[BugFix] Better comparison of tensorclasses

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Better comparison of tensorclasses #1137

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}34$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	44.4930μs	17.8770μs	55.9378 KOps/s	59.1072 KOps/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_plain_set_stack_nested	69.0590μs	18.1843μs	54.9926 KOps/s	57.8404 KOps/s	$\color{#d91a1a}-4.92\%$
test_plain_set_nested_inplace	0.1027ms	19.7520μs	50.6278 KOps/s	53.6332 KOps/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_plain_set_stack_nested_inplace	73.6970μs	19.9690μs	50.0776 KOps/s	53.3846 KOps/s	$\textbf{\color{#d91a1a}-6.19\%}$
test_items	43.8310μs	4.2455μs	235.5427 KOps/s	238.9171 KOps/s	$\color{#d91a1a}-1.41\%$
test_items_nested	0.5643ms	0.4114ms	2.4310 KOps/s	2.5291 KOps/s	$\color{#d91a1a}-3.88\%$
test_items_nested_locked	0.5088ms	0.4058ms	2.4645 KOps/s	2.5114 KOps/s	$\color{#d91a1a}-1.87\%$
test_items_nested_leaf	0.1300ms	71.1990μs	14.0451 KOps/s	14.0357 KOps/s	$\color{#35bf28}+0.07\%$
test_items_stack_nested	0.7349ms	0.4161ms	2.4030 KOps/s	2.5016 KOps/s	$\color{#d91a1a}-3.94\%$
test_items_stack_nested_leaf	0.1431ms	73.3390μs	13.6353 KOps/s	13.9635 KOps/s	$\color{#d91a1a}-2.35\%$
test_items_stack_nested_locked	0.6120ms	0.4090ms	2.4450 KOps/s	2.5135 KOps/s	$\color{#d91a1a}-2.73\%$
test_keys	42.9100μs	3.6446μs	274.3798 KOps/s	284.0541 KOps/s	$\color{#d91a1a}-3.41\%$
test_keys_nested	0.2459ms	0.1398ms	7.1538 KOps/s	7.3618 KOps/s	$\color{#d91a1a}-2.82\%$
test_keys_nested_locked	1.8638ms	0.1462ms	6.8412 KOps/s	6.9495 KOps/s	$\color{#d91a1a}-1.56\%$
test_keys_nested_leaf	0.2224ms	0.1216ms	8.2206 KOps/s	8.4559 KOps/s	$\color{#d91a1a}-2.78\%$
test_keys_stack_nested	0.2453ms	0.1411ms	7.0851 KOps/s	7.1183 KOps/s	$\color{#d91a1a}-0.47\%$
test_keys_stack_nested_leaf	0.1935ms	0.1201ms	8.3298 KOps/s	8.2516 KOps/s	$\color{#35bf28}+0.95\%$
test_keys_stack_nested_locked	0.2670ms	0.1461ms	6.8433 KOps/s	7.0011 KOps/s	$\color{#d91a1a}-2.25\%$
test_values	8.2034μs	1.0481μs	954.1489 KOps/s	940.8448 KOps/s	$\color{#35bf28}+1.41\%$
test_values_nested	0.1090ms	56.5359μs	17.6879 KOps/s	17.4122 KOps/s	$\color{#35bf28}+1.58\%$
test_values_nested_locked	0.1155ms	56.5265μs	17.6908 KOps/s	17.4572 KOps/s	$\color{#35bf28}+1.34\%$
test_values_nested_leaf	0.1209ms	60.6143μs	16.4978 KOps/s	16.0899 KOps/s	$\color{#35bf28}+2.54\%$
test_values_stack_nested	98.9740μs	57.6188μs	17.3554 KOps/s	17.7001 KOps/s	$\color{#d91a1a}-1.95\%$
test_values_stack_nested_leaf	0.1220ms	61.4513μs	16.2730 KOps/s	16.3074 KOps/s	$\color{#d91a1a}-0.21\%$
test_values_stack_nested_locked	0.1055ms	57.2608μs	17.4639 KOps/s	17.3874 KOps/s	$\color{#35bf28}+0.44\%$
test_membership	23.2330μs	0.9401μs	1.0638 MOps/s	1.0955 MOps/s	$\color{#d91a1a}-2.89\%$
test_membership_nested	52.0870μs	3.0542μs	327.4147 KOps/s	341.7034 KOps/s	$\color{#d91a1a}-4.18\%$
test_membership_nested_leaf	55.9150μs	3.0291μs	330.1295 KOps/s	339.2267 KOps/s	$\color{#d91a1a}-2.68\%$
test_membership_stacked_nested	20.8990μs	2.9909μs	334.3514 KOps/s	342.8317 KOps/s	$\color{#d91a1a}-2.47\%$
test_membership_stacked_nested_leaf	27.8010μs	3.0223μs	330.8783 KOps/s	340.5140 KOps/s	$\color{#d91a1a}-2.83\%$
test_membership_nested_last	54.1710μs	4.2645μs	234.4956 KOps/s	225.6845 KOps/s	$\color{#35bf28}+3.90\%$
test_membership_nested_leaf_last	28.1430μs	4.2998μs	232.5685 KOps/s	227.6255 KOps/s	$\color{#35bf28}+2.17\%$
test_membership_stacked_nested_last	53.2390μs	4.2779μs	233.7589 KOps/s	243.6463 KOps/s	$\color{#d91a1a}-4.06\%$
test_membership_stacked_nested_leaf_last	31.6990μs	4.3398μs	230.4268 KOps/s	242.9223 KOps/s	$\textbf{\color{#d91a1a}-5.14\%}$
test_nested_getleaf	58.1980μs	10.7368μs	93.1377 KOps/s	96.0261 KOps/s	$\color{#d91a1a}-3.01\%$
test_nested_get	54.0710μs	10.2056μs	97.9851 KOps/s	99.9877 KOps/s	$\color{#d91a1a}-2.00\%$
test_stacked_getleaf	41.3270μs	10.6487μs	93.9079 KOps/s	94.9396 KOps/s	$\color{#d91a1a}-1.09\%$
test_stacked_get	51.7160μs	10.1033μs	98.9773 KOps/s	99.2145 KOps/s	$\color{#d91a1a}-0.24\%$
test_nested_getitemleaf	57.3670μs	11.3336μs	88.2330 KOps/s	90.1961 KOps/s	$\color{#d91a1a}-2.18\%$
test_nested_getitem	32.9220μs	10.3536μs	96.5849 KOps/s	98.1268 KOps/s	$\color{#d91a1a}-1.57\%$
test_stacked_getitemleaf	61.5270μs	11.1960μs	89.3176 KOps/s	90.3725 KOps/s	$\color{#d91a1a}-1.17\%$
test_stacked_getitem	38.5600μs	10.2934μs	97.1494 KOps/s	96.2582 KOps/s	$\color{#35bf28}+0.93\%$
test_lock_nested	5.2927ms	0.4507ms	2.2189 KOps/s	2.2763 KOps/s	$\color{#d91a1a}-2.52\%$
test_lock_stack_nested	0.5129ms	0.4178ms	2.3934 KOps/s	2.4421 KOps/s	$\color{#d91a1a}-1.99\%$
test_unlock_nested	0.8256ms	0.3645ms	2.7437 KOps/s	2.7309 KOps/s	$\color{#35bf28}+0.47\%$
test_unlock_stack_nested	0.4299ms	0.3381ms	2.9577 KOps/s	3.0568 KOps/s	$\color{#d91a1a}-3.24\%$
test_flatten_speed	0.1743ms	94.6575μs	10.5644 KOps/s	10.5421 KOps/s	$\color{#35bf28}+0.21\%$
test_unflatten_speed	0.6283ms	0.4880ms	2.0493 KOps/s	2.0452 KOps/s	$\color{#35bf28}+0.20\%$
test_common_ops	4.9784ms	0.7924ms	1.2620 KOps/s	1.3576 KOps/s	$\textbf{\color{#d91a1a}-7.04\%}$
test_creation	76.0420μs	2.1023μs	475.6778 KOps/s	478.4002 KOps/s	$\color{#d91a1a}-0.57\%$
test_creation_empty	56.9160μs	11.4013μs	87.7094 KOps/s	104.3418 KOps/s	$\textbf{\color{#d91a1a}-15.94\%}$
test_creation_nested_1	63.2600μs	14.2857μs	70.0002 KOps/s	81.4639 KOps/s	$\textbf{\color{#d91a1a}-14.07\%}$
test_creation_nested_2	40.3060μs	18.7961μs	53.2024 KOps/s	60.5255 KOps/s	$\textbf{\color{#d91a1a}-12.10\%}$
test_clone	99.1450μs	14.6526μs	68.2475 KOps/s	76.0939 KOps/s	$\textbf{\color{#d91a1a}-10.31\%}$
test_getitem[int]	1.1706ms	12.8600μs	77.7604 KOps/s	78.2689 KOps/s	$\color{#d91a1a}-0.65\%$
test_getitem[slice_int]	0.1494ms	24.9228μs	40.1239 KOps/s	41.3616 KOps/s	$\color{#d91a1a}-2.99\%$
test_getitem[range]	0.1945ms	49.1203μs	20.3582 KOps/s	21.6921 KOps/s	$\textbf{\color{#d91a1a}-6.15\%}$
