[Feature] TensorClass shadow attributes

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] NonTensorData(*sequence_of_any) #1160
• -> [Feature] TensorClass shadow attributes #1159

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}16$. Worsened: $\large\color{#d91a1a}23$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	43.9720μs	21.9299μs	45.5999 KOps/s	49.3690 KOps/s	$\textbf{\color{#d91a1a}-7.63\%}$
test_plain_set_stack_nested	58.0480μs	22.1378μs	45.1717 KOps/s	48.2464 KOps/s	$\textbf{\color{#d91a1a}-6.37\%}$
test_plain_set_nested_inplace	57.9880μs	23.8550μs	41.9199 KOps/s	44.9955 KOps/s	$\textbf{\color{#d91a1a}-6.84\%}$
test_plain_set_stack_nested_inplace	67.3450μs	23.7042μs	42.1866 KOps/s	44.0407 KOps/s	$\color{#d91a1a}-4.21\%$
test_items	43.3100μs	4.1114μs	243.2260 KOps/s	240.0697 KOps/s	$\color{#35bf28}+1.31\%$
test_items_nested	0.6120ms	0.3993ms	2.5046 KOps/s	2.4799 KOps/s	$\color{#35bf28}+1.00\%$
test_items_nested_locked	0.5153ms	0.4036ms	2.4777 KOps/s	2.4645 KOps/s	$\color{#35bf28}+0.53\%$
test_items_nested_leaf	0.1345ms	76.5578μs	13.0620 KOps/s	12.8151 KOps/s	$\color{#35bf28}+1.93\%$
test_items_stack_nested	0.5042ms	0.4052ms	2.4678 KOps/s	2.4486 KOps/s	$\color{#35bf28}+0.78\%$
test_items_stack_nested_leaf	0.1315ms	78.3791μs	12.7585 KOps/s	12.7534 KOps/s	$\color{#35bf28}+0.04\%$
test_items_stack_nested_locked	0.5464ms	0.4073ms	2.4552 KOps/s	2.4564 KOps/s	$\color{#d91a1a}-0.05\%$
test_keys	40.3950μs	3.5239μs	283.7725 KOps/s	240.2102 KOps/s	$\textbf{\color{#35bf28}+18.14\%}$
test_keys_nested	0.2473ms	0.1668ms	5.9964 KOps/s	6.0014 KOps/s	$\color{#d91a1a}-0.08\%$
test_keys_nested_locked	0.7880ms	0.1752ms	5.7089 KOps/s	5.7909 KOps/s	$\color{#d91a1a}-1.41\%$
test_keys_nested_leaf	0.2290ms	0.1458ms	6.8572 KOps/s	6.8952 KOps/s	$\color{#d91a1a}-0.55\%$
test_keys_stack_nested	0.3114ms	0.1682ms	5.9438 KOps/s	6.0247 KOps/s	$\color{#d91a1a}-1.34\%$
test_keys_stack_nested_leaf	0.3827ms	0.1460ms	6.8501 KOps/s	6.8751 KOps/s	$\color{#d91a1a}-0.36\%$
test_keys_stack_nested_locked	0.2706ms	0.1724ms	5.8007 KOps/s	5.8626 KOps/s	$\color{#d91a1a}-1.05\%$
test_values	6.9230μs	1.0553μs	947.5992 KOps/s	912.2377 KOps/s	$\color{#35bf28}+3.88\%$
test_values_nested	0.1130ms	63.1759μs	15.8288 KOps/s	15.9002 KOps/s	$\color{#d91a1a}-0.45\%$
test_values_nested_locked	0.1090ms	63.1148μs	15.8441 KOps/s	15.8255 KOps/s	$\color{#35bf28}+0.12\%$
test_values_nested_leaf	0.3279ms	72.2142μs	13.8477 KOps/s	13.7716 KOps/s	$\color{#35bf28}+0.55\%$
test_values_stack_nested	0.1207ms	63.4064μs	15.7713 KOps/s	15.7883 KOps/s	$\color{#d91a1a}-0.11\%$
test_values_stack_nested_leaf	0.1391ms	71.5287μs	13.9804 KOps/s	13.8353 KOps/s	$\color{#35bf28}+1.05\%$
test_values_stack_nested_locked	0.1204ms	63.7043μs	15.6975 KOps/s	15.8658 KOps/s	$\color{#d91a1a}-1.06\%$
test_membership	4.1227μs	0.7078μs	1.4129 MOps/s	1.1263 MOps/s	$\textbf{\color{#35bf28}+25.44\%}$
test_membership_nested	24.0550μs	2.9739μs	336.2592 KOps/s	337.7055 KOps/s	$\color{#d91a1a}-0.43\%$
test_membership_nested_leaf	34.7250μs	3.0203μs	331.0932 KOps/s	338.5850 KOps/s	$\color{#d91a1a}-2.21\%$
test_membership_stacked_nested	27.3700μs	3.0161μs	331.5503 KOps/s	340.9667 KOps/s	$\color{#d91a1a}-2.76\%$
test_membership_stacked_nested_leaf	31.2880μs	3.0404μs	328.9054 KOps/s	343.8431 KOps/s	$\color{#d91a1a}-4.34\%$
test_membership_nested_last	22.0700μs	4.4408μs	225.1830 KOps/s	229.7333 KOps/s	$\color{#d91a1a}-1.98\%$
test_membership_nested_leaf_last	38.1410μs	4.4823μs	223.0982 KOps/s	226.3552 KOps/s	$\color{#d91a1a}-1.44\%$
test_membership_stacked_nested_last	27.6420μs	4.4159μs	226.4524 KOps/s	228.4539 KOps/s	$\color{#d91a1a}-0.88\%$
test_membership_stacked_nested_leaf_last	35.2160μs	4.5246μs	221.0117 KOps/s	232.5016 KOps/s	$\color{#d91a1a}-4.94\%$
test_nested_getleaf	94.0040μs	11.7781μs	84.9034 KOps/s	89.9518 KOps/s	$\textbf{\color{#d91a1a}-5.61\%}$
test_nested_get	51.2160μs	10.7066μs	93.4006 KOps/s	90.3789 KOps/s	$\color{#35bf28}+3.34\%$
test_stacked_getleaf	36.5580μs	11.3259μs	88.2934 KOps/s	87.8320 KOps/s	$\color{#35bf28}+0.53\%$
test_stacked_get	32.5510μs	10.6371μs	94.0107 KOps/s	92.5831 KOps/s	$\color{#35bf28}+1.54\%$
test_nested_getitemleaf	49.9750μs	11.6498μs	85.8386 KOps/s	84.7693 KOps/s	$\color{#35bf28}+1.26\%$
test_nested_getitem	39.1620μs	10.8789μs	91.9213 KOps/s	89.2878 KOps/s	$\color{#35bf28}+2.95\%$
test_stacked_getitemleaf	51.8170μs	11.4473μs	87.3567 KOps/s	83.4812 KOps/s	$\color{#35bf28}+4.64\%$
test_stacked_getitem	68.1470μs	10.7382μs	93.1255 KOps/s	87.6479 KOps/s	$\textbf{\color{#35bf28}+6.25\%}$
test_lock_nested	6.6502ms	0.4665ms	2.1437 KOps/s	2.1399 KOps/s	$\color{#35bf28}+0.18\%$
test_lock_stack_nested	0.6004ms	0.4306ms	2.3222 KOps/s	2.2816 KOps/s	$\color{#35bf28}+1.78\%$
test_unlock_nested	0.8123ms	0.3813ms	2.6227 KOps/s	2.5975 KOps/s	$\color{#35bf28}+0.97\%$
test_unlock_stack_nested	0.4375ms	0.3474ms	2.8786 KOps/s	2.8090 KOps/s	$\color{#35bf28}+2.48\%$
test_flatten_speed	0.1964ms	99.3767μs	10.0627 KOps/s	9.8590 KOps/s	$\color{#35bf28}+2.07\%$
test_unflatten_speed	0.6483ms	0.5396ms	1.8532 KOps/s	1.8377 KOps/s	$\color{#35bf28}+0.85\%$
test_common_ops	4.0722ms	0.8492ms	1.1776 KOps/s	1.2874 KOps/s	$\textbf{\color{#d91a1a}-8.53\%}$
test_creation	40.2060μs	2.6301μs	380.2112 KOps/s	407.8265 KOps/s	$\textbf{\color{#d91a1a}-6.77\%}$
test_creation_empty	42.3290μs	13.5534μs	73.7824 KOps/s	98.4425 KOps/s	$\textbf{\color{#d91a1a}-25.05\%}$
test_creation_nested_1	85.0670μs	16.5307μs	60.4934 KOps/s	76.8549 KOps/s	$\textbf{\color{#d91a1a}-21.29\%}$
test_creation_nested_2	54.5710μs	21.4832μs	46.5480 KOps/s	53.6592 KOps/s	$\textbf{\color{#d91a1a}-13.25\%}$
test_clone	0.1891ms	13.8265μs	72.3247 KOps/s	71.7401 KOps/s	$\color{#35bf28}+0.81\%$
test_getitem[int]	1.1871ms	12.9060μs	77.4830 KOps/s	75.6432 KOps/s	$\color{#35bf28}+2.43\%$
test_getitem[slice_int]	0.1426ms	25.9748μs	38.4989 KOps/s	40.0300 KOps/s	$\color{#d91a1a}-3.82\%$
