[Feature] TensorDict.clamp

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] broadcast pointwise ops for tensor/tensordict mixed inputs #1166
• -> [Feature] TensorDict.clamp #1165
• [BugFix] Fix unitary ops for tensorclass #1164
• [Feature] TensorDict.softmax #1163
• [Feature] TensorDict.logsumexp #1162

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}22$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	54.8520μs	22.6376μs	44.1742 KOps/s	45.9021 KOps/s	$\color{#d91a1a}-3.76\%$
test_plain_set_stack_nested	50.7150μs	23.0366μs	43.4092 KOps/s	45.0333 KOps/s	$\color{#d91a1a}-3.61\%$
test_plain_set_nested_inplace	73.4060μs	24.5438μs	40.7435 KOps/s	41.0886 KOps/s	$\color{#d91a1a}-0.84\%$
test_plain_set_stack_nested_inplace	70.6220μs	24.5562μs	40.7229 KOps/s	41.1509 KOps/s	$\color{#d91a1a}-1.04\%$
test_items	39.4530μs	4.1725μs	239.6636 KOps/s	237.6285 KOps/s	$\color{#35bf28}+0.86\%$
test_items_nested	0.7233ms	0.4045ms	2.4720 KOps/s	2.4687 KOps/s	$\color{#35bf28}+0.13\%$
test_items_nested_locked	0.5849ms	0.4030ms	2.4811 KOps/s	2.4735 KOps/s	$\color{#35bf28}+0.31\%$
test_items_nested_leaf	0.1353ms	77.8907μs	12.8385 KOps/s	12.9253 KOps/s	$\color{#d91a1a}-0.67\%$
test_items_stack_nested	0.7406ms	0.4066ms	2.4596 KOps/s	2.4551 KOps/s	$\color{#35bf28}+0.18\%$
test_items_stack_nested_leaf	0.1384ms	80.2497μs	12.4611 KOps/s	12.4964 KOps/s	$\color{#d91a1a}-0.28\%$
test_items_stack_nested_locked	0.6059ms	0.4029ms	2.4821 KOps/s	2.4283 KOps/s	$\color{#35bf28}+2.22\%$
test_keys	47.6990μs	3.4846μs	286.9778 KOps/s	282.5891 KOps/s	$\color{#35bf28}+1.55\%$
test_keys_nested	0.2550ms	0.1691ms	5.9149 KOps/s	5.8073 KOps/s	$\color{#35bf28}+1.85\%$
test_keys_nested_locked	1.7928ms	0.1795ms	5.5697 KOps/s	5.6335 KOps/s	$\color{#d91a1a}-1.13\%$
test_keys_nested_leaf	0.2355ms	0.1485ms	6.7320 KOps/s	6.6681 KOps/s	$\color{#35bf28}+0.96\%$
test_keys_stack_nested	0.3331ms	0.1675ms	5.9715 KOps/s	5.9650 KOps/s	$\color{#35bf28}+0.11\%$
test_keys_stack_nested_leaf	0.5040ms	0.1438ms	6.9561 KOps/s	6.9300 KOps/s	$\color{#35bf28}+0.38\%$
test_keys_stack_nested_locked	0.3705ms	0.1718ms	5.8204 KOps/s	5.7393 KOps/s	$\color{#35bf28}+1.41\%$
test_values	10.2872μs	1.0729μs	932.0876 KOps/s	969.6323 KOps/s	$\color{#d91a1a}-3.87\%$
test_values_nested	0.1223ms	66.9701μs	14.9320 KOps/s	15.3467 KOps/s	$\color{#d91a1a}-2.70\%$
test_values_nested_locked	0.1196ms	66.8310μs	14.9631 KOps/s	15.3422 KOps/s	$\color{#d91a1a}-2.47\%$
test_values_nested_leaf	0.1254ms	75.4501μs	13.2538 KOps/s	13.2948 KOps/s	$\color{#d91a1a}-0.31\%$
test_values_stack_nested	0.1233ms	67.0847μs	14.9065 KOps/s	14.2282 KOps/s	$\color{#35bf28}+4.77\%$
test_values_stack_nested_leaf	0.1346ms	75.9345μs	13.1693 KOps/s	13.4636 KOps/s	$\color{#d91a1a}-2.19\%$
test_values_stack_nested_locked	0.1306ms	67.1668μs	14.8883 KOps/s	15.2652 KOps/s	$\color{#d91a1a}-2.47\%$
test_membership	30.3370μs	0.8602μs	1.1626 MOps/s	1.1341 MOps/s	$\color{#35bf28}+2.52\%$
test_membership_nested	26.5190μs	2.9636μs	337.4267 KOps/s	333.1692 KOps/s	$\color{#35bf28}+1.28\%$
test_membership_nested_leaf	48.0090μs	2.9815μs	335.4041 KOps/s	332.1110 KOps/s	$\color{#35bf28}+0.99\%$
test_membership_stacked_nested	26.2690μs	2.9591μs	337.9406 KOps/s	332.9517 KOps/s	$\color{#35bf28}+1.50\%$
test_membership_stacked_nested_leaf	46.6780μs	2.9362μs	340.5747 KOps/s	328.4625 KOps/s	$\color{#35bf28}+3.69\%$
test_membership_nested_last	19.4260μs	4.5173μs	221.3732 KOps/s	219.7243 KOps/s	$\color{#35bf28}+0.75\%$
test_membership_nested_leaf_last	53.5100μs	4.5266μs	220.9178 KOps/s	216.5874 KOps/s	$\color{#35bf28}+2.00\%$
test_membership_stacked_nested_last	43.0300μs	13.8436μs	72.2355 KOps/s	71.6465 KOps/s	$\color{#35bf28}+0.82\%$
test_membership_stacked_nested_leaf_last	40.3250μs	13.7592μs	72.6788 KOps/s	72.2087 KOps/s	$\color{#35bf28}+0.65\%$
test_nested_getleaf	63.6490μs	10.8600μs	92.0812 KOps/s	92.3742 KOps/s	$\color{#d91a1a}-0.32\%$
test_nested_get	59.3810μs	10.2596μs	97.4697 KOps/s	97.6586 KOps/s	$\color{#d91a1a}-0.19\%$
test_stacked_getleaf	38.7520μs	10.7212μs	93.2727 KOps/s	92.5403 KOps/s	$\color{#35bf28}+0.79\%$
test_stacked_get	56.4550μs	10.2912μs	97.1702 KOps/s	97.9260 KOps/s	$\color{#d91a1a}-0.77\%$
test_nested_getitemleaf	48.3000μs	11.3819μs	87.8586 KOps/s	88.2396 KOps/s	$\color{#d91a1a}-0.43\%$
test_nested_getitem	59.7420μs	10.5682μs	94.6237 KOps/s	95.0260 KOps/s	$\color{#d91a1a}-0.42\%$
test_stacked_getitemleaf	37.4990μs	11.2283μs	89.0606 KOps/s	88.8484 KOps/s	$\color{#35bf28}+0.24\%$
test_stacked_getitem	58.0280μs	10.3767μs	96.3700 KOps/s	96.8315 KOps/s	$\color{#d91a1a}-0.48\%$
test_lock_nested	2.0126ms	0.4616ms	2.1662 KOps/s	1.7569 KOps/s	$\textbf{\color{#35bf28}+23.30\%}$
test_lock_stack_nested	0.6975ms	0.4211ms	2.3745 KOps/s	2.3787 KOps/s	$\color{#d91a1a}-0.18\%$
test_unlock_nested	0.8459ms	0.3760ms	2.6594 KOps/s	2.6348 KOps/s	$\color{#35bf28}+0.94\%$
test_unlock_stack_nested	0.6408ms	0.3365ms	2.9719 KOps/s	2.9618 KOps/s	$\color{#35bf28}+0.34\%$
test_flatten_speed	0.1886ms	0.1014ms	9.8582 KOps/s	9.8511 KOps/s	$\color{#35bf28}+0.07\%$
test_unflatten_speed	0.6902ms	0.5318ms	1.8805 KOps/s	1.8502 KOps/s	$\color{#35bf28}+1.64\%$
test_common_ops	1.5699ms	0.8126ms	1.2306 KOps/s	1.2787 KOps/s	$\color{#d91a1a}-3.77\%$
test_creation	32.3300μs	2.5550μs	391.3920 KOps/s	400.9198 KOps/s	$\color{#d91a1a}-2.38\%$
test_creation_empty	40.9060μs	12.7549μs	78.4011 KOps/s	90.5213 KOps/s	$\textbf{\color{#d91a1a}-13.39\%}$
test_creation_nested_1	63.2880μs	16.1093μs	62.0760 KOps/s	71.1110 KOps/s	$\textbf{\color{#d91a1a}-12.71\%}$
test_creation_nested_2	49.9640μs	21.1762μs	47.2228 KOps/s	52.6206 KOps/s	$\textbf{\color{#d91a1a}-10.26\%}$
test_clone	0.1010ms	13.4895μs	74.1318 KOps/s	74.2176 KOps/s	$\color{#d91a1a}-0.12\%$
test_getitem[int]	1.0989ms	13.0353μs	76.7148 KOps/s	76.8122 KOps/s	$\color{#d91a1a}-0.13\%$
test_getitem[slice_int]	0.1422ms	24.6834μs	40.5131 KOps/s	40.6257 KOps/s	$\color{#d91a1a}-0.28\%$