test_getitem[tuple]	0.1302ms	20.7140μs	48.2765 KOps/s	49.0500 KOps/s	$\color{#d91a1a}-1.58\%$
test_getitem[list]	0.2052ms	44.1296μs	22.6605 KOps/s	23.7618 KOps/s	$\color{#d91a1a}-4.63\%$
test_setitem_dim[int]	46.0660μs	26.2335μs	38.1191 KOps/s	39.7659 KOps/s	$\color{#d91a1a}-4.14\%$
test_setitem_dim[slice_int]	88.4950μs	53.8981μs	18.5535 KOps/s	19.1697 KOps/s	$\color{#d91a1a}-3.21\%$
test_setitem_dim[range]	0.1363ms	73.5884μs	13.5891 KOps/s	13.8917 KOps/s	$\color{#d91a1a}-2.18\%$
test_setitem_dim[tuple]	78.5870μs	42.4457μs	23.5595 KOps/s	24.4330 KOps/s	$\color{#d91a1a}-3.58\%$
test_setitem	82.5040μs	21.0630μs	47.4766 KOps/s	51.9602 KOps/s	$\textbf{\color{#d91a1a}-8.63\%}$
test_set	84.0970μs	20.4598μs	48.8764 KOps/s	53.1357 KOps/s	$\textbf{\color{#d91a1a}-8.02\%}$
test_set_shared	3.8560ms	0.1654ms	6.0442 KOps/s	5.9510 KOps/s	$\color{#35bf28}+1.57\%$
test_update	0.1495ms	23.2550μs	43.0014 KOps/s	48.7126 KOps/s	$\textbf{\color{#d91a1a}-11.72\%}$
test_update_nested	0.1198ms	34.3690μs	29.0959 KOps/s	33.0971 KOps/s	$\textbf{\color{#d91a1a}-12.09\%}$
test_update__nested	0.9216ms	33.0287μs	30.2767 KOps/s	31.5369 KOps/s	$\color{#d91a1a}-4.00\%$
test_set_nested	87.0320μs	22.7160μs	44.0218 KOps/s	48.2648 KOps/s	$\textbf{\color{#d91a1a}-8.79\%}$
test_set_nested_new	0.1009ms	27.2835μs	36.6522 KOps/s	39.5251 KOps/s	$\textbf{\color{#d91a1a}-7.27\%}$
test_select	0.1389ms	43.9150μs	22.7713 KOps/s	23.7762 KOps/s	$\color{#d91a1a}-4.23\%$
test_select_nested	0.1212ms	59.9687μs	16.6754 KOps/s	16.5324 KOps/s	$\color{#35bf28}+0.86\%$
test_exclude_nested	0.1729ms	80.4490μs	12.4302 KOps/s	12.7444 KOps/s	$\color{#d91a1a}-2.47\%$
test_empty[True]	0.5763ms	0.3844ms	2.6014 KOps/s	2.6234 KOps/s	$\color{#d91a1a}-0.84\%$
test_empty[False]	10.5323μs	1.3041μs	766.8288 KOps/s	811.0548 KOps/s	$\textbf{\color{#d91a1a}-5.45\%}$
test_unbind_speed	0.3668ms	0.2696ms	3.7095 KOps/s	3.8472 KOps/s	$\color{#d91a1a}-3.58\%$
test_unbind_speed_stack0	0.5902ms	0.2622ms	3.8135 KOps/s	3.9238 KOps/s	$\color{#d91a1a}-2.81\%$
test_unbind_speed_stack1	98.5437ms	0.7640ms	1.3089 KOps/s	1.4567 KOps/s	$\textbf{\color{#d91a1a}-10.14\%}$
test_split	1.7432ms	1.5646ms	639.1363 Ops/s	585.7770 Ops/s	$\textbf{\color{#35bf28}+9.11\%}$
test_chunk	98.6856ms	1.8799ms	531.9564 Ops/s	583.3770 Ops/s	$\textbf{\color{#d91a1a}-8.81\%}$
test_consolidate_njt[False-None]	8.2262ms	8.0359ms	124.4423 Ops/s	124.1124 Ops/s	$\color{#35bf28}+0.27\%$
test_creation[device0]	0.2755ms	88.8653μs	11.2530 KOps/s	10.9394 KOps/s	$\color{#35bf28}+2.87\%$
test_creation_from_tensor	4.1975ms	93.2938μs	10.7188 KOps/s	10.4469 KOps/s	$\color{#35bf28}+2.60\%$
test_add_one[memmap_tensor0]	0.1419ms	4.9938μs	200.2473 KOps/s	215.0742 KOps/s	$\textbf{\color{#d91a1a}-6.89\%}$
test_contiguous[memmap_tensor0]	12.1330μs	0.5200μs	1.9232 MOps/s	1.9593 MOps/s	$\color{#d91a1a}-1.85\%$
test_stack[memmap_tensor0]	35.0750μs	3.3489μs	298.6063 KOps/s	301.0359 KOps/s	$\color{#d91a1a}-0.81\%$
test_memmaptd_index	1.0074ms	0.2372ms	4.2158 KOps/s	4.2371 KOps/s	$\color{#d91a1a}-0.50\%$
test_memmaptd_index_astensor	0.6150ms	0.3155ms	3.1700 KOps/s	3.1999 KOps/s	$\color{#d91a1a}-0.93\%$
test_memmaptd_index_op	0.9850ms	0.5817ms	1.7190 KOps/s	1.8224 KOps/s	$\textbf{\color{#d91a1a}-5.67\%}$
test_serialize_model	0.1221s	0.1129s	8.8535 Ops/s	7.8052 Ops/s	$\textbf{\color{#35bf28}+13.43\%}$
test_serialize_model_pickle	0.4671s	0.3926s	2.5472 Ops/s	2.5262 Ops/s	$\color{#35bf28}+0.83\%$
test_serialize_weights	0.2246s	0.1271s	7.8701 Ops/s	8.8461 Ops/s	$\textbf{\color{#d91a1a}-11.03\%}$
test_serialize_weights_returnearly	0.1655s	0.1566s	6.3846 Ops/s	6.4112 Ops/s	$\color{#d91a1a}-0.41\%$
test_serialize_weights_pickle	0.5756s	0.4319s	2.3155 Ops/s	2.5385 Ops/s	$\textbf{\color{#d91a1a}-8.79\%}$
test_serialize_weights_filesystem	0.1533s	0.1452s	6.8890 Ops/s	6.4996 Ops/s	$\textbf{\color{#35bf28}+5.99\%}$
test_serialize_model_filesystem	0.2348s	0.1550s	6.4522 Ops/s	6.8173 Ops/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_reshape_pytree	59.7710μs	26.9006μs	37.1739 KOps/s	36.9823 KOps/s	$\color{#35bf28}+0.52\%$
test_reshape_td	93.3850μs	32.6107μs	30.6648 KOps/s	30.4394 KOps/s	$\color{#35bf28}+0.74\%$
test_view_pytree	64.0300μs	27.0364μs	36.9872 KOps/s	36.6019 KOps/s	$\color{#35bf28}+1.05\%$
test_view_td	93.3540μs	37.4172μs	26.7257 KOps/s	26.2795 KOps/s	$\color{#35bf28}+1.70\%$
test_unbind_pytree	82.4040μs	30.5796μs	32.7015 KOps/s	32.9509 KOps/s	$\color{#d91a1a}-0.76\%$
test_unbind_td	0.3236ms	38.5965μs	25.9091 KOps/s	26.3184 KOps/s	$\color{#d91a1a}-1.56\%$
test_split_pytree	66.3950μs	30.1337μs	33.1855 KOps/s	33.1825 KOps/s	$+0.01\%$
test_split_td	0.4849ms	44.8743μs	22.2845 KOps/s	22.7126 KOps/s	$\color{#d91a1a}-1.88\%$
test_add_pytree	98.0930μs	36.0233μs	27.7598 KOps/s	27.6428 KOps/s	$\color{#35bf28}+0.42\%$
test_add_td	0.1488ms	57.6692μs	17.3403 KOps/s	18.6575 KOps/s	$\textbf{\color{#d91a1a}-7.06\%}$
test_compile_add_one_nested[tensordict-compile]	0.1157ms	61.6092μs	16.2314 KOps/s	16.2816 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_add_one_nested[tensordict-eager]	0.3562ms	0.1606ms	6.2283 KOps/s	6.1214 KOps/s	$\color{#35bf28}+1.75\%$
test_compile_add_one_nested[pytree-compile]	0.1068ms	45.5007μs	21.9777 KOps/s	21.8353 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_add_one_nested[pytree-eager]	0.2212ms	0.1208ms	8.2812 KOps/s	8.4235 KOps/s	$\color{#d91a1a}-1.69\%$
test_compile_copy_nested[tensordict-compile]	66.9550μs	26.4222μs	37.8470 KOps/s	38.5679 KOps/s	$\color{#d91a1a}-1.87\%$
test_compile_copy_nested[tensordict-eager]	0.1188ms	53.7802μs	18.5942 KOps/s	18.6968 KOps/s	$\color{#d91a1a}-0.55\%$
test_compile_copy_nested[pytree-compile]	0.1791ms	79.0502μs	12.6502 KOps/s	12.4455 KOps/s	$\color{#35bf28}+1.65\%$
test_compile_copy_nested[pytree-eager]	0.1377ms	68.3117μs	14.6388 KOps/s	14.4662 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_add_one_flat[tensordict-compile]	0.1821ms	0.1029ms	9.7178 KOps/s	9.6014 KOps/s	$\color{#35bf28}+1.21\%$