test_getitem[range]	0.2335ms	50.4841μs	19.8082 KOps/s	20.3836 KOps/s	$\color{#d91a1a}-2.82\%$
test_getitem[tuple]	0.1425ms	20.9004μs	47.8461 KOps/s	48.0733 KOps/s	$\color{#d91a1a}-0.47\%$
test_getitem[list]	0.3067ms	45.7883μs	21.8397 KOps/s	22.1724 KOps/s	$\color{#d91a1a}-1.50\%$
test_setitem_dim[int]	58.6690μs	26.3811μs	37.9059 KOps/s	37.7025 KOps/s	$\color{#35bf28}+0.54\%$
test_setitem_dim[slice_int]	95.5480μs	55.3376μs	18.0709 KOps/s	18.6833 KOps/s	$\color{#d91a1a}-3.28\%$
test_setitem_dim[range]	0.1241ms	76.3209μs	13.1026 KOps/s	13.4831 KOps/s	$\color{#d91a1a}-2.82\%$
test_setitem_dim[tuple]	85.2690μs	42.8344μs	23.3457 KOps/s	23.5635 KOps/s	$\color{#d91a1a}-0.92\%$
test_setitem	0.1633ms	22.3387μs	44.7655 KOps/s	49.4336 KOps/s	$\textbf{\color{#d91a1a}-9.44\%}$
test_set	0.2205ms	21.7115μs	46.0585 KOps/s	50.8096 KOps/s	$\textbf{\color{#d91a1a}-9.35\%}$
test_set_shared	7.6501ms	0.1744ms	5.7327 KOps/s	5.5061 KOps/s	$\color{#35bf28}+4.12\%$
test_update	0.2270ms	25.2811μs	39.5553 KOps/s	46.0173 KOps/s	$\textbf{\color{#d91a1a}-14.04\%}$
test_update_nested	0.2412ms	36.1135μs	27.6905 KOps/s	31.0031 KOps/s	$\textbf{\color{#d91a1a}-10.68\%}$
test_update__nested	0.4864ms	34.7464μs	28.7799 KOps/s	28.6032 KOps/s	$\color{#35bf28}+0.62\%$
test_set_nested	0.1514ms	23.4187μs	42.7008 KOps/s	44.6418 KOps/s	$\color{#d91a1a}-4.35\%$
test_set_nested_new	0.1907ms	28.2909μs	35.3471 KOps/s	36.8721 KOps/s	$\color{#d91a1a}-4.14\%$
test_select	0.2310ms	44.8696μs	22.2868 KOps/s	22.8394 KOps/s	$\color{#d91a1a}-2.42\%$
test_select_nested	0.1236ms	63.8244μs	15.6680 KOps/s	15.7344 KOps/s	$\color{#d91a1a}-0.42\%$
test_exclude_nested	0.1735ms	83.3171μs	12.0023 KOps/s	11.9720 KOps/s	$\color{#35bf28}+0.25\%$
test_empty[True]	0.6228ms	0.4141ms	2.4147 KOps/s	2.3805 KOps/s	$\color{#35bf28}+1.44\%$
test_empty[False]	12.5112μs	1.3861μs	721.4646 KOps/s	715.8035 KOps/s	$\color{#35bf28}+0.79\%$
test_unbind_speed	0.3848ms	0.2707ms	3.6939 KOps/s	3.5663 KOps/s	$\color{#35bf28}+3.58\%$
test_unbind_speed_stack0	0.5266ms	0.2702ms	3.7006 KOps/s	3.6045 KOps/s	$\color{#35bf28}+2.67\%$
test_unbind_speed_stack1	0.1071s	0.8090ms	1.2361 KOps/s	1.2898 KOps/s	$\color{#d91a1a}-4.16\%$
test_split	0.1011s	1.7658ms	566.3063 Ops/s	550.7131 Ops/s	$\color{#35bf28}+2.83\%$
test_chunk	2.0270ms	1.6080ms	621.8901 Ops/s	549.4227 Ops/s	$\textbf{\color{#35bf28}+13.19\%}$
test_consolidate_njt[False-None]	8.7008ms	8.3793ms	119.3418 Ops/s	111.7626 Ops/s	$\textbf{\color{#35bf28}+6.78\%}$
test_creation[device0]	0.2552ms	91.7216μs	10.9026 KOps/s	10.2050 KOps/s	$\textbf{\color{#35bf28}+6.84\%}$
test_creation_from_tensor	3.6100ms	94.4417μs	10.5885 KOps/s	10.1254 KOps/s	$\color{#35bf28}+4.57\%$
test_add_one[memmap_tensor0]	0.1215ms	4.9112μs	203.6152 KOps/s	192.3018 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_contiguous[memmap_tensor0]	35.4260μs	0.5190μs	1.9268 MOps/s	1.9338 MOps/s	$\color{#d91a1a}-0.36\%$
test_stack[memmap_tensor0]	38.5720μs	3.3319μs	300.1294 KOps/s	276.9458 KOps/s	$\textbf{\color{#35bf28}+8.37\%}$
test_memmaptd_index	1.0956ms	0.2377ms	4.2074 KOps/s	3.9433 KOps/s	$\textbf{\color{#35bf28}+6.70\%}$
test_memmaptd_index_astensor	0.5761ms	0.3245ms	3.0814 KOps/s	2.9365 KOps/s	$\color{#35bf28}+4.93\%$
test_memmaptd_index_op	0.9878ms	0.6159ms	1.6238 KOps/s	1.6662 KOps/s	$\color{#d91a1a}-2.55\%$
test_serialize_model	0.1235s	0.1168s	8.5590 Ops/s	8.1577 Ops/s	$\color{#35bf28}+4.92\%$
test_serialize_model_pickle	0.4314s	0.3859s	2.5911 Ops/s	2.5518 Ops/s	$\color{#35bf28}+1.54\%$
test_serialize_weights	0.1209s	0.1146s	8.7274 Ops/s	8.2490 Ops/s	$\textbf{\color{#35bf28}+5.80\%}$
test_serialize_weights_returnearly	0.2293s	0.1606s	6.2254 Ops/s	6.3764 Ops/s	$\color{#d91a1a}-2.37\%$
test_serialize_weights_pickle	0.4449s	0.3917s	2.5532 Ops/s	1.1145 Ops/s	$\textbf{\color{#35bf28}+129.10\%}$
test_serialize_weights_filesystem	0.1468s	0.1405s	7.1182 Ops/s	7.0031 Ops/s	$\color{#35bf28}+1.64\%$
test_serialize_model_filesystem	0.1640s	0.1533s	6.5217 Ops/s	6.7519 Ops/s	$\color{#d91a1a}-3.41\%$
test_reshape_pytree	58.1790μs	26.6803μs	37.4809 KOps/s	36.7756 KOps/s	$\color{#35bf28}+1.92\%$
test_reshape_td	71.3630μs	33.5278μs	29.8260 KOps/s	28.8227 KOps/s	$\color{#35bf28}+3.48\%$
test_view_pytree	66.9650μs	26.5796μs	37.6228 KOps/s	36.4151 KOps/s	$\color{#35bf28}+3.32\%$
test_view_td	0.1030ms	39.3898μs	25.3873 KOps/s	25.1612 KOps/s	$\color{#35bf28}+0.90\%$
test_unbind_pytree	91.2980μs	29.8780μs	33.4694 KOps/s	33.3299 KOps/s	$\color{#35bf28}+0.42\%$
test_unbind_td	0.3799ms	40.4768μs	24.7055 KOps/s	24.3947 KOps/s	$\color{#35bf28}+1.27\%$
test_split_pytree	69.0490μs	29.6701μs	33.7040 KOps/s	33.4575 KOps/s	$\color{#35bf28}+0.74\%$
test_split_td	0.1019s	56.1230μs	17.8180 KOps/s	21.9604 KOps/s	$\textbf{\color{#d91a1a}-18.86\%}$
test_add_pytree	92.9530μs	34.8590μs	28.6870 KOps/s	27.0813 KOps/s	$\textbf{\color{#35bf28}+5.93\%}$
test_add_td	0.1375ms	60.6700μs	16.4826 KOps/s	16.8142 KOps/s	$\color{#d91a1a}-1.97\%$
test_compile_add_one_nested[tensordict-compile]	0.1698ms	62.3376μs	16.0417 KOps/s	16.1622 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_add_one_nested[tensordict-eager]	0.3715ms	0.1691ms	5.9133 KOps/s	5.7120 KOps/s	$\color{#35bf28}+3.52\%$
test_compile_add_one_nested[pytree-compile]	0.1827ms	46.6799μs	21.4225 KOps/s	21.6537 KOps/s	$\color{#d91a1a}-1.07\%$
test_compile_add_one_nested[pytree-eager]	0.2780ms	0.1180ms	8.4728 KOps/s	8.0541 KOps/s	$\textbf{\color{#35bf28}+5.20\%}$
test_compile_copy_nested[tensordict-compile]	82.3440μs	26.8408μs	37.2567 KOps/s	38.2357 KOps/s	$\color{#d91a1a}-2.56\%$
test_compile_copy_nested[tensordict-eager]	0.1324ms	59.8763μs	16.7011 KOps/s	16.8068 KOps/s	$\color{#d91a1a}-0.63\%$
test_compile_copy_nested[pytree-compile]	0.2077ms	76.1988μs	13.1236 KOps/s	12.6294 KOps/s	$\color{#35bf28}+3.91\%$
test_compile_copy_nested[pytree-eager]	0.1191ms	66.0287μs	15.1449 KOps/s	14.6363 KOps/s	$\color{#35bf28}+3.47\%$