test_getitem[range]	0.1696ms	48.3337μs	20.6895 KOps/s	19.9322 KOps/s	$\color{#35bf28}+3.80\%$
test_getitem[tuple]	0.1399ms	20.6268μs	48.4807 KOps/s	50.0095 KOps/s	$\color{#d91a1a}-3.06\%$
test_getitem[list]	0.1911ms	43.7878μs	22.8374 KOps/s	22.6488 KOps/s	$\color{#35bf28}+0.83\%$
test_setitem_dim[int]	43.3610μs	25.2059μs	39.6732 KOps/s	39.5766 KOps/s	$\color{#35bf28}+0.24\%$
test_setitem_dim[slice_int]	78.9270μs	50.9952μs	19.6097 KOps/s	19.5911 KOps/s	$\color{#35bf28}+0.09\%$
test_setitem_dim[range]	0.1026ms	72.2654μs	13.8379 KOps/s	13.5369 KOps/s	$\color{#35bf28}+2.22\%$
test_setitem_dim[tuple]	98.9750μs	42.7163μs	23.4103 KOps/s	24.5872 KOps/s	$\color{#d91a1a}-4.79\%$
test_setitem	0.3827ms	21.6429μs	46.2046 KOps/s	49.9020 KOps/s	$\textbf{\color{#d91a1a}-7.41\%}$
test_set	96.5700μs	21.1053μs	47.3814 KOps/s	51.0842 KOps/s	$\textbf{\color{#d91a1a}-7.25\%}$
test_set_shared	4.2304ms	0.1727ms	5.7896 KOps/s	5.9137 KOps/s	$\color{#d91a1a}-2.10\%$
test_update	0.2169ms	24.2084μs	41.3080 KOps/s	44.0005 KOps/s	$\textbf{\color{#d91a1a}-6.12\%}$
test_update_nested	0.1073ms	35.4992μs	28.1697 KOps/s	30.4417 KOps/s	$\textbf{\color{#d91a1a}-7.46\%}$
test_update__nested	1.0745ms	34.4493μs	29.0282 KOps/s	29.7765 KOps/s	$\color{#d91a1a}-2.51\%$
test_set_nested	0.3200ms	23.1418μs	43.2118 KOps/s	45.5489 KOps/s	$\textbf{\color{#d91a1a}-5.13\%}$
test_set_nested_new	85.9610μs	27.6008μs	36.2309 KOps/s	37.6710 KOps/s	$\color{#d91a1a}-3.82\%$
test_select	0.3328ms	44.1061μs	22.6726 KOps/s	23.3261 KOps/s	$\color{#d91a1a}-2.80\%$
test_select_nested	0.1283ms	63.8051μs	15.6727 KOps/s	15.6102 KOps/s	$\color{#35bf28}+0.40\%$
test_exclude_nested	0.3463ms	83.2085μs	12.0180 KOps/s	12.1365 KOps/s	$\color{#d91a1a}-0.98\%$
test_empty[True]	0.7439ms	0.4163ms	2.4019 KOps/s	2.4162 KOps/s	$\color{#d91a1a}-0.59\%$
test_empty[False]	11.0380μs	1.4351μs	696.8110 KOps/s	701.8006 KOps/s	$\color{#d91a1a}-0.71\%$
test_unbind_speed	0.3617ms	0.2694ms	3.7119 KOps/s	3.6245 KOps/s	$\color{#35bf28}+2.41\%$
test_unbind_speed_stack0	0.4865ms	0.2612ms	3.8279 KOps/s	3.8275 KOps/s	$\color{#35bf28}+0.01\%$
test_unbind_speed_stack1	0.1199s	0.8060ms	1.2407 KOps/s	1.3726 KOps/s	$\textbf{\color{#d91a1a}-9.61\%}$
test_split	2.5180ms	1.6100ms	621.1226 Ops/s	565.7713 Ops/s	$\textbf{\color{#35bf28}+9.78\%}$
test_chunk	0.1230s	2.0165ms	495.9179 Ops/s	563.6976 Ops/s	$\textbf{\color{#d91a1a}-12.02\%}$
test_consolidate_njt[False-None]	10.2302ms	8.3290ms	120.0619 Ops/s	124.9070 Ops/s	$\color{#d91a1a}-3.88\%$
test_creation[device0]	5.0040ms	94.1620μs	10.6200 KOps/s	10.9339 KOps/s	$\color{#d91a1a}-2.87\%$
test_creation_from_tensor	0.2812ms	93.7184μs	10.6703 KOps/s	10.4540 KOps/s	$\color{#35bf28}+2.07\%$
test_add_one[memmap_tensor0]	0.1127ms	4.6963μs	212.9324 KOps/s	205.0923 KOps/s	$\color{#35bf28}+3.82\%$
test_contiguous[memmap_tensor0]	11.9830μs	0.5276μs	1.8952 MOps/s	1.9617 MOps/s	$\color{#d91a1a}-3.39\%$
test_stack[memmap_tensor0]	38.3210μs	3.3374μs	299.6364 KOps/s	288.9800 KOps/s	$\color{#35bf28}+3.69\%$
test_memmaptd_index	1.0492ms	0.2431ms	4.1132 KOps/s	4.2178 KOps/s	$\color{#d91a1a}-2.48\%$
test_memmaptd_index_astensor	0.6360ms	0.3309ms	3.0220 KOps/s	3.0505 KOps/s	$\color{#d91a1a}-0.93\%$
test_memmaptd_index_op	1.0498ms	0.6115ms	1.6353 KOps/s	1.7009 KOps/s	$\color{#d91a1a}-3.86\%$
test_serialize_model	0.1257s	0.1185s	8.4354 Ops/s	8.2753 Ops/s	$\color{#35bf28}+1.93\%$
test_serialize_model_pickle	0.5000s	0.3990s	2.5064 Ops/s	2.5863 Ops/s	$\color{#d91a1a}-3.09\%$
test_serialize_weights	0.1238s	0.1155s	8.6552 Ops/s	7.5740 Ops/s	$\textbf{\color{#35bf28}+14.27\%}$
test_serialize_weights_returnearly	0.2576s	0.1814s	5.5112 Ops/s	6.2349 Ops/s	$\textbf{\color{#d91a1a}-11.61\%}$
test_serialize_weights_pickle	0.4608s	0.4103s	2.4375 Ops/s	2.4533 Ops/s	$\color{#d91a1a}-0.64\%$
test_serialize_weights_filesystem	0.1565s	0.1459s	6.8544 Ops/s	6.9204 Ops/s	$\color{#d91a1a}-0.95\%$
test_serialize_model_filesystem	0.1547s	0.1516s	6.5980 Ops/s	5.9262 Ops/s	$\textbf{\color{#35bf28}+11.34\%}$
test_reshape_pytree	80.0600μs	26.8810μs	37.2009 KOps/s	36.2051 KOps/s	$\color{#35bf28}+2.75\%$
test_reshape_td	70.0110μs	33.2153μs	30.1067 KOps/s	28.5437 KOps/s	$\textbf{\color{#35bf28}+5.48\%}$
test_view_pytree	81.1410μs	26.9371μs	37.1236 KOps/s	37.3162 KOps/s	$\color{#d91a1a}-0.52\%$
test_view_td	94.7770μs	38.9042μs	25.7042 KOps/s	25.5433 KOps/s	$\color{#35bf28}+0.63\%$
test_unbind_pytree	69.5500μs	29.8166μs	33.5384 KOps/s	33.2958 KOps/s	$\color{#35bf28}+0.73\%$
test_unbind_td	0.3321ms	39.9589μs	25.0257 KOps/s	25.0127 KOps/s	$\color{#35bf28}+0.05\%$
test_split_pytree	70.1210μs	29.6357μs	33.7431 KOps/s	33.5612 KOps/s	$\color{#35bf28}+0.54\%$
test_split_td	0.5586ms	45.6737μs	21.8945 KOps/s	22.2475 KOps/s	$\color{#d91a1a}-1.59\%$
test_add_pytree	88.3250μs	35.5839μs	28.1026 KOps/s	27.5323 KOps/s	$\color{#35bf28}+2.07\%$
test_add_td	0.1196ms	60.0852μs	16.6430 KOps/s	17.5472 KOps/s	$\textbf{\color{#d91a1a}-5.15\%}$
test_compile_add_one_nested[tensordict-compile]	0.1426ms	64.2531μs	15.5635 KOps/s	15.7214 KOps/s	$\color{#d91a1a}-1.00\%$
test_compile_add_one_nested[tensordict-eager]	0.4368ms	0.1744ms	5.7339 KOps/s	5.7167 KOps/s	$\color{#35bf28}+0.30\%$
test_compile_add_one_nested[pytree-compile]	0.1434ms	46.7603μs	21.3857 KOps/s	21.9558 KOps/s	$\color{#d91a1a}-2.60\%$
test_compile_add_one_nested[pytree-eager]	0.2781ms	0.1204ms	8.3034 KOps/s	8.4614 KOps/s	$\color{#d91a1a}-1.87\%$
test_compile_copy_nested[tensordict-compile]	64.2200μs	27.2907μs	36.6425 KOps/s	39.2591 KOps/s	$\textbf{\color{#d91a1a}-6.66\%}$
test_compile_copy_nested[tensordict-eager]	0.1356ms	59.5854μs	16.7826 KOps/s	16.8004 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_copy_nested[pytree-compile]	0.1626ms	80.9350μs	12.3556 KOps/s	12.5812 KOps/s	$\color{#d91a1a}-1.79\%$
test_compile_copy_nested[pytree-eager]	0.1417ms	68.4643μs	14.6061 KOps/s	14.6018 KOps/s	$\color{#35bf28}+0.03\%$
