[Performance] Faster copy of TDParams

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}34$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.9340μs	16.4075μs	60.9478 KOps/s	57.5544 KOps/s	$\textbf{\color{#35bf28}+5.90\%}$
test_plain_set_stack_nested	47.0790μs	16.7674μs	59.6395 KOps/s	57.0728 KOps/s	$\color{#35bf28}+4.50\%$
test_plain_set_nested_inplace	68.0560μs	18.3168μs	54.5947 KOps/s	52.3311 KOps/s	$\color{#35bf28}+4.33\%$
test_plain_set_stack_nested_inplace	67.1860μs	18.3636μs	54.4556 KOps/s	51.7642 KOps/s	$\textbf{\color{#35bf28}+5.20\%}$
test_items	0.1604ms	4.2693μs	234.2330 KOps/s	237.2934 KOps/s	$\color{#d91a1a}-1.29\%$
test_items_nested	0.5744ms	0.3387ms	2.9526 KOps/s	2.8966 KOps/s	$\color{#35bf28}+1.93\%$
test_items_nested_locked	0.6404ms	0.3422ms	2.9225 KOps/s	2.9316 KOps/s	$\color{#d91a1a}-0.31\%$
test_items_nested_leaf	0.1299ms	71.7923μs	13.9291 KOps/s	14.0357 KOps/s	$\color{#d91a1a}-0.76\%$
test_items_stack_nested	0.7922ms	0.3477ms	2.8756 KOps/s	2.9071 KOps/s	$\color{#d91a1a}-1.08\%$
test_items_stack_nested_leaf	0.3105ms	74.9423μs	13.3436 KOps/s	13.4479 KOps/s	$\color{#d91a1a}-0.78\%$
test_items_stack_nested_locked	0.4743ms	0.3404ms	2.9375 KOps/s	2.8848 KOps/s	$\color{#35bf28}+1.83\%$
test_keys	39.8530μs	3.5642μs	280.5713 KOps/s	280.9288 KOps/s	$\color{#d91a1a}-0.13\%$
test_keys_nested	0.2069ms	0.1361ms	7.3470 KOps/s	7.3606 KOps/s	$\color{#d91a1a}-0.18\%$
test_keys_nested_locked	1.7900ms	0.1434ms	6.9745 KOps/s	7.1206 KOps/s	$\color{#d91a1a}-2.05\%$
test_keys_nested_leaf	0.2673ms	0.1175ms	8.5125 KOps/s	8.4796 KOps/s	$\color{#35bf28}+0.39\%$
test_keys_stack_nested	0.2001ms	0.1365ms	7.3244 KOps/s	7.4868 KOps/s	$\color{#d91a1a}-2.17\%$
test_keys_stack_nested_leaf	0.2101ms	0.1176ms	8.5033 KOps/s	8.6686 KOps/s	$\color{#d91a1a}-1.91\%$
test_keys_stack_nested_locked	0.2426ms	0.1422ms	7.0314 KOps/s	7.1727 KOps/s	$\color{#d91a1a}-1.97\%$
test_values	8.7986μs	1.0712μs	933.5466 KOps/s	933.0029 KOps/s	$\color{#35bf28}+0.06\%$
test_values_nested	0.1100ms	55.3382μs	18.0707 KOps/s	17.7850 KOps/s	$\color{#35bf28}+1.61\%$
test_values_nested_locked	0.1086ms	55.3634μs	18.0625 KOps/s	18.0568 KOps/s	$\color{#35bf28}+0.03\%$
test_values_nested_leaf	0.1094ms	60.9151μs	16.4163 KOps/s	16.3443 KOps/s	$\color{#35bf28}+0.44\%$
test_values_stack_nested	0.1078ms	56.9476μs	17.5600 KOps/s	16.2191 KOps/s	$\textbf{\color{#35bf28}+8.27\%}$
test_values_stack_nested_leaf	0.1129ms	61.5324μs	16.2516 KOps/s	16.5554 KOps/s	$\color{#d91a1a}-1.84\%$
test_values_stack_nested_locked	0.1274ms	57.1346μs	17.5025 KOps/s	17.8103 KOps/s	$\color{#d91a1a}-1.73\%$
test_membership	6.5966μs	0.7644μs	1.3083 MOps/s	1.1033 MOps/s	$\textbf{\color{#35bf28}+18.57\%}$
test_membership_nested	35.7070μs	2.7535μs	363.1767 KOps/s	358.5908 KOps/s	$\color{#35bf28}+1.28\%$
test_membership_nested_leaf	44.1730μs	2.7842μs	359.1648 KOps/s	356.7595 KOps/s	$\color{#35bf28}+0.67\%$
test_membership_stacked_nested	54.7400μs	2.7558μs	362.8658 KOps/s	367.1706 KOps/s	$\color{#d91a1a}-1.17\%$
test_membership_stacked_nested_leaf	18.4040μs	2.8250μs	353.9877 KOps/s	363.0385 KOps/s	$\color{#d91a1a}-2.49\%$
test_membership_nested_last	23.2140μs	4.0731μs	245.5109 KOps/s	241.3717 KOps/s	$\color{#35bf28}+1.71\%$
test_membership_nested_leaf_last	24.3860μs	4.1436μs	241.3352 KOps/s	246.4473 KOps/s	$\color{#d91a1a}-2.07\%$
test_membership_stacked_nested_last	46.1980μs	4.6701μs	214.1298 KOps/s	167.2533 KOps/s	$\textbf{\color{#35bf28}+28.03\%}$
test_membership_stacked_nested_leaf_last	35.9670μs	4.7093μs	212.3465 KOps/s	174.8331 KOps/s	$\textbf{\color{#35bf28}+21.46\%}$
test_nested_getleaf	62.5980μs	10.8364μs	92.2812 KOps/s	94.1053 KOps/s	$\color{#d91a1a}-1.94\%$
test_nested_get	49.7330μs	10.2899μs	97.1824 KOps/s	99.1629 KOps/s	$\color{#d91a1a}-2.00\%$
test_stacked_getleaf	50.5630μs	10.6091μs	94.2586 KOps/s	94.8322 KOps/s	$\color{#d91a1a}-0.60\%$
test_stacked_get	47.5610μs	10.0983μs	99.0266 KOps/s	98.8330 KOps/s	$\color{#35bf28}+0.20\%$
test_nested_getitemleaf	54.9410μs	11.0380μs	90.5958 KOps/s	90.2159 KOps/s	$\color{#35bf28}+0.42\%$
test_nested_getitem	0.1365ms	10.3704μs	96.4286 KOps/s	96.2914 KOps/s	$\color{#35bf28}+0.14\%$
test_stacked_getitemleaf	73.3170μs	11.1559μs	89.6384 KOps/s	91.7882 KOps/s	$\color{#d91a1a}-2.34\%$
test_stacked_getitem	34.8050μs	10.4614μs	95.5891 KOps/s	96.6378 KOps/s	$\color{#d91a1a}-1.09\%$
test_lock_nested	3.1877ms	0.4423ms	2.2610 KOps/s	1.8688 KOps/s	$\textbf{\color{#35bf28}+20.99\%}$
test_lock_stack_nested	0.9089ms	0.4122ms	2.4258 KOps/s	2.4737 KOps/s	$\color{#d91a1a}-1.94\%$
test_unlock_nested	0.6911ms	0.3536ms	2.8283 KOps/s	2.7852 KOps/s	$\color{#35bf28}+1.55\%$
test_unlock_stack_nested	0.5609ms	0.3281ms	3.0476 KOps/s	3.0758 KOps/s	$\color{#d91a1a}-0.92\%$
test_flatten_speed	0.1608ms	91.2894μs	10.9542 KOps/s	10.7514 KOps/s	$\color{#35bf28}+1.89\%$
test_unflatten_speed	0.8281ms	0.4728ms	2.1151 KOps/s	2.0547 KOps/s	$\color{#35bf28}+2.94\%$
test_common_ops	3.6292ms	0.7205ms	1.3880 KOps/s	1.3144 KOps/s	$\textbf{\color{#35bf28}+5.60\%}$
test_creation	33.6330μs	2.0400μs	490.1978 KOps/s	473.5704 KOps/s	$\color{#35bf28}+3.51\%$
test_creation_empty	38.6420μs	8.6075μs	116.1782 KOps/s	96.6443 KOps/s	$\textbf{\color{#35bf28}+20.21\%}$
test_creation_nested_1	48.0900μs	11.2476μs	88.9082 KOps/s	78.4437 KOps/s	$\textbf{\color{#35bf28}+13.34\%}$
test_creation_nested_2	66.6050μs	15.6330μs	63.9673 KOps/s	57.3822 KOps/s	$\textbf{\color{#35bf28}+11.48\%}$
test_clone	0.1185ms	12.9312μs	77.3322 KOps/s	78.0149 KOps/s	$\color{#d91a1a}-0.88\%$
test_getitem[int]	1.4990ms	12.4798μs	80.1292 KOps/s	81.6276 KOps/s	$\color{#d91a1a}-1.84\%$
test_getitem[slice_int]	0.1390ms	24.4772μs	40.8544 KOps/s	41.9955 KOps/s	$\color{#d91a1a}-2.72\%$
test_getitem[range]	0.2631ms	48.2659μs	20.7186 KOps/s	20.8050 KOps/s	$\color{#d91a1a}-0.42\%$
test_getitem[tuple]	0.1389ms	19.9232μs	50.1927 KOps/s	50.9505 KOps/s	$\color{#d91a1a}-1.49\%$
