[Feature] Add missing __torch_function__

[vmoens]

Stack from ghstack (oldest at bottom):

• [Feature] UnbatchedTensor #1170
• -> [Feature] Add missing __torch_function__ #1169

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}6$. Worsened: $\large\color{#d91a1a}29$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	53.8000μs	20.7739μs	48.1374 KOps/s	49.7424 KOps/s	$\color{#d91a1a}-3.23\%$
test_plain_set_stack_nested	68.5770μs	20.8621μs	47.9338 KOps/s	49.7396 KOps/s	$\color{#d91a1a}-3.63\%$
test_plain_set_nested_inplace	56.8860μs	22.8246μs	43.8123 KOps/s	45.5541 KOps/s	$\color{#d91a1a}-3.82\%$
test_plain_set_stack_nested_inplace	61.7040μs	22.6793μs	44.0931 KOps/s	44.7143 KOps/s	$\color{#d91a1a}-1.39\%$
test_items	37.7200μs	4.3245μs	231.2389 KOps/s	236.8023 KOps/s	$\color{#d91a1a}-2.35\%$
test_items_nested	0.6921ms	0.3933ms	2.5424 KOps/s	2.4938 KOps/s	$\color{#35bf28}+1.95\%$
test_items_nested_locked	0.5134ms	0.3924ms	2.5487 KOps/s	2.4946 KOps/s	$\color{#35bf28}+2.17\%$
test_items_nested_leaf	0.1595ms	78.3559μs	12.7623 KOps/s	12.7155 KOps/s	$\color{#35bf28}+0.37\%$
test_items_stack_nested	0.5635ms	0.3929ms	2.5454 KOps/s	2.4747 KOps/s	$\color{#35bf28}+2.86\%$
test_items_stack_nested_leaf	0.1856ms	80.2289μs	12.4643 KOps/s	12.3833 KOps/s	$\color{#35bf28}+0.65\%$
test_items_stack_nested_locked	0.5747ms	0.3922ms	2.5495 KOps/s	2.4808 KOps/s	$\color{#35bf28}+2.77\%$
test_keys	29.2840μs	3.4484μs	289.9894 KOps/s	274.2992 KOps/s	$\textbf{\color{#35bf28}+5.72\%}$
test_keys_nested	0.2982ms	0.1625ms	6.1548 KOps/s	6.0843 KOps/s	$\color{#35bf28}+1.16\%$
test_keys_nested_locked	1.6589ms	0.1683ms	5.9418 KOps/s	5.8799 KOps/s	$\color{#35bf28}+1.05\%$
test_keys_nested_leaf	0.2596ms	0.1428ms	7.0030 KOps/s	6.8904 KOps/s	$\color{#35bf28}+1.64\%$
test_keys_stack_nested	0.2650ms	0.1582ms	6.3221 KOps/s	6.2590 KOps/s	$\color{#35bf28}+1.01\%$
test_keys_stack_nested_leaf	0.2958ms	0.1385ms	7.2182 KOps/s	7.1928 KOps/s	$\color{#35bf28}+0.35\%$
test_keys_stack_nested_locked	0.2633ms	0.1634ms	6.1185 KOps/s	6.0273 KOps/s	$\color{#35bf28}+1.51\%$
test_values	7.6322μs	1.0332μs	967.8645 KOps/s	963.8857 KOps/s	$\color{#35bf28}+0.41\%$
test_values_nested	0.1077ms	61.5967μs	16.2346 KOps/s	16.5157 KOps/s	$\color{#d91a1a}-1.70\%$
test_values_nested_locked	0.1089ms	61.2888μs	16.3162 KOps/s	16.4716 KOps/s	$\color{#d91a1a}-0.94\%$
test_values_nested_leaf	0.1351ms	70.7995μs	14.1244 KOps/s	13.8523 KOps/s	$\color{#35bf28}+1.96\%$
test_values_stack_nested	0.1175ms	62.4120μs	16.0226 KOps/s	15.7842 KOps/s	$\color{#35bf28}+1.51\%$
test_values_stack_nested_leaf	0.1828ms	69.8309μs	14.3203 KOps/s	14.4109 KOps/s	$\color{#d91a1a}-0.63\%$
test_values_stack_nested_locked	0.1220ms	62.4736μs	16.0068 KOps/s	15.8973 KOps/s	$\color{#35bf28}+0.69\%$
test_membership	39.3430μs	0.8702μs	1.1491 MOps/s	1.1835 MOps/s	$\color{#d91a1a}-2.90\%$
test_membership_nested	31.3780μs	2.9512μs	338.8498 KOps/s	346.5363 KOps/s	$\color{#d91a1a}-2.22\%$
test_membership_nested_leaf	39.0320μs	2.9639μs	337.3886 KOps/s	341.4360 KOps/s	$\color{#d91a1a}-1.19\%$
test_membership_stacked_nested	32.6010μs	2.9523μs	338.7199 KOps/s	348.7616 KOps/s	$\color{#d91a1a}-2.88\%$
test_membership_stacked_nested_leaf	23.0230μs	2.9200μs	342.4649 KOps/s	349.7886 KOps/s	$\color{#d91a1a}-2.09\%$
test_membership_nested_last	39.1630μs	4.3536μs	229.6934 KOps/s	230.3081 KOps/s	$\color{#d91a1a}-0.27\%$
test_membership_nested_leaf_last	35.7360μs	4.3965μs	227.4551 KOps/s	230.0698 KOps/s	$\color{#d91a1a}-1.14\%$
test_membership_stacked_nested_last	39.9640μs	13.4861μs	74.1506 KOps/s	228.7645 KOps/s	$\textbf{\color{#d91a1a}-67.59\%}$
test_membership_stacked_nested_leaf_last	48.5300μs	13.4808μs	74.1795 KOps/s	211.9089 KOps/s	$\textbf{\color{#d91a1a}-64.99\%}$
test_nested_getleaf	55.2530μs	10.5379μs	94.8955 KOps/s	93.6903 KOps/s	$\color{#35bf28}+1.29\%$
test_nested_get	54.7020μs	10.2049μs	97.9926 KOps/s	101.3345 KOps/s	$\color{#d91a1a}-3.30\%$
test_stacked_getleaf	48.2900μs	10.4828μs	95.3942 KOps/s	94.9327 KOps/s	$\color{#35bf28}+0.49\%$
test_stacked_get	45.2240μs	10.0224μs	99.7762 KOps/s	100.9444 KOps/s	$\color{#d91a1a}-1.16\%$
test_nested_getitemleaf	49.5220μs	11.2493μs	88.8944 KOps/s	89.9376 KOps/s	$\color{#d91a1a}-1.16\%$
test_nested_getitem	50.8640μs	10.5103μs	95.1446 KOps/s	95.2042 KOps/s	$\color{#d91a1a}-0.06\%$
test_stacked_getitemleaf	54.3910μs	10.8197μs	92.4242 KOps/s	91.2960 KOps/s	$\color{#35bf28}+1.24\%$
test_stacked_getitem	41.0160μs	10.3428μs	96.6852 KOps/s	97.3567 KOps/s	$\color{#d91a1a}-0.69\%$
test_lock_nested	1.9791ms	0.4557ms	2.1943 KOps/s	1.7857 KOps/s	$\textbf{\color{#35bf28}+22.88\%}$
test_lock_stack_nested	0.7108ms	0.4095ms	2.4419 KOps/s	2.3806 KOps/s	$\color{#35bf28}+2.57\%$
test_unlock_nested	0.8556ms	0.3756ms	2.6623 KOps/s	2.6499 KOps/s	$\color{#35bf28}+0.47\%$
test_unlock_stack_nested	0.5473ms	0.3310ms	3.0216 KOps/s	2.9883 KOps/s	$\color{#35bf28}+1.11\%$
test_flatten_speed	0.1803ms	99.9412μs	10.0059 KOps/s	9.9278 KOps/s	$\color{#35bf28}+0.79\%$
test_unflatten_speed	0.9143ms	0.5270ms	1.8976 KOps/s	1.9148 KOps/s	$\color{#d91a1a}-0.90\%$
test_common_ops	3.7463ms	0.8182ms	1.2223 KOps/s	1.2962 KOps/s	$\textbf{\color{#d91a1a}-5.70\%}$
test_creation	21.8800μs	2.5089μs	398.5801 KOps/s	412.8699 KOps/s	$\color{#d91a1a}-3.46\%$
test_creation_empty	63.7290μs	12.0638μs	82.8927 KOps/s	92.4108 KOps/s	$\textbf{\color{#d91a1a}-10.30\%}$
test_creation_nested_1	64.7710μs	14.8798μs	67.2054 KOps/s	71.7682 KOps/s	$\textbf{\color{#d91a1a}-6.36\%}$
test_creation_nested_2	57.0260μs	19.5417μs	51.1726 KOps/s	54.3036 KOps/s	$\textbf{\color{#d91a1a}-5.77\%}$
test_clone	0.1493ms	13.5620μs	73.7352 KOps/s	75.5527 KOps/s	$\color{#d91a1a}-2.41\%$
test_getitem[int]	1.2876ms	13.1642μs	75.9637 KOps/s	77.2159 KOps/s	$\color{#d91a1a}-1.62\%$
test_getitem[slice_int]	0.1523ms	24.4376μs	40.9205 KOps/s	38.5023 KOps/s	$\textbf{\color{#35bf28}+6.28\%}$
