[BugFix] Better check for TDModule

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Better check for TDModule #1248
• [Test] Test dataclasses.field as tensorclass default #1246

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}6$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	52.3290μs	20.3111μs	49.2342 KOps/s	48.8533 KOps/s	$\color{#35bf28}+0.78\%$
test_plain_set_stack_nested	41.5680μs	20.4273μs	48.9541 KOps/s	48.8191 KOps/s	$\color{#35bf28}+0.28\%$
test_plain_set_nested_inplace	54.0420μs	22.3077μs	44.8276 KOps/s	45.0328 KOps/s	$\color{#d91a1a}-0.46\%$
test_plain_set_stack_nested_inplace	58.7900μs	22.3878μs	44.6672 KOps/s	44.5891 KOps/s	$\color{#35bf28}+0.18\%$
test_items	27.5110μs	4.2053μs	237.7957 KOps/s	238.7460 KOps/s	$\color{#d91a1a}-0.40\%$
test_items_nested	0.5815ms	0.4097ms	2.4407 KOps/s	2.4403 KOps/s	$\color{#35bf28}+0.02\%$
test_items_nested_locked	0.7219ms	0.4110ms	2.4332 KOps/s	2.4261 KOps/s	$\color{#35bf28}+0.29\%$
test_items_nested_leaf	0.1454ms	76.8689μs	13.0092 KOps/s	12.6366 KOps/s	$\color{#35bf28}+2.95\%$
test_items_stack_nested	0.4943ms	0.4090ms	2.4448 KOps/s	2.4307 KOps/s	$\color{#35bf28}+0.58\%$
test_items_stack_nested_leaf	0.1372ms	77.4353μs	12.9140 KOps/s	12.6519 KOps/s	$\color{#35bf28}+2.07\%$
test_items_stack_nested_locked	0.4814ms	0.4089ms	2.4454 KOps/s	2.4285 KOps/s	$\color{#35bf28}+0.70\%$
test_keys	40.9660μs	3.4496μs	289.8846 KOps/s	281.1051 KOps/s	$\color{#35bf28}+3.12\%$
test_keys_nested	0.2263ms	0.1638ms	6.1033 KOps/s	6.0787 KOps/s	$\color{#35bf28}+0.41\%$
test_keys_nested_locked	1.7648ms	0.1712ms	5.8408 KOps/s	5.8393 KOps/s	$\color{#35bf28}+0.03\%$
test_keys_nested_leaf	0.2984ms	0.1428ms	7.0014 KOps/s	6.9367 KOps/s	$\color{#35bf28}+0.93\%$
test_keys_stack_nested	0.2553ms	0.1640ms	6.0984 KOps/s	6.0468 KOps/s	$\color{#35bf28}+0.85\%$
test_keys_stack_nested_leaf	0.2222ms	0.1435ms	6.9687 KOps/s	6.8369 KOps/s	$\color{#35bf28}+1.93\%$
test_keys_stack_nested_locked	0.2567ms	0.1708ms	5.8562 KOps/s	5.8052 KOps/s	$\color{#35bf28}+0.88\%$
test_values	5.4984μs	1.0422μs	959.5156 KOps/s	972.3919 KOps/s	$\color{#d91a1a}-1.32\%$
test_values_nested	0.1257ms	62.5908μs	15.9768 KOps/s	16.1369 KOps/s	$\color{#d91a1a}-0.99\%$
test_values_nested_locked	96.2110μs	62.3993μs	16.0258 KOps/s	16.1810 KOps/s	$\color{#d91a1a}-0.96\%$
test_values_nested_leaf	0.1518ms	71.3401μs	14.0174 KOps/s	13.5407 KOps/s	$\color{#35bf28}+3.52\%$
test_values_stack_nested	0.1089ms	61.6783μs	16.2132 KOps/s	15.9814 KOps/s	$\color{#35bf28}+1.45\%$
test_values_stack_nested_leaf	0.1226ms	71.1760μs	14.0497 KOps/s	14.0328 KOps/s	$\color{#35bf28}+0.12\%$
test_values_stack_nested_locked	0.2400ms	63.5881μs	15.7262 KOps/s	15.9740 KOps/s	$\color{#d91a1a}-1.55\%$
test_membership	26.6200μs	0.8698μs	1.1496 MOps/s	1.1639 MOps/s	$\color{#d91a1a}-1.22\%$
test_membership_nested	23.2130μs	2.9073μs	343.9562 KOps/s	346.5441 KOps/s	$\color{#d91a1a}-0.75\%$
test_membership_nested_leaf	36.7690μs	2.9338μs	340.8519 KOps/s	343.6232 KOps/s	$\color{#d91a1a}-0.81\%$
test_membership_stacked_nested	24.0060μs	2.8904μs	345.9676 KOps/s	341.5709 KOps/s	$\color{#35bf28}+1.29\%$
test_membership_stacked_nested_leaf	16.9620μs	2.8974μs	345.1383 KOps/s	346.6222 KOps/s	$\color{#d91a1a}-0.43\%$
test_membership_nested_last	25.0270μs	4.4182μs	226.3351 KOps/s	227.3534 KOps/s	$\color{#d91a1a}-0.45\%$
test_membership_nested_leaf_last	22.3820μs	4.4447μs	224.9872 KOps/s	226.3181 KOps/s	$\color{#d91a1a}-0.59\%$
test_membership_stacked_nested_last	30.5680μs	4.3740μs	228.6263 KOps/s	226.3402 KOps/s	$\color{#35bf28}+1.01\%$
test_membership_stacked_nested_leaf_last	22.8230μs	4.3508μs	229.8451 KOps/s	226.1675 KOps/s	$\color{#35bf28}+1.63\%$
test_nested_getleaf	40.4860μs	10.4861μs	95.3641 KOps/s	93.6499 KOps/s	$\color{#35bf28}+1.83\%$
test_nested_get	34.5960μs	10.0095μs	99.9055 KOps/s	99.2220 KOps/s	$\color{#35bf28}+0.69\%$
test_stacked_getleaf	28.3940μs	10.6652μs	93.7631 KOps/s	94.3540 KOps/s	$\color{#d91a1a}-0.63\%$
test_stacked_get	35.5070μs	10.0456μs	99.5457 KOps/s	99.1040 KOps/s	$\color{#35bf28}+0.45\%$
test_nested_getitemleaf	38.8530μs	11.3874μs	87.8167 KOps/s	88.5485 KOps/s	$\color{#d91a1a}-0.83\%$
test_nested_getitem	30.6570μs	10.8606μs	92.0756 KOps/s	93.0829 KOps/s	$\color{#d91a1a}-1.08\%$
test_stacked_getitemleaf	34.8660μs	11.3986μs	87.7300 KOps/s	89.5973 KOps/s	$\color{#d91a1a}-2.08\%$
test_stacked_getitem	33.3330μs	10.7206μs	93.2780 KOps/s	93.0942 KOps/s	$\color{#35bf28}+0.20\%$
test_lock_nested	0.6160ms	0.4158ms	2.4052 KOps/s	2.4410 KOps/s	$\color{#d91a1a}-1.47\%$
test_lock_stack_nested	0.5492ms	0.4236ms	2.3608 KOps/s	2.2639 KOps/s	$\color{#35bf28}+4.28\%$
test_unlock_nested	0.9532ms	0.3420ms	2.9237 KOps/s	2.9448 KOps/s	$\color{#d91a1a}-0.72\%$
test_unlock_stack_nested	0.4782ms	0.3444ms	2.9038 KOps/s	2.7994 KOps/s	$\color{#35bf28}+3.73\%$
test_flatten_speed	0.1972ms	0.1007ms	9.9272 KOps/s	9.7934 KOps/s	$\color{#35bf28}+1.37\%$
test_unflatten_speed	0.7731ms	0.5269ms	1.8977 KOps/s	1.8997 KOps/s	$\color{#d91a1a}-0.10\%$
test_common_ops	0.9291ms	0.7894ms	1.2667 KOps/s	1.2411 KOps/s	$\color{#35bf28}+2.07\%$
test_creation	32.2410μs	2.5433μs	393.1922 KOps/s	397.8078 KOps/s	$\color{#d91a1a}-1.16\%$
test_creation_empty	40.9370μs	11.3222μs	88.3223 KOps/s	87.7560 KOps/s	$\color{#35bf28}+0.65\%$
test_creation_nested_1	0.1502ms	14.4026μs	69.4320 KOps/s	70.1836 KOps/s	$\color{#d91a1a}-1.07\%$
test_creation_nested_2	46.8980μs	18.7086μs	53.4515 KOps/s	53.1064 KOps/s	$\color{#35bf28}+0.65\%$
test_clone	52.7490μs	13.9154μs	71.8630 KOps/s	74.3327 KOps/s	$\color{#d91a1a}-3.32\%$
test_getitem[int]	0.8527ms	12.8600μs	77.7605 KOps/s	75.2293 KOps/s	$\color{#35bf28}+3.36\%$
test_getitem[slice_int]	0.1296ms	24.5209μs	40.7815 KOps/s	39.5017 KOps/s	$\color{#35bf28}+3.24\%$
test_getitem[range]	0.1597ms	48.8126μs	20.4865 KOps/s	19.8825 KOps/s	$\color{#35bf28}+3.04\%$