test_getitem[list]	0.4124ms	43.7978μs	22.8322 KOps/s	23.2349 KOps/s	$\color{#d91a1a}-1.73\%$
test_setitem_dim[int]	84.8380μs	26.9490μs	37.1071 KOps/s	38.7809 KOps/s	$\color{#d91a1a}-4.32\%$
test_setitem_dim[slice_int]	0.1034ms	52.3427μs	19.1049 KOps/s	19.4999 KOps/s	$\color{#d91a1a}-2.03\%$
test_setitem_dim[range]	0.1270ms	73.7729μs	13.5551 KOps/s	13.3982 KOps/s	$\color{#35bf28}+1.17\%$
test_setitem_dim[tuple]	80.5610μs	40.4905μs	24.6971 KOps/s	24.6975 KOps/s	$-0.00\%$
test_setitem	0.1333ms	18.5675μs	53.8576 KOps/s	51.4862 KOps/s	$\color{#35bf28}+4.61\%$
test_set	0.1571ms	17.9651μs	55.6636 KOps/s	53.2340 KOps/s	$\color{#35bf28}+4.56\%$
test_set_shared	2.0953ms	0.1716ms	5.8266 KOps/s	5.9438 KOps/s	$\color{#d91a1a}-1.97\%$
test_update	0.1125ms	19.3383μs	51.7109 KOps/s	46.7011 KOps/s	$\textbf{\color{#35bf28}+10.73\%}$
test_update_nested	92.1950μs	27.9288μs	35.8053 KOps/s	32.3746 KOps/s	$\textbf{\color{#35bf28}+10.60\%}$
test_update__nested	0.8732ms	32.0801μs	31.1719 KOps/s	30.7899 KOps/s	$\color{#35bf28}+1.24\%$
test_set_nested	84.6190μs	19.9974μs	50.0066 KOps/s	46.7800 KOps/s	$\textbf{\color{#35bf28}+6.90\%}$
test_set_nested_new	80.9320μs	24.0272μs	41.6196 KOps/s	38.4121 KOps/s	$\textbf{\color{#35bf28}+8.35\%}$
test_select	99.6650μs	39.0106μs	25.6340 KOps/s	24.1011 KOps/s	$\textbf{\color{#35bf28}+6.36\%}$
test_select_nested	0.1193ms	59.8001μs	16.7224 KOps/s	16.8722 KOps/s	$\color{#d91a1a}-0.89\%$
test_exclude_nested	0.1464ms	75.1846μs	13.3006 KOps/s	13.4078 KOps/s	$\color{#d91a1a}-0.80\%$
test_empty[True]	0.7040ms	0.3510ms	2.8491 KOps/s	2.8504 KOps/s	$\color{#d91a1a}-0.04\%$
test_empty[False]	11.0807μs	1.2122μs	824.9398 KOps/s	820.7680 KOps/s	$\color{#35bf28}+0.51\%$
test_unbind_speed	0.3980ms	0.2618ms	3.8198 KOps/s	3.8462 KOps/s	$\color{#d91a1a}-0.68\%$
test_unbind_speed_stack0	0.5162ms	0.2582ms	3.8726 KOps/s	3.9714 KOps/s	$\color{#d91a1a}-2.49\%$
test_unbind_speed_stack1	0.1041s	0.7616ms	1.3131 KOps/s	1.4569 KOps/s	$\textbf{\color{#d91a1a}-9.87\%}$
test_split	0.1010s	1.7213ms	580.9665 Ops/s	570.1472 Ops/s	$\color{#35bf28}+1.90\%$
test_chunk	0.1084s	1.7371ms	575.6673 Ops/s	571.5657 Ops/s	$\color{#35bf28}+0.72\%$
test_consolidate_njt[False-None]	8.5916ms	8.0625ms	124.0312 Ops/s	120.7531 Ops/s	$\color{#35bf28}+2.71\%$
test_creation[device0]	0.2305ms	89.3481μs	11.1922 KOps/s	10.3730 KOps/s	$\textbf{\color{#35bf28}+7.90\%}$
test_creation_from_tensor	0.2232ms	92.3374μs	10.8299 KOps/s	10.5555 KOps/s	$\color{#35bf28}+2.60\%$
test_add_one[memmap_tensor0]	0.1224ms	5.0934μs	196.3338 KOps/s	197.7421 KOps/s	$\color{#d91a1a}-0.71\%$
test_contiguous[memmap_tensor0]	10.3290μs	0.5195μs	1.9249 MOps/s	1.8639 MOps/s	$\color{#35bf28}+3.27\%$
test_stack[memmap_tensor0]	49.4820μs	3.4257μs	291.9140 KOps/s	271.3112 KOps/s	$\textbf{\color{#35bf28}+7.59\%}$
test_memmaptd_index	1.1471ms	0.2372ms	4.2161 KOps/s	4.1704 KOps/s	$\color{#35bf28}+1.10\%$
test_memmaptd_index_astensor	0.5691ms	0.3138ms	3.1863 KOps/s	3.1371 KOps/s	$\color{#35bf28}+1.57\%$
test_memmaptd_index_op	1.0379ms	0.5531ms	1.8081 KOps/s	1.7321 KOps/s	$\color{#35bf28}+4.39\%$
test_serialize_model	0.1369s	0.1196s	8.3635 Ops/s	7.3003 Ops/s	$\textbf{\color{#35bf28}+14.56\%}$
test_serialize_model_pickle	0.4685s	0.4000s	2.5003 Ops/s	2.4933 Ops/s	$\color{#35bf28}+0.28\%$
test_serialize_weights	0.2150s	0.1284s	7.7897 Ops/s	8.7086 Ops/s	$\textbf{\color{#d91a1a}-10.55\%}$
test_serialize_weights_returnearly	0.1792s	0.1628s	6.1426 Ops/s	6.2516 Ops/s	$\color{#d91a1a}-1.74\%$
test_serialize_weights_pickle	0.4898s	0.4401s	2.2723 Ops/s	2.2923 Ops/s	$\color{#d91a1a}-0.87\%$
test_serialize_weights_filesystem	0.1448s	0.1402s	7.1343 Ops/s	7.1491 Ops/s	$\color{#d91a1a}-0.21\%$
test_serialize_model_filesystem	0.2401s	0.1610s	6.2095 Ops/s	5.8713 Ops/s	$\textbf{\color{#35bf28}+5.76\%}$
test_reshape_pytree	76.5930μs	26.5389μs	37.6805 KOps/s	37.3538 KOps/s	$\color{#35bf28}+0.87\%$
test_reshape_td	74.8900μs	32.3596μs	30.9027 KOps/s	31.1944 KOps/s	$\color{#d91a1a}-0.94\%$
test_view_pytree	78.2260μs	26.6012μs	37.5923 KOps/s	37.0224 KOps/s	$\color{#35bf28}+1.54\%$
test_view_td	0.1145ms	38.8238μs	25.7574 KOps/s	26.2810 KOps/s	$\color{#d91a1a}-1.99\%$
test_unbind_pytree	80.4910μs	30.5181μs	32.7675 KOps/s	33.2225 KOps/s	$\color{#d91a1a}-1.37\%$
test_unbind_td	0.3360ms	38.5822μs	25.9187 KOps/s	26.0543 KOps/s	$\color{#d91a1a}-0.52\%$
test_split_pytree	85.6210μs	29.9990μs	33.3345 KOps/s	33.4925 KOps/s	$\color{#d91a1a}-0.47\%$
test_split_td	0.1983ms	44.3526μs	22.5466 KOps/s	22.8950 KOps/s	$\color{#d91a1a}-1.52\%$
test_add_pytree	77.0140μs	35.8208μs	27.9167 KOps/s	27.4896 KOps/s	$\color{#35bf28}+1.55\%$
test_add_td	0.1428ms	53.0205μs	18.8606 KOps/s	17.0826 KOps/s	$\textbf{\color{#35bf28}+10.41\%}$
test_compile_add_one_nested[tensordict-compile]	0.1250ms	61.9784μs	16.1347 KOps/s	16.3361 KOps/s	$\color{#d91a1a}-1.23\%$
test_compile_add_one_nested[tensordict-eager]	0.3630ms	0.1622ms	6.1638 KOps/s	6.2653 KOps/s	$\color{#d91a1a}-1.62\%$
test_compile_add_one_nested[pytree-compile]	93.2240μs	45.4058μs	22.0236 KOps/s	21.6488 KOps/s	$\color{#35bf28}+1.73\%$
test_compile_add_one_nested[pytree-eager]	0.2264ms	0.1183ms	8.4529 KOps/s	8.3188 KOps/s	$\color{#35bf28}+1.61\%$
test_compile_copy_nested[tensordict-compile]	61.4350μs	26.1366μs	38.2605 KOps/s	39.3128 KOps/s	$\color{#d91a1a}-2.68\%$
test_compile_copy_nested[tensordict-eager]	0.1032ms	53.5781μs	18.6643 KOps/s	18.4940 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_copy_nested[pytree-compile]	0.1615ms	78.9573μs	12.6651 KOps/s	12.6957 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_copy_nested[pytree-eager]	0.1409ms	68.3646μs	14.6274 KOps/s	14.9449 KOps/s	$\color{#d91a1a}-2.12\%$