test_getitem[range]	0.1962ms	49.3625μs	20.2583 KOps/s	20.5451 KOps/s	$\color{#d91a1a}-1.40\%$
test_getitem[tuple]	0.1385ms	20.3687μs	49.0950 KOps/s	49.7488 KOps/s	$\color{#d91a1a}-1.31\%$
test_getitem[list]	0.4314ms	44.6295μs	22.4067 KOps/s	23.1564 KOps/s	$\color{#d91a1a}-3.24\%$
test_setitem_dim[int]	61.9150μs	24.8426μs	40.2534 KOps/s	42.2155 KOps/s	$\color{#d91a1a}-4.65\%$
test_setitem_dim[slice_int]	0.1185ms	53.2031μs	18.7959 KOps/s	20.2286 KOps/s	$\textbf{\color{#d91a1a}-7.08\%}$
test_setitem_dim[range]	0.1170ms	75.2769μs	13.2843 KOps/s	14.0284 KOps/s	$\textbf{\color{#d91a1a}-5.30\%}$
test_setitem_dim[tuple]	82.5740μs	40.1773μs	24.8897 KOps/s	25.4961 KOps/s	$\color{#d91a1a}-2.38\%$
test_setitem	0.2059ms	21.1078μs	47.3759 KOps/s	50.9039 KOps/s	$\textbf{\color{#d91a1a}-6.93\%}$
test_set	0.1475ms	20.2109μs	49.4782 KOps/s	52.1267 KOps/s	$\textbf{\color{#d91a1a}-5.08\%}$
test_set_shared	1.1855ms	0.1710ms	5.8493 KOps/s	5.8785 KOps/s	$\color{#d91a1a}-0.50\%$
test_update	0.2145ms	23.4218μs	42.6953 KOps/s	46.3485 KOps/s	$\textbf{\color{#d91a1a}-7.88\%}$
test_update_nested	0.1738ms	33.8584μs	29.5347 KOps/s	32.0270 KOps/s	$\textbf{\color{#d91a1a}-7.78\%}$
test_update__nested	0.4729ms	34.1753μs	29.2609 KOps/s	29.1592 KOps/s	$\color{#35bf28}+0.35\%$
test_set_nested	0.1512ms	22.7262μs	44.0021 KOps/s	46.4416 KOps/s	$\textbf{\color{#d91a1a}-5.25\%}$
test_set_nested_new	0.1924ms	27.4273μs	36.4600 KOps/s	38.4991 KOps/s	$\textbf{\color{#d91a1a}-5.30\%}$
test_select	0.1719ms	44.1160μs	22.6675 KOps/s	23.7116 KOps/s	$\color{#d91a1a}-4.40\%$
test_select_nested	0.1324ms	64.0812μs	15.6052 KOps/s	15.7166 KOps/s	$\color{#d91a1a}-0.71\%$
test_exclude_nested	0.2033ms	83.4915μs	11.9773 KOps/s	12.1414 KOps/s	$\color{#d91a1a}-1.35\%$
test_empty[True]	0.6246ms	0.4088ms	2.4462 KOps/s	2.4282 KOps/s	$\color{#35bf28}+0.74\%$
test_empty[False]	12.8465μs	1.3776μs	725.8800 KOps/s	742.3309 KOps/s	$\color{#d91a1a}-2.22\%$
test_unbind_speed	0.4215ms	0.2697ms	3.7077 KOps/s	3.7353 KOps/s	$\color{#d91a1a}-0.74\%$
test_unbind_speed_stack0	0.5271ms	0.2599ms	3.8475 KOps/s	3.8501 KOps/s	$\color{#d91a1a}-0.07\%$
test_unbind_speed_stack1	0.1068s	0.7804ms	1.2814 KOps/s	1.3963 KOps/s	$\textbf{\color{#d91a1a}-8.23\%}$
test_split	0.1061s	1.7960ms	556.7794 Ops/s	563.5085 Ops/s	$\color{#d91a1a}-1.19\%$
test_chunk	1.8110ms	1.6136ms	619.7180 Ops/s	568.6896 Ops/s	$\textbf{\color{#35bf28}+8.97\%}$
test_consolidate_njt[False-None]	0.1207s	9.1834ms	108.8923 Ops/s	124.0297 Ops/s	$\textbf{\color{#d91a1a}-12.20\%}$
test_creation[device0]	0.2849ms	92.3292μs	10.8308 KOps/s	11.1260 KOps/s	$\color{#d91a1a}-2.65\%$
test_creation_from_tensor	3.9012ms	95.4288μs	10.4790 KOps/s	10.8841 KOps/s	$\color{#d91a1a}-3.72\%$
test_add_one[memmap_tensor0]	0.2250ms	5.1722μs	193.3402 KOps/s	206.6561 KOps/s	$\textbf{\color{#d91a1a}-6.44\%}$
test_contiguous[memmap_tensor0]	23.6340μs	0.5260μs	1.9012 MOps/s	1.8672 MOps/s	$\color{#35bf28}+1.82\%$
test_stack[memmap_tensor0]	47.9300μs	3.4649μs	288.6117 KOps/s	289.1816 KOps/s	$\color{#d91a1a}-0.20\%$
test_memmaptd_index	1.0972ms	0.2397ms	4.1723 KOps/s	4.2399 KOps/s	$\color{#d91a1a}-1.59\%$
test_memmaptd_index_astensor	1.0611ms	0.3254ms	3.0727 KOps/s	3.0928 KOps/s	$\color{#d91a1a}-0.65\%$
test_memmaptd_index_op	1.0816ms	0.6009ms	1.6640 KOps/s	1.7365 KOps/s	$\color{#d91a1a}-4.17\%$
test_serialize_model	0.1288s	0.1182s	8.4574 Ops/s	8.4890 Ops/s	$\color{#d91a1a}-0.37\%$
test_serialize_model_pickle	0.4966s	0.4007s	2.4956 Ops/s	2.4912 Ops/s	$\color{#35bf28}+0.18\%$
test_serialize_weights	0.1267s	0.1163s	8.6015 Ops/s	8.3597 Ops/s	$\color{#35bf28}+2.89\%$
test_serialize_weights_returnearly	0.2740s	0.1760s	5.6806 Ops/s	6.2810 Ops/s	$\textbf{\color{#d91a1a}-9.56\%}$
test_serialize_weights_pickle	1.1618s	0.7406s	1.3502 Ops/s	2.5543 Ops/s	$\textbf{\color{#d91a1a}-47.14\%}$
test_serialize_weights_filesystem	0.1517s	0.1444s	6.9232 Ops/s	7.0670 Ops/s	$\color{#d91a1a}-2.03\%$
test_serialize_model_filesystem	0.1588s	0.1445s	6.9187 Ops/s	6.7235 Ops/s	$\color{#35bf28}+2.90\%$
test_reshape_pytree	61.1540μs	26.3337μs	37.9742 KOps/s	37.9883 KOps/s	$\color{#d91a1a}-0.04\%$
test_reshape_td	69.8800μs	32.5236μs	30.7469 KOps/s	30.5430 KOps/s	$\color{#35bf28}+0.67\%$
test_view_pytree	77.5050μs	26.2409μs	38.1085 KOps/s	37.9296 KOps/s	$\color{#35bf28}+0.47\%$
test_view_td	0.1178ms	37.4172μs	26.7257 KOps/s	25.7272 KOps/s	$\color{#35bf28}+3.88\%$
test_unbind_pytree	65.0110μs	29.5438μs	33.8480 KOps/s	34.0477 KOps/s	$\color{#d91a1a}-0.59\%$
test_unbind_td	0.3031ms	40.6402μs	24.6062 KOps/s	25.5758 KOps/s	$\color{#d91a1a}-3.79\%$
test_split_pytree	72.8160μs	29.1169μs	34.3443 KOps/s	34.0882 KOps/s	$\color{#35bf28}+0.75\%$
test_split_td	0.4518ms	45.5556μs	21.9512 KOps/s	22.2593 KOps/s	$\color{#d91a1a}-1.38\%$
test_add_pytree	0.1301ms	35.2958μs	28.3320 KOps/s	28.8523 KOps/s	$\color{#d91a1a}-1.80\%$
test_add_td	0.1343ms	57.2359μs	17.4716 KOps/s	18.6527 KOps/s	$\textbf{\color{#d91a1a}-6.33\%}$
test_compile_add_one_nested[tensordict-compile]	0.1237ms	63.2471μs	15.8110 KOps/s	15.4927 KOps/s	$\color{#35bf28}+2.05\%$
test_compile_add_one_nested[tensordict-eager]	0.4067ms	0.1722ms	5.8084 KOps/s	5.7794 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_add_one_nested[pytree-compile]	0.1198ms	46.2331μs	21.6295 KOps/s	21.3947 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_add_one_nested[pytree-eager]	0.1857ms	0.1172ms	8.5361 KOps/s	8.4895 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_copy_nested[tensordict-compile]	0.1071ms	26.1445μs	38.2489 KOps/s	39.1196 KOps/s	$\color{#d91a1a}-2.23\%$
test_compile_copy_nested[tensordict-eager]	0.1176ms	58.6414μs	17.0528 KOps/s	16.8583 KOps/s	$\color{#35bf28}+1.15\%$
test_compile_copy_nested[pytree-compile]	0.1537ms	77.6401μs	12.8799 KOps/s	12.0905 KOps/s	$\textbf{\color{#35bf28}+6.53\%}$
test_compile_copy_nested[pytree-eager]	0.1464ms	67.0333μs	14.9180 KOps/s	14.6902 KOps/s	$\color{#35bf28}+1.55\%$
