[Feature] Update batch-size

[vmoens]

Stack from ghstack (oldest at bottom):

• [BugFix] Fix compile during _check_keys #1239
• -> [Feature] Update batch-size #1237
• [BugFix] Fix serialization of stacks of Tensorclasses #1236

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}7$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.7820μs	21.1024μs	47.3880 KOps/s	47.9420 KOps/s	$\color{#d91a1a}-1.16\%$
test_plain_set_stack_nested	43.2810μs	20.9606μs	47.7085 KOps/s	47.7913 KOps/s	$\color{#d91a1a}-0.17\%$
test_plain_set_nested_inplace	69.8700μs	22.3650μs	44.7127 KOps/s	44.1670 KOps/s	$\color{#35bf28}+1.24\%$
test_plain_set_stack_nested_inplace	72.2340μs	22.3666μs	44.7095 KOps/s	43.8952 KOps/s	$\color{#35bf28}+1.86\%$
test_items	18.9950μs	4.2012μs	238.0286 KOps/s	239.3096 KOps/s	$\color{#d91a1a}-0.54\%$
test_items_nested	0.6712ms	0.4075ms	2.4541 KOps/s	2.4631 KOps/s	$\color{#d91a1a}-0.37\%$
test_items_nested_locked	0.5519ms	0.4108ms	2.4343 KOps/s	2.4379 KOps/s	$\color{#d91a1a}-0.15\%$
test_items_nested_leaf	0.1446ms	77.3147μs	12.9342 KOps/s	12.7832 KOps/s	$\color{#35bf28}+1.18\%$
test_items_stack_nested	0.6195ms	0.4099ms	2.4398 KOps/s	2.4182 KOps/s	$\color{#35bf28}+0.89\%$
test_items_stack_nested_leaf	0.1516ms	77.8964μs	12.8376 KOps/s	12.8970 KOps/s	$\color{#d91a1a}-0.46\%$
test_items_stack_nested_locked	0.8394ms	0.4055ms	2.4663 KOps/s	2.4404 KOps/s	$\color{#35bf28}+1.06\%$
test_keys	29.1140μs	3.4323μs	291.3457 KOps/s	284.1525 KOps/s	$\color{#35bf28}+2.53\%$
test_keys_nested	0.2736ms	0.1630ms	6.1367 KOps/s	6.0211 KOps/s	$\color{#35bf28}+1.92\%$
test_keys_nested_locked	1.7240ms	0.1687ms	5.9273 KOps/s	5.8130 KOps/s	$\color{#35bf28}+1.97\%$
test_keys_nested_leaf	0.2584ms	0.1422ms	7.0317 KOps/s	6.9378 KOps/s	$\color{#35bf28}+1.35\%$
test_keys_stack_nested	0.3953ms	0.1632ms	6.1260 KOps/s	5.9100 KOps/s	$\color{#35bf28}+3.66\%$
test_keys_stack_nested_leaf	0.2493ms	0.1429ms	6.9971 KOps/s	6.9314 KOps/s	$\color{#35bf28}+0.95\%$
test_keys_stack_nested_locked	0.3398ms	0.1682ms	5.9439 KOps/s	5.7796 KOps/s	$\color{#35bf28}+2.84\%$
test_values	6.1354μs	1.0425μs	959.2782 KOps/s	974.2341 KOps/s	$\color{#d91a1a}-1.54\%$
test_values_nested	0.1230ms	62.2033μs	16.0763 KOps/s	16.0168 KOps/s	$\color{#35bf28}+0.37\%$
test_values_nested_locked	0.1159ms	62.4589μs	16.0105 KOps/s	15.9607 KOps/s	$\color{#35bf28}+0.31\%$
test_values_nested_leaf	0.1548ms	74.3969μs	13.4414 KOps/s	13.9714 KOps/s	$\color{#d91a1a}-3.79\%$
test_values_stack_nested	0.1227ms	61.9749μs	16.1356 KOps/s	15.8839 KOps/s	$\color{#35bf28}+1.58\%$
test_values_stack_nested_leaf	0.1640ms	71.2866μs	14.0279 KOps/s	13.9039 KOps/s	$\color{#35bf28}+0.89\%$
test_values_stack_nested_locked	0.1431ms	63.0913μs	15.8500 KOps/s	15.9068 KOps/s	$\color{#d91a1a}-0.36\%$
test_membership	2.6820μs	0.6875μs	1.4546 MOps/s	1.1431 MOps/s	$\textbf{\color{#35bf28}+27.25\%}$
test_membership_nested	30.1760μs	2.9042μs	344.3339 KOps/s	345.6056 KOps/s	$\color{#d91a1a}-0.37\%$
test_membership_nested_leaf	43.3300μs	2.9331μs	340.9419 KOps/s	344.0575 KOps/s	$\color{#d91a1a}-0.91\%$
test_membership_stacked_nested	26.1790μs	2.9007μs	344.7423 KOps/s	347.4212 KOps/s	$\color{#d91a1a}-0.77\%$
test_membership_stacked_nested_leaf	26.5190μs	2.8828μs	346.8815 KOps/s	346.0047 KOps/s	$\color{#35bf28}+0.25\%$
test_membership_nested_last	25.9380μs	4.2975μs	232.6910 KOps/s	227.8457 KOps/s	$\color{#35bf28}+2.13\%$
test_membership_nested_leaf_last	0.1344ms	4.4066μs	226.9319 KOps/s	228.6534 KOps/s	$\color{#d91a1a}-0.75\%$
test_membership_stacked_nested_last	0.1825ms	4.5022μs	222.1122 KOps/s	229.9165 KOps/s	$\color{#d91a1a}-3.39\%$
test_membership_stacked_nested_leaf_last	44.4820μs	4.3353μs	230.6663 KOps/s	231.6057 KOps/s	$\color{#d91a1a}-0.41\%$
test_nested_getleaf	46.2360μs	10.5248μs	95.0134 KOps/s	95.7564 KOps/s	$\color{#d91a1a}-0.78\%$
test_nested_get	50.3830μs	10.2043μs	97.9982 KOps/s	100.1947 KOps/s	$\color{#d91a1a}-2.19\%$
test_stacked_getleaf	50.3540μs	10.5817μs	94.5032 KOps/s	94.6996 KOps/s	$\color{#d91a1a}-0.21\%$
test_stacked_get	35.7260μs	10.3299μs	96.8066 KOps/s	99.5334 KOps/s	$\color{#d91a1a}-2.74\%$
test_nested_getitemleaf	54.6320μs	11.3064μs	88.4453 KOps/s	89.6599 KOps/s	$\color{#d91a1a}-1.35\%$
test_nested_getitem	0.3387ms	10.7079μs	93.3891 KOps/s	92.5268 KOps/s	$\color{#35bf28}+0.93\%$
test_stacked_getitemleaf	29.7950μs	11.1584μs	89.6188 KOps/s	90.4044 KOps/s	$\color{#d91a1a}-0.87\%$
test_stacked_getitem	55.0220μs	10.7850μs	92.7215 KOps/s	93.6894 KOps/s	$\color{#d91a1a}-1.03\%$
test_lock_nested	0.7488ms	0.4067ms	2.4586 KOps/s	2.4518 KOps/s	$\color{#35bf28}+0.28\%$
test_lock_stack_nested	0.5619ms	0.4211ms	2.3747 KOps/s	2.3504 KOps/s	$\color{#35bf28}+1.03\%$
test_unlock_nested	0.7339ms	0.3325ms	3.0079 KOps/s	2.9764 KOps/s	$\color{#35bf28}+1.06\%$
test_unlock_stack_nested	0.6184ms	0.3393ms	2.9468 KOps/s	2.9292 KOps/s	$\color{#35bf28}+0.60\%$
test_flatten_speed	0.1916ms	99.6250μs	10.0376 KOps/s	9.9905 KOps/s	$\color{#35bf28}+0.47\%$
test_unflatten_speed	1.0653ms	0.5218ms	1.9166 KOps/s	1.9298 KOps/s	$\color{#d91a1a}-0.68\%$
test_common_ops	1.0136ms	0.8113ms	1.2326 KOps/s	1.2670 KOps/s	$\color{#d91a1a}-2.71\%$
test_creation	40.9570μs	2.5014μs	399.7742 KOps/s	407.4700 KOps/s	$\color{#d91a1a}-1.89\%$
test_creation_empty	36.7490μs	12.3906μs	80.7063 KOps/s	81.7519 KOps/s	$\color{#d91a1a}-1.28\%$
test_creation_nested_1	66.6150μs	15.1277μs	66.1037 KOps/s	65.7514 KOps/s	$\color{#35bf28}+0.54\%$
test_creation_nested_2	55.1230μs	19.5597μs	51.1255 KOps/s	51.0131 KOps/s	$\color{#35bf28}+0.22\%$
test_clone	66.6240μs	13.7512μs	72.7209 KOps/s	76.8391 KOps/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_getitem[int]	0.7811ms	12.7013μs	78.7320 KOps/s	79.9705 KOps/s	$\color{#d91a1a}-1.55\%$
test_getitem[slice_int]	0.1274ms	24.5975μs	40.6545 KOps/s	41.4608 KOps/s	$\color{#d91a1a}-1.94\%$