test_compile_add_one_flat[tensordict-compile]	0.1928ms	0.1040ms	9.6120 KOps/s	9.5595 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_add_one_flat[tensordict-eager]	0.3776ms	0.1983ms	5.0418 KOps/s	5.0574 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_add_one_flat[tensorclass-compile]	0.1065ms	44.3886μs	22.5283 KOps/s	21.4609 KOps/s	$\color{#35bf28}+4.97\%$
test_compile_add_one_flat[tensorclass-eager]	0.5210ms	61.4678μs	16.2687 KOps/s	16.4049 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_add_one_flat[pytree-compile]	0.1917ms	0.1027ms	9.7409 KOps/s	9.5889 KOps/s	$\color{#35bf28}+1.58\%$
test_compile_add_one_flat[pytree-eager]	0.4380ms	0.2005ms	4.9867 KOps/s	4.8914 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_add_self_flat[tensordict-eager]	0.4275ms	0.2101ms	4.7595 KOps/s	4.7407 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_add_self_flat[tensordict-compile]	0.2248ms	0.1043ms	9.5848 KOps/s	9.4137 KOps/s	$\color{#35bf28}+1.82\%$
test_compile_add_self_flat[tensorclass-eager]	0.1171ms	55.9406μs	17.8761 KOps/s	18.7188 KOps/s	$\color{#d91a1a}-4.50\%$
test_compile_add_self_flat[tensorclass-compile]	0.1518ms	47.6659μs	20.9794 KOps/s	21.7983 KOps/s	$\color{#d91a1a}-3.76\%$
test_compile_add_self_flat[pytree-eager]	0.3361ms	0.1593ms	6.2777 KOps/s	6.2643 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_add_self_flat[pytree-compile]	0.2049ms	0.1032ms	9.6854 KOps/s	9.6861 KOps/s	$-0.01\%$
test_compile_copy_flat[tensordict-compile]	66.0740μs	21.2060μs	47.1566 KOps/s	47.5589 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_copy_flat[tensordict-eager]	0.1609ms	59.2666μs	16.8729 KOps/s	17.0928 KOps/s	$\color{#d91a1a}-1.29\%$
test_compile_copy_flat[pytree-compile]	0.1785ms	81.4238μs	12.2814 KOps/s	12.1028 KOps/s	$\color{#35bf28}+1.48\%$
test_compile_copy_flat[pytree-eager]	0.1452ms	69.5699μs	14.3740 KOps/s	14.2666 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_assign_and_add[tensordict-compile]	0.2846ms	0.2093ms	4.7784 KOps/s	4.8158 KOps/s	$\color{#d91a1a}-0.78\%$
test_compile_assign_and_add[tensordict-eager]	2.1704ms	1.2892ms	775.6659 Ops/s	797.7010 Ops/s	$\color{#d91a1a}-2.76\%$
test_compile_assign_and_add[pytree-compile]	0.4645ms	0.2020ms	4.9502 KOps/s	4.9108 KOps/s	$\color{#35bf28}+0.80\%$
test_compile_assign_and_add[pytree-eager]	0.9369ms	0.7760ms	1.2887 KOps/s	1.2635 KOps/s	$\color{#35bf28}+2.00\%$
test_compile_assign_and_add_stack[compile]	0.5547ms	0.4583ms	2.1819 KOps/s	2.1519 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_assign_and_add_stack[eager]	4.0979ms	2.4709ms	404.7189 Ops/s	390.3909 Ops/s	$\color{#35bf28}+3.67\%$
test_compile_indexing[tensor-tensordict-compile]	0.1469ms	35.0629μs	28.5201 KOps/s	27.3903 KOps/s	$\color{#35bf28}+4.12\%$
test_compile_indexing[tensor-tensordict-eager]	0.8343ms	32.2903μs	30.9690 KOps/s	29.1441 KOps/s	$\textbf{\color{#35bf28}+6.26\%}$
test_compile_indexing[tensor-tensorclass-compile]	73.8780μs	28.7864μs	34.7386 KOps/s	32.9911 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_compile_indexing[tensor-tensorclass-eager]	78.6870μs	23.1633μs	43.1718 KOps/s	41.5548 KOps/s	$\color{#35bf28}+3.89\%$
test_compile_indexing[tensor-pytree-compile]	0.1077ms	29.9647μs	33.3726 KOps/s	32.2818 KOps/s	$\color{#35bf28}+3.38\%$
test_compile_indexing[tensor-pytree-eager]	79.8910μs	23.2733μs	42.9677 KOps/s	42.0353 KOps/s	$\color{#35bf28}+2.22\%$
test_compile_indexing[slice-tensordict-compile]	0.2393ms	52.4659μs	19.0600 KOps/s	18.8002 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_indexing[slice-tensordict-eager]	0.6119ms	19.4000μs	51.5464 KOps/s	51.0506 KOps/s	$\color{#35bf28}+0.97\%$
test_compile_indexing[slice-tensorclass-compile]	0.1260ms	44.1963μs	22.6263 KOps/s	21.9257 KOps/s	$\color{#35bf28}+3.20\%$
test_compile_indexing[slice-tensorclass-eager]	63.4290μs	18.7204μs	53.4177 KOps/s	51.9935 KOps/s	$\color{#35bf28}+2.74\%$
test_compile_indexing[slice-pytree-compile]	0.1275ms	45.1512μs	22.1478 KOps/s	21.2524 KOps/s	$\color{#35bf28}+4.21\%$
test_compile_indexing[slice-pytree-eager]	0.1709ms	18.9873μs	52.6667 KOps/s	52.3329 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_indexing[int-tensordict-compile]	0.1061ms	53.4934μs	18.6939 KOps/s	18.3612 KOps/s	$\color{#35bf28}+1.81\%$
test_compile_indexing[int-tensordict-eager]	1.0014ms	19.4722μs	51.3553 KOps/s	51.1196 KOps/s	$\color{#35bf28}+0.46\%$
test_compile_indexing[int-tensorclass-compile]	0.1048ms	45.4898μs	21.9829 KOps/s	21.2585 KOps/s	$\color{#35bf28}+3.41\%$
test_compile_indexing[int-tensorclass-eager]	78.8580μs	19.0196μs	52.5772 KOps/s	52.7284 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_indexing[int-pytree-compile]	0.1214ms	45.9088μs	21.7823 KOps/s	21.1854 KOps/s	$\color{#35bf28}+2.82\%$
test_compile_indexing[int-pytree-eager]	85.0580μs	18.8050μs	53.1774 KOps/s	52.0474 KOps/s	$\color{#35bf28}+2.17\%$
test_mod_add[eager]	83.1660μs	23.6258μs	42.3266 KOps/s	36.8722 KOps/s	$\textbf{\color{#35bf28}+14.79\%}$
test_mod_add[compile]	0.1213ms	44.6156μs	22.4137 KOps/s	21.2462 KOps/s	$\textbf{\color{#35bf28}+5.50\%}$
test_mod_add[compile-overhead]	0.1116ms	45.2132μs	22.1174 KOps/s	21.5231 KOps/s	$\color{#35bf28}+2.76\%$
test_mod_wrap[eager]	0.3500ms	0.2066ms	4.8404 KOps/s	4.5925 KOps/s	$\textbf{\color{#35bf28}+5.40\%}$
test_mod_wrap[compile]	1.9887ms	0.2051ms	4.8757 KOps/s	4.7373 KOps/s	$\color{#35bf28}+2.92\%$
test_mod_wrap[compile-overhead]	1.8411ms	0.2053ms	4.8703 KOps/s	4.7732 KOps/s	$\color{#35bf28}+2.03\%$
test_mod_wrap_and_backward[eager]	14.2706ms	11.5741ms	86.3997 Ops/s	91.0004 Ops/s	$\textbf{\color{#d91a1a}-5.06\%}$
test_mod_wrap_and_backward[compile]	16.4117ms	12.3571ms	80.9249 Ops/s	78.1459 Ops/s	$\color{#35bf28}+3.56\%$
test_mod_wrap_and_backward[compile-overhead]	15.8766ms	12.5505ms	79.6781 Ops/s	80.9790 Ops/s	$\color{#d91a1a}-1.61\%$
test_seq_add[eager]	0.1536ms	84.1887μs	11.8781 KOps/s	10.9720 KOps/s	$\textbf{\color{#35bf28}+8.26\%}$
test_seq_add[compile]	0.1170ms	60.7357μs	16.4648 KOps/s	16.4877 KOps/s	$\color{#d91a1a}-0.14\%$
