[Feature] from_any with UserDict

[vmoens]

Stack from ghstack (oldest at bottom):

• [BugFix] auto-batch-size in dispatch #1109
• [Feature] Better list casting in TensorDict.from_any #1108
• [Feature] NonTensorStack.from_list #1107
• -> [Feature] from_any with UserDict #1106

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}25$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	57.3870μs	17.7822μs	56.2360 KOps/s	58.2021 KOps/s	$\color{#d91a1a}-3.38\%$
test_plain_set_stack_nested	49.7730μs	17.6628μs	56.6163 KOps/s	56.8980 KOps/s	$\color{#d91a1a}-0.50\%$
test_plain_set_nested_inplace	48.9010μs	19.6903μs	50.7863 KOps/s	52.7786 KOps/s	$\color{#d91a1a}-3.77\%$
test_plain_set_stack_nested_inplace	47.0380μs	19.6782μs	50.8177 KOps/s	52.9472 KOps/s	$\color{#d91a1a}-4.02\%$
test_items	25.2570μs	4.2654μs	234.4466 KOps/s	238.0282 KOps/s	$\color{#d91a1a}-1.50\%$
test_items_nested	0.5435ms	0.3418ms	2.9253 KOps/s	2.9129 KOps/s	$\color{#35bf28}+0.43\%$
test_items_nested_locked	0.6466ms	0.3421ms	2.9227 KOps/s	2.8299 KOps/s	$\color{#35bf28}+3.28\%$
test_items_nested_leaf	0.1386ms	71.3025μs	14.0247 KOps/s	13.7932 KOps/s	$\color{#35bf28}+1.68\%$
test_items_stack_nested	0.6946ms	0.3478ms	2.8749 KOps/s	2.9076 KOps/s	$\color{#d91a1a}-1.12\%$
test_items_stack_nested_leaf	0.1471ms	75.1587μs	13.3052 KOps/s	13.0007 KOps/s	$\color{#35bf28}+2.34\%$
test_items_stack_nested_locked	0.5202ms	0.3439ms	2.9078 KOps/s	2.8892 KOps/s	$\color{#35bf28}+0.64\%$
test_keys	24.6460μs	3.5359μs	282.8096 KOps/s	280.3229 KOps/s	$\color{#35bf28}+0.89\%$
test_keys_nested	0.2473ms	0.1358ms	7.3616 KOps/s	7.2512 KOps/s	$\color{#35bf28}+1.52\%$
test_keys_nested_locked	0.7551ms	0.1432ms	6.9825 KOps/s	6.8360 KOps/s	$\color{#35bf28}+2.14\%$
test_keys_nested_leaf	0.2134ms	0.1169ms	8.5508 KOps/s	8.3942 KOps/s	$\color{#35bf28}+1.87\%$
test_keys_stack_nested	0.2751ms	0.1371ms	7.2923 KOps/s	7.2097 KOps/s	$\color{#35bf28}+1.14\%$
test_keys_stack_nested_leaf	0.2011ms	0.1170ms	8.5455 KOps/s	8.4688 KOps/s	$\color{#35bf28}+0.91\%$
test_keys_stack_nested_locked	0.2732ms	0.1416ms	7.0629 KOps/s	6.9050 KOps/s	$\color{#35bf28}+2.29\%$
test_values	7.0783μs	1.1396μs	877.4825 KOps/s	956.1549 KOps/s	$\textbf{\color{#d91a1a}-8.23\%}$
test_values_nested	0.1392ms	54.9236μs	18.2071 KOps/s	17.6170 KOps/s	$\color{#35bf28}+3.35\%$
test_values_nested_locked	0.1039ms	54.8790μs	18.2219 KOps/s	17.6402 KOps/s	$\color{#35bf28}+3.30\%$
test_values_nested_leaf	0.1151ms	59.7524μs	16.7357 KOps/s	16.2689 KOps/s	$\color{#35bf28}+2.87\%$
test_values_stack_nested	0.1104ms	57.4711μs	17.4000 KOps/s	17.3383 KOps/s	$\color{#35bf28}+0.36\%$
test_values_stack_nested_leaf	0.1157ms	61.0579μs	16.3779 KOps/s	15.2782 KOps/s	$\textbf{\color{#35bf28}+7.20\%}$
test_values_stack_nested_locked	0.1078ms	56.8840μs	17.5796 KOps/s	17.1693 KOps/s	$\color{#35bf28}+2.39\%$
test_membership	4.0017μs	0.7707μs	1.2975 MOps/s	1.4059 MOps/s	$\textbf{\color{#d91a1a}-7.71\%}$
test_membership_nested	20.9590μs	2.6782μs	373.3888 KOps/s	355.7604 KOps/s	$\color{#35bf28}+4.96\%$
test_membership_nested_leaf	22.7830μs	2.7231μs	367.2243 KOps/s	352.4923 KOps/s	$\color{#35bf28}+4.18\%$
test_membership_stacked_nested	24.4350μs	2.6706μs	374.4464 KOps/s	360.7544 KOps/s	$\color{#35bf28}+3.80\%$
test_membership_stacked_nested_leaf	14.5480μs	2.6972μs	370.7600 KOps/s	355.2781 KOps/s	$\color{#35bf28}+4.36\%$
test_membership_nested_last	24.9570μs	3.9874μs	250.7914 KOps/s	240.7302 KOps/s	$\color{#35bf28}+4.18\%$
test_membership_nested_leaf_last	34.3540μs	3.9677μs	252.0327 KOps/s	238.7743 KOps/s	$\textbf{\color{#35bf28}+5.55\%}$
test_membership_stacked_nested_last	25.5480μs	3.9473μs	253.3347 KOps/s	242.8791 KOps/s	$\color{#35bf28}+4.30\%$
test_membership_stacked_nested_leaf_last	18.6050μs	3.9560μs	252.7829 KOps/s	241.8810 KOps/s	$\color{#35bf28}+4.51\%$
test_nested_getleaf	0.1552ms	11.1093μs	90.0149 KOps/s	91.9300 KOps/s	$\color{#d91a1a}-2.08\%$
test_nested_get	33.3520μs	10.0217μs	99.7837 KOps/s	94.6217 KOps/s	$\textbf{\color{#35bf28}+5.46\%}$
test_stacked_getleaf	46.1660μs	10.4132μs	96.0322 KOps/s	96.8284 KOps/s	$\color{#d91a1a}-0.82\%$
test_stacked_get	35.5470μs	9.9285μs	100.7199 KOps/s	97.5766 KOps/s	$\color{#35bf28}+3.22\%$
test_nested_getitemleaf	30.4870μs	10.9078μs	91.6777 KOps/s	88.8190 KOps/s	$\color{#35bf28}+3.22\%$
test_nested_getitem	30.0360μs	10.0733μs	99.2719 KOps/s	96.6588 KOps/s	$\color{#35bf28}+2.70\%$
test_stacked_getitemleaf	0.1285ms	10.7902μs	92.6766 KOps/s	88.2876 KOps/s	$\color{#35bf28}+4.97\%$
test_stacked_getitem	32.5210μs	10.0356μs	99.6457 KOps/s	92.6576 KOps/s	$\textbf{\color{#35bf28}+7.54\%}$
test_lock_nested	0.8555ms	0.4429ms	2.2576 KOps/s	2.2126 KOps/s	$\color{#35bf28}+2.04\%$
test_lock_stack_nested	0.8786ms	0.4143ms	2.4136 KOps/s	2.3635 KOps/s	$\color{#35bf28}+2.12\%$
test_unlock_nested	0.6829ms	0.3623ms	2.7600 KOps/s	2.6922 KOps/s	$\color{#35bf28}+2.52\%$
test_unlock_stack_nested	0.6390ms	0.3302ms	3.0287 KOps/s	2.9542 KOps/s	$\color{#35bf28}+2.52\%$
test_flatten_speed	0.2075ms	95.2479μs	10.4989 KOps/s	10.4138 KOps/s	$\color{#35bf28}+0.82\%$
test_unflatten_speed	0.6249ms	0.4700ms	2.1276 KOps/s	1.9941 KOps/s	$\textbf{\color{#35bf28}+6.69\%}$
test_common_ops	4.6066ms	0.7693ms	1.2998 KOps/s	1.3603 KOps/s	$\color{#d91a1a}-4.44\%$
test_creation	75.4410μs	2.0526μs	487.1911 KOps/s	468.6242 KOps/s	$\color{#35bf28}+3.96\%$
test_creation_empty	39.2130μs	10.1803μs	98.2285 KOps/s	108.8066 KOps/s	$\textbf{\color{#d91a1a}-9.72\%}$
test_creation_nested_1	37.7810μs	13.0029μs	76.9060 KOps/s	82.1107 KOps/s	$\textbf{\color{#d91a1a}-6.34\%}$
test_creation_nested_2	48.1100μs	17.2734μs	57.8925 KOps/s	61.4911 KOps/s	$\textbf{\color{#d91a1a}-5.85\%}$
test_clone	0.1040ms	13.4414μs	74.3969 KOps/s	75.2674 KOps/s	$\color{#d91a1a}-1.16\%$
test_getitem[int]	0.8647ms	12.7860μs	78.2105 KOps/s	77.1435 KOps/s	$\color{#35bf28}+1.38\%$
