[Feature] Ensure all dists work with DETERMINSTIC type without warning

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [Feature] Ensure all dists work with DETERMINSTIC type without warning #1182

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}13$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.7330μs	20.8105μs	48.0526 KOps/s	47.7702 KOps/s	$\color{#35bf28}+0.59\%$
test_plain_set_stack_nested	71.1830μs	20.9467μs	47.7402 KOps/s	46.6584 KOps/s	$\color{#35bf28}+2.32\%$
test_plain_set_nested_inplace	70.2510μs	22.6839μs	44.0841 KOps/s	44.1236 KOps/s	$\color{#d91a1a}-0.09\%$
test_plain_set_stack_nested_inplace	82.0030μs	22.5425μs	44.3606 KOps/s	44.4539 KOps/s	$\color{#d91a1a}-0.21\%$
test_items	45.7460μs	4.1362μs	241.7655 KOps/s	240.6384 KOps/s	$\color{#35bf28}+0.47\%$
test_items_nested	0.7109ms	0.3932ms	2.5431 KOps/s	2.5158 KOps/s	$\color{#35bf28}+1.08\%$
test_items_nested_locked	0.8653ms	0.3932ms	2.5431 KOps/s	2.5268 KOps/s	$\color{#35bf28}+0.65\%$
test_items_nested_leaf	0.1499ms	77.2845μs	12.9392 KOps/s	12.9125 KOps/s	$\color{#35bf28}+0.21\%$
test_items_stack_nested	0.6066ms	0.3967ms	2.5206 KOps/s	2.5022 KOps/s	$\color{#35bf28}+0.73\%$
test_items_stack_nested_leaf	0.1411ms	79.7770μs	12.5349 KOps/s	12.6201 KOps/s	$\color{#d91a1a}-0.67\%$
test_items_stack_nested_locked	0.7476ms	0.3987ms	2.5085 KOps/s	2.4907 KOps/s	$\color{#35bf28}+0.71\%$
test_keys	44.3020μs	3.4534μs	289.5672 KOps/s	254.3927 KOps/s	$\textbf{\color{#35bf28}+13.83\%}$
test_keys_nested	0.3052ms	0.1614ms	6.1956 KOps/s	6.1176 KOps/s	$\color{#35bf28}+1.27\%$
test_keys_nested_locked	1.8955ms	0.1657ms	6.0348 KOps/s	5.8316 KOps/s	$\color{#35bf28}+3.48\%$
test_keys_nested_leaf	0.2631ms	0.1474ms	6.7838 KOps/s	7.0566 KOps/s	$\color{#d91a1a}-3.87\%$
test_keys_stack_nested	0.2373ms	0.1612ms	6.2035 KOps/s	6.2747 KOps/s	$\color{#d91a1a}-1.14\%$
test_keys_stack_nested_leaf	0.1966ms	0.1403ms	7.1273 KOps/s	7.3188 KOps/s	$\color{#d91a1a}-2.62\%$
test_keys_stack_nested_locked	0.3107ms	0.1675ms	5.9684 KOps/s	6.0028 KOps/s	$\color{#d91a1a}-0.57\%$
test_values	8.5860μs	1.0902μs	917.2520 KOps/s	955.8635 KOps/s	$\color{#d91a1a}-4.04\%$
test_values_nested	0.1358ms	61.8106μs	16.1784 KOps/s	16.1637 KOps/s	$\color{#35bf28}+0.09\%$
test_values_nested_locked	0.1107ms	62.0724μs	16.1102 KOps/s	16.2013 KOps/s	$\color{#d91a1a}-0.56\%$
test_values_nested_leaf	0.1228ms	71.1818μs	14.0485 KOps/s	13.5401 KOps/s	$\color{#35bf28}+3.75\%$
test_values_stack_nested	0.1188ms	62.7242μs	15.9428 KOps/s	15.6199 KOps/s	$\color{#35bf28}+2.07\%$
test_values_stack_nested_leaf	0.1237ms	70.7422μs	14.1358 KOps/s	14.4247 KOps/s	$\color{#d91a1a}-2.00\%$
test_values_stack_nested_locked	0.1112ms	62.8995μs	15.8984 KOps/s	15.7185 KOps/s	$\color{#35bf28}+1.14\%$
test_membership	16.2900μs	0.8691μs	1.1506 MOps/s	1.1363 MOps/s	$\color{#35bf28}+1.25\%$
test_membership_nested	47.7870μs	2.8351μs	352.7153 KOps/s	350.3034 KOps/s	$\color{#35bf28}+0.69\%$
test_membership_nested_leaf	26.3990μs	2.8584μs	349.8504 KOps/s	345.5403 KOps/s	$\color{#35bf28}+1.25\%$
test_membership_stacked_nested	24.4760μs	2.8515μs	350.6984 KOps/s	353.8042 KOps/s	$\color{#d91a1a}-0.88\%$
test_membership_stacked_nested_leaf	29.6350μs	2.8621μs	349.3926 KOps/s	331.3208 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_membership_nested_last	23.8650μs	4.2876μs	233.2315 KOps/s	236.2676 KOps/s	$\color{#d91a1a}-1.29\%$
test_membership_nested_leaf_last	47.5110μs	4.2809μs	233.5972 KOps/s	234.7507 KOps/s	$\color{#d91a1a}-0.49\%$
test_membership_stacked_nested_last	37.8300μs	4.2625μs	234.6033 KOps/s	234.5997 KOps/s	$+0.00\%$
test_membership_stacked_nested_leaf_last	48.4110μs	4.2710μs	234.1383 KOps/s	236.6069 KOps/s	$\color{#d91a1a}-1.04\%$
test_nested_getleaf	38.3110μs	10.5026μs	95.2144 KOps/s	93.7921 KOps/s	$\color{#35bf28}+1.52\%$
test_nested_get	55.4930μs	10.1384μs	98.6352 KOps/s	100.3198 KOps/s	$\color{#d91a1a}-1.68\%$
test_stacked_getleaf	39.1730μs	10.5409μs	94.8687 KOps/s	97.7277 KOps/s	$\color{#d91a1a}-2.93\%$
test_stacked_get	41.2970μs	10.0944μs	99.0645 KOps/s	102.8043 KOps/s	$\color{#d91a1a}-3.64\%$
test_nested_getitemleaf	78.9970μs	10.8763μs	91.9432 KOps/s	89.6313 KOps/s	$\color{#35bf28}+2.58\%$
test_nested_getitem	54.4310μs	10.4947μs	95.2859 KOps/s	94.7270 KOps/s	$\color{#35bf28}+0.59\%$
test_stacked_getitemleaf	38.0000μs	10.9614μs	91.2289 KOps/s	91.4821 KOps/s	$\color{#d91a1a}-0.28\%$
test_stacked_getitem	36.5580μs	10.4265μs	95.9091 KOps/s	95.3907 KOps/s	$\color{#35bf28}+0.54\%$
test_lock_nested	5.2853ms	0.4557ms	2.1946 KOps/s	1.8074 KOps/s	$\textbf{\color{#35bf28}+21.42\%}$
test_lock_stack_nested	0.7647ms	0.4229ms	2.3649 KOps/s	2.4135 KOps/s	$\color{#d91a1a}-2.01\%$
test_unlock_nested	0.7927ms	0.3733ms	2.6790 KOps/s	2.6920 KOps/s	$\color{#d91a1a}-0.48\%$
test_unlock_stack_nested	0.6560ms	0.3453ms	2.8961 KOps/s	3.0198 KOps/s	$\color{#d91a1a}-4.10\%$
test_flatten_speed	0.1999ms	0.1011ms	9.8873 KOps/s	9.9247 KOps/s	$\color{#d91a1a}-0.38\%$
test_unflatten_speed	0.6190ms	0.5147ms	1.9430 KOps/s	1.9201 KOps/s	$\color{#35bf28}+1.19\%$
test_common_ops	4.0381ms	0.8141ms	1.2284 KOps/s	1.2432 KOps/s	$\color{#d91a1a}-1.19\%$
test_creation	33.1010μs	2.4436μs	409.2311 KOps/s	399.7752 KOps/s	$\color{#35bf28}+2.37\%$
test_creation_empty	32.9710μs	12.3098μs	81.2362 KOps/s	80.8724 KOps/s	$\color{#35bf28}+0.45\%$
test_creation_nested_1	38.7120μs	15.2735μs	65.4728 KOps/s	65.3361 KOps/s	$\color{#35bf28}+0.21\%$
test_creation_nested_2	52.9190μs	19.5183μs	51.2339 KOps/s	50.3801 KOps/s	$\color{#35bf28}+1.69\%$
test_clone	72.5360μs	13.4130μs	74.5548 KOps/s	73.2991 KOps/s	$\color{#35bf28}+1.71\%$
test_getitem[int]	1.3341ms	13.1015μs	76.3271 KOps/s	78.3977 KOps/s	$\color{#d91a1a}-2.64\%$
test_getitem[slice_int]	0.1373ms	24.5234μs	40.7773 KOps/s	39.5636 KOps/s	$\color{#35bf28}+3.07\%$
test_getitem[range]	0.2531ms	49.0990μs	20.3670 KOps/s	20.4811 KOps/s	$\color{#d91a1a}-0.56\%$