test_getitem[range]	0.1652ms	51.5781μs	19.3881 KOps/s	19.6270 KOps/s	$\color{#d91a1a}-1.22\%$
test_getitem[tuple]	0.1227ms	20.4825μs	48.8223 KOps/s	50.3137 KOps/s	$\color{#d91a1a}-2.96\%$
test_getitem[list]	0.1531ms	45.9530μs	21.7614 KOps/s	21.9084 KOps/s	$\color{#d91a1a}-0.67\%$
test_setitem_dim[int]	44.9030μs	25.1302μs	39.7928 KOps/s	38.1304 KOps/s	$\color{#35bf28}+4.36\%$
test_setitem_dim[slice_int]	94.3460μs	51.7275μs	19.3321 KOps/s	19.1565 KOps/s	$\color{#35bf28}+0.92\%$
test_setitem_dim[range]	0.1303ms	77.7197μs	12.8667 KOps/s	12.7761 KOps/s	$\color{#35bf28}+0.71\%$
test_setitem_dim[tuple]	72.2750μs	41.0122μs	24.3830 KOps/s	23.8304 KOps/s	$\color{#35bf28}+2.32\%$
test_setitem	86.2700μs	21.5248μs	46.4581 KOps/s	48.2153 KOps/s	$\color{#d91a1a}-3.64\%$
test_set	83.5160μs	21.1119μs	47.3667 KOps/s	49.3455 KOps/s	$\color{#d91a1a}-4.01\%$
test_set_shared	4.1846ms	0.1820ms	5.4954 KOps/s	5.4243 KOps/s	$\color{#35bf28}+1.31\%$
test_update	0.1192ms	26.6314μs	37.5497 KOps/s	42.6315 KOps/s	$\textbf{\color{#d91a1a}-11.92\%}$
test_update_nested	0.1198ms	42.4217μs	23.5728 KOps/s	29.5986 KOps/s	$\textbf{\color{#d91a1a}-20.36\%}$
test_update__nested	0.4427ms	34.1743μs	29.2618 KOps/s	29.8542 KOps/s	$\color{#d91a1a}-1.98\%$
test_set_nested	80.8000μs	23.1744μs	43.1511 KOps/s	45.1416 KOps/s	$\color{#d91a1a}-4.41\%$
test_set_nested_new	84.2860μs	27.1999μs	36.7648 KOps/s	37.0853 KOps/s	$\color{#d91a1a}-0.86\%$
test_select	97.6010μs	44.2706μs	22.5884 KOps/s	22.9895 KOps/s	$\color{#d91a1a}-1.74\%$
test_select_nested	0.1231ms	63.0972μs	15.8486 KOps/s	15.8852 KOps/s	$\color{#d91a1a}-0.23\%$
test_exclude_nested	0.1483ms	80.5668μs	12.4121 KOps/s	12.3482 KOps/s	$\color{#35bf28}+0.52\%$
test_empty[True]	0.5177ms	0.4065ms	2.4602 KOps/s	2.4321 KOps/s	$\color{#35bf28}+1.16\%$
test_empty[False]	10.6873μs	1.3743μs	727.6631 KOps/s	721.7556 KOps/s	$\color{#35bf28}+0.82\%$
test_unbind_speed	0.6401ms	0.2663ms	3.7546 KOps/s	3.7141 KOps/s	$\color{#35bf28}+1.09\%$
test_unbind_speed_stack0	0.4145ms	0.2652ms	3.7700 KOps/s	3.7193 KOps/s	$\color{#35bf28}+1.36\%$
test_unbind_speed_stack1	0.1098s	0.7391ms	1.3530 KOps/s	1.2113 KOps/s	$\textbf{\color{#35bf28}+11.70\%}$
test_split	0.1022s	1.7610ms	567.8632 Ops/s	573.4723 Ops/s	$\color{#d91a1a}-0.98\%$
test_chunk	97.8720ms	1.7370ms	575.6976 Ops/s	630.4284 Ops/s	$\textbf{\color{#d91a1a}-8.68\%}$
test_consolidate_njt[False-None]	11.5692ms	8.2663ms	120.9735 Ops/s	105.6233 Ops/s	$\textbf{\color{#35bf28}+14.53\%}$
test_creation[device0]	0.2161ms	91.4440μs	10.9357 KOps/s	10.8518 KOps/s	$\color{#35bf28}+0.77\%$
test_creation_from_tensor	4.1434ms	96.1458μs	10.4009 KOps/s	10.3885 KOps/s	$\color{#35bf28}+0.12\%$
test_add_one[memmap_tensor0]	78.8670μs	5.1592μs	193.8274 KOps/s	207.9121 KOps/s	$\textbf{\color{#d91a1a}-6.77\%}$
test_contiguous[memmap_tensor0]	24.8360μs	0.5044μs	1.9826 MOps/s	1.9485 MOps/s	$\color{#35bf28}+1.75\%$
test_stack[memmap_tensor0]	29.0640μs	3.4435μs	290.4032 KOps/s	301.9203 KOps/s	$\color{#d91a1a}-3.81\%$
test_memmaptd_index	1.1775ms	0.2290ms	4.3677 KOps/s	4.2535 KOps/s	$\color{#35bf28}+2.68\%$
test_memmaptd_index_astensor	0.6761ms	0.3149ms	3.1757 KOps/s	3.1276 KOps/s	$\color{#35bf28}+1.54\%$
test_memmaptd_index_op	0.8074ms	0.5905ms	1.6935 KOps/s	1.6712 KOps/s	$\color{#35bf28}+1.34\%$
test_serialize_model	0.2248s	0.1309s	7.6385 Ops/s	8.6029 Ops/s	$\textbf{\color{#d91a1a}-11.21\%}$
test_serialize_model_pickle	0.4592s	0.3947s	2.5335 Ops/s	2.4571 Ops/s	$\color{#35bf28}+3.11\%$
test_serialize_weights	0.1273s	0.1182s	8.4602 Ops/s	8.8244 Ops/s	$\color{#d91a1a}-4.13\%$
test_serialize_weights_returnearly	0.1708s	0.1582s	6.3201 Ops/s	5.8005 Ops/s	$\textbf{\color{#35bf28}+8.96\%}$
test_serialize_weights_pickle	0.5983s	0.4303s	2.3239 Ops/s	2.4799 Ops/s	$\textbf{\color{#d91a1a}-6.29\%}$
test_serialize_weights_filesystem	0.2413s	0.1549s	6.4543 Ops/s	7.0182 Ops/s	$\textbf{\color{#d91a1a}-8.03\%}$
test_serialize_model_filesystem	0.1537s	0.1434s	6.9749 Ops/s	6.5852 Ops/s	$\textbf{\color{#35bf28}+5.92\%}$
test_reshape_pytree	74.4990μs	25.9619μs	38.5179 KOps/s	38.0025 KOps/s	$\color{#35bf28}+1.36\%$
test_reshape_td	76.4720μs	32.1307μs	31.1229 KOps/s	30.8585 KOps/s	$\color{#35bf28}+0.86\%$
test_view_pytree	73.1060μs	25.8564μs	38.6751 KOps/s	38.3810 KOps/s	$\color{#35bf28}+0.77\%$
test_view_td	0.1104ms	39.1544μs	25.5399 KOps/s	24.8938 KOps/s	$\color{#35bf28}+2.60\%$
test_unbind_pytree	78.0850μs	29.2066μs	34.2389 KOps/s	33.8340 KOps/s	$\color{#35bf28}+1.20\%$
test_unbind_td	0.3124ms	40.1071μs	24.9332 KOps/s	25.3711 KOps/s	$\color{#d91a1a}-1.73\%$
test_split_pytree	67.8070μs	28.5693μs	35.0026 KOps/s	34.5958 KOps/s	$\color{#35bf28}+1.18\%$
test_split_td	0.4875ms	45.3923μs	22.0302 KOps/s	22.3296 KOps/s	$\color{#d91a1a}-1.34\%$
test_add_pytree	91.4510μs	36.1981μs	27.6258 KOps/s	28.4486 KOps/s	$\color{#d91a1a}-2.89\%$
test_add_td	0.1328ms	55.7950μs	17.9228 KOps/s	17.6344 KOps/s	$\color{#35bf28}+1.64\%$
test_compile_add_one_nested[tensordict-compile]	0.1446ms	68.7787μs	14.5394 KOps/s	14.9655 KOps/s	$\color{#d91a1a}-2.85\%$
test_compile_add_one_nested[tensordict-eager]	0.3470ms	0.1710ms	5.8489 KOps/s	5.7159 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_add_one_nested[pytree-compile]	0.2715ms	46.7841μs	21.3748 KOps/s	22.1832 KOps/s	$\color{#d91a1a}-3.64\%$
test_compile_add_one_nested[pytree-eager]	0.2197ms	0.1194ms	8.3778 KOps/s	8.5217 KOps/s	$\color{#d91a1a}-1.69\%$
test_compile_copy_nested[tensordict-compile]	88.2240μs	28.7654μs	34.7640 KOps/s	35.8270 KOps/s	$\color{#d91a1a}-2.97\%$
test_compile_copy_nested[tensordict-eager]	0.1414ms	58.8423μs	16.9946 KOps/s	17.1753 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_copy_nested[pytree-compile]	0.1722ms	78.6327μs	12.7174 KOps/s	12.4111 KOps/s	$\color{#35bf28}+2.47\%$