test_getitem[slice_int]	0.1927ms	25.4168μs	39.3441 KOps/s	40.4638 KOps/s	$\color{#d91a1a}-2.77\%$
test_getitem[range]	0.1711ms	49.2400μs	20.3087 KOps/s	20.6835 KOps/s	$\color{#d91a1a}-1.81\%$
test_getitem[tuple]	0.1297ms	20.3071μs	49.2438 KOps/s	48.9413 KOps/s	$\color{#35bf28}+0.62\%$
test_getitem[list]	0.2598ms	44.2928μs	22.5770 KOps/s	22.6328 KOps/s	$\color{#d91a1a}-0.25\%$
test_setitem_dim[int]	48.1300μs	27.0573μs	36.9586 KOps/s	39.8789 KOps/s	$\textbf{\color{#d91a1a}-7.32\%}$
test_setitem_dim[slice_int]	98.0630μs	53.7657μs	18.5992 KOps/s	19.6353 KOps/s	$\textbf{\color{#d91a1a}-5.28\%}$
test_setitem_dim[range]	0.1430ms	73.9447μs	13.5236 KOps/s	13.4677 KOps/s	$\color{#35bf28}+0.42\%$
test_setitem_dim[tuple]	87.0130μs	41.9057μs	23.8631 KOps/s	24.6413 KOps/s	$\color{#d91a1a}-3.16\%$
test_setitem	0.1530ms	20.2671μs	49.3411 KOps/s	52.3778 KOps/s	$\textbf{\color{#d91a1a}-5.80\%}$
test_set	0.1187ms	19.6584μs	50.8690 KOps/s	53.8544 KOps/s	$\textbf{\color{#d91a1a}-5.54\%}$
test_set_shared	3.2370ms	0.1697ms	5.8923 KOps/s	5.8987 KOps/s	$\color{#d91a1a}-0.11\%$
test_update	0.1274ms	21.7546μs	45.9673 KOps/s	50.0952 KOps/s	$\textbf{\color{#d91a1a}-8.24\%}$
test_update_nested	0.1301ms	31.7802μs	31.4662 KOps/s	32.7887 KOps/s	$\color{#d91a1a}-4.03\%$
test_update__nested	0.6170ms	34.0876μs	29.3361 KOps/s	30.5029 KOps/s	$\color{#d91a1a}-3.83\%$
test_set_nested	0.1569ms	22.6486μs	44.1529 KOps/s	48.3250 KOps/s	$\textbf{\color{#d91a1a}-8.63\%}$
test_set_nested_new	0.3456ms	29.5811μs	33.8053 KOps/s	39.2721 KOps/s	$\textbf{\color{#d91a1a}-13.92\%}$
test_select	0.1772ms	45.3310μs	22.0599 KOps/s	24.4362 KOps/s	$\textbf{\color{#d91a1a}-9.72\%}$
test_select_nested	0.1195ms	60.0873μs	16.6425 KOps/s	16.0265 KOps/s	$\color{#35bf28}+3.84\%$
test_exclude_nested	0.4226ms	77.7534μs	12.8612 KOps/s	12.6777 KOps/s	$\color{#35bf28}+1.45\%$
test_empty[True]	0.5663ms	0.3485ms	2.8692 KOps/s	2.7438 KOps/s	$\color{#35bf28}+4.57\%$
test_empty[False]	11.9852μs	1.2807μs	780.8149 KOps/s	804.8554 KOps/s	$\color{#d91a1a}-2.99\%$
test_unbind_speed	0.4424ms	0.2679ms	3.7322 KOps/s	3.7146 KOps/s	$\color{#35bf28}+0.47\%$
test_unbind_speed_stack0	0.5353ms	0.2574ms	3.8846 KOps/s	3.8116 KOps/s	$\color{#35bf28}+1.92\%$
test_unbind_speed_stack1	97.9098ms	0.7605ms	1.3149 KOps/s	1.4369 KOps/s	$\textbf{\color{#d91a1a}-8.49\%}$
test_split	1.7712ms	1.5719ms	636.1874 Ops/s	566.4865 Ops/s	$\textbf{\color{#35bf28}+12.30\%}$
test_chunk	0.1003s	1.9060ms	524.6537 Ops/s	570.0143 Ops/s	$\textbf{\color{#d91a1a}-7.96\%}$
test_consolidate_njt[False-None]	14.5592ms	8.4283ms	118.6474 Ops/s	120.7610 Ops/s	$\color{#d91a1a}-1.75\%$
test_creation[device0]	0.2213ms	92.5308μs	10.8072 KOps/s	10.6843 KOps/s	$\color{#35bf28}+1.15\%$
test_creation_from_tensor	4.1571ms	96.1495μs	10.4005 KOps/s	10.3801 KOps/s	$\color{#35bf28}+0.20\%$
test_add_one[memmap_tensor0]	0.3098ms	5.1300μs	194.9317 KOps/s	204.9396 KOps/s	$\color{#d91a1a}-4.88\%$
test_contiguous[memmap_tensor0]	23.8050μs	0.5427μs	1.8425 MOps/s	1.9763 MOps/s	$\textbf{\color{#d91a1a}-6.77\%}$
test_stack[memmap_tensor0]	52.2770μs	3.4588μs	289.1209 KOps/s	295.1460 KOps/s	$\color{#d91a1a}-2.04\%$
test_memmaptd_index	0.9460ms	0.2393ms	4.1793 KOps/s	4.1668 KOps/s	$\color{#35bf28}+0.30\%$
test_memmaptd_index_astensor	0.5900ms	0.3197ms	3.1275 KOps/s	3.1318 KOps/s	$\color{#d91a1a}-0.14\%$
test_memmaptd_index_op	1.1066ms	0.5933ms	1.6854 KOps/s	1.8023 KOps/s	$\textbf{\color{#d91a1a}-6.49\%}$
test_serialize_model	0.1220s	0.1134s	8.8184 Ops/s	7.7004 Ops/s	$\textbf{\color{#35bf28}+14.52\%}$
test_serialize_model_pickle	0.4458s	0.3903s	2.5621 Ops/s	2.5860 Ops/s	$\color{#d91a1a}-0.92\%$
test_serialize_weights	0.1206s	0.1135s	8.8105 Ops/s	8.8786 Ops/s	$\color{#d91a1a}-0.77\%$
test_serialize_weights_returnearly	0.2668s	0.1735s	5.7623 Ops/s	6.4622 Ops/s	$\textbf{\color{#d91a1a}-10.83\%}$
test_serialize_weights_pickle	1.0555s	0.7365s	1.3578 Ops/s	2.4046 Ops/s	$\textbf{\color{#d91a1a}-43.53\%}$
test_serialize_weights_filesystem	0.1442s	0.1397s	7.1570 Ops/s	6.3735 Ops/s	$\textbf{\color{#35bf28}+12.29\%}$
test_serialize_model_filesystem	0.1452s	0.1404s	7.1231 Ops/s	6.7774 Ops/s	$\textbf{\color{#35bf28}+5.10\%}$
test_reshape_pytree	58.1090μs	27.1447μs	36.8396 KOps/s	34.9658 KOps/s	$\textbf{\color{#35bf28}+5.36\%}$
test_reshape_td	76.3430μs	33.9266μs	29.4754 KOps/s	29.1653 KOps/s	$\color{#35bf28}+1.06\%$
test_view_pytree	74.1990μs	27.1027μs	36.8967 KOps/s	35.8991 KOps/s	$\color{#35bf28}+2.78\%$
test_view_td	0.1071ms	39.0625μs	25.6000 KOps/s	25.4325 KOps/s	$\color{#35bf28}+0.66\%$
test_unbind_pytree	69.4000μs	30.3625μs	32.9353 KOps/s	32.0943 KOps/s	$\color{#35bf28}+2.62\%$
test_unbind_td	0.3442ms	39.7551μs	25.1540 KOps/s	24.8587 KOps/s	$\color{#35bf28}+1.19\%$
test_split_pytree	79.1180μs	30.1587μs	33.1580 KOps/s	32.1729 KOps/s	$\color{#35bf28}+3.06\%$
test_split_td	0.5311ms	45.1318μs	22.1573 KOps/s	21.6244 KOps/s	$\color{#35bf28}+2.46\%$
test_add_pytree	80.2000μs	36.7078μs	27.2422 KOps/s	27.4238 KOps/s	$\color{#d91a1a}-0.66\%$
test_add_td	0.1274ms	59.0517μs	16.9343 KOps/s	17.9700 KOps/s	$\textbf{\color{#d91a1a}-5.76\%}$
test_compile_add_one_nested[tensordict-compile]	0.1231ms	62.0631μs	16.1126 KOps/s	15.7986 KOps/s	$\color{#35bf28}+1.99\%$
test_compile_add_one_nested[tensordict-eager]	1.4829ms	0.1634ms	6.1194 KOps/s	6.0526 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_add_one_nested[pytree-compile]	99.4460μs	45.5803μs	21.9393 KOps/s	21.7224 KOps/s	$\color{#35bf28}+1.00\%$
test_compile_add_one_nested[pytree-eager]	0.2458ms	0.1204ms	8.3080 KOps/s	8.3057 KOps/s	$\color{#35bf28}+0.03\%$
test_compile_copy_nested[tensordict-compile]	59.0510μs	26.1376μs	38.2590 KOps/s	38.3274 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_copy_nested[tensordict-eager]	0.1109ms	53.7428μs	18.6072 KOps/s	17.9563 KOps/s	$\color{#35bf28}+3.62\%$
test_compile_copy_nested[pytree-compile]	0.1583ms	78.3860μs	12.7574 KOps/s	12.0340 KOps/s	$\textbf{\color{#35bf28}+6.01\%}$