test_getitem[tuple]	0.1330ms	20.4514μs	48.8964 KOps/s	47.9724 KOps/s	$\color{#35bf28}+1.93\%$
test_getitem[list]	0.3056ms	44.5911μs	22.4260 KOps/s	22.9823 KOps/s	$\color{#d91a1a}-2.42\%$
test_setitem_dim[int]	66.0230μs	25.3014μs	39.5235 KOps/s	37.9423 KOps/s	$\color{#35bf28}+4.17\%$
test_setitem_dim[slice_int]	0.1005ms	51.4352μs	19.4419 KOps/s	19.4130 KOps/s	$\color{#35bf28}+0.15\%$
test_setitem_dim[range]	0.1513ms	74.1387μs	13.4882 KOps/s	13.5911 KOps/s	$\color{#d91a1a}-0.76\%$
test_setitem_dim[tuple]	71.3840μs	39.9368μs	25.0396 KOps/s	24.2023 KOps/s	$\color{#35bf28}+3.46\%$
test_setitem	94.5070μs	20.8271μs	48.0145 KOps/s	46.9752 KOps/s	$\color{#35bf28}+2.21\%$
test_set	92.3420μs	20.4201μs	48.9713 KOps/s	47.9434 KOps/s	$\color{#35bf28}+2.14\%$
test_set_shared	1.2758ms	0.1707ms	5.8582 KOps/s	5.9022 KOps/s	$\color{#d91a1a}-0.75\%$
test_update	0.1903ms	23.5628μs	42.4398 KOps/s	41.9982 KOps/s	$\color{#35bf28}+1.05\%$
test_update_nested	0.1604ms	33.6614μs	29.7076 KOps/s	28.9951 KOps/s	$\color{#35bf28}+2.46\%$
test_update__nested	0.6060ms	33.6728μs	29.6976 KOps/s	29.0418 KOps/s	$\color{#35bf28}+2.26\%$
test_set_nested	78.6570μs	22.0999μs	45.2490 KOps/s	43.9688 KOps/s	$\color{#35bf28}+2.91\%$
test_set_nested_new	0.1042ms	26.4559μs	37.7987 KOps/s	36.4058 KOps/s	$\color{#35bf28}+3.83\%$
test_select	0.1552ms	42.6662μs	23.4378 KOps/s	22.6941 KOps/s	$\color{#35bf28}+3.28\%$
test_select_nested	0.1222ms	62.5495μs	15.9873 KOps/s	15.7702 KOps/s	$\color{#35bf28}+1.38\%$
test_exclude_nested	0.1585ms	79.4030μs	12.5940 KOps/s	12.1799 KOps/s	$\color{#35bf28}+3.40\%$
test_empty[True]	1.2836ms	0.4130ms	2.4215 KOps/s	2.4792 KOps/s	$\color{#d91a1a}-2.33\%$
test_empty[False]	10.1843μs	1.3465μs	742.6653 KOps/s	726.6446 KOps/s	$\color{#35bf28}+2.20\%$
test_unbind_speed	0.3798ms	0.2695ms	3.7102 KOps/s	3.6760 KOps/s	$\color{#35bf28}+0.93\%$
test_unbind_speed_stack0	0.4754ms	0.2665ms	3.7517 KOps/s	3.8602 KOps/s	$\color{#d91a1a}-2.81\%$
test_unbind_speed_stack1	0.1061s	0.8064ms	1.2401 KOps/s	1.4173 KOps/s	$\textbf{\color{#d91a1a}-12.51\%}$
test_split	0.1118s	1.8021ms	554.9034 Ops/s	559.2171 Ops/s	$\color{#d91a1a}-0.77\%$
test_chunk	0.1130s	1.8273ms	547.2566 Ops/s	561.6273 Ops/s	$\color{#d91a1a}-2.56\%$
test_consolidate_njt[False-None]	8.6260ms	8.1633ms	122.4996 Ops/s	123.0964 Ops/s	$\color{#d91a1a}-0.48\%$
test_creation[device0]	0.3080ms	90.9918μs	10.9900 KOps/s	11.2219 KOps/s	$\color{#d91a1a}-2.07\%$
test_creation_from_tensor	3.7096ms	96.1988μs	10.3951 KOps/s	10.5828 KOps/s	$\color{#d91a1a}-1.77\%$
test_add_one[memmap_tensor0]	0.1620ms	5.1637μs	193.6597 KOps/s	201.5399 KOps/s	$\color{#d91a1a}-3.91\%$
test_contiguous[memmap_tensor0]	14.2570μs	0.5416μs	1.8464 MOps/s	1.8908 MOps/s	$\color{#d91a1a}-2.35\%$
test_stack[memmap_tensor0]	48.4900μs	3.6521μs	273.8138 KOps/s	297.0321 KOps/s	$\textbf{\color{#d91a1a}-7.82\%}$
test_memmaptd_index	0.9983ms	0.2411ms	4.1468 KOps/s	4.2456 KOps/s	$\color{#d91a1a}-2.33\%$
test_memmaptd_index_astensor	0.6913ms	0.3281ms	3.0474 KOps/s	3.0927 KOps/s	$\color{#d91a1a}-1.47\%$
test_memmaptd_index_op	0.9926ms	0.6058ms	1.6507 KOps/s	1.6893 KOps/s	$\color{#d91a1a}-2.28\%$
test_serialize_model	0.1235s	0.1154s	8.6691 Ops/s	8.6140 Ops/s	$\color{#35bf28}+0.64\%$
test_serialize_model_pickle	0.4856s	0.3964s	2.5229 Ops/s	2.5177 Ops/s	$\color{#35bf28}+0.21\%$
test_serialize_weights	0.2187s	0.1290s	7.7503 Ops/s	8.5794 Ops/s	$\textbf{\color{#d91a1a}-9.66\%}$
test_serialize_weights_returnearly	0.1792s	0.1634s	6.1209 Ops/s	6.3368 Ops/s	$\color{#d91a1a}-3.41\%$
test_serialize_weights_pickle	0.5438s	0.4420s	2.2626 Ops/s	2.4678 Ops/s	$\textbf{\color{#d91a1a}-8.32\%}$
test_serialize_weights_filesystem	0.1443s	0.1389s	7.2007 Ops/s	7.0146 Ops/s	$\color{#35bf28}+2.65\%$
test_serialize_model_filesystem	0.1657s	0.1508s	6.6322 Ops/s	6.5081 Ops/s	$\color{#35bf28}+1.91\%$
test_reshape_pytree	68.9690μs	26.1336μs	38.2650 KOps/s	37.8850 KOps/s	$\color{#35bf28}+1.00\%$
test_reshape_td	66.8750μs	33.6120μs	29.7513 KOps/s	30.1513 KOps/s	$\color{#d91a1a}-1.33\%$
test_view_pytree	67.2350μs	26.5728μs	37.6325 KOps/s	38.3063 KOps/s	$\color{#d91a1a}-1.76\%$
test_view_td	0.1069ms	38.9059μs	25.7031 KOps/s	26.3102 KOps/s	$\color{#d91a1a}-2.31\%$
test_unbind_pytree	64.0290μs	29.7995μs	33.5576 KOps/s	33.0035 KOps/s	$\color{#35bf28}+1.68\%$
test_unbind_td	0.3369ms	39.7641μs	25.1483 KOps/s	25.2542 KOps/s	$\color{#d91a1a}-0.42\%$
test_split_pytree	88.7760μs	29.2467μs	34.1919 KOps/s	33.8873 KOps/s	$\color{#35bf28}+0.90\%$
test_split_td	0.1020s	57.2010μs	17.4822 KOps/s	22.0142 KOps/s	$\textbf{\color{#d91a1a}-20.59\%}$
test_add_pytree	74.3900μs	34.9598μs	28.6043 KOps/s	28.2405 KOps/s	$\color{#35bf28}+1.29\%$
test_add_td	0.3381ms	58.1187μs	17.2062 KOps/s	17.4836 KOps/s	$\color{#d91a1a}-1.59\%$
test_compile_add_one_nested[tensordict-compile]	0.1516ms	63.2179μs	15.8183 KOps/s	15.7081 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_add_one_nested[tensordict-eager]	0.3718ms	0.1732ms	5.7749 KOps/s	5.7749 KOps/s	$+0.00\%$
test_compile_add_one_nested[pytree-compile]	0.1197ms	45.8215μs	21.8238 KOps/s	22.1059 KOps/s	$\color{#d91a1a}-1.28\%$
test_compile_add_one_nested[pytree-eager]	0.1968ms	0.1186ms	8.4322 KOps/s	8.4018 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_copy_nested[tensordict-compile]	64.4700μs	27.5308μs	36.3230 KOps/s	38.5896 KOps/s	$\textbf{\color{#d91a1a}-5.87\%}$
test_compile_copy_nested[tensordict-eager]	0.1158ms	58.0168μs	17.2364 KOps/s	16.6038 KOps/s	$\color{#35bf28}+3.81\%$
test_compile_copy_nested[pytree-compile]	0.1667ms	77.5166μs	12.9005 KOps/s	13.0162 KOps/s	$\color{#d91a1a}-0.89\%$
test_compile_copy_nested[pytree-eager]	0.1787ms	65.9070μs	15.1729 KOps/s	15.1172 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_add_one_flat[tensordict-compile]	0.1837ms	0.1061ms	9.4293 KOps/s	9.4945 KOps/s	$\color{#d91a1a}-0.69\%$