test_compile_add_one_flat[tensordict-eager]	0.4453ms	0.2012ms	4.9689 KOps/s	4.9016 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_add_one_flat[tensorclass-compile]	96.2400μs	44.3299μs	22.5581 KOps/s	22.4856 KOps/s	$\color{#35bf28}+0.32\%$
test_compile_add_one_flat[tensorclass-eager]	0.4589ms	62.0274μs	16.1219 KOps/s	16.2810 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_add_one_flat[pytree-compile]	0.1752ms	0.1008ms	9.9218 KOps/s	9.7484 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_add_one_flat[pytree-eager]	0.2746ms	0.2005ms	4.9864 KOps/s	4.8431 KOps/s	$\color{#35bf28}+2.96\%$
test_compile_add_self_flat[tensordict-eager]	0.5226ms	0.2181ms	4.5850 KOps/s	4.6035 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_add_self_flat[tensordict-compile]	0.1746ms	0.1054ms	9.4898 KOps/s	9.5365 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_add_self_flat[tensorclass-eager]	0.1828ms	53.3205μs	18.7545 KOps/s	17.8027 KOps/s	$\textbf{\color{#35bf28}+5.35\%}$
test_compile_add_self_flat[tensorclass-compile]	91.0000μs	45.6027μs	21.9285 KOps/s	21.4025 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_add_self_flat[pytree-eager]	0.6381ms	0.1584ms	6.3147 KOps/s	6.2600 KOps/s	$\color{#35bf28}+0.87\%$
test_compile_add_self_flat[pytree-compile]	0.2179ms	0.1026ms	9.7513 KOps/s	9.7692 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_copy_flat[tensordict-compile]	54.1610μs	21.3862μs	46.7590 KOps/s	47.6258 KOps/s	$\color{#d91a1a}-1.82\%$
test_compile_copy_flat[tensordict-eager]	0.1488ms	59.3178μs	16.8583 KOps/s	16.8565 KOps/s	$\color{#35bf28}+0.01\%$
test_compile_copy_flat[pytree-compile]	0.1492ms	82.6226μs	12.1032 KOps/s	12.1148 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_copy_flat[pytree-eager]	0.1431ms	69.9820μs	14.2894 KOps/s	14.1974 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_assign_and_add[tensordict-compile]	0.3857ms	0.2098ms	4.7668 KOps/s	4.6561 KOps/s	$\color{#35bf28}+2.38\%$
test_compile_assign_and_add[tensordict-eager]	1.5247ms	1.2826ms	779.6806 Ops/s	760.1641 Ops/s	$\color{#35bf28}+2.57\%$
test_compile_assign_and_add[pytree-compile]	0.2688ms	0.2002ms	4.9962 KOps/s	4.9213 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_assign_and_add[pytree-eager]	1.0066ms	0.7731ms	1.2935 KOps/s	1.2730 KOps/s	$\color{#35bf28}+1.61\%$
test_compile_assign_and_add_stack[compile]	0.9009ms	0.4533ms	2.2058 KOps/s	2.1663 KOps/s	$\color{#35bf28}+1.83\%$
test_compile_assign_and_add_stack[eager]	4.1241ms	2.6651ms	375.2185 Ops/s	392.7567 Ops/s	$\color{#d91a1a}-4.47\%$
test_compile_indexing[tensor-tensordict-compile]	81.5420μs	36.5119μs	27.3883 KOps/s	28.1768 KOps/s	$\color{#d91a1a}-2.80\%$
test_compile_indexing[tensor-tensordict-eager]	0.6567ms	31.9263μs	31.3221 KOps/s	30.5929 KOps/s	$\color{#35bf28}+2.38\%$
test_compile_indexing[tensor-tensorclass-compile]	70.3620μs	28.9957μs	34.4879 KOps/s	34.0307 KOps/s	$\color{#35bf28}+1.34\%$
test_compile_indexing[tensor-tensorclass-eager]	66.5040μs	23.0052μs	43.4685 KOps/s	42.8864 KOps/s	$\color{#35bf28}+1.36\%$
test_compile_indexing[tensor-pytree-compile]	73.0770μs	29.8523μs	33.4983 KOps/s	32.9846 KOps/s	$\color{#35bf28}+1.56\%$
test_compile_indexing[tensor-pytree-eager]	63.1780μs	23.0831μs	43.3218 KOps/s	42.5165 KOps/s	$\color{#35bf28}+1.89\%$
test_compile_indexing[slice-tensordict-compile]	0.1011ms	52.1142μs	19.1886 KOps/s	19.6007 KOps/s	$\color{#d91a1a}-2.10\%$
test_compile_indexing[slice-tensordict-eager]	0.4994ms	19.5228μs	51.2220 KOps/s	48.8790 KOps/s	$\color{#35bf28}+4.79\%$
test_compile_indexing[slice-tensorclass-compile]	95.8290μs	44.4752μs	22.4844 KOps/s	22.8682 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_indexing[slice-tensorclass-eager]	80.3230μs	19.0712μs	52.4351 KOps/s	51.8512 KOps/s	$\color{#35bf28}+1.13\%$
test_compile_indexing[slice-pytree-compile]	0.1001ms	45.0881μs	22.1788 KOps/s	22.6462 KOps/s	$\color{#d91a1a}-2.06\%$
test_compile_indexing[slice-pytree-eager]	57.9680μs	18.9151μs	52.8678 KOps/s	51.2540 KOps/s	$\color{#35bf28}+3.15\%$
test_compile_indexing[int-tensordict-compile]	0.1037ms	53.2021μs	18.7963 KOps/s	19.4271 KOps/s	$\color{#d91a1a}-3.25\%$
test_compile_indexing[int-tensordict-eager]	0.8865ms	19.5807μs	51.0706 KOps/s	49.0439 KOps/s	$\color{#35bf28}+4.13\%$
test_compile_indexing[int-tensorclass-compile]	90.2480μs	45.0480μs	22.1985 KOps/s	22.7802 KOps/s	$\color{#d91a1a}-2.55\%$
test_compile_indexing[int-tensorclass-eager]	57.4370μs	19.0695μs	52.4397 KOps/s	52.0514 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_indexing[int-pytree-compile]	92.6430μs	45.0594μs	22.1929 KOps/s	22.7331 KOps/s	$\color{#d91a1a}-2.38\%$
test_compile_indexing[int-pytree-eager]	75.4610μs	18.8631μs	53.0135 KOps/s	51.9061 KOps/s	$\color{#35bf28}+2.13\%$
test_mod_add[eager]	71.6740μs	33.8369μs	29.5535 KOps/s	30.3264 KOps/s	$\color{#d91a1a}-2.55\%$
test_mod_add[compile]	94.5270μs	48.4369μs	20.6454 KOps/s	20.6314 KOps/s	$\color{#35bf28}+0.07\%$
test_mod_add[compile-overhead]	0.1215ms	47.5972μs	21.0096 KOps/s	20.4555 KOps/s	$\color{#35bf28}+2.71\%$
test_mod_wrap[eager]	0.4320ms	0.2276ms	4.3938 KOps/s	4.3910 KOps/s	$\color{#35bf28}+0.06\%$
test_mod_wrap[compile]	0.3434ms	0.2032ms	4.9210 KOps/s	4.8255 KOps/s	$\color{#35bf28}+1.98\%$
test_mod_wrap[compile-overhead]	0.2620ms	0.2020ms	4.9513 KOps/s	4.8614 KOps/s	$\color{#35bf28}+1.85\%$
test_mod_wrap_and_backward[eager]	12.3074ms	10.6820ms	93.6158 Ops/s	92.6525 Ops/s	$\color{#35bf28}+1.04\%$
test_mod_wrap_and_backward[compile]	22.1962ms	13.7475ms	72.7405 Ops/s	93.5862 Ops/s	$\textbf{\color{#d91a1a}-22.27\%}$
test_mod_wrap_and_backward[compile-overhead]	11.7539ms	10.5209ms	95.0492 Ops/s	93.8181 Ops/s	$\color{#35bf28}+1.31\%$
test_seq_add[eager]	0.2366ms	0.1112ms	8.9918 KOps/s	9.1603 KOps/s	$\color{#d91a1a}-1.84\%$
