[Feature] OrderedDict for TensorDictSequential

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] Plotting TensorDictSequential graphs #1144
• [Feature] Fix type assertion in Seq build #1143
• -> [Feature] OrderedDict for TensorDictSequential #1142

[github-actions]

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

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

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.2410μs	21.0067μs	47.6038 KOps/s	49.1444 KOps/s	$\color{#d91a1a}-3.13\%$
test_plain_set_stack_nested	42.5480μs	21.1305μs	47.3249 KOps/s	48.7377 KOps/s	$\color{#d91a1a}-2.90\%$
test_plain_set_nested_inplace	0.1034ms	23.1301μs	43.2337 KOps/s	44.7778 KOps/s	$\color{#d91a1a}-3.45\%$
test_plain_set_stack_nested_inplace	60.5830μs	22.7945μs	43.8702 KOps/s	44.8109 KOps/s	$\color{#d91a1a}-2.10\%$
test_items	29.9760μs	4.1894μs	238.6980 KOps/s	238.6961 KOps/s	$+0.00\%$
test_items_nested	0.7149ms	0.4001ms	2.4991 KOps/s	2.4522 KOps/s	$\color{#35bf28}+1.91\%$
test_items_nested_locked	0.5072ms	0.4002ms	2.4989 KOps/s	2.4423 KOps/s	$\color{#35bf28}+2.31\%$
test_items_nested_leaf	0.1228ms	76.7681μs	13.0262 KOps/s	13.1086 KOps/s	$\color{#d91a1a}-0.63\%$
test_items_stack_nested	0.6000ms	0.4017ms	2.4892 KOps/s	2.4186 KOps/s	$\color{#35bf28}+2.92\%$
test_items_stack_nested_leaf	0.1417ms	77.2472μs	12.9454 KOps/s	12.4325 KOps/s	$\color{#35bf28}+4.13\%$
test_items_stack_nested_locked	0.5844ms	0.4006ms	2.4960 KOps/s	2.4485 KOps/s	$\color{#35bf28}+1.94\%$
test_keys	25.1170μs	3.4934μs	286.2578 KOps/s	282.3420 KOps/s	$\color{#35bf28}+1.39\%$
test_keys_nested	0.3188ms	0.1650ms	6.0608 KOps/s	5.9750 KOps/s	$\color{#35bf28}+1.44\%$
test_keys_nested_locked	1.9396ms	0.1701ms	5.8773 KOps/s	5.7547 KOps/s	$\color{#35bf28}+2.13\%$
test_keys_nested_leaf	0.2514ms	0.1425ms	7.0200 KOps/s	6.9101 KOps/s	$\color{#35bf28}+1.59\%$
test_keys_stack_nested	0.2443ms	0.1649ms	6.0654 KOps/s	6.0810 KOps/s	$\color{#d91a1a}-0.26\%$
test_keys_stack_nested_leaf	0.2664ms	0.1444ms	6.9239 KOps/s	7.0458 KOps/s	$\color{#d91a1a}-1.73\%$
test_keys_stack_nested_locked	0.2665ms	0.1718ms	5.8211 KOps/s	5.9153 KOps/s	$\color{#d91a1a}-1.59\%$
test_values	8.0168μs	1.0518μs	950.7135 KOps/s	963.4188 KOps/s	$\color{#d91a1a}-1.32\%$
test_values_nested	0.1197ms	62.5079μs	15.9980 KOps/s	15.8828 KOps/s	$\color{#35bf28}+0.73\%$
test_values_nested_locked	0.1209ms	62.7899μs	15.9261 KOps/s	15.7224 KOps/s	$\color{#35bf28}+1.30\%$
test_values_nested_leaf	0.1282ms	71.8425μs	13.9193 KOps/s	13.2697 KOps/s	$\color{#35bf28}+4.90\%$
test_values_stack_nested	0.1236ms	63.7959μs	15.6750 KOps/s	15.6796 KOps/s	$\color{#d91a1a}-0.03\%$
test_values_stack_nested_leaf	0.1404ms	72.8941μs	13.7185 KOps/s	13.9063 KOps/s	$\color{#d91a1a}-1.35\%$
test_values_stack_nested_locked	0.1405ms	63.3704μs	15.7802 KOps/s	15.7522 KOps/s	$\color{#35bf28}+0.18\%$
test_membership	21.7610μs	0.9144μs	1.0936 MOps/s	1.1362 MOps/s	$\color{#d91a1a}-3.75\%$
test_membership_nested	23.7140μs	3.0026μs	333.0492 KOps/s	340.3915 KOps/s	$\color{#d91a1a}-2.16\%$
test_membership_nested_leaf	39.0030μs	3.0073μs	332.5253 KOps/s	337.8967 KOps/s	$\color{#d91a1a}-1.59\%$
test_membership_stacked_nested	23.6840μs	2.9492μs	339.0782 KOps/s	346.3278 KOps/s	$\color{#d91a1a}-2.09\%$
test_membership_stacked_nested_leaf	36.7680μs	3.0228μs	330.8195 KOps/s	332.3226 KOps/s	$\color{#d91a1a}-0.45\%$
test_membership_nested_last	35.5660μs	4.3665μs	229.0162 KOps/s	226.7968 KOps/s	$\color{#35bf28}+0.98\%$
test_membership_nested_leaf_last	31.3780μs	4.4041μs	227.0608 KOps/s	222.6049 KOps/s	$\color{#35bf28}+2.00\%$
test_membership_stacked_nested_last	40.3050μs	4.3550μs	229.6204 KOps/s	224.3194 KOps/s	$\color{#35bf28}+2.36\%$
test_membership_stacked_nested_leaf_last	23.0330μs	4.3297μs	230.9624 KOps/s	221.9158 KOps/s	$\color{#35bf28}+4.08\%$
test_nested_getleaf	40.9760μs	10.8956μs	91.7802 KOps/s	92.7169 KOps/s	$\color{#d91a1a}-1.01\%$
test_nested_get	39.4030μs	10.4294μs	95.8827 KOps/s	97.7346 KOps/s	$\color{#d91a1a}-1.89\%$
test_stacked_getleaf	42.8800μs	10.6914μs	93.5331 KOps/s	92.5988 KOps/s	$\color{#35bf28}+1.01\%$
test_stacked_get	53.6090μs	10.1405μs	98.6148 KOps/s	98.5663 KOps/s	$\color{#35bf28}+0.05\%$
test_nested_getitemleaf	39.7040μs	11.1003μs	90.0874 KOps/s	87.5151 KOps/s	$\color{#35bf28}+2.94\%$
test_nested_getitem	57.7570μs	10.4979μs	95.2570 KOps/s	92.3168 KOps/s	$\color{#35bf28}+3.18\%$
test_stacked_getitemleaf	36.2270μs	11.2841μs	88.6204 KOps/s	89.3659 KOps/s	$\color{#d91a1a}-0.83\%$
test_stacked_getitem	38.5420μs	10.3971μs	96.1805 KOps/s	95.3516 KOps/s	$\color{#35bf28}+0.87\%$
test_lock_nested	5.0118ms	0.4642ms	2.1543 KOps/s	2.1693 KOps/s	$\color{#d91a1a}-0.69\%$
test_lock_stack_nested	0.6737ms	0.4294ms	2.3287 KOps/s	2.3750 KOps/s	$\color{#d91a1a}-1.95\%$
test_unlock_nested	0.8089ms	0.3742ms	2.6727 KOps/s	2.6553 KOps/s	$\color{#35bf28}+0.66\%$
test_unlock_stack_nested	0.5193ms	0.3464ms	2.8869 KOps/s	2.9173 KOps/s	$\color{#d91a1a}-1.04\%$
test_flatten_speed	0.1993ms	99.5770μs	10.0425 KOps/s	10.0507 KOps/s	$\color{#d91a1a}-0.08\%$
test_unflatten_speed	0.9010ms	0.5226ms	1.9134 KOps/s	1.8402 KOps/s	$\color{#35bf28}+3.98\%$
test_common_ops	1.7294ms	0.7969ms	1.2549 KOps/s	1.2759 KOps/s	$\color{#d91a1a}-1.64\%$
test_creation	46.3870μs	2.5237μs	396.2480 KOps/s	401.4354 KOps/s	$\color{#d91a1a}-1.29\%$
test_creation_empty	40.5850μs	12.4048μs	80.6139 KOps/s	96.8223 KOps/s	$\textbf{\color{#d91a1a}-16.74\%}$
test_creation_nested_1	51.6850μs	15.4625μs	64.6728 KOps/s	74.1362 KOps/s	$\textbf{\color{#d91a1a}-12.76\%}$
test_creation_nested_2	53.6390μs	20.1573μs	49.6098 KOps/s	54.7949 KOps/s	$\textbf{\color{#d91a1a}-9.46\%}$
test_clone	73.4570μs	13.3395μs	74.9655 KOps/s	74.4358 KOps/s	$\color{#35bf28}+0.71\%$
test_getitem[int]	1.1212ms	12.8932μs	77.5601 KOps/s	77.9296 KOps/s	$\color{#d91a1a}-0.47\%$
test_getitem[slice_int]	0.1427ms	23.9952μs	41.6750 KOps/s	40.0126 KOps/s	$\color{#35bf28}+4.15\%$