test_compile_add_one_flat[tensordict-eager]	0.4273ms	0.2172ms	4.6043 KOps/s	4.6592 KOps/s	$\color{#d91a1a}-1.18\%$
test_compile_add_one_flat[tensorclass-compile]	90.6390μs	46.4299μs	21.5378 KOps/s	20.9366 KOps/s	$\color{#35bf28}+2.87\%$
test_compile_add_one_flat[tensorclass-eager]	0.4856ms	67.8004μs	14.7492 KOps/s	14.8002 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_add_one_flat[pytree-compile]	0.1943ms	0.1026ms	9.7424 KOps/s	9.8830 KOps/s	$\color{#d91a1a}-1.42\%$
test_compile_add_one_flat[pytree-eager]	0.4041ms	0.2045ms	4.8899 KOps/s	4.9549 KOps/s	$\color{#d91a1a}-1.31\%$
test_compile_add_self_flat[tensordict-eager]	0.3593ms	0.2336ms	4.2816 KOps/s	4.2667 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_add_self_flat[tensordict-compile]	0.1787ms	0.1042ms	9.5925 KOps/s	9.5126 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_add_self_flat[tensorclass-eager]	0.1527ms	64.4330μs	15.5200 KOps/s	15.7568 KOps/s	$\color{#d91a1a}-1.50\%$
test_compile_add_self_flat[tensorclass-compile]	0.1271ms	50.5829μs	19.7695 KOps/s	21.3749 KOps/s	$\textbf{\color{#d91a1a}-7.51\%}$
test_compile_add_self_flat[pytree-eager]	0.2915ms	0.1586ms	6.3062 KOps/s	6.3989 KOps/s	$\color{#d91a1a}-1.45\%$
test_compile_add_self_flat[pytree-compile]	0.1842ms	0.1038ms	9.6346 KOps/s	9.9352 KOps/s	$\color{#d91a1a}-3.03\%$
test_compile_copy_flat[tensordict-compile]	98.0230μs	21.9500μs	45.5581 KOps/s	45.8495 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_copy_flat[tensordict-eager]	0.1346ms	66.8765μs	14.9529 KOps/s	14.6211 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_copy_flat[pytree-compile]	0.1527ms	78.1710μs	12.7925 KOps/s	12.8563 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_copy_flat[pytree-eager]	0.1348ms	66.0590μs	15.1380 KOps/s	15.1138 KOps/s	$\color{#35bf28}+0.16\%$
test_compile_assign_and_add[tensordict-compile]	0.2756ms	0.2044ms	4.8915 KOps/s	4.7981 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_assign_and_add[tensordict-eager]	2.2526ms	1.3701ms	729.8888 Ops/s	748.6093 Ops/s	$\color{#d91a1a}-2.50\%$
test_compile_assign_and_add[pytree-compile]	0.2782ms	0.2005ms	4.9866 KOps/s	4.9164 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_assign_and_add[pytree-eager]	1.3076ms	0.8015ms	1.2476 KOps/s	1.2893 KOps/s	$\color{#d91a1a}-3.23\%$
test_compile_assign_and_add_stack[compile]	0.7976ms	0.4495ms	2.2247 KOps/s	2.2323 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_assign_and_add_stack[eager]	3.7106ms	2.7368ms	365.3878 Ops/s	276.8042 Ops/s	$\textbf{\color{#35bf28}+32.00\%}$
test_compile_indexing[tensor-tensordict-compile]	99.2660μs	36.1140μs	27.6901 KOps/s	25.7592 KOps/s	$\textbf{\color{#35bf28}+7.50\%}$
test_compile_indexing[tensor-tensordict-eager]	0.5094ms	33.3250μs	30.0075 KOps/s	29.4818 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_indexing[tensor-tensorclass-compile]	75.2900μs	29.0649μs	34.4058 KOps/s	33.6024 KOps/s	$\color{#35bf28}+2.39\%$
test_compile_indexing[tensor-tensorclass-eager]	84.9980μs	23.0078μs	43.4636 KOps/s	42.4209 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_indexing[tensor-pytree-compile]	83.2750μs	30.1379μs	33.1808 KOps/s	33.2051 KOps/s	$\color{#d91a1a}-0.07\%$
test_compile_indexing[tensor-pytree-eager]	78.5470μs	22.9339μs	43.6036 KOps/s	43.5311 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_indexing[slice-tensordict-compile]	0.1028ms	53.1193μs	18.8255 KOps/s	18.7972 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_indexing[slice-tensordict-eager]	0.6199ms	20.3848μs	49.0561 KOps/s	47.7578 KOps/s	$\color{#35bf28}+2.72\%$
test_compile_indexing[slice-tensorclass-compile]	98.6240μs	45.7455μs	21.8601 KOps/s	22.5628 KOps/s	$\color{#d91a1a}-3.11\%$
test_compile_indexing[slice-tensorclass-eager]	65.8130μs	18.7875μs	53.2270 KOps/s	53.7717 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_indexing[slice-pytree-compile]	0.1035ms	46.4533μs	21.5270 KOps/s	22.2218 KOps/s	$\color{#d91a1a}-3.13\%$
test_compile_indexing[slice-pytree-eager]	48.5900μs	18.7837μs	53.2377 KOps/s	53.4278 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_indexing[int-tensordict-compile]	0.1419ms	54.6688μs	18.2920 KOps/s	18.4038 KOps/s	$\color{#d91a1a}-0.61\%$
test_compile_indexing[int-tensordict-eager]	1.0606ms	20.4524μs	48.8940 KOps/s	48.8532 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_indexing[int-tensorclass-compile]	0.1022ms	45.6985μs	21.8826 KOps/s	22.3622 KOps/s	$\color{#d91a1a}-2.14\%$
test_compile_indexing[int-tensorclass-eager]	53.6300μs	18.6364μs	53.6585 KOps/s	51.9678 KOps/s	$\color{#35bf28}+3.25\%$
test_compile_indexing[int-pytree-compile]	0.1054ms	45.9891μs	21.7443 KOps/s	22.2421 KOps/s	$\color{#d91a1a}-2.24\%$
test_compile_indexing[int-pytree-eager]	61.4150μs	18.8084μs	53.1677 KOps/s	54.0100 KOps/s	$\color{#d91a1a}-1.56\%$
test_mod_add[eager]	90.6400μs	34.5178μs	28.9705 KOps/s	28.9488 KOps/s	$\color{#35bf28}+0.07\%$
test_mod_add[compile]	0.1093ms	48.6412μs	20.5587 KOps/s	20.6078 KOps/s	$\color{#d91a1a}-0.24\%$
test_mod_add[compile-overhead]	0.1026ms	47.5864μs	21.0144 KOps/s	19.9927 KOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_mod_wrap[eager]	0.3341ms	0.2218ms	4.5084 KOps/s	4.3256 KOps/s	$\color{#35bf28}+4.23\%$
test_mod_wrap[compile]	0.3141ms	0.2070ms	4.8316 KOps/s	4.8409 KOps/s	$\color{#d91a1a}-0.19\%$
test_mod_wrap[compile-overhead]	0.4440ms	0.2060ms	4.8535 KOps/s	4.8151 KOps/s	$\color{#35bf28}+0.80\%$
test_mod_wrap_and_backward[eager]	12.4025ms	10.8510ms	92.1575 Ops/s	85.7004 Ops/s	$\textbf{\color{#35bf28}+7.53\%}$
test_mod_wrap_and_backward[compile]	15.9639ms	11.2601ms	88.8093 Ops/s	82.0373 Ops/s	$\textbf{\color{#35bf28}+8.25\%}$
test_mod_wrap_and_backward[compile-overhead]	12.3284ms	10.8552ms	92.1220 Ops/s	90.8955 Ops/s	$\color{#35bf28}+1.35\%$
test_seq_add[eager]	0.1993ms	0.1164ms	8.5914 KOps/s	8.5513 KOps/s	$\color{#35bf28}+0.47\%$
test_seq_add[compile]	0.1193ms	65.9713μs	15.1581 KOps/s	15.4088 KOps/s	$\color{#d91a1a}-1.63\%$