test_compile_add_one_flat[tensordict-compile]	0.2183ms	0.1046ms	9.5618 KOps/s	9.4235 KOps/s	$\color{#35bf28}+1.47\%$
test_compile_add_one_flat[tensordict-eager]	1.5740ms	0.2173ms	4.6012 KOps/s	4.5810 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_one_flat[tensorclass-compile]	96.5500μs	44.1732μs	22.6382 KOps/s	22.3041 KOps/s	$\color{#35bf28}+1.50\%$
test_compile_add_one_flat[tensorclass-eager]	1.2644ms	64.2104μs	15.5738 KOps/s	14.9467 KOps/s	$\color{#35bf28}+4.20\%$
test_compile_add_one_flat[pytree-compile]	0.2204ms	0.1036ms	9.6523 KOps/s	9.6658 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_add_one_flat[pytree-eager]	0.3444ms	0.2027ms	4.9345 KOps/s	4.9722 KOps/s	$\color{#d91a1a}-0.76\%$
test_compile_add_self_flat[tensordict-eager]	0.3473ms	0.2344ms	4.2662 KOps/s	4.2749 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_add_self_flat[tensordict-compile]	0.2026ms	0.1056ms	9.4700 KOps/s	9.4296 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_add_self_flat[tensorclass-eager]	0.1286ms	60.3663μs	16.5655 KOps/s	16.7302 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_add_self_flat[tensorclass-compile]	0.7890ms	46.8321μs	21.3529 KOps/s	21.4853 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_add_self_flat[pytree-eager]	0.6518ms	0.1602ms	6.2414 KOps/s	6.2867 KOps/s	$\color{#d91a1a}-0.72\%$
test_compile_add_self_flat[pytree-compile]	0.2003ms	0.1036ms	9.6527 KOps/s	9.5018 KOps/s	$\color{#35bf28}+1.59\%$
test_compile_copy_flat[tensordict-compile]	83.0050μs	20.8815μs	47.8893 KOps/s	45.1711 KOps/s	$\textbf{\color{#35bf28}+6.02\%}$
test_compile_copy_flat[tensordict-eager]	0.1393ms	65.8479μs	15.1865 KOps/s	14.9804 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_copy_flat[pytree-compile]	0.1624ms	82.8753μs	12.0663 KOps/s	12.0121 KOps/s	$\color{#35bf28}+0.45\%$
test_compile_copy_flat[pytree-eager]	0.1369ms	67.8303μs	14.7427 KOps/s	14.4171 KOps/s	$\color{#35bf28}+2.26\%$
test_compile_assign_and_add[tensordict-compile]	0.3842ms	0.2062ms	4.8496 KOps/s	4.6882 KOps/s	$\color{#35bf28}+3.44\%$
test_compile_assign_and_add[tensordict-eager]	1.5497ms	1.3318ms	750.8845 Ops/s	747.1302 Ops/s	$\color{#35bf28}+0.50\%$
test_compile_assign_and_add[pytree-compile]	0.3930ms	0.2031ms	4.9238 KOps/s	4.7124 KOps/s	$\color{#35bf28}+4.49\%$
test_compile_assign_and_add[pytree-eager]	1.3456ms	0.7702ms	1.2984 KOps/s	1.2297 KOps/s	$\textbf{\color{#35bf28}+5.59\%}$
test_compile_assign_and_add_stack[compile]	0.8202ms	0.4543ms	2.2014 KOps/s	2.1259 KOps/s	$\color{#35bf28}+3.55\%$
test_compile_assign_and_add_stack[eager]	4.3499ms	2.7762ms	360.2080 Ops/s	371.1989 Ops/s	$\color{#d91a1a}-2.96\%$
test_compile_indexing[tensor-tensordict-compile]	0.1035ms	35.8966μs	27.8578 KOps/s	27.3310 KOps/s	$\color{#35bf28}+1.93\%$
test_compile_indexing[tensor-tensordict-eager]	0.5990ms	34.3970μs	29.0723 KOps/s	29.8729 KOps/s	$\color{#d91a1a}-2.68\%$
test_compile_indexing[tensor-tensorclass-compile]	91.8020μs	29.4911μs	33.9085 KOps/s	34.9022 KOps/s	$\color{#d91a1a}-2.85\%$
test_compile_indexing[tensor-tensorclass-eager]	87.2230μs	23.0849μs	43.3184 KOps/s	42.9632 KOps/s	$\color{#35bf28}+0.83\%$
test_compile_indexing[tensor-pytree-compile]	0.1191ms	30.6420μs	32.6349 KOps/s	33.2025 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_indexing[tensor-pytree-eager]	0.8271ms	23.1135μs	43.2649 KOps/s	42.3135 KOps/s	$\color{#35bf28}+2.25\%$
test_compile_indexing[slice-tensordict-compile]	0.1064ms	51.4405μs	19.4399 KOps/s	19.3058 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_indexing[slice-tensordict-eager]	0.7257ms	20.1508μs	49.6258 KOps/s	49.3500 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_indexing[slice-tensorclass-compile]	0.1230ms	45.1278μs	22.1593 KOps/s	22.7403 KOps/s	$\color{#d91a1a}-2.56\%$
test_compile_indexing[slice-tensorclass-eager]	0.1185ms	18.8713μs	52.9905 KOps/s	52.4299 KOps/s	$\color{#35bf28}+1.07\%$
test_compile_indexing[slice-pytree-compile]	0.1160ms	44.6978μs	22.3725 KOps/s	22.5597 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_indexing[slice-pytree-eager]	80.4100μs	18.9445μs	52.7857 KOps/s	51.8014 KOps/s	$\color{#35bf28}+1.90\%$
test_compile_indexing[int-tensordict-compile]	0.1180ms	52.7969μs	18.9405 KOps/s	19.1936 KOps/s	$\color{#d91a1a}-1.32\%$
test_compile_indexing[int-tensordict-eager]	1.0571ms	19.9281μs	50.1804 KOps/s	49.3842 KOps/s	$\color{#35bf28}+1.61\%$
test_compile_indexing[int-tensorclass-compile]	0.1119ms	44.9126μs	22.2655 KOps/s	22.3279 KOps/s	$\color{#d91a1a}-0.28\%$
test_compile_indexing[int-tensorclass-eager]	66.9950μs	18.6937μs	53.4939 KOps/s	52.5830 KOps/s	$\color{#35bf28}+1.73\%$
test_compile_indexing[int-pytree-compile]	0.1107ms	44.8437μs	22.2997 KOps/s	22.5322 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_indexing[int-pytree-eager]	77.6140μs	19.0120μs	52.5984 KOps/s	52.8526 KOps/s	$\color{#d91a1a}-0.48\%$
test_mod_add[eager]	0.1727ms	36.9092μs	27.0935 KOps/s	29.1360 KOps/s	$\textbf{\color{#d91a1a}-7.01\%}$
test_mod_add[compile]	0.1326ms	50.5984μs	19.7635 KOps/s	21.0480 KOps/s	$\textbf{\color{#d91a1a}-6.10\%}$
test_mod_add[compile-overhead]	0.1181ms	49.1353μs	20.3520 KOps/s	20.8824 KOps/s	$\color{#d91a1a}-2.54\%$
test_mod_wrap[eager]	0.3957ms	0.2309ms	4.3315 KOps/s	4.5016 KOps/s	$\color{#d91a1a}-3.78\%$
test_mod_wrap[compile]	0.3830ms	0.2092ms	4.7794 KOps/s	4.7927 KOps/s	$\color{#d91a1a}-0.28\%$
test_mod_wrap[compile-overhead]	0.3500ms	0.2071ms	4.8287 KOps/s	4.7908 KOps/s	$\color{#35bf28}+0.79\%$
test_mod_wrap_and_backward[eager]	22.6538ms	12.6512ms	79.0441 Ops/s	75.3084 Ops/s	$\color{#35bf28}+4.96\%$
test_mod_wrap_and_backward[compile]	18.6268ms	13.1479ms	76.0580 Ops/s	70.4928 Ops/s	$\textbf{\color{#35bf28}+7.89\%}$
test_mod_wrap_and_backward[compile-overhead]	14.0521ms	12.0034ms	83.3095 Ops/s	70.6661 Ops/s	$\textbf{\color{#35bf28}+17.89\%}$
test_seq_add[eager]	0.2198ms	0.1221ms	8.1926 KOps/s	8.7553 KOps/s	$\textbf{\color{#d91a1a}-6.43\%}$
