[Feature] tensorclass nocast

[vmoens]

Stack from ghstack (oldest at bottom):

• [Doc] Better docstring for to_module #1081
• [Feature] allow tensorclass to be customized #1080
• -> [Feature] tensorclass nocast #1079

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}26$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	35.7370μs	18.0181μs	55.4998 KOps/s	52.4439 KOps/s	$\textbf{\color{#35bf28}+5.83\%}$
test_plain_set_stack_nested	44.1330μs	18.1879μs	54.9815 KOps/s	53.1036 KOps/s	$\color{#35bf28}+3.54\%$
test_plain_set_nested_inplace	77.3650μs	19.3381μs	51.7114 KOps/s	47.7484 KOps/s	$\textbf{\color{#35bf28}+8.30\%}$
test_plain_set_stack_nested_inplace	52.4980μs	19.5185μs	51.2335 KOps/s	47.8676 KOps/s	$\textbf{\color{#35bf28}+7.03\%}$
test_items	31.7500μs	4.5180μs	221.3357 KOps/s	246.7659 KOps/s	$\textbf{\color{#d91a1a}-10.31\%}$
test_items_nested	0.4567ms	0.3480ms	2.8735 KOps/s	2.9142 KOps/s	$\color{#d91a1a}-1.40\%$
test_items_nested_locked	0.6132ms	0.3475ms	2.8777 KOps/s	2.9105 KOps/s	$\color{#d91a1a}-1.13\%$
test_items_nested_leaf	0.1496ms	72.5862μs	13.7767 KOps/s	14.0442 KOps/s	$\color{#d91a1a}-1.90\%$
test_items_stack_nested	0.4428ms	0.3520ms	2.8406 KOps/s	2.8728 KOps/s	$\color{#d91a1a}-1.12\%$
test_items_stack_nested_leaf	0.1756ms	75.2691μs	13.2857 KOps/s	13.4260 KOps/s	$\color{#d91a1a}-1.05\%$
test_items_stack_nested_locked	0.4245ms	0.3517ms	2.8432 KOps/s	2.8986 KOps/s	$\color{#d91a1a}-1.91\%$
test_keys	21.2900μs	3.5411μs	282.3969 KOps/s	286.4285 KOps/s	$\color{#d91a1a}-1.41\%$
test_keys_nested	0.2270ms	0.1355ms	7.3806 KOps/s	7.3986 KOps/s	$\color{#d91a1a}-0.24\%$
test_keys_nested_locked	2.2349ms	0.1412ms	7.0832 KOps/s	7.0934 KOps/s	$\color{#d91a1a}-0.14\%$
test_keys_nested_leaf	0.2274ms	0.1166ms	8.5791 KOps/s	8.6259 KOps/s	$\color{#d91a1a}-0.54\%$
test_keys_stack_nested	0.2201ms	0.1393ms	7.1763 KOps/s	7.4157 KOps/s	$\color{#d91a1a}-3.23\%$
test_keys_stack_nested_leaf	0.2738ms	0.1187ms	8.4274 KOps/s	8.7934 KOps/s	$\color{#d91a1a}-4.16\%$
test_keys_stack_nested_locked	0.2359ms	0.1422ms	7.0332 KOps/s	7.2016 KOps/s	$\color{#d91a1a}-2.34\%$
test_values	6.3080μs	1.0707μs	933.9942 KOps/s	937.3920 KOps/s	$\color{#d91a1a}-0.36\%$
test_values_nested	0.1102ms	56.5088μs	17.6964 KOps/s	17.9312 KOps/s	$\color{#d91a1a}-1.31\%$
test_values_nested_locked	0.1029ms	56.4108μs	17.7271 KOps/s	18.0758 KOps/s	$\color{#d91a1a}-1.93\%$
test_values_nested_leaf	0.1254ms	61.0482μs	16.3805 KOps/s	15.5969 KOps/s	$\textbf{\color{#35bf28}+5.02\%}$
test_values_stack_nested	0.1056ms	56.6767μs	17.6439 KOps/s	17.6217 KOps/s	$\color{#35bf28}+0.13\%$
test_values_stack_nested_leaf	0.1465ms	61.8227μs	16.1753 KOps/s	16.7037 KOps/s	$\color{#d91a1a}-3.16\%$
test_values_stack_nested_locked	0.1087ms	56.9754μs	17.5514 KOps/s	17.4581 KOps/s	$\color{#35bf28}+0.53\%$
test_membership	3.9374μs	0.7422μs	1.3473 MOps/s	1.0910 MOps/s	$\textbf{\color{#35bf28}+23.49\%}$
test_membership_nested	28.8440μs	2.7177μs	367.9604 KOps/s	355.9888 KOps/s	$\color{#35bf28}+3.36\%$
test_membership_nested_leaf	31.8690μs	2.7036μs	369.8782 KOps/s	352.2468 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_membership_stacked_nested	24.2250μs	2.6992μs	370.4805 KOps/s	355.8600 KOps/s	$\color{#35bf28}+4.11\%$
test_membership_stacked_nested_leaf	23.5640μs	2.7154μs	368.2653 KOps/s	357.8738 KOps/s	$\color{#35bf28}+2.90\%$
test_membership_nested_last	32.3610μs	4.0883μs	244.5975 KOps/s	237.3680 KOps/s	$\color{#35bf28}+3.05\%$
test_membership_nested_leaf_last	20.8790μs	4.0915μs	244.4117 KOps/s	239.7337 KOps/s	$\color{#35bf28}+1.95\%$
test_membership_stacked_nested_last	46.3570μs	4.0609μs	246.2497 KOps/s	74.6666 KOps/s	$\textbf{\color{#35bf28}+229.80\%}$
test_membership_stacked_nested_leaf_last	27.8730μs	4.0481μs	247.0281 KOps/s	75.0461 KOps/s	$\textbf{\color{#35bf28}+229.17\%}$
test_nested_getleaf	33.0520μs	10.7467μs	93.0516 KOps/s	92.8077 KOps/s	$\color{#35bf28}+0.26\%$
test_nested_get	36.5790μs	10.0614μs	99.3894 KOps/s	98.1616 KOps/s	$\color{#35bf28}+1.25\%$
test_stacked_getleaf	31.8000μs	10.8578μs	92.0998 KOps/s	94.1373 KOps/s	$\color{#d91a1a}-2.16\%$
test_stacked_get	32.3800μs	10.1540μs	98.4830 KOps/s	99.6276 KOps/s	$\color{#d91a1a}-1.15\%$
test_nested_getitemleaf	38.3620μs	10.8422μs	92.2326 KOps/s	88.5686 KOps/s	$\color{#35bf28}+4.14\%$
test_nested_getitem	34.5450μs	10.4400μs	95.7853 KOps/s	93.0703 KOps/s	$\color{#35bf28}+2.92\%$
test_stacked_getitemleaf	37.5910μs	10.9264μs	91.5211 KOps/s	90.0914 KOps/s	$\color{#35bf28}+1.59\%$
test_stacked_getitem	29.2150μs	9.9537μs	100.4649 KOps/s	94.9927 KOps/s	$\textbf{\color{#35bf28}+5.76\%}$
test_lock_nested	4.9633ms	0.4484ms	2.2301 KOps/s	2.2478 KOps/s	$\color{#d91a1a}-0.78\%$
test_lock_stack_nested	0.6533ms	0.4128ms	2.4223 KOps/s	2.4883 KOps/s	$\color{#d91a1a}-2.65\%$
test_unlock_nested	0.6685ms	0.3588ms	2.7867 KOps/s	2.7385 KOps/s	$\color{#35bf28}+1.76\%$
test_unlock_stack_nested	0.5045ms	0.3298ms	3.0321 KOps/s	3.0853 KOps/s	$\color{#d91a1a}-1.73\%$
test_flatten_speed	0.2007ms	93.2183μs	10.7275 KOps/s	10.7470 KOps/s	$\color{#d91a1a}-0.18\%$
test_unflatten_speed	0.6010ms	0.4680ms	2.1368 KOps/s	2.1287 KOps/s	$\color{#35bf28}+0.38\%$
test_common_ops	2.5378ms	0.7886ms	1.2681 KOps/s	1.2217 KOps/s	$\color{#35bf28}+3.80\%$
test_creation	16.2010μs	2.0777μs	481.3021 KOps/s	481.3593 KOps/s	$\color{#d91a1a}-0.01\%$
test_creation_empty	28.8140μs	10.4359μs	95.8230 KOps/s	76.2920 KOps/s	$\textbf{\color{#35bf28}+25.60\%}$
test_creation_nested_1	53.0290μs	13.4403μs	74.4030 KOps/s	61.2435 KOps/s	$\textbf{\color{#35bf28}+21.49\%}$
test_creation_nested_2	38.4620μs	17.4525μs	57.2984 KOps/s	49.9053 KOps/s	$\textbf{\color{#35bf28}+14.81\%}$
test_clone	51.0160μs	13.3779μs	74.7504 KOps/s	76.5482 KOps/s	$\color{#d91a1a}-2.35\%$
test_getitem[int]	1.5646ms	12.9895μs	76.9853 KOps/s	79.9085 KOps/s	$\color{#d91a1a}-3.66\%$
test_getitem[slice_int]	0.1511ms	24.3357μs	41.0919 KOps/s	42.2951 KOps/s	$\color{#d91a1a}-2.84\%$