test_seq_add[compile-overhead]	0.1414ms	62.5103μs	15.9974 KOps/s	15.7535 KOps/s	$\color{#35bf28}+1.55\%$
test_seq_wrap[eager]	0.8289ms	0.4411ms	2.2671 KOps/s	2.2089 KOps/s	$\color{#35bf28}+2.63\%$
test_seq_wrap[compile]	0.3905ms	0.2290ms	4.3672 KOps/s	4.2836 KOps/s	$\color{#35bf28}+1.95\%$
test_seq_wrap[compile-overhead]	0.4227ms	0.2293ms	4.3620 KOps/s	4.4044 KOps/s	$\color{#d91a1a}-0.96\%$
test_func_call_runtime[False-eager]	0.8244ms	0.5374ms	1.8606 KOps/s	1.8802 KOps/s	$\color{#d91a1a}-1.04\%$
test_func_call_runtime[False-compile]	0.6264ms	0.4346ms	2.3007 KOps/s	2.3560 KOps/s	$\color{#d91a1a}-2.35\%$
test_func_call_runtime[False-compile-overhead]	0.5799ms	0.4316ms	2.3172 KOps/s	2.3429 KOps/s	$\color{#d91a1a}-1.10\%$
test_func_call_runtime[True-eager]	1.1125ms	0.7490ms	1.3351 KOps/s	1.3218 KOps/s	$\color{#35bf28}+1.00\%$
test_func_call_runtime[True-compile]	0.8194ms	0.4759ms	2.1011 KOps/s	2.1456 KOps/s	$\color{#d91a1a}-2.08\%$
test_func_call_runtime[True-compile-overhead]	0.6185ms	0.4771ms	2.0960 KOps/s	2.1380 KOps/s	$\color{#d91a1a}-1.96\%$
test_func_call_cm_runtime[False-eager]	0.9392ms	0.5367ms	1.8631 KOps/s	1.8515 KOps/s	$\color{#35bf28}+0.62\%$
test_func_call_cm_runtime[False-compile]	0.8217ms	0.4318ms	2.3158 KOps/s	2.3602 KOps/s	$\color{#d91a1a}-1.88\%$
test_func_call_cm_runtime[False-compile-overhead]	0.9381ms	0.4348ms	2.2999 KOps/s	2.3372 KOps/s	$\color{#d91a1a}-1.59\%$
test_func_call_cm_runtime[True-eager]	1.4966ms	0.9018ms	1.1089 KOps/s	1.1008 KOps/s	$\color{#35bf28}+0.73\%$
test_func_call_cm_runtime[True-compile]	0.6603ms	0.4988ms	2.0048 KOps/s	2.0245 KOps/s	$\color{#d91a1a}-0.97\%$
test_func_call_cm_runtime[True-compile-overhead]	0.9056ms	0.5097ms	1.9619 KOps/s	2.0203 KOps/s	$\color{#d91a1a}-2.89\%$
test_vmap_func_call_cm_runtime[eager]	2.3287ms	1.9065ms	524.5246 Ops/s	515.1495 Ops/s	$\color{#35bf28}+1.82\%$
test_vmap_func_call_cm_runtime[compile]	0.8781ms	0.5337ms	1.8736 KOps/s	1.9142 KOps/s	$\color{#d91a1a}-2.12\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.7070ms	0.5315ms	1.8816 KOps/s	1.9077 KOps/s	$\color{#d91a1a}-1.37\%$
test_distributed	0.2861ms	0.1254ms	7.9749 KOps/s	7.7465 KOps/s	$\color{#35bf28}+2.95\%$
test_tdmodule	67.3960μs	26.2711μs	38.0646 KOps/s	37.8720 KOps/s	$\color{#35bf28}+0.51\%$
test_tdmodule_dispatch	83.8870μs	48.4150μs	20.6548 KOps/s	20.4409 KOps/s	$\color{#35bf28}+1.05\%$
test_tdseq	58.9400μs	29.3392μs	34.0841 KOps/s	33.9893 KOps/s	$\color{#35bf28}+0.28\%$
test_tdseq_dispatch	0.1028ms	53.5793μs	18.6639 KOps/s	18.6660 KOps/s	$\color{#d91a1a}-0.01\%$
test_instantiation_functorch	1.9362ms	1.5263ms	655.1930 Ops/s	651.7040 Ops/s	$\color{#35bf28}+0.54\%$
test_exec_functorch	0.3349ms	0.1771ms	5.6470 KOps/s	5.6447 KOps/s	$\color{#35bf28}+0.04\%$
test_exec_functional_call	0.4188ms	0.1721ms	5.8100 KOps/s	5.8903 KOps/s	$\color{#d91a1a}-1.36\%$
test_exec_td_decorator	0.4669ms	0.2250ms	4.4442 KOps/s	4.3231 KOps/s	$\color{#35bf28}+2.80\%$
test_vmap_mlp_speed_decorator[True-True]	0.8840ms	0.6560ms	1.5243 KOps/s	1.5307 KOps/s	$\color{#d91a1a}-0.42\%$
test_vmap_mlp_speed_decorator[True-False]	1.1651ms	0.6599ms	1.5153 KOps/s	1.5298 KOps/s	$\color{#d91a1a}-0.94\%$
test_vmap_mlp_speed_decorator[False-True]	0.8833ms	0.5275ms	1.8956 KOps/s	1.8938 KOps/s	$\color{#35bf28}+0.10\%$
test_vmap_mlp_speed_decorator[False-False]	0.9680ms	0.5325ms	1.8778 KOps/s	1.8959 KOps/s	$\color{#d91a1a}-0.95\%$
test_to_module_speed[True]	2.0995ms	1.3362ms	748.4002 Ops/s	748.2350 Ops/s	$\color{#35bf28}+0.02\%$
test_to_module_speed[False]	1.8729ms	1.2815ms	780.3257 Ops/s	764.5308 Ops/s	$\color{#35bf28}+2.07\%$
test_tc_init	99.5460μs	45.5385μs	21.9594 KOps/s	21.6985 KOps/s	$\color{#35bf28}+1.20\%$
test_tc_init_nested	0.2011ms	90.9191μs	10.9988 KOps/s	10.6704 KOps/s	$\color{#35bf28}+3.08\%$
test_tc_first_layer_tensor	16.1700μs	1.5122μs	661.2980 KOps/s	630.5702 KOps/s	$\color{#35bf28}+4.87\%$
test_tc_first_layer_nontensor	25.7480μs	4.6457μs	215.2524 KOps/s	208.2209 KOps/s	$\color{#35bf28}+3.38\%$
test_tc_second_layer_tensor	38.4120μs	2.7959μs	357.6620 KOps/s	339.1820 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_tc_second_layer_nontensor	26.8600μs	5.9594μs	167.8025 KOps/s	160.5988 KOps/s	$\color{#35bf28}+4.49\%$
test_unbind	0.2377s	13.4219ms	74.5052 Ops/s	77.4353 Ops/s	$\color{#d91a1a}-3.78\%$
test_full_like	8.8634ms	7.5293ms	132.8149 Ops/s	80.6383 Ops/s	$\textbf{\color{#35bf28}+64.70\%}$
test_zeros_like	3.3327ms	2.8845ms	346.6766 Ops/s	141.0201 Ops/s	$\textbf{\color{#35bf28}+145.83\%}$
test_ones_like	4.0809ms	3.4856ms	286.8965 Ops/s	128.9975 Ops/s	$\textbf{\color{#35bf28}+122.40\%}$
test_clone	6.4796ms	5.4119ms	184.7795 Ops/s	105.6547 Ops/s	$\textbf{\color{#35bf28}+74.89\%}$
test_squeeze	62.4960μs	12.1489μs	82.3119 KOps/s	80.6941 KOps/s	$\color{#35bf28}+2.00\%$
test_unsqueeze	0.3172ms	92.4545μs	10.8161 KOps/s	10.3921 KOps/s	$\color{#35bf28}+4.08\%$
test_split	0.3261ms	0.1919ms	5.2116 KOps/s	5.0228 KOps/s	$\color{#35bf28}+3.76\%$
test_permute	0.3127ms	0.1993ms	5.0178 KOps/s	4.8397 KOps/s	$\color{#35bf28}+3.68\%$
test_stack	31.3698ms	25.8905ms	38.6241 Ops/s	38.9169 Ops/s	$\color{#d91a1a}-0.75\%$
test_cat	34.0779ms	25.9056ms	38.6017 Ops/s	40.4320 Ops/s	$\color{#d91a1a}-4.53\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}39$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	25.3700μs	11.2249μs	89.0875 KOps/s	75.9002 KOps/s	$\textbf{\color{#35bf28}+17.37\%}$
test_plain_set_stack_nested	0.4077ms	11.4478μs	87.3530 KOps/s	74.7761 KOps/s	$\textbf{\color{#35bf28}+16.82\%}$
test_plain_set_nested_inplace	39.4200μs	12.4228μs	80.4970 KOps/s	69.8819 KOps/s	$\textbf{\color{#35bf28}+15.19\%}$
test_plain_set_stack_nested_inplace	0.4293ms	12.3987μs	80.6537 KOps/s	69.1866 KOps/s	$\textbf{\color{#35bf28}+16.57\%}$
test_items	0.4026ms	2.9113μs	343.4857 KOps/s	341.3482 KOps/s	$\color{#35bf28}+0.63\%$