test_seq_add[compile]	0.1623ms	63.4657μs	15.7565 KOps/s	16.2264 KOps/s	$\color{#d91a1a}-2.90\%$
test_seq_add[compile-overhead]	0.4616ms	63.4982μs	15.7485 KOps/s	16.5376 KOps/s	$\color{#d91a1a}-4.77\%$
test_seq_wrap[eager]	0.6745ms	0.4632ms	2.1588 KOps/s	2.0361 KOps/s	$\textbf{\color{#35bf28}+6.03\%}$
test_seq_wrap[compile]	0.4284ms	0.2337ms	4.2784 KOps/s	4.2606 KOps/s	$\color{#35bf28}+0.42\%$
test_seq_wrap[compile-overhead]	0.3333ms	0.2302ms	4.3444 KOps/s	4.3335 KOps/s	$\color{#35bf28}+0.25\%$
test_func_call_runtime[False-eager]	0.7831ms	0.5502ms	1.8176 KOps/s	1.8550 KOps/s	$\color{#d91a1a}-2.01\%$
test_func_call_runtime[False-compile]	0.6202ms	0.4326ms	2.3118 KOps/s	2.3250 KOps/s	$\color{#d91a1a}-0.57\%$
test_func_call_runtime[False-compile-overhead]	0.5808ms	0.4362ms	2.2924 KOps/s	2.3216 KOps/s	$\color{#d91a1a}-1.26\%$
test_func_call_runtime[True-eager]	1.0796ms	0.7788ms	1.2840 KOps/s	1.3163 KOps/s	$\color{#d91a1a}-2.46\%$
test_func_call_runtime[True-compile]	0.6517ms	0.4737ms	2.1110 KOps/s	2.1267 KOps/s	$\color{#d91a1a}-0.73\%$
test_func_call_runtime[True-compile-overhead]	0.6506ms	0.4838ms	2.0672 KOps/s	2.1463 KOps/s	$\color{#d91a1a}-3.69\%$
test_func_call_cm_runtime[False-eager]	1.1967ms	0.5697ms	1.7554 KOps/s	1.8189 KOps/s	$\color{#d91a1a}-3.49\%$
test_func_call_cm_runtime[False-compile]	0.5896ms	0.4343ms	2.3023 KOps/s	2.3186 KOps/s	$\color{#d91a1a}-0.70\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6089ms	0.4340ms	2.3041 KOps/s	2.3303 KOps/s	$\color{#d91a1a}-1.12\%$
test_func_call_cm_runtime[True-eager]	1.3316ms	0.9342ms	1.0704 KOps/s	1.1057 KOps/s	$\color{#d91a1a}-3.19\%$
test_func_call_cm_runtime[True-compile]	0.6731ms	0.4987ms	2.0050 KOps/s	2.0126 KOps/s	$\color{#d91a1a}-0.38\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6317ms	0.4989ms	2.0045 KOps/s	2.0065 KOps/s	$\color{#d91a1a}-0.10\%$
test_vmap_func_call_cm_runtime[eager]	3.6250ms	1.9952ms	501.2071 Ops/s	513.9526 Ops/s	$\color{#d91a1a}-2.48\%$
test_vmap_func_call_cm_runtime[compile]	0.7885ms	0.5300ms	1.8867 KOps/s	1.8766 KOps/s	$\color{#35bf28}+0.54\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6606ms	0.5324ms	1.8784 KOps/s	1.8944 KOps/s	$\color{#d91a1a}-0.84\%$
test_distributed	0.2885ms	0.1272ms	7.8636 KOps/s	7.6279 KOps/s	$\color{#35bf28}+3.09\%$
test_tdmodule	55.7940μs	28.5766μs	34.9937 KOps/s	39.7837 KOps/s	$\textbf{\color{#d91a1a}-12.04\%}$
test_tdmodule_dispatch	96.5700μs	51.4664μs	19.4302 KOps/s	21.5395 KOps/s	$\textbf{\color{#d91a1a}-9.79\%}$
test_tdseq	64.1900μs	31.5498μs	31.6959 KOps/s	35.7469 KOps/s	$\textbf{\color{#d91a1a}-11.33\%}$
test_tdseq_dispatch	99.6860μs	58.4497μs	17.1087 KOps/s	19.0807 KOps/s	$\textbf{\color{#d91a1a}-10.33\%}$
test_instantiation_functorch	2.4137ms	1.6015ms	624.4226 Ops/s	656.1869 Ops/s	$\color{#d91a1a}-4.84\%$
test_exec_functorch	0.3472ms	0.1824ms	5.4838 KOps/s	5.4396 KOps/s	$\color{#35bf28}+0.81\%$
test_exec_functional_call	0.3419ms	0.1782ms	5.6102 KOps/s	5.8424 KOps/s	$\color{#d91a1a}-3.97\%$
test_exec_td_decorator	0.6808ms	0.2408ms	4.1525 KOps/s	4.2329 KOps/s	$\color{#d91a1a}-1.90\%$
test_vmap_mlp_speed_decorator[True-True]	0.9020ms	0.6713ms	1.4896 KOps/s	1.5285 KOps/s	$\color{#d91a1a}-2.55\%$
test_vmap_mlp_speed_decorator[True-False]	0.9341ms	0.6695ms	1.4935 KOps/s	1.5259 KOps/s	$\color{#d91a1a}-2.12\%$
test_vmap_mlp_speed_decorator[False-True]	0.8297ms	0.5348ms	1.8699 KOps/s	1.9032 KOps/s	$\color{#d91a1a}-1.75\%$
test_vmap_mlp_speed_decorator[False-False]	0.7837ms	0.5313ms	1.8820 KOps/s	1.8883 KOps/s	$\color{#d91a1a}-0.33\%$
test_to_module_speed[True]	2.2334ms	1.3782ms	725.5985 Ops/s	736.4505 Ops/s	$\color{#d91a1a}-1.47\%$
test_to_module_speed[False]	2.2061ms	1.3509ms	740.2588 Ops/s	745.8348 Ops/s	$\color{#d91a1a}-0.75\%$
test_tc_init	0.1066ms	48.9274μs	20.4385 KOps/s	20.7925 KOps/s	$\color{#d91a1a}-1.70\%$
test_tc_init_nested	0.1981ms	98.0112μs	10.2029 KOps/s	10.3746 KOps/s	$\color{#d91a1a}-1.66\%$
test_tc_first_layer_tensor	30.4470μs	1.5683μs	637.6427 KOps/s	629.2118 KOps/s	$\color{#35bf28}+1.34\%$
test_tc_first_layer_nontensor	35.9670μs	4.8312μs	206.9882 KOps/s	205.8341 KOps/s	$\color{#35bf28}+0.56\%$
test_tc_second_layer_tensor	41.0770μs	2.9252μs	341.8550 KOps/s	337.3694 KOps/s	$\color{#35bf28}+1.33\%$
test_tc_second_layer_nontensor	38.6730μs	6.1983μs	161.3351 KOps/s	159.1127 KOps/s	$\color{#35bf28}+1.40\%$
test_unbind	0.2636s	16.1920ms	61.7590 Ops/s	75.0737 Ops/s	$\textbf{\color{#d91a1a}-17.74\%}$
test_full_like	12.9785ms	10.8933ms	91.7992 Ops/s	72.0770 Ops/s	$\textbf{\color{#35bf28}+27.36\%}$
test_zeros_like	4.2893ms	3.8999ms	256.4169 Ops/s	120.4092 Ops/s	$\textbf{\color{#35bf28}+112.95\%}$
test_ones_like	4.9273ms	4.5737ms	218.6418 Ops/s	122.5138 Ops/s	$\textbf{\color{#35bf28}+78.46\%}$
test_clone	7.2744ms	6.7138ms	148.9475 Ops/s	102.2662 Ops/s	$\textbf{\color{#35bf28}+45.65\%}$
test_squeeze	77.3640μs	12.4782μs	80.1400 KOps/s	81.0021 KOps/s	$\color{#d91a1a}-1.06\%$
test_unsqueeze	0.2093ms	97.8370μs	10.2211 KOps/s	10.4457 KOps/s	$\color{#d91a1a}-2.15\%$
test_split	0.5434ms	0.2007ms	4.9830 KOps/s	5.0505 KOps/s	$\color{#d91a1a}-1.34\%$
test_permute	0.4346ms	0.2116ms	4.7265 KOps/s	4.7305 KOps/s	$\color{#d91a1a}-0.08\%$
test_stack	34.9112ms	28.4459ms	35.1544 Ops/s	37.1514 Ops/s	$\textbf{\color{#d91a1a}-5.38\%}$
test_cat	34.3208ms	28.3540ms	35.2684 Ops/s	36.9050 Ops/s	$\color{#d91a1a}-4.43\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}53$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.3000μs	11.3633μs	88.0024 KOps/s	75.2640 KOps/s	$\textbf{\color{#35bf28}+16.93\%}$
test_plain_set_stack_nested	36.3000μs	11.4271μs	87.5111 KOps/s	74.4850 KOps/s	$\textbf{\color{#35bf28}+17.49\%}$
test_plain_set_nested_inplace	46.0500μs	12.3618μs	80.8944 KOps/s	69.2856 KOps/s	$\textbf{\color{#35bf28}+16.75\%}$
test_plain_set_stack_nested_inplace	60.2710μs	12.4437μs	80.3620 KOps/s	69.4986 KOps/s	$\textbf{\color{#35bf28}+15.63\%}$
