[Feature] NonTensorStack.from_list

[vmoens]

Stack from ghstack (oldest at bottom):

• [BugFix] auto-batch-size in dispatch #1109
• [Feature] Better list casting in TensorDict.from_any #1108
• -> [Feature] NonTensorStack.from_list #1107
• [Feature] from_any with UserDict #1106

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}39$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	34.2130μs	17.6071μs	56.7953 KOps/s	52.8006 KOps/s	$\textbf{\color{#35bf28}+7.57\%}$
test_plain_set_stack_nested	46.5770μs	17.5654μs	56.9302 KOps/s	52.2711 KOps/s	$\textbf{\color{#35bf28}+8.91\%}$
test_plain_set_nested_inplace	78.6610μs	19.0944μs	52.3714 KOps/s	48.0150 KOps/s	$\textbf{\color{#35bf28}+9.07\%}$
test_plain_set_stack_nested_inplace	49.2610μs	18.9100μs	52.8820 KOps/s	48.5064 KOps/s	$\textbf{\color{#35bf28}+9.02\%}$
test_items	50.4640μs	4.1633μs	240.1969 KOps/s	245.2319 KOps/s	$\color{#d91a1a}-2.05\%$
test_items_nested	0.4335ms	0.3438ms	2.9085 KOps/s	2.8972 KOps/s	$\color{#35bf28}+0.39\%$
test_items_nested_locked	0.4543ms	0.3442ms	2.9054 KOps/s	2.8890 KOps/s	$\color{#35bf28}+0.57\%$
test_items_nested_leaf	0.1394ms	72.7631μs	13.7432 KOps/s	13.8254 KOps/s	$\color{#d91a1a}-0.59\%$
test_items_stack_nested	0.4341ms	0.3449ms	2.8991 KOps/s	2.8756 KOps/s	$\color{#35bf28}+0.82\%$
test_items_stack_nested_leaf	0.1698ms	74.1685μs	13.4828 KOps/s	13.2387 KOps/s	$\color{#35bf28}+1.84\%$
test_items_stack_nested_locked	0.4496ms	0.3455ms	2.8945 KOps/s	2.9009 KOps/s	$\color{#d91a1a}-0.22\%$
test_keys	42.2190μs	3.5405μs	282.4479 KOps/s	288.1977 KOps/s	$\color{#d91a1a}-2.00\%$
test_keys_nested	0.2346ms	0.1376ms	7.2692 KOps/s	7.3586 KOps/s	$\color{#d91a1a}-1.21\%$
test_keys_nested_locked	1.8554ms	0.1442ms	6.9370 KOps/s	7.0581 KOps/s	$\color{#d91a1a}-1.72\%$
test_keys_nested_leaf	0.1929ms	0.1170ms	8.5467 KOps/s	8.5418 KOps/s	$\color{#35bf28}+0.06\%$
test_keys_stack_nested	0.1973ms	0.1382ms	7.2352 KOps/s	7.3582 KOps/s	$\color{#d91a1a}-1.67\%$
test_keys_stack_nested_leaf	0.2111ms	0.1202ms	8.3163 KOps/s	8.7134 KOps/s	$\color{#d91a1a}-4.56\%$
test_keys_stack_nested_locked	0.2664ms	0.1430ms	6.9939 KOps/s	7.0037 KOps/s	$\color{#d91a1a}-0.14\%$
test_values	6.3360μs	1.0312μs	969.7110 KOps/s	937.8332 KOps/s	$\color{#35bf28}+3.40\%$
test_values_nested	99.2860μs	55.2109μs	18.1124 KOps/s	18.0937 KOps/s	$\color{#35bf28}+0.10\%$
test_values_nested_locked	0.1071ms	55.7649μs	17.9324 KOps/s	18.1298 KOps/s	$\color{#d91a1a}-1.09\%$
test_values_nested_leaf	0.1171ms	60.1442μs	16.6267 KOps/s	16.2947 KOps/s	$\color{#35bf28}+2.04\%$
test_values_stack_nested	95.0680μs	55.8960μs	17.8904 KOps/s	17.4872 KOps/s	$\color{#35bf28}+2.31\%$
test_values_stack_nested_leaf	0.1167ms	61.2171μs	16.3353 KOps/s	16.5319 KOps/s	$\color{#d91a1a}-1.19\%$
test_values_stack_nested_locked	99.3860μs	56.0340μs	17.8463 KOps/s	17.7197 KOps/s	$\color{#35bf28}+0.71\%$
test_membership	21.2100μs	0.8955μs	1.1167 MOps/s	1.4242 MOps/s	$\textbf{\color{#d91a1a}-21.59\%}$