test_items_nested	0.4291ms	0.3618ms	2.7643 KOps/s	2.8068 KOps/s	$\color{#d91a1a}-1.51\%$
test_items_nested_locked	0.7649ms	0.3600ms	2.7775 KOps/s	2.7914 KOps/s	$\color{#d91a1a}-0.50\%$
test_items_nested_leaf	0.4724ms	57.9761μs	17.2485 KOps/s	17.2981 KOps/s	$\color{#d91a1a}-0.29\%$
test_items_stack_nested	0.7727ms	0.3586ms	2.7890 KOps/s	2.7918 KOps/s	$\color{#d91a1a}-0.10\%$
test_items_stack_nested_leaf	0.4670ms	58.1345μs	17.2015 KOps/s	17.0443 KOps/s	$\color{#35bf28}+0.92\%$
test_items_stack_nested_locked	0.7787ms	0.3575ms	2.7970 KOps/s	2.7669 KOps/s	$\color{#35bf28}+1.09\%$
test_keys	34.5000μs	3.4207μs	292.3393 KOps/s	290.6686 KOps/s	$\color{#35bf28}+0.57\%$
test_keys_nested	0.4874ms	89.0002μs	11.2359 KOps/s	11.3211 KOps/s	$\color{#d91a1a}-0.75\%$
test_keys_nested_locked	0.7659ms	94.4268μs	10.5902 KOps/s	10.6713 KOps/s	$\color{#d91a1a}-0.76\%$
test_keys_nested_leaf	0.4779ms	79.5207μs	12.5753 KOps/s	12.6183 KOps/s	$\color{#d91a1a}-0.34\%$
test_keys_stack_nested	0.4893ms	88.9386μs	11.2437 KOps/s	11.2776 KOps/s	$\color{#d91a1a}-0.30\%$
test_keys_stack_nested_leaf	0.4730ms	79.4379μs	12.5884 KOps/s	12.4550 KOps/s	$\color{#35bf28}+1.07\%$
test_keys_stack_nested_locked	0.4969ms	95.1667μs	10.5079 KOps/s	10.5530 KOps/s	$\color{#d91a1a}-0.43\%$
test_values	67.3343μs	0.8720μs	1.1468 MOps/s	1.1749 MOps/s	$\color{#d91a1a}-2.39\%$
test_values_nested	66.3310μs	38.0619μs	26.2730 KOps/s	26.4370 KOps/s	$\color{#d91a1a}-0.62\%$
test_values_nested_locked	0.4363ms	39.2039μs	25.5076 KOps/s	25.4258 KOps/s	$\color{#35bf28}+0.32\%$
test_values_nested_leaf	0.4405ms	42.3604μs	23.6070 KOps/s	23.6302 KOps/s	$\color{#d91a1a}-0.10\%$
test_values_stack_nested	0.4403ms	37.9768μs	26.3319 KOps/s	26.1810 KOps/s	$\color{#35bf28}+0.58\%$
test_values_stack_nested_leaf	0.4463ms	42.9337μs	23.2917 KOps/s	23.3687 KOps/s	$\color{#d91a1a}-0.33\%$
test_values_stack_nested_locked	0.4500ms	39.7055μs	25.1854 KOps/s	25.0091 KOps/s	$\color{#35bf28}+0.70\%$
test_membership	20.2288μs	0.5139μs	1.9460 MOps/s	2.0115 MOps/s	$\color{#d91a1a}-3.25\%$
test_membership_nested	17.0200μs	1.9601μs	510.1715 KOps/s	511.4579 KOps/s	$\color{#d91a1a}-0.25\%$
test_membership_nested_leaf	20.5000μs	1.9660μs	508.6566 KOps/s	520.2370 KOps/s	$\color{#d91a1a}-2.23\%$
test_membership_stacked_nested	37.2510μs	2.0612μs	485.1549 KOps/s	495.4854 KOps/s	$\color{#d91a1a}-2.08\%$
test_membership_stacked_nested_leaf	0.4166ms	2.0696μs	483.1761 KOps/s	481.5274 KOps/s	$\color{#35bf28}+0.34\%$
test_membership_nested_last	27.6710μs	3.0686μs	325.8848 KOps/s	329.7943 KOps/s	$\color{#d91a1a}-1.19\%$
test_membership_nested_leaf_last	0.4048ms	3.0583μs	326.9740 KOps/s	328.3270 KOps/s	$\color{#d91a1a}-0.41\%$
test_membership_stacked_nested_last	36.2600μs	3.0512μs	327.7430 KOps/s	327.1713 KOps/s	$\color{#35bf28}+0.17\%$
test_membership_stacked_nested_leaf_last	0.4015ms	3.0254μs	330.5322 KOps/s	330.0406 KOps/s	$\color{#35bf28}+0.15\%$
test_nested_getleaf	45.5710μs	6.0011μs	166.6364 KOps/s	164.5054 KOps/s	$\color{#35bf28}+1.30\%$
test_nested_get	51.7610μs	5.8339μs	171.4108 KOps/s	172.9944 KOps/s	$\color{#d91a1a}-0.92\%$
test_stacked_getleaf	39.1610μs	6.1183μs	163.4440 KOps/s	163.2870 KOps/s	$\color{#35bf28}+0.10\%$
test_stacked_get	60.4610μs	5.7121μs	175.0665 KOps/s	173.0155 KOps/s	$\color{#35bf28}+1.19\%$
test_nested_getitemleaf	44.3410μs	6.4497μs	155.0469 KOps/s	157.0839 KOps/s	$\color{#d91a1a}-1.30\%$
test_nested_getitem	0.3946ms	6.0653μs	164.8736 KOps/s	164.3385 KOps/s	$\color{#35bf28}+0.33\%$
test_stacked_getitemleaf	0.4044ms	6.3855μs	156.6038 KOps/s	155.3550 KOps/s	$\color{#35bf28}+0.80\%$
test_stacked_getitem	30.4910μs	5.9488μs	168.0999 KOps/s	162.9069 KOps/s	$\color{#35bf28}+3.19\%$
test_lock_nested	9.4806ms	0.3780ms	2.6455 KOps/s	2.6323 KOps/s	$\color{#35bf28}+0.50\%$
test_lock_stack_nested	0.4516ms	0.3378ms	2.9606 KOps/s	2.9019 KOps/s	$\color{#35bf28}+2.02\%$
test_unlock_nested	0.6230ms	0.3095ms	3.2307 KOps/s	3.2101 KOps/s	$\color{#35bf28}+0.64\%$
test_unlock_stack_nested	0.3156ms	0.2772ms	3.6078 KOps/s	3.5538 KOps/s	$\color{#35bf28}+1.52\%$
test_flatten_speed	0.1240ms	74.7878μs	13.3712 KOps/s	13.3655 KOps/s	$\color{#35bf28}+0.04\%$
test_unflatten_speed	0.4368ms	0.3149ms	3.1755 KOps/s	3.1811 KOps/s	$\color{#d91a1a}-0.18\%$
test_common_ops	1.6468ms	0.5639ms	1.7734 KOps/s	1.5602 KOps/s	$\textbf{\color{#35bf28}+13.67\%}$
test_creation	0.1091ms	1.7262μs	579.3187 KOps/s	574.7610 KOps/s	$\color{#35bf28}+0.79\%$
test_creation_empty	29.3600μs	6.3145μs	158.3657 KOps/s	98.6531 KOps/s	$\textbf{\color{#35bf28}+60.53\%}$
test_creation_nested_1	31.8310μs	7.9753μs	125.3871 KOps/s	84.7186 KOps/s	$\textbf{\color{#35bf28}+48.00\%}$
test_creation_nested_2	39.4310μs	10.5178μs	95.0769 KOps/s	69.1398 KOps/s	$\textbf{\color{#35bf28}+37.51\%}$
test_clone	78.6320μs	10.2966μs	97.1191 KOps/s	95.1704 KOps/s	$\color{#35bf28}+2.05\%$
test_getitem[int]	1.7525ms	10.4154μs	96.0121 KOps/s	94.5602 KOps/s	$\color{#35bf28}+1.54\%$
test_getitem[slice_int]	0.1113ms	20.1600μs	49.6031 KOps/s	49.8557 KOps/s	$\color{#d91a1a}-0.51\%$
test_getitem[range]	0.1294ms	35.3945μs	28.2530 KOps/s	28.0091 KOps/s	$\color{#35bf28}+0.87\%$
test_getitem[tuple]	0.1091ms	17.7938μs	56.1995 KOps/s	56.3563 KOps/s	$\color{#d91a1a}-0.28\%$
test_getitem[list]	0.2356ms	31.4393μs	31.8073 KOps/s	30.9168 KOps/s	$\color{#35bf28}+2.88\%$
test_setitem_dim[int]	40.7800μs	18.8518μs	53.0454 KOps/s	53.7195 KOps/s	$\color{#d91a1a}-1.25\%$
test_setitem_dim[slice_int]	0.1556ms	37.6611μs	26.5526 KOps/s	26.9558 KOps/s	$\color{#d91a1a}-1.50\%$
test_setitem_dim[range]	74.4810μs	51.2548μs	19.5104 KOps/s	19.4689 KOps/s	$\color{#35bf28}+0.21\%$
test_setitem_dim[tuple]	52.7710μs	30.7481μs	32.5223 KOps/s	32.9199 KOps/s	$\color{#d91a1a}-1.21\%$
test_setitem	88.1610μs	13.6192μs	73.4259 KOps/s	62.3792 KOps/s	$\textbf{\color{#35bf28}+17.71\%}$
test_set	95.4120μs	13.1904μs	75.8127 KOps/s	64.4005 KOps/s	$\textbf{\color{#35bf28}+17.72\%}$