test_compile_add_one_flat[tensordict-compile]	0.2118ms	0.1048ms	9.5416 KOps/s	9.4536 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_add_one_flat[tensordict-eager]	0.4566ms	0.2156ms	4.6378 KOps/s	4.7071 KOps/s	$\color{#d91a1a}-1.47\%$
test_compile_add_one_flat[tensorclass-compile]	0.1172ms	45.0135μs	22.2156 KOps/s	21.2490 KOps/s	$\color{#35bf28}+4.55\%$
test_compile_add_one_flat[tensorclass-eager]	0.4723ms	67.7106μs	14.7687 KOps/s	15.2002 KOps/s	$\color{#d91a1a}-2.84\%$
test_compile_add_one_flat[pytree-compile]	0.1926ms	0.1028ms	9.7244 KOps/s	9.6671 KOps/s	$\color{#35bf28}+0.59\%$
test_compile_add_one_flat[pytree-eager]	0.4572ms	0.2013ms	4.9687 KOps/s	5.0051 KOps/s	$\color{#d91a1a}-0.73\%$
test_compile_add_self_flat[tensordict-eager]	0.3467ms	0.2320ms	4.3101 KOps/s	4.3184 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_add_self_flat[tensordict-compile]	0.2182ms	0.1058ms	9.4487 KOps/s	9.4329 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_add_self_flat[tensorclass-eager]	0.1367ms	63.3176μs	15.7934 KOps/s	16.3281 KOps/s	$\color{#d91a1a}-3.27\%$
test_compile_add_self_flat[tensorclass-compile]	0.1107ms	46.6599μs	21.4317 KOps/s	21.4726 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_add_self_flat[pytree-eager]	0.5953ms	0.1590ms	6.2896 KOps/s	6.2984 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_add_self_flat[pytree-compile]	0.2056ms	0.1033ms	9.6822 KOps/s	9.7223 KOps/s	$\color{#d91a1a}-0.41\%$
test_compile_copy_flat[tensordict-compile]	72.4750μs	21.4373μs	46.6477 KOps/s	47.4646 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_copy_flat[tensordict-eager]	0.1345ms	65.6880μs	15.2235 KOps/s	14.7875 KOps/s	$\color{#35bf28}+2.95\%$
test_compile_copy_flat[pytree-compile]	0.1571ms	79.2740μs	12.6145 KOps/s	12.6214 KOps/s	$\color{#d91a1a}-0.06\%$
test_compile_copy_flat[pytree-eager]	0.1336ms	68.1989μs	14.6630 KOps/s	14.7334 KOps/s	$\color{#d91a1a}-0.48\%$
test_compile_assign_and_add[tensordict-compile]	0.4277ms	0.2147ms	4.6568 KOps/s	4.8119 KOps/s	$\color{#d91a1a}-3.22\%$
test_compile_assign_and_add[tensordict-eager]	2.0078ms	1.3274ms	753.3707 Ops/s	776.9458 Ops/s	$\color{#d91a1a}-3.03\%$
test_compile_assign_and_add[pytree-compile]	0.4094ms	0.2082ms	4.8029 KOps/s	4.8870 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_assign_and_add[pytree-eager]	1.3606ms	0.7933ms	1.2605 KOps/s	1.3099 KOps/s	$\color{#d91a1a}-3.77\%$
test_compile_assign_and_add_stack[compile]	0.6449ms	0.4590ms	2.1784 KOps/s	2.2389 KOps/s	$\color{#d91a1a}-2.70\%$
test_compile_assign_and_add_stack[eager]	3.8230ms	2.7816ms	359.5053 Ops/s	388.2824 Ops/s	$\textbf{\color{#d91a1a}-7.41\%}$
test_compile_indexing[tensor-tensordict-compile]	0.1324ms	38.1098μs	26.2400 KOps/s	27.4454 KOps/s	$\color{#d91a1a}-4.39\%$
test_compile_indexing[tensor-tensordict-eager]	0.7070ms	34.0577μs	29.3619 KOps/s	29.3161 KOps/s	$\color{#35bf28}+0.16\%$
test_compile_indexing[tensor-tensorclass-compile]	98.5430μs	29.7488μs	33.6147 KOps/s	33.4207 KOps/s	$\color{#35bf28}+0.58\%$
test_compile_indexing[tensor-tensorclass-eager]	67.0950μs	23.0171μs	43.4460 KOps/s	42.6776 KOps/s	$\color{#35bf28}+1.80\%$
test_compile_indexing[tensor-pytree-compile]	0.1074ms	30.8910μs	32.3719 KOps/s	32.8492 KOps/s	$\color{#d91a1a}-1.45\%$
test_compile_indexing[tensor-pytree-eager]	79.7180μs	23.0882μs	43.3121 KOps/s	42.7288 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_indexing[slice-tensordict-compile]	0.1179ms	53.1091μs	18.8292 KOps/s	19.6910 KOps/s	$\color{#d91a1a}-4.38\%$
test_compile_indexing[slice-tensordict-eager]	0.5802ms	20.4890μs	48.8068 KOps/s	51.2568 KOps/s	$\color{#d91a1a}-4.78\%$
test_compile_indexing[slice-tensorclass-compile]	0.1295ms	45.1207μs	22.1628 KOps/s	22.6796 KOps/s	$\color{#d91a1a}-2.28\%$
test_compile_indexing[slice-tensorclass-eager]	71.8050μs	18.7322μs	53.3840 KOps/s	54.2986 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_indexing[slice-pytree-compile]	0.1406ms	45.8062μs	21.8311 KOps/s	22.0855 KOps/s	$\color{#d91a1a}-1.15\%$
test_compile_indexing[slice-pytree-eager]	88.1040μs	18.5503μs	53.9074 KOps/s	53.6956 KOps/s	$\color{#35bf28}+0.39\%$
test_compile_indexing[int-tensordict-compile]	0.1282ms	53.4900μs	18.6951 KOps/s	19.3177 KOps/s	$\color{#d91a1a}-3.22\%$
test_compile_indexing[int-tensordict-eager]	0.9034ms	20.3666μs	49.1000 KOps/s	50.7605 KOps/s	$\color{#d91a1a}-3.27\%$
test_compile_indexing[int-tensorclass-compile]	95.3180μs	45.5694μs	21.9445 KOps/s	22.3883 KOps/s	$\color{#d91a1a}-1.98\%$
test_compile_indexing[int-tensorclass-eager]	75.4800μs	18.4332μs	54.2499 KOps/s	54.3743 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_indexing[int-pytree-compile]	0.1106ms	45.4656μs	21.9946 KOps/s	22.1675 KOps/s	$\color{#d91a1a}-0.78\%$
test_compile_indexing[int-pytree-eager]	84.4570μs	18.5723μs	53.8437 KOps/s	53.6164 KOps/s	$\color{#35bf28}+0.42\%$
test_mod_add[eager]	0.1463ms	35.3621μs	28.2788 KOps/s	29.1532 KOps/s	$\color{#d91a1a}-3.00\%$
test_mod_add[compile]	0.1019ms	48.6654μs	20.5485 KOps/s	20.6163 KOps/s	$\color{#d91a1a}-0.33\%$
test_mod_add[compile-overhead]	0.1255ms	49.6164μs	20.1546 KOps/s	20.2728 KOps/s	$\color{#d91a1a}-0.58\%$
test_mod_wrap[eager]	0.3630ms	0.2275ms	4.3962 KOps/s	4.5289 KOps/s	$\color{#d91a1a}-2.93\%$
test_mod_wrap[compile]	0.3282ms	0.2081ms	4.8047 KOps/s	4.8960 KOps/s	$\color{#d91a1a}-1.87\%$
test_mod_wrap[compile-overhead]	0.4161ms	0.2104ms	4.7522 KOps/s	4.8868 KOps/s	$\color{#d91a1a}-2.76\%$
test_mod_wrap_and_backward[eager]	22.7342ms	12.5257ms	79.8361 Ops/s	83.9039 Ops/s	$\color{#d91a1a}-4.85\%$
test_mod_wrap_and_backward[compile]	20.4316ms	14.0616ms	71.1159 Ops/s	75.0372 Ops/s	$\textbf{\color{#d91a1a}-5.23\%}$
test_mod_wrap_and_backward[compile-overhead]	15.3572ms	13.4165ms	74.5351 Ops/s	81.8808 Ops/s	$\textbf{\color{#d91a1a}-8.97\%}$
test_seq_add[eager]	0.2260ms	0.1187ms	8.4225 KOps/s	8.5666 KOps/s	$\color{#d91a1a}-1.68\%$
test_seq_add[compile]	0.1365ms	64.2160μs	15.5725 KOps/s	16.2304 KOps/s	$\color{#d91a1a}-4.05\%$