test_getitem[tuple]	0.1413ms	20.0762μs	49.8103 KOps/s	48.9942 KOps/s	$\color{#35bf28}+1.67\%$
test_getitem[list]	0.1634ms	44.1293μs	22.6607 KOps/s	21.9217 KOps/s	$\color{#35bf28}+3.37\%$
test_setitem_dim[int]	50.8360μs	24.9406μs	40.0953 KOps/s	39.6437 KOps/s	$\color{#35bf28}+1.14\%$
test_setitem_dim[slice_int]	75.1710μs	51.0542μs	19.5870 KOps/s	19.0705 KOps/s	$\color{#35bf28}+2.71\%$
test_setitem_dim[range]	0.1228ms	75.9069μs	13.1740 KOps/s	12.9809 KOps/s	$\color{#35bf28}+1.49\%$
test_setitem_dim[tuple]	76.5330μs	40.3213μs	24.8008 KOps/s	24.4948 KOps/s	$\color{#35bf28}+1.25\%$
test_setitem	78.5370μs	20.6993μs	48.3109 KOps/s	50.3734 KOps/s	$\color{#d91a1a}-4.09\%$
test_set	70.8930μs	19.9893μs	50.0267 KOps/s	51.5494 KOps/s	$\color{#d91a1a}-2.95\%$
test_set_shared	4.3900ms	0.1842ms	5.4285 KOps/s	5.3614 KOps/s	$\color{#35bf28}+1.25\%$
test_update	0.1207ms	25.6543μs	38.9798 KOps/s	40.4934 KOps/s	$\color{#d91a1a}-3.74\%$
test_update_nested	0.4130ms	40.3145μs	24.8050 KOps/s	25.0297 KOps/s	$\color{#d91a1a}-0.90\%$
test_update__nested	0.1143ms	34.1678μs	29.2673 KOps/s	29.9516 KOps/s	$\color{#d91a1a}-2.28\%$
test_set_nested	74.0400μs	22.2028μs	45.0394 KOps/s	46.7948 KOps/s	$\color{#d91a1a}-3.75\%$
test_set_nested_new	75.4520μs	26.6751μs	37.4882 KOps/s	38.1038 KOps/s	$\color{#d91a1a}-1.62\%$
test_select	0.1347ms	42.8962μs	23.3121 KOps/s	23.5897 KOps/s	$\color{#d91a1a}-1.18\%$
test_select_nested	0.1241ms	63.9127μs	15.6463 KOps/s	15.5445 KOps/s	$\color{#35bf28}+0.66\%$
test_exclude_nested	0.1517ms	83.1895μs	12.0208 KOps/s	12.1694 KOps/s	$\color{#d91a1a}-1.22\%$
test_empty[True]	0.6107ms	0.4121ms	2.4268 KOps/s	2.4280 KOps/s	$\color{#d91a1a}-0.05\%$
test_empty[False]	9.3525μs	1.4661μs	682.0940 KOps/s	710.3566 KOps/s	$\color{#d91a1a}-3.98\%$
test_unbind_speed	0.3472ms	0.2711ms	3.6889 KOps/s	3.7123 KOps/s	$\color{#d91a1a}-0.63\%$
test_unbind_speed_stack0	0.4587ms	0.2700ms	3.7038 KOps/s	3.7707 KOps/s	$\color{#d91a1a}-1.77\%$
test_unbind_speed_stack1	0.1030s	0.7359ms	1.3589 KOps/s	1.2412 KOps/s	$\textbf{\color{#35bf28}+9.48\%}$
test_split	0.1021s	1.7626ms	567.3568 Ops/s	624.2375 Ops/s	$\textbf{\color{#d91a1a}-9.11\%}$
test_chunk	0.1084s	1.7796ms	561.9342 Ops/s	545.7290 Ops/s	$\color{#35bf28}+2.97\%$
test_consolidate_njt[False-None]	9.1405ms	8.0254ms	124.6039 Ops/s	110.3505 Ops/s	$\textbf{\color{#35bf28}+12.92\%}$
test_creation[device0]	0.2210ms	92.7132μs	10.7859 KOps/s	10.7493 KOps/s	$\color{#35bf28}+0.34\%$
test_creation_from_tensor	4.2370ms	95.6709μs	10.4525 KOps/s	10.5760 KOps/s	$\color{#d91a1a}-1.17\%$
test_add_one[memmap_tensor0]	71.6840μs	5.2358μs	190.9929 KOps/s	197.4902 KOps/s	$\color{#d91a1a}-3.29\%$
test_contiguous[memmap_tensor0]	19.4760μs	0.5154μs	1.9402 MOps/s	1.9605 MOps/s	$\color{#d91a1a}-1.04\%$
test_stack[memmap_tensor0]	20.6480μs	3.4711μs	288.0923 KOps/s	296.6551 KOps/s	$\color{#d91a1a}-2.89\%$
test_memmaptd_index	1.3446ms	0.2259ms	4.4277 KOps/s	4.2980 KOps/s	$\color{#35bf28}+3.02\%$
test_memmaptd_index_astensor	0.4740ms	0.3121ms	3.2037 KOps/s	3.1222 KOps/s	$\color{#35bf28}+2.61\%$
test_memmaptd_index_op	1.1401ms	0.5742ms	1.7415 KOps/s	1.6675 KOps/s	$\color{#35bf28}+4.44\%$
test_serialize_model	0.2169s	0.1313s	7.6167 Ops/s	8.7069 Ops/s	$\textbf{\color{#d91a1a}-12.52\%}$
test_serialize_model_pickle	0.4408s	0.3995s	2.5031 Ops/s	2.4909 Ops/s	$\color{#35bf28}+0.49\%$
test_serialize_weights	0.1195s	0.1132s	8.8333 Ops/s	8.7334 Ops/s	$\color{#35bf28}+1.14\%$
test_serialize_weights_returnearly	0.2480s	0.1699s	5.8870 Ops/s	6.4686 Ops/s	$\textbf{\color{#d91a1a}-8.99\%}$
test_serialize_weights_pickle	0.4793s	0.3867s	2.5859 Ops/s	2.5314 Ops/s	$\color{#35bf28}+2.15\%$
test_serialize_weights_filesystem	0.2483s	0.1599s	6.2545 Ops/s	6.2426 Ops/s	$\color{#35bf28}+0.19\%$
test_serialize_model_filesystem	0.1549s	0.1496s	6.6829 Ops/s	6.5685 Ops/s	$\color{#35bf28}+1.74\%$
test_reshape_pytree	79.5000μs	25.7515μs	38.8327 KOps/s	37.8460 KOps/s	$\color{#35bf28}+2.61\%$
test_reshape_td	68.4090μs	33.3171μs	30.0146 KOps/s	28.3298 KOps/s	$\textbf{\color{#35bf28}+5.95\%}$
test_view_pytree	55.0530μs	25.4174μs	39.3431 KOps/s	37.9140 KOps/s	$\color{#35bf28}+3.77\%$
test_view_td	0.1199ms	39.3729μs	25.3982 KOps/s	23.2585 KOps/s	$\textbf{\color{#35bf28}+9.20\%}$
test_unbind_pytree	65.7740μs	28.9824μs	34.5036 KOps/s	33.8165 KOps/s	$\color{#35bf28}+2.03\%$
test_unbind_td	0.3459ms	40.2621μs	24.8373 KOps/s	24.8438 KOps/s	$\color{#d91a1a}-0.03\%$
test_split_pytree	59.7120μs	28.3652μs	35.2545 KOps/s	34.0023 KOps/s	$\color{#35bf28}+3.68\%$
test_split_td	0.2008ms	45.7883μs	21.8396 KOps/s	21.4267 KOps/s	$\color{#35bf28}+1.93\%$
test_add_pytree	0.1274ms	34.9887μs	28.5806 KOps/s	27.3858 KOps/s	$\color{#35bf28}+4.36\%$
test_add_td	0.1080ms	54.8541μs	18.2302 KOps/s	17.5835 KOps/s	$\color{#35bf28}+3.68\%$
test_compile_add_one_nested[tensordict-compile]	0.1558ms	68.8770μs	14.5186 KOps/s	15.0346 KOps/s	$\color{#d91a1a}-3.43\%$
test_compile_add_one_nested[tensordict-eager]	0.3262ms	0.1733ms	5.7715 KOps/s	5.7760 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_add_one_nested[pytree-compile]	0.1077ms	46.6753μs	21.4246 KOps/s	22.2858 KOps/s	$\color{#d91a1a}-3.86\%$
test_compile_add_one_nested[pytree-eager]	0.2285ms	0.1192ms	8.3890 KOps/s	8.3465 KOps/s	$\color{#35bf28}+0.51\%$
test_compile_copy_nested[tensordict-compile]	63.5800μs	28.2833μs	35.3566 KOps/s	35.9120 KOps/s	$\color{#d91a1a}-1.55\%$
test_compile_copy_nested[tensordict-eager]	0.1266ms	58.2295μs	17.1734 KOps/s	17.2652 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_copy_nested[pytree-compile]	0.1597ms	78.0164μs	12.8178 KOps/s	12.5295 KOps/s	$\color{#35bf28}+2.30\%$
test_compile_copy_nested[pytree-eager]	0.1837ms	66.0144μs	15.1482 KOps/s	14.8407 KOps/s	$\color{#35bf28}+2.07\%$
test_compile_add_one_flat[tensordict-compile]	0.2644ms	0.1084ms	9.2232 KOps/s	9.4660 KOps/s	$\color{#d91a1a}-2.57\%$