test_compile_copy_nested[pytree-eager]	0.1369ms	66.1390μs	15.1197 KOps/s	14.7569 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_add_one_flat[tensordict-compile]	0.2294ms	0.1093ms	9.1502 KOps/s	9.4173 KOps/s	$\color{#d91a1a}-2.84\%$
test_compile_add_one_flat[tensordict-eager]	0.4204ms	0.2162ms	4.6254 KOps/s	4.6533 KOps/s	$\color{#d91a1a}-0.60\%$
test_compile_add_one_flat[tensorclass-compile]	0.3343ms	47.8475μs	20.8997 KOps/s	21.7682 KOps/s	$\color{#d91a1a}-3.99\%$
test_compile_add_one_flat[tensorclass-eager]	0.1662ms	66.8820μs	14.9517 KOps/s	14.9889 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_one_flat[pytree-compile]	0.1857ms	0.1028ms	9.7294 KOps/s	10.0633 KOps/s	$\color{#d91a1a}-3.32\%$
test_compile_add_one_flat[pytree-eager]	0.3912ms	0.2068ms	4.8346 KOps/s	5.0068 KOps/s	$\color{#d91a1a}-3.44\%$
test_compile_add_self_flat[tensordict-eager]	0.4257ms	0.2373ms	4.2143 KOps/s	4.3302 KOps/s	$\color{#d91a1a}-2.68\%$
test_compile_add_self_flat[tensordict-compile]	0.2252ms	0.1082ms	9.2448 KOps/s	9.1936 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_add_self_flat[tensorclass-eager]	0.1459ms	62.3616μs	16.0355 KOps/s	15.9978 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_add_self_flat[tensorclass-compile]	0.1011ms	48.4373μs	20.6452 KOps/s	20.4983 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_add_self_flat[pytree-eager]	0.2891ms	0.1604ms	6.2351 KOps/s	6.3918 KOps/s	$\color{#d91a1a}-2.45\%$
test_compile_add_self_flat[pytree-compile]	0.1959ms	0.1026ms	9.7435 KOps/s	9.9457 KOps/s	$\color{#d91a1a}-2.03\%$
test_compile_copy_flat[tensordict-compile]	55.1030μs	22.3457μs	44.7513 KOps/s	46.5112 KOps/s	$\color{#d91a1a}-3.78\%$
test_compile_copy_flat[tensordict-eager]	0.1404ms	66.5311μs	15.0306 KOps/s	15.0057 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_copy_flat[pytree-compile]	0.1878ms	82.9481μs	12.0557 KOps/s	12.2811 KOps/s	$\color{#d91a1a}-1.84\%$
test_compile_copy_flat[pytree-eager]	0.1393ms	65.9737μs	15.1576 KOps/s	14.6207 KOps/s	$\color{#35bf28}+3.67\%$
test_compile_assign_and_add[tensordict-compile]	0.3187ms	0.2182ms	4.5830 KOps/s	4.6459 KOps/s	$\color{#d91a1a}-1.35\%$
test_compile_assign_and_add[tensordict-eager]	1.6873ms	1.3774ms	726.0146 Ops/s	728.4557 Ops/s	$\color{#d91a1a}-0.34\%$
test_compile_assign_and_add[pytree-compile]	0.3320ms	0.2142ms	4.6696 KOps/s	4.7609 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_assign_and_add[pytree-eager]	1.3211ms	0.8366ms	1.1953 KOps/s	1.2317 KOps/s	$\color{#d91a1a}-2.95\%$
test_compile_assign_and_add_stack[compile]	0.8714ms	0.4690ms	2.1324 KOps/s	2.1796 KOps/s	$\color{#d91a1a}-2.17\%$
test_compile_assign_and_add_stack[eager]	4.0772ms	2.7074ms	369.3591 Ops/s	359.5748 Ops/s	$\color{#35bf28}+2.72\%$
test_compile_indexing[tensor-tensordict-compile]	0.1087ms	38.6174μs	25.8951 KOps/s	25.9378 KOps/s	$\color{#d91a1a}-0.16\%$
test_compile_indexing[tensor-tensordict-eager]	0.5530ms	34.4664μs	29.0138 KOps/s	31.0271 KOps/s	$\textbf{\color{#d91a1a}-6.49\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.1392ms	31.7279μs	31.5180 KOps/s	32.9233 KOps/s	$\color{#d91a1a}-4.27\%$
test_compile_indexing[tensor-tensorclass-eager]	70.7420μs	23.2479μs	43.0146 KOps/s	43.4463 KOps/s	$\color{#d91a1a}-0.99\%$
test_compile_indexing[tensor-pytree-compile]	0.1043ms	31.9132μs	31.3350 KOps/s	32.0554 KOps/s	$\color{#d91a1a}-2.25\%$
test_compile_indexing[tensor-pytree-eager]	0.3292ms	23.3388μs	42.8472 KOps/s	44.2338 KOps/s	$\color{#d91a1a}-3.13\%$
test_compile_indexing[slice-tensordict-compile]	99.3240μs	52.7821μs	18.9458 KOps/s	18.9132 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_indexing[slice-tensordict-eager]	0.3862ms	20.0291μs	49.9274 KOps/s	50.3249 KOps/s	$\color{#d91a1a}-0.79\%$
test_compile_indexing[slice-tensorclass-compile]	98.6830μs	45.4663μs	21.9943 KOps/s	22.0454 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_indexing[slice-tensorclass-eager]	90.6590μs	18.4634μs	54.1613 KOps/s	54.2161 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_indexing[slice-pytree-compile]	93.4740μs	46.2581μs	21.6178 KOps/s	21.5428 KOps/s	$\color{#35bf28}+0.35\%$
test_compile_indexing[slice-pytree-eager]	60.6320μs	18.1772μs	55.0138 KOps/s	53.8413 KOps/s	$\color{#35bf28}+2.18\%$
test_compile_indexing[int-tensordict-compile]	0.1055ms	54.9532μs	18.1973 KOps/s	18.4699 KOps/s	$\color{#d91a1a}-1.48\%$
test_compile_indexing[int-tensordict-eager]	1.1555ms	19.8432μs	50.3952 KOps/s	51.4346 KOps/s	$\color{#d91a1a}-2.02\%$
test_compile_indexing[int-tensorclass-compile]	0.1137ms	46.8930μs	21.3252 KOps/s	21.6231 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_indexing[int-tensorclass-eager]	51.0950μs	18.3217μs	54.5799 KOps/s	54.2102 KOps/s	$\color{#35bf28}+0.68\%$
test_compile_indexing[int-pytree-compile]	0.1159ms	46.0445μs	21.7181 KOps/s	21.5868 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_indexing[int-pytree-eager]	57.5270μs	18.4521μs	54.1944 KOps/s	54.5200 KOps/s	$\color{#d91a1a}-0.60\%$
test_mod_add[eager]	84.7770μs	36.6059μs	27.3180 KOps/s	27.6588 KOps/s	$\color{#d91a1a}-1.23\%$
test_mod_add[compile]	0.1190ms	64.2407μs	15.5665 KOps/s	15.1748 KOps/s	$\color{#35bf28}+2.58\%$
test_mod_add[compile-overhead]	0.1162ms	62.8950μs	15.8995 KOps/s	15.1266 KOps/s	$\textbf{\color{#35bf28}+5.11\%}$
test_mod_wrap[eager]	0.4528ms	0.2265ms	4.4159 KOps/s	4.5585 KOps/s	$\color{#d91a1a}-3.13\%$
test_mod_wrap[compile]	1.6325ms	0.2265ms	4.4145 KOps/s	4.4063 KOps/s	$\color{#35bf28}+0.19\%$
test_mod_wrap[compile-overhead]	0.3426ms	0.2236ms	4.4732 KOps/s	4.4887 KOps/s	$\color{#d91a1a}-0.35\%$
test_mod_wrap_and_backward[eager]	15.6512ms	12.6512ms	79.0437 Ops/s	77.0078 Ops/s	$\color{#35bf28}+2.64\%$
test_mod_wrap_and_backward[compile]	13.6367ms	11.2872ms	88.5957 Ops/s	87.4800 Ops/s	$\color{#35bf28}+1.28\%$
test_mod_wrap_and_backward[compile-overhead]	13.3952ms	11.1602ms	89.6044 Ops/s	89.1992 Ops/s	$\color{#35bf28}+0.45\%$