test_membership_nested	53.4970μs	2.7392μs	365.0660 KOps/s	354.4542 KOps/s	$\color{#35bf28}+2.99\%$
test_membership_nested_leaf	20.5290μs	2.7944μs	357.8544 KOps/s	348.6260 KOps/s	$\color{#35bf28}+2.65\%$
test_membership_stacked_nested	25.6880μs	2.7460μs	364.1707 KOps/s	357.3659 KOps/s	$\color{#35bf28}+1.90\%$
test_membership_stacked_nested_leaf	42.5700μs	2.7985μs	357.3343 KOps/s	354.7293 KOps/s	$\color{#35bf28}+0.73\%$
test_membership_nested_last	29.2640μs	4.1195μs	242.7456 KOps/s	240.4442 KOps/s	$\color{#35bf28}+0.96\%$
test_membership_nested_leaf_last	48.1300μs	4.0704μs	245.6765 KOps/s	236.7923 KOps/s	$\color{#35bf28}+3.75\%$
test_membership_stacked_nested_last	28.1120μs	4.1117μs	243.2069 KOps/s	119.0010 KOps/s	$\textbf{\color{#35bf28}+104.37\%}$
test_membership_stacked_nested_leaf_last	53.9810μs	4.1025μs	243.7562 KOps/s	119.7516 KOps/s	$\textbf{\color{#35bf28}+103.55\%}$
test_nested_getleaf	41.6180μs	10.4429μs	95.7591 KOps/s	92.4936 KOps/s	$\color{#35bf28}+3.53\%$
test_nested_get	51.7470μs	10.0260μs	99.7404 KOps/s	97.4579 KOps/s	$\color{#35bf28}+2.34\%$
test_stacked_getleaf	47.4180μs	10.2908μs	97.1740 KOps/s	94.9869 KOps/s	$\color{#35bf28}+2.30\%$
test_stacked_get	37.8310μs	9.8822μs	101.1919 KOps/s	99.1595 KOps/s	$\color{#35bf28}+2.05\%$
test_nested_getitemleaf	62.0760μs	10.9447μs	91.3680 KOps/s	89.2949 KOps/s	$\color{#35bf28}+2.32\%$
test_nested_getitem	53.1900μs	10.3387μs	96.7239 KOps/s	96.5712 KOps/s	$\color{#35bf28}+0.16\%$
test_stacked_getitemleaf	39.6040μs	10.8805μs	91.9078 KOps/s	89.9902 KOps/s	$\color{#35bf28}+2.13\%$
test_stacked_getitem	54.2620μs	10.2123μs	97.9214 KOps/s	95.8201 KOps/s	$\color{#35bf28}+2.19\%$
test_lock_nested	2.9805ms	0.4438ms	2.2533 KOps/s	2.2549 KOps/s	$\color{#d91a1a}-0.07\%$
test_lock_stack_nested	0.7821ms	0.4120ms	2.4271 KOps/s	2.4720 KOps/s	$\color{#d91a1a}-1.82\%$
test_unlock_nested	0.9768ms	0.3558ms	2.8110 KOps/s	2.7681 KOps/s	$\color{#35bf28}+1.55\%$
test_unlock_stack_nested	0.5991ms	0.3300ms	3.0308 KOps/s	3.0805 KOps/s	$\color{#d91a1a}-1.62\%$
test_flatten_speed	0.1634ms	93.5959μs	10.6842 KOps/s	10.6121 KOps/s	$\color{#35bf28}+0.68\%$
test_unflatten_speed	0.9952ms	0.4779ms	2.0924 KOps/s	2.0675 KOps/s	$\color{#35bf28}+1.20\%$
test_common_ops	3.8612ms	0.7464ms	1.3397 KOps/s	1.2641 KOps/s	$\textbf{\color{#35bf28}+5.98\%}$
test_creation	24.6360μs	2.0957μs	477.1656 KOps/s	476.8400 KOps/s	$\color{#35bf28}+0.07\%$
test_creation_empty	35.9080μs	9.4892μs	105.3832 KOps/s	78.7941 KOps/s	$\textbf{\color{#35bf28}+33.75\%}$
test_creation_nested_1	41.0570μs	12.3871μs	80.7292 KOps/s	63.8492 KOps/s	$\textbf{\color{#35bf28}+26.44\%}$
test_creation_nested_2	49.7230μs	16.4740μs	60.7018 KOps/s	50.1899 KOps/s	$\textbf{\color{#35bf28}+20.94\%}$
test_clone	80.4300μs	12.9198μs	77.4004 KOps/s	76.0702 KOps/s	$\color{#35bf28}+1.75\%$
test_getitem[int]	1.4472ms	12.4138μs	80.5552 KOps/s	78.5778 KOps/s	$\color{#35bf28}+2.52\%$