test_compile_add_one_flat[tensordict-eager]	0.3540ms	0.2177ms	4.5933 KOps/s	4.6549 KOps/s	$\color{#d91a1a}-1.32\%$
test_compile_add_one_flat[tensorclass-compile]	0.1009ms	48.1004μs	20.7899 KOps/s	21.5920 KOps/s	$\color{#d91a1a}-3.71\%$
test_compile_add_one_flat[tensorclass-eager]	0.1948ms	67.2410μs	14.8719 KOps/s	14.9099 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_one_flat[pytree-compile]	0.1845ms	0.1015ms	9.8517 KOps/s	9.7069 KOps/s	$\color{#35bf28}+1.49\%$
test_compile_add_one_flat[pytree-eager]	0.3262ms	0.2040ms	4.9021 KOps/s	4.9217 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_add_self_flat[tensordict-eager]	0.5096ms	0.2354ms	4.2486 KOps/s	4.2823 KOps/s	$\color{#d91a1a}-0.79\%$
test_compile_add_self_flat[tensordict-compile]	0.1904ms	0.1079ms	9.2640 KOps/s	9.4378 KOps/s	$\color{#d91a1a}-1.84\%$
test_compile_add_self_flat[tensorclass-eager]	0.6815ms	65.0018μs	15.3842 KOps/s	15.9142 KOps/s	$\color{#d91a1a}-3.33\%$
test_compile_add_self_flat[tensorclass-compile]	0.1115ms	49.5502μs	20.1816 KOps/s	20.6279 KOps/s	$\color{#d91a1a}-2.16\%$
test_compile_add_self_flat[pytree-eager]	0.3563ms	0.1582ms	6.3214 KOps/s	6.3577 KOps/s	$\color{#d91a1a}-0.57\%$
test_compile_add_self_flat[pytree-compile]	0.2569ms	0.1025ms	9.7553 KOps/s	9.9359 KOps/s	$\color{#d91a1a}-1.82\%$
test_compile_copy_flat[tensordict-compile]	55.7140μs	21.6685μs	46.1499 KOps/s	47.6143 KOps/s	$\color{#d91a1a}-3.08\%$
test_compile_copy_flat[tensordict-eager]	0.1348ms	66.9235μs	14.9424 KOps/s	15.0730 KOps/s	$\color{#d91a1a}-0.87\%$
test_compile_copy_flat[pytree-compile]	0.1674ms	80.9027μs	12.3605 KOps/s	12.3052 KOps/s	$\color{#35bf28}+0.45\%$
test_compile_copy_flat[pytree-eager]	0.1297ms	67.5564μs	14.8024 KOps/s	14.6371 KOps/s	$\color{#35bf28}+1.13\%$
test_compile_assign_and_add[tensordict-compile]	0.3195ms	0.2181ms	4.5849 KOps/s	4.6285 KOps/s	$\color{#d91a1a}-0.94\%$
test_compile_assign_and_add[tensordict-eager]	1.7921ms	1.3739ms	727.8399 Ops/s	737.6546 Ops/s	$\color{#d91a1a}-1.33\%$
test_compile_assign_and_add[pytree-compile]	0.4236ms	0.2137ms	4.6794 KOps/s	4.6864 KOps/s	$\color{#d91a1a}-0.15\%$
test_compile_assign_and_add[pytree-eager]	1.4612ms	0.8368ms	1.1950 KOps/s	1.2043 KOps/s	$\color{#d91a1a}-0.77\%$
test_compile_assign_and_add_stack[compile]	0.6143ms	0.4700ms	2.1275 KOps/s	2.2258 KOps/s	$\color{#d91a1a}-4.41\%$
test_compile_assign_and_add_stack[eager]	4.0377ms	2.6225ms	381.3139 Ops/s	367.2013 Ops/s	$\color{#35bf28}+3.84\%$
test_compile_indexing[tensor-tensordict-compile]	89.1270μs	39.5485μs	25.2854 KOps/s	26.1493 KOps/s	$\color{#d91a1a}-3.30\%$
test_compile_indexing[tensor-tensordict-eager]	0.6156ms	32.4956μs	30.7734 KOps/s	30.9850 KOps/s	$\color{#d91a1a}-0.68\%$
test_compile_indexing[tensor-tensorclass-compile]	78.9490μs	30.9280μs	32.3332 KOps/s	33.3140 KOps/s	$\color{#d91a1a}-2.94\%$
test_compile_indexing[tensor-tensorclass-eager]	74.2490μs	22.8054μs	43.8492 KOps/s	42.4077 KOps/s	$\color{#35bf28}+3.40\%$
test_compile_indexing[tensor-pytree-compile]	76.2030μs	32.0305μs	31.2202 KOps/s	32.7956 KOps/s	$\color{#d91a1a}-4.80\%$
test_compile_indexing[tensor-pytree-eager]	83.8000μs	22.2434μs	44.9571 KOps/s	42.1066 KOps/s	$\textbf{\color{#35bf28}+6.77\%}$
test_compile_indexing[slice-tensordict-compile]	0.3768ms	54.6970μs	18.2825 KOps/s	19.3308 KOps/s	$\textbf{\color{#d91a1a}-5.42\%}$
test_compile_indexing[slice-tensordict-eager]	0.3784ms	20.2070μs	49.4877 KOps/s	48.0563 KOps/s	$\color{#35bf28}+2.98\%$
test_compile_indexing[slice-tensorclass-compile]	0.1040ms	46.3936μs	21.5547 KOps/s	22.1224 KOps/s	$\color{#d91a1a}-2.57\%$
test_compile_indexing[slice-tensorclass-eager]	63.0090μs	18.2741μs	54.7223 KOps/s	53.0918 KOps/s	$\color{#35bf28}+3.07\%$
test_compile_indexing[slice-pytree-compile]	0.1111ms	46.7204μs	21.4039 KOps/s	21.8382 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_indexing[slice-pytree-eager]	63.7500μs	18.2026μs	54.9373 KOps/s	53.1051 KOps/s	$\color{#35bf28}+3.45\%$
test_compile_indexing[int-tensordict-compile]	0.1690ms	56.9489μs	17.5596 KOps/s	19.1505 KOps/s	$\textbf{\color{#d91a1a}-8.31\%}$
test_compile_indexing[int-tensordict-eager]	1.0518ms	20.1064μs	49.7354 KOps/s	50.3732 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_indexing[int-tensorclass-compile]	0.1489ms	47.7848μs	20.9272 KOps/s	21.5614 KOps/s	$\color{#d91a1a}-2.94\%$
test_compile_indexing[int-tensorclass-eager]	0.6183ms	18.3183μs	54.5903 KOps/s	53.4115 KOps/s	$\color{#35bf28}+2.21\%$
test_compile_indexing[int-pytree-compile]	0.1112ms	46.8999μs	21.3220 KOps/s	21.7827 KOps/s	$\color{#d91a1a}-2.11\%$
test_compile_indexing[int-pytree-eager]	63.7700μs	18.4013μs	54.3439 KOps/s	53.5394 KOps/s	$\color{#35bf28}+1.50\%$
test_mod_add[eager]	87.6640μs	35.0573μs	28.5247 KOps/s	27.9442 KOps/s	$\color{#35bf28}+2.08\%$
test_mod_add[compile]	0.1319ms	66.4087μs	15.0583 KOps/s	15.2168 KOps/s	$\color{#d91a1a}-1.04\%$
test_mod_add[compile-overhead]	0.1342ms	64.9696μs	15.3918 KOps/s	15.1240 KOps/s	$\color{#35bf28}+1.77\%$
test_mod_wrap[eager]	0.3681ms	0.2274ms	4.3984 KOps/s	4.5333 KOps/s	$\color{#d91a1a}-2.97\%$
test_mod_wrap[compile]	2.0651ms	0.2293ms	4.3607 KOps/s	4.4225 KOps/s	$\color{#d91a1a}-1.40\%$
test_mod_wrap[compile-overhead]	0.3354ms	0.2224ms	4.4964 KOps/s	4.2919 KOps/s	$\color{#35bf28}+4.77\%$
test_mod_wrap_and_backward[eager]	12.4193ms	11.1348ms	89.8089 Ops/s	90.9877 Ops/s	$\color{#d91a1a}-1.30\%$
test_mod_wrap_and_backward[compile]	13.0898ms	10.9797ms	91.0769 Ops/s	89.2385 Ops/s	$\color{#35bf28}+2.06\%$
test_mod_wrap_and_backward[compile-overhead]	12.4797ms	10.9011ms	91.7336 Ops/s	92.3891 Ops/s	$\color{#d91a1a}-0.71\%$
test_seq_add[eager]	0.2510ms	0.1169ms	8.5528 KOps/s	8.5955 KOps/s	$\color{#d91a1a}-0.50\%$
test_seq_add[compile]	0.1400ms	75.9588μs	13.1650 KOps/s	13.1764 KOps/s	$\color{#d91a1a}-0.09\%$
