[CI] Add aarch64-linux wheels

[vmoens]

In order to close pytorch/rl#2430

[github-actions]

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

Total Benchmarks: 222. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	48.2300μs	20.5262μs	48.7182 KOps/s	47.2011 KOps/s	$\color{#35bf28}+3.21\%$
test_plain_set_stack_nested	0.1561ms	20.7442μs	48.2063 KOps/s	47.9267 KOps/s	$\color{#35bf28}+0.58\%$
test_plain_set_nested_inplace	0.1545ms	22.0230μs	45.4071 KOps/s	44.7434 KOps/s	$\color{#35bf28}+1.48\%$
test_plain_set_stack_nested_inplace	51.2860μs	21.9153μs	45.6301 KOps/s	44.8344 KOps/s	$\color{#35bf28}+1.77\%$
test_items	42.0590μs	4.2114μs	237.4481 KOps/s	214.5893 KOps/s	$\textbf{\color{#35bf28}+10.65\%}$
test_items_nested	0.5466ms	0.3284ms	3.0449 KOps/s	3.0103 KOps/s	$\color{#35bf28}+1.15\%$
test_items_nested_locked	0.4373ms	0.3285ms	3.0439 KOps/s	2.9882 KOps/s	$\color{#35bf28}+1.86\%$
test_items_nested_leaf	0.1661ms	84.4819μs	11.8369 KOps/s	11.7399 KOps/s	$\color{#35bf28}+0.83\%$
test_items_stack_nested	0.6824ms	0.3331ms	3.0021 KOps/s	3.0019 KOps/s	$+0.01\%$
test_items_stack_nested_leaf	0.1632ms	84.3468μs	11.8558 KOps/s	12.1570 KOps/s	$\color{#d91a1a}-2.48\%$
test_items_stack_nested_locked	0.5583ms	0.3303ms	3.0273 KOps/s	3.0049 KOps/s	$\color{#35bf28}+0.75\%$
test_keys	50.3040μs	3.4996μs	285.7447 KOps/s	244.2723 KOps/s	$\textbf{\color{#35bf28}+16.98\%}$
test_keys_nested	0.1926ms	98.2076μs	10.1825 KOps/s	9.8131 KOps/s	$\color{#35bf28}+3.76\%$
test_keys_nested_locked	1.6283ms	0.1026ms	9.7512 KOps/s	9.7874 KOps/s	$\color{#d91a1a}-0.37\%$
test_keys_nested_leaf	0.1435ms	82.4158μs	12.1336 KOps/s	12.2735 KOps/s	$\color{#d91a1a}-1.14\%$
test_keys_stack_nested	0.1809ms	97.5063μs	10.2557 KOps/s	10.5132 KOps/s	$\color{#d91a1a}-2.45\%$
test_keys_stack_nested_leaf	0.1404ms	80.1739μs	12.4729 KOps/s	12.7917 KOps/s	$\color{#d91a1a}-2.49\%$
test_keys_stack_nested_locked	0.2065ms	0.1007ms	9.9333 KOps/s	10.1375 KOps/s	$\color{#d91a1a}-2.01\%$
test_values	8.7864μs	1.1256μs	888.4225 KOps/s	969.3241 KOps/s	$\textbf{\color{#d91a1a}-8.35\%}$
test_values_nested	0.1022ms	46.7914μs	21.3714 KOps/s	20.6091 KOps/s	$\color{#35bf28}+3.70\%$
test_values_nested_locked	0.1042ms	47.1634μs	21.2029 KOps/s	20.2123 KOps/s	$\color{#35bf28}+4.90\%$
test_values_nested_leaf	92.2020μs	41.4856μs	24.1048 KOps/s	23.0178 KOps/s	$\color{#35bf28}+4.72\%$
test_values_stack_nested	0.1086ms	47.6990μs	20.9648 KOps/s	20.4368 KOps/s	$\color{#35bf28}+2.58\%$
test_values_stack_nested_leaf	98.9050μs	42.6886μs	23.4255 KOps/s	24.4353 KOps/s	$\color{#d91a1a}-4.13\%$
test_values_stack_nested_locked	0.1156ms	47.9631μs	20.8494 KOps/s	21.0059 KOps/s	$\color{#d91a1a}-0.75\%$
test_membership	39.3440μs	0.8222μs	1.2163 MOps/s	1.2025 MOps/s	$\color{#35bf28}+1.15\%$
test_membership_nested	30.6480μs	2.5830μs	387.1410 KOps/s	392.7944 KOps/s	$\color{#d91a1a}-1.44\%$
test_membership_nested_leaf	45.5160μs	2.5645μs	389.9347 KOps/s	390.3059 KOps/s	$\color{#d91a1a}-0.10\%$
test_membership_stacked_nested	36.1180μs	2.5327μs	394.8369 KOps/s	392.8159 KOps/s	$\color{#35bf28}+0.51\%$
test_membership_stacked_nested_leaf	27.9520μs	2.5295μs	395.3382 KOps/s	389.3862 KOps/s	$\color{#35bf28}+1.53\%$
test_membership_nested_last	46.2460μs	3.7060μs	269.8298 KOps/s	270.1274 KOps/s	$\color{#d91a1a}-0.11\%$
test_membership_nested_leaf_last	25.1580μs	3.7302μs	268.0802 KOps/s	265.6019 KOps/s	$\color{#35bf28}+0.93\%$
test_membership_stacked_nested_last	32.0890μs	4.8461μs	206.3533 KOps/s	78.0963 KOps/s	$\textbf{\color{#35bf28}+164.23\%}$
test_membership_stacked_nested_leaf_last	50.3140μs	4.7948μs	208.5604 KOps/s	77.8285 KOps/s	$\textbf{\color{#35bf28}+167.97\%}$
test_nested_getleaf	54.2520μs	10.4294μs	95.8827 KOps/s	92.4227 KOps/s	$\color{#35bf28}+3.74\%$
test_nested_get	41.0880μs	10.0203μs	99.7973 KOps/s	98.4495 KOps/s	$\color{#35bf28}+1.37\%$
test_stacked_getleaf	57.6070μs	10.4306μs	95.8715 KOps/s	93.3385 KOps/s	$\color{#35bf28}+2.71\%$
test_stacked_get	48.1700μs	9.9753μs	100.2474 KOps/s	99.3870 KOps/s	$\color{#35bf28}+0.87\%$
test_nested_getitemleaf	37.7110μs	10.8133μs	92.4789 KOps/s	89.9870 KOps/s	$\color{#35bf28}+2.77\%$
test_nested_getitem	47.6690μs	10.1312μs	98.7046 KOps/s	97.0664 KOps/s	$\color{#35bf28}+1.69\%$
test_stacked_getitemleaf	57.2070μs	10.8152μs	92.4626 KOps/s	90.1843 KOps/s	$\color{#35bf28}+2.53\%$
test_stacked_getitem	29.8460μs	10.0720μs	99.2855 KOps/s	96.3084 KOps/s	$\color{#35bf28}+3.09\%$
test_lock_nested	82.2688ms	0.5607ms	1.7835 KOps/s	2.1171 KOps/s	$\textbf{\color{#d91a1a}-15.76\%}$
test_lock_stack_nested	0.6733ms	0.4355ms	2.2963 KOps/s	2.3234 KOps/s	$\color{#d91a1a}-1.17\%$
test_unlock_nested	95.0182ms	0.4961ms	2.0158 KOps/s	2.5190 KOps/s	$\textbf{\color{#d91a1a}-19.98\%}$
test_unlock_stack_nested	0.5464ms	0.3547ms	2.8194 KOps/s	2.8692 KOps/s	$\color{#d91a1a}-1.74\%$
test_flatten_speed	0.2535ms	0.1038ms	9.6336 KOps/s	9.4653 KOps/s	$\color{#35bf28}+1.78\%$
test_unflatten_speed	0.8184ms	0.4652ms	2.1495 KOps/s	2.2090 KOps/s	$\color{#d91a1a}-2.69\%$
test_common_ops	5.4181ms	1.1164ms	895.7694 Ops/s	923.3836 Ops/s	$\color{#d91a1a}-2.99\%$
test_creation	27.8920μs	2.0277μs	493.1604 KOps/s	484.3552 KOps/s	$\color{#35bf28}+1.82\%$
test_creation_empty	73.4780μs	18.1588μs	55.0697 KOps/s	54.2803 KOps/s	$\color{#35bf28}+1.45\%$
test_creation_nested_1	65.8540μs	21.3287μs	46.8851 KOps/s	46.1207 KOps/s	$\color{#35bf28}+1.66\%$
test_creation_nested_2	93.7660μs	26.4536μs	37.8021 KOps/s	39.5534 KOps/s	$\color{#d91a1a}-4.43\%$
test_clone	82.9550μs	17.1059μs	58.4594 KOps/s	61.6253 KOps/s	$\textbf{\color{#d91a1a}-5.14\%}$
test_getitem[int]	1.1330ms	16.4044μs	60.9594 KOps/s	62.4437 KOps/s	$\color{#d91a1a}-2.38\%$
test_getitem[slice_int]	0.1585ms	30.1409μs	33.1775 KOps/s	34.1405 KOps/s	$\color{#d91a1a}-2.82\%$
test_getitem[range]	0.2550ms	56.9158μs	17.5698 KOps/s	17.5050 KOps/s	$\color{#35bf28}+0.37\%$