test_seq_add[eager]	0.2115ms	0.1197ms	8.3516 KOps/s	8.2842 KOps/s	$\color{#35bf28}+0.81\%$
test_seq_add[compile]	0.1568ms	75.9560μs	13.1655 KOps/s	13.1172 KOps/s	$\color{#35bf28}+0.37\%$
test_seq_add[compile-overhead]	0.1401ms	74.0360μs	13.5069 KOps/s	13.4115 KOps/s	$\color{#35bf28}+0.71\%$
test_seq_wrap[eager]	0.6542ms	0.4504ms	2.2201 KOps/s	2.2194 KOps/s	$\color{#35bf28}+0.03\%$
test_seq_wrap[compile]	0.3543ms	0.2413ms	4.1443 KOps/s	4.1693 KOps/s	$\color{#d91a1a}-0.60\%$
test_seq_wrap[compile-overhead]	0.4408ms	0.2382ms	4.1983 KOps/s	4.1917 KOps/s	$\color{#35bf28}+0.16\%$
test_func_call_runtime[False-eager]	0.6751ms	0.5468ms	1.8289 KOps/s	1.9087 KOps/s	$\color{#d91a1a}-4.18\%$
test_func_call_runtime[False-compile]	0.7304ms	0.4429ms	2.2578 KOps/s	2.2841 KOps/s	$\color{#d91a1a}-1.15\%$
test_func_call_runtime[False-compile-overhead]	0.5625ms	0.4415ms	2.2648 KOps/s	2.2840 KOps/s	$\color{#d91a1a}-0.84\%$
test_func_call_runtime[True-eager]	0.9485ms	0.7588ms	1.3179 KOps/s	1.3655 KOps/s	$\color{#d91a1a}-3.49\%$
test_func_call_runtime[True-compile]	0.5730ms	0.4655ms	2.1484 KOps/s	2.1784 KOps/s	$\color{#d91a1a}-1.38\%$
test_func_call_runtime[True-compile-overhead]	0.8364ms	0.4697ms	2.1291 KOps/s	2.1780 KOps/s	$\color{#d91a1a}-2.25\%$
test_func_call_cm_runtime[False-eager]	0.9251ms	0.5386ms	1.8566 KOps/s	1.8877 KOps/s	$\color{#d91a1a}-1.65\%$
test_func_call_cm_runtime[False-compile]	0.5464ms	0.4442ms	2.2512 KOps/s	2.2684 KOps/s	$\color{#d91a1a}-0.76\%$
test_func_call_cm_runtime[False-compile-overhead]	0.9295ms	0.4438ms	2.2531 KOps/s	2.2873 KOps/s	$\color{#d91a1a}-1.50\%$
test_func_call_cm_runtime[True-eager]	2.0291ms	0.9052ms	1.1048 KOps/s	1.1424 KOps/s	$\color{#d91a1a}-3.30\%$
test_func_call_cm_runtime[True-compile]	1.2075ms	0.8109ms	1.2332 KOps/s	1.2624 KOps/s	$\color{#d91a1a}-2.31\%$
test_func_call_cm_runtime[True-compile-overhead]	1.1359ms	0.8110ms	1.2331 KOps/s	1.2571 KOps/s	$\color{#d91a1a}-1.91\%$
test_vmap_func_call_cm_runtime[eager]	2.4980ms	1.9365ms	516.3995 Ops/s	521.5083 Ops/s	$\color{#d91a1a}-0.98\%$
test_vmap_func_call_cm_runtime[compile]	0.6639ms	0.5416ms	1.8463 KOps/s	1.8740 KOps/s	$\color{#d91a1a}-1.48\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9595ms	0.5396ms	1.8531 KOps/s	1.8736 KOps/s	$\color{#d91a1a}-1.09\%$
test_distributed	0.2927ms	0.1232ms	8.1139 KOps/s	7.8184 KOps/s	$\color{#35bf28}+3.78\%$
test_tdmodule	85.1680μs	28.4335μs	35.1698 KOps/s	35.0720 KOps/s	$\color{#35bf28}+0.28\%$
test_tdmodule_dispatch	70.0800μs	49.6598μs	20.1370 KOps/s	19.4398 KOps/s	$\color{#35bf28}+3.59\%$
test_tdseq	48.6410μs	29.4879μs	33.9122 KOps/s	33.3305 KOps/s	$\color{#35bf28}+1.75\%$
test_tdseq_dispatch	0.1066ms	56.3472μs	17.7471 KOps/s	17.7019 KOps/s	$\color{#35bf28}+0.26\%$
test_instantiation_functorch	1.7848ms	1.5173ms	659.0657 Ops/s	653.2233 Ops/s	$\color{#35bf28}+0.89\%$
test_exec_functorch	0.4152ms	0.1804ms	5.5422 KOps/s	5.5548 KOps/s	$\color{#d91a1a}-0.23\%$
test_exec_functional_call	0.3183ms	0.1705ms	5.8663 KOps/s	5.8720 KOps/s	$\color{#d91a1a}-0.10\%$
test_exec_td_decorator	0.4792ms	0.2320ms	4.3109 KOps/s	4.3650 KOps/s	$\color{#d91a1a}-1.24\%$
test_vmap_mlp_speed_decorator[True-True]	1.0249ms	0.6833ms	1.4634 KOps/s	1.5074 KOps/s	$\color{#d91a1a}-2.92\%$
test_vmap_mlp_speed_decorator[True-False]	1.1199ms	0.6704ms	1.4915 KOps/s	1.5061 KOps/s	$\color{#d91a1a}-0.97\%$
test_vmap_mlp_speed_decorator[False-True]	0.8593ms	0.5431ms	1.8414 KOps/s	1.8698 KOps/s	$\color{#d91a1a}-1.52\%$
test_vmap_mlp_speed_decorator[False-False]	0.6841ms	0.5440ms	1.8382 KOps/s	1.8657 KOps/s	$\color{#d91a1a}-1.48\%$
test_to_module_speed[True]	2.1777ms	1.3357ms	748.6658 Ops/s	745.9697 Ops/s	$\color{#35bf28}+0.36\%$
test_to_module_speed[False]	1.8069ms	1.2983ms	770.2349 Ops/s	770.1279 Ops/s	$\color{#35bf28}+0.01\%$
test_tc_init	79.4180μs	46.4744μs	21.5172 KOps/s	21.5043 KOps/s	$\color{#35bf28}+0.06\%$
test_tc_init_nested	0.2094ms	93.6418μs	10.6790 KOps/s	10.8170 KOps/s	$\color{#d91a1a}-1.28\%$
test_tc_first_layer_tensor	13.8260μs	1.5405μs	649.1237 KOps/s	632.9214 KOps/s	$\color{#35bf28}+2.56\%$
test_tc_first_layer_nontensor	45.8150μs	4.7423μs	210.8672 KOps/s	209.3304 KOps/s	$\color{#35bf28}+0.73\%$
test_tc_second_layer_tensor	38.0500μs	2.8730μs	348.0667 KOps/s	346.3609 KOps/s	$\color{#35bf28}+0.49\%$
test_tc_second_layer_nontensor	0.1256ms	6.0567μs	165.1075 KOps/s	162.7102 KOps/s	$\color{#35bf28}+1.47\%$
test_unbind	0.2573s	14.0226ms	71.3137 Ops/s	70.6906 Ops/s	$\color{#35bf28}+0.88\%$
test_full_like	9.8143ms	7.9821ms	125.2809 Ops/s	130.6380 Ops/s	$\color{#d91a1a}-4.10\%$
test_zeros_like	11.5573ms	3.0500ms	327.8708 Ops/s	355.1023 Ops/s	$\textbf{\color{#d91a1a}-7.67\%}$
test_ones_like	4.7205ms	3.3597ms	297.6457 Ops/s	316.3742 Ops/s	$\textbf{\color{#d91a1a}-5.92\%}$
test_clone	6.8610ms	5.1970ms	192.4205 Ops/s	136.8937 Ops/s	$\textbf{\color{#35bf28}+40.56\%}$
test_squeeze	62.5770μs	12.5020μs	79.9872 KOps/s	79.8474 KOps/s	$\color{#35bf28}+0.18\%$
test_unsqueeze	0.3066ms	92.3953μs	10.8231 KOps/s	10.4063 KOps/s	$\color{#35bf28}+4.01\%$
test_split	0.3605ms	0.1949ms	5.1297 KOps/s	5.0326 KOps/s	$\color{#35bf28}+1.93\%$
test_permute	0.4727ms	0.2057ms	4.8604 KOps/s	4.8004 KOps/s	$\color{#35bf28}+1.25\%$
test_stack	31.3461ms	25.3638ms	39.4262 Ops/s	39.2075 Ops/s	$\color{#35bf28}+0.56\%$
test_cat	30.7351ms	25.1481ms	39.7645 Ops/s	38.7323 Ops/s	$\color{#35bf28}+2.66\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}18$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.2500μs	12.4023μs	80.6302 KOps/s	78.3379 KOps/s	$\color{#35bf28}+2.93\%$
test_plain_set_stack_nested	55.9100μs	12.5742μs	79.5277 KOps/s	76.9144 KOps/s	$\color{#35bf28}+3.40\%$
test_plain_set_nested_inplace	39.4410μs	13.5418μs	73.8454 KOps/s	71.5265 KOps/s	$\color{#35bf28}+3.24\%$
test_plain_set_stack_nested_inplace	90.9220μs	13.4016μs	74.6181 KOps/s	71.9709 KOps/s	$\color{#35bf28}+3.68\%$
