flat[pytree-eager]	0.5766ms	0.5117ms	1.9541 KOps/s	2.0262 KOps/s	$\color{#d91a1a}-3.56\%$
test_compile_add_self_flat[tensordict-eager]	0.3786ms	0.2631ms	3.8014 KOps/s	3.8077 KOps/s	$\color{#d91a1a}-0.17\%$
test_compile_add_self_flat[tensordict-compile]	0.1938ms	0.1456ms	6.8674 KOps/s	7.0086 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_add_self_flat[tensorclass-eager]	0.1558ms	68.1256μs	14.6788 KOps/s	14.4918 KOps/s	$\color{#35bf28}+1.29\%$
test_compile_add_self_flat[tensorclass-compile]	0.1434ms	0.1002ms	9.9758 KOps/s	10.1230 KOps/s	$\color{#d91a1a}-1.45\%$
test_compile_add_self_flat[pytree-eager]	0.4836ms	0.4345ms	2.3017 KOps/s	2.4184 KOps/s	$\color{#d91a1a}-4.83\%$
test_compile_add_self_flat[pytree-compile]	0.1741ms	0.1361ms	7.3453 KOps/s	7.3484 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_copy_flat[tensordict-compile]	72.5610μs	25.1982μs	39.6853 KOps/s	52.7549 KOps/s	$\textbf{\color{#d91a1a}-24.77\%}$
test_compile_copy_flat[tensordict-eager]	59.0910μs	31.7651μs	31.4811 KOps/s	31.0066 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_copy_flat[pytree-compile]	0.1140ms	69.7330μs	14.3404 KOps/s	14.2722 KOps/s	$\color{#35bf28}+0.48\%$
test_compile_copy_flat[pytree-eager]	83.0220μs	52.3732μs	19.0937 KOps/s	19.1304 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_assign_and_add[tensordict-compile]	1.6646ms	0.4015ms	2.4909 KOps/s	2.1954 KOps/s	$\textbf{\color{#35bf28}+13.46\%}$
test_compile_assign_and_add[tensordict-eager]	2.8444ms	2.7622ms	362.0322 Ops/s	369.9182 Ops/s	$\color{#d91a1a}-2.13\%$
test_compile_assign_and_add[pytree-compile]	1.6403ms	0.3910ms	2.5573 KOps/s	2.2436 KOps/s	$\textbf{\color{#35bf28}+13.98\%}$
test_compile_assign_and_add[pytree-eager]	2.9837ms	2.8539ms	350.4031 Ops/s	366.3021 Ops/s	$\color{#d91a1a}-4.34\%$
test_compile_indexing[tensor-tensordict-compile]	0.6903ms	0.1210ms	8.2626 KOps/s	8.3689 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_indexing[tensor-tensordict-eager]	0.5819ms	83.5084μs	11.9748 KOps/s	11.5077 KOps/s	$\color{#35bf28}+4.06\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5231ms	0.1116ms	8.9598 KOps/s	8.8193 KOps/s	$\color{#35bf28}+1.59\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1828ms	70.4802μs	14.1884 KOps/s	13.6187 KOps/s	$\color{#35bf28}+4.18\%$
test_compile_indexing[tensor-pytree-compile]	0.1965ms	0.1133ms	8.8236 KOps/s	8.6383 KOps/s	$\color{#35bf28}+2.14\%$
test_compile_indexing[tensor-pytree-eager]	0.1326ms	69.6713μs	14.3531 KOps/s	13.4527 KOps/s	$\textbf{\color{#35bf28}+6.69\%}$
test_compile_indexing[slice-tensordict-compile]	0.1526ms	0.1053ms	9.4967 KOps/s	9.8244 KOps/s	$\color{#d91a1a}-3.33\%$