test_compile_copy_nested[pytree-eager]	0.1416ms	67.8191μs	14.7451 KOps/s	14.4554 KOps/s	$\color{#35bf28}+2.00\%$
test_compile_add_one_flat[tensordict-compile]	0.2173ms	0.1046ms	9.5580 KOps/s	9.5976 KOps/s	$\color{#d91a1a}-0.41\%$
test_compile_add_one_flat[tensordict-eager]	1.2195ms	0.1997ms	5.0076 KOps/s	4.9520 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_add_one_flat[tensorclass-compile]	0.1211ms	45.1091μs	22.1685 KOps/s	22.5752 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_add_one_flat[tensorclass-eager]	1.1759ms	64.0015μs	15.6246 KOps/s	15.8889 KOps/s	$\color{#d91a1a}-1.66\%$
test_compile_add_one_flat[pytree-compile]	0.2164ms	0.1018ms	9.8253 KOps/s	9.6646 KOps/s	$\color{#35bf28}+1.66\%$
test_compile_add_one_flat[pytree-eager]	0.4286ms	0.2102ms	4.7577 KOps/s	4.9523 KOps/s	$\color{#d91a1a}-3.93\%$
test_compile_add_self_flat[tensordict-eager]	0.7252ms	0.2126ms	4.7046 KOps/s	4.6869 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_add_self_flat[tensordict-compile]	0.2169ms	0.1044ms	9.5807 KOps/s	9.4475 KOps/s	$\color{#35bf28}+1.41\%$
test_compile_add_self_flat[tensorclass-eager]	0.1519ms	58.3513μs	17.1376 KOps/s	17.7934 KOps/s	$\color{#d91a1a}-3.69\%$
test_compile_add_self_flat[tensorclass-compile]	0.1140ms	45.8820μs	21.7950 KOps/s	21.7522 KOps/s	$\color{#35bf28}+0.20\%$
test_compile_add_self_flat[pytree-eager]	0.5812ms	0.1589ms	6.2946 KOps/s	6.2661 KOps/s	$\color{#35bf28}+0.46\%$
test_compile_add_self_flat[pytree-compile]	0.2030ms	0.1031ms	9.6992 KOps/s	9.4634 KOps/s	$\color{#35bf28}+2.49\%$
test_compile_copy_flat[tensordict-compile]	74.8700μs	20.5162μs	48.7420 KOps/s	48.3533 KOps/s	$\color{#35bf28}+0.80\%$
test_compile_copy_flat[tensordict-eager]	0.1226ms	58.1941μs	17.1839 KOps/s	16.3147 KOps/s	$\textbf{\color{#35bf28}+5.33\%}$
test_compile_copy_flat[pytree-compile]	0.1472ms	80.2840μs	12.4558 KOps/s	11.8275 KOps/s	$\textbf{\color{#35bf28}+5.31\%}$
test_compile_copy_flat[pytree-eager]	0.1446ms	67.8986μs	14.7279 KOps/s	14.1764 KOps/s	$\color{#35bf28}+3.89\%$
test_compile_assign_and_add[tensordict-compile]	0.4428ms	0.2059ms	4.8578 KOps/s	4.7928 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_assign_and_add[tensordict-eager]	1.4924ms	1.2584ms	794.6561 Ops/s	775.3628 Ops/s	$\color{#35bf28}+2.49\%$
test_compile_assign_and_add[pytree-compile]	0.3136ms	0.2022ms	4.9457 KOps/s	5.0355 KOps/s	$\color{#d91a1a}-1.78\%$
test_compile_assign_and_add[pytree-eager]	0.9678ms	0.7727ms	1.2942 KOps/s	1.2740 KOps/s	$\color{#35bf28}+1.59\%$
test_compile_assign_and_add_stack[compile]	0.5519ms	0.4477ms	2.2334 KOps/s	2.2107 KOps/s	$\color{#35bf28}+1.03\%$
test_compile_assign_and_add_stack[eager]	2.8231ms	2.5896ms	386.1639 Ops/s	396.8444 Ops/s	$\color{#d91a1a}-2.69\%$
test_compile_indexing[tensor-tensordict-compile]	90.3890μs	36.0228μs	27.7602 KOps/s	28.1718 KOps/s	$\color{#d91a1a}-1.46\%$
test_compile_indexing[tensor-tensordict-eager]	0.5245ms	34.2344μs	29.2104 KOps/s	29.5407 KOps/s	$\color{#d91a1a}-1.12\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1113ms	29.2724μs	34.1619 KOps/s	34.4201 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_indexing[tensor-tensorclass-eager]	69.9900μs	23.9930μs	41.6788 KOps/s	42.2956 KOps/s	$\color{#d91a1a}-1.46\%$
test_compile_indexing[tensor-pytree-compile]	0.1191ms	29.8289μs	33.5246 KOps/s	33.2800 KOps/s	$\color{#35bf28}+0.73\%$
test_compile_indexing[tensor-pytree-eager]	77.1750μs	23.8096μs	41.9998 KOps/s	42.1003 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_indexing[slice-tensordict-compile]	0.1350ms	52.4395μs	19.0696 KOps/s	19.3887 KOps/s	$\color{#d91a1a}-1.65\%$
test_compile_indexing[slice-tensordict-eager]	0.5805ms	20.6287μs	48.4761 KOps/s	48.2813 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_indexing[slice-tensorclass-compile]	0.1031ms	44.8347μs	22.3041 KOps/s	21.4660 KOps/s	$\color{#35bf28}+3.90\%$
test_compile_indexing[slice-tensorclass-eager]	89.6570μs	19.5680μs	51.1040 KOps/s	51.2594 KOps/s	$\color{#d91a1a}-0.30\%$
test_compile_indexing[slice-pytree-compile]	95.6890μs	45.5940μs	21.9327 KOps/s	20.8336 KOps/s	$\textbf{\color{#35bf28}+5.28\%}$
test_compile_indexing[slice-pytree-eager]	79.9400μs	19.4521μs	51.4084 KOps/s	50.8430 KOps/s	$\color{#35bf28}+1.11\%$
test_compile_indexing[int-tensordict-compile]	0.1119ms	53.1123μs	18.8280 KOps/s	18.4689 KOps/s	$\color{#35bf28}+1.94\%$
test_compile_indexing[int-tensordict-eager]	1.1093ms	20.2816μs	49.3057 KOps/s	48.5562 KOps/s	$\color{#35bf28}+1.54\%$
test_compile_indexing[int-tensorclass-compile]	0.1339ms	45.3401μs	22.0555 KOps/s	21.9398 KOps/s	$\color{#35bf28}+0.53\%$
test_compile_indexing[int-tensorclass-eager]	0.2059ms	20.0625μs	49.8441 KOps/s	51.8905 KOps/s	$\color{#d91a1a}-3.94\%$
test_compile_indexing[int-pytree-compile]	0.1243ms	45.4167μs	22.0184 KOps/s	21.8710 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_indexing[int-pytree-eager]	76.4830μs	19.1275μs	52.2807 KOps/s	51.1840 KOps/s	$\color{#35bf28}+2.14\%$
test_mod_add[eager]	79.7200μs	27.1571μs	36.8228 KOps/s	40.1377 KOps/s	$\textbf{\color{#d91a1a}-8.26\%}$
test_mod_add[compile]	0.1001ms	44.4916μs	22.4762 KOps/s	21.7504 KOps/s	$\color{#35bf28}+3.34\%$
test_mod_add[compile-overhead]	0.1182ms	44.6143μs	22.4143 KOps/s	21.9113 KOps/s	$\color{#35bf28}+2.30\%$
test_mod_wrap[eager]	0.4129ms	0.2166ms	4.6176 KOps/s	4.7206 KOps/s	$\color{#d91a1a}-2.18\%$
test_mod_wrap[compile]	1.6476ms	0.2043ms	4.8952 KOps/s	4.8752 KOps/s	$\color{#35bf28}+0.41\%$
test_mod_wrap[compile-overhead]	1.4703ms	0.2007ms	4.9831 KOps/s	4.9051 KOps/s	$\color{#35bf28}+1.59\%$
test_mod_wrap_and_backward[eager]	15.9074ms	12.1691ms	82.1750 Ops/s	82.3607 Ops/s	$\color{#d91a1a}-0.23\%$
test_mod_wrap_and_backward[compile]	16.3329ms	13.1628ms	75.9718 Ops/s	79.5492 Ops/s	$\color{#d91a1a}-4.50\%$
test_mod_wrap_and_backward[compile-overhead]	15.1946ms	13.5378ms	73.8672 Ops/s	85.8341 Ops/s	$\textbf{\color{#d91a1a}-13.94\%}$