test_seq_add[compile-overhead]	0.1257ms	57.4094μs	17.4188 KOps/s	16.9505 KOps/s	$\color{#35bf28}+2.76\%$
test_seq_wrap[eager]	0.4981ms	0.3713ms	2.6931 KOps/s	2.6160 KOps/s	$\color{#35bf28}+2.95\%$
test_seq_wrap[compile]	0.4055ms	0.2263ms	4.4194 KOps/s	4.3571 KOps/s	$\color{#35bf28}+1.43\%$
test_seq_wrap[compile-overhead]	0.4115ms	0.2286ms	4.3753 KOps/s	4.4520 KOps/s	$\color{#d91a1a}-1.72\%$
test_func_call_runtime[False-eager]	0.8134ms	0.5437ms	1.8394 KOps/s	1.8337 KOps/s	$\color{#35bf28}+0.31\%$
test_func_call_runtime[False-compile]	1.0050ms	0.4359ms	2.2939 KOps/s	2.3131 KOps/s	$\color{#d91a1a}-0.83\%$
test_func_call_runtime[False-compile-overhead]	0.5515ms	0.4345ms	2.3015 KOps/s	2.3264 KOps/s	$\color{#d91a1a}-1.07\%$
test_func_call_runtime[True-eager]	1.0731ms	0.7520ms	1.3298 KOps/s	1.3365 KOps/s	$\color{#d91a1a}-0.50\%$
test_func_call_runtime[True-compile]	0.7070ms	0.4703ms	2.1262 KOps/s	2.1326 KOps/s	$\color{#d91a1a}-0.30\%$
test_func_call_runtime[True-compile-overhead]	0.9219ms	0.4773ms	2.0951 KOps/s	2.1348 KOps/s	$\color{#d91a1a}-1.86\%$
test_func_call_cm_runtime[False-eager]	0.8577ms	0.5410ms	1.8486 KOps/s	1.8256 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_cm_runtime[False-compile]	0.7677ms	0.4354ms	2.2967 KOps/s	2.3025 KOps/s	$\color{#d91a1a}-0.25\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6917ms	0.4335ms	2.3070 KOps/s	2.3257 KOps/s	$\color{#d91a1a}-0.80\%$
test_func_call_cm_runtime[True-eager]	1.4463ms	0.8967ms	1.1152 KOps/s	1.1380 KOps/s	$\color{#d91a1a}-2.01\%$
test_func_call_cm_runtime[True-compile]	0.5912ms	0.4920ms	2.0327 KOps/s	2.0324 KOps/s	$\color{#35bf28}+0.01\%$
test_func_call_cm_runtime[True-compile-overhead]	0.9082ms	0.4998ms	2.0008 KOps/s	2.0303 KOps/s	$\color{#d91a1a}-1.45\%$
test_vmap_func_call_cm_runtime[eager]	3.1259ms	1.8923ms	528.4584 Ops/s	524.5319 Ops/s	$\color{#35bf28}+0.75\%$
test_vmap_func_call_cm_runtime[compile]	0.6738ms	0.5236ms	1.9098 KOps/s	1.8552 KOps/s	$\color{#35bf28}+2.94\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9001ms	0.5284ms	1.8924 KOps/s	1.9459 KOps/s	$\color{#d91a1a}-2.75\%$
test_distributed	0.3585ms	0.1263ms	7.9196 KOps/s	7.6640 KOps/s	$\color{#35bf28}+3.34\%$
test_tdmodule	52.5790μs	16.9899μs	58.8586 KOps/s	52.6840 KOps/s	$\textbf{\color{#35bf28}+11.72\%}$
test_tdmodule_dispatch	55.1630μs	32.7197μs	30.5626 KOps/s	27.2051 KOps/s	$\textbf{\color{#35bf28}+12.34\%}$
test_tdseq	48.8610μs	18.9425μs	52.7913 KOps/s	47.3859 KOps/s	$\textbf{\color{#35bf28}+11.41\%}$
test_tdseq_dispatch	80.0900μs	37.3074μs	26.8043 KOps/s	24.3984 KOps/s	$\textbf{\color{#35bf28}+9.86\%}$
test_instantiation_functorch	1.8485ms	1.5241ms	656.1327 Ops/s	623.9535 Ops/s	$\textbf{\color{#35bf28}+5.16\%}$
test_exec_functorch	0.2814ms	0.1795ms	5.5708 KOps/s	5.5443 KOps/s	$\color{#35bf28}+0.48\%$
test_exec_functional_call	0.3019ms	0.1688ms	5.9256 KOps/s	5.7171 KOps/s	$\color{#35bf28}+3.65\%$
test_exec_td_decorator	0.5559ms	0.2270ms	4.4051 KOps/s	4.3716 KOps/s	$\color{#35bf28}+0.77\%$
test_vmap_mlp_speed_decorator[True-True]	0.7757ms	0.6363ms	1.5716 KOps/s	1.5262 KOps/s	$\color{#35bf28}+2.97\%$
test_vmap_mlp_speed_decorator[True-False]	1.0182ms	0.6359ms	1.5726 KOps/s	1.5632 KOps/s	$\color{#35bf28}+0.60\%$
test_vmap_mlp_speed_decorator[False-True]	0.8377ms	0.5289ms	1.8906 KOps/s	1.8761 KOps/s	$\color{#35bf28}+0.77\%$
test_vmap_mlp_speed_decorator[False-False]	0.7360ms	0.5275ms	1.8957 KOps/s	1.8774 KOps/s	$\color{#35bf28}+0.98\%$
test_to_module_speed[True]	2.0682ms	1.3188ms	758.2804 Ops/s	784.2784 Ops/s	$\color{#d91a1a}-3.31\%$
test_to_module_speed[False]	1.8460ms	1.2924ms	773.7269 Ops/s	807.2439 Ops/s	$\color{#d91a1a}-4.15\%$
test_tc_init	0.1080ms	42.2898μs	23.6464 KOps/s	21.8100 KOps/s	$\textbf{\color{#35bf28}+8.42\%}$
test_tc_init_nested	0.1485ms	83.2507μs	12.0119 KOps/s	10.8806 KOps/s	$\textbf{\color{#35bf28}+10.40\%}$
test_tc_first_layer_tensor	24.4850μs	1.5322μs	652.6514 KOps/s	670.8925 KOps/s	$\color{#d91a1a}-2.72\%$
test_tc_first_layer_nontensor	59.0710μs	4.6135μs	216.7535 KOps/s	216.5018 KOps/s	$\color{#35bf28}+0.12\%$
test_tc_second_layer_tensor	31.7590μs	2.8396μs	352.1655 KOps/s	365.4092 KOps/s	$\color{#d91a1a}-3.62\%$
test_tc_second_layer_nontensor	50.2440μs	5.9646μs	167.6545 KOps/s	168.6621 KOps/s	$\color{#d91a1a}-0.60\%$
test_unbind	0.2416s	13.0820ms	76.4408 Ops/s	81.7615 Ops/s	$\textbf{\color{#d91a1a}-6.51\%}$
test_full_like	18.0985ms	13.8667ms	72.1154 Ops/s	122.0499 Ops/s	$\textbf{\color{#d91a1a}-40.91\%}$
test_zeros_like	14.3647ms	8.0280ms	124.5641 Ops/s	335.6527 Ops/s	$\textbf{\color{#d91a1a}-62.89\%}$
test_ones_like	13.7291ms	7.7377ms	129.2366 Ops/s	284.4030 Ops/s	$\textbf{\color{#d91a1a}-54.56\%}$
test_clone	14.1399ms	9.6139ms	104.0162 Ops/s	177.1764 Ops/s	$\textbf{\color{#d91a1a}-41.29\%}$
test_squeeze	60.0730μs	12.4198μs	80.5164 KOps/s	84.3989 KOps/s	$\color{#d91a1a}-4.60\%$
test_unsqueeze	0.1778ms	88.8223μs	11.2584 KOps/s	11.2391 KOps/s	$\color{#35bf28}+0.17\%$
test_split	0.5698ms	0.1906ms	5.2472 KOps/s	5.3094 KOps/s	$\color{#d91a1a}-1.17\%$
test_permute	0.4662ms	0.2241ms	4.4614 KOps/s	4.5828 KOps/s	$\color{#d91a1a}-2.65\%$
test_stack	30.2855ms	27.8996ms	35.8428 Ops/s	37.5281 Ops/s	$\color{#d91a1a}-4.49\%$
test_cat	31.6475ms	27.7583ms	36.0253 Ops/s	37.7820 Ops/s	$\color{#d91a1a}-4.65\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}26$. Worsened: $\large\color{#d91a1a}8$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	35.9720μs	10.8118μs	92.4916 KOps/s	86.7898 KOps/s	$\textbf{\color{#35bf28}+6.57\%}$
test_plain_set_stack_nested	31.7610μs	10.8716μs	91.9830 KOps/s	86.4394 KOps/s	$\textbf{\color{#35bf28}+6.41\%}$
test_plain_set_nested_inplace	40.7020μs	11.7402μs	85.1775 KOps/s	80.7435 KOps/s	$\textbf{\color{#35bf28}+5.49\%}$
test_plain_set_stack_nested_inplace	39.3920μs	11.6706μs	85.6857 KOps/s	80.4588 KOps/s	$\textbf{\color{#35bf28}+6.50\%}$