test_items	23.8300μs	2.8889μs	346.1490 KOps/s	349.8119 KOps/s	$\color{#d91a1a}-1.05\%$
test_items_nested	0.4235ms	0.3640ms	2.7471 KOps/s	2.7531 KOps/s	$\color{#d91a1a}-0.22\%$
test_items_nested_locked	0.4199ms	0.3639ms	2.7478 KOps/s	2.7184 KOps/s	$\color{#35bf28}+1.08\%$
test_items_nested_leaf	91.3420μs	60.7460μs	16.4620 KOps/s	16.6554 KOps/s	$\color{#d91a1a}-1.16\%$
test_items_stack_nested	0.4116ms	0.3657ms	2.7347 KOps/s	2.7498 KOps/s	$\color{#d91a1a}-0.55\%$
test_items_stack_nested_leaf	84.9310μs	60.2190μs	16.6061 KOps/s	16.6181 KOps/s	$\color{#d91a1a}-0.07\%$
test_items_stack_nested_locked	0.4418ms	0.3649ms	2.7404 KOps/s	2.7290 KOps/s	$\color{#35bf28}+0.41\%$
test_keys	26.4000μs	3.4448μs	290.2888 KOps/s	292.9828 KOps/s	$\color{#d91a1a}-0.92\%$
test_keys_nested	0.1231ms	88.2558μs	11.3307 KOps/s	11.4207 KOps/s	$\color{#d91a1a}-0.79\%$
test_keys_nested_locked	0.7358ms	93.8198μs	10.6587 KOps/s	10.8188 KOps/s	$\color{#d91a1a}-1.48\%$
test_keys_nested_leaf	96.7810μs	79.8286μs	12.5268 KOps/s	12.7161 KOps/s	$\color{#d91a1a}-1.49\%$
test_keys_stack_nested	0.1204ms	88.8353μs	11.2568 KOps/s	11.5023 KOps/s	$\color{#d91a1a}-2.13\%$
test_keys_stack_nested_leaf	0.1132ms	79.4414μs	12.5879 KOps/s	12.7056 KOps/s	$\color{#d91a1a}-0.93\%$
test_keys_stack_nested_locked	0.1421ms	93.3923μs	10.7075 KOps/s	10.7602 KOps/s	$\color{#d91a1a}-0.49\%$
test_values	5.3018μs	0.8510μs	1.1751 MOps/s	1.1778 MOps/s	$\color{#d91a1a}-0.23\%$
test_values_nested	66.0810μs	37.1940μs	26.8861 KOps/s	27.0521 KOps/s	$\color{#d91a1a}-0.61\%$
test_values_nested_locked	85.8510μs	39.1211μs	25.5616 KOps/s	25.7913 KOps/s	$\color{#d91a1a}-0.89\%$
test_values_nested_leaf	0.1357ms	42.0859μs	23.7609 KOps/s	23.7861 KOps/s	$\color{#d91a1a}-0.11\%$
test_values_stack_nested	65.3410μs	37.3113μs	26.8016 KOps/s	26.9897 KOps/s	$\color{#d91a1a}-0.70\%$
test_values_stack_nested_leaf	73.0410μs	42.1483μs	23.7257 KOps/s	23.6335 KOps/s	$\color{#35bf28}+0.39\%$
test_values_stack_nested_locked	68.8910μs	39.2510μs	25.4770 KOps/s	25.6439 KOps/s	$\color{#d91a1a}-0.65\%$
test_membership	9.1361μs	0.4984μs	2.0063 MOps/s	1.9965 MOps/s	$\color{#35bf28}+0.49\%$
test_membership_nested	31.7910μs	2.1086μs	474.2570 KOps/s	487.4818 KOps/s	$\color{#d91a1a}-2.71\%$
test_membership_nested_leaf	18.4105μs	2.0332μs	491.8270 KOps/s	501.7132 KOps/s	$\color{#d91a1a}-1.97\%$
test_membership_stacked_nested	38.6100μs	2.0935μs	477.6626 KOps/s	475.3054 KOps/s	$\color{#35bf28}+0.50\%$
test_membership_stacked_nested_leaf	32.7100μs	2.1018μs	475.7861 KOps/s	487.9396 KOps/s	$\color{#d91a1a}-2.49\%$
test_membership_nested_last	34.4500μs	3.1471μs	317.7557 KOps/s	322.0314 KOps/s	$\color{#d91a1a}-1.33\%$
test_membership_nested_leaf_last	34.3400μs	3.1245μs	320.0540 KOps/s	327.6313 KOps/s	$\color{#d91a1a}-2.31\%$
test_membership_stacked_nested_last	41.9200μs	3.0888μs	323.7521 KOps/s	328.0513 KOps/s	$\color{#d91a1a}-1.31\%$
test_membership_stacked_nested_leaf_last	44.9610μs	3.0775μs	324.9374 KOps/s	327.7344 KOps/s	$\color{#d91a1a}-0.85\%$
test_nested_getleaf	45.3010μs	6.2185μs	160.8107 KOps/s	159.0348 KOps/s	$\color{#35bf28}+1.12\%$
test_nested_get	48.8610μs	5.8441μs	171.1113 KOps/s	167.3499 KOps/s	$\color{#35bf28}+2.25\%$
test_stacked_getleaf	38.1810μs	6.1248μs	163.2717 KOps/s	162.2130 KOps/s	$\color{#35bf28}+0.65\%$
test_stacked_get	31.3100μs	5.7519μs	173.8551 KOps/s	172.0435 KOps/s	$\color{#35bf28}+1.05\%$
test_nested_getitemleaf	34.6500μs	6.4314μs	155.4862 KOps/s	156.3746 KOps/s	$\color{#d91a1a}-0.57\%$
test_nested_getitem	29.9210μs	6.1716μs	162.0337 KOps/s	164.0984 KOps/s	$\color{#d91a1a}-1.26\%$
test_stacked_getitemleaf	39.7200μs	6.4120μs	155.9573 KOps/s	158.2337 KOps/s	$\color{#d91a1a}-1.44\%$
test_stacked_getitem	27.7300μs	6.0238μs	166.0069 KOps/s	168.6239 KOps/s	$\color{#d91a1a}-1.55\%$
test_lock_nested	0.4050ms	0.3381ms	2.9575 KOps/s	2.8525 KOps/s	$\color{#35bf28}+3.68\%$
test_lock_stack_nested	0.4774ms	0.3488ms	2.8669 KOps/s	2.8387 KOps/s	$\color{#35bf28}+0.99\%$
test_unlock_nested	0.3519ms	0.2882ms	3.4700 KOps/s	3.5032 KOps/s	$\color{#d91a1a}-0.95\%$
test_unlock_stack_nested	0.3613ms	0.2885ms	3.4662 KOps/s	3.4810 KOps/s	$\color{#d91a1a}-0.43\%$
test_flatten_speed	0.1099ms	76.7665μs	13.0265 KOps/s	12.8635 KOps/s	$\color{#35bf28}+1.27\%$
test_unflatten_speed	0.3648ms	0.3228ms	3.0983 KOps/s	3.1292 KOps/s	$\color{#d91a1a}-0.99\%$
test_common_ops	0.8801ms	0.6644ms	1.5050 KOps/s	1.5392 KOps/s	$\color{#d91a1a}-2.22\%$
test_creation	72.9510μs	1.7399μs	574.7474 KOps/s	576.4006 KOps/s	$\color{#d91a1a}-0.29\%$
test_creation_empty	51.3110μs	8.3363μs	119.9568 KOps/s	106.5890 KOps/s	$\textbf{\color{#35bf28}+12.54\%}$
test_creation_nested_1	0.1289ms	10.1274μs	98.7417 KOps/s	90.2036 KOps/s	$\textbf{\color{#35bf28}+9.47\%}$
test_creation_nested_2	48.8810μs	12.6715μs	78.9170 KOps/s	71.6079 KOps/s	$\textbf{\color{#35bf28}+10.21\%}$
test_clone	62.2510μs	11.1298μs	89.8490 KOps/s	89.5468 KOps/s	$\color{#35bf28}+0.34\%$
test_getitem[int]	1.2670ms	10.8578μs	92.0993 KOps/s	91.2568 KOps/s	$\color{#35bf28}+0.92\%$
test_getitem[slice_int]	0.1145ms	21.0541μs	47.4967 KOps/s	46.3641 KOps/s	$\color{#35bf28}+2.44\%$
test_getitem[range]	0.1335ms	40.5742μs	24.6462 KOps/s	24.4327 KOps/s	$\color{#35bf28}+0.87\%$
test_getitem[tuple]	0.1170ms	19.6629μs	50.8572 KOps/s	53.7756 KOps/s	$\textbf{\color{#d91a1a}-5.43\%}$
test_getitem[list]	0.1239ms	34.6278μs	28.8785 KOps/s	28.6402 KOps/s	$\color{#35bf28}+0.83\%$
test_setitem_dim[int]	51.9500μs	21.2131μs	47.1407 KOps/s	49.7673 KOps/s	$\textbf{\color{#d91a1a}-5.28\%}$
test_setitem_dim[slice_int]	85.2310μs	42.6234μs	23.4613 KOps/s	25.4875 KOps/s	$\textbf{\color{#d91a1a}-7.95\%}$
test_setitem_dim[range]	0.1786ms	58.5054μs	17.0924 KOps/s	17.8966 KOps/s	$\color{#d91a1a}-4.49\%$
test_setitem_dim[tuple]	61.8510μs	34.4499μs	29.0276 KOps/s	28.2211 KOps/s	$\color{#35bf28}+2.86\%$