test_items	69.5810μs	2.8923μs	345.7470 KOps/s	339.4108 KOps/s	$\color{#35bf28}+1.87\%$
test_items_nested	0.4038ms	0.3557ms	2.8110 KOps/s	2.7626 KOps/s	$\color{#35bf28}+1.75\%$
test_items_nested_locked	0.4208ms	0.3577ms	2.7953 KOps/s	2.7773 KOps/s	$\color{#35bf28}+0.65\%$
test_items_nested_leaf	93.0810μs	58.0691μs	17.2209 KOps/s	17.1529 KOps/s	$\color{#35bf28}+0.40\%$
test_items_stack_nested	0.3942ms	0.3565ms	2.8050 KOps/s	2.7723 KOps/s	$\color{#35bf28}+1.18\%$
test_items_stack_nested_leaf	84.4510μs	60.1466μs	16.6260 KOps/s	17.0248 KOps/s	$\color{#d91a1a}-2.34\%$
test_items_stack_nested_locked	0.4707ms	0.3563ms	2.8064 KOps/s	2.7752 KOps/s	$\color{#35bf28}+1.12\%$
test_keys	31.3000μs	3.4319μs	291.3819 KOps/s	288.1301 KOps/s	$\color{#35bf28}+1.13\%$
test_keys_nested	0.1082ms	82.4754μs	12.1248 KOps/s	12.2039 KOps/s	$\color{#d91a1a}-0.65\%$
test_keys_nested_locked	0.7657ms	88.2032μs	11.3375 KOps/s	11.4096 KOps/s	$\color{#d91a1a}-0.63\%$
test_keys_nested_leaf	2.5998ms	74.3649μs	13.4472 KOps/s	13.8206 KOps/s	$\color{#d91a1a}-2.70\%$
test_keys_stack_nested	0.1418ms	84.8440μs	11.7863 KOps/s	12.1211 KOps/s	$\color{#d91a1a}-2.76\%$
test_keys_stack_nested_leaf	0.1130ms	75.6827μs	13.2131 KOps/s	13.7155 KOps/s	$\color{#d91a1a}-3.66\%$
test_keys_stack_nested_locked	0.5436ms	89.7696μs	11.1396 KOps/s	11.5460 KOps/s	$\color{#d91a1a}-3.52\%$
test_values	11.9352μs	0.8553μs	1.1691 MOps/s	1.1629 MOps/s	$\color{#35bf28}+0.54\%$
test_values_nested	61.2700μs	34.9226μs	28.6348 KOps/s	29.1208 KOps/s	$\color{#d91a1a}-1.67\%$
test_values_nested_locked	68.1510μs	36.7336μs	27.2230 KOps/s	27.9047 KOps/s	$\color{#d91a1a}-2.44\%$
test_values_nested_leaf	69.9510μs	39.8347μs	25.1037 KOps/s	25.7784 KOps/s	$\color{#d91a1a}-2.62\%$
test_values_stack_nested	69.2200μs	35.2317μs	28.3835 KOps/s	28.7541 KOps/s	$\color{#d91a1a}-1.29\%$
test_values_stack_nested_leaf	72.4610μs	40.0579μs	24.9639 KOps/s	25.7002 KOps/s	$\color{#d91a1a}-2.87\%$
test_values_stack_nested_locked	64.3210μs	37.3817μs	26.7511 KOps/s	27.4776 KOps/s	$\color{#d91a1a}-2.64\%$
test_membership	1.9545μs	0.5119μs	1.9535 MOps/s	1.9644 MOps/s	$\color{#d91a1a}-0.55\%$
test_membership_nested	20.5200μs	2.0240μs	494.0764 KOps/s	505.2648 KOps/s	$\color{#d91a1a}-2.21\%$
test_membership_nested_leaf	17.5300μs	1.9932μs	501.7107 KOps/s	497.7890 KOps/s	$\color{#35bf28}+0.79\%$
test_membership_stacked_nested	38.8200μs	2.0578μs	485.9600 KOps/s	482.6920 KOps/s	$\color{#35bf28}+0.68\%$
test_membership_stacked_nested_leaf	22.5600μs	2.0438μs	489.2810 KOps/s	489.1490 KOps/s	$\color{#35bf28}+0.03\%$
test_membership_nested_last	24.8700μs	3.0927μs	323.3447 KOps/s	326.5184 KOps/s	$\color{#d91a1a}-0.97\%$
test_membership_nested_leaf_last	26.0600μs	3.0735μs	325.3637 KOps/s	324.7407 KOps/s	$\color{#35bf28}+0.19\%$
test_membership_stacked_nested_last	20.7300μs	3.1379μs	318.6830 KOps/s	121.8870 KOps/s	$\textbf{\color{#35bf28}+161.46\%}$
test_membership_stacked_nested_leaf_last	32.7200μs	3.1052μs	322.0454 KOps/s	122.6076 KOps/s	$\textbf{\color{#35bf28}+162.66\%}$
test_nested_getleaf	28.2600μs	6.2487μs	160.0340 KOps/s	163.2573 KOps/s	$\color{#d91a1a}-1.97\%$
test_nested_get	43.8110μs	5.8754μs	170.2014 KOps/s	174.9593 KOps/s	$\color{#d91a1a}-2.72\%$
test_stacked_getleaf	32.2500μs	6.1961μs	161.3929 KOps/s	163.4724 KOps/s	$\color{#d91a1a}-1.27\%$
test_stacked_get	27.4200μs	5.8458μs	171.0618 KOps/s	172.3630 KOps/s	$\color{#d91a1a}-0.75\%$
test_nested_getitemleaf	38.9600μs	6.3244μs	158.1174 KOps/s	162.1917 KOps/s	$\color{#d91a1a}-2.51\%$
test_nested_getitem	33.3200μs	6.0331μs	165.7526 KOps/s	170.7811 KOps/s	$\color{#d91a1a}-2.94\%$
test_stacked_getitemleaf	34.7700μs	6.2547μs	159.8797 KOps/s	161.1184 KOps/s	$\color{#d91a1a}-0.77\%$
test_stacked_getitem	34.5110μs	5.9036μs	169.3875 KOps/s	169.1588 KOps/s	$\color{#35bf28}+0.14\%$
test_lock_nested	2.4426ms	0.3744ms	2.6708 KOps/s	2.6265 KOps/s	$\color{#35bf28}+1.69\%$
test_lock_stack_nested	0.3865ms	0.3460ms	2.8904 KOps/s	2.8763 KOps/s	$\color{#35bf28}+0.49\%$
test_unlock_nested	0.6434ms	0.3144ms	3.1807 KOps/s	3.0935 KOps/s	$\color{#35bf28}+2.82\%$
test_unlock_stack_nested	0.3164ms	0.2840ms	3.5210 KOps/s	3.5029 KOps/s	$\color{#35bf28}+0.52\%$
test_flatten_speed	0.1193ms	74.5185μs	13.4195 KOps/s	13.3378 KOps/s	$\color{#35bf28}+0.61\%$
test_unflatten_speed	0.3752ms	0.3242ms	3.0845 KOps/s	3.1624 KOps/s	$\color{#d91a1a}-2.46\%$
test_common_ops	1.5498ms	0.5747ms	1.7399 KOps/s	1.5066 KOps/s	$\textbf{\color{#35bf28}+15.49\%}$
test_creation	93.4310μs	1.7777μs	562.5091 KOps/s	577.8049 KOps/s	$\color{#d91a1a}-2.65\%$
test_creation_empty	30.0500μs	6.5056μs	153.7142 KOps/s	96.6963 KOps/s	$\textbf{\color{#35bf28}+58.97\%}$
test_creation_nested_1	1.6219ms	8.0845μs	123.6931 KOps/s	83.3374 KOps/s	$\textbf{\color{#35bf28}+48.42\%}$
test_creation_nested_2	37.5200μs	10.7792μs	92.7716 KOps/s	67.6008 KOps/s	$\textbf{\color{#35bf28}+37.23\%}$
test_clone	49.3410μs	10.8569μs	92.1075 KOps/s	87.0055 KOps/s	$\textbf{\color{#35bf28}+5.86\%}$
test_getitem[int]	1.2210ms	10.5905μs	94.4241 KOps/s	90.4984 KOps/s	$\color{#35bf28}+4.34\%$
test_getitem[slice_int]	0.1393ms	20.2009μs	49.5028 KOps/s	46.2026 KOps/s	$\textbf{\color{#35bf28}+7.14\%}$
test_getitem[range]	0.1248ms	37.5313μs	26.6444 KOps/s	25.4080 KOps/s	$\color{#35bf28}+4.87\%$
test_getitem[tuple]	0.1059ms	18.3246μs	54.5716 KOps/s	52.9074 KOps/s	$\color{#35bf28}+3.15\%$
test_getitem[list]	0.1271ms	33.6948μs	29.6782 KOps/s	28.2754 KOps/s	$\color{#35bf28}+4.96\%$
test_setitem_dim[int]	47.2800μs	18.9065μs	52.8919 KOps/s	49.5011 KOps/s	$\textbf{\color{#35bf28}+6.85\%}$
test_setitem_dim[slice_int]	75.0710μs	38.8742μs	25.7240 KOps/s	24.9798 KOps/s	$\color{#35bf28}+2.98\%$
test_setitem_dim[range]	94.1010μs	53.5166μs	18.6858 KOps/s	17.9599 KOps/s	$\color{#35bf28}+4.04\%$
test_setitem_dim[tuple]	54.4010μs	32.5005μs	30.7688 KOps/s	29.5378 KOps/s	$\color{#35bf28}+4.17\%$
test_setitem	0.1240ms	14.4368μs	69.2676 KOps/s	58.8040 KOps/s	$\textbf{\color{#35bf28}+17.79\%}$