test_seq_add[compile]	0.1490ms	62.1046μs	16.1019 KOps/s	16.4069 KOps/s	$\color{#d91a1a}-1.86\%$
test_seq_add[compile-overhead]	0.1194ms	59.3887μs	16.8382 KOps/s	17.0330 KOps/s	$\color{#d91a1a}-1.14\%$
test_seq_wrap[eager]	0.6022ms	0.4456ms	2.2440 KOps/s	2.3090 KOps/s	$\color{#d91a1a}-2.82\%$
test_seq_wrap[compile]	0.3424ms	0.2251ms	4.4433 KOps/s	4.4144 KOps/s	$\color{#35bf28}+0.66\%$
test_seq_wrap[compile-overhead]	0.3497ms	0.2229ms	4.4873 KOps/s	4.4187 KOps/s	$\color{#35bf28}+1.55\%$
test_func_call_runtime[False-eager]	0.8668ms	0.5594ms	1.7878 KOps/s	1.8313 KOps/s	$\color{#d91a1a}-2.38\%$
test_func_call_runtime[False-compile]	0.6245ms	0.4214ms	2.3731 KOps/s	2.3517 KOps/s	$\color{#35bf28}+0.91\%$
test_func_call_runtime[False-compile-overhead]	0.9345ms	0.4224ms	2.3673 KOps/s	2.3452 KOps/s	$\color{#35bf28}+0.94\%$
test_func_call_runtime[True-eager]	1.2961ms	0.7728ms	1.2941 KOps/s	1.3182 KOps/s	$\color{#d91a1a}-1.83\%$
test_func_call_runtime[True-compile]	0.5589ms	0.4593ms	2.1771 KOps/s	2.1365 KOps/s	$\color{#35bf28}+1.90\%$
test_func_call_runtime[True-compile-overhead]	0.8684ms	0.4648ms	2.1514 KOps/s	2.0942 KOps/s	$\color{#35bf28}+2.73\%$
test_func_call_cm_runtime[False-eager]	0.7791ms	0.5527ms	1.8093 KOps/s	1.8690 KOps/s	$\color{#d91a1a}-3.19\%$
test_func_call_cm_runtime[False-compile]	0.8770ms	0.4258ms	2.3483 KOps/s	2.3326 KOps/s	$\color{#35bf28}+0.67\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5227ms	0.4224ms	2.3675 KOps/s	2.3463 KOps/s	$\color{#35bf28}+0.91\%$
test_func_call_cm_runtime[True-eager]	1.1090ms	0.8939ms	1.1187 KOps/s	1.1039 KOps/s	$\color{#35bf28}+1.34\%$
test_func_call_cm_runtime[True-compile]	0.7915ms	0.4870ms	2.0534 KOps/s	2.0120 KOps/s	$\color{#35bf28}+2.06\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6263ms	0.4846ms	2.0637 KOps/s	2.0234 KOps/s	$\color{#35bf28}+1.99\%$
test_vmap_func_call_cm_runtime[eager]	3.0957ms	1.8808ms	531.6922 Ops/s	523.8237 Ops/s	$\color{#35bf28}+1.50\%$
test_vmap_func_call_cm_runtime[compile]	0.6380ms	0.5188ms	1.9275 KOps/s	1.9251 KOps/s	$\color{#35bf28}+0.12\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6824ms	0.5179ms	1.9308 KOps/s	1.9232 KOps/s	$\color{#35bf28}+0.40\%$
test_distributed	0.3268ms	0.1249ms	8.0038 KOps/s	7.8966 KOps/s	$\color{#35bf28}+1.36\%$
test_tdmodule	42.9410μs	26.1851μs	38.1896 KOps/s	41.2972 KOps/s	$\textbf{\color{#d91a1a}-7.52\%}$
test_tdmodule_dispatch	81.7730μs	47.9163μs	20.8697 KOps/s	21.8835 KOps/s	$\color{#d91a1a}-4.63\%$
test_tdseq	42.9910μs	25.9981μs	38.4643 KOps/s	40.7548 KOps/s	$\textbf{\color{#d91a1a}-5.62\%}$
test_tdseq_dispatch	84.2670μs	50.6926μs	19.7267 KOps/s	21.1760 KOps/s	$\textbf{\color{#d91a1a}-6.84\%}$
test_instantiation_functorch	2.5486ms	1.5672ms	638.0634 Ops/s	658.9787 Ops/s	$\color{#d91a1a}-3.17\%$
test_exec_functorch	0.3039ms	0.1798ms	5.5622 KOps/s	5.5393 KOps/s	$\color{#35bf28}+0.41\%$
test_exec_functional_call	0.3570ms	0.1743ms	5.7376 KOps/s	5.7571 KOps/s	$\color{#d91a1a}-0.34\%$
test_exec_td_decorator	0.4850ms	0.2309ms	4.3299 KOps/s	4.3017 KOps/s	$\color{#35bf28}+0.66\%$
test_vmap_mlp_speed_decorator[True-True]	0.8664ms	0.6449ms	1.5506 KOps/s	1.5629 KOps/s	$\color{#d91a1a}-0.79\%$
test_vmap_mlp_speed_decorator[True-False]	0.8942ms	0.6435ms	1.5539 KOps/s	1.5519 KOps/s	$\color{#35bf28}+0.13\%$
test_vmap_mlp_speed_decorator[False-True]	0.8156ms	0.5215ms	1.9177 KOps/s	1.9062 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_mlp_speed_decorator[False-False]	0.7119ms	0.5212ms	1.9186 KOps/s	1.9068 KOps/s	$\color{#35bf28}+0.62\%$
test_to_module_speed[True]	2.0810ms	1.3146ms	760.6735 Ops/s	770.3718 Ops/s	$\color{#d91a1a}-1.26\%$
test_to_module_speed[False]	2.0668ms	1.2728ms	785.6501 Ops/s	799.8780 Ops/s	$\color{#d91a1a}-1.78\%$
test_tc_init	85.1800μs	44.9949μs	22.2247 KOps/s	23.1743 KOps/s	$\color{#d91a1a}-4.10\%$
test_tc_init_nested	0.1759ms	92.8327μs	10.7721 KOps/s	11.3990 KOps/s	$\textbf{\color{#d91a1a}-5.50\%}$
test_tc_first_layer_tensor	27.8520μs	1.5213μs	657.3257 KOps/s	638.9879 KOps/s	$\color{#35bf28}+2.87\%$
test_tc_first_layer_nontensor	20.9590μs	4.7066μs	212.4659 KOps/s	210.7458 KOps/s	$\color{#35bf28}+0.82\%$
test_tc_second_layer_tensor	32.9010μs	2.8195μs	354.6734 KOps/s	351.7971 KOps/s	$\color{#35bf28}+0.82\%$
test_tc_second_layer_nontensor	28.7940μs	6.0394μs	165.5796 KOps/s	162.3833 KOps/s	$\color{#35bf28}+1.97\%$
test_unbind	0.2194s	12.6188ms	79.2470 Ops/s	79.4686 Ops/s	$\color{#d91a1a}-0.28\%$
test_full_like	17.6927ms	11.6699ms	85.6904 Ops/s	141.7829 Ops/s	$\textbf{\color{#d91a1a}-39.56\%}$
test_zeros_like	13.8497ms	6.8155ms	146.7242 Ops/s	370.5477 Ops/s	$\textbf{\color{#d91a1a}-60.40\%}$
test_ones_like	12.3430ms	7.3509ms	136.0375 Ops/s	320.7381 Ops/s	$\textbf{\color{#d91a1a}-57.59\%}$
test_clone	14.2047ms	8.9644ms	111.5523 Ops/s	204.6325 Ops/s	$\textbf{\color{#d91a1a}-45.49\%}$
test_squeeze	63.4280μs	11.7991μs	84.7521 KOps/s	85.4930 KOps/s	$\color{#d91a1a}-0.87\%$
test_unsqueeze	0.1873ms	88.9157μs	11.2466 KOps/s	11.3049 KOps/s	$\color{#d91a1a}-0.52\%$
test_split	0.5831ms	0.1961ms	5.1005 KOps/s	5.2136 KOps/s	$\color{#d91a1a}-2.17\%$
test_permute	0.3369ms	0.1982ms	5.0450 KOps/s	4.8899 KOps/s	$\color{#35bf28}+3.17\%$
test_stack	28.1564ms	24.0355ms	41.6051 Ops/s	41.3602 Ops/s	$\color{#35bf28}+0.59\%$
test_cat	28.9340ms	23.9189ms	41.8080 Ops/s	40.5601 Ops/s	$\color{#35bf28}+3.08\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}28$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.2920μs	10.7023μs	93.4381 KOps/s	98.7263 KOps/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_plain_set_stack_nested	34.3520μs	10.8950μs	91.7852 KOps/s	98.1602 KOps/s	$\textbf{\color{#d91a1a}-6.49\%}$
test_plain_set_nested_inplace	45.0530μs	11.6566μs	85.7882 KOps/s	90.6958 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_plain_set_stack_nested_inplace	46.7930μs	11.6353μs	85.9450 KOps/s	90.6613 KOps/s	$\textbf{\color{#d91a1a}-5.20\%}$