test_compile_indexing[slice-tensordict-eager]	0.1530ms	19.2875μs	51.8470 KOps/s	56.0801 KOps/s	$\textbf{\color{#d91a1a}-7.55\%}$
test_compile_indexing[slice-tensorclass-compile]	0.3916ms	97.2677μs	10.2809 KOps/s	9.6494 KOps/s	$\textbf{\color{#35bf28}+6.54\%}$
test_compile_indexing[slice-tensorclass-eager]	48.6410μs	16.4928μs	60.6324 KOps/s	62.7055 KOps/s	$\color{#d91a1a}-3.31\%$
test_compile_indexing[slice-pytree-compile]	0.1586ms	97.2121μs	10.2868 KOps/s	9.6321 KOps/s	$\textbf{\color{#35bf28}+6.80\%}$
test_compile_indexing[slice-pytree-eager]	60.8110μs	16.2577μs	61.5091 KOps/s	62.4868 KOps/s	$\color{#d91a1a}-1.56\%$
test_compile_indexing[int-tensordict-compile]	0.1537ms	0.1021ms	9.7939 KOps/s	9.4623 KOps/s	$\color{#35bf28}+3.50\%$
test_compile_indexing[int-tensordict-eager]	0.5417ms	18.0924μs	55.2718 KOps/s	55.7837 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_indexing[int-tensorclass-compile]	0.1710ms	97.2060μs	10.2874 KOps/s	10.2311 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_indexing[int-tensorclass-eager]	93.5510μs	16.3667μs	61.0997 KOps/s	62.2218 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_indexing[int-pytree-compile]	0.1457ms	97.3701μs	10.2701 KOps/s	9.6143 KOps/s	$\textbf{\color{#35bf28}+6.82\%}$
test_compile_indexing[int-pytree-eager]	79.7910μs	16.3930μs	61.0017 KOps/s	63.9586 KOps/s	$\color{#d91a1a}-4.62\%$
test_mod_add[eager]	0.1051ms	41.3374μs	24.1911 KOps/s	25.3360 KOps/s	$\color{#d91a1a}-4.52\%$
test_mod_add[compile]	0.1814ms	82.8221μs	12.0741 KOps/s	12.0091 KOps/s	$\color{#35bf28}+0.54\%$
test_mod_add[compile-overhead]	0.3358ms	0.1708ms	5.8549 KOps/s	5.5520 KOps/s	$\textbf{\color{#35bf28}+5.46\%}$
test_mod_wrap[eager]	0.3513ms	0.2600ms	3.8467 KOps/s	3.6862 KOps/s	$\color{#35bf28}+4.35\%$
test_mod_wrap[compile]	0.3882ms	0.2975ms	3.3610 KOps/s	3.4029 KOps/s	$\color{#d91a1a}-1.23\%$
test_mod_wrap[compile-overhead]	6.9504ms	3.7386ms	267.4787 Ops/s	263.8062 Ops/s	$\color{#35bf28}+1.39\%$
test_mod_wrap_and_backward[eager]	1.5169ms	1.3972ms	715.6933 Ops/s	677.4123 Ops/s	$\textbf{\color{#35bf28}+5.65\%}$
test_mod_wrap_and_backward[compile]	1.3814ms	1.3006ms	768.8984 Ops/s	769.8384 Ops/s	$\color{#d91a1a}-0.12\%$
test_mod_wrap_and_backward[compile-overhead]	1.3880ms	0.9308ms	1.0743 KOps/s	1.0692 KOps/s	$\color{#35bf28}+0.48\%$
test_seq_add[eager]	0.1683ms	0.1214ms	8.2383 KOps/s	8.3964 KOps/s	$\color{#d91a1a}-1.88\%$
test_seq_add[compile]	0.1355ms	90.5540μs	11.0431 KOps/s	10.8260 KOps/s	$\color{#35bf28}+2.01\%$
test_seq_add[compile-overhead]	0.1982ms	0.1314ms	7.6082 KOps/s	7.2288 KOps/s	$\textbf{\color{#35bf28}+5.25\%}$
test_seq_wrap[eager]	0.5461ms	0.4408ms	2.2684 KOps/s	2.2424 KOps/s	$\color{#35bf28}+1.16\%$
test_seq_wrap[compile]	0.3698ms	0.3084ms	3.2428 KOps/s	3.2202 KOps/s	$\color{#35bf28}+0.70\%$