test_seq_add[compile-overhead]	0.1772ms	74.7958μs	13.3697 KOps/s	13.3564 KOps/s	$\color{#35bf28}+0.10\%$
test_seq_wrap[eager]	0.5419ms	0.4412ms	2.2664 KOps/s	2.2635 KOps/s	$\color{#35bf28}+0.13\%$
test_seq_wrap[compile]	0.7977ms	0.2457ms	4.0707 KOps/s	4.0482 KOps/s	$\color{#35bf28}+0.56\%$
test_seq_wrap[compile-overhead]	0.3681ms	0.2410ms	4.1489 KOps/s	4.1482 KOps/s	$\color{#35bf28}+0.02\%$
test_func_call_runtime[False-eager]	0.9321ms	0.5428ms	1.8422 KOps/s	1.9049 KOps/s	$\color{#d91a1a}-3.29\%$
test_func_call_runtime[False-compile]	1.0179ms	0.4483ms	2.2304 KOps/s	2.2596 KOps/s	$\color{#d91a1a}-1.29\%$
test_func_call_runtime[False-compile-overhead]	0.9108ms	0.4467ms	2.2385 KOps/s	2.2547 KOps/s	$\color{#d91a1a}-0.72\%$
test_func_call_runtime[True-eager]	1.1049ms	0.7684ms	1.3013 KOps/s	1.3421 KOps/s	$\color{#d91a1a}-3.04\%$
test_func_call_runtime[True-compile]	0.6612ms	0.4688ms	2.1333 KOps/s	2.1379 KOps/s	$\color{#d91a1a}-0.22\%$
test_func_call_runtime[True-compile-overhead]	0.6622ms	0.4682ms	2.1358 KOps/s	2.1400 KOps/s	$\color{#d91a1a}-0.20\%$
test_func_call_cm_runtime[False-eager]	0.6545ms	0.5456ms	1.8327 KOps/s	1.9128 KOps/s	$\color{#d91a1a}-4.19\%$
test_func_call_cm_runtime[False-compile]	0.7746ms	0.4488ms	2.2282 KOps/s	2.2431 KOps/s	$\color{#d91a1a}-0.67\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7715ms	0.4479ms	2.2328 KOps/s	2.2593 KOps/s	$\color{#d91a1a}-1.17\%$
test_func_call_cm_runtime[True-eager]	2.7454ms	0.9207ms	1.0861 KOps/s	1.1301 KOps/s	$\color{#d91a1a}-3.89\%$
test_func_call_cm_runtime[True-compile]	0.9387ms	0.8026ms	1.2460 KOps/s	1.2591 KOps/s	$\color{#d91a1a}-1.05\%$
test_func_call_cm_runtime[True-compile-overhead]	1.3289ms	0.8197ms	1.2200 KOps/s	1.2600 KOps/s	$\color{#d91a1a}-3.18\%$
test_vmap_func_call_cm_runtime[eager]	2.6924ms	1.9431ms	514.6297 Ops/s	521.6021 Ops/s	$\color{#d91a1a}-1.34\%$
test_vmap_func_call_cm_runtime[compile]	1.0685ms	0.5394ms	1.8538 KOps/s	1.8695 KOps/s	$\color{#d91a1a}-0.84\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6984ms	0.5387ms	1.8562 KOps/s	1.8764 KOps/s	$\color{#d91a1a}-1.08\%$
test_distributed	0.2941ms	0.1273ms	7.8547 KOps/s	7.9221 KOps/s	$\color{#d91a1a}-0.85\%$
test_tdmodule	63.8400μs	26.2651μs	38.0733 KOps/s	36.7497 KOps/s	$\color{#35bf28}+3.60\%$
test_tdmodule_dispatch	86.4320μs	48.7629μs	20.5074 KOps/s	20.3732 KOps/s	$\color{#35bf28}+0.66\%$
test_tdseq	68.5180μs	28.0967μs	35.5913 KOps/s	34.7015 KOps/s	$\color{#35bf28}+2.56\%$
test_tdseq_dispatch	86.0720μs	53.0820μs	18.8388 KOps/s	18.6419 KOps/s	$\color{#35bf28}+1.06\%$
test_instantiation_functorch	1.7565ms	1.5518ms	644.4286 Ops/s	654.3876 Ops/s	$\color{#d91a1a}-1.52\%$
test_exec_functorch	0.2569ms	0.1771ms	5.6463 KOps/s	5.6465 KOps/s	$-0.00\%$
test_exec_functional_call	0.3329ms	0.1751ms	5.7106 KOps/s	5.8753 KOps/s	$\color{#d91a1a}-2.80\%$
test_exec_td_decorator	0.4935ms	0.2366ms	4.2272 KOps/s	4.3207 KOps/s	$\color{#d91a1a}-2.16\%$
test_vmap_mlp_speed_decorator[True-True]	0.9247ms	0.6712ms	1.4899 KOps/s	1.5294 KOps/s	$\color{#d91a1a}-2.58\%$
test_vmap_mlp_speed_decorator[True-False]	0.9648ms	0.6848ms	1.4602 KOps/s	1.5169 KOps/s	$\color{#d91a1a}-3.74\%$
test_vmap_mlp_speed_decorator[False-True]	0.8414ms	0.5458ms	1.8322 KOps/s	1.8757 KOps/s	$\color{#d91a1a}-2.32\%$
test_vmap_mlp_speed_decorator[False-False]	0.8022ms	0.5440ms	1.8383 KOps/s	1.8776 KOps/s	$\color{#d91a1a}-2.09\%$
test_to_module_speed[True]	1.8474ms	1.3536ms	738.7540 Ops/s	755.3086 Ops/s	$\color{#d91a1a}-2.19\%$
test_to_module_speed[False]	1.7910ms	1.3057ms	765.8803 Ops/s	769.0847 Ops/s	$\color{#d91a1a}-0.42\%$
test_tc_init	85.3400μs	46.0708μs	21.7057 KOps/s	22.1106 KOps/s	$\color{#d91a1a}-1.83\%$
test_tc_init_nested	0.1455ms	90.2342μs	11.0823 KOps/s	10.8933 KOps/s	$\color{#35bf28}+1.73\%$
test_tc_first_layer_tensor	16.2500μs	1.5459μs	646.8518 KOps/s	654.8060 KOps/s	$\color{#d91a1a}-1.21\%$
test_tc_first_layer_nontensor	21.8910μs	4.6621μs	214.4938 KOps/s	205.9891 KOps/s	$\color{#35bf28}+4.13\%$
test_tc_second_layer_tensor	41.8480μs	2.8705μs	348.3661 KOps/s	351.5389 KOps/s	$\color{#d91a1a}-0.90\%$
test_tc_second_layer_nontensor	44.3030μs	5.9937μs	166.8412 KOps/s	159.6052 KOps/s	$\color{#35bf28}+4.53\%$
test_unbind	0.2376s	13.4895ms	74.1316 Ops/s	81.3182 Ops/s	$\textbf{\color{#d91a1a}-8.84\%}$
test_full_like	10.2405ms	7.5458ms	132.5246 Ops/s	129.6398 Ops/s	$\color{#35bf28}+2.23\%$
test_zeros_like	5.4139ms	4.4040ms	227.0678 Ops/s	217.8449 Ops/s	$\color{#35bf28}+4.23\%$
test_ones_like	3.8437ms	3.1109ms	321.4537 Ops/s	194.8564 Ops/s	$\textbf{\color{#35bf28}+64.97\%}$
test_clone	6.0568ms	4.8785ms	204.9809 Ops/s	197.6984 Ops/s	$\color{#35bf28}+3.68\%$
test_squeeze	60.7840μs	12.4568μs	80.2773 KOps/s	79.1883 KOps/s	$\color{#35bf28}+1.38\%$
test_unsqueeze	0.3025ms	95.6735μs	10.4522 KOps/s	10.6116 KOps/s	$\color{#d91a1a}-1.50\%$
test_split	0.3120ms	0.1955ms	5.1154 KOps/s	5.1710 KOps/s	$\color{#d91a1a}-1.08\%$
test_permute	0.3327ms	0.1989ms	5.0268 KOps/s	5.0110 KOps/s	$\color{#35bf28}+0.32\%$
test_stack	29.1951ms	24.3494ms	41.0688 Ops/s	40.0692 Ops/s	$\color{#35bf28}+2.49\%$
test_cat	26.9454ms	24.1835ms	41.3505 Ops/s	40.5290 Ops/s	$\color{#35bf28}+2.03\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}27$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	34.7010μs	11.6116μs	86.1211 KOps/s	76.6752 KOps/s	$\textbf{\color{#35bf28}+12.32\%}$
test_plain_set_stack_nested	60.6420μs	11.6621μs	85.7480 KOps/s	76.4804 KOps/s	$\textbf{\color{#35bf28}+12.12\%}$
test_plain_set_nested_inplace	38.7110μs	12.7374μs	78.5087 KOps/s	71.2184 KOps/s	$\textbf{\color{#35bf28}+10.24\%}$
test_plain_set_stack_nested_inplace	36.8100μs	12.6286μs	79.1854 KOps/s	71.5294 KOps/s	$\textbf{\color{#35bf28}+10.70\%}$
test_items	22.3200μs	2.8653μs	348.9998 KOps/s	346.3076 KOps/s	$\color{#35bf28}+0.78\%$