test_getitem[tuple]	0.1418ms	25.1120μs	39.8216 KOps/s	41.1160 KOps/s	$\color{#d91a1a}-3.15\%$
test_getitem[list]	0.2067ms	52.1674μs	19.1690 KOps/s	19.2851 KOps/s	$\color{#d91a1a}-0.60\%$
test_setitem_dim[int]	59.1400μs	31.7092μs	31.5366 KOps/s	32.3493 KOps/s	$\color{#d91a1a}-2.51\%$
test_setitem_dim[slice_int]	0.1017ms	59.1111μs	16.9173 KOps/s	17.2045 KOps/s	$\color{#d91a1a}-1.67\%$
test_setitem_dim[range]	0.1556ms	82.8579μs	12.0689 KOps/s	11.7941 KOps/s	$\color{#35bf28}+2.33\%$
test_setitem_dim[tuple]	75.6620μs	47.3483μs	21.1201 KOps/s	21.2375 KOps/s	$\color{#d91a1a}-0.55\%$
test_setitem	87.0830μs	29.9982μs	33.3353 KOps/s	34.3416 KOps/s	$\color{#d91a1a}-2.93\%$
test_set	0.1454ms	29.7295μs	33.6366 KOps/s	34.7929 KOps/s	$\color{#d91a1a}-3.32\%$
test_set_shared	3.8404ms	0.2123ms	4.7111 KOps/s	4.6546 KOps/s	$\color{#35bf28}+1.21\%$
test_update	0.1392ms	35.9831μs	27.7908 KOps/s	28.6111 KOps/s	$\color{#d91a1a}-2.87\%$
test_update_nested	0.1464ms	46.3608μs	21.5699 KOps/s	21.8304 KOps/s	$\color{#d91a1a}-1.19\%$
test_update__nested	0.1418ms	34.5084μs	28.9784 KOps/s	29.1109 KOps/s	$\color{#d91a1a}-0.46\%$
test_set_nested	0.3458ms	35.6091μs	28.0827 KOps/s	32.6291 KOps/s	$\textbf{\color{#d91a1a}-13.93\%}$
test_set_nested_new	0.1897ms	37.4805μs	26.6806 KOps/s	27.6736 KOps/s	$\color{#d91a1a}-3.59\%$
test_select	0.2141ms	55.3676μs	18.0611 KOps/s	18.9148 KOps/s	$\color{#d91a1a}-4.51\%$
test_select_nested	0.1258ms	59.5947μs	16.7800 KOps/s	16.7813 KOps/s	$-0.01\%$
test_exclude_nested	0.1375ms	74.5001μs	13.4228 KOps/s	13.2720 KOps/s	$\color{#35bf28}+1.14\%$
test_empty[True]	0.4517ms	0.3110ms	3.2150 KOps/s	3.1878 KOps/s	$\color{#35bf28}+0.85\%$
test_empty[False]	8.1103μs	1.1969μs	835.4693 KOps/s	797.6019 KOps/s	$\color{#35bf28}+4.75\%$
test_unbind_speed	0.5368ms	0.2875ms	3.4782 KOps/s	3.4095 KOps/s	$\color{#35bf28}+2.01\%$
test_unbind_speed_stack0	0.7790ms	0.2854ms	3.5040 KOps/s	3.5627 KOps/s	$\color{#d91a1a}-1.65\%$
test_unbind_speed_stack1	90.9423ms	0.7681ms	1.3020 KOps/s	1.4082 KOps/s	$\textbf{\color{#d91a1a}-7.55\%}$
test_split	88.1736ms	2.1602ms	462.9138 Ops/s	473.0075 Ops/s	$\color{#d91a1a}-2.13\%$
test_chunk	2.3358ms	1.9919ms	502.0296 Ops/s	469.1218 Ops/s	$\textbf{\color{#35bf28}+7.01\%}$
test_creation[device0]	0.2259ms	0.1162ms	8.6043 KOps/s	8.6688 KOps/s	$\color{#d91a1a}-0.74\%$
test_creation_from_tensor	3.9849ms	0.1168ms	8.5638 KOps/s	8.5473 KOps/s	$\color{#35bf28}+0.19\%$
test_add_one[memmap_tensor0]	0.1446ms	7.1746μs	139.3805 KOps/s	143.3156 KOps/s	$\color{#d91a1a}-2.75\%$
test_contiguous[memmap_tensor0]	17.9430μs	1.8963μs	527.3539 KOps/s	528.2714 KOps/s	$\color{#d91a1a}-0.17\%$
test_stack[memmap_tensor0]	35.4960μs	5.6575μs	176.7571 KOps/s	182.2982 KOps/s	$\color{#d91a1a}-3.04\%$
test_memmaptd_index	1.0961ms	0.3845ms	2.6006 KOps/s	2.6217 KOps/s	$\color{#d91a1a}-0.80\%$
test_memmaptd_index_astensor	0.8456ms	0.4640ms	2.1551 KOps/s	2.1894 KOps/s	$\color{#d91a1a}-1.56\%$
test_memmaptd_index_op	1.6390ms	1.0068ms	993.2890 Ops/s	1.0268 KOps/s	$\color{#d91a1a}-3.26\%$
test_serialize_model	0.2232s	0.1353s	7.3936 Ops/s	8.3533 Ops/s	$\textbf{\color{#d91a1a}-11.49\%}$
test_serialize_model_pickle	0.4451s	0.3924s	2.5481 Ops/s	2.4407 Ops/s	$\color{#35bf28}+4.40\%$
test_serialize_weights	0.1363s	0.1188s	8.4145 Ops/s	7.6223 Ops/s	$\textbf{\color{#35bf28}+10.39\%}$
test_serialize_weights_returnearly	0.1838s	0.1586s	6.3043 Ops/s	6.3270 Ops/s	$\color{#d91a1a}-0.36\%$
test_serialize_weights_pickle	1.1044s	0.6928s	1.4435 Ops/s	2.2088 Ops/s	$\textbf{\color{#d91a1a}-34.65\%}$
test_serialize_weights_filesystem	0.1647s	0.1488s	6.7198 Ops/s	6.9491 Ops/s	$\color{#d91a1a}-3.30\%$
test_serialize_model_filesystem	0.1534s	0.1404s	7.1235 Ops/s	5.9416 Ops/s	$\textbf{\color{#35bf28}+19.89\%}$
test_reshape_pytree	89.4380μs	38.3748μs	26.0588 KOps/s	26.0518 KOps/s	$\color{#35bf28}+0.03\%$
test_reshape_td	0.1027ms	45.7438μs	21.8609 KOps/s	22.4414 KOps/s	$\color{#d91a1a}-2.59\%$
test_view_pytree	0.1499ms	37.9842μs	26.3267 KOps/s	26.4999 KOps/s	$\color{#d91a1a}-0.65\%$
test_view_td	0.1474ms	51.5581μs	19.3956 KOps/s	19.3745 KOps/s	$\color{#35bf28}+0.11\%$
test_unbind_pytree	0.1091ms	34.9921μs	28.5779 KOps/s	28.5668 KOps/s	$\color{#35bf28}+0.04\%$
test_unbind_td	0.3215ms	43.7489μs	22.8577 KOps/s	23.0797 KOps/s	$\color{#d91a1a}-0.96\%$
test_split_pytree	89.7080μs	37.6152μs	26.5850 KOps/s	27.1314 KOps/s	$\color{#d91a1a}-2.01\%$
test_split_td	0.5075ms	57.2402μs	17.4703 KOps/s	18.1418 KOps/s	$\color{#d91a1a}-3.70\%$
test_add_pytree	0.1107ms	43.0049μs	23.2532 KOps/s	23.3499 KOps/s	$\color{#d91a1a}-0.41\%$
test_add_td	0.1794ms	80.1432μs	12.4777 KOps/s	13.0644 KOps/s	$\color{#d91a1a}-4.49\%$
test_compile_add_one_nested[tensordict-compile]	0.1120ms	55.5990μs	17.9859 KOps/s	17.4774 KOps/s	$\color{#35bf28}+2.91\%$
test_compile_add_one_nested[tensordict-eager]	0.3343ms	0.1840ms	5.4344 KOps/s	5.2578 KOps/s	$\color{#35bf28}+3.36\%$
test_compile_add_one_nested[pytree-compile]	0.1209ms	55.1332μs	18.1379 KOps/s	17.6897 KOps/s	$\color{#35bf28}+2.53\%$
test_compile_add_one_nested[pytree-eager]	0.2538ms	0.1378ms	7.2568 KOps/s	7.2373 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_copy_nested[tensordict-compile]	90.3390μs	19.9884μs	50.0291 KOps/s	48.6682 KOps/s	$\color{#35bf28}+2.80\%$
test_compile_copy_nested[tensordict-eager]	0.1360ms	67.1273μs	14.8971 KOps/s	15.0574 KOps/s	$\color{#d91a1a}-1.06\%$
test_compile_copy_nested[pytree-compile]	0.1979ms	74.8935μs	13.3523 KOps/s	13.4854 KOps/s	$\color{#d91a1a}-0.99\%$
test_compile_copy_nested[pytree-eager]	0.1354ms	67.0053μs	14.9242 KOps/s	14.7641 KOps/s	$\color{#35bf28}+1.08\%$
test_compile_add_one_flat[tensordict-compile]	0.2975ms	0.1709ms	5.8530 KOps/s	5.8515 KOps/s	$\color{#35bf28}+0.03\%$
test_compile_add_one_flat[tensordict-eager]	0.3943ms	0.1866ms	5.3589 KOps/s	5.2605 KOps/s	$\color{#35bf28}+1.87\%$
test_compile_add_one_flat[tensorclass-compile]	0.1129ms	44.8779μs	22.2827 KOps/s	21.6919 KOps/s	$\color{#35bf28}+2.72\%$