test_seq_wrap[compile-overhead]	0.2796ms	0.2291ms	4.3648 KOps/s	4.2936 KOps/s	$\color{#35bf28}+1.66\%$
test_func_call_runtime[False-eager]	0.8570ms	0.7597ms	1.3162 KOps/s	1.2798 KOps/s	$\color{#35bf28}+2.85\%$
test_func_call_runtime[False-compile]	1.2804ms	0.7810ms	1.2804 KOps/s	1.3032 KOps/s	$\color{#d91a1a}-1.75\%$
test_func_call_runtime[False-compile-overhead]	0.4191ms	0.3724ms	2.6850 KOps/s	2.6899 KOps/s	$\color{#d91a1a}-0.19\%$
test_func_call_runtime[True-eager]	1.0262ms	0.9467ms	1.0563 KOps/s	1.0659 KOps/s	$\color{#d91a1a}-0.90\%$
test_func_call_runtime[True-compile]	0.8706ms	0.7976ms	1.2537 KOps/s	1.2712 KOps/s	$\color{#d91a1a}-1.38\%$
test_func_call_runtime[True-compile-overhead]	0.4518ms	0.3925ms	2.5480 KOps/s	2.5615 KOps/s	$\color{#d91a1a}-0.52\%$
test_func_call_cm_runtime[False-eager]	1.2580ms	0.7692ms	1.3001 KOps/s	1.3001 KOps/s	$-0.00\%$
test_func_call_cm_runtime[False-compile]	0.9259ms	0.7679ms	1.3023 KOps/s	1.2999 KOps/s	$\color{#35bf28}+0.18\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5988ms	0.3734ms	2.6784 KOps/s	2.6848 KOps/s	$\color{#d91a1a}-0.24\%$
test_func_call_cm_runtime[True-eager]	1.4457ms	1.0415ms	960.1217 Ops/s	959.0880 Ops/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[True-compile]	1.4349ms	1.0250ms	975.5976 Ops/s	972.1849 Ops/s	$\color{#35bf28}+0.35\%$
test_func_call_cm_runtime[True-compile-overhead]	1.4322ms	1.0266ms	974.0943 Ops/s	966.7811 Ops/s	$\color{#35bf28}+0.76\%$
test_vmap_func_call_cm_runtime[eager]	2.5518ms	2.1299ms	469.5019 Ops/s	462.6734 Ops/s	$\color{#35bf28}+1.48\%$
test_vmap_func_call_cm_runtime[compile]	0.9702ms	0.8349ms	1.1977 KOps/s	1.1888 KOps/s	$\color{#35bf28}+0.74\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4764ms	0.4232ms	2.3627 KOps/s	2.3547 KOps/s	$\color{#35bf28}+0.34\%$
test_distributed	3.2317ms	0.1786ms	5.6004 KOps/s	8.4117 KOps/s	$\textbf{\color{#d91a1a}-33.42\%}$
test_tdmodule	93.2610μs	22.1286μs	45.1904 KOps/s	48.6898 KOps/s	$\textbf{\color{#d91a1a}-7.19\%}$
test_tdmodule_dispatch	61.7210μs	40.9602μs	24.4139 KOps/s	27.2139 KOps/s	$\textbf{\color{#d91a1a}-10.29\%}$
test_tdseq	42.3200μs	22.1685μs	45.1090 KOps/s	47.8167 KOps/s	$\textbf{\color{#d91a1a}-5.66\%}$
test_tdseq_dispatch	61.8310μs	41.4786μs	24.1088 KOps/s	25.2680 KOps/s	$\color{#d91a1a}-4.59\%$
test_instantiation_functorch	1.7599ms	1.5898ms	629.0106 Ops/s	632.4824 Ops/s	$\color{#d91a1a}-0.55\%$
test_exec_functorch	0.2144ms	0.1494ms	6.6928 KOps/s	6.8768 KOps/s	$\color{#d91a1a}-2.68\%$
test_exec_functional_call	0.2102ms	0.1446ms	6.9178 KOps/s	7.1994 KOps/s	$\color{#d91a1a}-3.91\%$
test_exec_td_decorator	0.3835ms	0.1954ms	5.1172 KOps/s	5.2310 KOps/s	$\color{#d91a1a}-2.18\%$