test_items	25.1810μs	2.9552μs	338.3866 KOps/s	338.9254 KOps/s	$\color{#d91a1a}-0.16\%$
test_items_nested	0.4914ms	0.3660ms	2.7323 KOps/s	2.8140 KOps/s	$\color{#d91a1a}-2.90\%$
test_items_nested_locked	0.5381ms	0.3678ms	2.7189 KOps/s	2.8158 KOps/s	$\color{#d91a1a}-3.44\%$
test_items_nested_leaf	95.6050μs	58.4537μs	17.1076 KOps/s	17.3205 KOps/s	$\color{#d91a1a}-1.23\%$
test_items_stack_nested	0.6389ms	0.3692ms	2.7083 KOps/s	2.8303 KOps/s	$\color{#d91a1a}-4.31\%$
test_items_stack_nested_leaf	89.3450μs	59.3665μs	16.8445 KOps/s	16.9808 KOps/s	$\color{#d91a1a}-0.80\%$
test_items_stack_nested_locked	0.4266ms	0.3690ms	2.7098 KOps/s	2.8258 KOps/s	$\color{#d91a1a}-4.11\%$
test_keys	27.0120μs	3.6886μs	271.1047 KOps/s	289.3891 KOps/s	$\textbf{\color{#d91a1a}-6.32\%}$
test_keys_nested	0.1178ms	69.5321μs	14.3818 KOps/s	14.1145 KOps/s	$\color{#35bf28}+1.89\%$
test_keys_nested_locked	0.7206ms	75.4690μs	13.2505 KOps/s	12.9618 KOps/s	$\color{#35bf28}+2.23\%$
test_keys_nested_leaf	99.2250μs	61.4306μs	16.2785 KOps/s	16.3756 KOps/s	$\color{#d91a1a}-0.59\%$
test_keys_stack_nested	0.1104ms	70.2021μs	14.2446 KOps/s	13.9714 KOps/s	$\color{#35bf28}+1.96\%$
test_keys_stack_nested_leaf	95.0750μs	60.9395μs	16.4097 KOps/s	16.0412 KOps/s	$\color{#35bf28}+2.30\%$
test_keys_stack_nested_locked	0.1253ms	75.9818μs	13.1611 KOps/s	12.9936 KOps/s	$\color{#35bf28}+1.29\%$
test_values	4.8487μs	0.8666μs	1.1540 MOps/s	1.1763 MOps/s	$\color{#d91a1a}-1.90\%$
test_values_nested	60.9330μs	31.2094μs	32.0416 KOps/s	32.1268 KOps/s	$\color{#d91a1a}-0.27\%$
test_values_nested_locked	65.9940μs	32.8265μs	30.4632 KOps/s	30.4974 KOps/s	$\color{#d91a1a}-0.11\%$
test_values_nested_leaf	70.2340μs	33.7182μs	29.6575 KOps/s	29.8888 KOps/s	$\color{#d91a1a}-0.77\%$
test_values_stack_nested	59.7340μs	31.8212μs	31.4256 KOps/s	31.5586 KOps/s	$\color{#d91a1a}-0.42\%$
test_values_stack_nested_leaf	62.7930μs	34.1809μs	29.2561 KOps/s	29.5763 KOps/s	$\color{#d91a1a}-1.08\%$
test_values_stack_nested_locked	60.2530μs	33.3917μs	29.9476 KOps/s	30.1082 KOps/s	$\color{#d91a1a}-0.53\%$
test_membership	1.7201μs	0.5035μs	1.9863 MOps/s	1.9748 MOps/s	$\color{#35bf28}+0.58\%$
test_membership_nested	20.4960μs	2.0398μs	490.2521 KOps/s	496.3917 KOps/s	$\color{#d91a1a}-1.24\%$
test_membership_nested_leaf	15.6555μs	2.0456μs	488.8477 KOps/s	491.1596 KOps/s	$\color{#d91a1a}-0.47\%$
test_membership_stacked_nested	29.5220μs	2.1485μs	465.4366 KOps/s	486.2488 KOps/s	$\color{#d91a1a}-4.28\%$
test_membership_stacked_nested_leaf	23.0910μs	2.1038μs	475.3246 KOps/s	481.1275 KOps/s	$\color{#d91a1a}-1.21\%$
test_membership_nested_last	33.2210μs	3.0715μs	325.5786 KOps/s	336.3705 KOps/s	$\color{#d91a1a}-3.21\%$
test_membership_nested_leaf_last	32.2220μs	3.0734μs	325.3700 KOps/s	336.1607 KOps/s	$\color{#d91a1a}-3.21\%$
test_membership_stacked_nested_last	31.8810μs	5.0285μs	198.8650 KOps/s	332.3328 KOps/s	$\textbf{\color{#d91a1a}-40.16\%}$
test_membership_stacked_nested_leaf_last	28.0520μs	4.9570μs	201.7332 KOps/s	337.3157 KOps/s	$\textbf{\color{#d91a1a}-40.19\%}$
test_nested_getleaf	34.4120μs	6.1877μs	161.6102 KOps/s	163.2163 KOps/s	$\color{#d91a1a}-0.98\%$
test_nested_get	62.5730μs	5.8490μs	170.9701 KOps/s	172.2132 KOps/s	$\color{#d91a1a}-0.72\%$
test_stacked_getleaf	34.1320μs	6.1182μs	163.4458 KOps/s	163.1108 KOps/s	$\color{#35bf28}+0.21\%$
test_stacked_get	30.3920μs	5.8108μs	172.0945 KOps/s	172.4290 KOps/s	$\color{#d91a1a}-0.19\%$
test_nested_getitemleaf	25.6710μs	6.3392μs	157.7475 KOps/s	160.7833 KOps/s	$\color{#d91a1a}-1.89\%$
test_nested_getitem	32.4110μs	6.0647μs	164.8874 KOps/s	168.6592 KOps/s	$\color{#d91a1a}-2.24\%$
test_stacked_getitemleaf	31.3020μs	6.3836μs	156.6514 KOps/s	160.2105 KOps/s	$\color{#d91a1a}-2.22\%$
test_stacked_getitem	33.2020μs	6.0060μs	166.4995 KOps/s	168.6383 KOps/s	$\color{#d91a1a}-1.27\%$
test_lock_nested	9.2258ms	0.3882ms	2.5762 KOps/s	2.6721 KOps/s	$\color{#d91a1a}-3.59\%$
test_lock_stack_nested	0.4262ms	0.3392ms	2.9481 KOps/s	2.9351 KOps/s	$\color{#35bf28}+0.44\%$
test_unlock_nested	0.8010ms	0.3158ms	3.1662 KOps/s	3.2464 KOps/s	$\color{#d91a1a}-2.47\%$
test_unlock_stack_nested	0.3859ms	0.2784ms	3.5919 KOps/s	3.5837 KOps/s	$\color{#35bf28}+0.23\%$
test_flatten_speed	0.1235ms	76.3193μs	13.1028 KOps/s	13.1883 KOps/s	$\color{#d91a1a}-0.65\%$
test_unflatten_speed	0.4442ms	0.3129ms	3.1963 KOps/s	3.2679 KOps/s	$\color{#d91a1a}-2.19\%$
test_common_ops	1.7815ms	0.6181ms	1.6179 KOps/s	1.7182 KOps/s	$\textbf{\color{#d91a1a}-5.84\%}$
test_creation	0.1091ms	1.5050μs	664.4609 KOps/s	671.6914 KOps/s	$\color{#d91a1a}-1.08\%$
test_creation_empty	28.6710μs	7.7654μs	128.7766 KOps/s	152.6303 KOps/s	$\textbf{\color{#d91a1a}-15.63\%}$
test_creation_nested_1	45.3230μs	9.3943μs	106.4478 KOps/s	124.0134 KOps/s	$\textbf{\color{#d91a1a}-14.16\%}$
test_creation_nested_2	43.0820μs	12.1646μs	82.2058 KOps/s	93.9368 KOps/s	$\textbf{\color{#d91a1a}-12.49\%}$
test_clone	57.4530μs	11.1113μs	89.9987 KOps/s	94.9443 KOps/s	$\textbf{\color{#d91a1a}-5.21\%}$
test_getitem[int]	92.4187ms	16.1004μs	62.1102 KOps/s	93.1734 KOps/s	$\textbf{\color{#d91a1a}-33.34\%}$
test_getitem[slice_int]	0.1060ms	22.2588μs	44.9260 KOps/s	47.0895 KOps/s	$\color{#d91a1a}-4.59\%$
test_getitem[range]	0.1274ms	39.3926μs	25.3855 KOps/s	25.8211 KOps/s	$\color{#d91a1a}-1.69\%$
test_getitem[tuple]	0.1103ms	19.1771μs	52.1456 KOps/s	53.4112 KOps/s	$\color{#d91a1a}-2.37\%$
test_getitem[list]	0.2053ms	34.6471μs	28.8625 KOps/s	29.2993 KOps/s	$\color{#d91a1a}-1.49\%$
test_setitem_dim[int]	40.6120μs	19.8966μs	50.2598 KOps/s	53.0516 KOps/s	$\textbf{\color{#d91a1a}-5.26\%}$
test_setitem_dim[slice_int]	65.7340μs	40.3853μs	24.7615 KOps/s	25.6610 KOps/s	$\color{#d91a1a}-3.51\%$
test_setitem_dim[range]	82.3040μs	53.4868μs	18.6962 KOps/s	18.3806 KOps/s	$\color{#35bf28}+1.72\%$
test_setitem_dim[tuple]	65.3740μs	34.2438μs	29.2024 KOps/s	30.5291 KOps/s	$\color{#d91a1a}-4.35\%$