test_getitem[range]	0.1777ms	47.7756μs	20.9312 KOps/s	20.0662 KOps/s	$\color{#35bf28}+4.31\%$
test_getitem[tuple]	0.1361ms	19.9206μs	50.1993 KOps/s	45.6384 KOps/s	$\textbf{\color{#35bf28}+9.99\%}$
test_getitem[list]	0.1993ms	43.4672μs	23.0059 KOps/s	22.3212 KOps/s	$\color{#35bf28}+3.07\%$
test_setitem_dim[int]	56.5660μs	26.3314μs	37.9775 KOps/s	40.9247 KOps/s	$\textbf{\color{#d91a1a}-7.20\%}$
test_setitem_dim[slice_int]	92.1430μs	51.9253μs	19.2584 KOps/s	19.8486 KOps/s	$\color{#d91a1a}-2.97\%$
test_setitem_dim[range]	0.1191ms	73.7480μs	13.5597 KOps/s	13.1639 KOps/s	$\color{#35bf28}+3.01\%$
test_setitem_dim[tuple]	88.8370μs	41.8637μs	23.8870 KOps/s	25.0037 KOps/s	$\color{#d91a1a}-4.47\%$
test_setitem	76.7140μs	20.8926μs	47.8639 KOps/s	46.0013 KOps/s	$\color{#35bf28}+4.05\%$
test_set	69.4010μs	20.0432μs	49.8922 KOps/s	47.6191 KOps/s	$\color{#35bf28}+4.77\%$
test_set_shared	3.4956ms	0.1677ms	5.9638 KOps/s	5.8681 KOps/s	$\color{#35bf28}+1.63\%$
test_update	0.2046ms	22.6901μs	44.0721 KOps/s	39.9354 KOps/s	$\textbf{\color{#35bf28}+10.36\%}$
test_update_nested	83.4770μs	33.6634μs	29.7058 KOps/s	28.4225 KOps/s	$\color{#35bf28}+4.52\%$
test_update__nested	1.0879ms	33.6165μs	29.7473 KOps/s	31.1592 KOps/s	$\color{#d91a1a}-4.53\%$
test_set_nested	67.1260μs	22.4872μs	44.4698 KOps/s	42.9145 KOps/s	$\color{#35bf28}+3.62\%$
test_set_nested_new	68.8890μs	27.6203μs	36.2053 KOps/s	36.0056 KOps/s	$\color{#35bf28}+0.55\%$
test_select	0.1002ms	43.9074μs	22.7752 KOps/s	22.9555 KOps/s	$\color{#d91a1a}-0.79\%$
test_select_nested	0.1370ms	59.9419μs	16.6828 KOps/s	16.7120 KOps/s	$\color{#d91a1a}-0.17\%$
test_exclude_nested	0.3710ms	75.1667μs	13.3038 KOps/s	13.4006 KOps/s	$\color{#d91a1a}-0.72\%$
test_empty[True]	0.6592ms	0.3511ms	2.8479 KOps/s	2.8242 KOps/s	$\color{#35bf28}+0.84\%$
test_empty[False]	7.0408μs	1.2248μs	816.4775 KOps/s	817.3841 KOps/s	$\color{#d91a1a}-0.11\%$
test_unbind_speed	0.4559ms	0.2628ms	3.8058 KOps/s	3.7455 KOps/s	$\color{#35bf28}+1.61\%$
test_unbind_speed_stack0	0.5421ms	0.2597ms	3.8508 KOps/s	3.9276 KOps/s	$\color{#d91a1a}-1.96\%$
test_unbind_speed_stack1	0.1033s	0.7727ms	1.2942 KOps/s	1.4623 KOps/s	$\textbf{\color{#d91a1a}-11.49\%}$
test_split	2.4262ms	1.5543ms	643.3880 Ops/s	585.2477 Ops/s	$\textbf{\color{#35bf28}+9.93\%}$
test_chunk	0.1029s	1.8628ms	536.8221 Ops/s	576.8334 Ops/s	$\textbf{\color{#d91a1a}-6.94\%}$
test_consolidate_njt[False-None]	10.5504ms	8.3304ms	120.0429 Ops/s	119.8952 Ops/s	$\color{#35bf28}+0.12\%$
test_creation[device0]	3.3934ms	91.8440μs	10.8880 KOps/s	10.8246 KOps/s	$\color{#35bf28}+0.59\%$
test_creation_from_tensor	0.2199ms	92.1050μs	10.8572 KOps/s	10.5566 KOps/s	$\color{#35bf28}+2.85\%$
test_add_one[memmap_tensor0]	0.1278ms	4.8846μs	204.7261 KOps/s	196.6347 KOps/s	$\color{#35bf28}+4.11\%$
test_contiguous[memmap_tensor0]	11.2810μs	0.5308μs	1.8838 MOps/s	1.9696 MOps/s	$\color{#d91a1a}-4.36\%$
test_stack[memmap_tensor0]	39.9950μs	3.6128μs	276.7932 KOps/s	286.3311 KOps/s	$\color{#d91a1a}-3.33\%$
test_memmaptd_index	1.0595ms	0.2446ms	4.0884 KOps/s	4.2273 KOps/s	$\color{#d91a1a}-3.29\%$
test_memmaptd_index_astensor	0.5818ms	0.3215ms	3.1103 KOps/s	3.1892 KOps/s	$\color{#d91a1a}-2.47\%$
test_memmaptd_index_op	1.3277ms	0.5959ms	1.6780 KOps/s	1.6001 KOps/s	$\color{#35bf28}+4.87\%$
test_serialize_model	0.1233s	0.1142s	8.7575 Ops/s	7.7501 Ops/s	$\textbf{\color{#35bf28}+13.00\%}$
test_serialize_model_pickle	0.4459s	0.3909s	2.5582 Ops/s	2.5628 Ops/s	$\color{#d91a1a}-0.18\%$
test_serialize_weights	0.1184s	0.1127s	8.8735 Ops/s	8.7927 Ops/s	$\color{#35bf28}+0.92\%$
test_serialize_weights_returnearly	0.3712s	0.1838s	5.4421 Ops/s	6.3134 Ops/s	$\textbf{\color{#d91a1a}-13.80\%}$
test_serialize_weights_pickle	0.5584s	0.4259s	2.3477 Ops/s	2.4156 Ops/s	$\color{#d91a1a}-2.81\%$
test_serialize_weights_filesystem	0.1514s	0.1417s	7.0547 Ops/s	7.0884 Ops/s	$\color{#d91a1a}-0.48\%$
test_serialize_model_filesystem	0.2475s	0.1577s	6.3425 Ops/s	6.5121 Ops/s	$\color{#d91a1a}-2.61\%$
test_reshape_pytree	57.0270μs	27.7441μs	36.0438 KOps/s	37.2882 KOps/s	$\color{#d91a1a}-3.34\%$
test_reshape_td	66.2240μs	33.6085μs	29.7543 KOps/s	29.6472 KOps/s	$\color{#35bf28}+0.36\%$
test_view_pytree	69.2100μs	27.9172μs	35.8202 KOps/s	36.7701 KOps/s	$\color{#d91a1a}-2.58\%$
test_view_td	79.4390μs	38.7276μs	25.8214 KOps/s	25.8395 KOps/s	$\color{#d91a1a}-0.07\%$
test_unbind_pytree	64.9220μs	30.9231μs	32.3382 KOps/s	32.8046 KOps/s	$\color{#d91a1a}-1.42\%$
test_unbind_td	0.3695ms	38.8446μs	25.7436 KOps/s	25.2122 KOps/s	$\color{#35bf28}+2.11\%$
test_split_pytree	68.9390μs	30.4431μs	32.8481 KOps/s	33.2388 KOps/s	$\color{#d91a1a}-1.18\%$
test_split_td	0.2072ms	44.1554μs	22.6473 KOps/s	22.3341 KOps/s	$\color{#35bf28}+1.40\%$
test_add_pytree	83.9980μs	36.7528μs	27.2088 KOps/s	27.7711 KOps/s	$\color{#d91a1a}-2.02\%$
test_add_td	0.1105ms	57.5790μs	17.3675 KOps/s	16.4651 KOps/s	$\textbf{\color{#35bf28}+5.48\%}$
test_compile_add_one_nested[tensordict-compile]	0.1167ms	62.6911μs	15.9512 KOps/s	15.7283 KOps/s	$\color{#35bf28}+1.42\%$
test_compile_add_one_nested[tensordict-eager]	0.3589ms	0.1618ms	6.1798 KOps/s	6.1963 KOps/s	$\color{#d91a1a}-0.27\%$
test_compile_add_one_nested[pytree-compile]	0.1217ms	45.5715μs	21.9435 KOps/s	21.1877 KOps/s	$\color{#35bf28}+3.57\%$
test_compile_add_one_nested[pytree-eager]	0.2368ms	0.1208ms	8.2799 KOps/s	8.4510 KOps/s	$\color{#d91a1a}-2.02\%$
test_compile_copy_nested[tensordict-compile]	75.1110μs	26.4905μs	37.7493 KOps/s	38.2204 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_copy_nested[tensordict-eager]	0.1231ms	53.5416μs	18.6771 KOps/s	18.4240 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_copy_nested[pytree-compile]	0.1539ms	80.7230μs	12.3880 KOps/s	12.3112 KOps/s	$\color{#35bf28}+0.62\%$
test_compile_copy_nested[pytree-eager]	0.1398ms	70.0590μs	14.2737 KOps/s	14.5081 KOps/s	$\color{#d91a1a}-1.62\%$