test_getitem[range]	0.2179ms	48.0945μs	20.7924 KOps/s	20.4813 KOps/s	$\color{#35bf28}+1.52\%$
test_getitem[tuple]	0.1671ms	19.8648μs	50.3402 KOps/s	49.9962 KOps/s	$\color{#35bf28}+0.69\%$
test_getitem[list]	0.2996ms	43.4108μs	23.0357 KOps/s	22.4479 KOps/s	$\color{#35bf28}+2.62\%$
test_setitem_dim[int]	51.4150μs	24.8105μs	40.3054 KOps/s	37.4248 KOps/s	$\textbf{\color{#35bf28}+7.70\%}$
test_setitem_dim[slice_int]	0.1158ms	50.9172μs	19.6397 KOps/s	18.3843 KOps/s	$\textbf{\color{#35bf28}+6.83\%}$
test_setitem_dim[range]	0.1486ms	72.6851μs	13.7580 KOps/s	13.4679 KOps/s	$\color{#35bf28}+2.15\%$
test_setitem_dim[tuple]	0.1089ms	39.9670μs	25.0207 KOps/s	23.4090 KOps/s	$\textbf{\color{#35bf28}+6.88\%}$
test_setitem	0.1408ms	21.1043μs	47.3836 KOps/s	49.5398 KOps/s	$\color{#d91a1a}-4.35\%$
test_set	0.1321ms	20.5288μs	48.7120 KOps/s	51.0017 KOps/s	$\color{#d91a1a}-4.49\%$
test_set_shared	2.2572ms	0.1690ms	5.9170 KOps/s	5.8731 KOps/s	$\color{#35bf28}+0.75\%$
test_update	0.8333ms	23.1988μs	43.1056 KOps/s	44.5168 KOps/s	$\color{#d91a1a}-3.17\%$
test_update_nested	0.1157ms	33.0909μs	30.2198 KOps/s	29.8616 KOps/s	$\color{#35bf28}+1.20\%$
test_update__nested	0.1804ms	33.7805μs	29.6028 KOps/s	29.0142 KOps/s	$\color{#35bf28}+2.03\%$
test_set_nested	98.8830μs	22.4721μs	44.4997 KOps/s	45.2066 KOps/s	$\color{#d91a1a}-1.56\%$
test_set_nested_new	0.1031ms	27.1130μs	36.8827 KOps/s	36.6050 KOps/s	$\color{#35bf28}+0.76\%$
test_select	0.2103ms	43.6253μs	22.9225 KOps/s	22.7095 KOps/s	$\color{#35bf28}+0.94\%$
test_select_nested	0.1298ms	62.2911μs	16.0537 KOps/s	15.6796 KOps/s	$\color{#35bf28}+2.39\%$
test_exclude_nested	0.1621ms	80.6539μs	12.3987 KOps/s	12.1846 KOps/s	$\color{#35bf28}+1.76\%$
test_empty[True]	0.7020ms	0.4221ms	2.3691 KOps/s	2.4223 KOps/s	$\color{#d91a1a}-2.19\%$
test_empty[False]	12.3980μs	1.3877μs	720.5915 KOps/s	703.7374 KOps/s	$\color{#35bf28}+2.39\%$
test_unbind_speed	0.3260ms	0.2683ms	3.7272 KOps/s	3.6959 KOps/s	$\color{#35bf28}+0.85\%$
test_unbind_speed_stack0	0.5689ms	0.2679ms	3.7325 KOps/s	3.7538 KOps/s	$\color{#d91a1a}-0.57\%$
test_unbind_speed_stack1	0.1135s	0.8138ms	1.2288 KOps/s	1.3868 KOps/s	$\textbf{\color{#d91a1a}-11.39\%}$
test_split	0.1120s	1.7752ms	563.3109 Ops/s	549.2828 Ops/s	$\color{#35bf28}+2.55\%$
test_chunk	0.1109s	1.7836ms	560.6532 Ops/s	545.8042 Ops/s	$\color{#35bf28}+2.72\%$
test_consolidate_njt[False-None]	8.6703ms	8.1728ms	122.3568 Ops/s	120.9340 Ops/s	$\color{#35bf28}+1.18\%$
test_creation[device0]	0.2759ms	90.2850μs	11.0760 KOps/s	10.8648 KOps/s	$\color{#35bf28}+1.94\%$
test_creation_from_tensor	5.0047ms	95.7855μs	10.4400 KOps/s	9.8117 KOps/s	$\textbf{\color{#35bf28}+6.40\%}$
test_add_one[memmap_tensor0]	0.1873ms	4.7208μs	211.8286 KOps/s	210.0492 KOps/s	$\color{#35bf28}+0.85\%$
test_contiguous[memmap_tensor0]	16.1200μs	0.5140μs	1.9454 MOps/s	1.9263 MOps/s	$\color{#35bf28}+0.99\%$
test_stack[memmap_tensor0]	36.9690μs	3.3787μs	295.9681 KOps/s	298.6310 KOps/s	$\color{#d91a1a}-0.89\%$
test_memmaptd_index	0.8384ms	0.2349ms	4.2574 KOps/s	4.2351 KOps/s	$\color{#35bf28}+0.53\%$
test_memmaptd_index_astensor	0.5609ms	0.3232ms	3.0944 KOps/s	3.0731 KOps/s	$\color{#35bf28}+0.69\%$
test_memmaptd_index_op	1.1131ms	0.5956ms	1.6791 KOps/s	1.7206 KOps/s	$\color{#d91a1a}-2.41\%$
test_serialize_model	0.1296s	0.1177s	8.4926 Ops/s	8.5343 Ops/s	$\color{#d91a1a}-0.49\%$
test_serialize_model_pickle	0.4595s	0.3928s	2.5455 Ops/s	2.4264 Ops/s	$\color{#35bf28}+4.91\%$
test_serialize_weights	0.2223s	0.1329s	7.5218 Ops/s	7.1938 Ops/s	$\color{#35bf28}+4.56\%$
test_serialize_weights_returnearly	0.1732s	0.1570s	6.3702 Ops/s	6.0740 Ops/s	$\color{#35bf28}+4.88\%$
test_serialize_weights_pickle	0.5380s	0.4490s	2.2271 Ops/s	2.5392 Ops/s	$\textbf{\color{#d91a1a}-12.29\%}$
test_serialize_weights_filesystem	0.1560s	0.1451s	6.8916 Ops/s	6.9774 Ops/s	$\color{#d91a1a}-1.23\%$
test_serialize_model_filesystem	0.1634s	0.1463s	6.8370 Ops/s	5.9159 Ops/s	$\textbf{\color{#35bf28}+15.57\%}$
test_reshape_pytree	72.2540μs	27.1476μs	36.8356 KOps/s	37.5113 KOps/s	$\color{#d91a1a}-1.80\%$
test_reshape_td	96.8100μs	33.5982μs	29.7635 KOps/s	29.4207 KOps/s	$\color{#35bf28}+1.17\%$
test_view_pytree	61.4240μs	26.6600μs	37.5094 KOps/s	37.5052 KOps/s	$\color{#35bf28}+0.01\%$
test_view_td	81.9220μs	38.7997μs	25.7734 KOps/s	25.8760 KOps/s	$\color{#d91a1a}-0.40\%$
test_unbind_pytree	87.5430μs	30.1994μs	33.1132 KOps/s	33.7303 KOps/s	$\color{#d91a1a}-1.83\%$
test_unbind_td	0.3150ms	40.1409μs	24.9123 KOps/s	24.7797 KOps/s	$\color{#35bf28}+0.53\%$
test_split_pytree	73.2650μs	30.0180μs	33.3134 KOps/s	34.0501 KOps/s	$\color{#d91a1a}-2.16\%$
test_split_td	0.5027ms	44.9232μs	22.2602 KOps/s	22.3176 KOps/s	$\color{#d91a1a}-0.26\%$
test_add_pytree	70.3700μs	35.0717μs	28.5130 KOps/s	27.7969 KOps/s	$\color{#35bf28}+2.58\%$
test_add_td	0.1396ms	58.0627μs	17.2228 KOps/s	17.0775 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_add_one_nested[tensordict-compile]	0.1868ms	62.9824μs	15.8774 KOps/s	15.8014 KOps/s	$\color{#35bf28}+0.48\%$
test_compile_add_one_nested[tensordict-eager]	0.4321ms	0.1702ms	5.8758 KOps/s	5.7956 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_add_one_nested[pytree-compile]	0.1208ms	46.1640μs	21.6619 KOps/s	21.7343 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_add_one_nested[pytree-eager]	0.2535ms	0.1176ms	8.5002 KOps/s	8.4349 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_copy_nested[tensordict-compile]	58.9190μs	25.9031μs	38.6054 KOps/s	36.7929 KOps/s	$\color{#35bf28}+4.93\%$
test_compile_copy_nested[tensordict-eager]	0.1385ms	59.1374μs	16.9098 KOps/s	16.5420 KOps/s	$\color{#35bf28}+2.22\%$
test_compile_copy_nested[pytree-compile]	0.2024ms	78.0110μs	12.8187 KOps/s	12.6088 KOps/s	$\color{#35bf28}+1.66\%$