test_setitem	0.2022ms	17.1486μs	58.3137 KOps/s	60.7098 KOps/s	$\color{#d91a1a}-3.95\%$
test_set	91.8110μs	15.3611μs	65.0997 KOps/s	62.8221 KOps/s	$\color{#35bf28}+3.63\%$
test_set_shared	0.5111ms	0.1597ms	6.2613 KOps/s	6.0965 KOps/s	$\color{#35bf28}+2.70\%$
test_update	0.4248ms	20.6213μs	48.4935 KOps/s	50.8150 KOps/s	$\color{#d91a1a}-4.57\%$
test_update_nested	77.3910μs	28.7442μs	34.7896 KOps/s	38.9185 KOps/s	$\textbf{\color{#d91a1a}-10.61\%}$
test_update__nested	0.4961ms	26.8828μs	37.1985 KOps/s	38.2232 KOps/s	$\color{#d91a1a}-2.68\%$
test_set_nested	0.1488ms	16.5700μs	60.3502 KOps/s	57.4696 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_set_nested_new	52.1910μs	20.0300μs	49.9250 KOps/s	50.6702 KOps/s	$\color{#d91a1a}-1.47\%$
test_select	0.1831ms	32.8621μs	30.4302 KOps/s	30.9802 KOps/s	$\color{#d91a1a}-1.78\%$
test_select_nested	89.1510μs	43.8908μs	22.7838 KOps/s	22.6131 KOps/s	$\color{#35bf28}+0.75\%$
test_exclude_nested	0.1733ms	63.2527μs	15.8096 KOps/s	15.7826 KOps/s	$\color{#35bf28}+0.17\%$
test_empty[True]	0.3967ms	0.2913ms	3.4332 KOps/s	3.4361 KOps/s	$\color{#d91a1a}-0.08\%$
test_empty[False]	3.6820μs	0.8187μs	1.2214 MOps/s	1.2274 MOps/s	$\color{#d91a1a}-0.49\%$
test_to	90.4110μs	56.8464μs	17.5913 KOps/s	17.2573 KOps/s	$\color{#35bf28}+1.94\%$
test_to_nonblocking	0.1007ms	50.1998μs	19.9204 KOps/s	21.5083 KOps/s	$\textbf{\color{#d91a1a}-7.38\%}$
test_unbind_speed	0.4026ms	0.2640ms	3.7872 KOps/s	4.0282 KOps/s	$\textbf{\color{#d91a1a}-5.98\%}$
test_unbind_speed_stack0	0.3295ms	0.2516ms	3.9741 KOps/s	4.0739 KOps/s	$\color{#d91a1a}-2.45\%$
test_unbind_speed_stack1	95.0645ms	0.7444ms	1.3433 KOps/s	1.3282 KOps/s	$\color{#35bf28}+1.14\%$
test_split	97.2283ms	1.6051ms	623.0291 Ops/s	616.1161 Ops/s	$\color{#35bf28}+1.12\%$
test_chunk	98.7518ms	1.6207ms	617.0078 Ops/s	614.8887 Ops/s	$\color{#35bf28}+0.34\%$
test_consolidate[False-None]	97.4986ms	3.0435ms	328.5662 Ops/s	333.1507 Ops/s	$\color{#d91a1a}-1.38\%$
test_consolidate[default-None]	1.8522ms	1.7249ms	579.7357 Ops/s	572.9070 Ops/s	$\color{#35bf28}+1.19\%$
test_consolidate[reduce-overhead-None]	1.8917ms	1.7614ms	567.7272 Ops/s	555.3391 Ops/s	$\color{#35bf28}+2.23\%$
test_consolidate_njt[False-None]	6.8002ms	6.4500ms	155.0384 Ops/s	152.9350 Ops/s	$\color{#35bf28}+1.38\%$
test_to[False-False-None]	1.8819ms	1.7211ms	581.0095 Ops/s	581.4957 Ops/s	$\color{#d91a1a}-0.08\%$
test_to[True-False-None]	1.5658ms	1.3548ms	738.0941 Ops/s	724.1887 Ops/s	$\color{#35bf28}+1.92\%$
test_to[within-False-None]	4.7387ms	4.2163ms	237.1757 Ops/s	232.7361 Ops/s	$\color{#35bf28}+1.91\%$
test_to[True-default-None]	5.5401ms	5.3448ms	187.0987 Ops/s	192.8248 Ops/s	$\color{#d91a1a}-2.97\%$
test_to_njt[False-False-None]	7.4225ms	7.0994ms	140.8562 Ops/s	144.7297 Ops/s	$\color{#d91a1a}-2.68\%$
test_to_njt[True-False-None]	5.8523ms	5.5449ms	180.3469 Ops/s	180.6868 Ops/s	$\color{#d91a1a}-0.19\%$
test_to_njt[within-False-None]	12.4536ms	12.2490ms	81.6395 Ops/s	82.5738 Ops/s	$\color{#d91a1a}-1.13\%$
test_creation[device0]	0.4451ms	83.2468μs	12.0125 KOps/s	11.7182 KOps/s	$\color{#35bf28}+2.51\%$
test_creation_from_tensor	0.5476ms	86.5993μs	11.5474 KOps/s	11.3382 KOps/s	$\color{#35bf28}+1.85\%$
test_add_one[memmap_tensor0]	86.5410μs	7.0275μs	142.2971 KOps/s	140.9419 KOps/s	$\color{#35bf28}+0.96\%$
test_contiguous[memmap_tensor0]	5.3841μs	0.4270μs	2.3421 MOps/s	2.3815 MOps/s	$\color{#d91a1a}-1.66\%$
test_stack[memmap_tensor0]	32.6600μs	4.6573μs	214.7173 KOps/s	207.1545 KOps/s	$\color{#35bf28}+3.65\%$
test_memmaptd_index	1.4941ms	0.2447ms	4.0862 KOps/s	3.9969 KOps/s	$\color{#35bf28}+2.23\%$
test_memmaptd_index_astensor	0.4385ms	0.3044ms	3.2852 KOps/s	3.1838 KOps/s	$\color{#35bf28}+3.18\%$
test_memmaptd_index_op	0.7299ms	0.5820ms	1.7183 KOps/s	1.6508 KOps/s	$\color{#35bf28}+4.09\%$
test_serialize_model	0.1316s	0.1311s	7.6273 Ops/s	7.6209 Ops/s	$\color{#35bf28}+0.08\%$
test_serialize_model_pickle	1.3509s	1.1868s	0.8426 Ops/s	0.8400 Ops/s	$\color{#35bf28}+0.31\%$
test_serialize_weights	0.1319s	0.1305s	7.6652 Ops/s	7.6026 Ops/s	$\color{#35bf28}+0.82\%$
test_serialize_weights_returnearly	0.3485s	55.2986ms	18.0836 Ops/s	23.3958 Ops/s	$\textbf{\color{#d91a1a}-22.71\%}$
test_serialize_weights_pickle	1.3750s	1.2167s	0.8219 Ops/s	0.8242 Ops/s	$\color{#d91a1a}-0.27\%$
test_reshape_pytree	54.0810μs	22.3285μs	44.7857 KOps/s	45.8557 KOps/s	$\color{#d91a1a}-2.33\%$
test_reshape_td	0.1541ms	27.3629μs	36.5459 KOps/s	37.9590 KOps/s	$\color{#d91a1a}-3.72\%$
test_view_pytree	67.1410μs	22.0967μs	45.2555 KOps/s	46.2264 KOps/s	$\color{#d91a1a}-2.10\%$
test_view_td	76.3710μs	31.2411μs	32.0091 KOps/s	30.8025 KOps/s	$\color{#35bf28}+3.92\%$
test_unbind_pytree	85.9310μs	28.4654μs	35.1304 KOps/s	35.8248 KOps/s	$\color{#d91a1a}-1.94\%$
test_unbind_td	0.8470ms	38.2284μs	26.1585 KOps/s	26.9433 KOps/s	$\color{#d91a1a}-2.91\%$
test_split_pytree	0.1590ms	30.1338μs	33.1853 KOps/s	34.1242 KOps/s	$\color{#d91a1a}-2.75\%$
test_split_td	1.0059ms	38.6332μs	25.8845 KOps/s	25.5686 KOps/s	$\color{#35bf28}+1.24\%$
test_add_pytree	69.6600μs	35.1398μs	28.4578 KOps/s	28.6154 KOps/s	$\color{#d91a1a}-0.55\%$
test_add_td	0.1670ms	49.8865μs	20.0455 KOps/s	19.9528 KOps/s	$\color{#35bf28}+0.46\%$
test_compile_add_one_nested[tensordict-compile]	0.2709ms	0.1237ms	8.0859 KOps/s	8.0150 KOps/s	$\color{#35bf28}+0.88\%$
test_compile_add_one_nested[tensordict-eager]	0.2906ms	0.1361ms	7.3477 KOps/s	7.6766 KOps/s	$\color{#d91a1a}-4.28\%$
test_compile_add_one_nested[pytree-compile]	0.2462ms	97.7001μs	10.2354 KOps/s	10.4834 KOps/s	$\color{#d91a1a}-2.37\%$
test_compile_add_one_nested[pytree-eager]	1.4293ms	0.1523ms	6.5658 KOps/s	6.6712 KOps/s	$\color{#d91a1a}-1.58\%$
test_compile_copy_nested[tensordict-compile]	0.1873ms	34.4003μs	29.0695 KOps/s	43.4093 KOps/s	$\textbf{\color{#d91a1a}-33.03\%}$
test_compile_copy_nested[tensordict-eager]	0.1170ms	28.9765μs	34.5107 KOps/s	33.8366 KOps/s	$\color{#35bf28}+1.99\%$