test_compile_add_one_flat[tensordict-compile]	0.1683ms	0.1043ms	9.5873 KOps/s	9.3456 KOps/s	$\color{#35bf28}+2.59\%$
test_compile_add_one_flat[tensordict-eager]	0.3914ms	0.1980ms	5.0515 KOps/s	5.0330 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_add_one_flat[tensorclass-compile]	0.1005ms	45.1260μs	22.1602 KOps/s	21.6256 KOps/s	$\color{#35bf28}+2.47\%$
test_compile_add_one_flat[tensorclass-eager]	0.4575ms	63.1559μs	15.8338 KOps/s	16.0914 KOps/s	$\color{#d91a1a}-1.60\%$
test_compile_add_one_flat[pytree-compile]	0.2370ms	0.1058ms	9.4535 KOps/s	9.6551 KOps/s	$\color{#d91a1a}-2.09\%$
test_compile_add_one_flat[pytree-eager]	0.4524ms	0.2041ms	4.8988 KOps/s	4.9792 KOps/s	$\color{#d91a1a}-1.61\%$
test_compile_add_self_flat[tensordict-eager]	0.4236ms	0.2108ms	4.7433 KOps/s	4.7039 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_add_self_flat[tensordict-compile]	0.2052ms	0.1072ms	9.3244 KOps/s	9.4668 KOps/s	$\color{#d91a1a}-1.50\%$
test_compile_add_self_flat[tensorclass-eager]	0.1256ms	57.4110μs	17.4183 KOps/s	18.0221 KOps/s	$\color{#d91a1a}-3.35\%$
test_compile_add_self_flat[tensorclass-compile]	94.6480μs	45.5939μs	21.9328 KOps/s	21.5944 KOps/s	$\color{#35bf28}+1.57\%$
test_compile_add_self_flat[pytree-eager]	0.6358ms	0.1598ms	6.2592 KOps/s	6.2932 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_add_self_flat[pytree-compile]	0.1973ms	0.1029ms	9.7166 KOps/s	9.4741 KOps/s	$\color{#35bf28}+2.56\%$
test_compile_copy_flat[tensordict-compile]	59.4310μs	21.6171μs	46.2597 KOps/s	48.0648 KOps/s	$\color{#d91a1a}-3.76\%$
test_compile_copy_flat[tensordict-eager]	0.1080ms	58.6696μs	17.0446 KOps/s	16.8261 KOps/s	$\color{#35bf28}+1.30\%$
test_compile_copy_flat[pytree-compile]	0.1581ms	83.2630μs	12.0101 KOps/s	12.0159 KOps/s	$\color{#d91a1a}-0.05\%$
test_compile_copy_flat[pytree-eager]	0.1388ms	71.0704μs	14.0706 KOps/s	14.0953 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_assign_and_add[tensordict-compile]	0.3291ms	0.2080ms	4.8069 KOps/s	4.7692 KOps/s	$\color{#35bf28}+0.79\%$
test_compile_assign_and_add[tensordict-eager]	2.1753ms	1.3477ms	741.9890 Ops/s	746.0653 Ops/s	$\color{#d91a1a}-0.55\%$
test_compile_assign_and_add[pytree-compile]	0.4413ms	0.2025ms	4.9382 KOps/s	4.8180 KOps/s	$\color{#35bf28}+2.49\%$
test_compile_assign_and_add[pytree-eager]	1.0158ms	0.7850ms	1.2738 KOps/s	1.2798 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_assign_and_add_stack[compile]	0.7601ms	0.4571ms	2.1879 KOps/s	2.2218 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_assign_and_add_stack[eager]	2.8468ms	2.6271ms	380.6510 Ops/s	350.1844 Ops/s	$\textbf{\color{#35bf28}+8.70\%}$
test_compile_indexing[tensor-tensordict-compile]	88.8870μs	36.0294μs	27.7551 KOps/s	27.1504 KOps/s	$\color{#35bf28}+2.23\%$
test_compile_indexing[tensor-tensordict-eager]	0.5489ms	31.2384μs	32.0119 KOps/s	29.0147 KOps/s	$\textbf{\color{#35bf28}+10.33\%}$
test_compile_indexing[tensor-tensorclass-compile]	75.5620μs	29.5701μs	33.8179 KOps/s	32.9327 KOps/s	$\color{#35bf28}+2.69\%$
test_compile_indexing[tensor-tensorclass-eager]	91.7120μs	23.3451μs	42.8356 KOps/s	42.1537 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_indexing[tensor-pytree-compile]	2.2067ms	30.4486μs	32.8422 KOps/s	31.7888 KOps/s	$\color{#35bf28}+3.31\%$
test_compile_indexing[tensor-pytree-eager]	77.7860μs	23.2813μs	42.9529 KOps/s	42.1495 KOps/s	$\color{#35bf28}+1.91\%$
test_compile_indexing[slice-tensordict-compile]	0.4310ms	55.1900μs	18.1192 KOps/s	19.4207 KOps/s	$\textbf{\color{#d91a1a}-6.70\%}$
test_compile_indexing[slice-tensordict-eager]	0.5029ms	18.8396μs	53.0796 KOps/s	49.8230 KOps/s	$\textbf{\color{#35bf28}+6.54\%}$
test_compile_indexing[slice-tensorclass-compile]	91.1320μs	44.6250μs	22.4090 KOps/s	22.5956 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_indexing[slice-tensorclass-eager]	49.1530μs	18.9955μs	52.6439 KOps/s	52.7944 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_indexing[slice-pytree-compile]	0.1005ms	45.5552μs	21.9514 KOps/s	21.7221 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_indexing[slice-pytree-eager]	70.8830μs	19.0254μs	52.5612 KOps/s	52.5559 KOps/s	$\color{#35bf28}+0.01\%$
test_compile_indexing[int-tensordict-compile]	0.1208ms	52.7771μs	18.9476 KOps/s	18.8859 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_indexing[int-tensordict-eager]	0.8739ms	18.8672μs	53.0021 KOps/s	50.5193 KOps/s	$\color{#35bf28}+4.91\%$
test_compile_indexing[int-tensorclass-compile]	0.4028ms	46.0212μs	21.7291 KOps/s	21.8814 KOps/s	$\color{#d91a1a}-0.70\%$
test_compile_indexing[int-tensorclass-eager]	48.4300μs	19.0080μs	52.6094 KOps/s	53.1703 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_indexing[int-pytree-compile]	0.1201ms	45.3711μs	22.0405 KOps/s	21.2772 KOps/s	$\color{#35bf28}+3.59\%$
test_compile_indexing[int-pytree-eager]	60.1830μs	19.0096μs	52.6049 KOps/s	52.4292 KOps/s	$\color{#35bf28}+0.34\%$
test_mod_add[eager]	76.0930μs	26.7289μs	37.4127 KOps/s	36.0345 KOps/s	$\color{#35bf28}+3.82\%$
test_mod_add[compile]	99.0050μs	45.6401μs	21.9106 KOps/s	21.2927 KOps/s	$\color{#35bf28}+2.90\%$
test_mod_add[compile-overhead]	93.1640μs	46.0403μs	21.7201 KOps/s	21.2051 KOps/s	$\color{#35bf28}+2.43\%$
test_mod_wrap[eager]	0.3709ms	0.2140ms	4.6732 KOps/s	4.5774 KOps/s	$\color{#35bf28}+2.09\%$
test_mod_wrap[compile]	1.6736ms	0.2013ms	4.9676 KOps/s	4.8223 KOps/s	$\color{#35bf28}+3.01\%$
test_mod_wrap[compile-overhead]	1.3655ms	0.1992ms	5.0197 KOps/s	4.8011 KOps/s	$\color{#35bf28}+4.55\%$
test_mod_wrap_and_backward[eager]	13.0992ms	11.3379ms	88.1999 Ops/s	86.1489 Ops/s	$\color{#35bf28}+2.38\%$
test_mod_wrap_and_backward[compile]	14.5619ms	12.3259ms	81.1303 Ops/s	82.3767 Ops/s	$\color{#d91a1a}-1.51\%$
test_mod_wrap_and_backward[compile-overhead]	17.5416ms	12.0633ms	82.8957 Ops/s	79.2946 Ops/s	$\color{#35bf28}+4.54\%$