test_compile_add_one_flat[tensordict-compile]	0.1871ms	0.1057ms	9.4590 KOps/s	9.1924 KOps/s	$\color{#35bf28}+2.90\%$
test_compile_add_one_flat[tensordict-eager]	0.3896ms	0.2162ms	4.6261 KOps/s	4.5150 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_add_one_flat[tensorclass-compile]	97.2710μs	46.4210μs	21.5420 KOps/s	22.1544 KOps/s	$\color{#d91a1a}-2.76\%$
test_compile_add_one_flat[tensorclass-eager]	0.5149ms	66.0571μs	15.1384 KOps/s	14.8742 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_add_one_flat[pytree-compile]	0.2218ms	0.1053ms	9.5007 KOps/s	9.6054 KOps/s	$\color{#d91a1a}-1.09\%$
test_compile_add_one_flat[pytree-eager]	0.4118ms	0.1992ms	5.0207 KOps/s	4.8559 KOps/s	$\color{#35bf28}+3.40\%$
test_compile_add_self_flat[tensordict-eager]	0.4060ms	0.2312ms	4.3256 KOps/s	4.2544 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_add_self_flat[tensordict-compile]	0.1914ms	0.1063ms	9.4056 KOps/s	9.4905 KOps/s	$\color{#d91a1a}-0.89\%$
test_compile_add_self_flat[tensorclass-eager]	0.1121ms	59.3149μs	16.8592 KOps/s	16.8625 KOps/s	$\color{#d91a1a}-0.02\%$
test_compile_add_self_flat[tensorclass-compile]	1.0540ms	50.9606μs	19.6230 KOps/s	20.6191 KOps/s	$\color{#d91a1a}-4.83\%$
test_compile_add_self_flat[pytree-eager]	1.4488ms	0.1596ms	6.2654 KOps/s	6.1495 KOps/s	$\color{#35bf28}+1.89\%$
test_compile_add_self_flat[pytree-compile]	0.1833ms	0.1068ms	9.3621 KOps/s	9.7447 KOps/s	$\color{#d91a1a}-3.93\%$
test_compile_copy_flat[tensordict-compile]	69.6090μs	22.3839μs	44.6749 KOps/s	48.3124 KOps/s	$\textbf{\color{#d91a1a}-7.53\%}$
test_compile_copy_flat[tensordict-eager]	0.1358ms	66.7821μs	14.9741 KOps/s	15.0574 KOps/s	$\color{#d91a1a}-0.55\%$
test_compile_copy_flat[pytree-compile]	0.1948ms	82.0527μs	12.1873 KOps/s	12.1207 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_copy_flat[pytree-eager]	0.1562ms	67.7315μs	14.7642 KOps/s	14.3415 KOps/s	$\color{#35bf28}+2.95\%$
test_compile_assign_and_add[tensordict-compile]	0.4480ms	0.2101ms	4.7599 KOps/s	4.8757 KOps/s	$\color{#d91a1a}-2.37\%$
test_compile_assign_and_add[tensordict-eager]	2.5609ms	1.3223ms	756.2832 Ops/s	729.4857 Ops/s	$\color{#35bf28}+3.67\%$
test_compile_assign_and_add[pytree-compile]	0.3239ms	0.2054ms	4.8686 KOps/s	4.9199 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_assign_and_add[pytree-eager]	1.3404ms	0.7739ms	1.2921 KOps/s	1.2654 KOps/s	$\color{#35bf28}+2.11\%$
test_compile_assign_and_add_stack[compile]	0.7593ms	0.4614ms	2.1672 KOps/s	2.1899 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_assign_and_add_stack[eager]	3.7185ms	2.7952ms	357.7555 Ops/s	370.1376 Ops/s	$\color{#d91a1a}-3.35\%$
test_compile_indexing[tensor-tensordict-compile]	83.5960μs	36.1179μs	27.6871 KOps/s	26.6355 KOps/s	$\color{#35bf28}+3.95\%$
test_compile_indexing[tensor-tensordict-eager]	0.5236ms	34.4011μs	29.0689 KOps/s	28.2914 KOps/s	$\color{#35bf28}+2.75\%$
test_compile_indexing[tensor-tensorclass-compile]	72.2850μs	30.4669μs	32.8225 KOps/s	32.7469 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_indexing[tensor-tensorclass-eager]	63.8990μs	22.6399μs	44.1698 KOps/s	41.9382 KOps/s	$\textbf{\color{#35bf28}+5.32\%}$
test_compile_indexing[tensor-pytree-compile]	88.0540μs	31.0127μs	32.2448 KOps/s	32.5060 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_indexing[tensor-pytree-eager]	60.9030μs	22.6771μs	44.0973 KOps/s	42.2727 KOps/s	$\color{#35bf28}+4.32\%$
test_compile_indexing[slice-tensordict-compile]	0.1272ms	53.2183μs	18.7905 KOps/s	19.1053 KOps/s	$\color{#d91a1a}-1.65\%$
test_compile_indexing[slice-tensordict-eager]	0.5645ms	20.4898μs	48.8047 KOps/s	48.3233 KOps/s	$\color{#35bf28}+1.00\%$
test_compile_indexing[slice-tensorclass-compile]	94.2460μs	45.9016μs	21.7858 KOps/s	21.9174 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_indexing[slice-tensorclass-eager]	0.1276ms	19.1653μs	52.1776 KOps/s	52.6599 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_indexing[slice-pytree-compile]	0.1343ms	46.1484μs	21.6692 KOps/s	21.3444 KOps/s	$\color{#35bf28}+1.52\%$
test_compile_indexing[slice-pytree-eager]	53.4690μs	18.7600μs	53.3048 KOps/s	52.3044 KOps/s	$\color{#35bf28}+1.91\%$
test_compile_indexing[int-tensordict-compile]	0.1195ms	53.5973μs	18.6577 KOps/s	18.7785 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_indexing[int-tensordict-eager]	1.0120ms	20.8335μs	47.9996 KOps/s	49.7298 KOps/s	$\color{#d91a1a}-3.48\%$
test_compile_indexing[int-tensorclass-compile]	0.2141ms	46.6166μs	21.4516 KOps/s	21.6425 KOps/s	$\color{#d91a1a}-0.88\%$
test_compile_indexing[int-tensorclass-eager]	88.0260μs	18.5927μs	53.7845 KOps/s	52.7883 KOps/s	$\color{#35bf28}+1.89\%$
test_compile_indexing[int-pytree-compile]	91.1400μs	46.3582μs	21.5711 KOps/s	21.3856 KOps/s	$\color{#35bf28}+0.87\%$
test_compile_indexing[int-pytree-eager]	60.7930μs	18.3923μs	54.3707 KOps/s	52.4576 KOps/s	$\color{#35bf28}+3.65\%$
test_mod_add[eager]	77.6040μs	35.6828μs	28.0247 KOps/s	28.6516 KOps/s	$\color{#d91a1a}-2.19\%$
test_mod_add[compile]	0.1034ms	48.6722μs	20.5456 KOps/s	20.6248 KOps/s	$\color{#d91a1a}-0.38\%$
test_mod_add[compile-overhead]	0.1339ms	49.2532μs	20.3033 KOps/s	20.8520 KOps/s	$\color{#d91a1a}-2.63\%$
test_mod_wrap[eager]	0.4957ms	0.2357ms	4.2434 KOps/s	4.3174 KOps/s	$\color{#d91a1a}-1.71\%$
test_mod_wrap[compile]	0.2936ms	0.2095ms	4.7732 KOps/s	4.7578 KOps/s	$\color{#35bf28}+0.32\%$
test_mod_wrap[compile-overhead]	0.3454ms	0.2091ms	4.7834 KOps/s	4.8137 KOps/s	$\color{#d91a1a}-0.63\%$
test_mod_wrap_and_backward[eager]	13.6358ms	11.8548ms	84.3540 Ops/s	81.4667 Ops/s	$\color{#35bf28}+3.54\%$
test_mod_wrap_and_backward[compile]	16.2655ms	12.2480ms	81.6457 Ops/s	74.4031 Ops/s	$\textbf{\color{#35bf28}+9.73\%}$
test_mod_wrap_and_backward[compile-overhead]	19.0671ms	13.1710ms	75.9242 Ops/s	73.9699 Ops/s	$\color{#35bf28}+2.64\%$
test_seq_add[eager]	0.2028ms	0.1206ms	8.2929 KOps/s	8.5734 KOps/s	$\color{#d91a1a}-3.27\%$