test_seq_add[compile-overhead]	0.1564ms	63.0999μs	15.8479 KOps/s	16.6093 KOps/s	$\color{#d91a1a}-4.58\%$
test_seq_wrap[eager]	0.7734ms	0.4572ms	2.1875 KOps/s	2.2591 KOps/s	$\color{#d91a1a}-3.17\%$
test_seq_wrap[compile]	0.4466ms	0.2320ms	4.3105 KOps/s	4.3804 KOps/s	$\color{#d91a1a}-1.59\%$
test_seq_wrap[compile-overhead]	0.4625ms	0.2317ms	4.3165 KOps/s	4.3827 KOps/s	$\color{#d91a1a}-1.51\%$
test_func_call_runtime[False-eager]	0.9631ms	0.5500ms	1.8183 KOps/s	1.8791 KOps/s	$\color{#d91a1a}-3.24\%$
test_func_call_runtime[False-compile]	0.5792ms	0.4299ms	2.3263 KOps/s	2.3553 KOps/s	$\color{#d91a1a}-1.23\%$
test_func_call_runtime[False-compile-overhead]	0.5592ms	0.4293ms	2.3293 KOps/s	2.3320 KOps/s	$\color{#d91a1a}-0.11\%$
test_func_call_runtime[True-eager]	1.0911ms	0.7618ms	1.3127 KOps/s	1.3315 KOps/s	$\color{#d91a1a}-1.41\%$
test_func_call_runtime[True-compile]	0.5753ms	0.4639ms	2.1557 KOps/s	2.1295 KOps/s	$\color{#35bf28}+1.23\%$
test_func_call_runtime[True-compile-overhead]	0.5732ms	0.4660ms	2.1460 KOps/s	2.1379 KOps/s	$\color{#35bf28}+0.38\%$
test_func_call_cm_runtime[False-eager]	0.9208ms	0.5481ms	1.8246 KOps/s	1.8504 KOps/s	$\color{#d91a1a}-1.39\%$
test_func_call_cm_runtime[False-compile]	0.5433ms	0.4253ms	2.3515 KOps/s	2.3385 KOps/s	$\color{#35bf28}+0.56\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5577ms	0.4249ms	2.3532 KOps/s	2.3313 KOps/s	$\color{#35bf28}+0.94\%$
test_func_call_cm_runtime[True-eager]	2.0374ms	0.9092ms	1.0999 KOps/s	1.1062 KOps/s	$\color{#d91a1a}-0.57\%$
test_func_call_cm_runtime[True-compile]	0.8695ms	0.4946ms	2.0218 KOps/s	2.0241 KOps/s	$\color{#d91a1a}-0.11\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6255ms	0.4882ms	2.0483 KOps/s	2.0190 KOps/s	$\color{#35bf28}+1.45\%$
test_vmap_func_call_cm_runtime[eager]	2.5028ms	1.9144ms	522.3602 Ops/s	515.4130 Ops/s	$\color{#35bf28}+1.35\%$
test_vmap_func_call_cm_runtime[compile]	0.9088ms	0.5198ms	1.9237 KOps/s	1.9162 KOps/s	$\color{#35bf28}+0.39\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.7182ms	0.5274ms	1.8963 KOps/s	1.9222 KOps/s	$\color{#d91a1a}-1.35\%$
test_distributed	0.3233ms	0.1230ms	8.1327 KOps/s	7.8530 KOps/s	$\color{#35bf28}+3.56\%$
test_tdmodule	70.3810μs	26.8889μs	37.1900 KOps/s	38.0870 KOps/s	$\color{#d91a1a}-2.36\%$
test_tdmodule_dispatch	76.9530μs	48.2261μs	20.7357 KOps/s	21.4374 KOps/s	$\color{#d91a1a}-3.27\%$
test_tdseq	49.1410μs	28.9327μs	34.5630 KOps/s	34.0430 KOps/s	$\color{#35bf28}+1.53\%$
test_tdseq_dispatch	85.1280μs	53.2894μs	18.7655 KOps/s	18.5308 KOps/s	$\color{#35bf28}+1.27\%$
test_instantiation_functorch	2.0329ms	1.5451ms	647.2121 Ops/s	643.4857 Ops/s	$\color{#35bf28}+0.58\%$
test_exec_functorch	0.2770ms	0.1798ms	5.5624 KOps/s	5.5988 KOps/s	$\color{#d91a1a}-0.65\%$
test_exec_functional_call	0.3114ms	0.1753ms	5.7048 KOps/s	5.7835 KOps/s	$\color{#d91a1a}-1.36\%$
test_exec_td_decorator	0.5039ms	0.2333ms	4.2871 KOps/s	4.2947 KOps/s	$\color{#d91a1a}-0.18\%$
test_vmap_mlp_speed_decorator[True-True]	0.8594ms	0.6606ms	1.5139 KOps/s	1.5475 KOps/s	$\color{#d91a1a}-2.17\%$
test_vmap_mlp_speed_decorator[True-False]	1.0593ms	0.6587ms	1.5181 KOps/s	1.5551 KOps/s	$\color{#d91a1a}-2.38\%$
test_vmap_mlp_speed_decorator[False-True]	0.8599ms	0.5360ms	1.8658 KOps/s	1.9276 KOps/s	$\color{#d91a1a}-3.21\%$
test_vmap_mlp_speed_decorator[False-False]	0.7657ms	0.5323ms	1.8786 KOps/s	1.9271 KOps/s	$\color{#d91a1a}-2.52\%$
test_to_module_speed[True]	2.1339ms	1.3388ms	746.9354 Ops/s	743.0005 Ops/s	$\color{#35bf28}+0.53\%$
test_to_module_speed[False]	1.8235ms	1.2982ms	770.3113 Ops/s	761.9770 Ops/s	$\color{#35bf28}+1.09\%$
test_tc_init	84.6970μs	46.3936μs	21.5547 KOps/s	22.3586 KOps/s	$\color{#d91a1a}-3.60\%$
test_tc_init_nested	0.1503ms	92.7745μs	10.7788 KOps/s	10.9096 KOps/s	$\color{#d91a1a}-1.20\%$
test_tc_first_layer_tensor	19.0150μs	1.6242μs	615.6875 KOps/s	660.0352 KOps/s	$\textbf{\color{#d91a1a}-6.72\%}$
test_tc_first_layer_nontensor	32.0600μs	4.7531μs	210.3908 KOps/s	207.4478 KOps/s	$\color{#35bf28}+1.42\%$
test_tc_second_layer_tensor	24.0150μs	2.9695μs	336.7581 KOps/s	350.1884 KOps/s	$\color{#d91a1a}-3.84\%$
test_tc_second_layer_nontensor	41.8880μs	6.1717μs	162.0305 KOps/s	163.9613 KOps/s	$\color{#d91a1a}-1.18\%$
test_unbind	0.2320s	14.4359ms	69.2716 Ops/s	76.9822 Ops/s	$\textbf{\color{#d91a1a}-10.02\%}$
test_full_like	18.8815ms	13.1647ms	75.9605 Ops/s	124.6634 Ops/s	$\textbf{\color{#d91a1a}-39.07\%}$
test_zeros_like	13.7611ms	7.9810ms	125.2970 Ops/s	311.3710 Ops/s	$\textbf{\color{#d91a1a}-59.76\%}$
test_ones_like	11.6858ms	8.0307ms	124.5221 Ops/s	280.9763 Ops/s	$\textbf{\color{#d91a1a}-55.68\%}$
test_clone	14.1646ms	10.0785ms	99.2210 Ops/s	180.7739 Ops/s	$\textbf{\color{#d91a1a}-45.11\%}$
test_squeeze	63.3380μs	12.1812μs	82.0935 KOps/s	79.3255 KOps/s	$\color{#35bf28}+3.49\%$
test_unsqueeze	0.2024ms	90.6855μs	11.0271 KOps/s	11.0081 KOps/s	$\color{#35bf28}+0.17\%$
test_split	0.4790ms	0.1982ms	5.0460 KOps/s	5.1560 KOps/s	$\color{#d91a1a}-2.13\%$
test_permute	0.2863ms	0.1932ms	5.1748 KOps/s	5.1893 KOps/s	$\color{#d91a1a}-0.28\%$
test_stack	30.5786ms	25.6339ms	39.0109 Ops/s	37.5328 Ops/s	$\color{#35bf28}+3.94\%$
test_cat	29.8299ms	24.9813ms	40.0300 Ops/s	37.7729 Ops/s	$\textbf{\color{#35bf28}+5.98\%}$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}52$. Worsened: $\large\color{#d91a1a}3$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	32.0200μs	11.7966μs	84.7701 KOps/s	75.0659 KOps/s	$\textbf{\color{#35bf28}+12.93\%}$
test_plain_set_stack_nested	37.7610μs	12.0285μs	83.1361 KOps/s	73.8430 KOps/s	$\textbf{\color{#35bf28}+12.59\%}$
test_plain_set_nested_inplace	43.0110μs	12.8731μs	77.6817 KOps/s	69.3486 KOps/s	$\textbf{\color{#35bf28}+12.02\%}$
test_plain_set_stack_nested_inplace	39.4910μs	12.8711μs	77.6937 KOps/s	70.9745 KOps/s	$\textbf{\color{#35bf28}+9.47\%}$
test_items	33.4900μs	2.8429μs	351.7522 KOps/s	341.1462 KOps/s	$\color{#35bf28}+3.11\%$