test_items	33.7620μs	2.9193μs	342.5439 KOps/s	345.3822 KOps/s	$\color{#d91a1a}-0.82\%$
test_items_nested	0.4724ms	0.3187ms	3.1373 KOps/s	3.1507 KOps/s	$\color{#d91a1a}-0.43\%$
test_items_nested_locked	0.3769ms	0.3214ms	3.1115 KOps/s	3.1430 KOps/s	$\color{#d91a1a}-1.00\%$
test_items_nested_leaf	84.5850μs	58.9570μs	16.9615 KOps/s	16.8843 KOps/s	$\color{#35bf28}+0.46\%$
test_items_stack_nested	0.3813ms	0.3208ms	3.1170 KOps/s	3.1373 KOps/s	$\color{#d91a1a}-0.65\%$
test_items_stack_nested_leaf	93.9040μs	60.5336μs	16.5197 KOps/s	17.1711 KOps/s	$\color{#d91a1a}-3.79\%$
test_items_stack_nested_locked	0.4003ms	0.3230ms	3.0956 KOps/s	3.0951 KOps/s	$\color{#35bf28}+0.02\%$
test_keys	75.4540μs	3.4690μs	288.2635 KOps/s	289.2965 KOps/s	$\color{#d91a1a}-0.36\%$
test_keys_nested	0.1058ms	69.4865μs	14.3913 KOps/s	14.1941 KOps/s	$\color{#35bf28}+1.39\%$
test_keys_nested_locked	2.9828ms	75.1597μs	13.3050 KOps/s	13.1283 KOps/s	$\color{#35bf28}+1.35\%$
test_keys_nested_leaf	93.6340μs	61.5464μs	16.2479 KOps/s	16.1823 KOps/s	$\color{#35bf28}+0.41\%$
test_keys_stack_nested	0.1063ms	70.9397μs	14.0965 KOps/s	14.1673 KOps/s	$\color{#d91a1a}-0.50\%$
test_keys_stack_nested_leaf	95.0040μs	62.0630μs	16.1127 KOps/s	16.2620 KOps/s	$\color{#d91a1a}-0.92\%$
test_keys_stack_nested_locked	0.1071ms	76.0625μs	13.1471 KOps/s	13.1942 KOps/s	$\color{#d91a1a}-0.36\%$
test_values	5.5070μs	0.8440μs	1.1848 MOps/s	1.1826 MOps/s	$\color{#35bf28}+0.19\%$
test_values_nested	67.3030μs	31.2155μs	32.0353 KOps/s	32.1778 KOps/s	$\color{#d91a1a}-0.44\%$
test_values_nested_locked	63.4430μs	32.9622μs	30.3378 KOps/s	30.3160 KOps/s	$\color{#35bf28}+0.07\%$
test_values_nested_leaf	65.6530μs	34.1276μs	29.3018 KOps/s	29.6468 KOps/s	$\color{#d91a1a}-1.16\%$
test_values_stack_nested	66.3630μs	31.7251μs	31.5208 KOps/s	32.2408 KOps/s	$\color{#d91a1a}-2.23\%$
test_values_stack_nested_leaf	63.3130μs	34.5730μs	28.9243 KOps/s	29.7626 KOps/s	$\color{#d91a1a}-2.82\%$
test_values_stack_nested_locked	56.9130μs	33.3552μs	29.9804 KOps/s	30.2861 KOps/s	$\color{#d91a1a}-1.01\%$
test_membership	1.9366μs	0.5177μs	1.9318 MOps/s	1.9474 MOps/s	$\color{#d91a1a}-0.80\%$
test_membership_nested	13.3705μs	1.8878μs	529.7093 KOps/s	523.3712 KOps/s	$\color{#35bf28}+1.21\%$
test_membership_nested_leaf	18.7655μs	1.9034μs	525.3745 KOps/s	515.1535 KOps/s	$\color{#35bf28}+1.98\%$
test_membership_stacked_nested	37.9020μs	1.9607μs	510.0094 KOps/s	493.2736 KOps/s	$\color{#35bf28}+3.39\%$
test_membership_stacked_nested_leaf	37.6920μs	1.9603μs	510.1242 KOps/s	500.4893 KOps/s	$\color{#35bf28}+1.93\%$
test_membership_nested_last	30.2510μs	2.7858μs	358.9674 KOps/s	350.6043 KOps/s	$\color{#35bf28}+2.39\%$
test_membership_nested_leaf_last	35.3510μs	2.8264μs	353.8050 KOps/s	347.5504 KOps/s	$\color{#35bf28}+1.80\%$
test_membership_stacked_nested_last	38.6620μs	3.2367μs	308.9601 KOps/s	350.1974 KOps/s	$\textbf{\color{#d91a1a}-11.78\%}$
test_membership_stacked_nested_leaf_last	34.3720μs	3.2596μs	306.7880 KOps/s	351.8986 KOps/s	$\textbf{\color{#d91a1a}-12.82\%}$
test_nested_getleaf	36.6020μs	6.0335μs	165.7411 KOps/s	167.5535 KOps/s	$\color{#d91a1a}-1.08\%$
test_nested_get	35.9520μs	5.7097μs	175.1418 KOps/s	175.0355 KOps/s	$\color{#35bf28}+0.06\%$
test_stacked_getleaf	36.0020μs	6.0002μs	166.6619 KOps/s	166.5239 KOps/s	$\color{#35bf28}+0.08\%$
test_stacked_get	45.4320μs	5.7026μs	175.3581 KOps/s	175.5316 KOps/s	$\color{#d91a1a}-0.10\%$
test_nested_getitemleaf	29.9720μs	6.1613μs	162.3031 KOps/s	164.5527 KOps/s	$\color{#d91a1a}-1.37\%$
test_nested_getitem	37.9810μs	5.8117μs	172.0665 KOps/s	171.8239 KOps/s	$\color{#35bf28}+0.14\%$
test_stacked_getitemleaf	39.3720μs	6.1178μs	163.4580 KOps/s	163.8504 KOps/s	$\color{#d91a1a}-0.24\%$
test_stacked_getitem	41.9920μs	5.7567μs	173.7120 KOps/s	171.9548 KOps/s	$\color{#35bf28}+1.02\%$
test_lock_nested	7.2502ms	0.3725ms	2.6844 KOps/s	2.7134 KOps/s	$\color{#d91a1a}-1.07\%$
test_lock_stack_nested	0.3913ms	0.3333ms	3.0001 KOps/s	2.9951 KOps/s	$\color{#35bf28}+0.17\%$
test_unlock_nested	0.7503ms	0.3054ms	3.2745 KOps/s	3.3178 KOps/s	$\color{#d91a1a}-1.30\%$
test_unlock_stack_nested	0.3347ms	0.2742ms	3.6463 KOps/s	3.7026 KOps/s	$\color{#d91a1a}-1.52\%$
test_flatten_speed	0.1108ms	73.3224μs	13.6384 KOps/s	13.6665 KOps/s	$\color{#d91a1a}-0.21\%$
test_unflatten_speed	0.3836ms	0.2950ms	3.3902 KOps/s	3.4151 KOps/s	$\color{#d91a1a}-0.73\%$
test_common_ops	1.7696ms	0.5800ms	1.7241 KOps/s	1.6797 KOps/s	$\color{#35bf28}+2.64\%$
test_creation	0.1069ms	1.4751μs	677.9322 KOps/s	685.2447 KOps/s	$\color{#d91a1a}-1.07\%$
test_creation_empty	50.3430μs	7.7966μs	128.2605 KOps/s	107.9885 KOps/s	$\textbf{\color{#35bf28}+18.77\%}$
test_creation_nested_1	38.7520μs	9.2284μs	108.3611 KOps/s	93.1911 KOps/s	$\textbf{\color{#35bf28}+16.28\%}$
test_creation_nested_2	51.7420μs	11.8817μs	84.1629 KOps/s	76.2559 KOps/s	$\textbf{\color{#35bf28}+10.37\%}$
test_clone	56.4430μs	9.6631μs	103.4863 KOps/s	104.3942 KOps/s	$\color{#d91a1a}-0.87\%$
test_getitem[int]	92.3143ms	15.4716μs	64.6344 KOps/s	94.7482 KOps/s	$\textbf{\color{#d91a1a}-31.78\%}$
test_getitem[slice_int]	0.1105ms	20.6032μs	48.5362 KOps/s	48.5374 KOps/s	$-0.00\%$
test_getitem[range]	0.1326ms	36.0692μs	27.7245 KOps/s	27.6920 KOps/s	$\color{#35bf28}+0.12\%$
test_getitem[tuple]	0.1042ms	18.0521μs	55.3951 KOps/s	54.4916 KOps/s	$\color{#35bf28}+1.66\%$
test_getitem[list]	0.2361ms	31.7664μs	31.4798 KOps/s	31.2975 KOps/s	$\color{#35bf28}+0.58\%$
test_setitem_dim[int]	25.0010μs	17.0566μs	58.6283 KOps/s	57.1178 KOps/s	$\color{#35bf28}+2.64\%$
test_setitem_dim[slice_int]	59.5840μs	35.5149μs	28.1572 KOps/s	28.2020 KOps/s	$\color{#d91a1a}-0.16\%$
test_setitem_dim[range]	82.5040μs	51.5448μs	19.4006 KOps/s	19.5055 KOps/s	$\color{#d91a1a}-0.54\%$
test_setitem_dim[tuple]	57.1830μs	29.2575μs	34.1793 KOps/s	33.1783 KOps/s	$\color{#35bf28}+3.02\%$