test_compile_copy_nested[pytree-eager]	0.1414ms	68.3555μs	14.6294 KOps/s	14.7403 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_add_one_flat[tensordict-compile]	0.1991ms	0.1047ms	9.5497 KOps/s	9.4528 KOps/s	$\color{#35bf28}+1.02\%$
test_compile_add_one_flat[tensordict-eager]	0.4345ms	0.2151ms	4.6486 KOps/s	4.6545 KOps/s	$\color{#d91a1a}-0.13\%$
test_compile_add_one_flat[tensorclass-compile]	0.1170ms	45.5150μs	21.9708 KOps/s	21.2121 KOps/s	$\color{#35bf28}+3.58\%$
test_compile_add_one_flat[tensorclass-eager]	0.5307ms	65.1008μs	15.3608 KOps/s	15.1377 KOps/s	$\color{#35bf28}+1.47\%$
test_compile_add_one_flat[pytree-compile]	0.2492ms	0.1040ms	9.6126 KOps/s	9.8653 KOps/s	$\color{#d91a1a}-2.56\%$
test_compile_add_one_flat[pytree-eager]	0.3945ms	0.1998ms	5.0039 KOps/s	4.9375 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_add_self_flat[tensordict-eager]	0.4635ms	0.2328ms	4.2959 KOps/s	4.2654 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_add_self_flat[tensordict-compile]	0.2174ms	0.1063ms	9.4034 KOps/s	9.4601 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_add_self_flat[tensorclass-eager]	0.1429ms	59.4755μs	16.8136 KOps/s	16.9507 KOps/s	$\color{#d91a1a}-0.81\%$
test_compile_add_self_flat[tensorclass-compile]	0.5805ms	46.9837μs	21.2840 KOps/s	21.2132 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_add_self_flat[pytree-eager]	1.3405ms	0.1610ms	6.2129 KOps/s	6.1977 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_add_self_flat[pytree-compile]	0.3164ms	0.1066ms	9.3782 KOps/s	9.8276 KOps/s	$\color{#d91a1a}-4.57\%$
test_compile_copy_flat[tensordict-compile]	57.6570μs	21.4617μs	46.5947 KOps/s	48.1433 KOps/s	$\color{#d91a1a}-3.22\%$
test_compile_copy_flat[tensordict-eager]	0.1571ms	64.8621μs	15.4173 KOps/s	14.7267 KOps/s	$\color{#35bf28}+4.69\%$
test_compile_copy_flat[pytree-compile]	0.3432ms	80.9343μs	12.3557 KOps/s	12.2549 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_copy_flat[pytree-eager]	0.1484ms	68.7766μs	14.5398 KOps/s	14.4197 KOps/s	$\color{#35bf28}+0.83\%$
test_compile_assign_and_add[tensordict-compile]	0.4064ms	0.2090ms	4.7839 KOps/s	4.9334 KOps/s	$\color{#d91a1a}-3.03\%$
test_compile_assign_and_add[tensordict-eager]	2.4033ms	1.3331ms	750.1237 Ops/s	746.2972 Ops/s	$\color{#35bf28}+0.51\%$
test_compile_assign_and_add[pytree-compile]	0.4039ms	0.2050ms	4.8773 KOps/s	4.9393 KOps/s	$\color{#d91a1a}-1.26\%$
test_compile_assign_and_add[pytree-eager]	1.3134ms	0.7778ms	1.2857 KOps/s	1.2791 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_assign_and_add_stack[compile]	0.8036ms	0.4616ms	2.1663 KOps/s	2.2289 KOps/s	$\color{#d91a1a}-2.81\%$
test_compile_assign_and_add_stack[eager]	4.2872ms	2.7315ms	366.1026 Ops/s	383.1942 Ops/s	$\color{#d91a1a}-4.46\%$
test_compile_indexing[tensor-tensordict-compile]	86.2900μs	36.4504μs	27.4346 KOps/s	27.6810 KOps/s	$\color{#d91a1a}-0.89\%$
test_compile_indexing[tensor-tensordict-eager]	0.5837ms	32.9069μs	30.3887 KOps/s	30.2880 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_indexing[tensor-tensorclass-compile]	68.0170μs	28.6508μs	34.9031 KOps/s	34.0658 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_indexing[tensor-tensorclass-eager]	75.1790μs	23.2436μs	43.0226 KOps/s	42.8638 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_indexing[tensor-pytree-compile]	0.1369ms	29.8120μs	33.5435 KOps/s	33.2136 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_indexing[tensor-pytree-eager]	84.4970μs	23.0245μs	43.4320 KOps/s	43.4920 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_indexing[slice-tensordict-compile]	0.1313ms	52.2017μs	19.1564 KOps/s	19.3734 KOps/s	$\color{#d91a1a}-1.12\%$
test_compile_indexing[slice-tensordict-eager]	0.3772ms	20.3295μs	49.1895 KOps/s	51.0980 KOps/s	$\color{#d91a1a}-3.73\%$
test_compile_indexing[slice-tensorclass-compile]	88.0740μs	44.7969μs	22.3230 KOps/s	22.5106 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_indexing[slice-tensorclass-eager]	74.1380μs	19.0807μs	52.4091 KOps/s	52.7160 KOps/s	$\color{#d91a1a}-0.58\%$
test_compile_indexing[slice-pytree-compile]	0.1216ms	46.0944μs	21.6946 KOps/s	22.4190 KOps/s	$\color{#d91a1a}-3.23\%$
test_compile_indexing[slice-pytree-eager]	69.4990μs	18.7593μs	53.3068 KOps/s	54.0621 KOps/s	$\color{#d91a1a}-1.40\%$
test_compile_indexing[int-tensordict-compile]	0.1238ms	53.2438μs	18.7815 KOps/s	19.1292 KOps/s	$\color{#d91a1a}-1.82\%$
test_compile_indexing[int-tensordict-eager]	0.9250ms	20.1608μs	49.6013 KOps/s	51.2901 KOps/s	$\color{#d91a1a}-3.29\%$
test_compile_indexing[int-tensorclass-compile]	0.1318ms	45.6131μs	21.9235 KOps/s	22.0609 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_indexing[int-tensorclass-eager]	59.0300μs	18.9536μs	52.7604 KOps/s	54.2101 KOps/s	$\color{#d91a1a}-2.67\%$
test_compile_indexing[int-pytree-compile]	0.1175ms	45.8711μs	21.8002 KOps/s	22.0242 KOps/s	$\color{#d91a1a}-1.02\%$
test_compile_indexing[int-pytree-eager]	86.0600μs	19.0107μs	52.6020 KOps/s	53.8223 KOps/s	$\color{#d91a1a}-2.27\%$
test_mod_add[eager]	0.1088ms	34.9222μs	28.6351 KOps/s	29.5543 KOps/s	$\color{#d91a1a}-3.11\%$
test_mod_add[compile]	0.1190ms	48.2930μs	20.7069 KOps/s	20.9982 KOps/s	$\color{#d91a1a}-1.39\%$
test_mod_add[compile-overhead]	0.1242ms	48.8044μs	20.4899 KOps/s	20.9286 KOps/s	$\color{#d91a1a}-2.10\%$
test_mod_wrap[eager]	0.4121ms	0.2291ms	4.3642 KOps/s	4.3511 KOps/s	$\color{#35bf28}+0.30\%$
test_mod_wrap[compile]	0.3131ms	0.2071ms	4.8289 KOps/s	4.8833 KOps/s	$\color{#d91a1a}-1.12\%$
test_mod_wrap[compile-overhead]	0.4043ms	0.2051ms	4.8758 KOps/s	4.9104 KOps/s	$\color{#d91a1a}-0.71\%$
test_mod_wrap_and_backward[eager]	14.6355ms	11.6491ms	85.8435 Ops/s	85.8547 Ops/s	$\color{#d91a1a}-0.01\%$
test_mod_wrap_and_backward[compile]	20.5389ms	13.0657ms	76.5361 Ops/s	77.8908 Ops/s	$\color{#d91a1a}-1.74\%$
test_mod_wrap_and_backward[compile-overhead]	14.1885ms	12.2081ms	81.9128 Ops/s	76.0860 Ops/s	$\textbf{\color{#35bf28}+7.66\%}$
test_seq_add[eager]	0.2061ms	0.1201ms	8.3265 KOps/s	8.7010 KOps/s	$\color{#d91a1a}-4.30\%$