test_seq_add[eager]	0.1656ms	92.9535μs	10.7581 KOps/s	10.4442 KOps/s	$\color{#35bf28}+3.01\%$
test_seq_add[compile]	0.1330ms	61.3547μs	16.2987 KOps/s	16.1934 KOps/s	$\color{#35bf28}+0.65\%$
test_seq_add[compile-overhead]	0.1109ms	58.6007μs	17.0646 KOps/s	16.4461 KOps/s	$\color{#35bf28}+3.76\%$
test_seq_wrap[eager]	0.6324ms	0.3897ms	2.5663 KOps/s	2.4438 KOps/s	$\textbf{\color{#35bf28}+5.02\%}$
test_seq_wrap[compile]	0.3130ms	0.2234ms	4.4772 KOps/s	4.3869 KOps/s	$\color{#35bf28}+2.06\%$
test_seq_wrap[compile-overhead]	0.4161ms	0.2233ms	4.4789 KOps/s	4.3871 KOps/s	$\color{#35bf28}+2.09\%$
test_func_call_runtime[False-eager]	0.9098ms	0.5458ms	1.8322 KOps/s	1.8399 KOps/s	$\color{#d91a1a}-0.42\%$
test_func_call_runtime[False-compile]	0.8227ms	0.4225ms	2.3668 KOps/s	2.3196 KOps/s	$\color{#35bf28}+2.03\%$
test_func_call_runtime[False-compile-overhead]	0.9050ms	0.4243ms	2.3567 KOps/s	2.3294 KOps/s	$\color{#35bf28}+1.17\%$
test_func_call_runtime[True-eager]	1.4058ms	0.7626ms	1.3113 KOps/s	1.3280 KOps/s	$\color{#d91a1a}-1.25\%$
test_func_call_runtime[True-compile]	0.7734ms	0.4627ms	2.1612 KOps/s	2.1125 KOps/s	$\color{#35bf28}+2.31\%$
test_func_call_runtime[True-compile-overhead]	0.6923ms	0.4608ms	2.1702 KOps/s	2.1107 KOps/s	$\color{#35bf28}+2.82\%$
test_func_call_cm_runtime[False-eager]	0.8343ms	0.5409ms	1.8486 KOps/s	1.8493 KOps/s	$\color{#d91a1a}-0.04\%$
test_func_call_cm_runtime[False-compile]	0.8923ms	0.4171ms	2.3974 KOps/s	2.3182 KOps/s	$\color{#35bf28}+3.42\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7913ms	0.4188ms	2.3876 KOps/s	2.3157 KOps/s	$\color{#35bf28}+3.10\%$
test_func_call_cm_runtime[True-eager]	1.0546ms	0.8886ms	1.1254 KOps/s	1.1327 KOps/s	$\color{#d91a1a}-0.64\%$
test_func_call_cm_runtime[True-compile]	1.4139ms	0.5047ms	1.9813 KOps/s	1.9983 KOps/s	$\color{#d91a1a}-0.85\%$
test_func_call_cm_runtime[True-compile-overhead]	0.9966ms	0.4853ms	2.0608 KOps/s	2.0107 KOps/s	$\color{#35bf28}+2.49\%$
test_vmap_func_call_cm_runtime[eager]	2.5805ms	1.8579ms	538.2392 Ops/s	521.9543 Ops/s	$\color{#35bf28}+3.12\%$
test_vmap_func_call_cm_runtime[compile]	0.8836ms	0.5148ms	1.9427 KOps/s	1.7542 KOps/s	$\textbf{\color{#35bf28}+10.74\%}$
test_vmap_func_call_cm_runtime[compile-overhead]	0.8680ms	0.5175ms	1.9323 KOps/s	1.8975 KOps/s	$\color{#35bf28}+1.83\%$
test_distributed	0.3428ms	0.1257ms	7.9544 KOps/s	7.6433 KOps/s	$\color{#35bf28}+4.07\%$
test_tdmodule	80.9320μs	18.7006μs	53.4741 KOps/s	48.9255 KOps/s	$\textbf{\color{#35bf28}+9.30\%}$
test_tdmodule_dispatch	66.3150μs	37.0627μs	26.9813 KOps/s	24.8053 KOps/s	$\textbf{\color{#35bf28}+8.77\%}$
test_tdseq	45.2450μs	21.7565μs	45.9634 KOps/s	42.8222 KOps/s	$\textbf{\color{#35bf28}+7.34\%}$
test_tdseq_dispatch	70.1220μs	42.2478μs	23.6699 KOps/s	21.7059 KOps/s	$\textbf{\color{#35bf28}+9.05\%}$
test_instantiation_functorch	2.4428ms	1.5468ms	646.5121 Ops/s	638.1164 Ops/s	$\color{#35bf28}+1.32\%$
test_exec_functorch	0.6924ms	0.1935ms	5.1686 KOps/s	5.5144 KOps/s	$\textbf{\color{#d91a1a}-6.27\%}$
test_exec_functional_call	0.4521ms	0.1788ms	5.5938 KOps/s	5.7740 KOps/s	$\color{#d91a1a}-3.12\%$
test_exec_td_decorator	0.5067ms	0.2355ms	4.2456 KOps/s	4.3491 KOps/s	$\color{#d91a1a}-2.38\%$
test_vmap_mlp_speed_decorator[True-True]	0.8643ms	0.6235ms	1.6037 KOps/s	1.5516 KOps/s	$\color{#35bf28}+3.36\%$
test_vmap_mlp_speed_decorator[True-False]	0.9012ms	0.6246ms	1.6009 KOps/s	1.5290 KOps/s	$\color{#35bf28}+4.70\%$
test_vmap_mlp_speed_decorator[False-True]	0.7005ms	0.5113ms	1.9558 KOps/s	1.8840 KOps/s	$\color{#35bf28}+3.81\%$
test_vmap_mlp_speed_decorator[False-False]	0.8778ms	0.5129ms	1.9498 KOps/s	1.9004 KOps/s	$\color{#35bf28}+2.60\%$
test_to_module_speed[True]	1.7055ms	1.2822ms	779.9359 Ops/s	767.8285 Ops/s	$\color{#35bf28}+1.58\%$
test_to_module_speed[False]	1.7990ms	1.2503ms	799.7960 Ops/s	788.2808 Ops/s	$\color{#35bf28}+1.46\%$
test_tc_init	85.8920μs	46.1494μs	21.6688 KOps/s	21.3074 KOps/s	$\color{#35bf28}+1.70\%$
test_tc_init_nested	0.1596ms	92.3282μs	10.8309 KOps/s	10.5579 KOps/s	$\color{#35bf28}+2.59\%$
test_tc_first_layer_tensor	22.9030μs	1.5136μs	660.6686 KOps/s	648.0851 KOps/s	$\color{#35bf28}+1.94\%$
test_tc_first_layer_nontensor	21.8210μs	4.6741μs	213.9443 KOps/s	213.9693 KOps/s	$\color{#d91a1a}-0.01\%$
test_tc_second_layer_tensor	21.2790μs	2.8073μs	356.2084 KOps/s	356.1363 KOps/s	$\color{#35bf28}+0.02\%$
test_tc_second_layer_nontensor	27.7120μs	5.9634μs	167.6908 KOps/s	166.0582 KOps/s	$\color{#35bf28}+0.98\%$
test_unbind	0.2188s	12.3631ms	80.8860 Ops/s	82.4936 Ops/s	$\color{#d91a1a}-1.95\%$
test_full_like	7.6290ms	6.9082ms	144.7547 Ops/s	142.6469 Ops/s	$\color{#35bf28}+1.48\%$
test_zeros_like	3.1740ms	2.7324ms	365.9819 Ops/s	360.7920 Ops/s	$\color{#35bf28}+1.44\%$
test_ones_like	3.5748ms	3.1192ms	320.5949 Ops/s	323.2390 Ops/s	$\color{#d91a1a}-0.82\%$
test_clone	5.6219ms	5.0288ms	198.8557 Ops/s	201.1185 Ops/s	$\color{#d91a1a}-1.13\%$
test_squeeze	58.9110μs	12.0821μs	82.7670 KOps/s	81.4488 KOps/s	$\color{#35bf28}+1.62\%$
test_unsqueeze	0.2882ms	89.0995μs	11.2234 KOps/s	11.1497 KOps/s	$\color{#35bf28}+0.66\%$
test_split	0.3777ms	0.1887ms	5.3004 KOps/s	5.2787 KOps/s	$\color{#35bf28}+0.41\%$
test_permute	0.2892ms	0.2118ms	4.7215 KOps/s	4.5527 KOps/s	$\color{#35bf28}+3.71\%$
test_stack	27.7589ms	24.8751ms	40.2008 Ops/s	40.1956 Ops/s	$\color{#35bf28}+0.01\%$
test_cat	26.8506ms	24.6642ms	40.5446 Ops/s	40.6990 Ops/s	$\color{#d91a1a}-0.38\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}28$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	35.0310μs	10.1768μs	98.2627 KOps/s	82.2395 KOps/s	$\textbf{\color{#35bf28}+19.48\%}$
test_plain_set_stack_nested	96.4710μs	10.2586μs	97.4796 KOps/s	82.4650 KOps/s	$\textbf{\color{#35bf28}+18.21\%}$
test_plain_set_nested_inplace	39.4000μs	11.0025μs	90.8886 KOps/s	76.9744 KOps/s	$\textbf{\color{#35bf28}+18.08\%}$
test_plain_set_stack_nested_inplace	38.0210μs	11.0029μs	90.8855 KOps/s	76.2296 KOps/s	$\textbf{\color{#35bf28}+19.23\%}$