test_compile_add_one_flat[tensorclass-eager]	0.9384ms	66.3853μs	15.0636 KOps/s	14.9185 KOps/s	$\color{#35bf28}+0.97\%$
test_compile_add_one_flat[pytree-compile]	0.4028ms	0.1752ms	5.7069 KOps/s	5.7973 KOps/s	$\color{#d91a1a}-1.56\%$
test_compile_add_one_flat[pytree-eager]	0.3625ms	0.2806ms	3.5638 KOps/s	3.5748 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_add_self_flat[tensordict-eager]	0.2916ms	0.1969ms	5.0790 KOps/s	4.9071 KOps/s	$\color{#35bf28}+3.50\%$
test_compile_add_self_flat[tensordict-compile]	0.4763ms	0.1808ms	5.5312 KOps/s	5.8050 KOps/s	$\color{#d91a1a}-4.72\%$
test_compile_add_self_flat[tensorclass-eager]	1.0252ms	61.1426μs	16.3552 KOps/s	16.3329 KOps/s	$\color{#35bf28}+0.14\%$
test_compile_add_self_flat[tensorclass-compile]	0.1131ms	46.4854μs	21.5121 KOps/s	20.9739 KOps/s	$\color{#35bf28}+2.57\%$
test_compile_add_self_flat[pytree-eager]	0.4425ms	0.2294ms	4.3601 KOps/s	4.3232 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_add_self_flat[pytree-compile]	0.3256ms	0.1745ms	5.7322 KOps/s	5.6847 KOps/s	$\color{#35bf28}+0.83\%$
test_compile_copy_flat[tensordict-compile]	0.2314ms	0.1047ms	9.5532 KOps/s	9.9053 KOps/s	$\color{#d91a1a}-3.55\%$
test_compile_copy_flat[tensordict-eager]	0.1196ms	58.3422μs	17.1402 KOps/s	17.2427 KOps/s	$\color{#d91a1a}-0.59\%$
test_compile_copy_flat[pytree-compile]	0.1802ms	76.7187μs	13.0346 KOps/s	13.0692 KOps/s	$\color{#d91a1a}-0.26\%$
test_compile_copy_flat[pytree-eager]	0.1378ms	67.7499μs	14.7602 KOps/s	14.1562 KOps/s	$\color{#35bf28}+4.27\%$
test_compile_assign_and_add[tensordict-compile]	0.2799ms	0.1908ms	5.2403 KOps/s	5.1530 KOps/s	$\color{#35bf28}+1.69\%$
test_compile_assign_and_add[tensordict-eager]	2.6600ms	1.6044ms	623.2751 Ops/s	617.7446 Ops/s	$\color{#35bf28}+0.90\%$
test_compile_assign_and_add[pytree-compile]	0.7860ms	0.1920ms	5.2079 KOps/s	5.1521 KOps/s	$\color{#35bf28}+1.08\%$
test_compile_assign_and_add[pytree-eager]	1.2369ms	1.0762ms	929.1781 Ops/s	921.0995 Ops/s	$\color{#35bf28}+0.88\%$
test_compile_assign_and_add_stack[compile]	0.7164ms	0.4142ms	2.4141 KOps/s	2.4155 KOps/s	$\color{#d91a1a}-0.06\%$
test_compile_assign_and_add_stack[eager]	3.8072ms	3.6395ms	274.7612 Ops/s	266.3130 Ops/s	$\color{#35bf28}+3.17\%$
test_compile_indexing[tensor-tensordict-compile]	83.8570μs	33.6890μs	29.6833 KOps/s	30.1044 KOps/s	$\color{#d91a1a}-1.40\%$
test_compile_indexing[tensor-tensordict-eager]	0.6239ms	45.7042μs	21.8799 KOps/s	21.0860 KOps/s	$\color{#35bf28}+3.76\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1010ms	29.2285μs	34.2131 KOps/s	34.2154 KOps/s	$-0.01\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1047ms	27.8011μs	35.9698 KOps/s	35.5185 KOps/s	$\color{#35bf28}+1.27\%$
test_compile_indexing[tensor-pytree-compile]	0.1170ms	29.1125μs	34.3495 KOps/s	34.4516 KOps/s	$\color{#d91a1a}-0.30\%$
test_compile_indexing[tensor-pytree-eager]	90.1890μs	27.5813μs	36.2565 KOps/s	35.5388 KOps/s	$\color{#35bf28}+2.02\%$
test_compile_indexing[slice-tensordict-compile]	0.1440ms	71.4361μs	13.9985 KOps/s	13.8084 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_indexing[slice-tensordict-eager]	0.5183ms	27.6598μs	36.1535 KOps/s	37.6489 KOps/s	$\color{#d91a1a}-3.97\%$
test_compile_indexing[slice-tensorclass-compile]	0.1435ms	66.3617μs	15.0689 KOps/s	14.9837 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_indexing[slice-tensorclass-eager]	82.7050μs	23.0037μs	43.4713 KOps/s	45.1881 KOps/s	$\color{#d91a1a}-3.80\%$
test_compile_indexing[slice-pytree-compile]	0.1550ms	67.1988μs	14.8812 KOps/s	14.9891 KOps/s	$\color{#d91a1a}-0.72\%$
test_compile_indexing[slice-pytree-eager]	72.2450μs	22.8593μs	43.7459 KOps/s	45.1441 KOps/s	$\color{#d91a1a}-3.10\%$
test_compile_indexing[int-tensordict-compile]	0.1564ms	71.8812μs	13.9118 KOps/s	13.9181 KOps/s	$\color{#d91a1a}-0.05\%$
test_compile_indexing[int-tensordict-eager]	0.9489ms	27.0650μs	36.9481 KOps/s	38.1275 KOps/s	$\color{#d91a1a}-3.09\%$
test_compile_indexing[int-tensorclass-compile]	0.1436ms	66.2400μs	15.0966 KOps/s	15.0792 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_indexing[int-tensorclass-eager]	0.2905ms	22.8564μs	43.7515 KOps/s	44.9036 KOps/s	$\color{#d91a1a}-2.57\%$
test_compile_indexing[int-pytree-compile]	0.3308ms	68.7461μs	14.5463 KOps/s	15.0874 KOps/s	$\color{#d91a1a}-3.59\%$
test_compile_indexing[int-pytree-eager]	73.6180μs	22.7024μs	44.0482 KOps/s	44.7778 KOps/s	$\color{#d91a1a}-1.63\%$
test_mod_add[eager]	72.3370μs	23.5233μs	42.5111 KOps/s	40.0070 KOps/s	$\textbf{\color{#35bf28}+6.26\%}$
test_mod_add[compile]	0.1077ms	38.0159μs	26.3048 KOps/s	26.0107 KOps/s	$\color{#35bf28}+1.13\%$
test_mod_add[compile-overhead]	0.2845ms	37.7910μs	26.4613 KOps/s	26.4618 KOps/s	$-0.00\%$
test_mod_wrap[eager]	0.3977ms	0.2002ms	4.9955 KOps/s	4.8554 KOps/s	$\color{#35bf28}+2.89\%$
test_mod_wrap[compile]	0.3619ms	0.2277ms	4.3910 KOps/s	4.3603 KOps/s	$\color{#35bf28}+0.70\%$
test_mod_wrap[compile-overhead]	0.3876ms	0.2262ms	4.4210 KOps/s	4.3649 KOps/s	$\color{#35bf28}+1.28\%$
test_mod_wrap_and_backward[eager]	12.2011ms	10.7667ms	92.8791 Ops/s	90.0216 Ops/s	$\color{#35bf28}+3.17\%$
test_mod_wrap_and_backward[compile]	12.0426ms	10.5574ms	94.7203 Ops/s	80.7543 Ops/s	$\textbf{\color{#35bf28}+17.29\%}$
test_mod_wrap_and_backward[compile-overhead]	12.1782ms	10.8867ms	91.8553 Ops/s	82.8748 Ops/s	$\textbf{\color{#35bf28}+10.84\%}$
test_seq_add[eager]	0.2076ms	85.0314μs	11.7604 KOps/s	11.2218 KOps/s	$\color{#35bf28}+4.80\%$
test_seq_add[compile]	0.1510ms	61.5923μs	16.2358 KOps/s	16.1852 KOps/s	$\color{#35bf28}+0.31\%$
test_seq_add[compile-overhead]	0.1479ms	60.6722μs	16.4820 KOps/s	16.2390 KOps/s	$\color{#35bf28}+1.50\%$
test_seq_wrap[eager]	0.5473ms	0.3702ms	2.7010 KOps/s	2.6013 KOps/s	$\color{#35bf28}+3.83\%$
test_seq_wrap[compile]	0.3855ms	0.2597ms	3.8507 KOps/s	3.7539 KOps/s	$\color{#35bf28}+2.58\%$