test_set	0.1132ms	13.8783μs	72.0549 KOps/s	60.3603 KOps/s	$\textbf{\color{#35bf28}+19.37\%}$
test_set_shared	1.9054ms	0.1548ms	6.4612 KOps/s	6.3546 KOps/s	$\color{#35bf28}+1.68\%$
test_update	0.3465ms	15.8113μs	63.2459 KOps/s	48.8949 KOps/s	$\textbf{\color{#35bf28}+29.35\%}$
test_update_nested	0.1222ms	20.8555μs	47.9491 KOps/s	38.6339 KOps/s	$\textbf{\color{#35bf28}+24.11\%}$
test_update__nested	0.7770ms	26.1346μs	38.2635 KOps/s	38.7751 KOps/s	$\color{#d91a1a}-1.32\%$
test_set_nested	35.6000μs	15.1060μs	66.1987 KOps/s	56.4731 KOps/s	$\textbf{\color{#35bf28}+17.22\%}$
test_set_nested_new	0.1284ms	17.3557μs	57.6181 KOps/s	49.3256 KOps/s	$\textbf{\color{#35bf28}+16.81\%}$
test_select	66.2010μs	30.3694μs	32.9278 KOps/s	31.5370 KOps/s	$\color{#35bf28}+4.41\%$
test_select_nested	83.1610μs	43.2232μs	23.1357 KOps/s	22.8815 KOps/s	$\color{#35bf28}+1.11\%$
test_exclude_nested	89.8810μs	63.0404μs	15.8628 KOps/s	15.8868 KOps/s	$\color{#d91a1a}-0.15\%$
test_empty[True]	0.3589ms	0.2903ms	3.4452 KOps/s	3.4728 KOps/s	$\color{#d91a1a}-0.79\%$
test_empty[False]	2.9371μs	0.8244μs	1.2130 MOps/s	1.2208 MOps/s	$\color{#d91a1a}-0.64\%$
test_to	87.3810μs	56.9377μs	17.5631 KOps/s	16.9568 KOps/s	$\color{#35bf28}+3.58\%$
test_to_nonblocking	85.4710μs	48.5608μs	20.5927 KOps/s	18.9389 KOps/s	$\textbf{\color{#35bf28}+8.73\%}$
test_unbind_speed	0.2657ms	0.2368ms	4.2230 KOps/s	4.0137 KOps/s	$\textbf{\color{#35bf28}+5.22\%}$
test_unbind_speed_stack0	0.3079ms	0.2383ms	4.1971 KOps/s	4.1340 KOps/s	$\color{#35bf28}+1.53\%$
test_unbind_speed_stack1	93.5319ms	0.6707ms	1.4910 KOps/s	1.4864 KOps/s	$\color{#35bf28}+0.31\%$
test_split	93.3797ms	1.5800ms	632.8983 Ops/s	613.6553 Ops/s	$\color{#35bf28}+3.14\%$
test_chunk	95.7540ms	1.5894ms	629.1717 Ops/s	606.6456 Ops/s	$\color{#35bf28}+3.71\%$
test_consolidate[False-None]	96.6021ms	2.9189ms	342.5956 Ops/s	334.9664 Ops/s	$\color{#35bf28}+2.28\%$
test_consolidate[default-None]	1.7719ms	1.7056ms	586.3149 Ops/s	580.0271 Ops/s	$\color{#35bf28}+1.08\%$
test_consolidate[reduce-overhead-None]	1.8357ms	1.7412ms	574.3030 Ops/s	563.1216 Ops/s	$\color{#35bf28}+1.99\%$
test_consolidate_njt[False-None]	6.7064ms	6.6069ms	151.3561 Ops/s	111.5554 Ops/s	$\textbf{\color{#35bf28}+35.68\%}$
test_to[False-False-None]	1.8859ms	1.8059ms	553.7429 Ops/s	551.2749 Ops/s	$\color{#35bf28}+0.45\%$
test_to[True-False-None]	1.5901ms	1.3481ms	741.7658 Ops/s	710.7992 Ops/s	$\color{#35bf28}+4.36\%$
test_to[within-False-None]	0.2953s	5.3516ms	186.8599 Ops/s	232.4479 Ops/s	$\textbf{\color{#d91a1a}-19.61\%}$
test_to[True-default-None]	5.5471ms	5.1986ms	192.3579 Ops/s	182.0097 Ops/s	$\textbf{\color{#35bf28}+5.69\%}$
test_to_njt[False-False-None]	7.1218ms	6.9936ms	142.9883 Ops/s	138.9756 Ops/s	$\color{#35bf28}+2.89\%$
test_to_njt[True-False-None]	5.8725ms	5.5501ms	180.1757 Ops/s	177.5313 Ops/s	$\color{#35bf28}+1.49\%$
test_to_njt[within-False-None]	12.4188ms	12.2335ms	81.7424 Ops/s	80.1717 Ops/s	$\color{#35bf28}+1.96\%$
test_creation[device0]	0.5232ms	81.7207μs	12.2368 KOps/s	11.9559 KOps/s	$\color{#35bf28}+2.35\%$
test_creation_from_tensor	0.4404ms	85.4019μs	11.7093 KOps/s	11.2744 KOps/s	$\color{#35bf28}+3.86\%$
test_add_one[memmap_tensor0]	0.4873ms	6.8549μs	145.8813 KOps/s	134.9652 KOps/s	$\textbf{\color{#35bf28}+8.09\%}$
test_contiguous[memmap_tensor0]	2.1985μs	0.4285μs	2.3335 MOps/s	2.3119 MOps/s	$\color{#35bf28}+0.93\%$
test_stack[memmap_tensor0]	42.6310μs	4.3582μs	229.4508 KOps/s	208.7174 KOps/s	$\textbf{\color{#35bf28}+9.93\%}$
test_memmaptd_index	2.0679ms	0.2488ms	4.0193 KOps/s	3.7105 KOps/s	$\textbf{\color{#35bf28}+8.32\%}$
test_memmaptd_index_astensor	0.9868ms	0.3100ms	3.2254 KOps/s	3.0104 KOps/s	$\textbf{\color{#35bf28}+7.14\%}$
test_memmaptd_index_op	0.9565ms	0.5604ms	1.7845 KOps/s	1.5146 KOps/s	$\textbf{\color{#35bf28}+17.82\%}$
test_serialize_model	0.1325s	0.1311s	7.6260 Ops/s	7.6341 Ops/s	$\color{#d91a1a}-0.11\%$
test_serialize_model_pickle	1.3495s	1.2137s	0.8239 Ops/s	0.8429 Ops/s	$\color{#d91a1a}-2.25\%$
test_serialize_weights	0.2808s	0.1524s	6.5609 Ops/s	7.6969 Ops/s	$\textbf{\color{#d91a1a}-14.76\%}$
test_serialize_weights_returnearly	0.3389s	53.6858ms	18.6269 Ops/s	14.7953 Ops/s	$\textbf{\color{#35bf28}+25.90\%}$
test_serialize_weights_pickle	1.3752s	1.2153s	0.8229 Ops/s	0.8225 Ops/s	$\color{#35bf28}+0.04\%$
test_reshape_pytree	0.1219ms	21.8178μs	45.8342 KOps/s	43.6144 KOps/s	$\textbf{\color{#35bf28}+5.09\%}$
test_reshape_td	64.5910μs	26.0131μs	38.4422 KOps/s	36.7051 KOps/s	$\color{#35bf28}+4.73\%$
test_view_pytree	0.1122ms	21.9516μs	45.5548 KOps/s	44.6841 KOps/s	$\color{#35bf28}+1.95\%$
test_view_td	0.1163ms	28.8557μs	34.6552 KOps/s	31.7181 KOps/s	$\textbf{\color{#35bf28}+9.26\%}$
test_unbind_pytree	0.4437ms	27.6531μs	36.1624 KOps/s	35.2456 KOps/s	$\color{#35bf28}+2.60\%$
test_unbind_td	0.6386ms	35.5544μs	28.1259 KOps/s	26.5844 KOps/s	$\textbf{\color{#35bf28}+5.80\%}$
test_split_pytree	0.4420ms	30.0063μs	33.3264 KOps/s	33.3365 KOps/s	$\color{#d91a1a}-0.03\%$
test_split_td	0.7421ms	37.8426μs	26.4253 KOps/s	25.5673 KOps/s	$\color{#35bf28}+3.36\%$
test_add_pytree	0.4388ms	34.7528μs	28.7747 KOps/s	27.5847 KOps/s	$\color{#35bf28}+4.31\%$
test_add_td	84.1710μs	45.4959μs	21.9800 KOps/s	19.0663 KOps/s	$\textbf{\color{#35bf28}+15.28\%}$
test_compile_add_one_nested[tensordict-compile]	0.1723ms	0.1206ms	8.2938 KOps/s	7.9156 KOps/s	$\color{#35bf28}+4.78\%$
test_compile_add_one_nested[tensordict-eager]	0.2258ms	0.1309ms	7.6419 KOps/s	7.5984 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_add_one_nested[pytree-compile]	0.1386ms	97.2854μs	10.2790 KOps/s	9.8688 KOps/s	$\color{#35bf28}+4.16\%$
test_compile_add_one_nested[pytree-eager]	1.1223ms	0.1537ms	6.5052 KOps/s	6.3234 KOps/s	$\color{#35bf28}+2.88\%$
test_compile_copy_nested[tensordict-compile]	57.6210μs	23.0207μs	43.4392 KOps/s	48.0920 KOps/s	$\textbf{\color{#d91a1a}-9.67\%}$