test_setitem	0.3865ms	15.7546μs	63.4736 KOps/s	69.2779 KOps/s	$\textbf{\color{#d91a1a}-8.38\%}$
test_set	0.1006ms	15.3994μs	64.9377 KOps/s	70.8409 KOps/s	$\textbf{\color{#d91a1a}-8.33\%}$
test_set_shared	1.5925ms	0.1480ms	6.7555 KOps/s	6.8028 KOps/s	$\color{#d91a1a}-0.69\%$
test_update	0.3576ms	19.0370μs	52.5294 KOps/s	61.8159 KOps/s	$\textbf{\color{#d91a1a}-15.02\%}$
test_update_nested	0.2609ms	24.3255μs	41.1091 KOps/s	46.5011 KOps/s	$\textbf{\color{#d91a1a}-11.60\%}$
test_update__nested	0.1340ms	25.3509μs	39.4463 KOps/s	40.4086 KOps/s	$\color{#d91a1a}-2.38\%$
test_set_nested	84.5950μs	16.8159μs	59.4674 KOps/s	65.3212 KOps/s	$\textbf{\color{#d91a1a}-8.96\%}$
test_set_nested_new	0.1052ms	19.3395μs	51.7077 KOps/s	56.7209 KOps/s	$\textbf{\color{#d91a1a}-8.84\%}$
test_select	71.4240μs	31.6603μs	31.5853 KOps/s	32.9646 KOps/s	$\color{#d91a1a}-4.18\%$
test_select_nested	83.4440μs	42.1741μs	23.7113 KOps/s	23.7967 KOps/s	$\color{#d91a1a}-0.36\%$
test_exclude_nested	92.3850μs	63.5062μs	15.7465 KOps/s	16.0016 KOps/s	$\color{#d91a1a}-1.59\%$
test_empty[True]	0.4232ms	0.2835ms	3.5273 KOps/s	3.5789 KOps/s	$\color{#d91a1a}-1.44\%$
test_empty[False]	3.0852μs	0.7401μs	1.3511 MOps/s	1.3346 MOps/s	$\color{#35bf28}+1.24\%$
test_to	87.9250μs	55.6806μs	17.9596 KOps/s	17.6668 KOps/s	$\color{#35bf28}+1.66\%$
test_to_nonblocking	90.5050μs	46.6052μs	21.4568 KOps/s	21.8434 KOps/s	$\color{#d91a1a}-1.77\%$
test_unbind_speed	0.2986ms	0.2383ms	4.1973 KOps/s	4.2814 KOps/s	$\color{#d91a1a}-1.97\%$
test_unbind_speed_stack0	0.3352ms	0.2342ms	4.2705 KOps/s	4.2291 KOps/s	$\color{#35bf28}+0.98\%$
test_unbind_speed_stack1	0.6410ms	0.5949ms	1.6809 KOps/s	1.5187 KOps/s	$\textbf{\color{#35bf28}+10.68\%}$
test_split	94.3287ms	1.6258ms	615.0931 Ops/s	636.1440 Ops/s	$\color{#d91a1a}-3.31\%$
test_chunk	93.6998ms	1.6409ms	609.4049 Ops/s	583.6896 Ops/s	$\color{#35bf28}+4.41\%$
test_consolidate[False-None]	95.9819ms	2.8765ms	347.6403 Ops/s	373.0824 Ops/s	$\textbf{\color{#d91a1a}-6.82\%}$
test_consolidate[default-None]	1.7609ms	1.6838ms	593.8950 Ops/s	588.4710 Ops/s	$\color{#35bf28}+0.92\%$
test_consolidate[reduce-overhead-None]	1.8374ms	1.7106ms	584.5984 Ops/s	574.9906 Ops/s	$\color{#35bf28}+1.67\%$
test_consolidate_njt[False-None]	6.7380ms	6.6269ms	150.9005 Ops/s	153.3127 Ops/s	$\color{#d91a1a}-1.57\%$
test_to[False-False-None]	1.8446ms	1.7328ms	577.1133 Ops/s	588.5886 Ops/s	$\color{#d91a1a}-1.95\%$
test_to[True-False-None]	1.5840ms	1.3300ms	751.8805 Ops/s	759.8281 Ops/s	$\color{#d91a1a}-1.05\%$
test_to[within-False-None]	4.1615ms	4.0538ms	246.6827 Ops/s	177.7377 Ops/s	$\textbf{\color{#35bf28}+38.79\%}$
test_to[True-default-None]	5.4945ms	5.2927ms	188.9400 Ops/s	187.7985 Ops/s	$\color{#35bf28}+0.61\%$
test_to_njt[False-False-None]	7.2148ms	7.0968ms	140.9084 Ops/s	138.9320 Ops/s	$\color{#35bf28}+1.42\%$
test_to_njt[True-False-None]	5.8387ms	5.6103ms	178.2441 Ops/s	175.7405 Ops/s	$\color{#35bf28}+1.42\%$
test_to_njt[within-False-None]	12.5597ms	12.3311ms	81.0957 Ops/s	80.0966 Ops/s	$\color{#35bf28}+1.25\%$
test_creation[device0]	0.6739ms	79.8837μs	12.5182 KOps/s	11.9782 KOps/s	$\color{#35bf28}+4.51\%$
test_creation_from_tensor	0.4456ms	83.7637μs	11.9383 KOps/s	11.6978 KOps/s	$\color{#35bf28}+2.06\%$
test_add_one[memmap_tensor0]	0.4764ms	7.1306μs	140.2397 KOps/s	141.7947 KOps/s	$\color{#d91a1a}-1.10\%$
test_contiguous[memmap_tensor0]	2.2917μs	0.4113μs	2.4311 MOps/s	2.4307 MOps/s	$\color{#35bf28}+0.02\%$
test_stack[memmap_tensor0]	26.4420μs	4.6706μs	214.1045 KOps/s	219.7422 KOps/s	$\color{#d91a1a}-2.57\%$
test_memmaptd_index	1.8134ms	0.2525ms	3.9598 KOps/s	3.9687 KOps/s	$\color{#d91a1a}-0.22\%$
test_memmaptd_index_astensor	0.5941ms	0.3077ms	3.2501 KOps/s	3.2292 KOps/s	$\color{#35bf28}+0.65\%$
test_memmaptd_index_op	1.0370ms	0.5949ms	1.6810 KOps/s	1.7377 KOps/s	$\color{#d91a1a}-3.26\%$
test_serialize_model	0.1322s	0.1311s	7.6299 Ops/s	7.6589 Ops/s	$\color{#d91a1a}-0.38\%$
test_serialize_model_pickle	1.3688s	1.2228s	0.8178 Ops/s	0.8252 Ops/s	$\color{#d91a1a}-0.89\%$
test_serialize_weights	0.1315s	0.1305s	7.6647 Ops/s	7.6265 Ops/s	$\color{#35bf28}+0.50\%$
test_serialize_weights_returnearly	0.3465s	64.2160ms	15.5724 Ops/s	14.5441 Ops/s	$\textbf{\color{#35bf28}+7.07\%}$
test_serialize_weights_pickle	1.3785s	1.1912s	0.8395 Ops/s	0.8444 Ops/s	$\color{#d91a1a}-0.58\%$
test_reshape_pytree	54.0630μs	22.8180μs	43.8250 KOps/s	43.8496 KOps/s	$\color{#d91a1a}-0.06\%$
test_reshape_td	52.1930μs	27.1406μs	36.8451 KOps/s	37.3995 KOps/s	$\color{#d91a1a}-1.48\%$
test_view_pytree	51.8620μs	22.4273μs	44.5885 KOps/s	44.2711 KOps/s	$\color{#35bf28}+0.72\%$
test_view_td	69.5940μs	28.7181μs	34.8213 KOps/s	33.7960 KOps/s	$\color{#35bf28}+3.03\%$
test_unbind_pytree	54.0930μs	28.1115μs	35.5727 KOps/s	34.7230 KOps/s	$\color{#35bf28}+2.45\%$
test_unbind_td	0.5459ms	36.0857μs	27.7118 KOps/s	27.1597 KOps/s	$\color{#35bf28}+2.03\%$
test_split_pytree	58.7840μs	30.4302μs	32.8620 KOps/s	32.2766 KOps/s	$\color{#35bf28}+1.81\%$
test_split_td	0.8229ms	39.3507μs	25.4125 KOps/s	25.5710 KOps/s	$\color{#d91a1a}-0.62\%$
test_add_pytree	86.3450μs	35.5365μs	28.1401 KOps/s	27.9218 KOps/s	$\color{#35bf28}+0.78\%$
test_add_td	96.9050μs	46.9849μs	21.2835 KOps/s	21.7068 KOps/s	$\color{#d91a1a}-1.95\%$
test_compile_add_one_nested[tensordict-compile]	0.1780ms	0.1253ms	7.9790 KOps/s	8.0073 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_add_one_nested[tensordict-eager]	0.2315ms	0.1242ms	8.0534 KOps/s	7.8261 KOps/s	$\color{#35bf28}+2.90\%$
test_compile_add_one_nested[pytree-compile]	0.1465ms	98.4132μs	10.1612 KOps/s	10.1102 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_add_one_nested[pytree-eager]	1.1553ms	0.1532ms	6.5274 KOps/s	6.7183 KOps/s	$\color{#d91a1a}-2.84\%$
test_compile_copy_nested[tensordict-compile]	69.5240μs	23.8119μs	41.9958 KOps/s	42.1699 KOps/s	$\color{#d91a1a}-0.41\%$