test_items	39.9510μs	2.9485μs	339.1551 KOps/s	340.9414 KOps/s	$\color{#d91a1a}-0.52\%$
test_items_nested	0.3940ms	0.3255ms	3.0721 KOps/s	3.0274 KOps/s	$\color{#35bf28}+1.48\%$
test_items_nested_locked	0.3822ms	0.3285ms	3.0444 KOps/s	3.0071 KOps/s	$\color{#35bf28}+1.24\%$
test_items_nested_leaf	89.3720μs	59.0285μs	16.9410 KOps/s	16.8641 KOps/s	$\color{#35bf28}+0.46\%$
test_items_stack_nested	0.3722ms	0.3277ms	3.0517 KOps/s	3.0569 KOps/s	$\color{#d91a1a}-0.17\%$
test_items_stack_nested_leaf	87.5920μs	60.1244μs	16.6322 KOps/s	16.9709 KOps/s	$\color{#d91a1a}-2.00\%$
test_items_stack_nested_locked	0.3858ms	0.3304ms	3.0271 KOps/s	3.0534 KOps/s	$\color{#d91a1a}-0.86\%$
test_keys	26.2910μs	3.5471μs	281.9175 KOps/s	287.9638 KOps/s	$\color{#d91a1a}-2.10\%$
test_keys_nested	96.8120μs	69.5183μs	14.3847 KOps/s	14.2009 KOps/s	$\color{#35bf28}+1.29\%$
test_keys_nested_locked	0.7125ms	75.1058μs	13.3146 KOps/s	13.1526 KOps/s	$\color{#35bf28}+1.23\%$
test_keys_nested_leaf	94.1420μs	60.9425μs	16.4089 KOps/s	16.2243 KOps/s	$\color{#35bf28}+1.14\%$
test_keys_stack_nested	0.1083ms	69.7572μs	14.3354 KOps/s	14.1347 KOps/s	$\color{#35bf28}+1.42\%$
test_keys_stack_nested_leaf	0.1444ms	60.7577μs	16.4588 KOps/s	16.2880 KOps/s	$\color{#35bf28}+1.05\%$
test_keys_stack_nested_locked	0.1290ms	75.4536μs	13.2532 KOps/s	13.1615 KOps/s	$\color{#35bf28}+0.70\%$
test_values	5.9168μs	0.8585μs	1.1648 MOps/s	1.1518 MOps/s	$\color{#35bf28}+1.12\%$
test_values_nested	59.5110μs	31.7554μs	31.4907 KOps/s	31.6818 KOps/s	$\color{#d91a1a}-0.60\%$
test_values_nested_locked	61.7210μs	33.0517μs	30.2556 KOps/s	30.1854 KOps/s	$\color{#35bf28}+0.23\%$
test_values_nested_leaf	61.8210μs	33.9953μs	29.4158 KOps/s	29.6605 KOps/s	$\color{#d91a1a}-0.83\%$
test_values_stack_nested	63.5420μs	31.9471μs	31.3017 KOps/s	31.4970 KOps/s	$\color{#d91a1a}-0.62\%$
test_values_stack_nested_leaf	73.5820μs	34.2408μs	29.2050 KOps/s	29.5871 KOps/s	$\color{#d91a1a}-1.29\%$
test_values_stack_nested_locked	58.4110μs	33.4687μs	29.8787 KOps/s	30.0363 KOps/s	$\color{#d91a1a}-0.52\%$
test_membership	2.3406μs	0.5254μs	1.9035 MOps/s	1.8941 MOps/s	$\color{#35bf28}+0.49\%$
test_membership_nested	41.0310μs	2.0192μs	495.2451 KOps/s	506.5130 KOps/s	$\color{#d91a1a}-2.22\%$
test_membership_nested_leaf	22.8705μs	2.0006μs	499.8462 KOps/s	501.6784 KOps/s	$\color{#d91a1a}-0.37\%$
test_membership_stacked_nested	28.7210μs	2.0591μs	485.6496 KOps/s	488.3188 KOps/s	$\color{#d91a1a}-0.55\%$
test_membership_stacked_nested_leaf	37.8210μs	2.0712μs	482.8124 KOps/s	489.3576 KOps/s	$\color{#d91a1a}-1.34\%$
test_membership_nested_last	35.9010μs	2.9153μs	343.0234 KOps/s	344.3796 KOps/s	$\color{#d91a1a}-0.39\%$
test_membership_nested_leaf_last	24.8110μs	2.9394μs	340.2035 KOps/s	347.0322 KOps/s	$\color{#d91a1a}-1.97\%$
test_membership_stacked_nested_last	31.0300μs	4.6451μs	215.2824 KOps/s	343.2127 KOps/s	$\textbf{\color{#d91a1a}-37.27\%}$
test_membership_stacked_nested_leaf_last	28.0210μs	4.6600μs	214.5925 KOps/s	347.6389 KOps/s	$\textbf{\color{#d91a1a}-38.27\%}$
test_nested_getleaf	38.1910μs	6.1011μs	163.9049 KOps/s	162.7249 KOps/s	$\color{#35bf28}+0.73\%$
test_nested_get	44.7610μs	5.7404μs	174.2033 KOps/s	171.7809 KOps/s	$\color{#35bf28}+1.41\%$
test_stacked_getleaf	28.6710μs	6.0731μs	164.6608 KOps/s	162.9629 KOps/s	$\color{#35bf28}+1.04\%$
test_stacked_get	37.2910μs	5.7659μs	173.4324 KOps/s	172.1317 KOps/s	$\color{#35bf28}+0.76\%$
test_nested_getitemleaf	26.7810μs	6.1545μs	162.4840 KOps/s	162.2781 KOps/s	$\color{#35bf28}+0.13\%$
test_nested_getitem	32.9210μs	5.8659μs	170.4760 KOps/s	171.0022 KOps/s	$\color{#d91a1a}-0.31\%$
test_stacked_getitemleaf	49.3410μs	6.1309μs	163.1069 KOps/s	161.5343 KOps/s	$\color{#35bf28}+0.97\%$
test_stacked_getitem	37.4510μs	5.8210μs	171.7929 KOps/s	169.9485 KOps/s	$\color{#35bf28}+1.09\%$
test_lock_nested	0.7771ms	0.3776ms	2.6486 KOps/s	2.5840 KOps/s	$\color{#35bf28}+2.50\%$
test_lock_stack_nested	0.4044ms	0.3411ms	2.9319 KOps/s	2.8555 KOps/s	$\color{#35bf28}+2.67\%$
test_unlock_nested	0.6257ms	0.3170ms	3.1550 KOps/s	3.1356 KOps/s	$\color{#35bf28}+0.62\%$
test_unlock_stack_nested	0.3177ms	0.2795ms	3.5780 KOps/s	3.4697 KOps/s	$\color{#35bf28}+3.12\%$
test_flatten_speed	0.1170ms	73.2629μs	13.6495 KOps/s	13.6405 KOps/s	$\color{#35bf28}+0.07\%$
test_unflatten_speed	0.3556ms	0.2974ms	3.3621 KOps/s	3.3733 KOps/s	$\color{#d91a1a}-0.33\%$
test_common_ops	91.5859ms	0.6449ms	1.5506 KOps/s	1.5598 KOps/s	$\color{#d91a1a}-0.59\%$
test_creation	0.1710ms	1.4766μs	677.2189 KOps/s	667.4650 KOps/s	$\color{#35bf28}+1.46\%$
test_creation_empty	29.2800μs	6.3671μs	157.0585 KOps/s	98.4644 KOps/s	$\textbf{\color{#35bf28}+59.51\%}$
test_creation_nested_1	45.3510μs	7.9159μs	126.3281 KOps/s	84.3819 KOps/s	$\textbf{\color{#35bf28}+49.71\%}$
test_creation_nested_2	36.5210μs	10.3950μs	96.2001 KOps/s	69.5087 KOps/s	$\textbf{\color{#35bf28}+38.40\%}$
test_clone	76.9610μs	10.7934μs	92.6489 KOps/s	94.4487 KOps/s	$\color{#d91a1a}-1.91\%$
test_getitem[int]	1.4839ms	11.1269μs	89.8726 KOps/s	90.7909 KOps/s	$\color{#d91a1a}-1.01\%$
test_getitem[slice_int]	0.1111ms	21.9621μs	45.5330 KOps/s	44.5709 KOps/s	$\color{#35bf28}+2.16\%$
test_getitem[range]	0.1330ms	38.3623μs	26.0672 KOps/s	25.8478 KOps/s	$\color{#35bf28}+0.85\%$
test_getitem[tuple]	0.1066ms	18.4794μs	54.1144 KOps/s	53.2023 KOps/s	$\color{#35bf28}+1.71\%$
test_getitem[list]	0.2385ms	33.8670μs	29.5272 KOps/s	28.4877 KOps/s	$\color{#35bf28}+3.65\%$
test_setitem_dim[int]	45.0210μs	19.5314μs	51.1996 KOps/s	49.9684 KOps/s	$\color{#35bf28}+2.46\%$
test_setitem_dim[slice_int]	56.4610μs	39.1556μs	25.5391 KOps/s	24.7288 KOps/s	$\color{#35bf28}+3.28\%$
test_setitem_dim[range]	80.3710μs	53.7859μs	18.5922 KOps/s	18.4789 KOps/s	$\color{#35bf28}+0.61\%$
test_setitem_dim[tuple]	56.0310μs	32.4976μs	30.7715 KOps/s	29.4109 KOps/s	$\color{#35bf28}+4.63\%$