test_seq_wrap[compile-overhead]	0.3670ms	0.2602ms	3.8434 KOps/s	3.7952 KOps/s	$\color{#35bf28}+1.27\%$
test_func_call_runtime[False-eager]	0.9247ms	0.4991ms	2.0036 KOps/s	1.8937 KOps/s	$\textbf{\color{#35bf28}+5.80\%}$
test_func_call_runtime[False-compile]	0.6649ms	0.4930ms	2.0283 KOps/s	1.9740 KOps/s	$\color{#35bf28}+2.75\%$
test_func_call_runtime[False-compile-overhead]	0.6156ms	0.4885ms	2.0473 KOps/s	1.9790 KOps/s	$\color{#35bf28}+3.45\%$
test_func_call_runtime[True-eager]	1.1301ms	0.7125ms	1.4036 KOps/s	1.3383 KOps/s	$\color{#35bf28}+4.88\%$
test_func_call_runtime[True-compile]	0.8962ms	0.5014ms	1.9946 KOps/s	1.9572 KOps/s	$\color{#35bf28}+1.91\%$
test_func_call_runtime[True-compile-overhead]	1.0166ms	0.5055ms	1.9781 KOps/s	1.9676 KOps/s	$\color{#35bf28}+0.54\%$
test_func_call_cm_runtime[False-eager]	0.7580ms	0.5018ms	1.9929 KOps/s	1.9028 KOps/s	$\color{#35bf28}+4.73\%$
test_func_call_cm_runtime[False-compile]	0.8431ms	0.4964ms	2.0146 KOps/s	2.0102 KOps/s	$\color{#35bf28}+0.22\%$
test_func_call_cm_runtime[False-compile-overhead]	0.8140ms	0.4950ms	2.0204 KOps/s	2.0151 KOps/s	$\color{#35bf28}+0.26\%$
test_func_call_cm_runtime[True-eager]	1.0531ms	0.8373ms	1.1943 KOps/s	1.1323 KOps/s	$\textbf{\color{#35bf28}+5.48\%}$
test_func_call_cm_runtime[True-compile]	1.1997ms	0.7163ms	1.3960 KOps/s	1.3264 KOps/s	$\textbf{\color{#35bf28}+5.24\%}$
test_func_call_cm_runtime[True-compile-overhead]	1.0351ms	0.7171ms	1.3945 KOps/s	1.3478 KOps/s	$\color{#35bf28}+3.46\%$
test_vmap_func_call_cm_runtime[eager]	2.4624ms	1.7934ms	557.5867 Ops/s	540.9918 Ops/s	$\color{#35bf28}+3.07\%$
test_vmap_func_call_cm_runtime[compile]	2.7872ms	1.8417ms	542.9701 Ops/s	520.1077 Ops/s	$\color{#35bf28}+4.40\%$
test_vmap_func_call_cm_runtime[compile-overhead]	3.0442ms	1.8642ms	536.4329 Ops/s	524.6366 Ops/s	$\color{#35bf28}+2.25\%$
test_distributed	0.2808ms	0.1261ms	7.9300 KOps/s	7.8014 KOps/s	$\color{#35bf28}+1.65\%$
test_tdmodule	54.4720μs	17.5512μs	56.9762 KOps/s	54.2012 KOps/s	$\textbf{\color{#35bf28}+5.12\%}$
test_tdmodule_dispatch	51.3570μs	35.9355μs	27.8276 KOps/s	26.8980 KOps/s	$\color{#35bf28}+3.46\%$
test_tdseq	38.0210μs	20.2664μs	49.3429 KOps/s	47.4732 KOps/s	$\color{#35bf28}+3.94\%$
test_tdseq_dispatch	75.8220μs	41.6712μs	23.9974 KOps/s	23.9316 KOps/s	$\color{#35bf28}+0.28\%$
test_instantiation_functorch	2.3248ms	1.5881ms	629.6991 Ops/s	634.8444 Ops/s	$\color{#d91a1a}-0.81\%$
test_instantiation_td	2.0378ms	1.1650ms	858.3738 Ops/s	866.6390 Ops/s	$\color{#d91a1a}-0.95\%$
test_exec_functorch	0.2709ms	0.1844ms	5.4233 KOps/s	5.3950 KOps/s	$\color{#35bf28}+0.52\%$
test_exec_functional_call	0.4086ms	0.1734ms	5.7665 KOps/s	5.6729 KOps/s	$\color{#35bf28}+1.65\%$
test_exec_td	0.2703ms	0.1651ms	6.0559 KOps/s	5.8506 KOps/s	$\color{#35bf28}+3.51\%$
test_exec_td_decorator	0.3602ms	0.2179ms	4.5883 KOps/s	4.5074 KOps/s	$\color{#35bf28}+1.80\%$
test_vmap_mlp_speed[True-True]	1.3274ms	0.6262ms	1.5969 KOps/s	1.5209 KOps/s	$\color{#35bf28}+5.00\%$
test_vmap_mlp_speed[True-False]	0.8607ms	0.6223ms	1.6069 KOps/s	1.5892 KOps/s	$\color{#35bf28}+1.11\%$
test_vmap_mlp_speed[False-True]	0.6355ms	0.4776ms	2.0937 KOps/s	2.0387 KOps/s	$\color{#35bf28}+2.70\%$
test_vmap_mlp_speed[False-False]	0.7832ms	0.4814ms	2.0774 KOps/s	2.0492 KOps/s	$\color{#35bf28}+1.38\%$
test_vmap_mlp_speed_decorator[True-True]	1.3061ms	0.6019ms	1.6615 KOps/s	1.6402 KOps/s	$\color{#35bf28}+1.29\%$
test_vmap_mlp_speed_decorator[True-False]	0.8876ms	0.6032ms	1.6578 KOps/s	1.6247 KOps/s	$\color{#35bf28}+2.04\%$
test_vmap_mlp_speed_decorator[False-True]	0.7957ms	0.4934ms	2.0267 KOps/s	1.9816 KOps/s	$\color{#35bf28}+2.27\%$
test_vmap_mlp_speed_decorator[False-False]	0.6930ms	0.4917ms	2.0337 KOps/s	1.9718 KOps/s	$\color{#35bf28}+3.14\%$
test_to_module_speed[True]	2.0397ms	1.2765ms	783.3894 Ops/s	777.4475 Ops/s	$\color{#35bf28}+0.76\%$
test_to_module_speed[False]	1.3411ms	1.2338ms	810.5082 Ops/s	794.7533 Ops/s	$\color{#35bf28}+1.98\%$
test_tc_init	79.7290μs	44.4889μs	22.4775 KOps/s	22.4735 KOps/s	$\color{#35bf28}+0.02\%$
test_tc_init_nested	0.1933ms	91.9538μs	10.8750 KOps/s	11.2329 KOps/s	$\color{#d91a1a}-3.19\%$
test_tc_first_layer_tensor	19.5770μs	1.5137μs	660.6310 KOps/s	666.4287 KOps/s	$\color{#d91a1a}-0.87\%$
test_tc_first_layer_nontensor	43.3210μs	4.7050μs	212.5418 KOps/s	216.6259 KOps/s	$\color{#d91a1a}-1.89\%$
test_tc_second_layer_tensor	27.5420μs	2.8123μs	355.5869 KOps/s	350.0116 KOps/s	$\color{#35bf28}+1.59\%$
test_tc_second_layer_nontensor	34.6950μs	6.0092μs	166.4110 KOps/s	167.5062 KOps/s	$\color{#d91a1a}-0.65\%$
test_unbind	7.4495ms	7.1827ms	139.2241 Ops/s	75.5834 Ops/s	$\textbf{\color{#35bf28}+84.20\%}$
test_full_like	8.3013ms	7.0293ms	142.2623 Ops/s	137.0193 Ops/s	$\color{#35bf28}+3.83\%$
test_zeros_like	3.1398ms	2.7822ms	359.4342 Ops/s	148.0803 Ops/s	$\textbf{\color{#35bf28}+142.73\%}$
test_ones_like	3.7883ms	3.2886ms	304.0790 Ops/s	135.2732 Ops/s	$\textbf{\color{#35bf28}+124.79\%}$
test_clone	5.7379ms	4.9261ms	203.0006 Ops/s	110.1092 Ops/s	$\textbf{\color{#35bf28}+84.36\%}$
test_squeeze	59.0600μs	12.4075μs	80.5965 KOps/s	81.6743 KOps/s	$\color{#d91a1a}-1.32\%$
test_unsqueeze	0.2043ms	88.7963μs	11.2617 KOps/s	10.9711 KOps/s	$\color{#35bf28}+2.65\%$
test_split	0.5568ms	0.1921ms	5.2061 KOps/s	5.2450 KOps/s	$\color{#d91a1a}-0.74\%$
test_permute	0.3345ms	0.2154ms	4.6426 KOps/s	4.3535 KOps/s	$\textbf{\color{#35bf28}+6.64\%}$
test_stack	28.2218ms	24.3228ms	41.1137 Ops/s	39.3518 Ops/s	$\color{#35bf28}+4.48\%$
test_cat	28.1617ms	24.0936ms	41.5048 Ops/s	39.4884 Ops/s	$\textbf{\color{#35bf28}+5.11\%}$

[github-actions]

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

Total Benchmarks: 228. Improved: $\large\color{#35bf28}5$. Worsened: $\large\color{#d91a1a}15$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1515ms	14.8846μs	67.1835 KOps/s	70.1043 KOps/s	$\color{#d91a1a}-4.17\%$
test_plain_set_stack_nested	39.4210μs	14.8812μs	67.1988 KOps/s	69.0655 KOps/s	$\color{#d91a1a}-2.70\%$