test_items_nested	0.4300ms	0.3764ms	2.6565 KOps/s	2.7225 KOps/s	$\color{#d91a1a}-2.43\%$
test_items_nested_locked	0.4329ms	0.3764ms	2.6565 KOps/s	2.6975 KOps/s	$\color{#d91a1a}-1.52\%$
test_items_nested_leaf	0.1029ms	60.2129μs	16.6077 KOps/s	16.4152 KOps/s	$\color{#35bf28}+1.17\%$
test_items_stack_nested	0.4286ms	0.3721ms	2.6873 KOps/s	2.7324 KOps/s	$\color{#d91a1a}-1.65\%$
test_items_stack_nested_leaf	94.0120μs	61.0668μs	16.3755 KOps/s	16.5214 KOps/s	$\color{#d91a1a}-0.88\%$
test_items_stack_nested_locked	0.4440ms	0.3762ms	2.6583 KOps/s	2.7311 KOps/s	$\color{#d91a1a}-2.67\%$
test_keys	25.7900μs	3.4219μs	292.2323 KOps/s	289.5877 KOps/s	$\color{#35bf28}+0.91\%$
test_keys_nested	0.1176ms	89.7166μs	11.1462 KOps/s	11.1345 KOps/s	$\color{#35bf28}+0.11\%$
test_keys_nested_locked	0.7810ms	95.2798μs	10.4954 KOps/s	10.4793 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_nested_leaf	0.1087ms	80.5427μs	12.4158 KOps/s	12.4536 KOps/s	$\color{#d91a1a}-0.30\%$
test_keys_stack_nested	0.1122ms	89.6281μs	11.1572 KOps/s	11.2202 KOps/s	$\color{#d91a1a}-0.56\%$
test_keys_stack_nested_leaf	0.1120ms	80.5940μs	12.4079 KOps/s	12.5477 KOps/s	$\color{#d91a1a}-1.11\%$
test_keys_stack_nested_locked	0.1329ms	94.8870μs	10.5389 KOps/s	10.4335 KOps/s	$\color{#35bf28}+1.01\%$
test_values	4.6502μs	0.8512μs	1.1748 MOps/s	1.1716 MOps/s	$\color{#35bf28}+0.28\%$
test_values_nested	62.0110μs	38.0065μs	26.3113 KOps/s	26.6004 KOps/s	$\color{#d91a1a}-1.09\%$
test_values_nested_locked	63.1610μs	39.6263μs	25.2357 KOps/s	25.3146 KOps/s	$\color{#d91a1a}-0.31\%$
test_values_nested_leaf	67.7910μs	43.5217μs	22.9770 KOps/s	23.3639 KOps/s	$\color{#d91a1a}-1.66\%$
test_values_stack_nested	0.1067ms	38.0776μs	26.2622 KOps/s	26.3205 KOps/s	$\color{#d91a1a}-0.22\%$
test_values_stack_nested_leaf	77.2820μs	43.8688μs	22.7952 KOps/s	23.0705 KOps/s	$\color{#d91a1a}-1.19\%$
test_values_stack_nested_locked	70.2020μs	39.9714μs	25.0179 KOps/s	25.3137 KOps/s	$\color{#d91a1a}-1.17\%$
test_membership	1.6100μs	0.5010μs	1.9958 MOps/s	1.9934 MOps/s	$\color{#35bf28}+0.12\%$
test_membership_nested	18.0305μs	2.0081μs	497.9815 KOps/s	477.5331 KOps/s	$\color{#35bf28}+4.28\%$
test_membership_nested_leaf	15.5505μs	2.0169μs	495.8054 KOps/s	493.1493 KOps/s	$\color{#35bf28}+0.54\%$
test_membership_stacked_nested	35.4600μs	2.1011μs	475.9402 KOps/s	464.0376 KOps/s	$\color{#35bf28}+2.57\%$
test_membership_stacked_nested_leaf	28.9510μs	2.0840μs	479.8375 KOps/s	465.1463 KOps/s	$\color{#35bf28}+3.16\%$
test_membership_nested_last	23.9710μs	3.1171μs	320.8080 KOps/s	315.4772 KOps/s	$\color{#35bf28}+1.69\%$
test_membership_nested_leaf_last	36.5810μs	3.1290μs	319.5919 KOps/s	319.6677 KOps/s	$\color{#d91a1a}-0.02\%$
test_membership_stacked_nested_last	25.7310μs	3.1087μs	321.6799 KOps/s	325.7841 KOps/s	$\color{#d91a1a}-1.26\%$
test_membership_stacked_nested_leaf_last	29.1400μs	3.1004μs	322.5357 KOps/s	320.7879 KOps/s	$\color{#35bf28}+0.54\%$
test_nested_getleaf	28.4710μs	6.2129μs	160.9549 KOps/s	160.6112 KOps/s	$\color{#35bf28}+0.21\%$
test_nested_get	29.3300μs	5.9319μs	168.5801 KOps/s	165.0711 KOps/s	$\color{#35bf28}+2.13\%$
test_stacked_getleaf	27.6200μs	6.1082μs	163.7152 KOps/s	161.2054 KOps/s	$\color{#35bf28}+1.56\%$
test_stacked_get	30.7410μs	5.7508μs	173.8899 KOps/s	171.2580 KOps/s	$\color{#35bf28}+1.54\%$
test_nested_getitemleaf	35.1810μs	6.3784μs	156.7799 KOps/s	152.3824 KOps/s	$\color{#35bf28}+2.89\%$
test_nested_getitem	30.9200μs	6.0056μs	166.5117 KOps/s	161.8846 KOps/s	$\color{#35bf28}+2.86\%$
test_stacked_getitemleaf	30.7210μs	6.3793μs	156.7581 KOps/s	155.5627 KOps/s	$\color{#35bf28}+0.77\%$
test_stacked_getitem	40.8800μs	5.9341μs	168.5182 KOps/s	164.5026 KOps/s	$\color{#35bf28}+2.44\%$
test_lock_nested	9.1417ms	0.3564ms	2.8061 KOps/s	2.8922 KOps/s	$\color{#d91a1a}-2.98\%$
test_lock_stack_nested	0.4054ms	0.3535ms	2.8289 KOps/s	2.8483 KOps/s	$\color{#d91a1a}-0.68\%$
test_unlock_nested	0.4210ms	0.2946ms	3.3943 KOps/s	3.4574 KOps/s	$\color{#d91a1a}-1.82\%$
test_unlock_stack_nested	0.3355ms	0.2905ms	3.4419 KOps/s	3.4392 KOps/s	$\color{#35bf28}+0.08\%$
test_flatten_speed	0.1103ms	77.7566μs	12.8606 KOps/s	12.6976 KOps/s	$\color{#35bf28}+1.28\%$
test_unflatten_speed	0.3892ms	0.3275ms	3.0538 KOps/s	3.0954 KOps/s	$\color{#d91a1a}-1.34\%$
test_common_ops	0.7307ms	0.5923ms	1.6884 KOps/s	1.5603 KOps/s	$\textbf{\color{#35bf28}+8.21\%}$
test_creation	0.1215ms	1.7349μs	576.3896 KOps/s	577.5001 KOps/s	$\color{#d91a1a}-0.19\%$
test_creation_empty	38.3110μs	6.4441μs	155.1803 KOps/s	111.5667 KOps/s	$\textbf{\color{#35bf28}+39.09\%}$
test_creation_nested_1	34.3210μs	8.0500μs	124.2237 KOps/s	94.7314 KOps/s	$\textbf{\color{#35bf28}+31.13\%}$
test_creation_nested_2	33.7910μs	10.7026μs	93.4352 KOps/s	74.3410 KOps/s	$\textbf{\color{#35bf28}+25.68\%}$
test_clone	44.3210μs	11.0953μs	90.1281 KOps/s	89.6506 KOps/s	$\color{#35bf28}+0.53\%$
test_getitem[int]	1.2519ms	10.7434μs	93.0807 KOps/s	90.7623 KOps/s	$\color{#35bf28}+2.55\%$
test_getitem[slice_int]	0.1111ms	21.4925μs	46.5278 KOps/s	46.0558 KOps/s	$\color{#35bf28}+1.02\%$
test_getitem[range]	0.1361ms	38.5444μs	25.9441 KOps/s	26.5714 KOps/s	$\color{#d91a1a}-2.36\%$
test_getitem[tuple]	0.1169ms	18.4196μs	54.2899 KOps/s	53.6793 KOps/s	$\color{#35bf28}+1.14\%$
test_getitem[list]	0.1273ms	33.0418μs	30.2647 KOps/s	30.1278 KOps/s	$\color{#35bf28}+0.45\%$
test_setitem_dim[int]	43.3510μs	19.3978μs	51.5522 KOps/s	49.9502 KOps/s	$\color{#35bf28}+3.21\%$
test_setitem_dim[slice_int]	62.7410μs	38.9340μs	25.6845 KOps/s	25.5408 KOps/s	$\color{#35bf28}+0.56\%$
test_setitem_dim[range]	79.2810μs	52.2995μs	19.1206 KOps/s	18.8262 KOps/s	$\color{#35bf28}+1.56\%$
test_setitem_dim[tuple]	52.8210μs	31.9115μs	31.3366 KOps/s	31.0041 KOps/s	$\color{#35bf28}+1.07\%$
test_setitem	50.9310μs	14.5632μs	68.6662 KOps/s	60.6221 KOps/s	$\textbf{\color{#35bf28}+13.27\%}$