test_setitem	77.0310μs	14.5284μs	68.8306 KOps/s	59.9399 KOps/s	$\textbf{\color{#35bf28}+14.83\%}$
test_set	75.9810μs	14.0851μs	70.9969 KOps/s	62.1010 KOps/s	$\textbf{\color{#35bf28}+14.32\%}$
test_set_shared	1.4695ms	0.1499ms	6.6693 KOps/s	6.6685 KOps/s	$\color{#35bf28}+0.01\%$
test_update	0.4118ms	15.9398μs	62.7361 KOps/s	49.8767 KOps/s	$\textbf{\color{#35bf28}+25.78\%}$
test_update_nested	77.2610μs	21.0244μs	47.5637 KOps/s	39.3033 KOps/s	$\textbf{\color{#35bf28}+21.02\%}$
test_update__nested	1.1231ms	25.5784μs	39.0955 KOps/s	39.8933 KOps/s	$\color{#d91a1a}-2.00\%$
test_set_nested	74.9420μs	15.3244μs	65.2556 KOps/s	56.4335 KOps/s	$\textbf{\color{#35bf28}+15.63\%}$
test_set_nested_new	77.3010μs	17.5185μs	57.0825 KOps/s	50.5599 KOps/s	$\textbf{\color{#35bf28}+12.90\%}$
test_select	89.6420μs	29.3698μs	34.0486 KOps/s	31.4129 KOps/s	$\textbf{\color{#35bf28}+8.39\%}$
test_select_nested	82.4420μs	43.2206μs	23.1371 KOps/s	23.1033 KOps/s	$\color{#35bf28}+0.15\%$
test_exclude_nested	0.3945ms	60.5685μs	16.5102 KOps/s	16.3368 KOps/s	$\color{#35bf28}+1.06\%$
test_empty[True]	0.3092ms	0.2586ms	3.8676 KOps/s	3.8139 KOps/s	$\color{#35bf28}+1.41\%$
test_empty[False]	5.2931μs	0.7779μs	1.2854 MOps/s	1.2801 MOps/s	$\color{#35bf28}+0.42\%$
test_to	85.8410μs	54.7260μs	18.2728 KOps/s	17.6390 KOps/s	$\color{#35bf28}+3.59\%$
test_to_nonblocking	0.1009ms	47.8275μs	20.9085 KOps/s	20.9324 KOps/s	$\color{#d91a1a}-0.11\%$
test_unbind_speed	1.7604ms	0.2409ms	4.1503 KOps/s	4.0957 KOps/s	$\color{#35bf28}+1.33\%$
test_unbind_speed_stack0	0.2957ms	0.2359ms	4.2397 KOps/s	4.1663 KOps/s	$\color{#35bf28}+1.76\%$
test_unbind_speed_stack1	91.3214ms	0.6528ms	1.5319 KOps/s	1.6392 KOps/s	$\textbf{\color{#d91a1a}-6.54\%}$
test_split	92.9293ms	1.7033ms	587.0883 Ops/s	589.8618 Ops/s	$\color{#d91a1a}-0.47\%$
test_chunk	94.0945ms	1.6954ms	589.8378 Ops/s	586.0991 Ops/s	$\color{#35bf28}+0.64\%$
test_consolidate[False-None]	95.8091ms	2.9185ms	342.6473 Ops/s	345.0007 Ops/s	$\color{#d91a1a}-0.68\%$
test_consolidate[default-None]	1.8840ms	1.7165ms	582.5863 Ops/s	601.7089 Ops/s	$\color{#d91a1a}-3.18\%$
test_consolidate[reduce-overhead-None]	1.8960ms	1.7502ms	571.3515 Ops/s	584.7536 Ops/s	$\color{#d91a1a}-2.29\%$
test_consolidate_njt[False-None]	7.2592ms	6.8223ms	146.5791 Ops/s	148.3299 Ops/s	$\color{#d91a1a}-1.18\%$
test_to[False-False-None]	1.8122ms	1.7259ms	579.3928 Ops/s	573.1748 Ops/s	$\color{#35bf28}+1.08\%$
test_to[True-False-None]	1.4986ms	1.3874ms	720.7515 Ops/s	735.1736 Ops/s	$\color{#d91a1a}-1.96\%$
test_to[within-False-None]	4.4479ms	4.2211ms	236.9027 Ops/s	241.4695 Ops/s	$\color{#d91a1a}-1.89\%$
test_to[True-default-None]	5.4815ms	5.3087ms	188.3684 Ops/s	185.4414 Ops/s	$\color{#35bf28}+1.58\%$
test_to_njt[False-False-None]	7.1902ms	7.0945ms	140.9540 Ops/s	137.2819 Ops/s	$\color{#35bf28}+2.67\%$
test_to_njt[True-False-None]	5.8558ms	5.7297ms	174.5285 Ops/s	166.6167 Ops/s	$\color{#35bf28}+4.75\%$
test_to_njt[within-False-None]	12.7574ms	12.5823ms	79.4766 Ops/s	55.3953 Ops/s	$\textbf{\color{#35bf28}+43.47\%}$
test_creation[device0]	0.5490ms	80.9387μs	12.3550 KOps/s	11.9120 KOps/s	$\color{#35bf28}+3.72\%$
test_creation_from_tensor	0.6139ms	84.5892μs	11.8218 KOps/s	11.3050 KOps/s	$\color{#35bf28}+4.57\%$
test_add_one[memmap_tensor0]	0.3788ms	7.0275μs	142.2990 KOps/s	147.3025 KOps/s	$\color{#d91a1a}-3.40\%$
test_contiguous[memmap_tensor0]	1.9650μs	0.4327μs	2.3108 MOps/s	2.2607 MOps/s	$\color{#35bf28}+2.22\%$
test_stack[memmap_tensor0]	41.5700μs	4.7362μs	211.1419 KOps/s	215.9946 KOps/s	$\color{#d91a1a}-2.25\%$
test_memmaptd_index	1.9945ms	0.2605ms	3.8392 KOps/s	3.8493 KOps/s	$\color{#d91a1a}-0.26\%$
test_memmaptd_index_astensor	0.5760ms	0.3212ms	3.1132 KOps/s	3.1546 KOps/s	$\color{#d91a1a}-1.31\%$
test_memmaptd_index_op	0.9951ms	0.5846ms	1.7106 KOps/s	1.5513 KOps/s	$\textbf{\color{#35bf28}+10.26\%}$
test_serialize_model	0.1322s	0.1310s	7.6365 Ops/s	7.6184 Ops/s	$\color{#35bf28}+0.24\%$
test_serialize_model_pickle	1.3642s	1.2192s	0.8202 Ops/s	0.8435 Ops/s	$\color{#d91a1a}-2.77\%$
test_serialize_weights	0.1305s	0.1299s	7.6992 Ops/s	7.6227 Ops/s	$\color{#35bf28}+1.00\%$
test_serialize_weights_returnearly	0.3808s	56.5784ms	17.6746 Ops/s	23.4963 Ops/s	$\textbf{\color{#d91a1a}-24.78\%}$
test_serialize_weights_pickle	1.3484s	1.2115s	0.8254 Ops/s	0.8223 Ops/s	$\color{#35bf28}+0.38\%$
test_reshape_pytree	55.3810μs	23.8292μs	41.9654 KOps/s	43.7552 KOps/s	$\color{#d91a1a}-4.09\%$
test_reshape_td	58.5310μs	27.6495μs	36.1670 KOps/s	36.5508 KOps/s	$\color{#d91a1a}-1.05\%$
test_view_pytree	64.6410μs	23.7894μs	42.0355 KOps/s	43.6796 KOps/s	$\color{#d91a1a}-3.76\%$
test_view_td	58.3510μs	30.6823μs	32.5921 KOps/s	33.5884 KOps/s	$\color{#d91a1a}-2.97\%$
test_unbind_pytree	61.0220μs	29.3364μs	34.0874 KOps/s	34.9760 KOps/s	$\color{#d91a1a}-2.54\%$
test_unbind_td	0.6365ms	36.5960μs	27.3254 KOps/s	26.6769 KOps/s	$\color{#35bf28}+2.43\%$
test_split_pytree	69.7220μs	31.9858μs	31.2639 KOps/s	31.8591 KOps/s	$\color{#d91a1a}-1.87\%$
test_split_td	0.7549ms	41.3214μs	24.2005 KOps/s	23.4899 KOps/s	$\color{#35bf28}+3.03\%$
test_add_pytree	62.3110μs	35.1413μs	28.4566 KOps/s	28.5887 KOps/s	$\color{#d91a1a}-0.46\%$
test_add_td	80.2220μs	45.6729μs	21.8948 KOps/s	19.4589 KOps/s	$\textbf{\color{#35bf28}+12.52\%}$
test_compile_add_one_nested[tensordict-compile]	0.1708ms	0.1238ms	8.0756 KOps/s	8.0449 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_add_one_nested[tensordict-eager]	0.2263ms	0.1280ms	7.8134 KOps/s	7.5251 KOps/s	$\color{#35bf28}+3.83\%$
test_compile_add_one_nested[pytree-compile]	0.1432ms	0.1001ms	9.9910 KOps/s	9.8305 KOps/s	$\color{#35bf28}+1.63\%$
test_compile_add_one_nested[pytree-eager]	1.5173ms	0.1570ms	6.3679 KOps/s	6.4103 KOps/s	$\color{#d91a1a}-0.66\%$
test_compile_copy_nested[tensordict-compile]	58.6910μs	22.7975μs	43.8644 KOps/s	42.6051 KOps/s	$\color{#35bf28}+2.96\%$