test_seq_add[eager]	0.1671ms	93.1053μs	10.7405 KOps/s	11.0995 KOps/s	$\color{#d91a1a}-3.23\%$
test_seq_add[compile]	0.1242ms	58.3725μs	17.1313 KOps/s	16.4146 KOps/s	$\color{#35bf28}+4.37\%$
test_seq_add[compile-overhead]	0.1317ms	58.0815μs	17.2172 KOps/s	16.2661 KOps/s	$\textbf{\color{#35bf28}+5.85\%}$
test_seq_wrap[eager]	0.6789ms	0.3943ms	2.5362 KOps/s	2.5683 KOps/s	$\color{#d91a1a}-1.25\%$
test_seq_wrap[compile]	0.4198ms	0.2244ms	4.4555 KOps/s	4.3739 KOps/s	$\color{#35bf28}+1.87\%$
test_seq_wrap[compile-overhead]	0.4222ms	0.2228ms	4.4874 KOps/s	4.3883 KOps/s	$\color{#35bf28}+2.26\%$
test_func_call_runtime[False-eager]	0.7168ms	0.5572ms	1.7946 KOps/s	1.8208 KOps/s	$\color{#d91a1a}-1.43\%$
test_func_call_runtime[False-compile]	0.5661ms	0.4250ms	2.3529 KOps/s	2.3290 KOps/s	$\color{#35bf28}+1.03\%$
test_func_call_runtime[False-compile-overhead]	0.5545ms	0.4213ms	2.3734 KOps/s	2.3885 KOps/s	$\color{#d91a1a}-0.63\%$
test_func_call_runtime[True-eager]	1.2502ms	0.7764ms	1.2879 KOps/s	1.3087 KOps/s	$\color{#d91a1a}-1.59\%$
test_func_call_runtime[True-compile]	0.9487ms	0.4735ms	2.1119 KOps/s	2.1671 KOps/s	$\color{#d91a1a}-2.55\%$
test_func_call_runtime[True-compile-overhead]	0.7033ms	0.4622ms	2.1635 KOps/s	2.1620 KOps/s	$\color{#35bf28}+0.07\%$
test_func_call_cm_runtime[False-eager]	0.7033ms	0.5554ms	1.8005 KOps/s	1.8494 KOps/s	$\color{#d91a1a}-2.64\%$
test_func_call_cm_runtime[False-compile]	0.7775ms	0.4213ms	2.3738 KOps/s	2.3622 KOps/s	$\color{#35bf28}+0.49\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6221ms	0.4240ms	2.3587 KOps/s	2.3915 KOps/s	$\color{#d91a1a}-1.37\%$
test_func_call_cm_runtime[True-eager]	1.4697ms	0.9090ms	1.1001 KOps/s	1.1057 KOps/s	$\color{#d91a1a}-0.50\%$
test_func_call_cm_runtime[True-compile]	0.9693ms	0.4894ms	2.0434 KOps/s	2.0379 KOps/s	$\color{#35bf28}+0.27\%$
test_func_call_cm_runtime[True-compile-overhead]	0.8724ms	0.4925ms	2.0306 KOps/s	2.0497 KOps/s	$\color{#d91a1a}-0.93\%$
test_vmap_func_call_cm_runtime[eager]	2.7617ms	1.8963ms	527.3419 Ops/s	526.3827 Ops/s	$\color{#35bf28}+0.18\%$
test_vmap_func_call_cm_runtime[compile]	0.6876ms	0.5193ms	1.9256 KOps/s	1.8987 KOps/s	$\color{#35bf28}+1.42\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.7854ms	0.5163ms	1.9368 KOps/s	1.3823 KOps/s	$\textbf{\color{#35bf28}+40.11\%}$
test_distributed	0.2332ms	0.1264ms	7.9120 KOps/s	7.8289 KOps/s	$\color{#35bf28}+1.06\%$
test_tdmodule	76.6830μs	19.2913μs	51.8369 KOps/s	55.5619 KOps/s	$\textbf{\color{#d91a1a}-6.70\%}$
test_tdmodule_dispatch	91.0410μs	40.6721μs	24.5869 KOps/s	26.0660 KOps/s	$\textbf{\color{#d91a1a}-5.67\%}$
test_tdseq	44.2320μs	20.8820μs	47.8881 KOps/s	48.4898 KOps/s	$\color{#d91a1a}-1.24\%$
test_tdseq_dispatch	72.6360μs	43.3988μs	23.0421 KOps/s	23.0953 KOps/s	$\color{#d91a1a}-0.23\%$
test_instantiation_functorch	2.4106ms	1.5738ms	635.4149 Ops/s	634.2244 Ops/s	$\color{#35bf28}+0.19\%$
test_exec_functorch	0.3474ms	0.1839ms	5.4372 KOps/s	5.4770 KOps/s	$\color{#d91a1a}-0.73\%$
test_exec_functional_call	0.4692ms	0.1798ms	5.5610 KOps/s	5.6655 KOps/s	$\color{#d91a1a}-1.84\%$
test_exec_td_decorator	0.5185ms	0.2349ms	4.2574 KOps/s	4.2471 KOps/s	$\color{#35bf28}+0.24\%$
test_vmap_mlp_speed_decorator[True-True]	0.9292ms	0.6485ms	1.5421 KOps/s	1.5826 KOps/s	$\color{#d91a1a}-2.56\%$
test_vmap_mlp_speed_decorator[True-False]	0.8705ms	0.6679ms	1.4973 KOps/s	1.5183 KOps/s	$\color{#d91a1a}-1.39\%$
test_vmap_mlp_speed_decorator[False-True]	0.6763ms	0.5308ms	1.8841 KOps/s	1.9036 KOps/s	$\color{#d91a1a}-1.02\%$
test_vmap_mlp_speed_decorator[False-False]	0.9652ms	0.5347ms	1.8701 KOps/s	1.8954 KOps/s	$\color{#d91a1a}-1.33\%$
test_to_module_speed[True]	1.4238ms	1.2965ms	771.2991 Ops/s	738.7619 Ops/s	$\color{#35bf28}+4.40\%$
test_to_module_speed[False]	1.4112ms	1.2601ms	793.6102 Ops/s	762.1204 Ops/s	$\color{#35bf28}+4.13\%$
test_tc_init	0.1030ms	46.6389μs	21.4413 KOps/s	22.5263 KOps/s	$\color{#d91a1a}-4.82\%$
test_tc_init_nested	0.1837ms	90.8174μs	11.0111 KOps/s	11.1619 KOps/s	$\color{#d91a1a}-1.35\%$
test_tc_first_layer_tensor	22.0620μs	1.5530μs	643.9265 KOps/s	643.1178 KOps/s	$\color{#35bf28}+0.13\%$
test_tc_first_layer_nontensor	28.6130μs	4.7196μs	211.8833 KOps/s	204.0030 KOps/s	$\color{#35bf28}+3.86\%$
test_tc_second_layer_tensor	20.9790μs	2.8726μs	348.1208 KOps/s	335.7234 KOps/s	$\color{#35bf28}+3.69\%$
test_tc_second_layer_nontensor	32.3000μs	6.0326μs	165.7646 KOps/s	157.4445 KOps/s	$\textbf{\color{#35bf28}+5.28\%}$
test_unbind	0.2139s	14.4931ms	68.9983 Ops/s	84.2556 Ops/s	$\textbf{\color{#d91a1a}-18.11\%}$
test_full_like	12.2111ms	7.0792ms	141.2599 Ops/s	142.9454 Ops/s	$\color{#d91a1a}-1.18\%$
test_zeros_like	3.1914ms	2.6496ms	377.4160 Ops/s	347.5464 Ops/s	$\textbf{\color{#35bf28}+8.59\%}$
test_ones_like	3.3217ms	3.0482ms	328.0656 Ops/s	294.3728 Ops/s	$\textbf{\color{#35bf28}+11.45\%}$
test_clone	5.2922ms	4.8082ms	207.9766 Ops/s	196.3739 Ops/s	$\textbf{\color{#35bf28}+5.91\%}$
test_squeeze	59.2810μs	12.5142μs	79.9093 KOps/s	81.7410 KOps/s	$\color{#d91a1a}-2.24\%$
test_unsqueeze	0.1875ms	90.4301μs	11.0583 KOps/s	10.8955 KOps/s	$\color{#35bf28}+1.49\%$
test_split	0.5082ms	0.1910ms	5.2345 KOps/s	5.1627 KOps/s	$\color{#35bf28}+1.39\%$
test_permute	0.3277ms	0.2212ms	4.5217 KOps/s	4.4424 KOps/s	$\color{#35bf28}+1.78\%$
test_stack	30.3543ms	24.1378ms	41.4289 Ops/s	40.5928 Ops/s	$\color{#35bf28}+2.06\%$
test_cat	28.0407ms	23.9099ms	41.8237 Ops/s	41.2915 Ops/s	$\color{#35bf28}+1.29\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}19$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	26.0200μs	10.6983μs	93.4729 KOps/s	96.4488 KOps/s	$\color{#d91a1a}-3.09\%$
test_plain_set_stack_nested	34.9910μs	11.1135μs	89.9805 KOps/s	96.9595 KOps/s	$\textbf{\color{#d91a1a}-7.20\%}$
test_plain_set_nested_inplace	41.3900μs	12.0719μs	82.8372 KOps/s	89.4227 KOps/s	$\textbf{\color{#d91a1a}-7.36\%}$