test_plain_set_nested_inplace	52.3710μs	15.8895μs	62.9347 KOps/s	64.3970 KOps/s	$\color{#d91a1a}-2.27\%$
test_plain_set_stack_nested_inplace	48.0310μs	15.9280μs	62.7825 KOps/s	65.1745 KOps/s	$\color{#d91a1a}-3.67\%$
test_items	22.7300μs	2.8669μs	348.8050 KOps/s	346.6128 KOps/s	$\color{#35bf28}+0.63\%$
test_items_nested	0.4667ms	0.3106ms	3.2193 KOps/s	3.1585 KOps/s	$\color{#35bf28}+1.92\%$
test_items_nested_locked	0.4516ms	0.3111ms	3.2142 KOps/s	3.1606 KOps/s	$\color{#35bf28}+1.70\%$
test_items_nested_leaf	0.1023ms	63.6169μs	15.7191 KOps/s	16.0892 KOps/s	$\color{#d91a1a}-2.30\%$
test_items_stack_nested	0.4148ms	0.3137ms	3.1881 KOps/s	3.1758 KOps/s	$\color{#35bf28}+0.39\%$
test_items_stack_nested_leaf	0.1127ms	64.9246μs	15.4025 KOps/s	15.6295 KOps/s	$\color{#d91a1a}-1.45\%$
test_items_stack_nested_locked	0.4553ms	0.3162ms	3.1627 KOps/s	3.1509 KOps/s	$\color{#35bf28}+0.38\%$
test_keys	38.4310μs	3.3971μs	294.3694 KOps/s	294.9511 KOps/s	$\color{#d91a1a}-0.20\%$
test_keys_nested	96.6320μs	55.0712μs	18.1583 KOps/s	18.6444 KOps/s	$\color{#d91a1a}-2.61\%$
test_keys_nested_locked	2.2959ms	60.2882μs	16.5870 KOps/s	16.4264 KOps/s	$\color{#35bf28}+0.98\%$
test_keys_nested_leaf	80.9810μs	46.5425μs	21.4857 KOps/s	22.1116 KOps/s	$\color{#d91a1a}-2.83\%$
test_keys_stack_nested	80.6720μs	54.6788μs	18.2886 KOps/s	18.0673 KOps/s	$\color{#35bf28}+1.22\%$
test_keys_stack_nested_leaf	0.1015ms	46.8223μs	21.3573 KOps/s	21.1659 KOps/s	$\color{#35bf28}+0.90\%$
test_keys_stack_nested_locked	97.8420μs	59.6440μs	16.7662 KOps/s	16.5940 KOps/s	$\color{#35bf28}+1.04\%$
test_values	5.7171μs	0.8011μs	1.2483 MOps/s	1.2268 MOps/s	$\color{#35bf28}+1.76\%$
test_values_nested	59.3810μs	27.2456μs	36.7031 KOps/s	36.2835 KOps/s	$\color{#35bf28}+1.16\%$
test_values_nested_locked	58.8910μs	29.1648μs	34.2879 KOps/s	34.1053 KOps/s	$\color{#35bf28}+0.54\%$
test_values_nested_leaf	62.0720μs	24.0211μs	41.6301 KOps/s	41.2249 KOps/s	$\color{#35bf28}+0.98\%$
test_values_stack_nested	63.8110μs	28.0165μs	35.6932 KOps/s	35.2427 KOps/s	$\color{#35bf28}+1.28\%$
test_values_stack_nested_leaf	61.6210μs	24.7096μs	40.4700 KOps/s	40.0489 KOps/s	$\color{#35bf28}+1.05\%$
test_values_stack_nested_locked	90.1810μs	29.8757μs	33.4720 KOps/s	33.1240 KOps/s	$\color{#35bf28}+1.05\%$
test_membership	2.3371μs	0.4710μs	2.1231 MOps/s	2.1259 MOps/s	$\color{#d91a1a}-0.13\%$
test_membership_nested	16.3905μs	1.7585μs	568.6810 KOps/s	558.8681 KOps/s	$\color{#35bf28}+1.76\%$
test_membership_nested_leaf	12.1767μs	1.7183μs	581.9841 KOps/s	577.3452 KOps/s	$\color{#35bf28}+0.80\%$
test_membership_stacked_nested	37.6010μs	1.8198μs	549.5073 KOps/s	563.3504 KOps/s	$\color{#d91a1a}-2.46\%$
test_membership_stacked_nested_leaf	32.8910μs	1.8074μs	553.2961 KOps/s	554.5971 KOps/s	$\color{#d91a1a}-0.23\%$
test_membership_nested_last	38.9310μs	2.6029μs	384.1834 KOps/s	380.7995 KOps/s	$\color{#35bf28}+0.89\%$
test_membership_nested_leaf_last	32.2710μs	2.6277μs	380.5626 KOps/s	382.1472 KOps/s	$\color{#d91a1a}-0.41\%$
test_membership_stacked_nested_last	32.2500μs	3.7227μs	268.6236 KOps/s	381.1374 KOps/s	$\textbf{\color{#d91a1a}-29.52\%}$
test_membership_stacked_nested_leaf_last	39.9900μs	3.7209μs	268.7532 KOps/s	380.4217 KOps/s	$\textbf{\color{#d91a1a}-29.35\%}$
test_nested_getleaf	50.4110μs	6.0855μs	164.3243 KOps/s	166.1783 KOps/s	$\color{#d91a1a}-1.12\%$
test_nested_get	31.4110μs	5.8019μs	172.3577 KOps/s	175.9977 KOps/s	$\color{#d91a1a}-2.07\%$
test_stacked_getleaf	59.2710μs	6.1011μs	163.9056 KOps/s	166.0210 KOps/s	$\color{#d91a1a}-1.27\%$
test_stacked_get	35.5310μs	5.6812μs	176.0189 KOps/s	176.5631 KOps/s	$\color{#d91a1a}-0.31\%$
test_nested_getitemleaf	41.4010μs	6.0960μs	164.0408 KOps/s	163.4554 KOps/s	$\color{#35bf28}+0.36\%$
test_nested_getitem	37.3310μs	5.7635μs	173.5048 KOps/s	175.5235 KOps/s	$\color{#d91a1a}-1.15\%$
test_stacked_getitemleaf	50.1410μs	6.1157μs	163.5141 KOps/s	163.6994 KOps/s	$\color{#d91a1a}-0.11\%$
test_stacked_getitem	33.1700μs	5.7561μs	173.7299 KOps/s	176.4079 KOps/s	$\color{#d91a1a}-1.52\%$
test_lock_nested	5.1733ms	0.4244ms	2.3565 KOps/s	2.4260 KOps/s	$\color{#d91a1a}-2.87\%$
test_lock_stack_nested	0.4886ms	0.3835ms	2.6078 KOps/s	2.6404 KOps/s	$\color{#d91a1a}-1.24\%$
test_unlock_nested	0.8140ms	0.3587ms	2.7877 KOps/s	2.8225 KOps/s	$\color{#d91a1a}-1.23\%$
test_unlock_stack_nested	0.3601ms	0.3216ms	3.1095 KOps/s	3.1352 KOps/s	$\color{#d91a1a}-0.82\%$
test_flatten_speed	0.2938ms	79.9076μs	12.5145 KOps/s	12.6430 KOps/s	$\color{#d91a1a}-1.02\%$
test_unflatten_speed	0.3218ms	0.2804ms	3.5657 KOps/s	3.5635 KOps/s	$\color{#35bf28}+0.06\%$
test_common_ops	1.5580ms	1.3200ms	757.5671 Ops/s	766.5461 Ops/s	$\color{#d91a1a}-1.17\%$
test_creation	24.8110μs	1.4702μs	680.1827 KOps/s	680.0993 KOps/s	$\color{#35bf28}+0.01\%$
test_creation_empty	47.3910μs	17.4104μs	57.4369 KOps/s	61.6600 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_creation_nested_1	59.0910μs	19.2545μs	51.9359 KOps/s	55.9961 KOps/s	$\textbf{\color{#d91a1a}-7.25\%}$
test_creation_nested_2	54.7010μs	21.6793μs	46.1269 KOps/s	49.5272 KOps/s	$\textbf{\color{#d91a1a}-6.87\%}$
test_clone	83.2320μs	29.3014μs	34.1281 KOps/s	34.3716 KOps/s	$\color{#d91a1a}-0.71\%$
test_getitem[int]	1.2647ms	16.2104μs	61.6888 KOps/s	63.3171 KOps/s	$\color{#d91a1a}-2.57\%$
test_getitem[slice_int]	0.1194ms	28.1670μs	35.5025 KOps/s	35.8090 KOps/s	$\color{#d91a1a}-0.86\%$
test_getitem[range]	0.2312ms	0.1112ms	8.9911 KOps/s	9.1808 KOps/s	$\color{#d91a1a}-2.07\%$
test_getitem[tuple]	0.1174ms	24.0090μs	41.6510 KOps/s	41.6091 KOps/s	$\color{#35bf28}+0.10\%$
test_getitem[list]	0.1932ms	0.1002ms	9.9834 KOps/s	10.0795 KOps/s	$\color{#d91a1a}-0.95\%$
test_setitem_dim[int]	70.8710μs	45.7027μs	21.8806 KOps/s	22.0346 KOps/s	$\color{#d91a1a}-0.70\%$
test_setitem_dim[slice_int]	97.2620μs	68.3716μs	14.6260 KOps/s	14.1742 KOps/s	$\color{#35bf28}+3.19\%$