test_compile_copy_nested[tensordict-eager]	67.4320μs	27.2609μs	36.6826 KOps/s	35.7335 KOps/s	$\color{#35bf28}+2.66\%$
test_compile_copy_nested[pytree-compile]	0.2249ms	65.2031μs	15.3367 KOps/s	15.2446 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_copy_nested[pytree-eager]	98.1420μs	50.0983μs	19.9607 KOps/s	19.8803 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_add_one_flat[tensordict-compile]	0.1972ms	0.1445ms	6.9207 KOps/s	6.9496 KOps/s	$\color{#d91a1a}-0.42\%$
test_compile_add_one_flat[tensordict-eager]	0.2920ms	0.2118ms	4.7209 KOps/s	4.7715 KOps/s	$\color{#d91a1a}-1.06\%$
test_compile_add_one_flat[tensorclass-compile]	0.1628ms	0.1003ms	9.9749 KOps/s	10.1325 KOps/s	$\color{#d91a1a}-1.56\%$
test_compile_add_one_flat[tensorclass-eager]	0.1107ms	53.6690μs	18.6327 KOps/s	18.8449 KOps/s	$\color{#d91a1a}-1.13\%$
test_compile_add_one_flat[pytree-compile]	0.1952ms	0.1512ms	6.6149 KOps/s	6.9073 KOps/s	$\color{#d91a1a}-4.23\%$
test_compile_add_one_flat[pytree-eager]	0.5981ms	0.5141ms	1.9450 KOps/s	1.9699 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_add_self_flat[tensordict-eager]	0.3511ms	0.2488ms	4.0193 KOps/s	3.9901 KOps/s	$\color{#35bf28}+0.73\%$
test_compile_add_self_flat[tensordict-compile]	0.2219ms	0.1473ms	6.7893 KOps/s	6.6924 KOps/s	$\color{#35bf28}+1.45\%$
test_compile_add_self_flat[tensorclass-eager]	0.1462ms	62.9927μs	15.8748 KOps/s	15.5044 KOps/s	$\color{#35bf28}+2.39\%$
test_compile_add_self_flat[tensorclass-compile]	0.1636ms	0.1044ms	9.5789 KOps/s	9.7961 KOps/s	$\color{#d91a1a}-2.22\%$
test_compile_add_self_flat[pytree-eager]	0.5073ms	0.4181ms	2.3919 KOps/s	2.3978 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_self_flat[pytree-compile]	0.1850ms	0.1451ms	6.8895 KOps/s	7.0175 KOps/s	$\color{#d91a1a}-1.82\%$
test_compile_copy_flat[tensordict-compile]	66.0110μs	19.5632μs	51.1163 KOps/s	42.5040 KOps/s	$\textbf{\color{#35bf28}+20.26\%}$
test_compile_copy_flat[tensordict-eager]	59.8810μs	27.4902μs	36.3766 KOps/s	36.6348 KOps/s	$\color{#d91a1a}-0.70\%$
test_compile_copy_flat[pytree-compile]	0.1699ms	70.1934μs	14.2463 KOps/s	14.2942 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_copy_flat[pytree-eager]	83.2720μs	52.2484μs	19.1393 KOps/s	19.2605 KOps/s	$\color{#d91a1a}-0.63\%$
test_compile_assign_and_add[tensordict-compile]	1.6704ms	0.4557ms	2.1943 KOps/s	2.1833 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_assign_and_add[tensordict-eager]	2.7729ms	2.6753ms	373.7913 Ops/s	372.3847 Ops/s	$\color{#35bf28}+0.38\%$
test_compile_assign_and_add[pytree-compile]	1.6442ms	0.4420ms	2.2624 KOps/s	2.2310 KOps/s	$\color{#35bf28}+1.41\%$
test_compile_assign_and_add[pytree-eager]	2.8987ms	2.7779ms	359.9804 Ops/s	358.5840 Ops/s	$\color{#35bf28}+0.39\%$
test_compile_indexing[tensor-tensordict-compile]	0.1985ms	0.1222ms	8.1821 KOps/s	8.5858 KOps/s	$\color{#d91a1a}-4.70\%$
test_compile_indexing[tensor-tensordict-eager]	0.5592ms	84.7839μs	11.7947 KOps/s	12.1795 KOps/s	$\color{#d91a1a}-3.16\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1708ms	0.1143ms	8.7467 KOps/s	9.1977 KOps/s	$\color{#d91a1a}-4.90\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1540ms	73.4673μs	13.6115 KOps/s	13.4155 KOps/s	$\color{#35bf28}+1.46\%$
test_compile_indexing[tensor-pytree-compile]	0.1771ms	0.1093ms	9.1481 KOps/s	8.6147 KOps/s	$\textbf{\color{#35bf28}+6.19\%}$
test_compile_indexing[tensor-pytree-eager]	0.1412ms	72.6122μs	13.7718 KOps/s	13.5268 KOps/s	$\color{#35bf28}+1.81\%$
test_compile_indexing[slice-tensordict-compile]	0.1567ms	0.1038ms	9.6307 KOps/s	9.6758 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_indexing[slice-tensordict-eager]	0.1444ms	18.3245μs	54.5719 KOps/s	53.2916 KOps/s	$\color{#35bf28}+2.40\%$
test_compile_indexing[slice-tensorclass-compile]	0.1500ms	0.1003ms	9.9657 KOps/s	10.0149 KOps/s	$\color{#d91a1a}-0.49\%$
test_compile_indexing[slice-tensorclass-eager]	56.8410μs	17.7205μs	56.4318 KOps/s	57.9026 KOps/s	$\color{#d91a1a}-2.54\%$
test_compile_indexing[slice-pytree-compile]	0.2226ms	0.1050ms	9.5213 KOps/s	9.9814 KOps/s	$\color{#d91a1a}-4.61\%$
test_compile_indexing[slice-pytree-eager]	46.9710μs	17.6228μs	56.7446 KOps/s	57.4485 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_indexing[int-tensordict-compile]	0.1657ms	0.1085ms	9.2194 KOps/s	9.6135 KOps/s	$\color{#d91a1a}-4.10\%$
test_compile_indexing[int-tensordict-eager]	0.5431ms	18.2247μs	54.8707 KOps/s	54.2229 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_indexing[int-tensorclass-compile]	0.1580ms	0.1052ms	9.5037 KOps/s	10.0208 KOps/s	$\textbf{\color{#d91a1a}-5.16\%}$
test_compile_indexing[int-tensorclass-eager]	47.4410μs	17.8516μs	56.0175 KOps/s	58.1033 KOps/s	$\color{#d91a1a}-3.59\%$
test_compile_indexing[int-pytree-compile]	0.2066ms	0.1008ms	9.9187 KOps/s	10.0110 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_indexing[int-pytree-eager]	50.8310μs	17.7363μs	56.3816 KOps/s	58.3072 KOps/s	$\color{#d91a1a}-3.30\%$
test_mod_add[eager]	71.5310μs	30.5685μs	32.7134 KOps/s	29.0886 KOps/s	$\textbf{\color{#35bf28}+12.46\%}$
test_mod_add[compile]	0.1403ms	82.1176μs	12.1777 KOps/s	12.5724 KOps/s	$\color{#d91a1a}-3.14\%$
test_mod_add[compile-overhead]	0.3129ms	0.1655ms	6.0429 KOps/s	5.8337 KOps/s	$\color{#35bf28}+3.59\%$
test_mod_wrap[eager]	0.3579ms	0.2493ms	4.0110 KOps/s	3.8367 KOps/s	$\color{#35bf28}+4.54\%$
test_mod_wrap[compile]	1.6258ms	0.3040ms	3.2893 KOps/s	3.4525 KOps/s	$\color{#d91a1a}-4.73\%$
test_mod_wrap[compile-overhead]	7.8212ms	4.0297ms	248.1548 Ops/s	244.9692 Ops/s	$\color{#35bf28}+1.30\%$
test_mod_wrap_and_backward[eager]	1.7882ms	1.4625ms	683.7544 Ops/s	673.8177 Ops/s	$\color{#35bf28}+1.47\%$
test_mod_wrap_and_backward[compile]	1.5790ms	1.3883ms	720.3134 Ops/s	718.4868 Ops/s	$\color{#35bf28}+0.25\%$
test_mod_wrap_and_backward[compile-overhead]	1.5047ms	1.0378ms	963.5707 Ops/s	964.2210 Ops/s	$\color{#d91a1a}-0.07\%$
test_seq_add[eager]	0.2622ms	98.5454μs	10.1476 KOps/s	9.8234 KOps/s	$\color{#35bf28}+3.30\%$