test_plain_set_stack_nested_inplace	36.0310μs	11.9072μs	83.9831 KOps/s	89.2814 KOps/s	$\textbf{\color{#d91a1a}-5.93\%}$
test_items	42.1600μs	2.9033μs	344.4383 KOps/s	343.0405 KOps/s	$\color{#35bf28}+0.41\%$
test_items_nested	0.3602ms	0.3188ms	3.1366 KOps/s	3.1309 KOps/s	$\color{#35bf28}+0.18\%$
test_items_nested_locked	0.3696ms	0.3219ms	3.1064 KOps/s	3.1445 KOps/s	$\color{#d91a1a}-1.21\%$
test_items_nested_leaf	89.5710μs	57.9211μs	17.2649 KOps/s	17.2414 KOps/s	$\color{#35bf28}+0.14\%$
test_items_stack_nested	0.3546ms	0.3204ms	3.1212 KOps/s	3.1874 KOps/s	$\color{#d91a1a}-2.08\%$
test_items_stack_nested_leaf	84.5410μs	58.4352μs	17.1130 KOps/s	17.1026 KOps/s	$\color{#35bf28}+0.06\%$
test_items_stack_nested_locked	0.3596ms	0.3179ms	3.1457 KOps/s	3.1255 KOps/s	$\color{#35bf28}+0.65\%$
test_keys	31.3700μs	3.4735μs	287.8931 KOps/s	289.5593 KOps/s	$\color{#d91a1a}-0.58\%$
test_keys_nested	0.1058ms	69.4785μs	14.3929 KOps/s	14.0468 KOps/s	$\color{#35bf28}+2.46\%$
test_keys_nested_locked	0.8106ms	75.4253μs	13.2581 KOps/s	13.1919 KOps/s	$\color{#35bf28}+0.50\%$
test_keys_nested_leaf	0.1025ms	61.1121μs	16.3634 KOps/s	16.2501 KOps/s	$\color{#35bf28}+0.70\%$
test_keys_stack_nested	0.1006ms	69.9639μs	14.2931 KOps/s	14.0681 KOps/s	$\color{#35bf28}+1.60\%$
test_keys_stack_nested_leaf	0.1031ms	61.8127μs	16.1779 KOps/s	16.1537 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_stack_nested_locked	0.1061ms	75.3181μs	13.2770 KOps/s	13.2087 KOps/s	$\color{#35bf28}+0.52\%$
test_values	5.1485μs	0.8387μs	1.1923 MOps/s	1.1823 MOps/s	$\color{#35bf28}+0.85\%$
test_values_nested	61.1310μs	31.0829μs	32.1720 KOps/s	32.3436 KOps/s	$\color{#d91a1a}-0.53\%$
test_values_nested_locked	58.5810μs	32.5590μs	30.7135 KOps/s	30.9025 KOps/s	$\color{#d91a1a}-0.61\%$
test_values_nested_leaf	69.2210μs	33.4353μs	29.9085 KOps/s	30.1703 KOps/s	$\color{#d91a1a}-0.87\%$
test_values_stack_nested	58.6410μs	31.6671μs	31.5785 KOps/s	32.0455 KOps/s	$\color{#d91a1a}-1.46\%$
test_values_stack_nested_leaf	68.9810μs	33.8159μs	29.5719 KOps/s	29.8640 KOps/s	$\color{#d91a1a}-0.98\%$
test_values_stack_nested_locked	75.2110μs	33.2105μs	30.1110 KOps/s	30.5276 KOps/s	$\color{#d91a1a}-1.36\%$
test_membership	1.8271μs	0.5065μs	1.9742 MOps/s	1.9684 MOps/s	$\color{#35bf28}+0.30\%$
test_membership_nested	18.1000μs	1.8521μs	539.9167 KOps/s	545.9229 KOps/s	$\color{#d91a1a}-1.10\%$
test_membership_nested_leaf	18.8403μs	1.8274μs	547.2351 KOps/s	542.8733 KOps/s	$\color{#35bf28}+0.80\%$
test_membership_stacked_nested	39.9400μs	1.9363μs	516.4389 KOps/s	522.7868 KOps/s	$\color{#d91a1a}-1.21\%$
test_membership_stacked_nested_leaf	26.6400μs	1.9167μs	521.7245 KOps/s	525.0018 KOps/s	$\color{#d91a1a}-0.62\%$
test_membership_nested_last	35.9310μs	2.7776μs	360.0256 KOps/s	360.5685 KOps/s	$\color{#d91a1a}-0.15\%$
test_membership_nested_leaf_last	32.7010μs	2.8269μs	353.7424 KOps/s	356.1633 KOps/s	$\color{#d91a1a}-0.68\%$
test_membership_stacked_nested_last	39.5300μs	3.2083μs	311.6885 KOps/s	358.7104 KOps/s	$\textbf{\color{#d91a1a}-13.11\%}$
test_membership_stacked_nested_leaf_last	29.0200μs	3.2062μs	311.8939 KOps/s	356.9462 KOps/s	$\textbf{\color{#d91a1a}-12.62\%}$
test_nested_getleaf	40.4410μs	5.9878μs	167.0060 KOps/s	166.0993 KOps/s	$\color{#35bf28}+0.55\%$
test_nested_get	33.1110μs	5.6332μs	177.5179 KOps/s	175.7842 KOps/s	$\color{#35bf28}+0.99\%$
test_stacked_getleaf	48.7510μs	5.9930μs	166.8618 KOps/s	166.5489 KOps/s	$\color{#35bf28}+0.19\%$
test_stacked_get	34.5200μs	5.6542μs	176.8607 KOps/s	174.3134 KOps/s	$\color{#35bf28}+1.46\%$
test_nested_getitemleaf	45.8810μs	6.0472μs	165.3657 KOps/s	165.3409 KOps/s	$\color{#35bf28}+0.02\%$
test_nested_getitem	37.8910μs	5.7586μs	173.6532 KOps/s	172.6569 KOps/s	$\color{#35bf28}+0.58\%$
test_stacked_getitemleaf	40.5800μs	6.0490μs	165.3154 KOps/s	164.5790 KOps/s	$\color{#35bf28}+0.45\%$
test_stacked_getitem	34.5800μs	5.7159μs	174.9514 KOps/s	174.3544 KOps/s	$\color{#35bf28}+0.34\%$
test_lock_nested	0.6942ms	0.3537ms	2.8271 KOps/s	2.7388 KOps/s	$\color{#35bf28}+3.22\%$
test_lock_stack_nested	0.4053ms	0.3298ms	3.0317 KOps/s	3.0107 KOps/s	$\color{#35bf28}+0.70\%$
test_unlock_nested	0.6003ms	0.2988ms	3.3465 KOps/s	3.3119 KOps/s	$\color{#35bf28}+1.04\%$
test_unlock_stack_nested	0.3214ms	0.2716ms	3.6812 KOps/s	3.6693 KOps/s	$\color{#35bf28}+0.33\%$
test_flatten_speed	0.1159ms	72.4324μs	13.8060 KOps/s	13.8431 KOps/s	$\color{#d91a1a}-0.27\%$
test_unflatten_speed	0.3357ms	0.2855ms	3.5025 KOps/s	3.4733 KOps/s	$\color{#35bf28}+0.84\%$
test_common_ops	1.5719ms	0.5863ms	1.7057 KOps/s	1.7205 KOps/s	$\color{#d91a1a}-0.86\%$
test_creation	0.1721ms	1.4320μs	698.3303 KOps/s	700.6844 KOps/s	$\color{#d91a1a}-0.34\%$
test_creation_empty	25.1000μs	7.8069μs	128.0921 KOps/s	146.2737 KOps/s	$\textbf{\color{#d91a1a}-12.43\%}$
test_creation_nested_1	46.3910μs	9.3291μs	107.1913 KOps/s	118.3659 KOps/s	$\textbf{\color{#d91a1a}-9.44\%}$
test_creation_nested_2	40.0510μs	11.7887μs	84.8268 KOps/s	94.2886 KOps/s	$\textbf{\color{#d91a1a}-10.03\%}$
test_clone	29.0800μs	10.4886μs	95.3417 KOps/s	91.3641 KOps/s	$\color{#35bf28}+4.35\%$
test_getitem[int]	2.0147ms	10.3087μs	97.0056 KOps/s	95.5156 KOps/s	$\color{#35bf28}+1.56\%$
test_getitem[slice_int]	0.1111ms	19.8531μs	50.3698 KOps/s	49.4926 KOps/s	$\color{#35bf28}+1.77\%$
test_getitem[range]	0.1361ms	36.2956μs	27.5516 KOps/s	26.9642 KOps/s	$\color{#35bf28}+2.18\%$
test_getitem[tuple]	0.1107ms	17.4085μs	57.4433 KOps/s	56.6049 KOps/s	$\color{#35bf28}+1.48\%$
test_getitem[list]	0.3120ms	32.7275μs	30.5553 KOps/s	29.8367 KOps/s	$\color{#35bf28}+2.41\%$
test_setitem_dim[int]	38.6110μs	18.3329μs	54.5469 KOps/s	53.4084 KOps/s	$\color{#35bf28}+2.13\%$
test_setitem_dim[slice_int]	72.1110μs	37.1316μs	26.9313 KOps/s	26.8536 KOps/s	$\color{#35bf28}+0.29\%$
test_setitem_dim[range]	91.2120μs	51.4360μs	19.4417 KOps/s	19.3330 KOps/s	$\color{#35bf28}+0.56\%$
test_setitem_dim[tuple]	52.9200μs	30.8884μs	32.3746 KOps/s	32.1566 KOps/s	$\color{#35bf28}+0.68\%$