test_vmap_mlp_speed_decorator[True-True]	0.8168ms	0.7043ms	1.4198 KOps/s	1.4163 KOps/s	$\color{#35bf28}+0.25\%$
test_vmap_mlp_speed_decorator[True-False]	0.8686ms	0.7046ms	1.4193 KOps/s	1.4179 KOps/s	$\color{#35bf28}+0.10\%$
test_vmap_mlp_speed_decorator[False-True]	0.9966ms	0.6084ms	1.6436 KOps/s	1.6246 KOps/s	$\color{#35bf28}+1.17\%$
test_vmap_mlp_speed_decorator[False-False]	0.7577ms	0.6096ms	1.6405 KOps/s	1.6175 KOps/s	$\color{#35bf28}+1.42\%$
test_vmap_transformer_speed_decorator[True-True]	20.5111ms	19.6106ms	50.9929 Ops/s	50.8153 Ops/s	$\color{#35bf28}+0.35\%$
test_vmap_transformer_speed_decorator[True-False]	19.9402ms	19.5769ms	51.0805 Ops/s	50.7953 Ops/s	$\color{#35bf28}+0.56\%$
test_vmap_transformer_speed_decorator[False-True]	19.5937ms	19.4154ms	51.5055 Ops/s	51.4167 Ops/s	$\color{#35bf28}+0.17\%$
test_vmap_transformer_speed_decorator[False-False]	19.5964ms	19.4064ms	51.5293 Ops/s	51.4861 Ops/s	$\color{#35bf28}+0.08\%$
test_to_module_speed[True]	1.4668ms	0.9903ms	1.0098 KOps/s	1.0124 KOps/s	$\color{#d91a1a}-0.26\%$
test_to_module_speed[False]	1.3414ms	0.9582ms	1.0437 KOps/s	1.0352 KOps/s	$\color{#35bf28}+0.82\%$
test_tc_init	69.1610μs	37.1393μs	26.9257 KOps/s	27.1839 KOps/s	$\color{#d91a1a}-0.95\%$
test_tc_init_nested	0.1258ms	76.1957μs	13.1241 KOps/s	13.8663 KOps/s	$\textbf{\color{#d91a1a}-5.35\%}$
test_tc_first_layer_tensor	4.1344μs	0.7000μs	1.4286 MOps/s	1.4073 MOps/s	$\color{#35bf28}+1.51\%$
test_tc_first_layer_nontensor	0.4104ms	2.2411μs	446.2087 KOps/s	443.4371 KOps/s	$\color{#35bf28}+0.63\%$
test_tc_second_layer_tensor	10.3833μs	1.4156μs	706.3998 KOps/s	706.0608 KOps/s	$\color{#35bf28}+0.05\%$
test_tc_second_layer_nontensor	0.4125ms	2.9673μs	337.0045 KOps/s	336.3363 KOps/s	$\color{#35bf28}+0.20\%$
test_unbind	0.2275s	9.9010ms	100.9995 Ops/s	140.0228 Ops/s	$\textbf{\color{#d91a1a}-27.87\%}$
test_full_like	10.1977ms	9.2311ms	108.3300 Ops/s	108.2203 Ops/s	$\color{#35bf28}+0.10\%$
test_zeros_like	4.9105ms	4.3196ms	231.5046 Ops/s	233.8228 Ops/s	$\color{#d91a1a}-0.99\%$
test_ones_like	4.7100ms	4.3201ms	231.4768 Ops/s	236.4982 Ops/s	$\color{#d91a1a}-2.12\%$
test_clone	6.7694ms	6.4077ms	156.0621 Ops/s	108.9287 Ops/s	$\textbf{\color{#35bf28}+43.27\%}$
test_squeeze	80.2810μs	10.0628μs	99.3764 KOps/s	101.6485 KOps/s	$\color{#d91a1a}-2.24\%$
test_unsqueeze	0.1253ms	74.7791μs	13.3727 KOps/s	13.1845 KOps/s	$\color{#35bf28}+1.43\%$
test_split	0.5690ms	0.1634ms	6.1188 KOps/s	6.2265 KOps/s	$\color{#d91a1a}-1.73\%$
test_permute	0.2437ms	0.1769ms	5.6520 KOps/s	5.5071 KOps/s	$\color{#35bf28}+2.63\%$
test_stack	51.2908ms	50.8172ms	19.6784 Ops/s	19.6944 Ops/s	$\color{#d91a1a}-0.08\%$
test_cat	51.0912ms	50.6771ms	19.7328 Ops/s	19.7262 Ops/s	$\color{#35bf28}+0.03\%$