test_items_nested	0.4175ms	0.3639ms	2.7484 KOps/s	2.8007 KOps/s	$\color{#d91a1a}-1.87\%$
test_items_nested_locked	0.4413ms	0.3647ms	2.7419 KOps/s	2.8221 KOps/s	$\color{#d91a1a}-2.84\%$
test_items_nested_leaf	88.7710μs	58.7197μs	17.0301 KOps/s	17.1140 KOps/s	$\color{#d91a1a}-0.49\%$
test_items_stack_nested	0.4141ms	0.3603ms	2.7751 KOps/s	2.7724 KOps/s	$\color{#35bf28}+0.10\%$
test_items_stack_nested_leaf	94.2410μs	60.2529μs	16.5967 KOps/s	16.3451 KOps/s	$\color{#35bf28}+1.54\%$
test_items_stack_nested_locked	0.4188ms	0.3594ms	2.7823 KOps/s	2.7794 KOps/s	$\color{#35bf28}+0.10\%$
test_keys	34.8110μs	3.4338μs	291.2205 KOps/s	288.9931 KOps/s	$\color{#35bf28}+0.77\%$
test_keys_nested	0.1302ms	81.5420μs	12.2636 KOps/s	12.1925 KOps/s	$\color{#35bf28}+0.58\%$
test_keys_nested_locked	2.6347ms	87.9103μs	11.3752 KOps/s	11.4023 KOps/s	$\color{#d91a1a}-0.24\%$
test_keys_nested_leaf	0.1061ms	72.3465μs	13.8224 KOps/s	13.7853 KOps/s	$\color{#35bf28}+0.27\%$
test_keys_stack_nested	0.1280ms	82.8037μs	12.0768 KOps/s	12.0016 KOps/s	$\color{#35bf28}+0.63\%$
test_keys_stack_nested_leaf	0.1485ms	72.7071μs	13.7538 KOps/s	13.5332 KOps/s	$\color{#35bf28}+1.63\%$
test_keys_stack_nested_locked	0.1292ms	88.2155μs	11.3359 KOps/s	11.2478 KOps/s	$\color{#35bf28}+0.78\%$
test_values	7.4602μs	0.8474μs	1.1800 MOps/s	1.1849 MOps/s	$\color{#d91a1a}-0.42\%$
test_values_nested	79.1420μs	34.4084μs	29.0627 KOps/s	29.1286 KOps/s	$\color{#d91a1a}-0.23\%$
test_values_nested_locked	65.2810μs	36.3887μs	27.4811 KOps/s	27.7020 KOps/s	$\color{#d91a1a}-0.80\%$
test_values_nested_leaf	66.9810μs	39.2058μs	25.5064 KOps/s	25.2994 KOps/s	$\color{#35bf28}+0.82\%$
test_values_stack_nested	72.9010μs	34.5581μs	28.9367 KOps/s	28.6493 KOps/s	$\color{#35bf28}+1.00\%$
test_values_stack_nested_leaf	64.6920μs	39.7803μs	25.1381 KOps/s	24.9956 KOps/s	$\color{#35bf28}+0.57\%$
test_values_stack_nested_locked	78.0220μs	36.9530μs	27.0614 KOps/s	27.0381 KOps/s	$\color{#35bf28}+0.09\%$
test_membership	1.6695μs	0.4995μs	2.0022 MOps/s	1.9430 MOps/s	$\color{#35bf28}+3.05\%$
test_membership_nested	26.8100μs	2.0687μs	483.3937 KOps/s	508.4304 KOps/s	$\color{#d91a1a}-4.92\%$
test_membership_nested_leaf	20.1505μs	1.9947μs	501.3262 KOps/s	509.3883 KOps/s	$\color{#d91a1a}-1.58\%$
test_membership_stacked_nested	36.2910μs	2.0812μs	480.4924 KOps/s	492.5789 KOps/s	$\color{#d91a1a}-2.45\%$
test_membership_stacked_nested_leaf	32.4210μs	2.0557μs	486.4630 KOps/s	488.4536 KOps/s	$\color{#d91a1a}-0.41\%$
test_membership_nested_last	46.9610μs	3.0715μs	325.5693 KOps/s	328.5118 KOps/s	$\color{#d91a1a}-0.90\%$
test_membership_nested_leaf_last	38.2710μs	3.1051μs	322.0468 KOps/s	325.3612 KOps/s	$\color{#d91a1a}-1.02\%$
test_membership_stacked_nested_last	35.7500μs	3.5735μs	279.8414 KOps/s	281.7871 KOps/s	$\color{#d91a1a}-0.69\%$
test_membership_stacked_nested_leaf_last	43.6410μs	3.5232μs	283.8343 KOps/s	282.1248 KOps/s	$\color{#35bf28}+0.61\%$
test_nested_getleaf	32.3100μs	6.0630μs	164.9348 KOps/s	162.5365 KOps/s	$\color{#35bf28}+1.48\%$
test_nested_get	35.4710μs	5.7492μs	173.9369 KOps/s	171.4239 KOps/s	$\color{#35bf28}+1.47\%$
test_stacked_getleaf	27.4000μs	6.0778μs	164.5345 KOps/s	162.5118 KOps/s	$\color{#35bf28}+1.24\%$
test_stacked_get	51.3310μs	5.8541μs	170.8210 KOps/s	172.3957 KOps/s	$\color{#d91a1a}-0.91\%$
test_nested_getitemleaf	36.5900μs	6.1629μs	162.2603 KOps/s	160.3481 KOps/s	$\color{#35bf28}+1.19\%$
test_nested_getitem	45.9210μs	5.9311μs	168.6023 KOps/s	169.9301 KOps/s	$\color{#d91a1a}-0.78\%$
test_stacked_getitemleaf	35.8810μs	6.1811μs	161.7825 KOps/s	162.2352 KOps/s	$\color{#d91a1a}-0.28\%$
test_stacked_getitem	43.2300μs	5.8643μs	170.5228 KOps/s	167.9578 KOps/s	$\color{#35bf28}+1.53\%$
test_lock_nested	4.4931ms	0.3729ms	2.6819 KOps/s	2.6236 KOps/s	$\color{#35bf28}+2.22\%$
test_lock_stack_nested	0.4045ms	0.3424ms	2.9202 KOps/s	2.8355 KOps/s	$\color{#35bf28}+2.99\%$
test_unlock_nested	0.7509ms	0.3115ms	3.2100 KOps/s	3.0922 KOps/s	$\color{#35bf28}+3.81\%$
test_unlock_stack_nested	0.3198ms	0.2803ms	3.5679 KOps/s	3.4430 KOps/s	$\color{#35bf28}+3.63\%$
test_flatten_speed	0.1071ms	74.8190μs	13.3656 KOps/s	13.0720 KOps/s	$\color{#35bf28}+2.25\%$
test_unflatten_speed	0.3928ms	0.3239ms	3.0875 KOps/s	3.0791 KOps/s	$\color{#35bf28}+0.27\%$
test_common_ops	1.6388ms	0.5903ms	1.6941 KOps/s	1.5261 KOps/s	$\textbf{\color{#35bf28}+11.00\%}$
test_creation	0.1867ms	1.7469μs	572.4531 KOps/s	584.5570 KOps/s	$\color{#d91a1a}-2.07\%$
test_creation_empty	35.6200μs	7.2909μs	137.1569 KOps/s	97.6263 KOps/s	$\textbf{\color{#35bf28}+40.49\%}$
test_creation_nested_1	46.3800μs	8.9783μs	111.3791 KOps/s	84.0858 KOps/s	$\textbf{\color{#35bf28}+32.46\%}$
test_creation_nested_2	36.0200μs	11.6623μs	85.7465 KOps/s	68.3968 KOps/s	$\textbf{\color{#35bf28}+25.37\%}$
test_clone	0.1260ms	10.4319μs	95.8600 KOps/s	88.2050 KOps/s	$\textbf{\color{#35bf28}+8.68\%}$
test_getitem[int]	72.4510μs	10.4837μs	95.3861 KOps/s	93.2847 KOps/s	$\color{#35bf28}+2.25\%$
test_getitem[slice_int]	1.1669ms	20.8444μs	47.9746 KOps/s	47.0398 KOps/s	$\color{#35bf28}+1.99\%$
test_getitem[range]	0.1283ms	36.6675μs	27.2721 KOps/s	26.1615 KOps/s	$\color{#35bf28}+4.25\%$
test_getitem[tuple]	0.1113ms	18.2118μs	54.9095 KOps/s	53.7871 KOps/s	$\color{#35bf28}+2.09\%$
test_getitem[list]	0.1262ms	31.6957μs	31.5500 KOps/s	29.6013 KOps/s	$\textbf{\color{#35bf28}+6.58\%}$
test_setitem_dim[int]	37.4810μs	18.0178μs	55.5006 KOps/s	51.8179 KOps/s	$\textbf{\color{#35bf28}+7.11\%}$
test_setitem_dim[slice_int]	65.6610μs	37.8281μs	26.4354 KOps/s	25.6835 KOps/s	$\color{#35bf28}+2.93\%$
test_setitem_dim[range]	72.6110μs	50.3261μs	19.8704 KOps/s	18.8356 KOps/s	$\textbf{\color{#35bf28}+5.49\%}$
test_setitem_dim[tuple]	60.3810μs	31.0971μs	32.1574 KOps/s	30.4546 KOps/s	$\textbf{\color{#35bf28}+5.59\%}$
test_setitem	0.1145ms	14.4893μs	69.0162 KOps/s	58.8688 KOps/s	$\textbf{\color{#35bf28}+17.24\%}$