test_seq_add[compile]	0.1589ms	63.4588μs	15.7582 KOps/s	15.9591 KOps/s	$\color{#d91a1a}-1.26\%$
test_seq_add[compile-overhead]	0.1611ms	61.7424μs	16.1963 KOps/s	16.3633 KOps/s	$\color{#d91a1a}-1.02\%$
test_seq_wrap[eager]	1.2802ms	0.4543ms	2.2011 KOps/s	2.2730 KOps/s	$\color{#d91a1a}-3.16\%$
test_seq_wrap[compile]	0.4060ms	0.2330ms	4.2915 KOps/s	4.4298 KOps/s	$\color{#d91a1a}-3.12\%$
test_seq_wrap[compile-overhead]	0.4312ms	0.2293ms	4.3612 KOps/s	4.3843 KOps/s	$\color{#d91a1a}-0.53\%$
test_func_call_runtime[False-eager]	1.0002ms	0.5560ms	1.7986 KOps/s	1.7903 KOps/s	$\color{#35bf28}+0.47\%$
test_func_call_runtime[False-compile]	0.8768ms	0.4337ms	2.3056 KOps/s	2.3275 KOps/s	$\color{#d91a1a}-0.94\%$
test_func_call_runtime[False-compile-overhead]	0.7567ms	0.4258ms	2.3486 KOps/s	2.3530 KOps/s	$\color{#d91a1a}-0.18\%$
test_func_call_runtime[True-eager]	0.9094ms	0.7746ms	1.2909 KOps/s	1.2944 KOps/s	$\color{#d91a1a}-0.27\%$
test_func_call_runtime[True-compile]	0.9532ms	0.4681ms	2.1365 KOps/s	2.1350 KOps/s	$\color{#35bf28}+0.07\%$
test_func_call_runtime[True-compile-overhead]	0.5740ms	0.4623ms	2.1629 KOps/s	2.1499 KOps/s	$\color{#35bf28}+0.60\%$
test_func_call_cm_runtime[False-eager]	0.6632ms	0.5500ms	1.8183 KOps/s	1.8037 KOps/s	$\color{#35bf28}+0.81\%$
test_func_call_cm_runtime[False-compile]	0.5338ms	0.4245ms	2.3558 KOps/s	2.3568 KOps/s	$\color{#d91a1a}-0.04\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5516ms	0.4256ms	2.3495 KOps/s	2.3523 KOps/s	$\color{#d91a1a}-0.12\%$
test_func_call_cm_runtime[True-eager]	1.1475ms	0.9125ms	1.0959 KOps/s	1.0957 KOps/s	$\color{#35bf28}+0.02\%$
test_func_call_cm_runtime[True-compile]	0.6560ms	0.4944ms	2.0226 KOps/s	2.0271 KOps/s	$\color{#d91a1a}-0.22\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6180ms	0.4893ms	2.0438 KOps/s	2.0474 KOps/s	$\color{#d91a1a}-0.18\%$
test_vmap_func_call_cm_runtime[eager]	2.6217ms	1.9284ms	518.5553 Ops/s	522.1539 Ops/s	$\color{#d91a1a}-0.69\%$
test_vmap_func_call_cm_runtime[compile]	0.8966ms	0.5258ms	1.9017 KOps/s	1.9516 KOps/s	$\color{#d91a1a}-2.56\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9696ms	0.5259ms	1.9016 KOps/s	1.8871 KOps/s	$\color{#35bf28}+0.77\%$
test_distributed	0.3138ms	0.1318ms	7.5846 KOps/s	7.6055 KOps/s	$\color{#d91a1a}-0.27\%$
test_tdmodule	0.1026ms	27.9581μs	35.7678 KOps/s	37.5079 KOps/s	$\color{#d91a1a}-4.64\%$
test_tdmodule_dispatch	0.1216ms	49.0247μs	20.3979 KOps/s	20.2328 KOps/s	$\color{#35bf28}+0.82\%$
test_tdseq	56.5250μs	30.5623μs	32.7200 KOps/s	38.7305 KOps/s	$\textbf{\color{#d91a1a}-15.52\%}$
test_tdseq_dispatch	83.6450μs	56.4193μs	17.7244 KOps/s	19.6336 KOps/s	$\textbf{\color{#d91a1a}-9.72\%}$
test_instantiation_functorch	2.3798ms	1.5296ms	653.7473 Ops/s	649.0638 Ops/s	$\color{#35bf28}+0.72\%$
test_exec_functorch	0.4595ms	0.1810ms	5.5242 KOps/s	5.5687 KOps/s	$\color{#d91a1a}-0.80\%$
test_exec_functional_call	0.3005ms	0.1739ms	5.7503 KOps/s	5.7122 KOps/s	$\color{#35bf28}+0.67\%$
test_exec_td_decorator	0.5320ms	0.2359ms	4.2398 KOps/s	4.2192 KOps/s	$\color{#35bf28}+0.49\%$
test_vmap_mlp_speed_decorator[True-True]	0.9470ms	0.6530ms	1.5314 KOps/s	1.5130 KOps/s	$\color{#35bf28}+1.22\%$
test_vmap_mlp_speed_decorator[True-False]	0.9859ms	0.6596ms	1.5160 KOps/s	1.5155 KOps/s	$\color{#35bf28}+0.03\%$
test_vmap_mlp_speed_decorator[False-True]	0.9856ms	0.5360ms	1.8657 KOps/s	1.8762 KOps/s	$\color{#d91a1a}-0.56\%$
test_vmap_mlp_speed_decorator[False-False]	0.8146ms	0.5304ms	1.8853 KOps/s	1.8943 KOps/s	$\color{#d91a1a}-0.47\%$
test_to_module_speed[True]	2.1320ms	1.3481ms	741.7802 Ops/s	733.4602 Ops/s	$\color{#35bf28}+1.13\%$
test_to_module_speed[False]	1.7030ms	1.3142ms	760.9466 Ops/s	759.9506 Ops/s	$\color{#35bf28}+0.13\%$
test_tc_init	0.1065ms	49.0194μs	20.4001 KOps/s	20.7558 KOps/s	$\color{#d91a1a}-1.71\%$
test_tc_init_nested	0.1749ms	94.7998μs	10.5485 KOps/s	10.6274 KOps/s	$\color{#d91a1a}-0.74\%$
test_tc_first_layer_tensor	33.1620μs	1.5307μs	653.3135 KOps/s	566.6876 KOps/s	$\textbf{\color{#35bf28}+15.29\%}$
test_tc_first_layer_nontensor	52.9490μs	4.7160μs	212.0463 KOps/s	209.3287 KOps/s	$\color{#35bf28}+1.30\%$
test_tc_second_layer_tensor	31.8390μs	2.8360μs	352.6042 KOps/s	301.2706 KOps/s	$\textbf{\color{#35bf28}+17.04\%}$
test_tc_second_layer_nontensor	54.7510μs	6.0222μs	166.0515 KOps/s	162.0209 KOps/s	$\color{#35bf28}+2.49\%$
test_unbind	0.2360s	15.9162ms	62.8290 Ops/s	67.9996 Ops/s	$\textbf{\color{#d91a1a}-7.60\%}$
test_full_like	9.4367ms	8.1525ms	122.6625 Ops/s	120.3113 Ops/s	$\color{#35bf28}+1.95\%$
test_zeros_like	4.0443ms	3.1327ms	319.2102 Ops/s	317.0574 Ops/s	$\color{#35bf28}+0.68\%$
test_ones_like	4.1311ms	3.6484ms	274.0918 Ops/s	274.9187 Ops/s	$\color{#d91a1a}-0.30\%$
test_clone	6.8566ms	5.7426ms	174.1358 Ops/s	179.0243 Ops/s	$\color{#d91a1a}-2.73\%$
test_squeeze	62.7970μs	12.0796μs	82.7839 KOps/s	79.2372 KOps/s	$\color{#35bf28}+4.48\%$
test_unsqueeze	0.1543ms	92.2018μs	10.8458 KOps/s	10.6660 KOps/s	$\color{#35bf28}+1.69\%$
test_split	0.5247ms	0.1984ms	5.0393 KOps/s	4.9853 KOps/s	$\color{#35bf28}+1.08\%$
test_permute	0.2854ms	0.2068ms	4.8355 KOps/s	4.7067 KOps/s	$\color{#35bf28}+2.74\%$
test_stack	32.8539ms	25.9337ms	38.5598 Ops/s	36.9071 Ops/s	$\color{#35bf28}+4.48\%$
test_cat	32.4229ms	26.2306ms	38.1234 Ops/s	37.9093 Ops/s	$\color{#35bf28}+0.56\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}25$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	36.6500μs	11.4404μs	87.4099 KOps/s	78.9826 KOps/s	$\textbf{\color{#35bf28}+10.67\%}$
test_plain_set_stack_nested	32.7300μs	11.6939μs	85.5147 KOps/s	77.7886 KOps/s	$\textbf{\color{#35bf28}+9.93\%}$
test_plain_set_nested_inplace	37.9900μs	12.5089μs	79.9432 KOps/s	72.3482 KOps/s	$\textbf{\color{#35bf28}+10.50\%}$
test_plain_set_stack_nested_inplace	39.7910μs	12.6546μs	79.0224 KOps/s	71.6456 KOps/s	$\textbf{\color{#35bf28}+10.30\%}$