test_compile_copy_nested[pytree-compile]	0.4761ms	64.0470μs	15.6135 KOps/s	15.8186 KOps/s	$\color{#d91a1a}-1.30\%$
test_compile_copy_nested[pytree-eager]	99.2020μs	48.5864μs	20.5819 KOps/s	20.5932 KOps/s	$\color{#d91a1a}-0.06\%$
test_compile_add_one_flat[tensordict-compile]	0.2357ms	0.1430ms	6.9927 KOps/s	7.0769 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_add_one_flat[tensordict-eager]	0.3529ms	0.2183ms	4.5800 KOps/s	4.6600 KOps/s	$\color{#d91a1a}-1.72\%$
test_compile_add_one_flat[tensorclass-compile]	0.2417ms	98.7179μs	10.1299 KOps/s	10.4078 KOps/s	$\color{#d91a1a}-2.67\%$
test_compile_add_one_flat[tensorclass-eager]	0.2180ms	58.3525μs	17.1372 KOps/s	18.4333 KOps/s	$\textbf{\color{#d91a1a}-7.03\%}$
test_compile_add_one_flat[pytree-compile]	0.2769ms	0.1445ms	6.9189 KOps/s	7.3398 KOps/s	$\textbf{\color{#d91a1a}-5.73\%}$
test_compile_add_one_flat[pytree-eager]	0.6981ms	0.4947ms	2.0214 KOps/s	2.0622 KOps/s	$\color{#d91a1a}-1.98\%$
test_compile_add_self_flat[tensordict-eager]	0.3954ms	0.2613ms	3.8276 KOps/s	3.8388 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_add_self_flat[tensordict-compile]	0.2950ms	0.1461ms	6.8442 KOps/s	7.0500 KOps/s	$\color{#d91a1a}-2.92\%$
test_compile_add_self_flat[tensorclass-eager]	0.2587ms	70.3457μs	14.2155 KOps/s	15.1430 KOps/s	$\textbf{\color{#d91a1a}-6.13\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1472ms	0.1011ms	9.8950 KOps/s	10.0379 KOps/s	$\color{#d91a1a}-1.42\%$
test_compile_add_self_flat[pytree-eager]	0.6196ms	0.4221ms	2.3689 KOps/s	2.4368 KOps/s	$\color{#d91a1a}-2.79\%$
test_compile_add_self_flat[pytree-compile]	0.1836ms	0.1365ms	7.3247 KOps/s	7.4907 KOps/s	$\color{#d91a1a}-2.22\%$
test_compile_copy_flat[tensordict-compile]	0.1961ms	18.9809μs	52.6844 KOps/s	51.1662 KOps/s	$\color{#35bf28}+2.97\%$
test_compile_copy_flat[tensordict-eager]	0.2054ms	31.0005μs	32.2575 KOps/s	31.7948 KOps/s	$\color{#35bf28}+1.46\%$
test_compile_copy_flat[pytree-compile]	0.1131ms	70.0451μs	14.2765 KOps/s	14.4954 KOps/s	$\color{#d91a1a}-1.51\%$
test_compile_copy_flat[pytree-eager]	0.1542ms	51.4831μs	19.4238 KOps/s	19.1192 KOps/s	$\color{#35bf28}+1.59\%$
test_compile_assign_and_add[tensordict-compile]	1.6777ms	0.4005ms	2.4970 KOps/s	2.2324 KOps/s	$\textbf{\color{#35bf28}+11.85\%}$
test_compile_assign_and_add[tensordict-eager]	2.8988ms	2.6617ms	375.6933 Ops/s	370.5610 Ops/s	$\color{#35bf28}+1.39\%$
test_compile_assign_and_add[pytree-compile]	1.6403ms	0.4447ms	2.2489 KOps/s	2.2838 KOps/s	$\color{#d91a1a}-1.53\%$
test_compile_assign_and_add[pytree-eager]	2.8702ms	2.6948ms	371.0815 Ops/s	373.1878 Ops/s	$\color{#d91a1a}-0.56\%$
test_compile_indexing[tensor-tensordict-compile]	0.2792ms	0.1168ms	8.5641 KOps/s	8.8786 KOps/s	$\color{#d91a1a}-3.54\%$
test_compile_indexing[tensor-tensordict-eager]	0.5912ms	80.8391μs	12.3703 KOps/s	12.4941 KOps/s	$\color{#d91a1a}-0.99\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2630ms	0.1124ms	8.8983 KOps/s	9.4679 KOps/s	$\textbf{\color{#d91a1a}-6.02\%}$
test_compile_indexing[tensor-tensorclass-eager]	0.2223ms	72.2211μs	13.8464 KOps/s	14.2659 KOps/s	$\color{#d91a1a}-2.94\%$
test_compile_indexing[tensor-pytree-compile]	0.2966ms	0.1157ms	8.6459 KOps/s	9.3858 KOps/s	$\textbf{\color{#d91a1a}-7.88\%}$
test_compile_indexing[tensor-pytree-eager]	0.2229ms	72.7442μs	13.7468 KOps/s	14.7394 KOps/s	$\textbf{\color{#d91a1a}-6.73\%}$
test_compile_indexing[slice-tensordict-compile]	0.2406ms	0.1034ms	9.6688 KOps/s	9.9543 KOps/s	$\color{#d91a1a}-2.87\%$
test_compile_indexing[slice-tensordict-eager]	0.2374ms	17.3312μs	57.6993 KOps/s	58.7673 KOps/s	$\color{#d91a1a}-1.82\%$
test_compile_indexing[slice-tensorclass-compile]	0.2442ms	97.0515μs	10.3038 KOps/s	10.2938 KOps/s	$\color{#35bf28}+0.10\%$
test_compile_indexing[slice-tensorclass-eager]	0.1006ms	15.7928μs	63.3199 KOps/s	63.5440 KOps/s	$\color{#d91a1a}-0.35\%$
test_compile_indexing[slice-pytree-compile]	0.2470ms	0.1006ms	9.9433 KOps/s	10.2285 KOps/s	$\color{#d91a1a}-2.79\%$
test_compile_indexing[slice-pytree-eager]	0.1491ms	15.6216μs	64.0140 KOps/s	57.7881 KOps/s	$\textbf{\color{#35bf28}+10.77\%}$
test_compile_indexing[int-tensordict-compile]	0.2335ms	0.1065ms	9.3908 KOps/s	9.8717 KOps/s	$\color{#d91a1a}-4.87\%$
test_compile_indexing[int-tensordict-eager]	0.6010ms	17.2393μs	58.0069 KOps/s	56.9002 KOps/s	$\color{#35bf28}+1.94\%$
test_compile_indexing[int-tensorclass-compile]	0.2476ms	99.9180μs	10.0082 KOps/s	10.2800 KOps/s	$\color{#d91a1a}-2.64\%$
test_compile_indexing[int-tensorclass-eager]	49.4100μs	15.7176μs	63.6230 KOps/s	64.0819 KOps/s	$\color{#d91a1a}-0.72\%$
test_compile_indexing[int-pytree-compile]	0.2467ms	99.7056μs	10.0295 KOps/s	10.2547 KOps/s	$\color{#d91a1a}-2.20\%$
test_compile_indexing[int-pytree-eager]	0.1603ms	15.7275μs	63.5829 KOps/s	63.7773 KOps/s	$\color{#d91a1a}-0.30\%$
test_mod_add[eager]	0.1959ms	39.6936μs	25.1930 KOps/s	24.8848 KOps/s	$\color{#35bf28}+1.24\%$
test_mod_add[compile]	0.2274ms	82.6139μs	12.1045 KOps/s	12.1249 KOps/s	$\color{#d91a1a}-0.17\%$
test_mod_add[compile-overhead]	0.3357ms	0.1691ms	5.9128 KOps/s	5.6948 KOps/s	$\color{#35bf28}+3.83\%$
test_mod_wrap[eager]	0.4317ms	0.2565ms	3.8981 KOps/s	3.7958 KOps/s	$\color{#35bf28}+2.69\%$
test_mod_wrap[compile]	0.4845ms	0.2853ms	3.5048 KOps/s	3.4545 KOps/s	$\color{#35bf28}+1.46\%$
test_mod_wrap[compile-overhead]	7.0646ms	3.7569ms	266.1744 Ops/s	261.4097 Ops/s	$\color{#35bf28}+1.82\%$
test_mod_wrap_and_backward[eager]	1.5930ms	1.3911ms	718.8323 Ops/s	690.3346 Ops/s	$\color{#35bf28}+4.13\%$
test_mod_wrap_and_backward[compile]	1.3996ms	1.2914ms	774.3684 Ops/s	726.2191 Ops/s	$\textbf{\color{#35bf28}+6.63\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3933ms	0.9305ms	1.0747 KOps/s	929.7138 Ops/s	$\textbf{\color{#35bf28}+15.60\%}$
test_seq_add[eager]	0.2939ms	0.1175ms	8.5096 KOps/s	8.3172 KOps/s	$\color{#35bf28}+2.31\%$