test_set	0.1291ms	14.2249μs	70.2991 KOps/s	59.9512 KOps/s	$\textbf{\color{#35bf28}+17.26\%}$
test_set_shared	1.5047ms	0.1511ms	6.6163 KOps/s	6.5235 KOps/s	$\color{#35bf28}+1.42\%$
test_update	0.3056ms	16.2118μs	61.6834 KOps/s	48.9977 KOps/s	$\textbf{\color{#35bf28}+25.89\%}$
test_update_nested	63.5810μs	21.9212μs	45.6179 KOps/s	38.2943 KOps/s	$\textbf{\color{#35bf28}+19.12\%}$
test_update__nested	0.4426ms	26.3620μs	37.9334 KOps/s	36.7790 KOps/s	$\color{#35bf28}+3.14\%$
test_set_nested	0.1096ms	16.4001μs	60.9753 KOps/s	56.5433 KOps/s	$\textbf{\color{#35bf28}+7.84\%}$
test_set_nested_new	0.1114ms	18.1625μs	55.0585 KOps/s	49.8896 KOps/s	$\textbf{\color{#35bf28}+10.36\%}$
test_select	0.1235ms	28.9487μs	34.5439 KOps/s	31.1269 KOps/s	$\textbf{\color{#35bf28}+10.98\%}$
test_select_nested	73.6710μs	44.5303μs	22.4566 KOps/s	22.9048 KOps/s	$\color{#d91a1a}-1.96\%$
test_exclude_nested	96.0310μs	63.6664μs	15.7069 KOps/s	15.8586 KOps/s	$\color{#d91a1a}-0.96\%$
test_empty[True]	0.3714ms	0.2895ms	3.4542 KOps/s	3.4880 KOps/s	$\color{#d91a1a}-0.97\%$
test_empty[False]	3.3311μs	0.8281μs	1.2075 MOps/s	1.2091 MOps/s	$\color{#d91a1a}-0.13\%$
test_to	84.9020μs	55.1585μs	18.1296 KOps/s	17.3097 KOps/s	$\color{#35bf28}+4.74\%$
test_to_nonblocking	95.8220μs	47.4503μs	21.0747 KOps/s	20.5877 KOps/s	$\color{#35bf28}+2.37\%$
test_unbind_speed	1.7349ms	0.2361ms	4.2349 KOps/s	4.1678 KOps/s	$\color{#35bf28}+1.61\%$
test_unbind_speed_stack0	0.3102ms	0.2356ms	4.2449 KOps/s	4.0897 KOps/s	$\color{#35bf28}+3.79\%$
test_unbind_speed_stack1	92.8162ms	0.6645ms	1.5050 KOps/s	1.4772 KOps/s	$\color{#35bf28}+1.88\%$
test_split	93.4323ms	1.5627ms	639.9080 Ops/s	622.5076 Ops/s	$\color{#35bf28}+2.80\%$
test_chunk	95.7041ms	1.5751ms	634.8895 Ops/s	620.6517 Ops/s	$\color{#35bf28}+2.29\%$
test_consolidate[False-None]	96.5292ms	2.9073ms	343.9607 Ops/s	343.4048 Ops/s	$\color{#35bf28}+0.16\%$
test_consolidate[default-None]	1.8019ms	1.6875ms	592.5809 Ops/s	600.0391 Ops/s	$\color{#d91a1a}-1.24\%$
test_consolidate[reduce-overhead-None]	1.8656ms	1.7254ms	579.5636 Ops/s	590.1325 Ops/s	$\color{#d91a1a}-1.79\%$
test_consolidate_njt[False-None]	6.9741ms	6.5857ms	151.8438 Ops/s	156.3867 Ops/s	$\color{#d91a1a}-2.90\%$
test_to[False-False-None]	1.7666ms	1.6968ms	589.3332 Ops/s	584.8969 Ops/s	$\color{#35bf28}+0.76\%$
test_to[True-False-None]	1.5301ms	1.2928ms	773.5433 Ops/s	762.3038 Ops/s	$\color{#35bf28}+1.47\%$
test_to[within-False-None]	4.3253ms	4.0499ms	246.9171 Ops/s	241.6282 Ops/s	$\color{#35bf28}+2.19\%$
test_to[True-default-None]	5.4370ms	5.1917ms	192.6163 Ops/s	192.9919 Ops/s	$\color{#d91a1a}-0.19\%$
test_to_njt[False-False-None]	6.9437ms	6.8309ms	146.3930 Ops/s	146.6726 Ops/s	$\color{#d91a1a}-0.19\%$
test_to_njt[True-False-None]	5.9479ms	5.4157ms	184.6490 Ops/s	181.8270 Ops/s	$\color{#35bf28}+1.55\%$
test_to_njt[within-False-None]	12.2815ms	12.0208ms	83.1894 Ops/s	82.5779 Ops/s	$\color{#35bf28}+0.74\%$
test_creation[device0]	0.6385ms	80.7020μs	12.3913 KOps/s	12.1392 KOps/s	$\color{#35bf28}+2.08\%$
test_creation_from_tensor	0.5003ms	83.4912μs	11.9773 KOps/s	12.0013 KOps/s	$\color{#d91a1a}-0.20\%$
test_add_one[memmap_tensor0]	0.4292ms	6.4740μs	154.4645 KOps/s	143.0150 KOps/s	$\textbf{\color{#35bf28}+8.01\%}$
test_contiguous[memmap_tensor0]	2.4580μs	0.4031μs	2.4810 MOps/s	2.3757 MOps/s	$\color{#35bf28}+4.44\%$
test_stack[memmap_tensor0]	40.6210μs	4.3221μs	231.3710 KOps/s	225.2567 KOps/s	$\color{#35bf28}+2.71\%$
test_memmaptd_index	2.0382ms	0.2467ms	4.0541 KOps/s	3.9673 KOps/s	$\color{#35bf28}+2.19\%$
test_memmaptd_index_astensor	0.9516ms	0.3065ms	3.2627 KOps/s	3.1875 KOps/s	$\color{#35bf28}+2.36\%$
test_memmaptd_index_op	0.9960ms	0.5523ms	1.8106 KOps/s	1.5912 KOps/s	$\textbf{\color{#35bf28}+13.79\%}$
test_serialize_model	0.1325s	0.1317s	7.5932 Ops/s	7.6085 Ops/s	$\color{#d91a1a}-0.20\%$
test_serialize_model_pickle	1.3591s	1.2160s	0.8224 Ops/s	0.8248 Ops/s	$\color{#d91a1a}-0.29\%$
test_serialize_weights	0.1307s	0.1298s	7.7024 Ops/s	7.6750 Ops/s	$\color{#35bf28}+0.36\%$
test_serialize_weights_returnearly	0.3284s	53.2505ms	18.7792 Ops/s	14.3516 Ops/s	$\textbf{\color{#35bf28}+30.85\%}$
test_serialize_weights_pickle	1.3766s	1.2168s	0.8218 Ops/s	0.8234 Ops/s	$\color{#d91a1a}-0.19\%$
test_reshape_pytree	65.3510μs	21.7350μs	46.0087 KOps/s	42.9900 KOps/s	$\textbf{\color{#35bf28}+7.02\%}$
test_reshape_td	57.2710μs	26.4972μs	37.7399 KOps/s	36.1774 KOps/s	$\color{#35bf28}+4.32\%$
test_view_pytree	51.4110μs	21.6890μs	46.1063 KOps/s	44.7535 KOps/s	$\color{#35bf28}+3.02\%$
test_view_td	64.5320μs	29.7061μs	33.6631 KOps/s	30.9914 KOps/s	$\textbf{\color{#35bf28}+8.62\%}$
test_unbind_pytree	68.9120μs	27.7236μs	36.0704 KOps/s	35.4405 KOps/s	$\color{#35bf28}+1.78\%$
test_unbind_td	0.5786ms	35.9478μs	27.8181 KOps/s	26.7701 KOps/s	$\color{#35bf28}+3.91\%$
test_split_pytree	61.5910μs	29.2164μs	34.2274 KOps/s	33.3392 KOps/s	$\color{#35bf28}+2.66\%$
test_split_td	0.7434ms	37.5755μs	26.6131 KOps/s	25.2341 KOps/s	$\textbf{\color{#35bf28}+5.46\%}$
test_add_pytree	61.4220μs	33.4491μs	29.8962 KOps/s	28.0275 KOps/s	$\textbf{\color{#35bf28}+6.67\%}$
test_add_td	96.0210μs	47.2575μs	21.1607 KOps/s	18.1022 KOps/s	$\textbf{\color{#35bf28}+16.90\%}$
test_compile_add_one_nested[tensordict-compile]	0.1719ms	0.1195ms	8.3655 KOps/s	8.1092 KOps/s	$\color{#35bf28}+3.16\%$
test_compile_add_one_nested[tensordict-eager]	0.2279ms	0.1321ms	7.5675 KOps/s	7.3289 KOps/s	$\color{#35bf28}+3.26\%$
test_compile_add_one_nested[pytree-compile]	0.1942ms	97.1268μs	10.2958 KOps/s	10.2663 KOps/s	$\color{#35bf28}+0.29\%$
test_compile_add_one_nested[pytree-eager]	1.3633ms	0.1487ms	6.7255 KOps/s	6.4380 KOps/s	$\color{#35bf28}+4.47\%$
test_compile_copy_nested[tensordict-compile]	66.5210μs	23.1496μs	43.1973 KOps/s	43.4256 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_copy_nested[tensordict-eager]	0.1018ms	29.4728μs	33.9296 KOps/s	33.7882 KOps/s	$\color{#35bf28}+0.42\%$