test_seq_add[compile]	0.1193ms	63.4485μs	15.7608 KOps/s	16.0333 KOps/s	$\color{#d91a1a}-1.70\%$
test_seq_add[compile-overhead]	0.1126ms	60.9203μs	16.4149 KOps/s	16.0621 KOps/s	$\color{#35bf28}+2.20\%$
test_seq_wrap[eager]	1.0709ms	0.4630ms	2.1599 KOps/s	2.1749 KOps/s	$\color{#d91a1a}-0.69\%$
test_seq_wrap[compile]	0.4414ms	0.2346ms	4.2623 KOps/s	4.2778 KOps/s	$\color{#d91a1a}-0.36\%$
test_seq_wrap[compile-overhead]	0.4402ms	0.2320ms	4.3108 KOps/s	4.2507 KOps/s	$\color{#35bf28}+1.41\%$
test_func_call_runtime[False-eager]	0.8517ms	0.5596ms	1.7870 KOps/s	1.7974 KOps/s	$\color{#d91a1a}-0.58\%$
test_func_call_runtime[False-compile]	0.5716ms	0.4313ms	2.3185 KOps/s	2.2846 KOps/s	$\color{#35bf28}+1.48\%$
test_func_call_runtime[False-compile-overhead]	0.5655ms	0.4299ms	2.3261 KOps/s	2.2801 KOps/s	$\color{#35bf28}+2.02\%$
test_func_call_runtime[True-eager]	1.2811ms	0.7681ms	1.3020 KOps/s	1.2603 KOps/s	$\color{#35bf28}+3.30\%$
test_func_call_runtime[True-compile]	0.7996ms	0.4688ms	2.1333 KOps/s	2.1088 KOps/s	$\color{#35bf28}+1.16\%$
test_func_call_runtime[True-compile-overhead]	0.8713ms	0.4732ms	2.1131 KOps/s	2.1095 KOps/s	$\color{#35bf28}+0.17\%$
test_func_call_cm_runtime[False-eager]	0.9603ms	0.5550ms	1.8018 KOps/s	1.7993 KOps/s	$\color{#35bf28}+0.14\%$
test_func_call_cm_runtime[False-compile]	0.6376ms	0.4324ms	2.3128 KOps/s	2.3202 KOps/s	$\color{#d91a1a}-0.32\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5241ms	0.4301ms	2.3249 KOps/s	2.3055 KOps/s	$\color{#35bf28}+0.84\%$
test_func_call_cm_runtime[True-eager]	1.2373ms	0.9149ms	1.0930 KOps/s	1.0664 KOps/s	$\color{#35bf28}+2.49\%$
test_func_call_cm_runtime[True-compile]	0.6053ms	0.4940ms	2.0244 KOps/s	2.0038 KOps/s	$\color{#35bf28}+1.03\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5901ms	0.4932ms	2.0275 KOps/s	1.9888 KOps/s	$\color{#35bf28}+1.95\%$
test_vmap_func_call_cm_runtime[eager]	2.6593ms	1.9500ms	512.8264 Ops/s	497.2357 Ops/s	$\color{#35bf28}+3.14\%$
test_vmap_func_call_cm_runtime[compile]	0.9490ms	0.5215ms	1.9175 KOps/s	1.9251 KOps/s	$\color{#d91a1a}-0.40\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.8227ms	0.5235ms	1.9103 KOps/s	1.8855 KOps/s	$\color{#35bf28}+1.32\%$
test_distributed	0.2784ms	0.1267ms	7.8900 KOps/s	7.6709 KOps/s	$\color{#35bf28}+2.86\%$
test_tdmodule	51.8370μs	28.2164μs	35.4404 KOps/s	38.6273 KOps/s	$\textbf{\color{#d91a1a}-8.25\%}$
test_tdmodule_dispatch	79.0070μs	52.2052μs	19.1552 KOps/s	20.9480 KOps/s	$\textbf{\color{#d91a1a}-8.56\%}$
test_tdseq	53.0790μs	31.7706μs	31.4757 KOps/s	34.2315 KOps/s	$\textbf{\color{#d91a1a}-8.05\%}$
test_tdseq_dispatch	89.1860μs	58.7498μs	17.0213 KOps/s	18.3228 KOps/s	$\textbf{\color{#d91a1a}-7.10\%}$
test_instantiation_functorch	1.7237ms	1.5463ms	646.6941 Ops/s	636.3368 Ops/s	$\color{#35bf28}+1.63\%$
test_exec_functorch	0.3324ms	0.1837ms	5.4443 KOps/s	5.3507 KOps/s	$\color{#35bf28}+1.75\%$
test_exec_functional_call	0.4529ms	0.1791ms	5.5843 KOps/s	5.5686 KOps/s	$\color{#35bf28}+0.28\%$
test_exec_td_decorator	0.4733ms	0.2388ms	4.1873 KOps/s	4.1415 KOps/s	$\color{#35bf28}+1.11\%$
test_vmap_mlp_speed_decorator[True-True]	1.1725ms	0.6818ms	1.4667 KOps/s	1.4167 KOps/s	$\color{#35bf28}+3.53\%$
test_vmap_mlp_speed_decorator[True-False]	1.0122ms	0.6794ms	1.4718 KOps/s	1.4630 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_mlp_speed_decorator[False-True]	0.8638ms	0.5560ms	1.7984 KOps/s	1.8321 KOps/s	$\color{#d91a1a}-1.84\%$
test_vmap_mlp_speed_decorator[False-False]	0.9280ms	0.5424ms	1.8437 KOps/s	1.8274 KOps/s	$\color{#35bf28}+0.89\%$
test_to_module_speed[True]	2.0458ms	1.3541ms	738.5250 Ops/s	735.9055 Ops/s	$\color{#35bf28}+0.36\%$
test_to_module_speed[False]	1.4390ms	1.3294ms	752.2328 Ops/s	742.0862 Ops/s	$\color{#35bf28}+1.37\%$
test_tc_init	91.4710μs	51.1410μs	19.5538 KOps/s	22.0972 KOps/s	$\textbf{\color{#d91a1a}-11.51\%}$
test_tc_init_nested	0.1711ms	0.1034ms	9.6709 KOps/s	10.9201 KOps/s	$\textbf{\color{#d91a1a}-11.44\%}$
test_tc_first_layer_tensor	16.9920μs	1.5591μs	641.4066 KOps/s	649.4535 KOps/s	$\color{#d91a1a}-1.24\%$
test_tc_first_layer_nontensor	26.6900μs	4.7088μs	212.3686 KOps/s	212.0873 KOps/s	$\color{#35bf28}+0.13\%$
test_tc_second_layer_tensor	48.0400μs	2.8077μs	356.1585 KOps/s	360.6884 KOps/s	$\color{#d91a1a}-1.26\%$
test_tc_second_layer_nontensor	52.2200μs	5.9582μs	167.8350 KOps/s	165.8096 KOps/s	$\color{#35bf28}+1.22\%$
test_unbind	0.2276s	15.5323ms	64.3820 Ops/s	75.0398 Ops/s	$\textbf{\color{#d91a1a}-14.20\%}$
test_full_like	8.8898ms	7.8011ms	128.1868 Ops/s	117.1174 Ops/s	$\textbf{\color{#35bf28}+9.45\%}$
test_zeros_like	3.6547ms	3.1308ms	319.4087 Ops/s	327.5350 Ops/s	$\color{#d91a1a}-2.48\%$
test_ones_like	3.8900ms	3.4247ms	291.9962 Ops/s	298.0450 Ops/s	$\color{#d91a1a}-2.03\%$
test_clone	6.4289ms	5.6522ms	176.9231 Ops/s	175.3990 Ops/s	$\color{#35bf28}+0.87\%$
test_squeeze	62.9080μs	12.8086μs	78.0726 KOps/s	81.3826 KOps/s	$\color{#d91a1a}-4.07\%$
test_unsqueeze	0.3212ms	95.7965μs	10.4388 KOps/s	10.4541 KOps/s	$\color{#d91a1a}-0.15\%$
test_split	0.3726ms	0.2049ms	4.8793 KOps/s	4.9086 KOps/s	$\color{#d91a1a}-0.60\%$
test_permute	0.4318ms	0.2184ms	4.5790 KOps/s	4.8013 KOps/s	$\color{#d91a1a}-4.63\%$
test_stack	34.4658ms	27.4914ms	36.3750 Ops/s	38.6498 Ops/s	$\textbf{\color{#d91a1a}-5.89\%}$
test_cat	35.5788ms	28.0480ms	35.6532 Ops/s	37.2192 Ops/s	$\color{#d91a1a}-4.21\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}33$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	40.5120μs	11.2027μs	89.2641 KOps/s	77.7410 KOps/s	$\textbf{\color{#35bf28}+14.82\%}$
test_plain_set_stack_nested	39.4320μs	11.2597μs	88.8125 KOps/s	75.9059 KOps/s	$\textbf{\color{#35bf28}+17.00\%}$
test_plain_set_nested_inplace	47.6330μs	12.3599μs	80.9065 KOps/s	70.1271 KOps/s	$\textbf{\color{#35bf28}+15.37\%}$
test_plain_set_stack_nested_inplace	1.0239ms	12.2560μs	81.5926 KOps/s	70.9381 KOps/s	$\textbf{\color{#35bf28}+15.02\%}$