test_set_shared	1.5200ms	0.1501ms	6.6627 KOps/s	6.6530 KOps/s	$\color{#35bf28}+0.15\%$
test_update	0.3640ms	14.8810μs	67.1998 KOps/s	51.5770 KOps/s	$\textbf{\color{#35bf28}+30.29\%}$
test_update_nested	91.8020μs	20.0382μs	49.9047 KOps/s	39.8128 KOps/s	$\textbf{\color{#35bf28}+25.35\%}$
test_update__nested	1.0424ms	25.3175μs	39.4984 KOps/s	39.9973 KOps/s	$\color{#d91a1a}-1.25\%$
test_set_nested	95.5910μs	14.3034μs	69.9132 KOps/s	59.8384 KOps/s	$\textbf{\color{#35bf28}+16.84\%}$
test_set_nested_new	0.1031ms	16.8510μs	59.3437 KOps/s	51.9967 KOps/s	$\textbf{\color{#35bf28}+14.13\%}$
test_select	0.1031ms	28.4268μs	35.1781 KOps/s	32.3066 KOps/s	$\textbf{\color{#35bf28}+8.89\%}$
test_select_nested	93.8020μs	42.5086μs	23.5247 KOps/s	22.6740 KOps/s	$\color{#35bf28}+3.75\%$
test_exclude_nested	98.9610μs	61.2569μs	16.3247 KOps/s	16.1569 KOps/s	$\color{#35bf28}+1.04\%$
test_empty[True]	0.7182ms	0.2942ms	3.3995 KOps/s	3.4173 KOps/s	$\color{#d91a1a}-0.52\%$
test_empty[False]	3.4410μs	0.8252μs	1.2118 MOps/s	1.2129 MOps/s	$\color{#d91a1a}-0.09\%$
test_to	85.7510μs	54.8466μs	18.2327 KOps/s	17.0214 KOps/s	$\textbf{\color{#35bf28}+7.12\%}$
test_to_nonblocking	0.1952ms	46.3932μs	21.5549 KOps/s	21.4370 KOps/s	$\color{#35bf28}+0.55\%$
test_unbind_speed	0.8328ms	0.2342ms	4.2707 KOps/s	4.2174 KOps/s	$\color{#35bf28}+1.26\%$
test_unbind_speed_stack0	0.2918ms	0.2319ms	4.3123 KOps/s	4.2719 KOps/s	$\color{#35bf28}+0.95\%$
test_unbind_speed_stack1	94.3248ms	0.6691ms	1.4946 KOps/s	1.4889 KOps/s	$\color{#35bf28}+0.38\%$
test_split	96.6717ms	1.6988ms	588.6498 Ops/s	635.2133 Ops/s	$\textbf{\color{#d91a1a}-7.33\%}$
test_chunk	1.5317ms	1.4190ms	704.7175 Ops/s	582.6259 Ops/s	$\textbf{\color{#35bf28}+20.96\%}$
test_consolidate[False-None]	99.7775ms	2.8710ms	348.3139 Ops/s	380.1582 Ops/s	$\textbf{\color{#d91a1a}-8.38\%}$
test_consolidate[default-None]	1.7093ms	1.6342ms	611.9359 Ops/s	610.1202 Ops/s	$\color{#35bf28}+0.30\%$
test_consolidate[reduce-overhead-None]	1.7681ms	1.6760ms	596.6525 Ops/s	596.5023 Ops/s	$\color{#35bf28}+0.03\%$
test_consolidate_njt[False-None]	6.8008ms	6.2438ms	160.1595 Ops/s	157.5777 Ops/s	$\color{#35bf28}+1.64\%$
test_to[False-False-None]	1.8732ms	1.6962ms	589.5660 Ops/s	594.5990 Ops/s	$\color{#d91a1a}-0.85\%$
test_to[True-False-None]	1.7567ms	1.2896ms	775.4274 Ops/s	789.5375 Ops/s	$\color{#d91a1a}-1.79\%$
test_to[within-False-None]	4.2970ms	3.9771ms	251.4415 Ops/s	246.1598 Ops/s	$\color{#35bf28}+2.15\%$
test_to[True-default-None]	5.4769ms	5.1511ms	194.1334 Ops/s	194.0557 Ops/s	$\color{#35bf28}+0.04\%$
test_to_njt[False-False-None]	6.8022ms	6.7280ms	148.6336 Ops/s	146.8174 Ops/s	$\color{#35bf28}+1.24\%$
test_to_njt[True-False-None]	5.5193ms	5.2622ms	190.0356 Ops/s	188.6348 Ops/s	$\color{#35bf28}+0.74\%$
test_to_njt[within-False-None]	12.1520ms	11.6356ms	85.9433 Ops/s	83.3940 Ops/s	$\color{#35bf28}+3.06\%$
test_creation[device0]	0.6167ms	81.9100μs	12.2085 KOps/s	12.2935 KOps/s	$\color{#d91a1a}-0.69\%$
test_creation_from_tensor	0.5064ms	83.7087μs	11.9462 KOps/s	12.0569 KOps/s	$\color{#d91a1a}-0.92\%$
test_add_one[memmap_tensor0]	0.4099ms	6.5701μs	152.2050 KOps/s	154.4669 KOps/s	$\color{#d91a1a}-1.46\%$
test_contiguous[memmap_tensor0]	1.9305μs	0.3973μs	2.5172 MOps/s	2.4705 MOps/s	$\color{#35bf28}+1.89\%$
test_stack[memmap_tensor0]	22.7510μs	4.1533μs	240.7745 KOps/s	239.1793 KOps/s	$\color{#35bf28}+0.67\%$
test_memmaptd_index	1.7298ms	0.2461ms	4.0633 KOps/s	4.1012 KOps/s	$\color{#d91a1a}-0.92\%$
test_memmaptd_index_astensor	0.8251ms	0.3075ms	3.2518 KOps/s	3.2974 KOps/s	$\color{#d91a1a}-1.38\%$
test_memmaptd_index_op	0.9693ms	0.5347ms	1.8701 KOps/s	1.6873 KOps/s	$\textbf{\color{#35bf28}+10.83\%}$
test_serialize_model	0.1315s	0.1308s	7.6437 Ops/s	7.6190 Ops/s	$\color{#35bf28}+0.32\%$
test_serialize_model_pickle	1.3601s	1.2179s	0.8211 Ops/s	0.8251 Ops/s	$\color{#d91a1a}-0.48\%$
test_serialize_weights	0.1314s	0.1305s	7.6604 Ops/s	7.6895 Ops/s	$\color{#d91a1a}-0.38\%$
test_serialize_weights_returnearly	0.3324s	53.3915ms	18.7296 Ops/s	15.7178 Ops/s	$\textbf{\color{#35bf28}+19.16\%}$
test_serialize_weights_pickle	1.3731s	1.2181s	0.8210 Ops/s	0.8231 Ops/s	$\color{#d91a1a}-0.27\%$
test_reshape_pytree	54.9910μs	21.7697μs	45.9354 KOps/s	45.4610 KOps/s	$\color{#35bf28}+1.04\%$
test_reshape_td	67.0910μs	26.2266μs	38.1292 KOps/s	36.1645 KOps/s	$\textbf{\color{#35bf28}+5.43\%}$
test_view_pytree	61.7610μs	21.6052μs	46.2852 KOps/s	46.7027 KOps/s	$\color{#d91a1a}-0.89\%$
test_view_td	0.1405ms	29.6005μs	33.7832 KOps/s	32.0473 KOps/s	$\textbf{\color{#35bf28}+5.42\%}$
test_unbind_pytree	63.3910μs	27.4238μs	36.4647 KOps/s	36.1925 KOps/s	$\color{#35bf28}+0.75\%$
test_unbind_td	0.7646ms	35.4841μs	28.1817 KOps/s	26.2170 KOps/s	$\textbf{\color{#35bf28}+7.49\%}$
test_split_pytree	74.5610μs	29.7783μs	33.5815 KOps/s	34.0436 KOps/s	$\color{#d91a1a}-1.36\%$
test_split_td	0.9668ms	37.6509μs	26.5598 KOps/s	26.7961 KOps/s	$\color{#d91a1a}-0.88\%$
test_add_pytree	75.1110μs	33.4730μs	29.8748 KOps/s	30.2324 KOps/s	$\color{#d91a1a}-1.18\%$
test_add_td	0.4435ms	43.6074μs	22.9319 KOps/s	20.2545 KOps/s	$\textbf{\color{#35bf28}+13.22\%}$
test_compile_add_one_nested[tensordict-compile]	0.1800ms	0.1247ms	8.0216 KOps/s	8.1753 KOps/s	$\color{#d91a1a}-1.88\%$
test_compile_add_one_nested[tensordict-eager]	0.2262ms	0.1315ms	7.6030 KOps/s	7.6478 KOps/s	$\color{#d91a1a}-0.59\%$
test_compile_add_one_nested[pytree-compile]	0.1391ms	93.7941μs	10.6616 KOps/s	10.5652 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_add_one_nested[pytree-eager]	0.2733ms	0.1485ms	6.7362 KOps/s	6.8535 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_copy_nested[tensordict-compile]	97.5820μs	21.8380μs	45.7917 KOps/s	45.4833 KOps/s	$\color{#35bf28}+0.68\%$
test_compile_copy_nested[tensordict-eager]	70.9510μs	28.7526μs	34.7795 KOps/s	33.8247 KOps/s	$\color{#35bf28}+2.82\%$