test_compile_copy_nested[pytree-compile]	0.1055ms	65.0206μs	15.3797 KOps/s	15.2035 KOps/s	$\color{#35bf28}+1.16\%$
test_compile_copy_nested[pytree-eager]	83.2910μs	49.0406μs	20.3913 KOps/s	19.9989 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_add_one_flat[tensordict-compile]	0.1822ms	0.1420ms	7.0403 KOps/s	7.0647 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_add_one_flat[tensordict-eager]	0.3187ms	0.2171ms	4.6070 KOps/s	4.6524 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_add_one_flat[tensorclass-compile]	0.1371ms	98.9602μs	10.1051 KOps/s	9.8751 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_add_one_flat[tensorclass-eager]	0.1124ms	54.6572μs	18.2959 KOps/s	17.9259 KOps/s	$\color{#35bf28}+2.06\%$
test_compile_add_one_flat[pytree-compile]	0.1735ms	0.1350ms	7.4048 KOps/s	7.3526 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_add_one_flat[pytree-eager]	0.5475ms	0.4818ms	2.0757 KOps/s	1.9742 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_compile_add_self_flat[tensordict-eager]	0.3837ms	0.2602ms	3.8436 KOps/s	3.8179 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_add_self_flat[tensordict-compile]	0.2293ms	0.1442ms	6.9352 KOps/s	6.9785 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_add_self_flat[tensorclass-eager]	0.1657ms	65.7095μs	15.2185 KOps/s	14.4351 KOps/s	$\textbf{\color{#35bf28}+5.43\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1768ms	99.6340μs	10.0367 KOps/s	10.2186 KOps/s	$\color{#d91a1a}-1.78\%$
test_compile_add_self_flat[pytree-eager]	0.4528ms	0.4062ms	2.4618 KOps/s	2.4215 KOps/s	$\color{#35bf28}+1.67\%$
test_compile_add_self_flat[pytree-compile]	0.1773ms	0.1357ms	7.3687 KOps/s	7.4643 KOps/s	$\color{#d91a1a}-1.28\%$
test_compile_copy_flat[tensordict-compile]	54.5010μs	18.7877μs	53.2264 KOps/s	55.3228 KOps/s	$\color{#d91a1a}-3.79\%$
test_compile_copy_flat[tensordict-eager]	59.4810μs	31.2555μs	31.9944 KOps/s	31.7527 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_copy_flat[pytree-compile]	0.1746ms	69.4981μs	14.3889 KOps/s	14.2975 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_copy_flat[pytree-eager]	74.8310μs	51.1521μs	19.5496 KOps/s	19.1530 KOps/s	$\color{#35bf28}+2.07\%$
test_compile_assign_and_add[tensordict-compile]	1.6514ms	0.3952ms	2.5306 KOps/s	2.2094 KOps/s	$\textbf{\color{#35bf28}+14.54\%}$
test_compile_assign_and_add[tensordict-eager]	2.6996ms	2.5643ms	389.9701 Ops/s	376.3144 Ops/s	$\color{#35bf28}+3.63\%$
test_compile_assign_and_add[pytree-compile]	1.5592ms	0.4250ms	2.3529 KOps/s	2.2839 KOps/s	$\color{#35bf28}+3.02\%$
test_compile_assign_and_add[pytree-eager]	2.7214ms	2.6123ms	382.8091 Ops/s	373.7672 Ops/s	$\color{#35bf28}+2.42\%$
test_compile_indexing[tensor-tensordict-compile]	0.5201ms	0.1158ms	8.6387 KOps/s	8.4568 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_indexing[tensor-tensordict-eager]	0.5686ms	77.6105μs	12.8848 KOps/s	11.8335 KOps/s	$\textbf{\color{#35bf28}+8.88\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.4405ms	0.1076ms	9.2908 KOps/s	9.0625 KOps/s	$\color{#35bf28}+2.52\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1097ms	68.0457μs	14.6960 KOps/s	14.1117 KOps/s	$\color{#35bf28}+4.14\%$
test_compile_indexing[tensor-pytree-compile]	0.1582ms	0.1080ms	9.2595 KOps/s	9.0284 KOps/s	$\color{#35bf28}+2.56\%$
test_compile_indexing[tensor-pytree-eager]	0.1428ms	71.3303μs	14.0193 KOps/s	13.9844 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_indexing[slice-tensordict-compile]	0.1454ms	0.1004ms	9.9581 KOps/s	9.6243 KOps/s	$\color{#35bf28}+3.47\%$
test_compile_indexing[slice-tensordict-eager]	0.1419ms	17.9505μs	55.7087 KOps/s	57.3932 KOps/s	$\color{#d91a1a}-2.93\%$
test_compile_indexing[slice-tensorclass-compile]	0.1368ms	94.6133μs	10.5693 KOps/s	10.3344 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_indexing[slice-tensorclass-eager]	46.2810μs	15.7461μs	63.5077 KOps/s	60.0322 KOps/s	$\textbf{\color{#35bf28}+5.79\%}$
test_compile_indexing[slice-pytree-compile]	0.1431ms	94.4123μs	10.5918 KOps/s	10.0584 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_compile_indexing[slice-pytree-eager]	45.1800μs	15.6225μs	64.0104 KOps/s	62.7336 KOps/s	$\color{#35bf28}+2.04\%$
test_compile_indexing[int-tensordict-compile]	0.1665ms	99.1223μs	10.0885 KOps/s	9.5418 KOps/s	$\textbf{\color{#35bf28}+5.73\%}$
test_compile_indexing[int-tensordict-eager]	0.5840ms	17.3788μs	57.5415 KOps/s	58.5034 KOps/s	$\color{#d91a1a}-1.64\%$
test_compile_indexing[int-tensorclass-compile]	0.1492ms	97.5183μs	10.2545 KOps/s	9.9986 KOps/s	$\color{#35bf28}+2.56\%$
test_compile_indexing[int-tensorclass-eager]	0.1653ms	15.6211μs	64.0161 KOps/s	62.3401 KOps/s	$\color{#35bf28}+2.69\%$
test_compile_indexing[int-pytree-compile]	0.1388ms	96.5968μs	10.3523 KOps/s	9.9823 KOps/s	$\color{#35bf28}+3.71\%$
test_compile_indexing[int-pytree-eager]	0.1182ms	17.7892μs	56.2138 KOps/s	63.5156 KOps/s	$\textbf{\color{#d91a1a}-11.50\%}$
test_mod_add[eager]	78.6910μs	36.5454μs	27.3633 KOps/s	24.1129 KOps/s	$\textbf{\color{#35bf28}+13.48\%}$
test_mod_add[compile]	0.1315ms	78.4092μs	12.7536 KOps/s	12.4465 KOps/s	$\color{#35bf28}+2.47\%$
test_mod_add[compile-overhead]	0.3313ms	0.1747ms	5.7225 KOps/s	5.6843 KOps/s	$\color{#35bf28}+0.67\%$
test_mod_wrap[eager]	0.3186ms	0.2439ms	4.1008 KOps/s	3.9717 KOps/s	$\color{#35bf28}+3.25\%$
test_mod_wrap[compile]	0.3484ms	0.2875ms	3.4779 KOps/s	3.5379 KOps/s	$\color{#d91a1a}-1.70\%$
test_mod_wrap[compile-overhead]	7.0850ms	3.7695ms	265.2882 Ops/s	278.7667 Ops/s	$\color{#d91a1a}-4.84\%$
test_mod_wrap_and_backward[eager]	1.4821ms	1.3438ms	744.1623 Ops/s	687.4526 Ops/s	$\textbf{\color{#35bf28}+8.25\%}$
test_mod_wrap_and_backward[compile]	1.4024ms	1.2576ms	795.1784 Ops/s	732.2027 Ops/s	$\textbf{\color{#35bf28}+8.60\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3885ms	0.9303ms	1.0749 KOps/s	964.1262 Ops/s	$\textbf{\color{#35bf28}+11.49\%}$
test_seq_add[eager]	0.5227ms	0.1149ms	8.7048 KOps/s	8.2191 KOps/s	$\textbf{\color{#35bf28}+5.91\%}$