test_items	33.7910μs	2.8819μs	346.9972 KOps/s	341.3361 KOps/s	$\color{#35bf28}+1.66\%$
test_items_nested	0.4068ms	0.3542ms	2.8236 KOps/s	2.8065 KOps/s	$\color{#35bf28}+0.61\%$
test_items_nested_locked	0.4364ms	0.3547ms	2.8190 KOps/s	2.7991 KOps/s	$\color{#35bf28}+0.71\%$
test_items_nested_leaf	0.1058ms	58.0938μs	17.2135 KOps/s	17.0314 KOps/s	$\color{#35bf28}+1.07\%$
test_items_stack_nested	0.3992ms	0.3587ms	2.7882 KOps/s	2.7981 KOps/s	$\color{#d91a1a}-0.36\%$
test_items_stack_nested_leaf	85.1140μs	58.2314μs	17.1729 KOps/s	16.6323 KOps/s	$\color{#35bf28}+3.25\%$
test_items_stack_nested_locked	0.4123ms	0.3586ms	2.7887 KOps/s	2.7845 KOps/s	$\color{#35bf28}+0.15\%$
test_keys	32.2310μs	3.4789μs	287.4461 KOps/s	284.2802 KOps/s	$\color{#35bf28}+1.11\%$
test_keys_nested	0.1307ms	80.7455μs	12.3846 KOps/s	12.2332 KOps/s	$\color{#35bf28}+1.24\%$
test_keys_nested_locked	0.8279ms	87.0605μs	11.4863 KOps/s	11.4548 KOps/s	$\color{#35bf28}+0.27\%$
test_keys_nested_leaf	0.1082ms	71.7156μs	13.9440 KOps/s	13.8131 KOps/s	$\color{#35bf28}+0.95\%$
test_keys_stack_nested	0.1445ms	80.8378μs	12.3705 KOps/s	12.0156 KOps/s	$\color{#35bf28}+2.95\%$
test_keys_stack_nested_leaf	0.1040ms	71.7904μs	13.9294 KOps/s	13.4244 KOps/s	$\color{#35bf28}+3.76\%$
test_keys_stack_nested_locked	0.1445ms	87.4984μs	11.4288 KOps/s	11.2162 KOps/s	$\color{#35bf28}+1.90\%$
test_values	6.6887μs	0.8504μs	1.1759 MOps/s	1.1598 MOps/s	$\color{#35bf28}+1.39\%$
test_values_nested	92.6050μs	33.8097μs	29.5773 KOps/s	29.1250 KOps/s	$\color{#35bf28}+1.55\%$
test_values_nested_locked	70.3240μs	35.7437μs	27.9769 KOps/s	27.4592 KOps/s	$\color{#35bf28}+1.89\%$
test_values_nested_leaf	65.4230μs	38.4026μs	26.0399 KOps/s	25.6016 KOps/s	$\color{#35bf28}+1.71\%$
test_values_stack_nested	70.6340μs	33.9603μs	29.4461 KOps/s	28.7431 KOps/s	$\color{#35bf28}+2.45\%$
test_values_stack_nested_leaf	71.2240μs	38.9403μs	25.6803 KOps/s	25.3495 KOps/s	$\color{#35bf28}+1.30\%$
test_values_stack_nested_locked	0.1013ms	36.0294μs	27.7551 KOps/s	27.1054 KOps/s	$\color{#35bf28}+2.40\%$
test_membership	1.9351μs	0.5317μs	1.8808 MOps/s	1.8708 MOps/s	$\color{#35bf28}+0.53\%$
test_membership_nested	16.8055μs	2.0366μs	491.0152 KOps/s	490.4527 KOps/s	$\color{#35bf28}+0.11\%$
test_membership_nested_leaf	21.4410μs	2.0497μs	487.8792 KOps/s	481.4325 KOps/s	$\color{#35bf28}+1.34\%$
test_membership_stacked_nested	39.3220μs	2.1404μs	467.2060 KOps/s	477.3158 KOps/s	$\color{#d91a1a}-2.12\%$
test_membership_stacked_nested_leaf	30.0920μs	2.0983μs	476.5761 KOps/s	473.8606 KOps/s	$\color{#35bf28}+0.57\%$
test_membership_nested_last	24.1420μs	3.1566μs	316.7921 KOps/s	312.6731 KOps/s	$\color{#35bf28}+1.32\%$
test_membership_nested_leaf_last	54.2330μs	3.1402μs	318.4462 KOps/s	312.8868 KOps/s	$\color{#35bf28}+1.78\%$
test_membership_stacked_nested_last	27.0010μs	3.1526μs	317.1950 KOps/s	251.9898 KOps/s	$\textbf{\color{#35bf28}+25.88\%}$
test_membership_stacked_nested_leaf_last	62.5430μs	3.1274μs	319.7568 KOps/s	254.6853 KOps/s	$\textbf{\color{#35bf28}+25.55\%}$
test_nested_getleaf	30.7120μs	6.0928μs	164.1289 KOps/s	161.4943 KOps/s	$\color{#35bf28}+1.63\%$
test_nested_get	50.5330μs	5.7967μs	172.5118 KOps/s	169.3085 KOps/s	$\color{#35bf28}+1.89\%$
test_stacked_getleaf	31.4110μs	6.1114μs	163.6276 KOps/s	163.1401 KOps/s	$\color{#35bf28}+0.30\%$
test_stacked_get	46.0830μs	5.8126μs	172.0387 KOps/s	171.3833 KOps/s	$\color{#35bf28}+0.38\%$
test_nested_getitemleaf	70.9940μs	6.2647μs	159.6254 KOps/s	157.8944 KOps/s	$\color{#35bf28}+1.10\%$
test_nested_getitem	35.8520μs	6.0073μs	166.4654 KOps/s	165.9028 KOps/s	$\color{#35bf28}+0.34\%$
test_stacked_getitemleaf	32.9310μs	6.3299μs	157.9809 KOps/s	159.9671 KOps/s	$\color{#d91a1a}-1.24\%$
test_stacked_getitem	42.5320μs	5.9018μs	169.4398 KOps/s	168.2967 KOps/s	$\color{#35bf28}+0.68\%$
test_lock_nested	0.7537ms	0.3722ms	2.6865 KOps/s	2.6599 KOps/s	$\color{#35bf28}+1.00\%$
test_lock_stack_nested	0.4139ms	0.3450ms	2.8981 KOps/s	2.8685 KOps/s	$\color{#35bf28}+1.03\%$
test_unlock_nested	0.6110ms	0.3110ms	3.2153 KOps/s	3.1862 KOps/s	$\color{#35bf28}+0.92\%$
test_unlock_stack_nested	0.3432ms	0.2810ms	3.5592 KOps/s	3.5387 KOps/s	$\color{#35bf28}+0.58\%$
test_flatten_speed	0.1537ms	73.8275μs	13.5451 KOps/s	13.4422 KOps/s	$\color{#35bf28}+0.77\%$
test_unflatten_speed	0.3847ms	0.3225ms	3.1005 KOps/s	3.1179 KOps/s	$\color{#d91a1a}-0.56\%$
test_common_ops	1.5257ms	0.5796ms	1.7253 KOps/s	1.5781 KOps/s	$\textbf{\color{#35bf28}+9.33\%}$
test_creation	0.1079ms	1.7456μs	572.8654 KOps/s	561.1653 KOps/s	$\color{#35bf28}+2.08\%$
test_creation_empty	34.7220μs	6.5725μs	152.1495 KOps/s	102.9990 KOps/s	$\textbf{\color{#35bf28}+47.72\%}$
test_creation_nested_1	41.3820μs	8.1914μs	122.0789 KOps/s	88.6700 KOps/s	$\textbf{\color{#35bf28}+37.68\%}$
test_creation_nested_2	60.7430μs	11.0720μs	90.3180 KOps/s	71.3953 KOps/s	$\textbf{\color{#35bf28}+26.50\%}$
test_clone	93.5850μs	10.5048μs	95.1945 KOps/s	93.7954 KOps/s	$\color{#35bf28}+1.49\%$
test_getitem[int]	1.7992ms	10.5183μs	95.0727 KOps/s	92.6676 KOps/s	$\color{#35bf28}+2.60\%$
test_getitem[slice_int]	0.1074ms	20.5758μs	48.6009 KOps/s	47.7402 KOps/s	$\color{#35bf28}+1.80\%$
test_getitem[range]	0.1260ms	37.4303μs	26.7163 KOps/s	26.9541 KOps/s	$\color{#d91a1a}-0.88\%$
test_getitem[tuple]	0.1053ms	18.1989μs	54.9485 KOps/s	54.0758 KOps/s	$\color{#35bf28}+1.61\%$
test_getitem[list]	0.2498ms	33.1360μs	30.1787 KOps/s	30.1104 KOps/s	$\color{#35bf28}+0.23\%$
test_setitem_dim[int]	28.3020μs	17.7766μs	56.2536 KOps/s	53.6008 KOps/s	$\color{#35bf28}+4.95\%$
test_setitem_dim[slice_int]	66.0340μs	37.5323μs	26.6437 KOps/s	26.1791 KOps/s	$\color{#35bf28}+1.77\%$
test_setitem_dim[range]	84.1140μs	52.0460μs	19.2138 KOps/s	19.3618 KOps/s	$\color{#d91a1a}-0.76\%$
test_setitem_dim[tuple]	52.5930μs	32.6042μs	30.6709 KOps/s	31.5994 KOps/s	$\color{#d91a1a}-2.94\%$