test_setitem	81.6350μs	13.5130μs	74.0027 KOps/s	67.7075 KOps/s	$\textbf{\color{#35bf28}+9.30\%}$
test_set	92.1950μs	13.3198μs	75.0762 KOps/s	71.0424 KOps/s	$\textbf{\color{#35bf28}+5.68\%}$
test_set_shared	1.6487ms	0.1436ms	6.9649 KOps/s	6.9537 KOps/s	$\color{#35bf28}+0.16\%$
test_update	0.3386ms	16.2197μs	61.6534 KOps/s	56.1272 KOps/s	$\textbf{\color{#35bf28}+9.85\%}$
test_update_nested	1.1225ms	20.9525μs	47.7270 KOps/s	44.6071 KOps/s	$\textbf{\color{#35bf28}+6.99\%}$
test_update__nested	81.2440μs	23.5511μs	42.4608 KOps/s	43.3704 KOps/s	$\color{#d91a1a}-2.10\%$
test_set_nested	83.1650μs	14.3771μs	69.5550 KOps/s	66.3022 KOps/s	$\color{#35bf28}+4.91\%$
test_set_nested_new	72.2640μs	16.3688μs	61.0919 KOps/s	58.1001 KOps/s	$\textbf{\color{#35bf28}+5.15\%}$
test_select	62.0940μs	28.5941μs	34.9722 KOps/s	33.2868 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_select_nested	72.5740μs	41.5138μs	24.0883 KOps/s	24.0411 KOps/s	$\color{#35bf28}+0.20\%$
test_exclude_nested	0.1127ms	59.5023μs	16.8061 KOps/s	16.9944 KOps/s	$\color{#d91a1a}-1.11\%$
test_empty[True]	0.3150ms	0.2542ms	3.9335 KOps/s	3.9361 KOps/s	$\color{#d91a1a}-0.07\%$
test_empty[False]	3.4271μs	0.7419μs	1.3479 MOps/s	1.3520 MOps/s	$\color{#d91a1a}-0.30\%$
test_to	86.1950μs	54.1027μs	18.4834 KOps/s	18.3974 KOps/s	$\color{#35bf28}+0.47\%$
test_to_nonblocking	95.3350μs	45.2303μs	22.1091 KOps/s	22.5853 KOps/s	$\color{#d91a1a}-2.11\%$
test_unbind_speed	1.6743ms	0.2307ms	4.3340 KOps/s	4.3456 KOps/s	$\color{#d91a1a}-0.27\%$
test_unbind_speed_stack0	0.2800ms	0.2311ms	4.3264 KOps/s	4.3151 KOps/s	$\color{#35bf28}+0.26\%$
test_unbind_speed_stack1	92.9464ms	0.6427ms	1.5560 KOps/s	1.5513 KOps/s	$\color{#35bf28}+0.31\%$
test_split	93.6500ms	1.5872ms	630.0279 Ops/s	640.9608 Ops/s	$\color{#d91a1a}-1.71\%$
test_chunk	95.7835ms	1.6012ms	624.5151 Ops/s	587.8331 Ops/s	$\textbf{\color{#35bf28}+6.24\%}$
test_consolidate[False-None]	96.2208ms	2.8902ms	345.9914 Ops/s	383.4067 Ops/s	$\textbf{\color{#d91a1a}-9.76\%}$
test_consolidate[default-None]	1.8450ms	1.6980ms	588.9128 Ops/s	596.9174 Ops/s	$\color{#d91a1a}-1.34\%$
test_consolidate[reduce-overhead-None]	1.8652ms	1.6982ms	588.8714 Ops/s	577.3676 Ops/s	$\color{#35bf28}+1.99\%$
test_consolidate_njt[False-None]	6.6552ms	6.3377ms	157.7862 Ops/s	151.6324 Ops/s	$\color{#35bf28}+4.06\%$
test_to[False-False-None]	1.7951ms	1.6481ms	606.7465 Ops/s	613.9869 Ops/s	$\color{#d91a1a}-1.18\%$
test_to[True-False-None]	1.6510ms	1.2922ms	773.8788 Ops/s	777.2406 Ops/s	$\color{#d91a1a}-0.43\%$
test_to[within-False-None]	4.5622ms	4.0257ms	248.4068 Ops/s	253.2489 Ops/s	$\color{#d91a1a}-1.91\%$
test_to[True-default-None]	5.5862ms	5.1776ms	193.1407 Ops/s	199.9816 Ops/s	$\color{#d91a1a}-3.42\%$
test_to_njt[False-False-None]	7.3806ms	6.9411ms	144.0695 Ops/s	146.8333 Ops/s	$\color{#d91a1a}-1.88\%$
test_to_njt[True-False-None]	5.8206ms	5.4220ms	184.4350 Ops/s	188.4751 Ops/s	$\color{#d91a1a}-2.14\%$
test_to_njt[within-False-None]	12.3872ms	11.9945ms	83.3714 Ops/s	85.1553 Ops/s	$\color{#d91a1a}-2.09\%$
test_creation[device0]	0.5826ms	79.2740μs	12.6145 KOps/s	12.8214 KOps/s	$\color{#d91a1a}-1.61\%$
test_creation_from_tensor	0.4697ms	81.2200μs	12.3122 KOps/s	12.4794 KOps/s	$\color{#d91a1a}-1.34\%$
test_add_one[memmap_tensor0]	0.3078ms	5.9783μs	167.2706 KOps/s	167.0615 KOps/s	$\color{#35bf28}+0.13\%$
test_contiguous[memmap_tensor0]	1.7421μs	0.4103μs	2.4373 MOps/s	2.4112 MOps/s	$\color{#35bf28}+1.08\%$
test_stack[memmap_tensor0]	38.6720μs	4.3693μs	228.8672 KOps/s	231.9893 KOps/s	$\color{#d91a1a}-1.35\%$
test_memmaptd_index	1.7822ms	0.2386ms	4.1905 KOps/s	4.1593 KOps/s	$\color{#35bf28}+0.75\%$
test_memmaptd_index_astensor	0.9544ms	0.2962ms	3.3764 KOps/s	3.3630 KOps/s	$\color{#35bf28}+0.40\%$
test_memmaptd_index_op	0.9763ms	0.5392ms	1.8547 KOps/s	1.7711 KOps/s	$\color{#35bf28}+4.72\%$
test_serialize_model	0.1328s	0.1311s	7.6292 Ops/s	7.6410 Ops/s	$\color{#d91a1a}-0.15\%$
test_serialize_model_pickle	1.3505s	1.2121s	0.8250 Ops/s	0.8244 Ops/s	$\color{#35bf28}+0.07\%$
test_serialize_weights	0.1311s	0.1301s	7.6842 Ops/s	7.6826 Ops/s	$\color{#35bf28}+0.02\%$
test_serialize_weights_returnearly	47.4668ms	41.2615ms	24.2357 Ops/s	14.5642 Ops/s	$\textbf{\color{#35bf28}+66.41\%}$
test_serialize_weights_pickle	1.3483s	1.2111s	0.8257 Ops/s	0.8351 Ops/s	$\color{#d91a1a}-1.12\%$
test_reshape_pytree	44.9110μs	22.2840μs	44.8752 KOps/s	43.2435 KOps/s	$\color{#35bf28}+3.77\%$
test_reshape_td	57.5020μs	26.2096μs	38.1539 KOps/s	36.9473 KOps/s	$\color{#35bf28}+3.27\%$
test_view_pytree	52.7620μs	22.2982μs	44.8467 KOps/s	44.9154 KOps/s	$\color{#d91a1a}-0.15\%$
test_view_td	53.8520μs	28.9234μs	34.5741 KOps/s	32.2934 KOps/s	$\textbf{\color{#35bf28}+7.06\%}$
test_unbind_pytree	56.9320μs	27.3736μs	36.5316 KOps/s	35.9214 KOps/s	$\color{#35bf28}+1.70\%$
test_unbind_td	0.8847ms	35.9118μs	27.8460 KOps/s	27.6272 KOps/s	$\color{#35bf28}+0.79\%$
test_split_pytree	68.0130μs	30.2922μs	33.0118 KOps/s	33.5186 KOps/s	$\color{#d91a1a}-1.51\%$
test_split_td	1.0545ms	39.4590μs	25.3427 KOps/s	25.6437 KOps/s	$\color{#d91a1a}-1.17\%$
test_add_pytree	75.7430μs	32.9817μs	30.3198 KOps/s	31.0402 KOps/s	$\color{#d91a1a}-2.32\%$
test_add_td	85.3640μs	46.7498μs	21.3904 KOps/s	21.6153 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_add_one_nested[tensordict-compile]	0.1766ms	0.1210ms	8.2673 KOps/s	8.0140 KOps/s	$\color{#35bf28}+3.16\%$
test_compile_add_one_nested[tensordict-eager]	0.2332ms	0.1237ms	8.0858 KOps/s	7.9891 KOps/s	$\color{#35bf28}+1.21\%$
test_compile_add_one_nested[pytree-compile]	0.1512ms	96.0847μs	10.4075 KOps/s	10.1955 KOps/s	$\color{#35bf28}+2.08\%$
test_compile_add_one_nested[pytree-eager]	0.2063ms	0.1458ms	6.8593 KOps/s	6.8912 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_copy_nested[tensordict-compile]	52.8330μs	23.6786μs	42.2323 KOps/s	32.1528 KOps/s	$\textbf{\color{#35bf28}+31.35\%}$