test_items	24.3510μs	2.9374μs	340.4388 KOps/s	346.3393 KOps/s	$\color{#d91a1a}-1.70\%$
test_items_nested	0.4193ms	0.3608ms	2.7719 KOps/s	2.7988 KOps/s	$\color{#d91a1a}-0.96\%$
test_items_nested_locked	0.4193ms	0.3634ms	2.7519 KOps/s	2.7802 KOps/s	$\color{#d91a1a}-1.02\%$
test_items_nested_leaf	86.6110μs	59.0272μs	16.9413 KOps/s	17.2488 KOps/s	$\color{#d91a1a}-1.78\%$
test_items_stack_nested	0.4259ms	0.3641ms	2.7469 KOps/s	2.7453 KOps/s	$\color{#35bf28}+0.06\%$
test_items_stack_nested_leaf	0.1107ms	60.3563μs	16.5683 KOps/s	16.9713 KOps/s	$\color{#d91a1a}-2.37\%$
test_items_stack_nested_locked	0.3998ms	0.3680ms	2.7176 KOps/s	2.7642 KOps/s	$\color{#d91a1a}-1.69\%$
test_keys	30.9610μs	3.4907μs	286.4785 KOps/s	289.8298 KOps/s	$\color{#d91a1a}-1.16\%$
test_keys_nested	0.1279ms	82.3020μs	12.1504 KOps/s	12.2379 KOps/s	$\color{#d91a1a}-0.71\%$
test_keys_nested_locked	0.6954ms	88.1450μs	11.3449 KOps/s	11.3893 KOps/s	$\color{#d91a1a}-0.39\%$
test_keys_nested_leaf	0.1162ms	73.5933μs	13.5882 KOps/s	13.7661 KOps/s	$\color{#d91a1a}-1.29\%$
test_keys_stack_nested	0.1220ms	82.4269μs	12.1320 KOps/s	12.1476 KOps/s	$\color{#d91a1a}-0.13\%$
test_keys_stack_nested_leaf	0.1206ms	73.6395μs	13.5797 KOps/s	13.5921 KOps/s	$\color{#d91a1a}-0.09\%$
test_keys_stack_nested_locked	0.1471ms	88.4517μs	11.3056 KOps/s	11.3171 KOps/s	$\color{#d91a1a}-0.10\%$
test_values	5.8333μs	0.8494μs	1.1773 MOps/s	1.1771 MOps/s	$\color{#35bf28}+0.02\%$
test_values_nested	92.2310μs	35.1405μs	28.4572 KOps/s	29.0452 KOps/s	$\color{#d91a1a}-2.02\%$
test_values_nested_locked	64.5800μs	36.3778μs	27.4893 KOps/s	27.8336 KOps/s	$\color{#d91a1a}-1.24\%$
test_values_nested_leaf	61.4210μs	39.7204μs	25.1760 KOps/s	25.1896 KOps/s	$\color{#d91a1a}-0.05\%$
test_values_stack_nested	63.1800μs	35.0913μs	28.4971 KOps/s	28.5492 KOps/s	$\color{#d91a1a}-0.18\%$
test_values_stack_nested_leaf	0.1124ms	39.9315μs	25.0429 KOps/s	24.9816 KOps/s	$\color{#35bf28}+0.25\%$
test_values_stack_nested_locked	64.0910μs	36.7566μs	27.2060 KOps/s	27.3210 KOps/s	$\color{#d91a1a}-0.42\%$
test_membership	1.8420μs	0.5035μs	1.9860 MOps/s	1.9558 MOps/s	$\color{#35bf28}+1.54\%$
test_membership_nested	30.3600μs	2.0988μs	476.4521 KOps/s	494.3690 KOps/s	$\color{#d91a1a}-3.62\%$
test_membership_nested_leaf	20.2805μs	2.0541μs	486.8200 KOps/s	484.1413 KOps/s	$\color{#35bf28}+0.55\%$
test_membership_stacked_nested	30.4800μs	2.0989μs	476.4345 KOps/s	476.0627 KOps/s	$\color{#35bf28}+0.08\%$
test_membership_stacked_nested_leaf	34.7500μs	2.0689μs	483.3583 KOps/s	482.6695 KOps/s	$\color{#35bf28}+0.14\%$
test_membership_nested_last	39.9400μs	3.1606μs	316.3961 KOps/s	320.5465 KOps/s	$\color{#d91a1a}-1.29\%$
test_membership_nested_leaf_last	37.7500μs	3.1619μs	316.2680 KOps/s	322.5746 KOps/s	$\color{#d91a1a}-1.96\%$
test_membership_stacked_nested_last	62.9910μs	8.1878μs	122.1322 KOps/s	323.9833 KOps/s	$\textbf{\color{#d91a1a}-62.30\%}$
test_membership_stacked_nested_leaf_last	52.3100μs	8.1870μs	122.1446 KOps/s	323.1845 KOps/s	$\textbf{\color{#d91a1a}-62.21\%}$
test_nested_getleaf	39.4010μs	6.1191μs	163.4240 KOps/s	160.0595 KOps/s	$\color{#35bf28}+2.10\%$
test_nested_get	34.4900μs	5.8834μs	169.9684 KOps/s	172.1625 KOps/s	$\color{#d91a1a}-1.27\%$
test_stacked_getleaf	36.2800μs	6.1410μs	162.8386 KOps/s	162.7804 KOps/s	$\color{#35bf28}+0.04\%$
test_stacked_get	28.7210μs	5.8367μs	171.3292 KOps/s	169.2304 KOps/s	$\color{#35bf28}+1.24\%$
test_nested_getitemleaf	27.7610μs	6.3269μs	158.0561 KOps/s	161.1561 KOps/s	$\color{#d91a1a}-1.92\%$
test_nested_getitem	29.7500μs	5.9595μs	167.7998 KOps/s	169.7736 KOps/s	$\color{#d91a1a}-1.16\%$
test_stacked_getitemleaf	41.1100μs	6.2214μs	160.7358 KOps/s	159.5481 KOps/s	$\color{#35bf28}+0.74\%$
test_stacked_getitem	26.3410μs	5.9010μs	169.4628 KOps/s	168.3725 KOps/s	$\color{#35bf28}+0.65\%$
test_lock_nested	9.0866ms	0.3937ms	2.5400 KOps/s	2.5768 KOps/s	$\color{#d91a1a}-1.43\%$
test_lock_stack_nested	0.4238ms	0.3480ms	2.8739 KOps/s	2.8447 KOps/s	$\color{#35bf28}+1.03\%$
test_unlock_nested	0.6563ms	0.3225ms	3.1010 KOps/s	3.0828 KOps/s	$\color{#35bf28}+0.59\%$
test_unlock_stack_nested	0.3510ms	0.2847ms	3.5121 KOps/s	3.4159 KOps/s	$\color{#35bf28}+2.82\%$
test_flatten_speed	0.1427ms	75.7360μs	13.2038 KOps/s	13.0860 KOps/s	$\color{#35bf28}+0.90\%$
test_unflatten_speed	0.3857ms	0.3211ms	3.1146 KOps/s	3.1470 KOps/s	$\color{#d91a1a}-1.03\%$
test_common_ops	1.7208ms	0.6004ms	1.6655 KOps/s	1.5534 KOps/s	$\textbf{\color{#35bf28}+7.22\%}$
test_creation	37.4610μs	1.7390μs	575.0508 KOps/s	568.1263 KOps/s	$\color{#35bf28}+1.22\%$
test_creation_empty	1.3271ms	6.9320μs	144.2588 KOps/s	110.6424 KOps/s	$\textbf{\color{#35bf28}+30.38\%}$
test_creation_nested_1	40.5600μs	8.6175μs	116.0436 KOps/s	93.0814 KOps/s	$\textbf{\color{#35bf28}+24.67\%}$
test_creation_nested_2	45.8500μs	11.3869μs	87.8204 KOps/s	74.6889 KOps/s	$\textbf{\color{#35bf28}+17.58\%}$
test_clone	79.2910μs	11.6558μs	85.7942 KOps/s	86.6154 KOps/s	$\color{#d91a1a}-0.95\%$
test_getitem[int]	1.0596ms	11.3207μs	88.3339 KOps/s	89.0858 KOps/s	$\color{#d91a1a}-0.84\%$
test_getitem[slice_int]	0.1153ms	22.2727μs	44.8981 KOps/s	45.7757 KOps/s	$\color{#d91a1a}-1.92\%$
test_getitem[range]	0.1312ms	39.7102μs	25.1824 KOps/s	25.3006 KOps/s	$\color{#d91a1a}-0.47\%$
test_getitem[tuple]	0.1103ms	19.1629μs	52.1842 KOps/s	53.5852 KOps/s	$\color{#d91a1a}-2.61\%$
test_getitem[list]	0.2310ms	34.7596μs	28.7690 KOps/s	28.5719 KOps/s	$\color{#35bf28}+0.69\%$
test_setitem_dim[int]	28.4600μs	19.7241μs	50.6993 KOps/s	49.1318 KOps/s	$\color{#35bf28}+3.19\%$
test_setitem_dim[slice_int]	61.9510μs	39.5091μs	25.3106 KOps/s	24.8727 KOps/s	$\color{#35bf28}+1.76\%$
test_setitem_dim[range]	87.8610μs	53.6630μs	18.6348 KOps/s	18.1803 KOps/s	$\color{#35bf28}+2.50\%$
test_setitem_dim[tuple]	65.2710μs	34.5249μs	28.9646 KOps/s	29.4994 KOps/s	$\color{#d91a1a}-1.81\%$