test_setitem	87.8340μs	14.0790μs	71.0277 KOps/s	62.8219 KOps/s	$\textbf{\color{#35bf28}+13.06\%}$
test_set	90.9140μs	13.5216μs	73.9556 KOps/s	64.6680 KOps/s	$\textbf{\color{#35bf28}+14.36\%}$
test_set_shared	1.7270ms	0.1539ms	6.4962 KOps/s	6.5426 KOps/s	$\color{#d91a1a}-0.71\%$
test_update	0.4705ms	15.3644μs	65.0854 KOps/s	51.8033 KOps/s	$\textbf{\color{#35bf28}+25.64\%}$
test_update_nested	90.1440μs	21.0751μs	47.4494 KOps/s	40.0545 KOps/s	$\textbf{\color{#35bf28}+18.46\%}$
test_update__nested	0.5097ms	25.7751μs	38.7971 KOps/s	38.9322 KOps/s	$\color{#d91a1a}-0.35\%$
test_set_nested	80.3540μs	14.9527μs	66.8776 KOps/s	59.5495 KOps/s	$\textbf{\color{#35bf28}+12.31\%}$
test_set_nested_new	78.7840μs	17.0494μs	58.6532 KOps/s	52.4034 KOps/s	$\textbf{\color{#35bf28}+11.93\%}$
test_select	0.2184ms	31.2005μs	32.0508 KOps/s	31.9337 KOps/s	$\color{#35bf28}+0.37\%$
test_select_nested	72.5240μs	43.9737μs	22.7408 KOps/s	22.4113 KOps/s	$\color{#35bf28}+1.47\%$
test_exclude_nested	0.1368ms	63.4949μs	15.7493 KOps/s	15.6555 KOps/s	$\color{#35bf28}+0.60\%$
test_empty[True]	0.3490ms	0.2870ms	3.4838 KOps/s	3.4810 KOps/s	$\color{#35bf28}+0.08\%$
test_empty[False]	3.4422μs	0.8657μs	1.1552 MOps/s	1.1508 MOps/s	$\color{#35bf28}+0.38\%$
test_to	86.7040μs	58.4959μs	17.0952 KOps/s	18.0333 KOps/s	$\textbf{\color{#d91a1a}-5.20\%}$
test_to_nonblocking	92.2750μs	48.0903μs	20.7942 KOps/s	21.3273 KOps/s	$\color{#d91a1a}-2.50\%$
test_unbind_speed	0.2900ms	0.2363ms	4.2322 KOps/s	4.2943 KOps/s	$\color{#d91a1a}-1.45\%$
test_unbind_speed_stack0	0.3740ms	0.2384ms	4.1941 KOps/s	4.2604 KOps/s	$\color{#d91a1a}-1.56\%$
test_unbind_speed_stack1	97.3006ms	0.6654ms	1.5028 KOps/s	1.4799 KOps/s	$\color{#35bf28}+1.55\%$
test_split	99.4393ms	1.7082ms	585.4281 Ops/s	632.2556 Ops/s	$\textbf{\color{#d91a1a}-7.41\%}$
test_chunk	0.1016s	1.5857ms	630.6188 Ops/s	629.8363 Ops/s	$\color{#35bf28}+0.12\%$
test_consolidate[False-None]	3.3594ms	2.6531ms	376.9106 Ops/s	326.3413 Ops/s	$\textbf{\color{#35bf28}+15.50\%}$
test_consolidate[default-None]	1.7418ms	1.6677ms	599.6221 Ops/s	591.5615 Ops/s	$\color{#35bf28}+1.36\%$
test_consolidate[reduce-overhead-None]	1.9261ms	1.7068ms	585.9010 Ops/s	576.5160 Ops/s	$\color{#35bf28}+1.63\%$
test_consolidate_njt[False-None]	6.7163ms	6.6214ms	151.0264 Ops/s	150.7099 Ops/s	$\color{#35bf28}+0.21\%$
test_to[False-False-None]	1.9383ms	1.7455ms	572.8864 Ops/s	596.4426 Ops/s	$\color{#d91a1a}-3.95\%$
test_to[True-False-None]	1.5379ms	1.2856ms	777.8640 Ops/s	763.1708 Ops/s	$\color{#35bf28}+1.93\%$
test_to[within-False-None]	4.1523ms	4.0473ms	247.0753 Ops/s	240.3654 Ops/s	$\color{#35bf28}+2.79\%$
test_to[True-default-None]	5.6143ms	5.3529ms	186.8144 Ops/s	187.8142 Ops/s	$\color{#d91a1a}-0.53\%$
test_to_njt[False-False-None]	7.0653ms	6.9627ms	143.6229 Ops/s	146.0411 Ops/s	$\color{#d91a1a}-1.66\%$
test_to_njt[True-False-None]	5.6432ms	5.4610ms	183.1170 Ops/s	182.2364 Ops/s	$\color{#35bf28}+0.48\%$
test_to_njt[within-False-None]	12.3281ms	12.2355ms	81.7291 Ops/s	81.4788 Ops/s	$\color{#35bf28}+0.31\%$
test_creation[device0]	0.4646ms	80.3939μs	12.4388 KOps/s	12.0030 KOps/s	$\color{#35bf28}+3.63\%$
test_creation_from_tensor	0.4359ms	86.4740μs	11.5642 KOps/s	11.9430 KOps/s	$\color{#d91a1a}-3.17\%$
test_add_one[memmap_tensor0]	0.4080ms	6.7084μs	149.0673 KOps/s	150.3664 KOps/s	$\color{#d91a1a}-0.86\%$
test_contiguous[memmap_tensor0]	2.7641μs	0.4030μs	2.4813 MOps/s	2.4385 MOps/s	$\color{#35bf28}+1.76\%$
test_stack[memmap_tensor0]	22.4010μs	4.3260μs	231.1585 KOps/s	227.7971 KOps/s	$\color{#35bf28}+1.48\%$
test_memmaptd_index	1.6204ms	0.2482ms	4.0291 KOps/s	4.0545 KOps/s	$\color{#d91a1a}-0.63\%$
test_memmaptd_index_astensor	0.6040ms	0.3072ms	3.2549 KOps/s	3.2279 KOps/s	$\color{#35bf28}+0.84\%$
test_memmaptd_index_op	0.9927ms	0.5572ms	1.7946 KOps/s	1.6585 KOps/s	$\textbf{\color{#35bf28}+8.21\%}$
test_serialize_model	0.1315s	0.1309s	7.6398 Ops/s	7.6404 Ops/s	$-0.01\%$
test_serialize_model_pickle	1.3553s	1.2121s	0.8250 Ops/s	0.8229 Ops/s	$\color{#35bf28}+0.26\%$
test_serialize_weights	0.1330s	0.1310s	7.6351 Ops/s	7.6524 Ops/s	$\color{#d91a1a}-0.23\%$
test_serialize_weights_returnearly	0.3602s	56.9364ms	17.5635 Ops/s	14.3210 Ops/s	$\textbf{\color{#35bf28}+22.64\%}$
test_serialize_weights_pickle	1.3741s	1.2175s	0.8213 Ops/s	0.8181 Ops/s	$\color{#35bf28}+0.40\%$
test_reshape_pytree	62.1030μs	22.0061μs	45.4420 KOps/s	44.8524 KOps/s	$\color{#35bf28}+1.31\%$
test_reshape_td	66.1930μs	27.1647μs	36.8124 KOps/s	35.6717 KOps/s	$\color{#35bf28}+3.20\%$
test_view_pytree	56.1830μs	21.8501μs	45.7664 KOps/s	44.8764 KOps/s	$\color{#35bf28}+1.98\%$
test_view_td	68.0640μs	31.9942μs	31.2556 KOps/s	30.2404 KOps/s	$\color{#35bf28}+3.36\%$
test_unbind_pytree	70.9340μs	29.5490μs	33.8421 KOps/s	35.1554 KOps/s	$\color{#d91a1a}-3.74\%$
test_unbind_td	0.8484ms	36.6218μs	27.3061 KOps/s	27.2071 KOps/s	$\color{#35bf28}+0.36\%$
test_split_pytree	0.1624ms	29.7804μs	33.5792 KOps/s	33.3868 KOps/s	$\color{#35bf28}+0.58\%$
test_split_td	0.9526ms	38.1188μs	26.2338 KOps/s	25.4769 KOps/s	$\color{#35bf28}+2.97\%$
test_add_pytree	79.2340μs	34.6713μs	28.8423 KOps/s	29.0864 KOps/s	$\color{#d91a1a}-0.84\%$
test_add_td	0.1823ms	44.2475μs	22.6002 KOps/s	19.7076 KOps/s	$\textbf{\color{#35bf28}+14.68\%}$
test_compile_add_one_nested[tensordict-compile]	0.1728ms	0.1203ms	8.3147 KOps/s	8.1088 KOps/s	$\color{#35bf28}+2.54\%$
test_compile_add_one_nested[tensordict-eager]	0.2772ms	0.1297ms	7.7085 KOps/s	7.6189 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_add_one_nested[pytree-compile]	0.1593ms	96.1975μs	10.3953 KOps/s	10.3279 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_add_one_nested[pytree-eager]	1.5720ms	0.1500ms	6.6657 KOps/s	6.8582 KOps/s	$\color{#d91a1a}-2.81\%$
test_compile_copy_nested[tensordict-compile]	67.6940μs	22.3865μs	44.6699 KOps/s	42.5093 KOps/s	$\textbf{\color{#35bf28}+5.08\%}$