test_seq_add[compile]	0.2065ms	91.6315μs	10.9133 KOps/s	11.1644 KOps/s	$\color{#d91a1a}-2.25\%$
test_seq_add[compile-overhead]	0.2554ms	0.1307ms	7.6523 KOps/s	7.6539 KOps/s	$\color{#d91a1a}-0.02\%$
test_seq_wrap[eager]	0.5729ms	0.4320ms	2.3148 KOps/s	2.2365 KOps/s	$\color{#35bf28}+3.50\%$
test_seq_wrap[compile]	0.4488ms	0.3040ms	3.2894 KOps/s	3.1596 KOps/s	$\color{#35bf28}+4.11\%$
test_seq_wrap[compile-overhead]	0.3767ms	0.2274ms	4.3972 KOps/s	4.3434 KOps/s	$\color{#35bf28}+1.24\%$
test_func_call_runtime[False-eager]	0.9572ms	0.7667ms	1.3043 KOps/s	1.3467 KOps/s	$\color{#d91a1a}-3.15\%$
test_func_call_runtime[False-compile]	1.0240ms	0.7544ms	1.3256 KOps/s	1.3320 KOps/s	$\color{#d91a1a}-0.48\%$
test_func_call_runtime[False-compile-overhead]	0.5064ms	0.3678ms	2.7189 KOps/s	2.7055 KOps/s	$\color{#35bf28}+0.50\%$
test_func_call_runtime[True-eager]	1.0327ms	0.9069ms	1.1027 KOps/s	1.0993 KOps/s	$\color{#35bf28}+0.30\%$
test_func_call_runtime[True-compile]	0.9820ms	0.7772ms	1.2867 KOps/s	1.2767 KOps/s	$\color{#35bf28}+0.78\%$
test_func_call_runtime[True-compile-overhead]	0.5480ms	0.3910ms	2.5574 KOps/s	2.5669 KOps/s	$\color{#d91a1a}-0.37\%$
test_func_call_cm_runtime[False-eager]	0.8801ms	0.7384ms	1.3543 KOps/s	1.3375 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_cm_runtime[False-compile]	0.9075ms	0.7575ms	1.3201 KOps/s	1.3122 KOps/s	$\color{#35bf28}+0.60\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4826ms	0.3671ms	2.7240 KOps/s	2.6921 KOps/s	$\color{#35bf28}+1.19\%$
test_func_call_cm_runtime[True-eager]	1.2287ms	1.0094ms	990.7113 Ops/s	976.3917 Ops/s	$\color{#35bf28}+1.47\%$
test_func_call_cm_runtime[True-compile]	1.1415ms	0.9991ms	1.0009 KOps/s	994.7598 Ops/s	$\color{#35bf28}+0.62\%$
test_func_call_cm_runtime[True-compile-overhead]	1.2260ms	1.0305ms	970.3893 Ops/s	991.0205 Ops/s	$\color{#d91a1a}-2.08\%$
test_vmap_func_call_cm_runtime[eager]	2.5463ms	2.1137ms	473.1031 Ops/s	470.4754 Ops/s	$\color{#35bf28}+0.56\%$
test_vmap_func_call_cm_runtime[compile]	0.9844ms	0.8250ms	1.2121 KOps/s	1.1469 KOps/s	$\textbf{\color{#35bf28}+5.68\%}$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5643ms	0.4207ms	2.3767 KOps/s	2.3825 KOps/s	$\color{#d91a1a}-0.24\%$
test_distributed	2.6389ms	0.2528ms	3.9565 KOps/s	7.8485 KOps/s	$\textbf{\color{#d91a1a}-49.59\%}$
test_tdmodule	39.7210μs	20.6520μs	48.4215 KOps/s	47.2521 KOps/s	$\color{#35bf28}+2.47\%$
test_tdmodule_dispatch	0.3133ms	36.8986μs	27.1013 KOps/s	26.2086 KOps/s	$\color{#35bf28}+3.41\%$
test_tdseq	40.9500μs	21.1988μs	47.1725 KOps/s	46.2214 KOps/s	$\color{#35bf28}+2.06\%$
test_tdseq_dispatch	60.4110μs	39.1511μs	25.5421 KOps/s	24.6052 KOps/s	$\color{#35bf28}+3.81\%$
test_instantiation_functorch	1.9200ms	1.5453ms	647.1348 Ops/s	632.2418 Ops/s	$\color{#35bf28}+2.36\%$
test_exec_functorch	0.2173ms	0.1466ms	6.8197 KOps/s	6.7155 KOps/s	$\color{#35bf28}+1.55\%$
test_exec_functional_call	0.5410ms	0.1431ms	6.9873 KOps/s	6.9949 KOps/s	$\color{#d91a1a}-0.11\%$
test_exec_td_decorator	0.4098ms	0.2005ms	4.9871 KOps/s	5.1264 KOps/s	$\color{#d91a1a}-2.72\%$
test_vmap_mlp_speed_decorator[True-True]	1.1076ms	0.7114ms	1.4056 KOps/s	1.4481 KOps/s	$\color{#d91a1a}-2.93\%$
test_vmap_mlp_speed_decorator[True-False]	1.1144ms	0.6950ms	1.4389 KOps/s	1.4102 KOps/s	$\color{#35bf28}+2.03\%$
test_vmap_mlp_speed_decorator[False-True]	1.0271ms	0.6175ms	1.6194 KOps/s	1.6010 KOps/s	$\color{#35bf28}+1.15\%$
test_vmap_mlp_speed_decorator[False-False]	1.0189ms	0.6026ms	1.6594 KOps/s	1.6066 KOps/s	$\color{#35bf28}+3.29\%$
test_vmap_transformer_speed_decorator[True-True]	20.0732ms	19.5188ms	51.2326 Ops/s	51.2764 Ops/s	$\color{#d91a1a}-0.09\%$
test_vmap_transformer_speed_decorator[True-False]	19.9327ms	19.4875ms	51.3149 Ops/s	51.7396 Ops/s	$\color{#d91a1a}-0.82\%$
test_vmap_transformer_speed_decorator[False-True]	20.5949ms	19.4997ms	51.2827 Ops/s	50.5930 Ops/s	$\color{#35bf28}+1.36\%$
test_vmap_transformer_speed_decorator[False-False]	20.2915ms	19.6066ms	51.0033 Ops/s	51.3773 Ops/s	$\color{#d91a1a}-0.73\%$
test_to_module_speed[True]	1.5180ms	0.9624ms	1.0391 KOps/s	1.0380 KOps/s	$\color{#35bf28}+0.11\%$
test_to_module_speed[False]	1.4261ms	0.9453ms	1.0579 KOps/s	1.0637 KOps/s	$\color{#d91a1a}-0.55\%$
test_tc_init	0.1772ms	34.0621μs	29.3582 KOps/s	28.1250 KOps/s	$\color{#35bf28}+4.38\%$
test_tc_init_nested	0.4685ms	67.8163μs	14.7457 KOps/s	13.7500 KOps/s	$\textbf{\color{#35bf28}+7.24\%}$
test_tc_first_layer_tensor	24.5500μs	0.7925μs	1.2618 MOps/s	1.2432 MOps/s	$\color{#35bf28}+1.50\%$
test_tc_first_layer_nontensor	25.8510μs	2.2340μs	447.6303 KOps/s	459.0809 KOps/s	$\color{#d91a1a}-2.49\%$
test_tc_second_layer_tensor	9.2378μs	1.3967μs	715.9950 KOps/s	708.7822 KOps/s	$\color{#35bf28}+1.02\%$
test_tc_second_layer_nontensor	0.3875ms	2.9389μs	340.2643 KOps/s	343.6123 KOps/s	$\color{#d91a1a}-0.97\%$
test_unbind	0.2354s	10.0940ms	99.0688 Ops/s	145.4760 Ops/s	$\textbf{\color{#d91a1a}-31.90\%}$
test_full_like	10.3245ms	9.7083ms	103.0044 Ops/s	103.5905 Ops/s	$\color{#d91a1a}-0.57\%$
test_zeros_like	5.4436ms	4.4110ms	226.7079 Ops/s	114.5448 Ops/s	$\textbf{\color{#35bf28}+97.92\%}$
test_ones_like	9.3673ms	7.2734ms	137.4874 Ops/s	228.3569 Ops/s	$\textbf{\color{#d91a1a}-39.79\%}$
test_clone	7.5651ms	6.7695ms	147.7211 Ops/s	147.1171 Ops/s	$\color{#35bf28}+0.41\%$
test_squeeze	98.9710μs	9.8520μs	101.5018 KOps/s	102.0188 KOps/s	$\color{#d91a1a}-0.51\%$
test_unsqueeze	0.2065ms	77.2383μs	12.9469 KOps/s	12.6867 KOps/s	$\color{#35bf28}+2.05\%$
test_split	0.6664ms	0.1635ms	6.1145 KOps/s	6.1189 KOps/s	$\color{#d91a1a}-0.07\%$
test_permute	0.3862ms	0.1930ms	5.1817 KOps/s	5.1854 KOps/s	$\color{#d91a1a}-0.07\%$
test_stack	52.6359ms	51.5933ms	19.3824 Ops/s	19.5552 Ops/s	$\color{#d91a1a}-0.88\%$
test_cat	52.0645ms	51.5279ms	19.4069 Ops/s	23.2624 Ops/s	$\textbf{\color{#d91a1a}-16.57\%}$