test_compile_copy_nested[tensordict-eager]	0.1558ms	27.0708μs	36.9402 KOps/s	36.9819 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_copy_nested[pytree-compile]	0.3305ms	64.5064μs	15.5023 KOps/s	15.0244 KOps/s	$\color{#35bf28}+3.18\%$
test_compile_copy_nested[pytree-eager]	83.4140μs	49.0404μs	20.3914 KOps/s	20.0173 KOps/s	$\color{#35bf28}+1.87\%$
test_compile_add_one_flat[tensordict-compile]	0.1818ms	0.1431ms	6.9894 KOps/s	6.8913 KOps/s	$\color{#35bf28}+1.42\%$
test_compile_add_one_flat[tensordict-eager]	0.2939ms	0.2060ms	4.8548 KOps/s	4.9437 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_add_one_flat[tensorclass-compile]	0.1487ms	98.8316μs	10.1182 KOps/s	10.0865 KOps/s	$\color{#35bf28}+0.31\%$
test_compile_add_one_flat[tensorclass-eager]	0.1069ms	52.9333μs	18.8917 KOps/s	19.7497 KOps/s	$\color{#d91a1a}-4.34\%$
test_compile_add_one_flat[pytree-compile]	0.1827ms	0.1364ms	7.3314 KOps/s	7.2912 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_add_one_flat[pytree-eager]	0.5578ms	0.4647ms	2.1520 KOps/s	2.1426 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_self_flat[tensordict-eager]	0.3872ms	0.2465ms	4.0560 KOps/s	4.1280 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_add_self_flat[tensordict-compile]	0.1926ms	0.1441ms	6.9406 KOps/s	6.9402 KOps/s	$+0.01\%$
test_compile_add_self_flat[tensorclass-eager]	0.1355ms	63.0568μs	15.8587 KOps/s	16.4427 KOps/s	$\color{#d91a1a}-3.55\%$
test_compile_add_self_flat[tensorclass-compile]	0.1553ms	99.7416μs	10.0259 KOps/s	10.1384 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_add_self_flat[pytree-eager]	0.4656ms	0.4006ms	2.4964 KOps/s	2.5515 KOps/s	$\color{#d91a1a}-2.16\%$
test_compile_add_self_flat[pytree-compile]	0.1989ms	0.1367ms	7.3163 KOps/s	7.3278 KOps/s	$\color{#d91a1a}-0.16\%$
test_compile_copy_flat[tensordict-compile]	64.8730μs	19.3870μs	51.5811 KOps/s	55.8420 KOps/s	$\textbf{\color{#d91a1a}-7.63\%}$
test_compile_copy_flat[tensordict-eager]	55.1530μs	26.9482μs	37.1082 KOps/s	36.9472 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_copy_flat[pytree-compile]	0.1171ms	69.9369μs	14.2986 KOps/s	14.1441 KOps/s	$\color{#35bf28}+1.09\%$
test_compile_copy_flat[pytree-eager]	95.7350μs	51.4669μs	19.4300 KOps/s	19.2706 KOps/s	$\color{#35bf28}+0.83\%$
test_compile_assign_and_add[tensordict-compile]	1.6466ms	0.3937ms	2.5403 KOps/s	2.1716 KOps/s	$\textbf{\color{#35bf28}+16.98\%}$
test_compile_assign_and_add[tensordict-eager]	2.6878ms	2.5431ms	393.2154 Ops/s	396.9514 Ops/s	$\color{#d91a1a}-0.94\%$
test_compile_assign_and_add[pytree-compile]	1.6114ms	0.4366ms	2.2905 KOps/s	2.2183 KOps/s	$\color{#35bf28}+3.25\%$
test_compile_assign_and_add[pytree-eager]	2.6610ms	2.5288ms	395.4467 Ops/s	394.0710 Ops/s	$\color{#35bf28}+0.35\%$
test_compile_indexing[tensor-tensordict-compile]	0.1757ms	0.1107ms	9.0294 KOps/s	9.0522 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_indexing[tensor-tensordict-eager]	0.5769ms	76.0322μs	13.1523 KOps/s	13.1180 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2454ms	0.1036ms	9.6529 KOps/s	9.8426 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1602ms	66.7969μs	14.9708 KOps/s	15.3697 KOps/s	$\color{#d91a1a}-2.60\%$
test_compile_indexing[tensor-pytree-compile]	0.2162ms	0.1041ms	9.6037 KOps/s	9.6389 KOps/s	$\color{#d91a1a}-0.37\%$
test_compile_indexing[tensor-pytree-eager]	0.1892ms	67.2061μs	14.8796 KOps/s	15.5617 KOps/s	$\color{#d91a1a}-4.38\%$
test_compile_indexing[slice-tensordict-compile]	0.2161ms	0.1000ms	9.9991 KOps/s	9.8601 KOps/s	$\color{#35bf28}+1.41\%$
test_compile_indexing[slice-tensordict-eager]	0.1477ms	16.8737μs	59.2638 KOps/s	59.4377 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_indexing[slice-tensorclass-compile]	0.1598ms	95.4252μs	10.4794 KOps/s	10.5524 KOps/s	$\color{#d91a1a}-0.69\%$
test_compile_indexing[slice-tensorclass-eager]	0.1084ms	15.9885μs	62.5450 KOps/s	64.4333 KOps/s	$\color{#d91a1a}-2.93\%$
test_compile_indexing[slice-pytree-compile]	0.2053ms	95.7604μs	10.4427 KOps/s	10.4359 KOps/s	$\color{#35bf28}+0.07\%$
test_compile_indexing[slice-pytree-eager]	98.5840μs	15.7850μs	63.3513 KOps/s	64.7694 KOps/s	$\color{#d91a1a}-2.19\%$
test_compile_indexing[int-tensordict-compile]	0.1951ms	0.1012ms	9.8766 KOps/s	9.9869 KOps/s	$\color{#d91a1a}-1.10\%$
test_compile_indexing[int-tensordict-eager]	0.5645ms	16.6845μs	59.9359 KOps/s	60.5537 KOps/s	$\color{#d91a1a}-1.02\%$
test_compile_indexing[int-tensorclass-compile]	0.1935ms	96.2237μs	10.3924 KOps/s	10.4474 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_indexing[int-tensorclass-eager]	0.1523ms	15.8605μs	63.0497 KOps/s	64.6856 KOps/s	$\color{#d91a1a}-2.53\%$
test_compile_indexing[int-pytree-compile]	0.1875ms	96.3209μs	10.3820 KOps/s	10.4864 KOps/s	$\color{#d91a1a}-1.00\%$
test_compile_indexing[int-pytree-eager]	97.6240μs	15.8723μs	63.0027 KOps/s	64.6225 KOps/s	$\color{#d91a1a}-2.51\%$
test_mod_add[eager]	0.1338ms	30.3562μs	32.9422 KOps/s	32.5019 KOps/s	$\color{#35bf28}+1.35\%$
test_mod_add[compile]	0.1867ms	77.8205μs	12.8501 KOps/s	12.5743 KOps/s	$\color{#35bf28}+2.19\%$
test_mod_add[compile-overhead]	0.3134ms	0.1641ms	6.0929 KOps/s	5.7693 KOps/s	$\textbf{\color{#35bf28}+5.61\%}$
test_mod_wrap[eager]	0.3347ms	0.2298ms	4.3514 KOps/s	4.3107 KOps/s	$\color{#35bf28}+0.94\%$
test_mod_wrap[compile]	1.6067ms	0.2822ms	3.5436 KOps/s	3.5561 KOps/s	$\color{#d91a1a}-0.35\%$
test_mod_wrap[compile-overhead]	7.4127ms	3.8386ms	260.5113 Ops/s	264.0540 Ops/s	$\color{#d91a1a}-1.34\%$
test_mod_wrap_and_backward[eager]	1.5361ms	1.3258ms	754.2898 Ops/s	718.1627 Ops/s	$\textbf{\color{#35bf28}+5.03\%}$
test_mod_wrap_and_backward[compile]	1.4667ms	1.2863ms	777.4345 Ops/s	726.4200 Ops/s	$\textbf{\color{#35bf28}+7.02\%}$
test_mod_wrap_and_backward[compile-overhead]	1.4137ms	0.9820ms	1.0184 KOps/s	972.0550 Ops/s	$\color{#35bf28}+4.77\%$