test_compile_copy_nested[tensordict-eager]	51.2700μs	30.1830μs	33.1312 KOps/s	33.9257 KOps/s	$\color{#d91a1a}-2.34\%$
test_compile_copy_nested[pytree-compile]	0.1024ms	65.3298μs	15.3069 KOps/s	15.0453 KOps/s	$\color{#35bf28}+1.74\%$
test_compile_copy_nested[pytree-eager]	0.1042ms	49.4037μs	20.2414 KOps/s	20.1536 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_one_flat[tensordict-compile]	0.1819ms	0.1426ms	7.0126 KOps/s	7.0200 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_add_one_flat[tensordict-eager]	0.3037ms	0.2152ms	4.6468 KOps/s	4.6835 KOps/s	$\color{#d91a1a}-0.78\%$
test_compile_add_one_flat[tensorclass-compile]	0.1356ms	97.9850μs	10.2056 KOps/s	10.2050 KOps/s	$+0.01\%$
test_compile_add_one_flat[tensorclass-eager]	0.1239ms	54.1739μs	18.4591 KOps/s	18.5488 KOps/s	$\color{#d91a1a}-0.48\%$
test_compile_add_one_flat[pytree-compile]	0.1777ms	0.1365ms	7.3245 KOps/s	7.3294 KOps/s	$\color{#d91a1a}-0.07\%$
test_compile_add_one_flat[pytree-eager]	0.5395ms	0.4995ms	2.0019 KOps/s	1.9241 KOps/s	$\color{#35bf28}+4.04\%$
test_compile_add_self_flat[tensordict-eager]	0.3619ms	0.2594ms	3.8554 KOps/s	3.9016 KOps/s	$\color{#d91a1a}-1.18\%$
test_compile_add_self_flat[tensordict-compile]	0.2474ms	0.1440ms	6.9428 KOps/s	7.0489 KOps/s	$\color{#d91a1a}-1.51\%$
test_compile_add_self_flat[tensorclass-eager]	0.1436ms	65.6194μs	15.2394 KOps/s	14.9583 KOps/s	$\color{#35bf28}+1.88\%$
test_compile_add_self_flat[tensorclass-compile]	0.1437ms	99.3156μs	10.0689 KOps/s	9.9036 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_add_self_flat[pytree-eager]	0.4685ms	0.4239ms	2.3591 KOps/s	2.3235 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_add_self_flat[pytree-compile]	0.1810ms	0.1360ms	7.3536 KOps/s	7.1820 KOps/s	$\color{#35bf28}+2.39\%$
test_compile_copy_flat[tensordict-compile]	55.1510μs	18.9072μs	52.8899 KOps/s	57.9880 KOps/s	$\textbf{\color{#d91a1a}-8.79\%}$
test_compile_copy_flat[tensordict-eager]	0.1326ms	31.5912μs	31.6544 KOps/s	32.2905 KOps/s	$\color{#d91a1a}-1.97\%$
test_compile_copy_flat[pytree-compile]	0.1135ms	71.2771μs	14.0298 KOps/s	13.4762 KOps/s	$\color{#35bf28}+4.11\%$
test_compile_copy_flat[pytree-eager]	83.5210μs	51.9374μs	19.2539 KOps/s	18.5292 KOps/s	$\color{#35bf28}+3.91\%$
test_compile_assign_and_add[tensordict-compile]	1.6498ms	0.3964ms	2.5224 KOps/s	2.1547 KOps/s	$\textbf{\color{#35bf28}+17.06\%}$
test_compile_assign_and_add[tensordict-eager]	3.1896ms	2.6615ms	375.7277 Ops/s	363.4511 Ops/s	$\color{#35bf28}+3.38\%$
test_compile_assign_and_add[pytree-compile]	1.6046ms	0.3826ms	2.6135 KOps/s	2.2367 KOps/s	$\textbf{\color{#35bf28}+16.84\%}$
test_compile_assign_and_add[pytree-eager]	3.0125ms	2.8286ms	353.5370 Ops/s	351.3375 Ops/s	$\color{#35bf28}+0.63\%$
test_compile_indexing[tensor-tensordict-compile]	0.6055ms	0.1174ms	8.5153 KOps/s	8.3757 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_indexing[tensor-tensordict-eager]	0.5653ms	86.4384μs	11.5689 KOps/s	12.1567 KOps/s	$\color{#d91a1a}-4.84\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5417ms	0.1114ms	8.9768 KOps/s	8.9203 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1202ms	74.1720μs	13.4822 KOps/s	13.7022 KOps/s	$\color{#d91a1a}-1.61\%$
test_compile_indexing[tensor-pytree-compile]	0.2091ms	0.1109ms	9.0168 KOps/s	8.8771 KOps/s	$\color{#35bf28}+1.57\%$
test_compile_indexing[tensor-pytree-eager]	0.1138ms	74.8437μs	13.3612 KOps/s	13.8049 KOps/s	$\color{#d91a1a}-3.21\%$
test_compile_indexing[slice-tensordict-compile]	0.1529ms	0.1027ms	9.7328 KOps/s	9.5405 KOps/s	$\color{#35bf28}+2.02\%$
test_compile_indexing[slice-tensordict-eager]	0.1417ms	17.3951μs	57.4873 KOps/s	55.4503 KOps/s	$\color{#35bf28}+3.67\%$
test_compile_indexing[slice-tensorclass-compile]	0.1766ms	0.1011ms	9.8925 KOps/s	10.3635 KOps/s	$\color{#d91a1a}-4.54\%$
test_compile_indexing[slice-tensorclass-eager]	59.4000μs	15.8555μs	63.0696 KOps/s	59.9364 KOps/s	$\textbf{\color{#35bf28}+5.23\%}$
test_compile_indexing[slice-pytree-compile]	0.1460ms	0.1018ms	9.8258 KOps/s	10.0912 KOps/s	$\color{#d91a1a}-2.63\%$
test_compile_indexing[slice-pytree-eager]	51.7110μs	15.7281μs	63.5805 KOps/s	60.0246 KOps/s	$\textbf{\color{#35bf28}+5.92\%}$
test_compile_indexing[int-tensordict-compile]	0.2051ms	0.1045ms	9.5712 KOps/s	9.4284 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_indexing[int-tensordict-eager]	0.5633ms	18.0110μs	55.5217 KOps/s	55.3820 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_indexing[int-tensorclass-compile]	0.1581ms	0.1028ms	9.7281 KOps/s	9.9338 KOps/s	$\color{#d91a1a}-2.07\%$
test_compile_indexing[int-tensorclass-eager]	0.1896ms	16.3276μs	61.2460 KOps/s	59.9578 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_indexing[int-pytree-compile]	0.1918ms	0.1005ms	9.9466 KOps/s	9.9787 KOps/s	$\color{#d91a1a}-0.32\%$
test_compile_indexing[int-pytree-eager]	56.4110μs	16.1640μs	61.8660 KOps/s	59.9465 KOps/s	$\color{#35bf28}+3.20\%$
test_mod_add[eager]	80.4010μs	37.2490μs	26.8464 KOps/s	24.1423 KOps/s	$\textbf{\color{#35bf28}+11.20\%}$
test_mod_add[compile]	0.4622ms	83.5305μs	11.9717 KOps/s	12.2972 KOps/s	$\color{#d91a1a}-2.65\%$
test_mod_add[compile-overhead]	0.3249ms	0.1672ms	5.9793 KOps/s	5.5945 KOps/s	$\textbf{\color{#35bf28}+6.88\%}$
test_mod_wrap[eager]	0.3363ms	0.2519ms	3.9694 KOps/s	3.7490 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_mod_wrap[compile]	0.3752ms	0.2859ms	3.4977 KOps/s	3.3339 KOps/s	$\color{#35bf28}+4.91\%$
test_mod_wrap[compile-overhead]	7.0833ms	3.7547ms	266.3333 Ops/s	275.3940 Ops/s	$\color{#d91a1a}-3.29\%$
test_mod_wrap_and_backward[eager]	1.5621ms	1.4827ms	674.4364 Ops/s	669.9767 Ops/s	$\color{#35bf28}+0.67\%$
test_mod_wrap_and_backward[compile]	1.4751ms	1.3771ms	726.1586 Ops/s	711.0268 Ops/s	$\color{#35bf28}+2.13\%$
test_mod_wrap_and_backward[compile-overhead]	1.5444ms	1.0450ms	956.9312 Ops/s	929.7506 Ops/s	$\color{#35bf28}+2.92\%$
test_seq_add[eager]	0.2040ms	0.1120ms	8.9272 KOps/s	8.3379 KOps/s	$\textbf{\color{#35bf28}+7.07\%}$