test_compile_copy_nested[tensordict-eager]	53.1820μs	26.3479μs	37.9538 KOps/s	36.2159 KOps/s	$\color{#35bf28}+4.80\%$
test_compile_copy_nested[pytree-compile]	0.1120ms	63.6274μs	15.7165 KOps/s	14.9670 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_compile_copy_nested[pytree-eager]	90.1450μs	49.6111μs	20.1568 KOps/s	19.7232 KOps/s	$\color{#35bf28}+2.20\%$
test_compile_add_one_flat[tensordict-compile]	0.2042ms	0.1432ms	6.9828 KOps/s	7.0267 KOps/s	$\color{#d91a1a}-0.63\%$
test_compile_add_one_flat[tensordict-eager]	0.2955ms	0.2076ms	4.8162 KOps/s	4.7914 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_add_one_flat[tensorclass-compile]	0.1625ms	99.5542μs	10.0448 KOps/s	10.1374 KOps/s	$\color{#d91a1a}-0.91\%$
test_compile_add_one_flat[tensorclass-eager]	0.1150ms	52.2383μs	19.1430 KOps/s	19.1036 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_add_one_flat[pytree-compile]	0.1950ms	0.1373ms	7.2811 KOps/s	7.2913 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_add_one_flat[pytree-eager]	0.5751ms	0.5030ms	1.9881 KOps/s	2.1249 KOps/s	$\textbf{\color{#d91a1a}-6.44\%}$
test_compile_add_self_flat[tensordict-eager]	0.3846ms	0.2471ms	4.0466 KOps/s	4.0272 KOps/s	$\color{#35bf28}+0.48\%$
test_compile_add_self_flat[tensordict-compile]	0.1848ms	0.1446ms	6.9170 KOps/s	6.8728 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_add_self_flat[tensorclass-eager]	0.1719ms	63.0264μs	15.8664 KOps/s	15.8305 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_add_self_flat[tensorclass-compile]	0.1409ms	99.2958μs	10.0709 KOps/s	10.0753 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_add_self_flat[pytree-eager]	0.5022ms	0.4256ms	2.3494 KOps/s	2.4539 KOps/s	$\color{#d91a1a}-4.26\%$
test_compile_add_self_flat[pytree-compile]	0.1890ms	0.1386ms	7.2132 KOps/s	7.3150 KOps/s	$\color{#d91a1a}-1.39\%$
test_compile_copy_flat[tensordict-compile]	55.6330μs	19.1315μs	52.2699 KOps/s	44.9782 KOps/s	$\textbf{\color{#35bf28}+16.21\%}$
test_compile_copy_flat[tensordict-eager]	62.9530μs	26.5590μs	37.6520 KOps/s	37.4191 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_copy_flat[pytree-compile]	0.1103ms	70.7766μs	14.1290 KOps/s	14.0805 KOps/s	$\color{#35bf28}+0.34\%$
test_compile_copy_flat[pytree-eager]	0.1335ms	52.6526μs	18.9924 KOps/s	18.6736 KOps/s	$\color{#35bf28}+1.71\%$
test_compile_assign_and_add[tensordict-compile]	1.7103ms	0.4050ms	2.4689 KOps/s	2.1536 KOps/s	$\textbf{\color{#35bf28}+14.64\%}$
test_compile_assign_and_add[tensordict-eager]	3.2085ms	2.7847ms	359.1098 Ops/s	379.6635 Ops/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_compile_assign_and_add[pytree-compile]	1.6041ms	0.4344ms	2.3020 KOps/s	2.2136 KOps/s	$\color{#35bf28}+3.99\%$
test_compile_assign_and_add[pytree-eager]	2.9066ms	2.7828ms	359.3464 Ops/s	376.7644 Ops/s	$\color{#d91a1a}-4.62\%$
test_compile_indexing[tensor-tensordict-compile]	0.1898ms	0.1166ms	8.5765 KOps/s	8.5911 KOps/s	$\color{#d91a1a}-0.17\%$
test_compile_indexing[tensor-tensordict-eager]	0.5550ms	84.5929μs	11.8213 KOps/s	11.9508 KOps/s	$\color{#d91a1a}-1.08\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1720ms	0.1100ms	9.0904 KOps/s	9.1626 KOps/s	$\color{#d91a1a}-0.79\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1384ms	68.4470μs	14.6098 KOps/s	14.1323 KOps/s	$\color{#35bf28}+3.38\%$
test_compile_indexing[tensor-pytree-compile]	0.1657ms	0.1105ms	9.0496 KOps/s	9.0866 KOps/s	$\color{#d91a1a}-0.41\%$
test_compile_indexing[tensor-pytree-eager]	0.1276ms	70.3373μs	14.2172 KOps/s	14.4212 KOps/s	$\color{#d91a1a}-1.41\%$
test_compile_indexing[slice-tensordict-compile]	0.1490ms	0.1022ms	9.7874 KOps/s	9.8270 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_indexing[slice-tensordict-eager]	0.1405ms	18.2020μs	54.9391 KOps/s	53.2948 KOps/s	$\color{#35bf28}+3.09\%$
test_compile_indexing[slice-tensorclass-compile]	0.1763ms	99.1887μs	10.0818 KOps/s	10.2502 KOps/s	$\color{#d91a1a}-1.64\%$
test_compile_indexing[slice-tensorclass-eager]	73.0540μs	16.2486μs	61.5438 KOps/s	62.1299 KOps/s	$\color{#d91a1a}-0.94\%$
test_compile_indexing[slice-pytree-compile]	0.1511ms	98.9207μs	10.1091 KOps/s	10.1611 KOps/s	$\color{#d91a1a}-0.51\%$
test_compile_indexing[slice-pytree-eager]	73.4140μs	16.2499μs	61.5387 KOps/s	62.2900 KOps/s	$\color{#d91a1a}-1.21\%$
test_compile_indexing[int-tensordict-compile]	0.1740ms	0.1033ms	9.6788 KOps/s	9.7569 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_indexing[int-tensordict-eager]	0.5661ms	18.5357μs	53.9499 KOps/s	55.2466 KOps/s	$\color{#d91a1a}-2.35\%$
test_compile_indexing[int-tensorclass-compile]	0.1584ms	0.1015ms	9.8569 KOps/s	9.8489 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_indexing[int-tensorclass-eager]	62.7040μs	16.3695μs	61.0893 KOps/s	62.2642 KOps/s	$\color{#d91a1a}-1.89\%$
test_compile_indexing[int-pytree-compile]	0.1531ms	0.1002ms	9.9829 KOps/s	10.2153 KOps/s	$\color{#d91a1a}-2.28\%$
test_compile_indexing[int-pytree-eager]	0.3895ms	16.2717μs	61.4564 KOps/s	61.9440 KOps/s	$\color{#d91a1a}-0.79\%$
test_mod_add[eager]	92.0150μs	39.9579μs	25.0263 KOps/s	27.4026 KOps/s	$\textbf{\color{#d91a1a}-8.67\%}$
test_mod_add[compile]	0.1353ms	81.0163μs	12.3432 KOps/s	12.3801 KOps/s	$\color{#d91a1a}-0.30\%$
test_mod_add[compile-overhead]	0.3270ms	0.1696ms	5.8976 KOps/s	5.6379 KOps/s	$\color{#35bf28}+4.61\%$
test_mod_wrap[eager]	0.3487ms	0.2642ms	3.7857 KOps/s	3.9397 KOps/s	$\color{#d91a1a}-3.91\%$
test_mod_wrap[compile]	0.3339ms	0.2895ms	3.4538 KOps/s	3.4629 KOps/s	$\color{#d91a1a}-0.26\%$
test_mod_wrap[compile-overhead]	7.1320ms	3.7838ms	264.2826 Ops/s	271.4624 Ops/s	$\color{#d91a1a}-2.64\%$
test_mod_wrap_and_backward[eager]	1.4972ms	1.3841ms	722.4925 Ops/s	684.0040 Ops/s	$\textbf{\color{#35bf28}+5.63\%}$
test_mod_wrap_and_backward[compile]	1.3752ms	1.2849ms	778.3009 Ops/s	779.3076 Ops/s	$\color{#d91a1a}-0.13\%$
test_mod_wrap_and_backward[compile-overhead]	1.3810ms	0.9303ms	1.0749 KOps/s	1.0688 KOps/s	$\color{#35bf28}+0.57\%$