test_compile_copy_nested[tensordict-eager]	67.1130μs	29.4361μs	33.9719 KOps/s	33.0760 KOps/s	$\color{#35bf28}+2.71\%$
test_compile_copy_nested[pytree-compile]	0.4306ms	64.3356μs	15.5435 KOps/s	15.4294 KOps/s	$\color{#35bf28}+0.74\%$
test_compile_copy_nested[pytree-eager]	81.9750μs	49.0162μs	20.4014 KOps/s	20.4043 KOps/s	$\color{#d91a1a}-0.01\%$
test_compile_add_one_flat[tensordict-compile]	0.1811ms	0.1417ms	7.0554 KOps/s	6.9327 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_add_one_flat[tensordict-eager]	0.3460ms	0.2164ms	4.6220 KOps/s	4.6727 KOps/s	$\color{#d91a1a}-1.08\%$
test_compile_add_one_flat[tensorclass-compile]	0.1603ms	98.4688μs	10.1555 KOps/s	10.2212 KOps/s	$\color{#d91a1a}-0.64\%$
test_compile_add_one_flat[tensorclass-eager]	0.1212ms	53.5894μs	18.6604 KOps/s	18.6792 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_add_one_flat[pytree-compile]	0.2462ms	0.1367ms	7.3171 KOps/s	7.3232 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_add_one_flat[pytree-eager]	0.6364ms	0.4798ms	2.0843 KOps/s	2.1420 KOps/s	$\color{#d91a1a}-2.69\%$
test_compile_add_self_flat[tensordict-eager]	0.4343ms	0.2589ms	3.8626 KOps/s	3.8786 KOps/s	$\color{#d91a1a}-0.41\%$
test_compile_add_self_flat[tensordict-compile]	0.2058ms	0.1465ms	6.8258 KOps/s	6.9782 KOps/s	$\color{#d91a1a}-2.18\%$
test_compile_add_self_flat[tensorclass-eager]	0.1818ms	65.0052μs	15.3834 KOps/s	15.6223 KOps/s	$\color{#d91a1a}-1.53\%$
test_compile_add_self_flat[tensorclass-compile]	0.1490ms	0.1022ms	9.7892 KOps/s	10.1305 KOps/s	$\color{#d91a1a}-3.37\%$
test_compile_add_self_flat[pytree-eager]	0.5000ms	0.4036ms	2.4777 KOps/s	2.5264 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_add_self_flat[pytree-compile]	0.1958ms	0.1366ms	7.3196 KOps/s	7.3129 KOps/s	$\color{#35bf28}+0.09\%$
test_compile_copy_flat[tensordict-compile]	0.1599ms	19.2093μs	52.0581 KOps/s	51.9408 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_copy_flat[tensordict-eager]	99.9150μs	31.9382μs	31.3105 KOps/s	30.3396 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_copy_flat[pytree-compile]	0.1826ms	70.5547μs	14.1734 KOps/s	14.0728 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_copy_flat[pytree-eager]	85.0140μs	51.8236μs	19.2962 KOps/s	19.4576 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_assign_and_add[tensordict-compile]	1.6047ms	0.3874ms	2.5814 KOps/s	2.2174 KOps/s	$\textbf{\color{#35bf28}+16.42\%}$
test_compile_assign_and_add[tensordict-eager]	2.8615ms	2.6321ms	379.9259 Ops/s	368.4427 Ops/s	$\color{#35bf28}+3.12\%$
test_compile_assign_and_add[pytree-compile]	1.5763ms	0.4282ms	2.3351 KOps/s	2.2831 KOps/s	$\color{#35bf28}+2.28\%$
test_compile_assign_and_add[pytree-eager]	2.8505ms	2.6377ms	379.1251 Ops/s	392.1930 Ops/s	$\color{#d91a1a}-3.33\%$
test_compile_indexing[tensor-tensordict-compile]	0.5397ms	0.1172ms	8.5332 KOps/s	8.8490 KOps/s	$\color{#d91a1a}-3.57\%$
test_compile_indexing[tensor-tensordict-eager]	0.5607ms	82.6012μs	12.1064 KOps/s	12.1601 KOps/s	$\color{#d91a1a}-0.44\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5544ms	0.1119ms	8.9350 KOps/s	9.1469 KOps/s	$\color{#d91a1a}-2.32\%$
test_compile_indexing[tensor-tensorclass-eager]	0.4929ms	72.4508μs	13.8025 KOps/s	14.3422 KOps/s	$\color{#d91a1a}-3.76\%$
test_compile_indexing[tensor-pytree-compile]	0.5364ms	0.1140ms	8.7732 KOps/s	9.1628 KOps/s	$\color{#d91a1a}-4.25\%$
test_compile_indexing[tensor-pytree-eager]	0.5107ms	72.3754μs	13.8169 KOps/s	14.1814 KOps/s	$\color{#d91a1a}-2.57\%$
test_compile_indexing[slice-tensordict-compile]	0.1633ms	0.1034ms	9.6676 KOps/s	9.8970 KOps/s	$\color{#d91a1a}-2.32\%$
test_compile_indexing[slice-tensordict-eager]	0.4248ms	17.5223μs	57.0701 KOps/s	53.6544 KOps/s	$\textbf{\color{#35bf28}+6.37\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1593ms	0.1017ms	9.8291 KOps/s	10.3545 KOps/s	$\textbf{\color{#d91a1a}-5.07\%}$
test_compile_indexing[slice-tensorclass-eager]	0.4192ms	16.0163μs	62.4364 KOps/s	64.4039 KOps/s	$\color{#d91a1a}-3.05\%$
test_compile_indexing[slice-pytree-compile]	0.5467ms	0.1043ms	9.5860 KOps/s	10.2499 KOps/s	$\textbf{\color{#d91a1a}-6.48\%}$
test_compile_indexing[slice-pytree-eager]	0.4222ms	15.6323μs	63.9699 KOps/s	63.3193 KOps/s	$\color{#35bf28}+1.03\%$
test_compile_indexing[int-tensordict-compile]	0.2435ms	0.1057ms	9.4586 KOps/s	9.8299 KOps/s	$\color{#d91a1a}-3.78\%$
test_compile_indexing[int-tensordict-eager]	0.6227ms	16.8256μs	59.4332 KOps/s	57.8398 KOps/s	$\color{#35bf28}+2.75\%$
test_compile_indexing[int-tensorclass-compile]	0.2463ms	99.1660μs	10.0841 KOps/s	10.2638 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_indexing[int-tensorclass-eager]	75.2940μs	16.1687μs	61.8480 KOps/s	64.1620 KOps/s	$\color{#d91a1a}-3.61\%$
test_compile_indexing[int-pytree-compile]	0.1646ms	99.2280μs	10.0778 KOps/s	10.2771 KOps/s	$\color{#d91a1a}-1.94\%$
test_compile_indexing[int-pytree-eager]	51.0030μs	16.0583μs	62.2732 KOps/s	63.7848 KOps/s	$\color{#d91a1a}-2.37\%$
test_mod_add[eager]	0.1304ms	35.8296μs	27.9099 KOps/s	25.8785 KOps/s	$\textbf{\color{#35bf28}+7.85\%}$
test_mod_add[compile]	0.1498ms	80.0693μs	12.4892 KOps/s	12.4516 KOps/s	$\color{#35bf28}+0.30\%$
test_mod_add[compile-overhead]	0.3352ms	0.1715ms	5.8310 KOps/s	5.7200 KOps/s	$\color{#35bf28}+1.94\%$
test_mod_wrap[eager]	0.3895ms	0.2483ms	4.0276 KOps/s	4.0252 KOps/s	$\color{#35bf28}+0.06\%$
test_mod_wrap[compile]	0.3586ms	0.2782ms	3.5952 KOps/s	3.3080 KOps/s	$\textbf{\color{#35bf28}+8.68\%}$
test_mod_wrap[compile-overhead]	6.5653ms	3.6232ms	276.0017 Ops/s	269.3096 Ops/s	$\color{#35bf28}+2.48\%$
test_mod_wrap_and_backward[eager]	1.4795ms	1.3582ms	736.2617 Ops/s	743.6184 Ops/s	$\color{#d91a1a}-0.99\%$
test_mod_wrap_and_backward[compile]	1.3539ms	1.2526ms	798.3610 Ops/s	789.3003 Ops/s	$\color{#35bf28}+1.15\%$
test_mod_wrap_and_backward[compile-overhead]	1.3777ms	0.9281ms	1.0774 KOps/s	1.0621 KOps/s	$\color{#35bf28}+1.45\%$