test_seq_add[compile]	0.1312ms	87.9928μs	11.3646 KOps/s	11.1677 KOps/s	$\color{#35bf28}+1.76\%$
test_seq_add[compile-overhead]	0.1735ms	0.1299ms	7.6965 KOps/s	7.4135 KOps/s	$\color{#35bf28}+3.82\%$
test_seq_wrap[eager]	0.4929ms	0.4271ms	2.3413 KOps/s	2.1906 KOps/s	$\textbf{\color{#35bf28}+6.88\%}$
test_seq_wrap[compile]	0.3544ms	0.2973ms	3.3635 KOps/s	3.1730 KOps/s	$\textbf{\color{#35bf28}+6.00\%}$
test_seq_wrap[compile-overhead]	0.2839ms	0.2258ms	4.4296 KOps/s	4.3585 KOps/s	$\color{#35bf28}+1.63\%$
test_func_call_runtime[False-eager]	0.8656ms	0.7942ms	1.2592 KOps/s	1.2673 KOps/s	$\color{#d91a1a}-0.64\%$
test_func_call_runtime[False-compile]	0.8166ms	0.7428ms	1.3463 KOps/s	1.2708 KOps/s	$\textbf{\color{#35bf28}+5.94\%}$
test_func_call_runtime[False-compile-overhead]	0.4178ms	0.3670ms	2.7246 KOps/s	2.6363 KOps/s	$\color{#35bf28}+3.35\%$
test_func_call_runtime[True-eager]	0.9977ms	0.9190ms	1.0882 KOps/s	1.0404 KOps/s	$\color{#35bf28}+4.59\%$
test_func_call_runtime[True-compile]	0.8270ms	0.7671ms	1.3037 KOps/s	1.2424 KOps/s	$\color{#35bf28}+4.93\%$
test_func_call_runtime[True-compile-overhead]	0.4478ms	0.3871ms	2.5833 KOps/s	2.5614 KOps/s	$\color{#35bf28}+0.85\%$
test_func_call_cm_runtime[False-eager]	0.8275ms	0.7434ms	1.3452 KOps/s	1.2283 KOps/s	$\textbf{\color{#35bf28}+9.51\%}$
test_func_call_cm_runtime[False-compile]	0.8321ms	0.7572ms	1.3206 KOps/s	1.2900 KOps/s	$\color{#35bf28}+2.37\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4375ms	0.3709ms	2.6958 KOps/s	2.6795 KOps/s	$\color{#35bf28}+0.61\%$
test_func_call_cm_runtime[True-eager]	1.1275ms	1.0275ms	973.2252 Ops/s	958.6806 Ops/s	$\color{#35bf28}+1.52\%$
test_func_call_cm_runtime[True-compile]	0.9649ms	0.7932ms	1.2607 KOps/s	1.2152 KOps/s	$\color{#35bf28}+3.75\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4628ms	0.4151ms	2.4092 KOps/s	2.3772 KOps/s	$\color{#35bf28}+1.35\%$
test_vmap_func_call_cm_runtime[eager]	2.6723ms	2.1192ms	471.8802 Ops/s	468.5931 Ops/s	$\color{#35bf28}+0.70\%$
test_vmap_func_call_cm_runtime[compile]	0.8911ms	0.8090ms	1.2361 KOps/s	1.1819 KOps/s	$\color{#35bf28}+4.58\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4757ms	0.4170ms	2.3981 KOps/s	2.3623 KOps/s	$\color{#35bf28}+1.51\%$
test_distributed	2.0049ms	0.2785ms	3.5908 KOps/s	8.5094 KOps/s	$\textbf{\color{#d91a1a}-57.80\%}$
test_tdmodule	52.8800μs	19.2489μs	51.9510 KOps/s	44.8333 KOps/s	$\textbf{\color{#35bf28}+15.88\%}$
test_tdmodule_dispatch	62.8510μs	33.4948μs	29.8554 KOps/s	26.0740 KOps/s	$\textbf{\color{#35bf28}+14.50\%}$
test_tdseq	45.2500μs	19.5065μs	51.2649 KOps/s	45.1645 KOps/s	$\textbf{\color{#35bf28}+13.51\%}$
test_tdseq_dispatch	64.8310μs	35.7324μs	27.9858 KOps/s	24.0186 KOps/s	$\textbf{\color{#35bf28}+16.52\%}$
test_instantiation_functorch	1.7137ms	1.5887ms	629.4357 Ops/s	627.3630 Ops/s	$\color{#35bf28}+0.33\%$
test_exec_functorch	0.2171ms	0.1471ms	6.7983 KOps/s	6.5941 KOps/s	$\color{#35bf28}+3.10\%$
test_exec_functional_call	0.2045ms	0.1438ms	6.9535 KOps/s	6.9005 KOps/s	$\color{#35bf28}+0.77\%$
test_exec_td_decorator	0.3799ms	0.1932ms	5.1759 KOps/s	5.1091 KOps/s	$\color{#35bf28}+1.31\%$
test_vmap_mlp_speed_decorator[True-True]	0.8591ms	0.7015ms	1.4255 KOps/s	1.4188 KOps/s	$\color{#35bf28}+0.47\%$
test_vmap_mlp_speed_decorator[True-False]	0.8381ms	0.6964ms	1.4359 KOps/s	1.4196 KOps/s	$\color{#35bf28}+1.14\%$
test_vmap_mlp_speed_decorator[False-True]	0.7284ms	0.6066ms	1.6486 KOps/s	1.5888 KOps/s	$\color{#35bf28}+3.76\%$
test_vmap_mlp_speed_decorator[False-False]	0.7471ms	0.6165ms	1.6219 KOps/s	1.5672 KOps/s	$\color{#35bf28}+3.49\%$
test_vmap_transformer_speed_decorator[True-True]	20.6705ms	19.6115ms	50.9905 Ops/s	49.4686 Ops/s	$\color{#35bf28}+3.08\%$
test_vmap_transformer_speed_decorator[True-False]	20.3658ms	20.0226ms	49.9434 Ops/s	50.8279 Ops/s	$\color{#d91a1a}-1.74\%$
test_vmap_transformer_speed_decorator[False-True]	19.4800ms	19.3785ms	51.6036 Ops/s	50.7653 Ops/s	$\color{#35bf28}+1.65\%$
test_vmap_transformer_speed_decorator[False-False]	20.0117ms	19.3992ms	51.5486 Ops/s	51.0217 Ops/s	$\color{#35bf28}+1.03\%$
test_to_module_speed[True]	1.1536ms	0.9804ms	1.0200 KOps/s	1.0217 KOps/s	$\color{#d91a1a}-0.16\%$
test_to_module_speed[False]	1.5215ms	0.9492ms	1.0536 KOps/s	1.0595 KOps/s	$\color{#d91a1a}-0.56\%$
test_tc_init	0.1303ms	33.8136μs	29.5739 KOps/s	26.2003 KOps/s	$\textbf{\color{#35bf28}+12.88\%}$
test_tc_init_nested	0.1132ms	68.6564μs	14.5653 KOps/s	13.1768 KOps/s	$\textbf{\color{#35bf28}+10.54\%}$
test_tc_first_layer_tensor	22.7500μs	0.8167μs	1.2244 MOps/s	1.2185 MOps/s	$\color{#35bf28}+0.48\%$
test_tc_first_layer_nontensor	25.5700μs	2.2716μs	440.2111 KOps/s	436.9428 KOps/s	$\color{#35bf28}+0.75\%$
test_tc_second_layer_tensor	14.8500μs	1.4459μs	691.6051 KOps/s	696.3324 KOps/s	$\color{#d91a1a}-0.68\%$
test_tc_second_layer_nontensor	34.2300μs	3.0144μs	331.7395 KOps/s	327.7644 KOps/s	$\color{#35bf28}+1.21\%$
test_unbind	0.2367s	10.2088ms	97.9544 Ops/s	140.6716 Ops/s	$\textbf{\color{#d91a1a}-30.37\%}$
test_full_like	10.1784ms	9.2097ms	108.5807 Ops/s	107.4314 Ops/s	$\color{#35bf28}+1.07\%$
test_zeros_like	9.3825ms	7.2479ms	137.9706 Ops/s	229.5144 Ops/s	$\textbf{\color{#d91a1a}-39.89\%}$
test_ones_like	4.9440ms	4.3224ms	231.3535 Ops/s	139.0868 Ops/s	$\textbf{\color{#35bf28}+66.34\%}$
test_clone	7.2756ms	6.4374ms	155.3412 Ops/s	108.5870 Ops/s	$\textbf{\color{#35bf28}+43.06\%}$
test_squeeze	95.1110μs	9.8681μs	101.3365 KOps/s	106.5290 KOps/s	$\color{#d91a1a}-4.87\%$
test_unsqueeze	0.1592ms	70.8538μs	14.1136 KOps/s	13.8872 KOps/s	$\color{#35bf28}+1.63\%$
test_split	0.3715ms	0.1601ms	6.2451 KOps/s	6.1799 KOps/s	$\color{#35bf28}+1.05\%$
test_permute	0.3206ms	0.1764ms	5.6700 KOps/s	5.5727 KOps/s	$\color{#35bf28}+1.75\%$
test_stack	51.0211ms	50.5659ms	19.7762 Ops/s	19.7059 Ops/s	$\color{#35bf28}+0.36\%$
test_cat	51.0839ms	50.4134ms	19.8360 Ops/s	19.7104 Ops/s	$\color{#35bf28}+0.64\%$