test_seq_add[eager]	0.2541ms	0.1161ms	8.6116 KOps/s	8.8821 KOps/s	$\color{#d91a1a}-3.05\%$
test_seq_add[compile]	0.1387ms	90.3190μs	11.0719 KOps/s	11.1568 KOps/s	$\color{#d91a1a}-0.76\%$
test_seq_add[compile-overhead]	0.2458ms	0.1299ms	7.6953 KOps/s	7.7585 KOps/s	$\color{#d91a1a}-0.82\%$
test_seq_wrap[eager]	0.5046ms	0.4189ms	2.3872 KOps/s	2.3910 KOps/s	$\color{#d91a1a}-0.16\%$
test_seq_wrap[compile]	0.3854ms	0.3055ms	3.2736 KOps/s	3.2692 KOps/s	$\color{#35bf28}+0.13\%$
test_seq_wrap[compile-overhead]	0.2851ms	0.2255ms	4.4339 KOps/s	4.4180 KOps/s	$\color{#35bf28}+0.36\%$
test_func_call_runtime[False-eager]	0.8477ms	0.7611ms	1.3138 KOps/s	1.3154 KOps/s	$\color{#d91a1a}-0.12\%$
test_func_call_runtime[False-compile]	0.8525ms	0.7593ms	1.3170 KOps/s	1.3331 KOps/s	$\color{#d91a1a}-1.20\%$
test_func_call_runtime[False-compile-overhead]	0.4188ms	0.3638ms	2.7484 KOps/s	2.7439 KOps/s	$\color{#35bf28}+0.16\%$
test_func_call_runtime[True-eager]	0.9988ms	0.9324ms	1.0725 KOps/s	1.0817 KOps/s	$\color{#d91a1a}-0.85\%$
test_func_call_runtime[True-compile]	0.8840ms	0.8135ms	1.2292 KOps/s	1.2941 KOps/s	$\textbf{\color{#d91a1a}-5.01\%}$
test_func_call_runtime[True-compile-overhead]	0.4472ms	0.3865ms	2.5873 KOps/s	2.6019 KOps/s	$\color{#d91a1a}-0.56\%$
test_func_call_cm_runtime[False-eager]	0.8440ms	0.7572ms	1.3207 KOps/s	1.3378 KOps/s	$\color{#d91a1a}-1.28\%$
test_func_call_cm_runtime[False-compile]	0.8430ms	0.7607ms	1.3146 KOps/s	1.3272 KOps/s	$\color{#d91a1a}-0.94\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4653ms	0.3649ms	2.7402 KOps/s	2.7170 KOps/s	$\color{#35bf28}+0.86\%$
test_func_call_cm_runtime[True-eager]	1.1309ms	1.0205ms	979.9316 Ops/s	973.9781 Ops/s	$\color{#35bf28}+0.61\%$
test_func_call_cm_runtime[True-compile]	0.9007ms	0.8068ms	1.2394 KOps/s	1.2509 KOps/s	$\color{#d91a1a}-0.92\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4621ms	0.4144ms	2.4134 KOps/s	2.4118 KOps/s	$\color{#35bf28}+0.06\%$
test_vmap_func_call_cm_runtime[eager]	2.5765ms	2.0980ms	476.6488 Ops/s	470.0593 Ops/s	$\color{#35bf28}+1.40\%$
test_vmap_func_call_cm_runtime[compile]	0.8823ms	0.8271ms	1.2090 KOps/s	1.2185 KOps/s	$\color{#d91a1a}-0.78\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4680ms	0.4137ms	2.4173 KOps/s	2.4117 KOps/s	$\color{#35bf28}+0.23\%$
test_distributed	0.6794ms	0.1616ms	6.1893 KOps/s	8.5854 KOps/s	$\textbf{\color{#d91a1a}-27.91\%}$
test_tdmodule	56.4740μs	19.1954μs	52.0957 KOps/s	52.8807 KOps/s	$\color{#d91a1a}-1.48\%$
test_tdmodule_dispatch	79.1040μs	34.5348μs	28.9563 KOps/s	30.2094 KOps/s	$\color{#d91a1a}-4.15\%$
test_tdseq	39.6120μs	18.8888μs	52.9415 KOps/s	54.5982 KOps/s	$\color{#d91a1a}-3.03\%$
test_tdseq_dispatch	64.6140μs	36.5121μs	27.3882 KOps/s	28.5108 KOps/s	$\color{#d91a1a}-3.94\%$
test_instantiation_functorch	1.6559ms	1.5766ms	634.2695 Ops/s	632.7563 Ops/s	$\color{#35bf28}+0.24\%$
test_exec_functorch	0.1919ms	0.1499ms	6.6726 KOps/s	6.9050 KOps/s	$\color{#d91a1a}-3.37\%$
test_exec_functional_call	0.1835ms	0.1435ms	6.9666 KOps/s	7.0240 KOps/s	$\color{#d91a1a}-0.82\%$
test_exec_td_decorator	0.3716ms	0.1903ms	5.2552 KOps/s	5.3262 KOps/s	$\color{#d91a1a}-1.33\%$
test_vmap_mlp_speed_decorator[True-True]	0.7535ms	0.6877ms	1.4541 KOps/s	1.4651 KOps/s	$\color{#d91a1a}-0.76\%$
test_vmap_mlp_speed_decorator[True-False]	0.8793ms	0.6889ms	1.4515 KOps/s	1.4577 KOps/s	$\color{#d91a1a}-0.43\%$
test_vmap_mlp_speed_decorator[False-True]	0.7239ms	0.5971ms	1.6747 KOps/s	1.6679 KOps/s	$\color{#35bf28}+0.41\%$
test_vmap_mlp_speed_decorator[False-False]	0.7663ms	0.5997ms	1.6676 KOps/s	1.6697 KOps/s	$\color{#d91a1a}-0.13\%$
test_vmap_transformer_speed_decorator[True-True]	19.5027ms	19.4554ms	51.3995 Ops/s	51.9704 Ops/s	$\color{#d91a1a}-1.10\%$
test_vmap_transformer_speed_decorator[True-False]	20.6865ms	19.5422ms	51.1713 Ops/s	51.7725 Ops/s	$\color{#d91a1a}-1.16\%$
test_vmap_transformer_speed_decorator[False-True]	19.5136ms	19.4013ms	51.5429 Ops/s	52.2925 Ops/s	$\color{#d91a1a}-1.43\%$
test_vmap_transformer_speed_decorator[False-False]	19.4401ms	19.3600ms	51.6530 Ops/s	52.1389 Ops/s	$\color{#d91a1a}-0.93\%$
test_to_module_speed[True]	1.0225ms	0.9380ms	1.0661 KOps/s	1.0506 KOps/s	$\color{#35bf28}+1.47\%$
test_to_module_speed[False]	1.3697ms	0.9264ms	1.0795 KOps/s	1.0807 KOps/s	$\color{#d91a1a}-0.11\%$
test_tc_init	77.6050μs	36.4012μs	27.4716 KOps/s	28.7906 KOps/s	$\color{#d91a1a}-4.58\%$
test_tc_init_nested	0.1676ms	73.4685μs	13.6113 KOps/s	14.7457 KOps/s	$\textbf{\color{#d91a1a}-7.69\%}$
test_tc_first_layer_tensor	4.6374μs	0.7054μs	1.4176 MOps/s	1.4540 MOps/s	$\color{#d91a1a}-2.51\%$
test_tc_first_layer_nontensor	25.7410μs	2.3561μs	424.4268 KOps/s	429.4530 KOps/s	$\color{#d91a1a}-1.17\%$
test_tc_second_layer_tensor	21.9263μs	1.4049μs	711.7952 KOps/s	695.6890 KOps/s	$\color{#35bf28}+2.32\%$
test_tc_second_layer_nontensor	42.3420μs	3.0551μs	327.3198 KOps/s	327.2140 KOps/s	$\color{#35bf28}+0.03\%$
test_unbind	0.2231s	9.8266ms	101.7647 Ops/s	150.9935 Ops/s	$\textbf{\color{#d91a1a}-32.60\%}$
test_full_like	10.1091ms	9.3841ms	106.5628 Ops/s	99.7438 Ops/s	$\textbf{\color{#35bf28}+6.84\%}$
test_zeros_like	4.9008ms	4.3525ms	229.7504 Ops/s	230.0126 Ops/s	$\color{#d91a1a}-0.11\%$
test_ones_like	4.6441ms	4.3849ms	228.0568 Ops/s	226.9086 Ops/s	$\color{#35bf28}+0.51\%$
test_clone	7.3753ms	6.7802ms	147.4874 Ops/s	145.0582 Ops/s	$\color{#35bf28}+1.67\%$
test_squeeze	56.5140μs	9.5221μs	105.0187 KOps/s	107.2510 KOps/s	$\color{#d91a1a}-2.08\%$
test_unsqueeze	0.1275ms	71.7409μs	13.9391 KOps/s	13.2676 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_split	0.3740ms	0.1651ms	6.0563 KOps/s	6.1429 KOps/s	$\color{#d91a1a}-1.41\%$
test_permute	0.2316ms	0.1788ms	5.5913 KOps/s	5.6027 KOps/s	$\color{#d91a1a}-0.20\%$
test_stack	51.8542ms	51.5196ms	19.4101 Ops/s	19.0091 Ops/s	$\color{#35bf28}+2.11\%$
test_cat	51.8796ms	51.3111ms	19.4889 Ops/s	18.6443 Ops/s	$\color{#35bf28}+4.53\%$