test_setitem	0.1280ms	16.3169μs	61.2862 KOps/s	67.7756 KOps/s	$\textbf{\color{#d91a1a}-9.57\%}$
test_set	48.7600μs	16.4059μs	60.9538 KOps/s	69.5604 KOps/s	$\textbf{\color{#d91a1a}-12.37\%}$
test_set_shared	1.5992ms	0.1447ms	6.9093 KOps/s	6.8540 KOps/s	$\color{#35bf28}+0.81\%$
test_update	0.2990ms	17.2020μs	58.1326 KOps/s	59.9778 KOps/s	$\color{#d91a1a}-3.08\%$
test_update_nested	56.3010μs	22.2383μs	44.9674 KOps/s	46.3161 KOps/s	$\color{#d91a1a}-2.91\%$
test_update__nested	1.0400ms	24.6429μs	40.5797 KOps/s	40.2962 KOps/s	$\color{#35bf28}+0.70\%$
test_set_nested	45.0910μs	15.6049μs	64.0822 KOps/s	64.5413 KOps/s	$\color{#d91a1a}-0.71\%$
test_set_nested_new	0.1260ms	17.7984μs	56.1849 KOps/s	56.8085 KOps/s	$\color{#d91a1a}-1.10\%$
test_select	66.1710μs	29.2191μs	34.2242 KOps/s	33.9127 KOps/s	$\color{#35bf28}+0.92\%$
test_select_nested	77.9210μs	41.1416μs	24.3063 KOps/s	24.1012 KOps/s	$\color{#35bf28}+0.85\%$
test_exclude_nested	0.1194ms	56.9188μs	17.5689 KOps/s	17.3931 KOps/s	$\color{#35bf28}+1.01\%$
test_empty[True]	0.3193ms	0.2536ms	3.9425 KOps/s	3.9622 KOps/s	$\color{#d91a1a}-0.50\%$
test_empty[False]	3.5961μs	0.7333μs	1.3636 MOps/s	1.3415 MOps/s	$\color{#35bf28}+1.65\%$
test_to	0.1009ms	55.4770μs	18.0255 KOps/s	18.5442 KOps/s	$\color{#d91a1a}-2.80\%$
test_to_nonblocking	85.3110μs	44.8579μs	22.2926 KOps/s	22.2296 KOps/s	$\color{#35bf28}+0.28\%$
test_unbind_speed	0.8426ms	0.2264ms	4.4178 KOps/s	4.3648 KOps/s	$\color{#35bf28}+1.22\%$
test_unbind_speed_stack0	0.2695ms	0.2262ms	4.4214 KOps/s	4.3240 KOps/s	$\color{#35bf28}+2.25\%$
test_unbind_speed_stack1	92.9921ms	0.6504ms	1.5375 KOps/s	1.5377 KOps/s	$\color{#d91a1a}-0.01\%$
test_split	93.2102ms	1.5451ms	647.2193 Ops/s	633.4498 Ops/s	$\color{#35bf28}+2.17\%$
test_chunk	95.8969ms	1.6778ms	596.0129 Ops/s	580.7165 Ops/s	$\color{#35bf28}+2.63\%$
test_consolidate[False-None]	3.3485ms	2.5349ms	394.4931 Ops/s	385.7857 Ops/s	$\color{#35bf28}+2.26\%$
test_consolidate[default-None]	1.7159ms	1.6561ms	603.8209 Ops/s	585.9597 Ops/s	$\color{#35bf28}+3.05\%$
test_consolidate[reduce-overhead-None]	1.7653ms	1.6973ms	589.1667 Ops/s	571.0596 Ops/s	$\color{#35bf28}+3.17\%$
test_consolidate_njt[False-None]	6.8153ms	6.4419ms	155.2338 Ops/s	155.5066 Ops/s	$\color{#d91a1a}-0.18\%$
test_to[False-False-None]	1.7575ms	1.6973ms	589.1671 Ops/s	582.8556 Ops/s	$\color{#35bf28}+1.08\%$
test_to[True-False-None]	1.4681ms	1.2660ms	789.8914 Ops/s	781.9872 Ops/s	$\color{#35bf28}+1.01\%$
test_to[within-False-None]	4.1013ms	3.8890ms	257.1358 Ops/s	253.0411 Ops/s	$\color{#35bf28}+1.62\%$
test_to[True-default-None]	5.3355ms	5.2154ms	191.7396 Ops/s	195.1570 Ops/s	$\color{#d91a1a}-1.75\%$
test_to_njt[False-False-None]	7.1467ms	6.9091ms	144.7360 Ops/s	146.8578 Ops/s	$\color{#d91a1a}-1.44\%$
test_to_njt[True-False-None]	5.5503ms	5.4014ms	185.1382 Ops/s	184.8089 Ops/s	$\color{#35bf28}+0.18\%$
test_to_njt[within-False-None]	12.0169ms	11.9209ms	83.8861 Ops/s	83.7402 Ops/s	$\color{#35bf28}+0.17\%$
test_creation[device0]	0.4548ms	78.1290μs	12.7993 KOps/s	12.8334 KOps/s	$\color{#d91a1a}-0.27\%$
test_creation_from_tensor	0.5293ms	81.2358μs	12.3098 KOps/s	12.0052 KOps/s	$\color{#35bf28}+2.54\%$
test_add_one[memmap_tensor0]	0.4115ms	6.7195μs	148.8197 KOps/s	134.6841 KOps/s	$\textbf{\color{#35bf28}+10.50\%}$
test_contiguous[memmap_tensor0]	1.8865μs	0.3966μs	2.5212 MOps/s	2.5347 MOps/s	$\color{#d91a1a}-0.53\%$
test_stack[memmap_tensor0]	38.7710μs	4.1964μs	238.3003 KOps/s	231.6431 KOps/s	$\color{#35bf28}+2.87\%$
test_memmaptd_index	1.4711ms	0.2440ms	4.0976 KOps/s	4.1010 KOps/s	$\color{#d91a1a}-0.08\%$
test_memmaptd_index_astensor	0.5529ms	0.2971ms	3.3661 KOps/s	3.3436 KOps/s	$\color{#35bf28}+0.67\%$
test_memmaptd_index_op	1.0016ms	0.5709ms	1.7517 KOps/s	1.7644 KOps/s	$\color{#d91a1a}-0.72\%$
test_serialize_model	0.1318s	0.1301s	7.6866 Ops/s	7.6987 Ops/s	$\color{#d91a1a}-0.16\%$
test_serialize_model_pickle	1.3475s	1.1894s	0.8407 Ops/s	0.8248 Ops/s	$\color{#35bf28}+1.94\%$
test_serialize_weights	0.1301s	0.1292s	7.7373 Ops/s	5.4996 Ops/s	$\textbf{\color{#35bf28}+40.69\%}$
test_serialize_weights_returnearly	0.3199s	53.1833ms	18.8029 Ops/s	23.8283 Ops/s	$\textbf{\color{#d91a1a}-21.09\%}$
test_serialize_weights_pickle	1.3501s	1.2158s	0.8225 Ops/s	0.8415 Ops/s	$\color{#d91a1a}-2.26\%$
test_reshape_pytree	88.1120μs	21.6402μs	46.2103 KOps/s	45.2286 KOps/s	$\color{#35bf28}+2.17\%$
test_reshape_td	63.0810μs	25.9125μs	38.5914 KOps/s	38.1531 KOps/s	$\color{#35bf28}+1.15\%$
test_view_pytree	53.7810μs	21.4786μs	46.5581 KOps/s	45.6605 KOps/s	$\color{#35bf28}+1.97\%$
test_view_td	65.3210μs	29.0365μs	34.4394 KOps/s	32.8476 KOps/s	$\color{#35bf28}+4.85\%$
test_unbind_pytree	62.7610μs	27.4081μs	36.4855 KOps/s	35.8185 KOps/s	$\color{#35bf28}+1.86\%$
test_unbind_td	0.7828ms	34.2963μs	29.1577 KOps/s	28.4646 KOps/s	$\color{#35bf28}+2.43\%$
test_split_pytree	67.3610μs	29.3974μs	34.0166 KOps/s	32.9309 KOps/s	$\color{#35bf28}+3.30\%$
test_split_td	1.0053ms	37.3163μs	26.7979 KOps/s	26.3235 KOps/s	$\color{#35bf28}+1.80\%$
test_add_pytree	64.4410μs	34.1080μs	29.3186 KOps/s	29.2517 KOps/s	$\color{#35bf28}+0.23\%$
test_add_td	80.2810μs	48.3608μs	20.6779 KOps/s	21.8929 KOps/s	$\textbf{\color{#d91a1a}-5.55\%}$
test_compile_add_one_nested[tensordict-compile]	0.1781ms	0.1197ms	8.3546 KOps/s	8.0545 KOps/s	$\color{#35bf28}+3.73\%$
test_compile_add_one_nested[tensordict-eager]	0.2173ms	0.1239ms	8.0735 KOps/s	8.0751 KOps/s	$\color{#d91a1a}-0.02\%$
test_compile_add_one_nested[pytree-compile]	0.1322ms	95.7126μs	10.4479 KOps/s	10.4313 KOps/s	$\color{#35bf28}+0.16\%$
test_compile_add_one_nested[pytree-eager]	0.2032ms	0.1477ms	6.7683 KOps/s	6.7029 KOps/s	$\color{#35bf28}+0.98\%$
test_compile_copy_nested[tensordict-compile]	73.6610μs	23.1859μs	43.1297 KOps/s	42.6520 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_copy_nested[tensordict-eager]	57.1310μs	25.8720μs	38.6519 KOps/s	37.0243 KOps/s	$\color{#35bf28}+4.40\%$