test_setitem_dim[range]	0.1785ms	0.1291ms	7.7484 KOps/s	7.7506 KOps/s	$\color{#d91a1a}-0.03\%$
test_setitem_dim[tuple]	85.6420μs	62.1585μs	16.0879 KOps/s	16.2725 KOps/s	$\color{#d91a1a}-1.13\%$
test_setitem	90.5410μs	42.6279μs	23.4588 KOps/s	23.4793 KOps/s	$\color{#d91a1a}-0.09\%$
test_set	80.0310μs	41.9102μs	23.8606 KOps/s	23.9001 KOps/s	$\color{#d91a1a}-0.17\%$
test_set_shared	0.3261ms	51.3190μs	19.4859 KOps/s	19.5374 KOps/s	$\color{#d91a1a}-0.26\%$
test_update	94.7220μs	51.4062μs	19.4529 KOps/s	19.6557 KOps/s	$\color{#d91a1a}-1.03\%$
test_update_nested	99.7720μs	59.0724μs	16.9284 KOps/s	17.2353 KOps/s	$\color{#d91a1a}-1.78\%$
test_update__nested	0.1019ms	60.3721μs	16.5639 KOps/s	16.8759 KOps/s	$\color{#d91a1a}-1.85\%$
test_set_nested	82.6920μs	44.4573μs	22.4935 KOps/s	23.0666 KOps/s	$\color{#d91a1a}-2.48\%$
test_set_nested_new	0.1034ms	48.6101μs	20.5719 KOps/s	21.1390 KOps/s	$\color{#d91a1a}-2.68\%$
test_select	0.1198ms	65.3149μs	15.3104 KOps/s	16.5258 KOps/s	$\textbf{\color{#d91a1a}-7.35\%}$
test_select_nested	0.5734ms	42.2280μs	23.6810 KOps/s	24.0326 KOps/s	$\color{#d91a1a}-1.46\%$
test_exclude_nested	95.5110μs	58.7826μs	17.0118 KOps/s	16.9110 KOps/s	$\color{#35bf28}+0.60\%$
test_empty[True]	0.3577ms	0.2391ms	4.1815 KOps/s	4.1332 KOps/s	$\color{#35bf28}+1.17\%$
test_empty[False]	4.4561μs	0.7413μs	1.3489 MOps/s	1.3541 MOps/s	$\color{#d91a1a}-0.38\%$
test_to	46.4310μs	25.3332μs	39.4739 KOps/s	39.1639 KOps/s	$\color{#35bf28}+0.79\%$
test_to_nonblocking	61.6010μs	24.4971μs	40.8212 KOps/s	41.3061 KOps/s	$\color{#d91a1a}-1.17\%$
test_unbind_speed	1.1052ms	0.2774ms	3.6047 KOps/s	3.5967 KOps/s	$\color{#35bf28}+0.22\%$
test_unbind_speed_stack0	0.3910ms	0.2734ms	3.6575 KOps/s	3.6541 KOps/s	$\color{#35bf28}+0.09\%$
test_unbind_speed_stack1	92.7688ms	0.7022ms	1.4240 KOps/s	1.4335 KOps/s	$\color{#d91a1a}-0.67\%$
test_split	93.7569ms	2.2367ms	447.0931 Ops/s	454.8908 Ops/s	$\color{#d91a1a}-1.71\%$
test_chunk	95.8807ms	2.2402ms	446.3980 Ops/s	452.9296 Ops/s	$\color{#d91a1a}-1.44\%$
test_creation[device0]	0.3507ms	0.1298ms	7.7034 KOps/s	7.8894 KOps/s	$\color{#d91a1a}-2.36\%$
test_creation_from_tensor	0.3526ms	0.1351ms	7.3997 KOps/s	7.7035 KOps/s	$\color{#d91a1a}-3.94\%$
test_add_one[memmap_tensor0]	0.1727ms	9.0649μs	110.3154 KOps/s	112.7212 KOps/s	$\color{#d91a1a}-2.13\%$
test_contiguous[memmap_tensor0]	20.4300μs	2.1867μs	457.3114 KOps/s	453.8269 KOps/s	$\color{#35bf28}+0.77\%$
test_stack[memmap_tensor0]	38.6810μs	6.9749μs	143.3719 KOps/s	136.6639 KOps/s	$\color{#35bf28}+4.91\%$
test_memmaptd_index	1.1108ms	0.4343ms	2.3025 KOps/s	2.3027 KOps/s	$-0.01\%$
test_memmaptd_index_astensor	0.7380ms	0.4877ms	2.0504 KOps/s	2.0206 KOps/s	$\color{#35bf28}+1.48\%$
test_memmaptd_index_op	1.4574ms	1.0742ms	930.9380 Ops/s	956.5722 Ops/s	$\color{#d91a1a}-2.68\%$
test_serialize_model	0.1297s	0.1293s	7.7346 Ops/s	7.7234 Ops/s	$\color{#35bf28}+0.15\%$
test_serialize_model_pickle	1.3471s	1.2124s	0.8248 Ops/s	0.8244 Ops/s	$\color{#35bf28}+0.05\%$
test_serialize_weights	0.1303s	0.1292s	7.7380 Ops/s	7.7806 Ops/s	$\color{#d91a1a}-0.55\%$
test_serialize_weights_returnearly	0.2345s	61.6664ms	16.2163 Ops/s	16.2943 Ops/s	$\color{#d91a1a}-0.48\%$
test_serialize_weights_pickle	1.3524s	1.2135s	0.8240 Ops/s	0.8209 Ops/s	$\color{#35bf28}+0.38\%$
test_reshape_pytree	66.3910μs	36.8316μs	27.1506 KOps/s	27.6182 KOps/s	$\color{#d91a1a}-1.69\%$
test_reshape_td	89.9720μs	45.2668μs	22.0913 KOps/s	23.2430 KOps/s	$\color{#d91a1a}-4.96\%$
test_view_pytree	68.2410μs	35.9189μs	27.8405 KOps/s	28.1817 KOps/s	$\color{#d91a1a}-1.21\%$
test_view_td	95.4510μs	47.9515μs	20.8544 KOps/s	21.2784 KOps/s	$\color{#d91a1a}-1.99\%$
test_unbind_pytree	75.2720μs	34.4896μs	28.9942 KOps/s	29.7519 KOps/s	$\color{#d91a1a}-2.55\%$
test_unbind_td	0.3871ms	44.9867μs	22.2288 KOps/s	23.8940 KOps/s	$\textbf{\color{#d91a1a}-6.97\%}$
test_split_pytree	85.8310μs	46.5420μs	21.4860 KOps/s	21.4458 KOps/s	$\color{#35bf28}+0.19\%$
test_split_td	0.4824ms	57.8054μs	17.2994 KOps/s	17.5927 KOps/s	$\color{#d91a1a}-1.67\%$
test_add_pytree	98.2920μs	57.4744μs	17.3991 KOps/s	17.5601 KOps/s	$\color{#d91a1a}-0.92\%$
test_add_td	0.1805ms	0.1017ms	9.8284 KOps/s	11.0713 KOps/s	$\textbf{\color{#d91a1a}-11.23\%}$
test_compile_add_one_nested[tensordict-compile]	0.4676ms	0.2160ms	4.6304 KOps/s	4.6248 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_add_one_nested[tensordict-eager]	0.2539ms	0.1537ms	6.5080 KOps/s	6.3412 KOps/s	$\color{#35bf28}+2.63\%$
test_compile_add_one_nested[pytree-compile]	0.1923ms	0.1448ms	6.9082 KOps/s	6.9459 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_add_one_nested[pytree-eager]	0.2371ms	0.1821ms	5.4923 KOps/s	5.4446 KOps/s	$\color{#35bf28}+0.88\%$
test_compile_copy_nested[tensordict-compile]	50.5210μs	19.8318μs	50.4240 KOps/s	51.0212 KOps/s	$\color{#d91a1a}-1.17\%$
test_compile_copy_nested[tensordict-eager]	74.9120μs	43.3423μs	23.0722 KOps/s	22.7127 KOps/s	$\color{#35bf28}+1.58\%$
test_compile_copy_nested[pytree-compile]	0.2109ms	63.5327μs	15.7399 KOps/s	15.6124 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_copy_nested[pytree-eager]	74.2210μs	49.6362μs	20.1466 KOps/s	20.3724 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_add_one_flat[tensordict-compile]	0.4740ms	0.3177ms	3.1475 KOps/s	3.0958 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_add_one_flat[tensordict-eager]	0.2845ms	0.2083ms	4.7999 KOps/s	4.7380 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_add_one_flat[tensorclass-compile]	0.1647ms	0.1272ms	7.8637 KOps/s	7.6951 KOps/s	$\color{#35bf28}+2.19\%$
test_compile_add_one_flat[tensorclass-eager]	0.1196ms	62.4608μs	16.0101 KOps/s	16.5492 KOps/s	$\color{#d91a1a}-3.26\%$
test_compile_add_one_flat[pytree-compile]	0.3562ms	0.3199ms	3.1262 KOps/s	3.0897 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_add_one_flat[pytree-eager]	0.6996ms	0.6321ms	1.5820 KOps/s	1.6118 KOps/s	$\color{#d91a1a}-1.85\%$