test_setitem	75.6710μs	15.3010μs	65.3551 KOps/s	61.1852 KOps/s	$\textbf{\color{#35bf28}+6.82\%}$
test_set	89.9010μs	15.1823μs	65.8662 KOps/s	62.6706 KOps/s	$\textbf{\color{#35bf28}+5.10\%}$
test_set_shared	1.8910ms	0.1516ms	6.5983 KOps/s	6.5886 KOps/s	$\color{#35bf28}+0.15\%$
test_update	0.2453ms	16.9731μs	58.9166 KOps/s	52.2546 KOps/s	$\textbf{\color{#35bf28}+12.75\%}$
test_update_nested	91.4710μs	23.1580μs	43.1816 KOps/s	40.0535 KOps/s	$\textbf{\color{#35bf28}+7.81\%}$
test_update__nested	1.1432ms	27.0644μs	36.9490 KOps/s	37.7265 KOps/s	$\color{#d91a1a}-2.06\%$
test_set_nested	89.0010μs	16.2728μs	61.4521 KOps/s	57.3014 KOps/s	$\textbf{\color{#35bf28}+7.24\%}$
test_set_nested_new	94.0210μs	18.4492μs	54.2030 KOps/s	50.9368 KOps/s	$\textbf{\color{#35bf28}+6.41\%}$
test_select	88.4810μs	30.2565μs	33.0507 KOps/s	31.3806 KOps/s	$\textbf{\color{#35bf28}+5.32\%}$
test_select_nested	71.5510μs	44.1152μs	22.6679 KOps/s	22.5355 KOps/s	$\color{#35bf28}+0.59\%$
test_exclude_nested	88.5500μs	63.3344μs	15.7892 KOps/s	15.8475 KOps/s	$\color{#d91a1a}-0.37\%$
test_empty[True]	0.3462ms	0.2882ms	3.4701 KOps/s	3.4733 KOps/s	$\color{#d91a1a}-0.09\%$
test_empty[False]	3.2410μs	0.8344μs	1.1985 MOps/s	1.2060 MOps/s	$\color{#d91a1a}-0.62\%$
test_to	88.2410μs	58.4890μs	17.0972 KOps/s	16.7079 KOps/s	$\color{#35bf28}+2.33\%$
test_to_nonblocking	85.2010μs	49.5970μs	20.1625 KOps/s	20.4135 KOps/s	$\color{#d91a1a}-1.23\%$
test_unbind_speed	0.2870ms	0.2449ms	4.0834 KOps/s	4.1010 KOps/s	$\color{#d91a1a}-0.43\%$
test_unbind_speed_stack0	0.3038ms	0.2422ms	4.1294 KOps/s	4.0710 KOps/s	$\color{#35bf28}+1.43\%$
test_unbind_speed_stack1	92.7437ms	0.6725ms	1.4869 KOps/s	1.4650 KOps/s	$\color{#35bf28}+1.50\%$
test_split	93.2377ms	1.6450ms	607.8902 Ops/s	613.1543 Ops/s	$\color{#d91a1a}-0.86\%$
test_chunk	95.4767ms	1.6513ms	605.5670 Ops/s	611.2966 Ops/s	$\color{#d91a1a}-0.94\%$
test_consolidate[False-None]	95.6935ms	2.9776ms	335.8402 Ops/s	335.5113 Ops/s	$\color{#35bf28}+0.10\%$
test_consolidate[default-None]	2.1654ms	1.7592ms	568.4290 Ops/s	576.0310 Ops/s	$\color{#d91a1a}-1.32\%$
test_consolidate[reduce-overhead-None]	1.9233ms	1.8020ms	554.9488 Ops/s	566.9519 Ops/s	$\color{#d91a1a}-2.12\%$
test_consolidate_njt[False-None]	6.8773ms	6.7334ms	148.5129 Ops/s	150.9138 Ops/s	$\color{#d91a1a}-1.59\%$
test_to[False-False-None]	1.9234ms	1.7750ms	563.3651 Ops/s	568.9246 Ops/s	$\color{#d91a1a}-0.98\%$
test_to[True-False-None]	1.6591ms	1.4143ms	707.0568 Ops/s	735.6691 Ops/s	$\color{#d91a1a}-3.89\%$
test_to[within-False-None]	4.5085ms	4.2643ms	234.5055 Ops/s	237.1416 Ops/s	$\color{#d91a1a}-1.11\%$
test_to[True-default-None]	5.9233ms	5.4564ms	183.2723 Ops/s	185.3888 Ops/s	$\color{#d91a1a}-1.14\%$
test_to_njt[False-False-None]	7.4990ms	7.0761ms	141.3206 Ops/s	144.0432 Ops/s	$\color{#d91a1a}-1.89\%$
test_to_njt[True-False-None]	5.9018ms	5.6029ms	178.4794 Ops/s	180.9249 Ops/s	$\color{#d91a1a}-1.35\%$
test_to_njt[within-False-None]	12.8590ms	12.4435ms	80.3634 Ops/s	82.4939 Ops/s	$\color{#d91a1a}-2.58\%$
test_creation[device0]	0.6274ms	80.5117μs	12.4206 KOps/s	12.4550 KOps/s	$\color{#d91a1a}-0.28\%$
test_creation_from_tensor	0.7256ms	83.9303μs	11.9146 KOps/s	11.7539 KOps/s	$\color{#35bf28}+1.37\%$
test_add_one[memmap_tensor0]	0.2362ms	7.3928μs	135.2664 KOps/s	138.8046 KOps/s	$\color{#d91a1a}-2.55\%$
test_contiguous[memmap_tensor0]	22.0817μs	0.4201μs	2.3806 MOps/s	2.3796 MOps/s	$\color{#35bf28}+0.04\%$
test_stack[memmap_tensor0]	33.9110μs	4.7827μs	209.0852 KOps/s	223.7509 KOps/s	$\textbf{\color{#d91a1a}-6.55\%}$
test_memmaptd_index	1.7405ms	0.2697ms	3.7074 KOps/s	3.8574 KOps/s	$\color{#d91a1a}-3.89\%$
test_memmaptd_index_astensor	0.9506ms	0.3328ms	3.0050 KOps/s	3.1160 KOps/s	$\color{#d91a1a}-3.56\%$
test_memmaptd_index_op	1.0796ms	0.6094ms	1.6409 KOps/s	1.5913 KOps/s	$\color{#35bf28}+3.12\%$
test_serialize_model	0.1313s	0.1304s	7.6694 Ops/s	7.6068 Ops/s	$\color{#35bf28}+0.82\%$
test_serialize_model_pickle	1.3504s	1.2124s	0.8248 Ops/s	0.8187 Ops/s	$\color{#35bf28}+0.74\%$
test_serialize_weights	0.1322s	0.1302s	7.6831 Ops/s	7.6809 Ops/s	$\color{#35bf28}+0.03\%$
test_serialize_weights_returnearly	0.3073s	54.0323ms	18.5074 Ops/s	22.7403 Ops/s	$\textbf{\color{#d91a1a}-18.61\%}$
test_serialize_weights_pickle	1.3533s	1.2108s	0.8259 Ops/s	0.8232 Ops/s	$\color{#35bf28}+0.33\%$
test_reshape_pytree	55.2510μs	23.0505μs	43.3830 KOps/s	43.1665 KOps/s	$\color{#35bf28}+0.50\%$
test_reshape_td	57.1210μs	27.4685μs	36.4054 KOps/s	34.1811 KOps/s	$\textbf{\color{#35bf28}+6.51\%}$
test_view_pytree	0.4414ms	22.6904μs	44.0715 KOps/s	44.0984 KOps/s	$\color{#d91a1a}-0.06\%$
test_view_td	80.9410μs	30.2264μs	33.0836 KOps/s	30.1822 KOps/s	$\textbf{\color{#35bf28}+9.61\%}$
test_unbind_pytree	61.3810μs	29.0855μs	34.3814 KOps/s	34.7172 KOps/s	$\color{#d91a1a}-0.97\%$
test_unbind_td	0.7717ms	37.6709μs	26.5457 KOps/s	26.7091 KOps/s	$\color{#d91a1a}-0.61\%$
test_split_pytree	63.4600μs	30.6313μs	32.6463 KOps/s	32.3334 KOps/s	$\color{#35bf28}+0.97\%$
test_split_td	0.9525ms	40.6013μs	24.6298 KOps/s	24.9644 KOps/s	$\color{#d91a1a}-1.34\%$
test_add_pytree	73.7310μs	36.8867μs	27.1100 KOps/s	27.8667 KOps/s	$\color{#d91a1a}-2.72\%$
test_add_td	0.4700ms	49.4831μs	20.2089 KOps/s	19.6854 KOps/s	$\color{#35bf28}+2.66\%$
test_compile_add_one_nested[tensordict-compile]	0.1821ms	0.1216ms	8.2259 KOps/s	8.0482 KOps/s	$\color{#35bf28}+2.21\%$
test_compile_add_one_nested[tensordict-eager]	0.2256ms	0.1303ms	7.6774 KOps/s	7.6204 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_add_one_nested[pytree-compile]	0.1324ms	96.7393μs	10.3371 KOps/s	10.3396 KOps/s	$\color{#d91a1a}-0.02\%$
test_compile_add_one_nested[pytree-eager]	0.5554ms	0.1536ms	6.5090 KOps/s	6.5557 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_copy_nested[tensordict-compile]	48.6800μs	23.1231μs	43.2469 KOps/s	43.7455 KOps/s	$\color{#d91a1a}-1.14\%$