test_seq_add[compile]	0.1669ms	88.6702μs	11.2777 KOps/s	11.1171 KOps/s	$\color{#35bf28}+1.45\%$
test_seq_add[compile-overhead]	0.1995ms	0.1291ms	7.7445 KOps/s	7.6782 KOps/s	$\color{#35bf28}+0.86\%$
test_seq_wrap[eager]	0.4799ms	0.3823ms	2.6160 KOps/s	2.3620 KOps/s	$\textbf{\color{#35bf28}+10.76\%}$
test_seq_wrap[compile]	0.3641ms	0.3055ms	3.2734 KOps/s	3.2493 KOps/s	$\color{#35bf28}+0.74\%$
test_seq_wrap[compile-overhead]	0.2794ms	0.2225ms	4.4947 KOps/s	4.4709 KOps/s	$\color{#35bf28}+0.53\%$
test_func_call_runtime[False-eager]	0.8943ms	0.7747ms	1.2909 KOps/s	1.2780 KOps/s	$\color{#35bf28}+1.01\%$
test_func_call_runtime[False-compile]	1.0675ms	0.7644ms	1.3082 KOps/s	1.3091 KOps/s	$\color{#d91a1a}-0.07\%$
test_func_call_runtime[False-compile-overhead]	0.4494ms	0.3683ms	2.7152 KOps/s	2.7596 KOps/s	$\color{#d91a1a}-1.61\%$
test_func_call_runtime[True-eager]	1.0802ms	0.9650ms	1.0363 KOps/s	1.0571 KOps/s	$\color{#d91a1a}-1.97\%$
test_func_call_runtime[True-compile]	0.8497ms	0.7851ms	1.2738 KOps/s	1.2792 KOps/s	$\color{#d91a1a}-0.42\%$
test_func_call_runtime[True-compile-overhead]	0.4591ms	0.3811ms	2.6240 KOps/s	2.6188 KOps/s	$\color{#35bf28}+0.19\%$
test_func_call_cm_runtime[False-eager]	0.9216ms	0.7950ms	1.2578 KOps/s	1.2869 KOps/s	$\color{#d91a1a}-2.26\%$
test_func_call_cm_runtime[False-compile]	0.9232ms	0.7655ms	1.3064 KOps/s	1.3086 KOps/s	$\color{#d91a1a}-0.17\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4249ms	0.3640ms	2.7471 KOps/s	2.7432 KOps/s	$\color{#35bf28}+0.14\%$
test_func_call_cm_runtime[True-eager]	1.1345ms	1.0305ms	970.3570 Ops/s	955.8643 Ops/s	$\color{#35bf28}+1.52\%$
test_func_call_cm_runtime[True-compile]	1.1008ms	0.8287ms	1.2068 KOps/s	1.2375 KOps/s	$\color{#d91a1a}-2.48\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5113ms	0.4091ms	2.4447 KOps/s	2.4428 KOps/s	$\color{#35bf28}+0.08\%$
test_vmap_func_call_cm_runtime[eager]	2.5245ms	2.0736ms	482.2443 Ops/s	472.9845 Ops/s	$\color{#35bf28}+1.96\%$
test_vmap_func_call_cm_runtime[compile]	0.8923ms	0.8221ms	1.2164 KOps/s	1.1985 KOps/s	$\color{#35bf28}+1.50\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4818ms	0.4122ms	2.4260 KOps/s	2.4188 KOps/s	$\color{#35bf28}+0.30\%$
test_distributed	1.8840ms	0.1726ms	5.7932 KOps/s	8.5428 KOps/s	$\textbf{\color{#d91a1a}-32.19\%}$
test_tdmodule	19.8410μs	12.6664μs	78.9490 KOps/s	60.0057 KOps/s	$\textbf{\color{#35bf28}+31.57\%}$
test_tdmodule_dispatch	0.5671ms	25.4580μs	39.2804 KOps/s	33.1475 KOps/s	$\textbf{\color{#35bf28}+18.50\%}$
test_tdseq	34.4000μs	14.4061μs	69.4151 KOps/s	59.8132 KOps/s	$\textbf{\color{#35bf28}+16.05\%}$
test_tdseq_dispatch	47.3900μs	28.0356μs	35.6689 KOps/s	29.7759 KOps/s	$\textbf{\color{#35bf28}+19.79\%}$
test_instantiation_functorch	1.7156ms	1.5878ms	629.7985 Ops/s	622.2150 Ops/s	$\color{#35bf28}+1.22\%$
test_exec_functorch	0.1992ms	0.1531ms	6.5322 KOps/s	6.5260 KOps/s	$\color{#35bf28}+0.10\%$
test_exec_functional_call	0.2884ms	0.1465ms	6.8282 KOps/s	6.7970 KOps/s	$\color{#35bf28}+0.46\%$
test_exec_td_decorator	0.3703ms	0.1912ms	5.2297 KOps/s	5.1960 KOps/s	$\color{#35bf28}+0.65\%$
test_vmap_mlp_speed_decorator[True-True]	0.8194ms	0.6710ms	1.4902 KOps/s	1.4712 KOps/s	$\color{#35bf28}+1.29\%$
test_vmap_mlp_speed_decorator[True-False]	0.8247ms	0.6694ms	1.4939 KOps/s	1.4698 KOps/s	$\color{#35bf28}+1.64\%$
test_vmap_mlp_speed_decorator[False-True]	0.6946ms	0.5939ms	1.6839 KOps/s	1.6813 KOps/s	$\color{#35bf28}+0.16\%$
test_vmap_mlp_speed_decorator[False-False]	0.7053ms	0.5897ms	1.6958 KOps/s	1.6833 KOps/s	$\color{#35bf28}+0.75\%$
test_vmap_transformer_speed_decorator[True-True]	20.0912ms	19.4778ms	51.3404 Ops/s	51.4546 Ops/s	$\color{#d91a1a}-0.22\%$
test_vmap_transformer_speed_decorator[True-False]	20.0861ms	19.3992ms	51.5486 Ops/s	51.3207 Ops/s	$\color{#35bf28}+0.44\%$
test_vmap_transformer_speed_decorator[False-True]	19.8472ms	19.2748ms	51.8812 Ops/s	51.9377 Ops/s	$\color{#d91a1a}-0.11\%$
test_vmap_transformer_speed_decorator[False-False]	19.3483ms	19.2320ms	51.9968 Ops/s	51.8194 Ops/s	$\color{#35bf28}+0.34\%$
test_to_module_speed[True]	1.0529ms	0.9492ms	1.0535 KOps/s	1.0479 KOps/s	$\color{#35bf28}+0.54\%$
test_to_module_speed[False]	1.3328ms	0.9349ms	1.0697 KOps/s	1.0705 KOps/s	$\color{#d91a1a}-0.08\%$
test_tc_init	66.6910μs	33.1237μs	30.1899 KOps/s	25.5730 KOps/s	$\textbf{\color{#35bf28}+18.05\%}$
test_tc_init_nested	0.1558ms	68.0515μs	14.6948 KOps/s	12.6578 KOps/s	$\textbf{\color{#35bf28}+16.09\%}$
test_tc_first_layer_tensor	6.9416μs	0.6988μs	1.4311 MOps/s	1.3771 MOps/s	$\color{#35bf28}+3.92\%$
test_tc_first_layer_nontensor	26.5810μs	2.3378μs	427.7589 KOps/s	431.1265 KOps/s	$\color{#d91a1a}-0.78\%$
test_tc_second_layer_tensor	16.7002μs	1.4291μs	699.7492 KOps/s	691.9296 KOps/s	$\color{#35bf28}+1.13\%$
test_tc_second_layer_nontensor	27.4110μs	3.0725μs	325.4707 KOps/s	325.8801 KOps/s	$\color{#d91a1a}-0.13\%$
test_unbind	0.2372s	10.1452ms	98.5684 Ops/s	149.2625 Ops/s	$\textbf{\color{#d91a1a}-33.96\%}$
test_full_like	9.6740ms	9.0997ms	109.8940 Ops/s	105.6444 Ops/s	$\color{#35bf28}+4.02\%$
test_zeros_like	9.2343ms	7.2806ms	137.3519 Ops/s	114.3920 Ops/s	$\textbf{\color{#35bf28}+20.07\%}$
test_ones_like	5.2493ms	4.3184ms	231.5694 Ops/s	233.6542 Ops/s	$\color{#d91a1a}-0.89\%$
test_clone	6.6839ms	6.3527ms	157.4129 Ops/s	158.2744 Ops/s	$\color{#d91a1a}-0.54\%$
test_squeeze	58.1110μs	9.4484μs	105.8382 KOps/s	106.1178 KOps/s	$\color{#d91a1a}-0.26\%$
test_unsqueeze	0.1591ms	74.5539μs	13.4131 KOps/s	13.6494 KOps/s	$\color{#d91a1a}-1.73\%$
test_split	0.3982ms	0.1700ms	5.8810 KOps/s	5.8153 KOps/s	$\color{#35bf28}+1.13\%$
test_permute	0.2534ms	0.1920ms	5.2092 KOps/s	5.2861 KOps/s	$\color{#d91a1a}-1.45\%$
test_stack	50.9252ms	50.5664ms	19.7760 Ops/s	19.7380 Ops/s	$\color{#35bf28}+0.19\%$
test_cat	50.5639ms	49.9669ms	20.0133 Ops/s	19.9372 Ops/s	$\color{#35bf28}+0.38\%$