test_compile_add_self_flat[tensordict-eager]	0.4022ms	0.2494ms	4.0095 KOps/s	3.9653 KOps/s	$\color{#35bf28}+1.11\%$
test_compile_add_self_flat[tensordict-compile]	0.4373ms	0.3289ms	3.0400 KOps/s	3.0656 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_add_self_flat[tensorclass-eager]	0.1820ms	72.3910μs	13.8139 KOps/s	14.1534 KOps/s	$\color{#d91a1a}-2.40\%$
test_compile_add_self_flat[tensorclass-compile]	0.1805ms	0.1284ms	7.7869 KOps/s	7.8208 KOps/s	$\color{#d91a1a}-0.43\%$
test_compile_add_self_flat[pytree-eager]	0.6563ms	0.5328ms	1.8770 KOps/s	1.9155 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_add_self_flat[pytree-compile]	0.4631ms	0.3186ms	3.1386 KOps/s	3.1371 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_copy_flat[tensordict-compile]	72.7710μs	16.7072μs	59.8543 KOps/s	59.5324 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_copy_flat[tensordict-eager]	67.9310μs	27.0106μs	37.0225 KOps/s	36.2575 KOps/s	$\color{#35bf28}+2.11\%$
test_compile_copy_flat[pytree-compile]	0.1328ms	67.9670μs	14.7130 KOps/s	14.7101 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_copy_flat[pytree-eager]	79.4410μs	51.1467μs	19.5516 KOps/s	19.7707 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_assign_and_add[tensordict-compile]	2.4491ms	0.8582ms	1.1652 KOps/s	1.1208 KOps/s	$\color{#35bf28}+3.96\%$
test_compile_assign_and_add[tensordict-eager]	3.8024ms	3.4208ms	292.3261 Ops/s	293.7915 Ops/s	$\color{#d91a1a}-0.50\%$
test_compile_assign_and_add[pytree-compile]	2.3143ms	0.8057ms	1.2412 KOps/s	1.1239 KOps/s	$\textbf{\color{#35bf28}+10.44\%}$
test_compile_assign_and_add[pytree-eager]	3.6209ms	3.3590ms	297.7042 Ops/s	295.4177 Ops/s	$\color{#35bf28}+0.77\%$
test_compile_indexing[tensor-tensordict-compile]	0.1599ms	0.1096ms	9.1216 KOps/s	8.6873 KOps/s	$\textbf{\color{#35bf28}+5.00\%}$
test_compile_indexing[tensor-tensordict-eager]	0.1878ms	62.4773μs	16.0058 KOps/s	15.6076 KOps/s	$\color{#35bf28}+2.55\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2137ms	0.1047ms	9.5501 KOps/s	9.3457 KOps/s	$\color{#35bf28}+2.19\%$
test_compile_indexing[tensor-tensorclass-eager]	85.6720μs	45.5404μs	21.9585 KOps/s	21.1983 KOps/s	$\color{#35bf28}+3.59\%$
test_compile_indexing[tensor-pytree-compile]	0.1476ms	0.1098ms	9.1051 KOps/s	9.3881 KOps/s	$\color{#d91a1a}-3.01\%$
test_compile_indexing[tensor-pytree-eager]	89.2610μs	46.4198μs	21.5425 KOps/s	23.3121 KOps/s	$\textbf{\color{#d91a1a}-7.59\%}$
test_compile_indexing[slice-tensordict-compile]	0.1843ms	0.1389ms	7.2000 KOps/s	7.3642 KOps/s	$\color{#d91a1a}-2.23\%$
test_compile_indexing[slice-tensordict-eager]	0.1602ms	26.9663μs	37.0833 KOps/s	39.7159 KOps/s	$\textbf{\color{#d91a1a}-6.63\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1695ms	0.1328ms	7.5278 KOps/s	7.7232 KOps/s	$\color{#d91a1a}-2.53\%$
test_compile_indexing[slice-tensorclass-eager]	54.8010μs	21.5358μs	46.4344 KOps/s	47.4724 KOps/s	$\color{#d91a1a}-2.19\%$
test_compile_indexing[slice-pytree-compile]	0.1884ms	0.1333ms	7.5017 KOps/s	7.5800 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_indexing[slice-pytree-eager]	64.2110μs	21.9878μs	45.4797 KOps/s	45.4421 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_indexing[int-tensordict-compile]	0.1983ms	0.1442ms	6.9339 KOps/s	7.3019 KOps/s	$\textbf{\color{#d91a1a}-5.04\%}$
test_compile_indexing[int-tensordict-eager]	0.5172ms	26.8010μs	37.3120 KOps/s	40.1243 KOps/s	$\textbf{\color{#d91a1a}-7.01\%}$
test_compile_indexing[int-tensorclass-compile]	0.2727ms	0.1396ms	7.1655 KOps/s	7.6465 KOps/s	$\textbf{\color{#d91a1a}-6.29\%}$
test_compile_indexing[int-tensorclass-eager]	54.1810μs	21.6374μs	46.2163 KOps/s	47.6381 KOps/s	$\color{#d91a1a}-2.98\%$
test_compile_indexing[int-pytree-compile]	0.1767ms	0.1335ms	7.4889 KOps/s	7.6184 KOps/s	$\color{#d91a1a}-1.70\%$
test_compile_indexing[int-pytree-eager]	0.3153ms	21.5385μs	46.4284 KOps/s	46.4633 KOps/s	$\color{#d91a1a}-0.07\%$
test_mod_add[eager]	69.6710μs	32.3219μs	30.9388 KOps/s	32.5283 KOps/s	$\color{#d91a1a}-4.89\%$
test_mod_add[compile]	0.3198ms	73.2024μs	13.6608 KOps/s	14.3438 KOps/s	$\color{#d91a1a}-4.76\%$
test_mod_add[compile-overhead]	0.2588ms	0.1345ms	7.4330 KOps/s	6.6647 KOps/s	$\textbf{\color{#35bf28}+11.53\%}$
test_mod_wrap[eager]	0.3282ms	0.2538ms	3.9406 KOps/s	4.0316 KOps/s	$\color{#d91a1a}-2.26\%$
test_mod_wrap[compile]	0.4661ms	0.2889ms	3.4618 KOps/s	3.4930 KOps/s	$\color{#d91a1a}-0.89\%$
test_mod_wrap[compile-overhead]	7.8104ms	4.1177ms	242.8522 Ops/s	246.4263 Ops/s	$\color{#d91a1a}-1.45\%$
test_mod_wrap_and_backward[eager]	1.5738ms	1.4813ms	675.0685 Ops/s	692.3900 Ops/s	$\color{#d91a1a}-2.50\%$
test_mod_wrap_and_backward[compile]	1.8891ms	1.3584ms	736.1752 Ops/s	704.0151 Ops/s	$\color{#35bf28}+4.57\%$
test_mod_wrap_and_backward[compile-overhead]	1.3017ms	0.8965ms	1.1155 KOps/s	922.4403 Ops/s	$\textbf{\color{#35bf28}+20.92\%}$
test_seq_add[eager]	0.2458ms	96.2563μs	10.3889 KOps/s	10.3370 KOps/s	$\color{#35bf28}+0.50\%$
test_seq_add[compile]	0.5410ms	82.0713μs	12.1845 KOps/s	12.3534 KOps/s	$\color{#d91a1a}-1.37\%$
test_seq_add[compile-overhead]	0.1499ms	0.1135ms	8.8082 KOps/s	8.7691 KOps/s	$\color{#35bf28}+0.45\%$
test_seq_wrap[eager]	0.4476ms	0.3776ms	2.6485 KOps/s	2.4915 KOps/s	$\textbf{\color{#35bf28}+6.30\%}$
test_seq_wrap[compile]	0.3567ms	0.3023ms	3.3080 KOps/s	3.2934 KOps/s	$\color{#35bf28}+0.45\%$
test_seq_wrap[compile-overhead]	0.2526ms	0.2074ms	4.8220 KOps/s	4.7604 KOps/s	$\color{#35bf28}+1.29\%$
test_func_call_runtime[False-eager]	0.8973ms	0.7590ms	1.3175 KOps/s	1.3573 KOps/s	$\color{#d91a1a}-2.93\%$
test_func_call_runtime[False-compile]	1.1762ms	0.7844ms	1.2748 KOps/s	1.2901 KOps/s	$\color{#d91a1a}-1.19\%$
test_func_call_runtime[False-compile-overhead]	0.4277ms	0.3485ms	2.8694 KOps/s	2.8675 KOps/s	$\color{#35bf28}+0.07\%$
test_func_call_runtime[True-eager]	1.1520ms	0.8872ms	1.1271 KOps/s	1.1309 KOps/s	$\color{#d91a1a}-0.33\%$