test_compile_copy_nested[pytree-compile]	0.2978ms	63.7169μs	15.6944 KOps/s	15.4844 KOps/s	$\color{#35bf28}+1.36\%$
test_compile_copy_nested[pytree-eager]	73.1710μs	48.8106μs	20.4874 KOps/s	20.3416 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_add_one_flat[tensordict-compile]	0.2844ms	0.1417ms	7.0579 KOps/s	6.9043 KOps/s	$\color{#35bf28}+2.23\%$
test_compile_add_one_flat[tensordict-eager]	0.3078ms	0.2049ms	4.8796 KOps/s	4.8681 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_add_one_flat[tensorclass-compile]	0.1447ms	96.7002μs	10.3412 KOps/s	10.1699 KOps/s	$\color{#35bf28}+1.68\%$
test_compile_add_one_flat[tensorclass-eager]	0.2077ms	50.7082μs	19.7207 KOps/s	19.9882 KOps/s	$\color{#d91a1a}-1.34\%$
test_compile_add_one_flat[pytree-compile]	0.1809ms	0.1362ms	7.3413 KOps/s	7.2836 KOps/s	$\color{#35bf28}+0.79\%$
test_compile_add_one_flat[pytree-eager]	0.5333ms	0.4760ms	2.1009 KOps/s	2.0464 KOps/s	$\color{#35bf28}+2.66\%$
test_compile_add_self_flat[tensordict-eager]	0.3729ms	0.2458ms	4.0681 KOps/s	4.0691 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_add_self_flat[tensordict-compile]	0.1885ms	0.1437ms	6.9600 KOps/s	6.9627 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_add_self_flat[tensorclass-eager]	0.1591ms	61.2704μs	16.3211 KOps/s	16.5020 KOps/s	$\color{#d91a1a}-1.10\%$
test_compile_add_self_flat[tensorclass-compile]	0.1433ms	0.1004ms	9.9579 KOps/s	10.1361 KOps/s	$\color{#d91a1a}-1.76\%$
test_compile_add_self_flat[pytree-eager]	0.5314ms	0.4174ms	2.3960 KOps/s	2.4313 KOps/s	$\color{#d91a1a}-1.45\%$
test_compile_add_self_flat[pytree-compile]	0.1816ms	0.1356ms	7.3761 KOps/s	7.3606 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_copy_flat[tensordict-compile]	54.1010μs	19.2122μs	52.0502 KOps/s	55.2847 KOps/s	$\textbf{\color{#d91a1a}-5.85\%}$
test_compile_copy_flat[tensordict-eager]	55.7910μs	27.1305μs	36.8589 KOps/s	37.5038 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_copy_flat[pytree-compile]	0.1186ms	69.1517μs	14.4610 KOps/s	14.3962 KOps/s	$\color{#35bf28}+0.45\%$
test_compile_copy_flat[pytree-eager]	0.1648ms	50.8911μs	19.6498 KOps/s	19.7563 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_assign_and_add[tensordict-compile]	1.5890ms	0.3842ms	2.6029 KOps/s	2.2001 KOps/s	$\textbf{\color{#35bf28}+18.31\%}$
test_compile_assign_and_add[tensordict-eager]	2.7559ms	2.6119ms	382.8579 Ops/s	366.9610 Ops/s	$\color{#35bf28}+4.33\%$
test_compile_assign_and_add[pytree-compile]	1.5710ms	0.4261ms	2.3469 KOps/s	2.2742 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_assign_and_add[pytree-eager]	2.8628ms	2.6978ms	370.6725 Ops/s	370.6924 Ops/s	$-0.01\%$
test_compile_indexing[tensor-tensordict-compile]	0.5343ms	0.1174ms	8.5173 KOps/s	8.5146 KOps/s	$\color{#35bf28}+0.03\%$
test_compile_indexing[tensor-tensordict-eager]	0.5493ms	81.7842μs	12.2273 KOps/s	12.1774 KOps/s	$\color{#35bf28}+0.41\%$
test_compile_indexing[tensor-tensorclass-compile]	0.6076ms	0.1112ms	8.9903 KOps/s	9.0498 KOps/s	$\color{#d91a1a}-0.66\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1779ms	67.9169μs	14.7239 KOps/s	14.5817 KOps/s	$\color{#35bf28}+0.97\%$
test_compile_indexing[tensor-pytree-compile]	0.4795ms	0.1111ms	9.0032 KOps/s	9.3063 KOps/s	$\color{#d91a1a}-3.26\%$
test_compile_indexing[tensor-pytree-eager]	0.4401ms	68.8342μs	14.5277 KOps/s	14.6975 KOps/s	$\color{#d91a1a}-1.16\%$
test_compile_indexing[slice-tensordict-compile]	0.1744ms	0.1016ms	9.8423 KOps/s	9.9105 KOps/s	$\color{#d91a1a}-0.69\%$
test_compile_indexing[slice-tensordict-eager]	0.1462ms	17.4061μs	57.4513 KOps/s	56.4634 KOps/s	$\color{#35bf28}+1.75\%$
test_compile_indexing[slice-tensorclass-compile]	0.1552ms	95.1919μs	10.5051 KOps/s	10.3471 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_indexing[slice-tensorclass-eager]	82.5010μs	15.5453μs	64.3280 KOps/s	63.1998 KOps/s	$\color{#35bf28}+1.79\%$
test_compile_indexing[slice-pytree-compile]	0.1511ms	95.7368μs	10.4453 KOps/s	10.2790 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_indexing[slice-pytree-eager]	0.1106ms	15.3635μs	65.0892 KOps/s	62.9987 KOps/s	$\color{#35bf28}+3.32\%$
test_compile_indexing[int-tensordict-compile]	0.4792ms	0.1045ms	9.5736 KOps/s	9.8898 KOps/s	$\color{#d91a1a}-3.20\%$
test_compile_indexing[int-tensordict-eager]	0.7388ms	16.1601μs	61.8808 KOps/s	56.8769 KOps/s	$\textbf{\color{#35bf28}+8.80\%}$
test_compile_indexing[int-tensorclass-compile]	0.4966ms	99.4884μs	10.0514 KOps/s	9.7386 KOps/s	$\color{#35bf28}+3.21\%$
test_compile_indexing[int-tensorclass-eager]	53.8710μs	15.4691μs	64.6449 KOps/s	61.8715 KOps/s	$\color{#35bf28}+4.48\%$
test_compile_indexing[int-pytree-compile]	0.4899ms	95.9856μs	10.4182 KOps/s	10.2805 KOps/s	$\color{#35bf28}+1.34\%$
test_compile_indexing[int-pytree-eager]	0.4226ms	15.5649μs	64.2470 KOps/s	62.5087 KOps/s	$\color{#35bf28}+2.78\%$
test_mod_add[eager]	0.4457ms	31.4614μs	31.7850 KOps/s	32.5670 KOps/s	$\color{#d91a1a}-2.40\%$
test_mod_add[compile]	0.2858ms	76.8852μs	13.0064 KOps/s	12.9223 KOps/s	$\color{#35bf28}+0.65\%$
test_mod_add[compile-overhead]	0.3081ms	0.1630ms	6.1366 KOps/s	5.5884 KOps/s	$\textbf{\color{#35bf28}+9.81\%}$
test_mod_wrap[eager]	0.3191ms	0.2396ms	4.1744 KOps/s	4.2116 KOps/s	$\color{#d91a1a}-0.88\%$
test_mod_wrap[compile]	1.5447ms	0.2771ms	3.6094 KOps/s	3.5898 KOps/s	$\color{#35bf28}+0.54\%$
test_mod_wrap[compile-overhead]	7.3163ms	3.8364ms	260.6582 Ops/s	260.1118 Ops/s	$\color{#35bf28}+0.21\%$
test_mod_wrap_and_backward[eager]	1.5157ms	1.3513ms	740.0294 Ops/s	693.2815 Ops/s	$\textbf{\color{#35bf28}+6.74\%}$
test_mod_wrap_and_backward[compile]	1.3417ms	1.2418ms	805.2741 Ops/s	735.1177 Ops/s	$\textbf{\color{#35bf28}+9.54\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3608ms	0.9083ms	1.1010 KOps/s	968.8107 Ops/s	$\textbf{\color{#35bf28}+13.64\%}$