test_seq_add[compile]	0.4855ms	87.7108μs	11.4011 KOps/s	11.4089 KOps/s	$\color{#d91a1a}-0.07\%$
test_seq_add[compile-overhead]	0.1824ms	0.1292ms	7.7408 KOps/s	7.3493 KOps/s	$\textbf{\color{#35bf28}+5.33\%}$
test_seq_wrap[eager]	0.8440ms	0.4111ms	2.4325 KOps/s	2.2947 KOps/s	$\textbf{\color{#35bf28}+6.01\%}$
test_seq_wrap[compile]	0.7152ms	0.2930ms	3.4129 KOps/s	3.2977 KOps/s	$\color{#35bf28}+3.50\%$
test_seq_wrap[compile-overhead]	0.2764ms	0.2235ms	4.4741 KOps/s	4.4243 KOps/s	$\color{#35bf28}+1.13\%$
test_func_call_runtime[False-eager]	1.2205ms	0.7446ms	1.3431 KOps/s	1.3559 KOps/s	$\color{#d91a1a}-0.94\%$
test_func_call_runtime[False-compile]	1.1619ms	0.7305ms	1.3689 KOps/s	1.3084 KOps/s	$\color{#35bf28}+4.62\%$
test_func_call_runtime[False-compile-overhead]	0.8220ms	0.3614ms	2.7673 KOps/s	2.7620 KOps/s	$\color{#35bf28}+0.19\%$
test_func_call_runtime[True-eager]	1.3205ms	0.8900ms	1.1236 KOps/s	1.1081 KOps/s	$\color{#35bf28}+1.40\%$
test_func_call_runtime[True-compile]	1.1848ms	0.7511ms	1.3314 KOps/s	1.3149 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_runtime[True-compile-overhead]	0.5165ms	0.3792ms	2.6371 KOps/s	2.6048 KOps/s	$\color{#35bf28}+1.24\%$
test_func_call_cm_runtime[False-eager]	0.8330ms	0.7192ms	1.3904 KOps/s	1.3541 KOps/s	$\color{#35bf28}+2.69\%$
test_func_call_cm_runtime[False-compile]	0.8946ms	0.7319ms	1.3662 KOps/s	1.3410 KOps/s	$\color{#35bf28}+1.88\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4374ms	0.3605ms	2.7743 KOps/s	2.7391 KOps/s	$\color{#35bf28}+1.28\%$
test_func_call_cm_runtime[True-eager]	1.1379ms	0.9899ms	1.0102 KOps/s	984.7168 Ops/s	$\color{#35bf28}+2.59\%$
test_func_call_cm_runtime[True-compile]	0.9213ms	0.7746ms	1.2910 KOps/s	1.2505 KOps/s	$\color{#35bf28}+3.24\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4832ms	0.4069ms	2.4576 KOps/s	2.4324 KOps/s	$\color{#35bf28}+1.04\%$
test_vmap_func_call_cm_runtime[eager]	2.5443ms	2.0746ms	482.0159 Ops/s	478.7462 Ops/s	$\color{#35bf28}+0.68\%$
test_vmap_func_call_cm_runtime[compile]	1.2398ms	0.7902ms	1.2656 KOps/s	1.2409 KOps/s	$\color{#35bf28}+1.99\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.8395ms	0.4097ms	2.4407 KOps/s	2.4207 KOps/s	$\color{#35bf28}+0.83\%$
test_distributed	4.1250ms	0.3156ms	3.1690 KOps/s	8.4237 KOps/s	$\textbf{\color{#d91a1a}-62.38\%}$
test_tdmodule	52.2110μs	19.2327μs	51.9949 KOps/s	46.1873 KOps/s	$\textbf{\color{#35bf28}+12.57\%}$
test_tdmodule_dispatch	78.6810μs	34.8571μs	28.6885 KOps/s	26.1000 KOps/s	$\textbf{\color{#35bf28}+9.92\%}$
test_tdseq	50.6100μs	20.2736μs	49.3253 KOps/s	45.1637 KOps/s	$\textbf{\color{#35bf28}+9.21\%}$
test_tdseq_dispatch	68.4210μs	37.3348μs	26.7847 KOps/s	24.7387 KOps/s	$\textbf{\color{#35bf28}+8.27\%}$
test_instantiation_functorch	1.7576ms	1.5603ms	640.9050 Ops/s	635.8051 Ops/s	$\color{#35bf28}+0.80\%$
test_exec_functorch	0.1877ms	0.1424ms	7.0235 KOps/s	6.7321 KOps/s	$\color{#35bf28}+4.33\%$
test_exec_functional_call	0.1708ms	0.1336ms	7.4860 KOps/s	7.1473 KOps/s	$\color{#35bf28}+4.74\%$
test_exec_td_decorator	0.3684ms	0.1808ms	5.5307 KOps/s	5.2908 KOps/s	$\color{#35bf28}+4.53\%$
test_vmap_mlp_speed_decorator[True-True]	0.7866ms	0.6937ms	1.4414 KOps/s	1.4519 KOps/s	$\color{#d91a1a}-0.72\%$
test_vmap_mlp_speed_decorator[True-False]	0.8474ms	0.6851ms	1.4595 KOps/s	1.4464 KOps/s	$\color{#35bf28}+0.91\%$
test_vmap_mlp_speed_decorator[False-True]	0.7044ms	0.5932ms	1.6857 KOps/s	1.6748 KOps/s	$\color{#35bf28}+0.65\%$
test_vmap_mlp_speed_decorator[False-False]	0.7099ms	0.5944ms	1.6824 KOps/s	1.6779 KOps/s	$\color{#35bf28}+0.27\%$
test_vmap_transformer_speed_decorator[True-True]	19.1699ms	19.0453ms	52.5063 Ops/s	52.2652 Ops/s	$\color{#35bf28}+0.46\%$
test_vmap_transformer_speed_decorator[True-False]	19.7993ms	19.0641ms	52.4547 Ops/s	52.3178 Ops/s	$\color{#35bf28}+0.26\%$
test_vmap_transformer_speed_decorator[False-True]	19.1539ms	19.0008ms	52.6295 Ops/s	52.8868 Ops/s	$\color{#d91a1a}-0.49\%$
test_vmap_transformer_speed_decorator[False-False]	19.7736ms	18.9958ms	52.6431 Ops/s	52.8790 Ops/s	$\color{#d91a1a}-0.45\%$
test_to_module_speed[True]	1.1615ms	0.9763ms	1.0243 KOps/s	1.0253 KOps/s	$\color{#d91a1a}-0.11\%$
test_to_module_speed[False]	1.0647ms	0.9566ms	1.0454 KOps/s	1.0389 KOps/s	$\color{#35bf28}+0.62\%$
test_tc_init	69.8110μs	35.8312μs	27.9086 KOps/s	25.7467 KOps/s	$\textbf{\color{#35bf28}+8.40\%}$
test_tc_init_nested	0.2026ms	70.8514μs	14.1140 KOps/s	13.2924 KOps/s	$\textbf{\color{#35bf28}+6.18\%}$
test_tc_first_layer_tensor	6.1586μs	0.6796μs	1.4715 MOps/s	1.4010 MOps/s	$\textbf{\color{#35bf28}+5.03\%}$
test_tc_first_layer_nontensor	26.8810μs	2.2428μs	445.8740 KOps/s	442.1256 KOps/s	$\color{#35bf28}+0.85\%$
test_tc_second_layer_tensor	11.4203μs	1.4383μs	695.2615 KOps/s	703.1631 KOps/s	$\color{#d91a1a}-1.12\%$
test_tc_second_layer_nontensor	31.5710μs	2.9949μs	333.8968 KOps/s	330.5956 KOps/s	$\color{#35bf28}+1.00\%$
test_unbind	0.2220s	11.6128ms	86.1119 Ops/s	140.4174 Ops/s	$\textbf{\color{#d91a1a}-38.67\%}$
test_full_like	9.2787ms	9.0471ms	110.5325 Ops/s	110.0388 Ops/s	$\color{#35bf28}+0.45\%$
test_zeros_like	5.4230ms	4.3145ms	231.7750 Ops/s	237.7590 Ops/s	$\color{#d91a1a}-2.52\%$
test_ones_like	4.9830ms	4.2679ms	234.3080 Ops/s	231.3949 Ops/s	$\color{#35bf28}+1.26\%$
test_clone	6.6128ms	6.2884ms	159.0228 Ops/s	110.9145 Ops/s	$\textbf{\color{#35bf28}+43.37\%}$
test_squeeze	92.7520μs	9.5601μs	104.6016 KOps/s	100.1645 KOps/s	$\color{#35bf28}+4.43\%$
test_unsqueeze	0.1232ms	70.5027μs	14.1839 KOps/s	14.0340 KOps/s	$\color{#35bf28}+1.07\%$
test_split	0.2920ms	0.1570ms	6.3705 KOps/s	6.0278 KOps/s	$\textbf{\color{#35bf28}+5.69\%}$
test_permute	0.2487ms	0.1736ms	5.7592 KOps/s	5.4819 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_stack	50.4990ms	50.1713ms	19.9317 Ops/s	19.8444 Ops/s	$\color{#35bf28}+0.44\%$
test_cat	50.4541ms	50.1047ms	19.9582 Ops/s	19.9433 Ops/s	$\color{#35bf28}+0.07\%$