test_set	56.0810μs	13.8587μs	72.1567 KOps/s	62.3675 KOps/s	$\textbf{\color{#35bf28}+15.70\%}$
test_set_shared	0.5138ms	0.1593ms	6.2756 KOps/s	6.2590 KOps/s	$\color{#35bf28}+0.27\%$
test_update	0.4135ms	17.3837μs	57.5250 KOps/s	47.4522 KOps/s	$\textbf{\color{#35bf28}+21.23\%}$
test_update_nested	76.7710μs	26.5433μs	37.6744 KOps/s	33.2422 KOps/s	$\textbf{\color{#35bf28}+13.33\%}$
test_update__nested	0.4823ms	25.8514μs	38.6826 KOps/s	38.3309 KOps/s	$\color{#35bf28}+0.92\%$
test_set_nested	64.2810μs	15.4240μs	64.8340 KOps/s	57.3906 KOps/s	$\textbf{\color{#35bf28}+12.97\%}$
test_set_nested_new	52.5910μs	17.6263μs	56.7333 KOps/s	50.5187 KOps/s	$\textbf{\color{#35bf28}+12.30\%}$
test_select	59.7910μs	29.1847μs	34.2646 KOps/s	31.3049 KOps/s	$\textbf{\color{#35bf28}+9.45\%}$
test_select_nested	72.4410μs	44.1547μs	22.6477 KOps/s	22.9345 KOps/s	$\color{#d91a1a}-1.25\%$
test_exclude_nested	89.8020μs	63.7760μs	15.6799 KOps/s	15.6641 KOps/s	$\color{#35bf28}+0.10\%$
test_empty[True]	0.3981ms	0.2954ms	3.3847 KOps/s	3.3453 KOps/s	$\color{#35bf28}+1.18\%$
test_empty[False]	3.6471μs	0.8203μs	1.2190 MOps/s	1.2116 MOps/s	$\color{#35bf28}+0.61\%$
test_to	91.4620μs	56.9141μs	17.5703 KOps/s	18.0445 KOps/s	$\color{#d91a1a}-2.63\%$
test_to_nonblocking	79.9120μs	47.5494μs	21.0308 KOps/s	21.2231 KOps/s	$\color{#d91a1a}-0.91\%$
test_unbind_speed	0.2867ms	0.2504ms	3.9938 KOps/s	4.0393 KOps/s	$\color{#d91a1a}-1.13\%$
test_unbind_speed_stack0	0.3192ms	0.2459ms	4.0659 KOps/s	4.0337 KOps/s	$\color{#35bf28}+0.80\%$
test_unbind_speed_stack1	93.3927ms	0.7566ms	1.3217 KOps/s	1.3329 KOps/s	$\color{#d91a1a}-0.84\%$
test_split	96.0199ms	1.6145ms	619.3939 Ops/s	625.2970 Ops/s	$\color{#d91a1a}-0.94\%$
test_chunk	95.4910ms	1.6085ms	621.6920 Ops/s	620.2804 Ops/s	$\color{#35bf28}+0.23\%$
test_consolidate[False-None]	3.3884ms	2.7195ms	367.7152 Ops/s	365.6155 Ops/s	$\color{#35bf28}+0.57\%$
test_consolidate[default-None]	1.7867ms	1.6872ms	592.6961 Ops/s	588.3386 Ops/s	$\color{#35bf28}+0.74\%$
test_consolidate[reduce-overhead-None]	1.8074ms	1.7264ms	579.2349 Ops/s	572.9935 Ops/s	$\color{#35bf28}+1.09\%$
test_consolidate_njt[False-None]	6.8981ms	6.5231ms	153.3012 Ops/s	154.3583 Ops/s	$\color{#d91a1a}-0.68\%$
test_to[False-False-None]	1.8352ms	1.7370ms	575.7026 Ops/s	582.3525 Ops/s	$\color{#d91a1a}-1.14\%$
test_to[True-False-None]	1.6006ms	1.3857ms	721.6372 Ops/s	745.9979 Ops/s	$\color{#d91a1a}-3.27\%$
test_to[within-False-None]	4.4632ms	4.2200ms	236.9649 Ops/s	238.1173 Ops/s	$\color{#d91a1a}-0.48\%$
test_to[True-default-None]	5.5349ms	5.1750ms	193.2370 Ops/s	189.1101 Ops/s	$\color{#35bf28}+2.18\%$
test_to_njt[False-False-None]	7.1052ms	6.9211ms	144.4852 Ops/s	145.6563 Ops/s	$\color{#d91a1a}-0.80\%$
test_to_njt[True-False-None]	6.1730ms	5.5530ms	180.0837 Ops/s	184.0407 Ops/s	$\color{#d91a1a}-2.15\%$
test_to_njt[within-False-None]	12.4064ms	12.0953ms	82.6767 Ops/s	82.7111 Ops/s	$\color{#d91a1a}-0.04\%$
test_creation[device0]	0.4607ms	79.4542μs	12.5859 KOps/s	12.5805 KOps/s	$\color{#35bf28}+0.04\%$
test_creation_from_tensor	0.5387ms	83.1333μs	12.0289 KOps/s	11.8738 KOps/s	$\color{#35bf28}+1.31\%$
test_add_one[memmap_tensor0]	0.3374ms	7.1251μs	140.3480 KOps/s	142.0584 KOps/s	$\color{#d91a1a}-1.20\%$
test_contiguous[memmap_tensor0]	1.7526μs	0.4043μs	2.4733 MOps/s	2.4071 MOps/s	$\color{#35bf28}+2.75\%$
test_stack[memmap_tensor0]	41.2610μs	4.6812μs	213.6210 KOps/s	211.1546 KOps/s	$\color{#35bf28}+1.17\%$
test_memmaptd_index	1.4697ms	0.2448ms	4.0845 KOps/s	4.0891 KOps/s	$\color{#d91a1a}-0.11\%$
test_memmaptd_index_astensor	0.4573ms	0.3084ms	3.2422 KOps/s	3.2489 KOps/s	$\color{#d91a1a}-0.21\%$
test_memmaptd_index_op	0.8230ms	0.5597ms	1.7868 KOps/s	1.6702 KOps/s	$\textbf{\color{#35bf28}+6.98\%}$
test_serialize_model	0.1321s	0.1309s	7.6377 Ops/s	7.6407 Ops/s	$\color{#d91a1a}-0.04\%$
test_serialize_model_pickle	1.3505s	1.2145s	0.8234 Ops/s	0.8434 Ops/s	$\color{#d91a1a}-2.37\%$
test_serialize_weights	0.1319s	0.1310s	7.6318 Ops/s	7.6691 Ops/s	$\color{#d91a1a}-0.49\%$
test_serialize_weights_returnearly	0.3358s	53.8080ms	18.5846 Ops/s	23.2255 Ops/s	$\textbf{\color{#d91a1a}-19.98\%}$
test_serialize_weights_pickle	1.3881s	1.2211s	0.8189 Ops/s	0.8188 Ops/s	$\color{#35bf28}+0.01\%$
test_reshape_pytree	66.2710μs	21.9634μs	45.5304 KOps/s	44.9075 KOps/s	$\color{#35bf28}+1.39\%$
test_reshape_td	65.3920μs	26.6172μs	37.5697 KOps/s	36.9911 KOps/s	$\color{#35bf28}+1.56\%$
test_view_pytree	54.0510μs	21.8391μs	45.7894 KOps/s	45.2590 KOps/s	$\color{#35bf28}+1.17\%$
test_view_td	70.1210μs	32.0206μs	31.2299 KOps/s	31.7138 KOps/s	$\color{#d91a1a}-1.53\%$
test_unbind_pytree	75.1620μs	28.8606μs	34.6493 KOps/s	34.4112 KOps/s	$\color{#35bf28}+0.69\%$
test_unbind_td	0.7805ms	37.8676μs	26.4078 KOps/s	26.3323 KOps/s	$\color{#35bf28}+0.29\%$
test_split_pytree	63.5810μs	29.4403μs	33.9670 KOps/s	32.7729 KOps/s	$\color{#35bf28}+3.64\%$
test_split_td	0.9251ms	40.1517μs	24.9056 KOps/s	24.9784 KOps/s	$\color{#d91a1a}-0.29\%$
test_add_pytree	0.1227ms	36.0347μs	27.7510 KOps/s	27.6656 KOps/s	$\color{#35bf28}+0.31\%$
test_add_td	90.9820μs	47.7264μs	20.9528 KOps/s	19.6334 KOps/s	$\textbf{\color{#35bf28}+6.72\%}$
test_compile_add_one_nested[tensordict-compile]	0.1763ms	0.1238ms	8.0792 KOps/s	7.6959 KOps/s	$\color{#35bf28}+4.98\%$
test_compile_add_one_nested[tensordict-eager]	0.2268ms	0.1338ms	7.4758 KOps/s	7.4623 KOps/s	$\color{#35bf28}+0.18\%$
test_compile_add_one_nested[pytree-compile]	0.2014ms	97.1515μs	10.2932 KOps/s	10.2413 KOps/s	$\color{#35bf28}+0.51\%$
test_compile_add_one_nested[pytree-eager]	1.3573ms	0.1523ms	6.5640 KOps/s	6.6637 KOps/s	$\color{#d91a1a}-1.50\%$
test_compile_copy_nested[tensordict-compile]	64.2610μs	25.7476μs	38.8386 KOps/s	36.1531 KOps/s	$\textbf{\color{#35bf28}+7.43\%}$