test_compile_copy_nested[pytree-compile]	0.4813ms	63.1677μs	15.8309 KOps/s	15.6077 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_copy_nested[pytree-eager]	96.0910μs	49.1460μs	20.3475 KOps/s	20.3020 KOps/s	$\color{#35bf28}+0.22\%$
test_compile_add_one_flat[tensordict-compile]	0.1928ms	0.1388ms	7.2029 KOps/s	7.0113 KOps/s	$\color{#35bf28}+2.73\%$
test_compile_add_one_flat[tensordict-eager]	0.6495ms	0.2176ms	4.5963 KOps/s	4.6157 KOps/s	$\color{#d91a1a}-0.42\%$
test_compile_add_one_flat[tensorclass-compile]	0.1587ms	0.1013ms	9.8749 KOps/s	10.3995 KOps/s	$\textbf{\color{#d91a1a}-5.04\%}$
test_compile_add_one_flat[tensorclass-eager]	0.1369ms	56.6995μs	17.6368 KOps/s	18.3783 KOps/s	$\color{#d91a1a}-4.03\%$
test_compile_add_one_flat[pytree-compile]	0.2443ms	0.1333ms	7.4996 KOps/s	7.4279 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_add_one_flat[pytree-eager]	0.5964ms	0.4711ms	2.1228 KOps/s	2.1598 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_add_self_flat[tensordict-eager]	0.4082ms	0.2630ms	3.8030 KOps/s	3.8135 KOps/s	$\color{#d91a1a}-0.28\%$
test_compile_add_self_flat[tensordict-compile]	0.2054ms	0.1483ms	6.7422 KOps/s	7.0437 KOps/s	$\color{#d91a1a}-4.28\%$
test_compile_add_self_flat[tensorclass-eager]	0.2070ms	68.0047μs	14.7049 KOps/s	14.7840 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_add_self_flat[tensorclass-compile]	0.1574ms	98.2882μs	10.1742 KOps/s	10.2789 KOps/s	$\color{#d91a1a}-1.02\%$
test_compile_add_self_flat[pytree-eager]	0.5655ms	0.4016ms	2.4903 KOps/s	2.5362 KOps/s	$\color{#d91a1a}-1.81\%$
test_compile_add_self_flat[pytree-compile]	0.1926ms	0.1338ms	7.4760 KOps/s	7.5024 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_copy_flat[tensordict-compile]	82.6820μs	18.8044μs	53.1791 KOps/s	56.8736 KOps/s	$\textbf{\color{#d91a1a}-6.50\%}$
test_compile_copy_flat[tensordict-eager]	74.2510μs	30.9811μs	32.2777 KOps/s	31.5116 KOps/s	$\color{#35bf28}+2.43\%$
test_compile_copy_flat[pytree-compile]	0.2033ms	69.0521μs	14.4818 KOps/s	14.4642 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_copy_flat[pytree-eager]	94.4910μs	50.9989μs	19.6083 KOps/s	19.6018 KOps/s	$\color{#35bf28}+0.03\%$
test_compile_assign_and_add[tensordict-compile]	1.5752ms	0.3797ms	2.6337 KOps/s	2.2670 KOps/s	$\textbf{\color{#35bf28}+16.18\%}$
test_compile_assign_and_add[tensordict-eager]	2.7215ms	2.6048ms	383.9085 Ops/s	393.5432 Ops/s	$\color{#d91a1a}-2.45\%$
test_compile_assign_and_add[pytree-compile]	1.5483ms	0.4219ms	2.3705 KOps/s	2.2875 KOps/s	$\color{#35bf28}+3.63\%$
test_compile_assign_and_add[pytree-eager]	2.7288ms	2.5707ms	389.0025 Ops/s	395.8501 Ops/s	$\color{#d91a1a}-1.73\%$
test_compile_indexing[tensor-tensordict-compile]	0.5945ms	0.1122ms	8.9134 KOps/s	8.9962 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_indexing[tensor-tensordict-eager]	0.5536ms	76.1705μs	13.1284 KOps/s	12.1970 KOps/s	$\textbf{\color{#35bf28}+7.64\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.5052ms	0.1080ms	9.2557 KOps/s	9.2997 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1257ms	65.9402μs	15.1653 KOps/s	14.4496 KOps/s	$\color{#35bf28}+4.95\%$
test_compile_indexing[tensor-pytree-compile]	0.1662ms	0.1078ms	9.2790 KOps/s	9.5750 KOps/s	$\color{#d91a1a}-3.09\%$
test_compile_indexing[tensor-pytree-eager]	0.1848ms	70.1893μs	14.2472 KOps/s	15.3147 KOps/s	$\textbf{\color{#d91a1a}-6.97\%}$
test_compile_indexing[slice-tensordict-compile]	0.1637ms	0.1030ms	9.7065 KOps/s	10.2110 KOps/s	$\color{#d91a1a}-4.94\%$
test_compile_indexing[slice-tensordict-eager]	0.1418ms	16.7187μs	59.8134 KOps/s	59.0632 KOps/s	$\color{#35bf28}+1.27\%$
test_compile_indexing[slice-tensorclass-compile]	0.1376ms	98.6446μs	10.1374 KOps/s	10.5915 KOps/s	$\color{#d91a1a}-4.29\%$
test_compile_indexing[slice-tensorclass-eager]	99.6720μs	15.4475μs	64.7354 KOps/s	59.4909 KOps/s	$\textbf{\color{#35bf28}+8.82\%}$
test_compile_indexing[slice-pytree-compile]	0.5134ms	93.9089μs	10.6486 KOps/s	10.5371 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_indexing[slice-pytree-eager]	0.4033ms	15.5857μs	64.1613 KOps/s	64.8637 KOps/s	$\color{#d91a1a}-1.08\%$
test_compile_indexing[int-tensordict-compile]	0.5042ms	98.3138μs	10.1715 KOps/s	10.1401 KOps/s	$\color{#35bf28}+0.31\%$
test_compile_indexing[int-tensordict-eager]	0.6567ms	16.2725μs	61.4532 KOps/s	61.3075 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_indexing[int-tensorclass-compile]	0.5043ms	93.9307μs	10.6461 KOps/s	10.5252 KOps/s	$\color{#35bf28}+1.15\%$
test_compile_indexing[int-tensorclass-eager]	76.0110μs	15.4304μs	64.8070 KOps/s	65.2716 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_indexing[int-pytree-compile]	0.5421ms	93.6742μs	10.6753 KOps/s	10.4833 KOps/s	$\color{#35bf28}+1.83\%$
test_compile_indexing[int-pytree-eager]	0.4186ms	15.5464μs	64.3235 KOps/s	64.6702 KOps/s	$\color{#d91a1a}-0.54\%$
test_mod_add[eager]	0.4705ms	35.4487μs	28.2098 KOps/s	26.0567 KOps/s	$\textbf{\color{#35bf28}+8.26\%}$
test_mod_add[compile]	0.3533ms	77.8096μs	12.8519 KOps/s	12.5972 KOps/s	$\color{#35bf28}+2.02\%$
test_mod_add[compile-overhead]	0.3236ms	0.1630ms	6.1346 KOps/s	5.7670 KOps/s	$\textbf{\color{#35bf28}+6.37\%}$
test_mod_wrap[eager]	0.6679ms	0.2429ms	4.1176 KOps/s	4.0331 KOps/s	$\color{#35bf28}+2.09\%$
test_mod_wrap[compile]	0.3417ms	0.2729ms	3.6644 KOps/s	3.5903 KOps/s	$\color{#35bf28}+2.06\%$
test_mod_wrap[compile-overhead]	7.0426ms	3.7349ms	267.7482 Ops/s	266.5366 Ops/s	$\color{#35bf28}+0.45\%$
test_mod_wrap_and_backward[eager]	1.7076ms	1.3449ms	743.5351 Ops/s	691.7374 Ops/s	$\textbf{\color{#35bf28}+7.49\%}$
test_mod_wrap_and_backward[compile]	1.3603ms	1.2252ms	816.1986 Ops/s	745.0939 Ops/s	$\textbf{\color{#35bf28}+9.54\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3505ms	0.9096ms	1.0993 KOps/s	960.1592 Ops/s	$\textbf{\color{#35bf28}+14.50\%}$
test_seq_add[eager]	0.1800ms	0.1073ms	9.3231 KOps/s	8.5519 KOps/s	$\textbf{\color{#35bf28}+9.02\%}$