test_seq_add[eager]	0.1318ms	96.3023μs	10.3840 KOps/s	10.5298 KOps/s	$\color{#d91a1a}-1.39\%$
test_seq_add[compile]	0.2707ms	87.8249μs	11.3863 KOps/s	11.7119 KOps/s	$\color{#d91a1a}-2.78\%$
test_seq_add[compile-overhead]	0.1834ms	0.1252ms	7.9896 KOps/s	7.9215 KOps/s	$\color{#35bf28}+0.86\%$
test_seq_wrap[eager]	0.4372ms	0.3774ms	2.6494 KOps/s	2.6794 KOps/s	$\color{#d91a1a}-1.12\%$
test_seq_wrap[compile]	1.5743ms	0.3086ms	3.2403 KOps/s	3.3846 KOps/s	$\color{#d91a1a}-4.26\%$
test_seq_wrap[compile-overhead]	0.2637ms	0.2179ms	4.5897 KOps/s	4.5373 KOps/s	$\color{#35bf28}+1.15\%$
test_func_call_runtime[False-eager]	0.8027ms	0.7371ms	1.3567 KOps/s	1.3132 KOps/s	$\color{#35bf28}+3.31\%$
test_func_call_runtime[False-compile]	0.7857ms	0.7339ms	1.3626 KOps/s	1.3571 KOps/s	$\color{#35bf28}+0.40\%$
test_func_call_runtime[False-compile-overhead]	0.4269ms	0.3519ms	2.8417 KOps/s	2.7779 KOps/s	$\color{#35bf28}+2.30\%$
test_func_call_runtime[True-eager]	1.0232ms	0.9029ms	1.1076 KOps/s	1.0887 KOps/s	$\color{#35bf28}+1.74\%$
test_func_call_runtime[True-compile]	0.8346ms	0.7570ms	1.3211 KOps/s	1.2459 KOps/s	$\textbf{\color{#35bf28}+6.03\%}$
test_func_call_runtime[True-compile-overhead]	0.4542ms	0.3774ms	2.6500 KOps/s	2.6350 KOps/s	$\color{#35bf28}+0.57\%$
test_func_call_cm_runtime[False-eager]	0.7877ms	0.7304ms	1.3690 KOps/s	1.3184 KOps/s	$\color{#35bf28}+3.84\%$
test_func_call_cm_runtime[False-compile]	0.7979ms	0.7381ms	1.3549 KOps/s	1.3450 KOps/s	$\color{#35bf28}+0.73\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4025ms	0.3552ms	2.8150 KOps/s	2.7881 KOps/s	$\color{#35bf28}+0.96\%$
test_func_call_cm_runtime[True-eager]	1.0797ms	0.9936ms	1.0064 KOps/s	999.8448 Ops/s	$\color{#35bf28}+0.66\%$
test_func_call_cm_runtime[True-compile]	0.8583ms	0.7882ms	1.2687 KOps/s	1.2750 KOps/s	$\color{#d91a1a}-0.49\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4575ms	0.4025ms	2.4845 KOps/s	2.4785 KOps/s	$\color{#35bf28}+0.24\%$
test_vmap_func_call_cm_runtime[eager]	2.5301ms	2.0819ms	480.3354 Ops/s	479.9845 Ops/s	$\color{#35bf28}+0.07\%$
test_vmap_func_call_cm_runtime[compile]	0.8420ms	0.7949ms	1.2580 KOps/s	1.2105 KOps/s	$\color{#35bf28}+3.92\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4496ms	0.4016ms	2.4897 KOps/s	2.4526 KOps/s	$\color{#35bf28}+1.51\%$
test_distributed	3.9857ms	0.3051ms	3.2775 KOps/s	8.7363 KOps/s	$\textbf{\color{#d91a1a}-62.48\%}$
test_tdmodule	37.5100μs	14.3275μs	69.7960 KOps/s	72.6056 KOps/s	$\color{#d91a1a}-3.87\%$
test_tdmodule_dispatch	70.4110μs	30.2246μs	33.0856 KOps/s	35.8349 KOps/s	$\textbf{\color{#d91a1a}-7.67\%}$
test_tdseq	26.2610μs	15.9144μs	62.8361 KOps/s	66.6519 KOps/s	$\textbf{\color{#d91a1a}-5.72\%}$
test_tdseq_dispatch	54.1410μs	33.6591μs	29.7096 KOps/s	32.3565 KOps/s	$\textbf{\color{#d91a1a}-8.18\%}$
test_instantiation_functorch	1.6404ms	1.5297ms	653.7190 Ops/s	651.2350 Ops/s	$\color{#35bf28}+0.38\%$
test_exec_functorch	0.1914ms	0.1437ms	6.9569 KOps/s	6.9906 KOps/s	$\color{#d91a1a}-0.48\%$
test_exec_functional_call	0.2009ms	0.1405ms	7.1188 KOps/s	7.2231 KOps/s	$\color{#d91a1a}-1.44\%$
test_exec_td_decorator	0.3565ms	0.1825ms	5.4792 KOps/s	5.4817 KOps/s	$\color{#d91a1a}-0.04\%$
test_vmap_mlp_speed_decorator[True-True]	0.7786ms	0.6712ms	1.4898 KOps/s	1.4927 KOps/s	$\color{#d91a1a}-0.19\%$
test_vmap_mlp_speed_decorator[True-False]	0.8484ms	0.6702ms	1.4922 KOps/s	1.4265 KOps/s	$\color{#35bf28}+4.60\%$
test_vmap_mlp_speed_decorator[False-True]	0.7168ms	0.5918ms	1.6898 KOps/s	1.6121 KOps/s	$\color{#35bf28}+4.82\%$
test_vmap_mlp_speed_decorator[False-False]	0.7436ms	0.6202ms	1.6123 KOps/s	1.6179 KOps/s	$\color{#d91a1a}-0.34\%$
test_vmap_transformer_speed_decorator[True-True]	19.2694ms	19.1592ms	52.1944 Ops/s	51.9709 Ops/s	$\color{#35bf28}+0.43\%$
test_vmap_transformer_speed_decorator[True-False]	20.2857ms	19.3913ms	51.5695 Ops/s	51.2639 Ops/s	$\color{#35bf28}+0.60\%$
test_vmap_transformer_speed_decorator[False-True]	20.0982ms	19.1353ms	52.2594 Ops/s	51.8065 Ops/s	$\color{#35bf28}+0.87\%$
test_vmap_transformer_speed_decorator[False-False]	19.1226ms	19.0654ms	52.4511 Ops/s	52.2188 Ops/s	$\color{#35bf28}+0.44\%$
test_to_module_speed[True]	1.0103ms	0.9187ms	1.0885 KOps/s	1.0560 KOps/s	$\color{#35bf28}+3.08\%$
test_to_module_speed[False]	1.3522ms	0.8992ms	1.1120 KOps/s	1.0874 KOps/s	$\color{#35bf28}+2.27\%$
test_tc_init	63.6110μs	35.4187μs	28.2337 KOps/s	29.3272 KOps/s	$\color{#d91a1a}-3.73\%$
test_tc_init_nested	0.1104ms	70.9907μs	14.0864 KOps/s	14.7273 KOps/s	$\color{#d91a1a}-4.35\%$
test_tc_first_layer_tensor	4.0387μs	0.6863μs	1.4570 MOps/s	1.4491 MOps/s	$\color{#35bf28}+0.54\%$
test_tc_first_layer_nontensor	0.1153ms	2.2582μs	442.8402 KOps/s	436.3284 KOps/s	$\color{#35bf28}+1.49\%$
test_tc_second_layer_tensor	11.7327μs	1.4131μs	707.6655 KOps/s	724.0750 KOps/s	$\color{#d91a1a}-2.27\%$
test_tc_second_layer_nontensor	28.1010μs	2.9908μs	334.3586 KOps/s	335.2029 KOps/s	$\color{#d91a1a}-0.25\%$
test_unbind	0.2310s	9.8196ms	101.8372 Ops/s	154.2980 Ops/s	$\textbf{\color{#d91a1a}-34.00\%}$
test_full_like	9.7146ms	9.2027ms	108.6633 Ops/s	106.9659 Ops/s	$\color{#35bf28}+1.59\%$
test_zeros_like	4.9146ms	4.3200ms	231.4819 Ops/s	114.6139 Ops/s	$\textbf{\color{#35bf28}+101.97\%}$
test_ones_like	4.4327ms	4.3089ms	232.0803 Ops/s	230.8994 Ops/s	$\color{#35bf28}+0.51\%$
test_clone	6.7347ms	6.4322ms	155.4666 Ops/s	153.9783 Ops/s	$\color{#35bf28}+0.97\%$
test_squeeze	57.3310μs	9.2054μs	108.6316 KOps/s	108.8402 KOps/s	$\color{#d91a1a}-0.19\%$
test_unsqueeze	0.1139ms	70.6168μs	14.1609 KOps/s	14.4943 KOps/s	$\color{#d91a1a}-2.30\%$
test_split	0.3776ms	0.1568ms	6.3770 KOps/s	6.4887 KOps/s	$\color{#d91a1a}-1.72\%$
test_permute	0.2168ms	0.1766ms	5.6621 KOps/s	5.8353 KOps/s	$\color{#d91a1a}-2.97\%$
test_stack	51.7003ms	51.1666ms	19.5440 Ops/s	19.6573 Ops/s	$\color{#d91a1a}-0.58\%$
test_cat	51.4663ms	50.9451ms	19.6290 Ops/s	23.4553 Ops/s	$\textbf{\color{#d91a1a}-16.31\%}$