test_compile_copy_nested[tensordict-eager]	62.8220μs	29.1327μs	34.3256 KOps/s	34.3197 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_copy_nested[pytree-compile]	0.4254ms	64.6143μs	15.4765 KOps/s	15.4526 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_copy_nested[pytree-eager]	0.1203ms	48.9423μs	20.4322 KOps/s	20.1198 KOps/s	$\color{#35bf28}+1.55\%$
test_compile_add_one_flat[tensordict-compile]	0.1909ms	0.1490ms	6.7110 KOps/s	6.9606 KOps/s	$\color{#d91a1a}-3.58\%$
test_compile_add_one_flat[tensordict-eager]	0.3345ms	0.2200ms	4.5462 KOps/s	4.5734 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_add_one_flat[tensorclass-compile]	0.1460ms	98.8178μs	10.1196 KOps/s	9.9978 KOps/s	$\color{#35bf28}+1.22\%$
test_compile_add_one_flat[tensorclass-eager]	0.1121ms	55.0241μs	18.1739 KOps/s	17.6660 KOps/s	$\color{#35bf28}+2.87\%$
test_compile_add_one_flat[pytree-compile]	0.2498ms	0.1377ms	7.2614 KOps/s	7.1908 KOps/s	$\color{#35bf28}+0.98\%$
test_compile_add_one_flat[pytree-eager]	0.5507ms	0.4932ms	2.0277 KOps/s	2.0858 KOps/s	$\color{#d91a1a}-2.78\%$
test_compile_add_self_flat[tensordict-eager]	0.4105ms	0.2696ms	3.7086 KOps/s	3.7893 KOps/s	$\color{#d91a1a}-2.13\%$
test_compile_add_self_flat[tensordict-compile]	0.2067ms	0.1523ms	6.5649 KOps/s	6.9695 KOps/s	$\textbf{\color{#d91a1a}-5.81\%}$
test_compile_add_self_flat[tensorclass-eager]	0.1706ms	69.9526μs	14.2954 KOps/s	13.9427 KOps/s	$\color{#35bf28}+2.53\%$
test_compile_add_self_flat[tensorclass-compile]	0.2329ms	0.1029ms	9.7205 KOps/s	10.0610 KOps/s	$\color{#d91a1a}-3.38\%$
test_compile_add_self_flat[pytree-eager]	0.5525ms	0.4224ms	2.3673 KOps/s	2.4678 KOps/s	$\color{#d91a1a}-4.07\%$
test_compile_add_self_flat[pytree-compile]	0.2191ms	0.1426ms	7.0123 KOps/s	7.2865 KOps/s	$\color{#d91a1a}-3.76\%$
test_compile_copy_flat[tensordict-compile]	0.1184ms	19.5879μs	51.0520 KOps/s	52.4616 KOps/s	$\color{#d91a1a}-2.69\%$
test_compile_copy_flat[tensordict-eager]	0.1224ms	31.8128μs	31.4338 KOps/s	31.5588 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_copy_flat[pytree-compile]	0.2057ms	70.0114μs	14.2834 KOps/s	14.2812 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_copy_flat[pytree-eager]	0.1594ms	51.9882μs	19.2351 KOps/s	19.0330 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_assign_and_add[tensordict-compile]	1.6148ms	0.3897ms	2.5662 KOps/s	2.2266 KOps/s	$\textbf{\color{#35bf28}+15.25\%}$
test_compile_assign_and_add[tensordict-eager]	3.1067ms	2.7187ms	367.8164 Ops/s	373.5853 Ops/s	$\color{#d91a1a}-1.54\%$
test_compile_assign_and_add[pytree-compile]	1.5885ms	0.4366ms	2.2902 KOps/s	2.2854 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_assign_and_add[pytree-eager]	2.9587ms	2.8179ms	354.8720 Ops/s	379.5435 Ops/s	$\textbf{\color{#d91a1a}-6.50\%}$
test_compile_indexing[tensor-tensordict-compile]	0.5172ms	0.1183ms	8.4538 KOps/s	8.4052 KOps/s	$\color{#35bf28}+0.58\%$
test_compile_indexing[tensor-tensordict-eager]	0.5817ms	80.1653μs	12.4742 KOps/s	11.8649 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.6739ms	0.1109ms	9.0170 KOps/s	8.9853 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1235ms	74.1089μs	13.4937 KOps/s	13.7949 KOps/s	$\color{#d91a1a}-2.18\%$
test_compile_indexing[tensor-pytree-compile]	0.2334ms	0.1126ms	8.8788 KOps/s	9.3671 KOps/s	$\textbf{\color{#d91a1a}-5.21\%}$
test_compile_indexing[tensor-pytree-eager]	0.1304ms	72.6901μs	13.7570 KOps/s	14.3702 KOps/s	$\color{#d91a1a}-4.27\%$
test_compile_indexing[slice-tensordict-compile]	0.1641ms	0.1025ms	9.7574 KOps/s	9.8996 KOps/s	$\color{#d91a1a}-1.44\%$
test_compile_indexing[slice-tensordict-eager]	0.1421ms	18.3233μs	54.5753 KOps/s	57.3264 KOps/s	$\color{#d91a1a}-4.80\%$
test_compile_indexing[slice-tensorclass-compile]	0.1473ms	97.7321μs	10.2320 KOps/s	10.2676 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_indexing[slice-tensorclass-eager]	48.7310μs	17.1538μs	58.2960 KOps/s	58.8353 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_indexing[slice-pytree-compile]	0.1370ms	96.5605μs	10.3562 KOps/s	10.2001 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_indexing[slice-pytree-eager]	52.4410μs	16.0795μs	62.1912 KOps/s	62.3874 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_indexing[int-tensordict-compile]	0.1557ms	0.1062ms	9.4137 KOps/s	9.2498 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_indexing[int-tensordict-eager]	0.5544ms	17.3702μs	57.5699 KOps/s	56.7432 KOps/s	$\color{#35bf28}+1.46\%$
test_compile_indexing[int-tensorclass-compile]	0.1719ms	99.3987μs	10.0605 KOps/s	10.2162 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_indexing[int-tensorclass-eager]	52.6710μs	17.2276μs	58.0466 KOps/s	62.5678 KOps/s	$\textbf{\color{#d91a1a}-7.23\%}$
test_compile_indexing[int-pytree-compile]	0.2050ms	96.5931μs	10.3527 KOps/s	10.1721 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_indexing[int-pytree-eager]	53.0710μs	16.0894μs	62.1527 KOps/s	62.7430 KOps/s	$\color{#d91a1a}-0.94\%$
test_mod_add[eager]	74.8920μs	37.9080μs	26.3797 KOps/s	25.0197 KOps/s	$\textbf{\color{#35bf28}+5.44\%}$
test_mod_add[compile]	0.3844ms	81.2656μs	12.3053 KOps/s	12.0955 KOps/s	$\color{#35bf28}+1.74\%$
test_mod_add[compile-overhead]	0.3272ms	0.1692ms	5.9094 KOps/s	5.6476 KOps/s	$\color{#35bf28}+4.64\%$
test_mod_wrap[eager]	0.3568ms	0.2524ms	3.9622 KOps/s	3.8070 KOps/s	$\color{#35bf28}+4.08\%$
test_mod_wrap[compile]	0.4221ms	0.3008ms	3.3244 KOps/s	3.4163 KOps/s	$\color{#d91a1a}-2.69\%$
test_mod_wrap[compile-overhead]	7.4274ms	3.8560ms	259.3381 Ops/s	262.7778 Ops/s	$\color{#d91a1a}-1.31\%$
test_mod_wrap_and_backward[eager]	1.6029ms	1.5141ms	660.4627 Ops/s	676.8315 Ops/s	$\color{#d91a1a}-2.42\%$
test_mod_wrap_and_backward[compile]	1.5246ms	1.4274ms	700.5638 Ops/s	712.8089 Ops/s	$\color{#d91a1a}-1.72\%$
test_mod_wrap_and_backward[compile-overhead]	1.4755ms	1.0047ms	995.3573 Ops/s	968.5917 Ops/s	$\color{#35bf28}+2.76\%$
test_seq_add[eager]	0.1630ms	0.1164ms	8.5924 KOps/s	8.1408 KOps/s	$\textbf{\color{#35bf28}+5.55\%}$