test_seq_add[eager]	0.1611ms	96.3985μs	10.3736 KOps/s	10.2988 KOps/s	$\color{#35bf28}+0.73\%$
test_seq_add[compile]	0.1813ms	87.4066μs	11.4408 KOps/s	11.6527 KOps/s	$\color{#d91a1a}-1.82\%$
test_seq_add[compile-overhead]	0.1844ms	0.1284ms	7.7869 KOps/s	7.8182 KOps/s	$\color{#d91a1a}-0.40\%$
test_seq_wrap[eager]	0.4513ms	0.3720ms	2.6885 KOps/s	2.6666 KOps/s	$\color{#35bf28}+0.82\%$
test_seq_wrap[compile]	0.4208ms	0.2990ms	3.3448 KOps/s	3.3431 KOps/s	$\color{#35bf28}+0.05\%$
test_seq_wrap[compile-overhead]	0.2986ms	0.2221ms	4.5017 KOps/s	4.4929 KOps/s	$\color{#35bf28}+0.20\%$
test_func_call_runtime[False-eager]	0.8043ms	0.7105ms	1.4074 KOps/s	1.4183 KOps/s	$\color{#d91a1a}-0.77\%$
test_func_call_runtime[False-compile]	0.8246ms	0.7436ms	1.3448 KOps/s	1.3685 KOps/s	$\color{#d91a1a}-1.73\%$
test_func_call_runtime[False-compile-overhead]	0.4746ms	0.3636ms	2.7505 KOps/s	2.7170 KOps/s	$\color{#35bf28}+1.23\%$
test_func_call_runtime[True-eager]	0.9982ms	0.8735ms	1.1448 KOps/s	1.1339 KOps/s	$\color{#35bf28}+0.96\%$
test_func_call_runtime[True-compile]	0.8392ms	0.7681ms	1.3019 KOps/s	1.3035 KOps/s	$\color{#d91a1a}-0.12\%$
test_func_call_runtime[True-compile-overhead]	0.4589ms	0.3845ms	2.6005 KOps/s	2.5976 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[False-eager]	0.8330ms	0.7429ms	1.3461 KOps/s	1.3012 KOps/s	$\color{#35bf28}+3.45\%$
test_func_call_cm_runtime[False-compile]	0.9138ms	0.7545ms	1.3253 KOps/s	1.3286 KOps/s	$\color{#d91a1a}-0.24\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4223ms	0.3650ms	2.7394 KOps/s	2.7332 KOps/s	$\color{#35bf28}+0.23\%$
test_func_call_cm_runtime[True-eager]	1.0796ms	0.9699ms	1.0310 KOps/s	1.0270 KOps/s	$\color{#35bf28}+0.39\%$
test_func_call_cm_runtime[True-compile]	0.8601ms	0.7954ms	1.2572 KOps/s	1.2557 KOps/s	$\color{#35bf28}+0.12\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4615ms	0.4101ms	2.4383 KOps/s	2.4273 KOps/s	$\color{#35bf28}+0.45\%$
test_vmap_func_call_cm_runtime[eager]	2.4884ms	1.9833ms	504.1984 Ops/s	498.3557 Ops/s	$\color{#35bf28}+1.17\%$
test_vmap_func_call_cm_runtime[compile]	0.8893ms	0.8092ms	1.2358 KOps/s	1.1694 KOps/s	$\textbf{\color{#35bf28}+5.67\%}$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4668ms	0.4144ms	2.4131 KOps/s	2.4129 KOps/s	$+0.01\%$
test_distributed	1.8704ms	0.1853ms	5.3962 KOps/s	8.7547 KOps/s	$\textbf{\color{#d91a1a}-38.36\%}$
test_tdmodule	48.0830μs	14.0397μs	71.2266 KOps/s	68.8997 KOps/s	$\color{#35bf28}+3.38\%$
test_tdmodule_dispatch	67.4630μs	27.6255μs	36.1984 KOps/s	34.0742 KOps/s	$\textbf{\color{#35bf28}+6.23\%}$
test_tdseq	47.3820μs	15.5224μs	64.4231 KOps/s	61.9709 KOps/s	$\color{#35bf28}+3.96\%$
test_tdseq_dispatch	52.1920μs	30.2898μs	33.0144 KOps/s	31.2162 KOps/s	$\textbf{\color{#35bf28}+5.76\%}$
test_instantiation_functorch	1.6298ms	1.5419ms	648.5319 Ops/s	650.8334 Ops/s	$\color{#d91a1a}-0.35\%$
test_exec_functorch	0.1974ms	0.1386ms	7.2136 KOps/s	7.0092 KOps/s	$\color{#35bf28}+2.91\%$
test_exec_functional_call	0.1686ms	0.1314ms	7.6126 KOps/s	7.5589 KOps/s	$\color{#35bf28}+0.71\%$
test_exec_td_decorator	0.3829ms	0.1767ms	5.6603 KOps/s	5.6767 KOps/s	$\color{#d91a1a}-0.29\%$
test_vmap_mlp_speed_decorator[True-True]	0.7845ms	0.6478ms	1.5438 KOps/s	1.5255 KOps/s	$\color{#35bf28}+1.20\%$
test_vmap_mlp_speed_decorator[True-False]	0.7470ms	0.6475ms	1.5444 KOps/s	1.5304 KOps/s	$\color{#35bf28}+0.91\%$
test_vmap_mlp_speed_decorator[False-True]	0.7163ms	0.5648ms	1.7706 KOps/s	1.7542 KOps/s	$\color{#35bf28}+0.93\%$
test_vmap_mlp_speed_decorator[False-False]	0.6727ms	0.5679ms	1.7608 KOps/s	1.7544 KOps/s	$\color{#35bf28}+0.37\%$
test_vmap_transformer_speed_decorator[True-True]	18.5943ms	18.4224ms	54.2818 Ops/s	54.4962 Ops/s	$\color{#d91a1a}-0.39\%$
test_vmap_transformer_speed_decorator[True-False]	18.5803ms	18.4536ms	54.1898 Ops/s	54.3603 Ops/s	$\color{#d91a1a}-0.31\%$
test_vmap_transformer_speed_decorator[False-True]	18.4517ms	18.2705ms	54.7331 Ops/s	55.0474 Ops/s	$\color{#d91a1a}-0.57\%$
test_vmap_transformer_speed_decorator[False-False]	18.5464ms	18.3361ms	54.5372 Ops/s	54.8736 Ops/s	$\color{#d91a1a}-0.61\%$
test_to_module_speed[True]	1.2946ms	0.9244ms	1.0817 KOps/s	1.0487 KOps/s	$\color{#35bf28}+3.15\%$
test_to_module_speed[False]	1.4830ms	0.9197ms	1.0873 KOps/s	1.0781 KOps/s	$\color{#35bf28}+0.85\%$
test_tc_init	61.7030μs	34.7133μs	28.8074 KOps/s	27.1292 KOps/s	$\textbf{\color{#35bf28}+6.19\%}$
test_tc_init_nested	0.1733ms	71.2934μs	14.0266 KOps/s	13.2134 KOps/s	$\textbf{\color{#35bf28}+6.15\%}$
test_tc_first_layer_tensor	5.2360μs	0.6982μs	1.4322 MOps/s	1.4249 MOps/s	$\color{#35bf28}+0.51\%$
test_tc_first_layer_nontensor	53.9730μs	2.2889μs	436.8872 KOps/s	427.2113 KOps/s	$\color{#35bf28}+2.26\%$
test_tc_second_layer_tensor	9.3703μs	1.4271μs	700.6983 KOps/s	699.2530 KOps/s	$\color{#35bf28}+0.21\%$
test_tc_second_layer_nontensor	29.1020μs	3.0469μs	328.2033 KOps/s	323.9309 KOps/s	$\color{#35bf28}+1.32\%$
test_unbind	0.2376s	10.0010ms	99.9899 Ops/s	152.2043 Ops/s	$\textbf{\color{#d91a1a}-34.31\%}$
test_full_like	12.9254ms	9.1288ms	109.5428 Ops/s	109.4642 Ops/s	$\color{#35bf28}+0.07\%$
test_zeros_like	5.2746ms	4.3234ms	231.2984 Ops/s	231.8210 Ops/s	$\color{#d91a1a}-0.23\%$
test_ones_like	9.2153ms	7.2211ms	138.4839 Ops/s	231.5149 Ops/s	$\textbf{\color{#d91a1a}-40.18\%}$
test_clone	6.8334ms	6.3153ms	158.3466 Ops/s	157.9370 Ops/s	$\color{#35bf28}+0.26\%$
test_squeeze	57.6530μs	9.4195μs	106.1630 KOps/s	107.4572 KOps/s	$\color{#d91a1a}-1.20\%$
test_unsqueeze	0.1204ms	71.0344μs	14.0777 KOps/s	13.8297 KOps/s	$\color{#35bf28}+1.79\%$
test_split	0.3762ms	0.1587ms	6.3003 KOps/s	6.2757 KOps/s	$\color{#35bf28}+0.39\%$
test_permute	0.2253ms	0.1812ms	5.5179 KOps/s	5.3962 KOps/s	$\color{#35bf28}+2.26\%$
test_stack	51.1818ms	50.7562ms	19.7020 Ops/s	19.6945 Ops/s	$\color{#35bf28}+0.04\%$
test_cat	51.3135ms	50.5710ms	19.7742 Ops/s	19.7232 Ops/s	$\color{#35bf28}+0.26\%$