test_seq_add[compile]	0.2606ms	88.2349μs	11.3334 KOps/s	11.7096 KOps/s	$\color{#d91a1a}-3.21\%$
test_seq_add[compile-overhead]	0.1728ms	0.1320ms	7.5785 KOps/s	7.9616 KOps/s	$\color{#d91a1a}-4.81\%$
test_seq_wrap[eager]	0.4759ms	0.4039ms	2.4760 KOps/s	2.3632 KOps/s	$\color{#35bf28}+4.77\%$
test_seq_wrap[compile]	0.3614ms	0.2978ms	3.3584 KOps/s	3.3858 KOps/s	$\color{#d91a1a}-0.81\%$
test_seq_wrap[compile-overhead]	0.2717ms	0.2200ms	4.5448 KOps/s	4.5077 KOps/s	$\color{#35bf28}+0.82\%$
test_func_call_runtime[False-eager]	0.8144ms	0.7421ms	1.3474 KOps/s	1.3752 KOps/s	$\color{#d91a1a}-2.02\%$
test_func_call_runtime[False-compile]	0.9147ms	0.7365ms	1.3577 KOps/s	1.3876 KOps/s	$\color{#d91a1a}-2.15\%$
test_func_call_runtime[False-compile-overhead]	0.4258ms	0.3514ms	2.8458 KOps/s	2.7835 KOps/s	$\color{#35bf28}+2.24\%$
test_func_call_runtime[True-eager]	1.0102ms	0.8980ms	1.1136 KOps/s	1.1144 KOps/s	$\color{#d91a1a}-0.08\%$
test_func_call_runtime[True-compile]	0.7955ms	0.7360ms	1.3587 KOps/s	1.3338 KOps/s	$\color{#35bf28}+1.87\%$
test_func_call_runtime[True-compile-overhead]	0.4227ms	0.3726ms	2.6840 KOps/s	2.6311 KOps/s	$\color{#35bf28}+2.01\%$
test_func_call_cm_runtime[False-eager]	1.1001ms	0.7239ms	1.3815 KOps/s	1.3714 KOps/s	$\color{#35bf28}+0.73\%$
test_func_call_cm_runtime[False-compile]	0.8320ms	0.7176ms	1.3936 KOps/s	1.3802 KOps/s	$\color{#35bf28}+0.97\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4458ms	0.3564ms	2.8057 KOps/s	2.7852 KOps/s	$\color{#35bf28}+0.74\%$
test_func_call_cm_runtime[True-eager]	1.0793ms	0.9872ms	1.0130 KOps/s	1.0008 KOps/s	$\color{#35bf28}+1.21\%$
test_func_call_cm_runtime[True-compile]	0.8397ms	0.7645ms	1.3080 KOps/s	1.3013 KOps/s	$\color{#35bf28}+0.52\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4591ms	0.3966ms	2.5211 KOps/s	2.4557 KOps/s	$\color{#35bf28}+2.66\%$
test_vmap_func_call_cm_runtime[eager]	2.5512ms	2.0666ms	483.8790 Ops/s	475.4932 Ops/s	$\color{#35bf28}+1.76\%$
test_vmap_func_call_cm_runtime[compile]	0.8596ms	0.7786ms	1.2844 KOps/s	1.2626 KOps/s	$\color{#35bf28}+1.73\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4518ms	0.4021ms	2.4871 KOps/s	2.4547 KOps/s	$\color{#35bf28}+1.32\%$
test_distributed	6.1247ms	0.4029ms	2.4821 KOps/s	8.6087 KOps/s	$\textbf{\color{#d91a1a}-71.17\%}$
test_tdmodule	0.1444ms	18.6990μs	53.4788 KOps/s	47.1474 KOps/s	$\textbf{\color{#35bf28}+13.43\%}$
test_tdmodule_dispatch	70.7510μs	33.2551μs	30.0706 KOps/s	26.3942 KOps/s	$\textbf{\color{#35bf28}+13.93\%}$
test_tdseq	46.4410μs	19.7598μs	50.6077 KOps/s	46.0257 KOps/s	$\textbf{\color{#35bf28}+9.96\%}$
test_tdseq_dispatch	62.7410μs	36.0684μs	27.7251 KOps/s	24.4357 KOps/s	$\textbf{\color{#35bf28}+13.46\%}$
test_instantiation_functorch	1.6120ms	1.4961ms	668.3873 Ops/s	655.2125 Ops/s	$\color{#35bf28}+2.01\%$
test_exec_functorch	0.1810ms	0.1405ms	7.1186 KOps/s	7.1350 KOps/s	$\color{#d91a1a}-0.23\%$
test_exec_functional_call	0.1767ms	0.1321ms	7.5682 KOps/s	7.4882 KOps/s	$\color{#35bf28}+1.07\%$
test_exec_td_decorator	0.3651ms	0.1835ms	5.4505 KOps/s	5.4582 KOps/s	$\color{#d91a1a}-0.14\%$
test_vmap_mlp_speed_decorator[True-True]	0.8533ms	0.6804ms	1.4697 KOps/s	1.4540 KOps/s	$\color{#35bf28}+1.08\%$
test_vmap_mlp_speed_decorator[True-False]	0.8306ms	0.6682ms	1.4965 KOps/s	1.4594 KOps/s	$\color{#35bf28}+2.54\%$
test_vmap_mlp_speed_decorator[False-True]	0.7501ms	0.5874ms	1.7023 KOps/s	1.6931 KOps/s	$\color{#35bf28}+0.55\%$
test_vmap_mlp_speed_decorator[False-False]	0.7094ms	0.5868ms	1.7041 KOps/s	1.6911 KOps/s	$\color{#35bf28}+0.77\%$
test_vmap_transformer_speed_decorator[True-True]	19.2014ms	19.0913ms	52.3800 Ops/s	52.0396 Ops/s	$\color{#35bf28}+0.65\%$
test_vmap_transformer_speed_decorator[True-False]	19.7912ms	19.1267ms	52.2831 Ops/s	51.8711 Ops/s	$\color{#35bf28}+0.79\%$
test_vmap_transformer_speed_decorator[False-True]	19.1782ms	19.0042ms	52.6199 Ops/s	52.3386 Ops/s	$\color{#35bf28}+0.54\%$
test_vmap_transformer_speed_decorator[False-False]	19.1487ms	18.9615ms	52.7384 Ops/s	52.5008 Ops/s	$\color{#35bf28}+0.45\%$
test_to_module_speed[True]	1.1131ms	0.9515ms	1.0510 KOps/s	1.0445 KOps/s	$\color{#35bf28}+0.62\%$
test_to_module_speed[False]	1.3313ms	0.9401ms	1.0637 KOps/s	1.0489 KOps/s	$\color{#35bf28}+1.41\%$
test_tc_init	64.3910μs	34.7808μs	28.7515 KOps/s	27.0786 KOps/s	$\textbf{\color{#35bf28}+6.18\%}$
test_tc_init_nested	0.1126ms	68.0670μs	14.6914 KOps/s	13.6866 KOps/s	$\textbf{\color{#35bf28}+7.34\%}$
test_tc_first_layer_tensor	22.8800μs	0.8039μs	1.2439 MOps/s	1.2329 MOps/s	$\color{#35bf28}+0.89\%$
test_tc_first_layer_nontensor	24.5900μs	2.2307μs	448.2838 KOps/s	443.3242 KOps/s	$\color{#35bf28}+1.12\%$
test_tc_second_layer_tensor	9.3270μs	1.4082μs	710.1064 KOps/s	697.8133 KOps/s	$\color{#35bf28}+1.76\%$
test_tc_second_layer_nontensor	23.2600μs	2.9868μs	334.8104 KOps/s	335.2055 KOps/s	$\color{#d91a1a}-0.12\%$
test_unbind	0.2306s	10.2659ms	97.4099 Ops/s	142.3727 Ops/s	$\textbf{\color{#d91a1a}-31.58\%}$
test_full_like	12.6108ms	9.3678ms	106.7484 Ops/s	105.6935 Ops/s	$\color{#35bf28}+1.00\%$
test_zeros_like	9.1582ms	7.1583ms	139.6985 Ops/s	140.0183 Ops/s	$\color{#d91a1a}-0.23\%$
test_ones_like	4.9057ms	4.3483ms	229.9724 Ops/s	229.5974 Ops/s	$\color{#35bf28}+0.16\%$
test_clone	14.1105ms	10.6405ms	93.9804 Ops/s	151.4699 Ops/s	$\textbf{\color{#d91a1a}-37.95\%}$
test_squeeze	57.9910μs	9.4999μs	105.2644 KOps/s	105.9291 KOps/s	$\color{#d91a1a}-0.63\%$
test_unsqueeze	0.1203ms	68.7825μs	14.5386 KOps/s	13.9543 KOps/s	$\color{#35bf28}+4.19\%$
test_split	0.3952ms	0.1563ms	6.3989 KOps/s	6.3612 KOps/s	$\color{#35bf28}+0.59\%$
test_permute	0.3031ms	0.1725ms	5.7958 KOps/s	5.6049 KOps/s	$\color{#35bf28}+3.41\%$
test_stack	51.2113ms	50.8631ms	19.6606 Ops/s	18.7135 Ops/s	$\textbf{\color{#35bf28}+5.06\%}$
test_cat	51.5649ms	50.9444ms	19.6292 Ops/s	18.7319 Ops/s	$\color{#35bf28}+4.79\%$