test_seq_add[compile]	0.1346ms	88.8012μs	11.2611 KOps/s	10.5511 KOps/s	$\textbf{\color{#35bf28}+6.73\%}$
test_seq_add[compile-overhead]	0.1828ms	0.1299ms	7.6988 KOps/s	7.5934 KOps/s	$\color{#35bf28}+1.39\%$
test_seq_wrap[eager]	0.4869ms	0.4237ms	2.3600 KOps/s	2.2661 KOps/s	$\color{#35bf28}+4.14\%$
test_seq_wrap[compile]	0.3759ms	0.3034ms	3.2954 KOps/s	3.2382 KOps/s	$\color{#35bf28}+1.77\%$
test_seq_wrap[compile-overhead]	0.2838ms	0.2239ms	4.4661 KOps/s	4.3696 KOps/s	$\color{#35bf28}+2.21\%$
test_func_call_runtime[False-eager]	0.8024ms	0.7514ms	1.3308 KOps/s	1.3277 KOps/s	$\color{#35bf28}+0.23\%$
test_func_call_runtime[False-compile]	0.8130ms	0.7437ms	1.3446 KOps/s	1.3200 KOps/s	$\color{#35bf28}+1.86\%$
test_func_call_runtime[False-compile-overhead]	0.4129ms	0.3649ms	2.7407 KOps/s	2.7162 KOps/s	$\color{#35bf28}+0.90\%$
test_func_call_runtime[True-eager]	0.9768ms	0.9178ms	1.0896 KOps/s	1.0906 KOps/s	$\color{#d91a1a}-0.10\%$
test_func_call_runtime[True-compile]	0.8659ms	0.7800ms	1.2821 KOps/s	1.2865 KOps/s	$\color{#d91a1a}-0.34\%$
test_func_call_runtime[True-compile-overhead]	0.5250ms	0.3844ms	2.6017 KOps/s	2.5845 KOps/s	$\color{#35bf28}+0.66\%$
test_func_call_cm_runtime[False-eager]	0.8461ms	0.7454ms	1.3415 KOps/s	1.3449 KOps/s	$\color{#d91a1a}-0.25\%$
test_func_call_cm_runtime[False-compile]	0.8005ms	0.7565ms	1.3220 KOps/s	1.3148 KOps/s	$\color{#35bf28}+0.55\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4386ms	0.3676ms	2.7203 KOps/s	2.7222 KOps/s	$\color{#d91a1a}-0.07\%$
test_func_call_cm_runtime[True-eager]	1.1157ms	1.0136ms	986.5453 Ops/s	973.2682 Ops/s	$\color{#35bf28}+1.36\%$
test_func_call_cm_runtime[True-compile]	1.1016ms	1.0050ms	995.0644 Ops/s	995.0395 Ops/s	$+0.00\%$
test_func_call_cm_runtime[True-compile-overhead]	1.0768ms	1.0033ms	996.7254 Ops/s	988.2776 Ops/s	$\color{#35bf28}+0.85\%$
test_vmap_func_call_cm_runtime[eager]	2.5282ms	2.1203ms	471.6352 Ops/s	473.4547 Ops/s	$\color{#d91a1a}-0.38\%$
test_vmap_func_call_cm_runtime[compile]	0.8735ms	0.8161ms	1.2253 KOps/s	1.2093 KOps/s	$\color{#35bf28}+1.33\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4547ms	0.4146ms	2.4120 KOps/s	2.3748 KOps/s	$\color{#35bf28}+1.57\%$
test_distributed	3.1668ms	0.1867ms	5.3556 KOps/s	8.3825 KOps/s	$\textbf{\color{#d91a1a}-36.11\%}$
test_tdmodule	0.3969ms	20.1480μs	49.6328 KOps/s	46.3040 KOps/s	$\textbf{\color{#35bf28}+7.19\%}$
test_tdmodule_dispatch	56.0910μs	34.6251μs	28.8807 KOps/s	26.2127 KOps/s	$\textbf{\color{#35bf28}+10.18\%}$
test_tdseq	27.8110μs	19.9936μs	50.0161 KOps/s	46.2918 KOps/s	$\textbf{\color{#35bf28}+8.05\%}$
test_tdseq_dispatch	67.4110μs	38.0974μs	26.2485 KOps/s	24.8509 KOps/s	$\textbf{\color{#35bf28}+5.62\%}$
test_instantiation_functorch	1.7202ms	1.6266ms	614.7885 Ops/s	637.2970 Ops/s	$\color{#d91a1a}-3.53\%$
test_exec_functorch	0.1953ms	0.1472ms	6.7920 KOps/s	6.8060 KOps/s	$\color{#d91a1a}-0.21\%$
test_exec_functional_call	0.1827ms	0.1402ms	7.1305 KOps/s	7.0663 KOps/s	$\color{#35bf28}+0.91\%$
test_exec_td_decorator	0.3752ms	0.1912ms	5.2293 KOps/s	5.2085 KOps/s	$\color{#35bf28}+0.40\%$
test_vmap_mlp_speed_decorator[True-True]	0.8430ms	0.7056ms	1.4173 KOps/s	1.4469 KOps/s	$\color{#d91a1a}-2.05\%$
test_vmap_mlp_speed_decorator[True-False]	0.8094ms	0.6852ms	1.4594 KOps/s	1.4512 KOps/s	$\color{#35bf28}+0.57\%$
test_vmap_mlp_speed_decorator[False-True]	0.7161ms	0.6005ms	1.6652 KOps/s	1.6765 KOps/s	$\color{#d91a1a}-0.68\%$
test_vmap_mlp_speed_decorator[False-False]	0.7369ms	0.6078ms	1.6453 KOps/s	1.6748 KOps/s	$\color{#d91a1a}-1.77\%$
test_vmap_transformer_speed_decorator[True-True]	19.5983ms	19.4726ms	51.3541 Ops/s	52.0957 Ops/s	$\color{#d91a1a}-1.42\%$
test_vmap_transformer_speed_decorator[True-False]	19.9331ms	19.4627ms	51.3804 Ops/s	52.0437 Ops/s	$\color{#d91a1a}-1.27\%$
test_vmap_transformer_speed_decorator[False-True]	20.1067ms	19.3688ms	51.6293 Ops/s	52.4486 Ops/s	$\color{#d91a1a}-1.56\%$
test_vmap_transformer_speed_decorator[False-False]	19.4235ms	19.3152ms	51.7726 Ops/s	52.5416 Ops/s	$\color{#d91a1a}-1.46\%$
test_to_module_speed[True]	1.4507ms	0.9747ms	1.0260 KOps/s	1.0286 KOps/s	$\color{#d91a1a}-0.26\%$
test_to_module_speed[False]	1.1046ms	0.9493ms	1.0534 KOps/s	1.0529 KOps/s	$\color{#35bf28}+0.05\%$
test_tc_init	63.7210μs	34.4924μs	28.9919 KOps/s	28.6459 KOps/s	$\color{#35bf28}+1.21\%$
test_tc_init_nested	0.1149ms	69.8295μs	14.3206 KOps/s	13.8096 KOps/s	$\color{#35bf28}+3.70\%$
test_tc_first_layer_tensor	19.1010μs	0.8033μs	1.2448 MOps/s	1.4283 MOps/s	$\textbf{\color{#d91a1a}-12.84\%}$
test_tc_first_layer_nontensor	24.6610μs	2.2056μs	453.3862 KOps/s	448.3332 KOps/s	$\color{#35bf28}+1.13\%$
test_tc_second_layer_tensor	26.3310μs	1.4999μs	666.7122 KOps/s	699.2673 KOps/s	$\color{#d91a1a}-4.66\%$
test_tc_second_layer_nontensor	24.1210μs	2.9339μs	340.8459 KOps/s	333.4364 KOps/s	$\color{#35bf28}+2.22\%$
test_unbind	7.2074ms	6.9710ms	143.4506 Ops/s	142.4282 Ops/s	$\color{#35bf28}+0.72\%$
test_full_like	12.1539ms	9.3104ms	107.4070 Ops/s	107.7254 Ops/s	$\color{#d91a1a}-0.30\%$
test_zeros_like	5.9012ms	4.2578ms	234.8613 Ops/s	230.8062 Ops/s	$\color{#35bf28}+1.76\%$
test_ones_like	4.4968ms	4.3213ms	231.4142 Ops/s	230.6611 Ops/s	$\color{#35bf28}+0.33\%$
test_clone	11.3713ms	9.0947ms	109.9539 Ops/s	156.4913 Ops/s	$\textbf{\color{#d91a1a}-29.74\%}$
test_squeeze	47.1410μs	9.7577μs	102.4835 KOps/s	103.0415 KOps/s	$\color{#d91a1a}-0.54\%$
test_unsqueeze	0.1137ms	73.9509μs	13.5225 KOps/s	13.1322 KOps/s	$\color{#35bf28}+2.97\%$
test_split	0.2099s	0.2211ms	4.5222 KOps/s	6.0173 KOps/s	$\textbf{\color{#d91a1a}-24.85\%}$
test_permute	0.2338ms	0.1855ms	5.3915 KOps/s	5.2793 KOps/s	$\color{#35bf28}+2.13\%$
test_stack	51.7408ms	50.2445ms	19.9027 Ops/s	19.8132 Ops/s	$\color{#35bf28}+0.45\%$
test_cat	50.6412ms	49.8984ms	20.0407 Ops/s	19.8848 Ops/s	$\color{#35bf28}+0.78\%$