test_func_call_runtime[True-compile]	0.9000ms	0.8132ms	1.2297 KOps/s	1.2223 KOps/s	$\color{#35bf28}+0.61\%$
test_func_call_runtime[True-compile-overhead]	0.4237ms	0.3847ms	2.5994 KOps/s	2.6080 KOps/s	$\color{#d91a1a}-0.33\%$
test_func_call_cm_runtime[False-eager]	0.7750ms	0.7234ms	1.3823 KOps/s	1.3707 KOps/s	$\color{#35bf28}+0.85\%$
test_func_call_cm_runtime[False-compile]	0.8270ms	0.7769ms	1.2871 KOps/s	1.2843 KOps/s	$\color{#35bf28}+0.22\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4008ms	0.3507ms	2.8518 KOps/s	2.8523 KOps/s	$\color{#d91a1a}-0.02\%$
test_func_call_cm_runtime[True-eager]	1.1124ms	0.9788ms	1.0217 KOps/s	1.0111 KOps/s	$\color{#35bf28}+1.05\%$
test_func_call_cm_runtime[True-compile]	0.8913ms	0.8403ms	1.1901 KOps/s	1.1821 KOps/s	$\color{#35bf28}+0.68\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4666ms	0.4110ms	2.4330 KOps/s	2.4421 KOps/s	$\color{#d91a1a}-0.37\%$
test_vmap_func_call_cm_runtime[eager]	2.5911ms	2.0585ms	485.7826 Ops/s	479.7826 Ops/s	$\color{#35bf28}+1.25\%$
test_vmap_func_call_cm_runtime[compile]	0.9354ms	0.8554ms	1.1690 KOps/s	1.1646 KOps/s	$\color{#35bf28}+0.38\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4734ms	0.4152ms	2.4087 KOps/s	2.4023 KOps/s	$\color{#35bf28}+0.27\%$
test_distributed	6.7244ms	0.2536ms	3.9429 KOps/s	8.9439 KOps/s	$\textbf{\color{#d91a1a}-55.91\%}$
test_tdmodule	0.3119ms	15.8159μs	63.2276 KOps/s	63.4843 KOps/s	$\color{#d91a1a}-0.40\%$
test_tdmodule_dispatch	53.8210μs	31.3779μs	31.8696 KOps/s	33.2553 KOps/s	$\color{#d91a1a}-4.17\%$
test_tdseq	23.7200μs	15.7014μs	63.6884 KOps/s	65.0632 KOps/s	$\color{#d91a1a}-2.11\%$
test_tdseq_dispatch	67.0610μs	33.4382μs	29.9059 KOps/s	31.1460 KOps/s	$\color{#d91a1a}-3.98\%$
test_instantiation_functorch	1.8989ms	1.8340ms	545.2546 Ops/s	538.4933 Ops/s	$\color{#35bf28}+1.26\%$
test_instantiation_td	1.7502ms	1.1716ms	853.5102 Ops/s	842.3176 Ops/s	$\color{#35bf28}+1.33\%$
test_exec_functorch	0.2395ms	0.2048ms	4.8818 KOps/s	4.9074 KOps/s	$\color{#d91a1a}-0.52\%$
test_exec_functional_call	0.2489ms	0.2025ms	4.9392 KOps/s	4.9449 KOps/s	$\color{#d91a1a}-0.12\%$
test_exec_td	0.2665ms	0.2080ms	4.8073 KOps/s	4.7482 KOps/s	$\color{#35bf28}+1.24\%$
test_exec_td_decorator	0.6509ms	0.2501ms	3.9979 KOps/s	3.9888 KOps/s	$\color{#35bf28}+0.23\%$
test_vmap_mlp_speed[True-True]	0.7339ms	0.6838ms	1.4624 KOps/s	1.4444 KOps/s	$\color{#35bf28}+1.25\%$
test_vmap_mlp_speed[True-False]	0.7244ms	0.6829ms	1.4642 KOps/s	1.4483 KOps/s	$\color{#35bf28}+1.10\%$
test_vmap_mlp_speed[False-True]	0.6785ms	0.5703ms	1.7535 KOps/s	1.7267 KOps/s	$\color{#35bf28}+1.55\%$
test_vmap_mlp_speed[False-False]	1.1091ms	0.5734ms	1.7441 KOps/s	1.7264 KOps/s	$\color{#35bf28}+1.02\%$
test_vmap_mlp_speed_decorator[True-True]	0.8075ms	0.6678ms	1.4975 KOps/s	1.4877 KOps/s	$\color{#35bf28}+0.66\%$
test_vmap_mlp_speed_decorator[True-False]	0.8379ms	0.6686ms	1.4957 KOps/s	1.4840 KOps/s	$\color{#35bf28}+0.78\%$
test_vmap_mlp_speed_decorator[False-True]	0.6972ms	0.5838ms	1.7129 KOps/s	1.6919 KOps/s	$\color{#35bf28}+1.24\%$
test_vmap_mlp_speed_decorator[False-False]	0.7169ms	0.5849ms	1.7097 KOps/s	1.6897 KOps/s	$\color{#35bf28}+1.18\%$
test_vmap_transformer_speed[True-True]	8.3816ms	8.3235ms	120.1417 Ops/s	118.2983 Ops/s	$\color{#35bf28}+1.56\%$
test_vmap_transformer_speed[True-False]	8.3503ms	8.2847ms	120.7040 Ops/s	118.7160 Ops/s	$\color{#35bf28}+1.67\%$
test_vmap_transformer_speed[False-True]	8.3206ms	8.1616ms	122.5257 Ops/s	121.8498 Ops/s	$\color{#35bf28}+0.55\%$
test_vmap_transformer_speed[False-False]	9.4850ms	8.1492ms	122.7108 Ops/s	121.9846 Ops/s	$\color{#35bf28}+0.60\%$
test_vmap_transformer_speed_decorator[True-True]	19.9014ms	19.4481ms	51.4189 Ops/s	51.0155 Ops/s	$\color{#35bf28}+0.79\%$
test_vmap_transformer_speed_decorator[True-False]	19.5552ms	19.4986ms	51.2858 Ops/s	51.0679 Ops/s	$\color{#35bf28}+0.43\%$
test_vmap_transformer_speed_decorator[False-True]	19.4039ms	19.3477ms	51.6858 Ops/s	51.5867 Ops/s	$\color{#35bf28}+0.19\%$
test_vmap_transformer_speed_decorator[False-False]	19.3963ms	19.3460ms	51.6902 Ops/s	51.4161 Ops/s	$\color{#35bf28}+0.53\%$
test_to_module_speed[True]	1.4585ms	0.9321ms	1.0729 KOps/s	1.0949 KOps/s	$\color{#d91a1a}-2.01\%$
test_to_module_speed[False]	1.2956ms	0.9102ms	1.0986 KOps/s	1.1164 KOps/s	$\color{#d91a1a}-1.59\%$
test_tc_init	62.6310μs	35.2937μs	28.3337 KOps/s	28.5388 KOps/s	$\color{#d91a1a}-0.72\%$
test_tc_init_nested	0.1124ms	69.7971μs	14.3272 KOps/s	14.1951 KOps/s	$\color{#35bf28}+0.93\%$
test_tc_first_layer_tensor	4.4501μs	0.6731μs	1.4856 MOps/s	1.4854 MOps/s	$\color{#35bf28}+0.01\%$
test_tc_first_layer_nontensor	18.1700μs	2.2334μs	447.7573 KOps/s	443.4329 KOps/s	$\color{#35bf28}+0.98\%$
test_tc_second_layer_tensor	7.7328μs	1.3586μs	736.0419 KOps/s	731.5705 KOps/s	$\color{#35bf28}+0.61\%$
test_tc_second_layer_nontensor	22.3900μs	2.9337μs	340.8611 KOps/s	337.8062 KOps/s	$\color{#35bf28}+0.90\%$
test_unbind	0.1936s	11.9639ms	83.5849 Ops/s	93.6045 Ops/s	$\textbf{\color{#d91a1a}-10.70\%}$
test_full_like	0.6525ms	0.5749ms	1.7393 KOps/s	1.7429 KOps/s	$\color{#d91a1a}-0.20\%$
test_zeros_like	0.2605ms	0.1979ms	5.0533 KOps/s	5.0499 KOps/s	$\color{#35bf28}+0.07\%$
test_ones_like	0.2370ms	0.1978ms	5.0561 KOps/s	5.0547 KOps/s	$\color{#35bf28}+0.03\%$
test_clone	0.4423ms	0.4144ms	2.4131 KOps/s	2.4117 KOps/s	$\color{#35bf28}+0.06\%$
test_squeeze	34.6700μs	9.3590μs	106.8487 KOps/s	106.5762 KOps/s	$\color{#35bf28}+0.26\%$
test_unsqueeze	0.2164ms	69.8174μs	14.3231 KOps/s	13.6426 KOps/s	$\color{#35bf28}+4.99\%$
test_split	0.3844ms	0.1538ms	6.5002 KOps/s	6.4517 KOps/s	$\color{#35bf28}+0.75\%$
test_permute	0.2239ms	0.1788ms	5.5923 KOps/s	5.6280 KOps/s	$\color{#d91a1a}-0.63\%$
test_stack	1.2461ms	0.8665ms	1.1541 KOps/s	1.1522 KOps/s	$\color{#35bf28}+0.16\%$
test_cat	1.2503ms	1.2319ms	811.7634 Ops/s	812.0474 Ops/s	$\color{#d91a1a}-0.03\%$