test_seq_add[eager]	0.1697ms	0.1131ms	8.8453 KOps/s	8.3533 KOps/s	$\textbf{\color{#35bf28}+5.89\%}$
test_seq_add[compile]	0.2816ms	88.0325μs	11.3594 KOps/s	11.4922 KOps/s	$\color{#d91a1a}-1.16\%$
test_seq_add[compile-overhead]	0.1960ms	0.1288ms	7.7657 KOps/s	7.7986 KOps/s	$\color{#d91a1a}-0.42\%$
test_seq_wrap[eager]	0.4740ms	0.4091ms	2.4443 KOps/s	2.3442 KOps/s	$\color{#35bf28}+4.27\%$
test_seq_wrap[compile]	0.4391ms	0.2971ms	3.3662 KOps/s	3.3172 KOps/s	$\color{#35bf28}+1.48\%$
test_seq_wrap[compile-overhead]	0.2784ms	0.2244ms	4.4560 KOps/s	4.4130 KOps/s	$\color{#35bf28}+0.98\%$
test_func_call_runtime[False-eager]	0.8273ms	0.7302ms	1.3694 KOps/s	1.3849 KOps/s	$\color{#d91a1a}-1.12\%$
test_func_call_runtime[False-compile]	0.7893ms	0.7271ms	1.3753 KOps/s	1.3615 KOps/s	$\color{#35bf28}+1.01\%$
test_func_call_runtime[False-compile-overhead]	0.4192ms	0.3633ms	2.7522 KOps/s	2.7373 KOps/s	$\color{#35bf28}+0.55\%$
test_func_call_runtime[True-eager]	1.0375ms	0.8967ms	1.1152 KOps/s	1.1243 KOps/s	$\color{#d91a1a}-0.80\%$
test_func_call_runtime[True-compile]	0.8921ms	0.7564ms	1.3220 KOps/s	1.3279 KOps/s	$\color{#d91a1a}-0.45\%$
test_func_call_runtime[True-compile-overhead]	0.4336ms	0.3833ms	2.6092 KOps/s	2.6074 KOps/s	$\color{#35bf28}+0.07\%$
test_func_call_cm_runtime[False-eager]	0.8601ms	0.7257ms	1.3781 KOps/s	1.3907 KOps/s	$\color{#d91a1a}-0.91\%$
test_func_call_cm_runtime[False-compile]	0.8203ms	0.7327ms	1.3648 KOps/s	1.3512 KOps/s	$\color{#35bf28}+1.01\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4217ms	0.3651ms	2.7389 KOps/s	2.7178 KOps/s	$\color{#35bf28}+0.78\%$
test_func_call_cm_runtime[True-eager]	1.1038ms	0.9937ms	1.0063 KOps/s	997.9730 Ops/s	$\color{#35bf28}+0.84\%$
test_func_call_cm_runtime[True-compile]	0.8690ms	0.7821ms	1.2785 KOps/s	1.2775 KOps/s	$\color{#35bf28}+0.09\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5523ms	0.4088ms	2.4462 KOps/s	2.4137 KOps/s	$\color{#35bf28}+1.35\%$
test_vmap_func_call_cm_runtime[eager]	2.5381ms	2.0864ms	479.2830 Ops/s	477.0301 Ops/s	$\color{#35bf28}+0.47\%$
test_vmap_func_call_cm_runtime[compile]	0.9733ms	0.8020ms	1.2469 KOps/s	1.2479 KOps/s	$\color{#d91a1a}-0.08\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4696ms	0.4125ms	2.4240 KOps/s	2.4046 KOps/s	$\color{#35bf28}+0.80\%$
test_distributed	6.6873ms	0.1903ms	5.2559 KOps/s	8.5075 KOps/s	$\textbf{\color{#d91a1a}-38.22\%}$
test_tdmodule	39.0520μs	18.8636μs	53.0121 KOps/s	48.2937 KOps/s	$\textbf{\color{#35bf28}+9.77\%}$
test_tdmodule_dispatch	62.5730μs	33.4414μs	29.9031 KOps/s	26.8759 KOps/s	$\textbf{\color{#35bf28}+11.26\%}$
test_tdseq	55.6530μs	19.7098μs	50.7362 KOps/s	45.4292 KOps/s	$\textbf{\color{#35bf28}+11.68\%}$
test_tdseq_dispatch	67.3930μs	36.3421μs	27.5163 KOps/s	24.4830 KOps/s	$\textbf{\color{#35bf28}+12.39\%}$
test_instantiation_functorch	1.6567ms	1.5557ms	642.8008 Ops/s	635.0326 Ops/s	$\color{#35bf28}+1.22\%$
test_exec_functorch	0.1888ms	0.1452ms	6.8850 KOps/s	7.0304 KOps/s	$\color{#d91a1a}-2.07\%$
test_exec_functional_call	0.1953ms	0.1364ms	7.3324 KOps/s	7.3691 KOps/s	$\color{#d91a1a}-0.50\%$
test_exec_td_decorator	0.3950ms	0.1854ms	5.3941 KOps/s	5.3977 KOps/s	$\color{#d91a1a}-0.07\%$
test_vmap_mlp_speed_decorator[True-True]	0.8147ms	0.6892ms	1.4510 KOps/s	1.4570 KOps/s	$\color{#d91a1a}-0.41\%$
test_vmap_mlp_speed_decorator[True-False]	0.8743ms	0.6858ms	1.4581 KOps/s	1.4545 KOps/s	$\color{#35bf28}+0.24\%$
test_vmap_mlp_speed_decorator[False-True]	0.7423ms	0.5997ms	1.6676 KOps/s	1.6862 KOps/s	$\color{#d91a1a}-1.10\%$
test_vmap_mlp_speed_decorator[False-False]	0.7153ms	0.6007ms	1.6649 KOps/s	1.6874 KOps/s	$\color{#d91a1a}-1.34\%$
test_vmap_transformer_speed_decorator[True-True]	21.1457ms	19.3539ms	51.6693 Ops/s	52.3269 Ops/s	$\color{#d91a1a}-1.26\%$
test_vmap_transformer_speed_decorator[True-False]	20.0667ms	19.3391ms	51.7087 Ops/s	52.2234 Ops/s	$\color{#d91a1a}-0.99\%$
test_vmap_transformer_speed_decorator[False-True]	19.3003ms	19.1952ms	52.0964 Ops/s	52.7962 Ops/s	$\color{#d91a1a}-1.33\%$
test_vmap_transformer_speed_decorator[False-False]	19.5114ms	19.2012ms	52.0801 Ops/s	52.5698 Ops/s	$\color{#d91a1a}-0.93\%$
test_to_module_speed[True]	1.0608ms	0.9679ms	1.0332 KOps/s	1.0202 KOps/s	$\color{#35bf28}+1.27\%$
test_to_module_speed[False]	1.0519ms	0.9515ms	1.0509 KOps/s	1.0352 KOps/s	$\color{#35bf28}+1.52\%$
test_tc_init	60.7430μs	34.8861μs	28.6647 KOps/s	26.3805 KOps/s	$\textbf{\color{#35bf28}+8.66\%}$
test_tc_init_nested	0.1057ms	68.1989μs	14.6630 KOps/s	13.1850 KOps/s	$\textbf{\color{#35bf28}+11.21\%}$
test_tc_first_layer_tensor	4.4274μs	0.7149μs	1.3987 MOps/s	1.3987 MOps/s	$+0.00\%$
test_tc_first_layer_nontensor	0.1277ms	2.2762μs	439.3290 KOps/s	421.2691 KOps/s	$\color{#35bf28}+4.29\%$
test_tc_second_layer_tensor	10.2973μs	1.4544μs	687.5895 KOps/s	695.8408 KOps/s	$\color{#d91a1a}-1.19\%$
test_tc_second_layer_nontensor	70.6040μs	3.0207μs	331.0546 KOps/s	321.9338 KOps/s	$\color{#35bf28}+2.83\%$
test_unbind	0.2281s	11.1606ms	89.6008 Ops/s	141.8442 Ops/s	$\textbf{\color{#d91a1a}-36.83\%}$
test_full_like	12.3229ms	10.3087ms	97.0055 Ops/s	96.3653 Ops/s	$\color{#35bf28}+0.66\%$
test_zeros_like	5.5395ms	4.5406ms	220.2376 Ops/s	218.3931 Ops/s	$\color{#35bf28}+0.84\%$
test_ones_like	5.2828ms	4.4813ms	223.1508 Ops/s	214.1781 Ops/s	$\color{#35bf28}+4.19\%$
test_clone	9.0446ms	7.1127ms	140.5931 Ops/s	136.3155 Ops/s	$\color{#35bf28}+3.14\%$
test_squeeze	60.8840μs	9.8226μs	101.8061 KOps/s	101.6964 KOps/s	$\color{#35bf28}+0.11\%$
test_unsqueeze	0.1241ms	71.7465μs	13.9380 KOps/s	13.6436 KOps/s	$\color{#35bf28}+2.16\%$
test_split	0.3932ms	0.1580ms	6.3288 KOps/s	6.2148 KOps/s	$\color{#35bf28}+1.83\%$
test_permute	0.2372ms	0.1798ms	5.5627 KOps/s	5.2960 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_stack	54.1820ms	52.2205ms	19.1496 Ops/s	19.0241 Ops/s	$\color{#35bf28}+0.66\%$
test_cat	53.4870ms	52.1520ms	19.1747 Ops/s	19.1332 Ops/s	$\color{#35bf28}+0.22\%$