test_compile_copy_nested[tensordict-eager]	0.4462ms	29.5046μs	33.8930 KOps/s	33.7529 KOps/s	$\color{#35bf28}+0.42\%$
test_compile_copy_nested[pytree-compile]	0.4573ms	64.7393μs	15.4466 KOps/s	15.2059 KOps/s	$\color{#35bf28}+1.58\%$
test_compile_copy_nested[pytree-eager]	0.4310ms	49.1680μs	20.3384 KOps/s	20.3473 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_add_one_flat[tensordict-compile]	0.1960ms	0.1435ms	6.9689 KOps/s	6.9541 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_add_one_flat[tensordict-eager]	0.6447ms	0.2165ms	4.6185 KOps/s	4.6145 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_add_one_flat[tensorclass-compile]	0.5985ms	99.1273μs	10.0880 KOps/s	10.0752 KOps/s	$\color{#35bf28}+0.13\%$
test_compile_add_one_flat[tensorclass-eager]	0.4470ms	54.5602μs	18.3284 KOps/s	18.0567 KOps/s	$\color{#35bf28}+1.50\%$
test_compile_add_one_flat[pytree-compile]	0.1764ms	0.1374ms	7.2775 KOps/s	7.3353 KOps/s	$\color{#d91a1a}-0.79\%$
test_compile_add_one_flat[pytree-eager]	0.6238ms	0.4979ms	2.0083 KOps/s	2.0287 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_add_self_flat[tensordict-eager]	0.6887ms	0.2593ms	3.8559 KOps/s	3.8697 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_add_self_flat[tensordict-compile]	0.1840ms	0.1446ms	6.9140 KOps/s	6.9673 KOps/s	$\color{#d91a1a}-0.77\%$
test_compile_add_self_flat[tensorclass-eager]	0.5027ms	65.0252μs	15.3787 KOps/s	15.4380 KOps/s	$\color{#d91a1a}-0.38\%$
test_compile_add_self_flat[tensorclass-compile]	0.1458ms	0.1003ms	9.9736 KOps/s	10.0199 KOps/s	$\color{#d91a1a}-0.46\%$
test_compile_add_self_flat[pytree-eager]	0.8156ms	0.4142ms	2.4143 KOps/s	2.4455 KOps/s	$\color{#d91a1a}-1.28\%$
test_compile_add_self_flat[pytree-compile]	0.1768ms	0.1389ms	7.1991 KOps/s	7.3799 KOps/s	$\color{#d91a1a}-2.45\%$
test_compile_copy_flat[tensordict-compile]	61.2100μs	19.6025μs	51.0138 KOps/s	50.8451 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_copy_flat[tensordict-eager]	0.4221ms	31.7821μs	31.4642 KOps/s	32.1228 KOps/s	$\color{#d91a1a}-2.05\%$
test_compile_copy_flat[pytree-compile]	0.4746ms	69.8544μs	14.3155 KOps/s	14.1834 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_copy_flat[pytree-eager]	0.4731ms	51.1286μs	19.5585 KOps/s	19.3110 KOps/s	$\color{#35bf28}+1.28\%$
test_compile_assign_and_add[tensordict-compile]	1.6609ms	0.3983ms	2.5109 KOps/s	2.1964 KOps/s	$\textbf{\color{#35bf28}+14.32\%}$
test_compile_assign_and_add[tensordict-eager]	3.1113ms	2.7179ms	367.9368 Ops/s	373.8932 Ops/s	$\color{#d91a1a}-1.59\%$
test_compile_assign_and_add[pytree-compile]	1.6287ms	0.3906ms	2.5605 KOps/s	2.1618 KOps/s	$\textbf{\color{#35bf28}+18.44\%}$
test_compile_assign_and_add[pytree-eager]	2.8601ms	2.7700ms	361.0144 Ops/s	360.2360 Ops/s	$\color{#35bf28}+0.22\%$
test_compile_indexing[tensor-tensordict-compile]	0.5858ms	0.1213ms	8.2427 KOps/s	8.2702 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_indexing[tensor-tensordict-eager]	0.5754ms	83.3100μs	12.0034 KOps/s	11.7141 KOps/s	$\color{#35bf28}+2.47\%$
test_compile_indexing[tensor-tensorclass-compile]	0.4799ms	0.1140ms	8.7750 KOps/s	9.2974 KOps/s	$\textbf{\color{#d91a1a}-5.62\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.1853ms	70.5735μs	14.1696 KOps/s	13.6296 KOps/s	$\color{#35bf28}+3.96\%$
test_compile_indexing[tensor-pytree-compile]	0.2200ms	0.1167ms	8.5686 KOps/s	9.2118 KOps/s	$\textbf{\color{#d91a1a}-6.98\%}$
test_compile_indexing[tensor-pytree-eager]	0.1291ms	74.3433μs	13.4511 KOps/s	14.1770 KOps/s	$\textbf{\color{#d91a1a}-5.12\%}$
test_compile_indexing[slice-tensordict-compile]	0.1491ms	0.1026ms	9.7438 KOps/s	9.6060 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_indexing[slice-tensordict-eager]	0.1421ms	18.1762μs	55.0170 KOps/s	55.7583 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_indexing[slice-tensorclass-compile]	0.1834ms	98.9289μs	10.1083 KOps/s	9.8647 KOps/s	$\color{#35bf28}+2.47\%$
test_compile_indexing[slice-tensorclass-eager]	44.2600μs	16.3775μs	61.0592 KOps/s	60.9724 KOps/s	$\color{#35bf28}+0.14\%$
test_compile_indexing[slice-pytree-compile]	0.1616ms	0.1028ms	9.7268 KOps/s	9.8214 KOps/s	$\color{#d91a1a}-0.96\%$
test_compile_indexing[slice-pytree-eager]	54.3610μs	16.3570μs	61.1359 KOps/s	61.6564 KOps/s	$\color{#d91a1a}-0.84\%$
test_compile_indexing[int-tensordict-compile]	0.1506ms	0.1068ms	9.3649 KOps/s	9.3568 KOps/s	$\color{#35bf28}+0.09\%$
test_compile_indexing[int-tensordict-eager]	0.5918ms	17.8918μs	55.8914 KOps/s	57.3958 KOps/s	$\color{#d91a1a}-2.62\%$
test_compile_indexing[int-tensorclass-compile]	0.1420ms	98.2904μs	10.1739 KOps/s	9.7930 KOps/s	$\color{#35bf28}+3.89\%$
test_compile_indexing[int-tensorclass-eager]	44.3400μs	16.6807μs	59.9496 KOps/s	62.2486 KOps/s	$\color{#d91a1a}-3.69\%$
test_compile_indexing[int-pytree-compile]	0.2107ms	99.5206μs	10.0482 KOps/s	9.8450 KOps/s	$\color{#35bf28}+2.06\%$
test_compile_indexing[int-pytree-eager]	57.7510μs	16.3054μs	61.3293 KOps/s	61.3078 KOps/s	$\color{#35bf28}+0.03\%$
test_mod_add[eager]	79.1510μs	38.4803μs	25.9873 KOps/s	25.5507 KOps/s	$\color{#35bf28}+1.71\%$
test_mod_add[compile]	0.3524ms	80.9162μs	12.3585 KOps/s	12.2329 KOps/s	$\color{#35bf28}+1.03\%$
test_mod_add[compile-overhead]	0.3301ms	0.1689ms	5.9216 KOps/s	5.6964 KOps/s	$\color{#35bf28}+3.95\%$
test_mod_wrap[eager]	0.3365ms	0.2525ms	3.9602 KOps/s	3.8726 KOps/s	$\color{#35bf28}+2.26\%$
test_mod_wrap[compile]	0.3789ms	0.2982ms	3.3534 KOps/s	3.2666 KOps/s	$\color{#35bf28}+2.66\%$
test_mod_wrap[compile-overhead]	7.1289ms	3.7116ms	269.4254 Ops/s	269.8335 Ops/s	$\color{#d91a1a}-0.15\%$
test_mod_wrap_and_backward[eager]	1.5266ms	1.4108ms	708.8359 Ops/s	681.3686 Ops/s	$\color{#35bf28}+4.03\%$
test_mod_wrap_and_backward[compile]	1.3987ms	1.3077ms	764.7233 Ops/s	714.4238 Ops/s	$\textbf{\color{#35bf28}+7.04\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3882ms	0.9346ms	1.0700 KOps/s	948.7086 Ops/s	$\textbf{\color{#35bf28}+12.79\%}$
test_seq_add[eager]	0.2094ms	0.1150ms	8.6964 KOps/s	8.1673 KOps/s	$\textbf{\color{#35bf28}+6.48\%}$
test_seq_add[compile]	0.1405ms	90.0658μs	11.1030 KOps/s	11.1704 KOps/s	$\color{#d91a1a}-0.60\%$
test_seq_add[compile-overhead]	0.1881ms	0.1305ms	7.6653 KOps/s	7.5727 KOps/s	$\color{#35bf28}+1.22\%$
test_seq_wrap[eager]	0.4932ms	0.4201ms	2.3803 KOps/s	2.2842 KOps/s	$\color{#35bf28}+4.20\%$
test_seq_wrap[compile]	0.3787ms	0.3063ms	3.2650 KOps/s	3.1283 KOps/s	$\color{#35bf28}+4.37\%$
test_seq_wrap[compile-overhead]	0.2992ms	0.2270ms	4.4056 KOps/s	4.3973 KOps/s	$\color{#35bf28}+0.19\%$
test_func_call_runtime[False-eager]	0.9735ms	0.7792ms	1.2834 KOps/s	1.3171 KOps/s	$\color{#d91a1a}-2.56\%$
test_func_call_runtime[False-compile]	1.0336ms	0.7622ms	1.3120 KOps/s	1.3157 KOps/s	$\color{#d91a1a}-0.28\%$
test_func_call_runtime[False-compile-overhead]	0.4137ms	0.3666ms	2.7281 KOps/s	2.7029 KOps/s	$\color{#35bf28}+0.93\%$
test_func_call_runtime[True-eager]	1.0080ms	0.9218ms	1.0848 KOps/s	1.0838 KOps/s	$\color{#35bf28}+0.10\%$
test_func_call_runtime[True-compile]	0.8719ms	0.7979ms	1.2533 KOps/s	1.2811 KOps/s	$\color{#d91a1a}-2.17\%$
test_func_call_runtime[True-compile-overhead]	0.4437ms	0.3883ms	2.5752 KOps/s	2.5659 KOps/s	$\color{#35bf28}+0.36\%$
test_func_call_cm_runtime[False-eager]	0.8060ms	0.7475ms	1.3377 KOps/s	1.3272 KOps/s	$\color{#35bf28}+0.79\%$
test_func_call_cm_runtime[False-compile]	0.8339ms	0.7655ms	1.3063 KOps/s	1.3081 KOps/s	$\color{#d91a1a}-0.14\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4189ms	0.3700ms	2.7027 KOps/s	2.6860 KOps/s	$\color{#35bf28}+0.62\%$
test_func_call_cm_runtime[True-eager]	1.1269ms	1.0261ms	974.5389 Ops/s	972.0207 Ops/s	$\color{#35bf28}+0.26\%$
test_func_call_cm_runtime[True-compile]	0.8762ms	0.8119ms	1.2317 KOps/s	1.2304 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4649ms	0.4132ms	2.4202 KOps/s	2.3947 KOps/s	$\color{#35bf28}+1.06\%$
test_vmap_func_call_cm_runtime[eager]	2.5678ms	2.1054ms	474.9709 Ops/s	471.8077 Ops/s	$\color{#35bf28}+0.67\%$
test_vmap_func_call_cm_runtime[compile]	0.9774ms	0.8258ms	1.2109 KOps/s	1.2117 KOps/s	$\color{#d91a1a}-0.06\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5022ms	0.4193ms	2.3848 KOps/s	2.3862 KOps/s	$\color{#d91a1a}-0.06\%$
test_distributed	5.3899ms	0.1816ms	5.5069 KOps/s	8.6124 KOps/s	$\textbf{\color{#d91a1a}-36.06\%}$
test_tdmodule	55.0500μs	18.7521μs	53.3273 KOps/s	50.2035 KOps/s	$\textbf{\color{#35bf28}+6.22\%}$
test_tdmodule_dispatch	82.1810μs	33.9557μs	29.4502 KOps/s	27.5687 KOps/s	$\textbf{\color{#35bf28}+6.82\%}$
test_tdseq	39.3500μs	19.9130μs	50.2185 KOps/s	49.7618 KOps/s	$\color{#35bf28}+0.92\%$
test_tdseq_dispatch	58.2300μs	37.0736μs	26.9734 KOps/s	26.0288 KOps/s	$\color{#35bf28}+3.63\%$
test_instantiation_functorch	2.4498ms	1.6139ms	619.6347 Ops/s	628.3074 Ops/s	$\color{#d91a1a}-1.38\%$
test_exec_functorch	0.2332ms	0.1507ms	6.6339 KOps/s	6.7353 KOps/s	$\color{#d91a1a}-1.50\%$
test_exec_functional_call	0.5649ms	0.1439ms	6.9494 KOps/s	6.9883 KOps/s	$\color{#d91a1a}-0.56\%$
test_exec_td_decorator	0.5886ms	0.1924ms	5.1968 KOps/s	5.2271 KOps/s	$\color{#d91a1a}-0.58\%$
test_vmap_mlp_speed_decorator[True-True]	1.1244ms	0.6922ms	1.4447 KOps/s	1.4400 KOps/s	$\color{#35bf28}+0.33\%$
test_vmap_mlp_speed_decorator[True-False]	1.1124ms	0.6891ms	1.4512 KOps/s	1.4356 KOps/s	$\color{#35bf28}+1.08\%$
test_vmap_mlp_speed_decorator[False-True]	1.0441ms	0.6058ms	1.6507 KOps/s	1.6540 KOps/s	$\color{#d91a1a}-0.20\%$
test_vmap_mlp_speed_decorator[False-False]	1.0179ms	0.6053ms	1.6520 KOps/s	1.6531 KOps/s	$\color{#d91a1a}-0.06\%$
test_vmap_transformer_speed_decorator[True-True]	19.9994ms	19.4714ms	51.3575 Ops/s	51.5289 Ops/s	$\color{#d91a1a}-0.33\%$
test_vmap_transformer_speed_decorator[True-False]	19.8563ms	19.4660ms	51.3717 Ops/s	51.5031 Ops/s	$\color{#d91a1a}-0.26\%$
test_vmap_transformer_speed_decorator[False-True]	19.8990ms	19.3698ms	51.6268 Ops/s	52.0121 Ops/s	$\color{#d91a1a}-0.74\%$
test_vmap_transformer_speed_decorator[False-False]	19.7196ms	19.3270ms	51.7411 Ops/s	51.7585 Ops/s	$\color{#d91a1a}-0.03\%$
test_to_module_speed[True]	1.0605ms	0.9802ms	1.0202 KOps/s	1.0244 KOps/s	$\color{#d91a1a}-0.41\%$
test_to_module_speed[False]	1.5015ms	0.9463ms	1.0567 KOps/s	1.0386 KOps/s	$\color{#35bf28}+1.75\%$
test_tc_init	68.8510μs	37.5688μs	26.6179 KOps/s	26.0931 KOps/s	$\color{#35bf28}+2.01\%$
test_tc_init_nested	0.4632ms	75.9536μs	13.1659 KOps/s	13.2634 KOps/s	$\color{#d91a1a}-0.73\%$
test_tc_first_layer_tensor	23.9900μs	0.7958μs	1.2566 MOps/s	1.4241 MOps/s	$\textbf{\color{#d91a1a}-11.76\%}$
test_tc_first_layer_nontensor	26.6500μs	2.3304μs	429.1041 KOps/s	423.6641 KOps/s	$\color{#35bf28}+1.28\%$
test_tc_second_layer_tensor	0.3938ms	1.5267μs	654.9986 KOps/s	704.1222 KOps/s	$\textbf{\color{#d91a1a}-6.98\%}$
test_tc_second_layer_nontensor	25.5300μs	3.0745μs	325.2565 KOps/s	329.0563 KOps/s	$\color{#d91a1a}-1.15\%$
test_unbind	0.2291s	10.2543ms	97.5201 Ops/s	142.8736 Ops/s	$\textbf{\color{#d91a1a}-31.74\%}$
test_full_like	9.3780ms	9.1231ms	109.6113 Ops/s	107.8687 Ops/s	$\color{#35bf28}+1.62\%$
test_zeros_like	4.9502ms	4.3172ms	231.6324 Ops/s	235.7975 Ops/s	$\color{#d91a1a}-1.77\%$
test_ones_like	4.9357ms	4.3248ms	231.2221 Ops/s	235.9793 Ops/s	$\color{#d91a1a}-2.02\%$
test_clone	6.7546ms	6.3860ms	156.5915 Ops/s	109.5051 Ops/s	$\textbf{\color{#35bf28}+43.00\%}$
test_squeeze	62.2000μs	9.8514μs	101.5080 KOps/s	105.5551 KOps/s	$\color{#d91a1a}-3.83\%$
test_unsqueeze	0.5257ms	76.7039μs	13.0372 KOps/s	13.6576 KOps/s	$\color{#d91a1a}-4.54\%$
test_split	0.2735ms	0.1640ms	6.0976 KOps/s	5.9747 KOps/s	$\color{#35bf28}+2.06\%$
test_permute	0.6248ms	0.1887ms	5.3000 KOps/s	5.3915 KOps/s	$\color{#d91a1a}-1.70\%$
test_stack	51.1787ms	50.5457ms	19.7841 Ops/s	19.7778 Ops/s	$\color{#35bf28}+0.03\%$
test_cat	53.0448ms	50.8939ms	19.6487 Ops/s	19.7066 Ops/s	$\color{#d91a1a}-0.29\%$