test_getitem[slice_int]	0.1402ms	24.9111μs	40.1428 KOps/s	41.2607 KOps/s	$\color{#d91a1a}-2.71\%$
test_getitem[range]	0.1740ms	48.2158μs	20.7401 KOps/s	19.8169 KOps/s	$\color{#35bf28}+4.66\%$
test_getitem[tuple]	0.1262ms	20.1615μs	49.5994 KOps/s	50.3068 KOps/s	$\color{#d91a1a}-1.41\%$
test_getitem[list]	0.1768ms	43.6199μs	22.9253 KOps/s	21.8802 KOps/s	$\color{#35bf28}+4.78\%$
test_setitem_dim[int]	49.5230μs	24.9568μs	40.0692 KOps/s	38.9498 KOps/s	$\color{#35bf28}+2.87\%$
test_setitem_dim[slice_int]	87.8540μs	51.1625μs	19.5455 KOps/s	19.4306 KOps/s	$\color{#35bf28}+0.59\%$
test_setitem_dim[range]	0.1180ms	74.6258μs	13.4002 KOps/s	13.4540 KOps/s	$\color{#d91a1a}-0.40\%$
test_setitem_dim[tuple]	66.6850μs	40.7265μs	24.5540 KOps/s	24.6741 KOps/s	$\color{#d91a1a}-0.49\%$
test_setitem	70.1310μs	19.2617μs	51.9165 KOps/s	47.6767 KOps/s	$\textbf{\color{#35bf28}+8.89\%}$
test_set	82.9260μs	18.4588μs	54.1748 KOps/s	48.7505 KOps/s	$\textbf{\color{#35bf28}+11.13\%}$
test_set_shared	1.2546ms	0.1662ms	6.0186 KOps/s	5.9182 KOps/s	$\color{#35bf28}+1.70\%$
test_update	0.1999ms	20.7197μs	48.2632 KOps/s	40.2898 KOps/s	$\textbf{\color{#35bf28}+19.79\%}$
test_update_nested	92.8940μs	31.0676μs	32.1879 KOps/s	29.5252 KOps/s	$\textbf{\color{#35bf28}+9.02\%}$
test_update__nested	0.5156ms	31.3218μs	31.9266 KOps/s	30.5907 KOps/s	$\color{#35bf28}+4.37\%$
test_set_nested	66.3640μs	20.3862μs	49.0527 KOps/s	44.5087 KOps/s	$\textbf{\color{#35bf28}+10.21\%}$
test_set_nested_new	91.2210μs	25.2116μs	39.6643 KOps/s	36.3357 KOps/s	$\textbf{\color{#35bf28}+9.16\%}$
test_select	0.1025ms	41.1387μs	24.3080 KOps/s	22.5799 KOps/s	$\textbf{\color{#35bf28}+7.65\%}$
test_select_nested	0.1198ms	60.4402μs	16.5453 KOps/s	16.5589 KOps/s	$\color{#d91a1a}-0.08\%$
test_exclude_nested	0.1397ms	75.6149μs	13.2249 KOps/s	13.1108 KOps/s	$\color{#35bf28}+0.87\%$
test_empty[True]	0.4564ms	0.3516ms	2.8444 KOps/s	2.8581 KOps/s	$\color{#d91a1a}-0.48\%$
test_empty[False]	10.2015μs	1.2533μs	797.9199 KOps/s	810.5400 KOps/s	$\color{#d91a1a}-1.56\%$
test_unbind_speed	0.4331ms	0.2614ms	3.8254 KOps/s	3.6608 KOps/s	$\color{#35bf28}+4.50\%$
test_unbind_speed_stack0	0.3875ms	0.2594ms	3.8546 KOps/s	3.8794 KOps/s	$\color{#d91a1a}-0.64\%$
test_unbind_speed_stack1	98.0568ms	0.7615ms	1.3131 KOps/s	1.4860 KOps/s	$\textbf{\color{#d91a1a}-11.63\%}$
test_split	96.3091ms	1.7239ms	580.0810 Ops/s	562.3111 Ops/s	$\color{#35bf28}+3.16\%$
test_chunk	99.1518ms	1.7389ms	575.0865 Ops/s	568.2321 Ops/s	$\color{#35bf28}+1.21\%$
test_consolidate_njt[False-None]	14.5029ms	8.0197ms	124.6936 Ops/s	122.0791 Ops/s	$\color{#35bf28}+2.14\%$
test_creation[device0]	3.5571ms	91.0493μs	10.9831 KOps/s	10.6502 KOps/s	$\color{#35bf28}+3.13\%$
test_creation_from_tensor	0.2196ms	91.6303μs	10.9134 KOps/s	10.5345 KOps/s	$\color{#35bf28}+3.60\%$
test_add_one[memmap_tensor0]	0.1461ms	4.7277μs	211.5187 KOps/s	204.0405 KOps/s	$\color{#35bf28}+3.67\%$
test_contiguous[memmap_tensor0]	17.0720μs	0.5101μs	1.9603 MOps/s	1.9459 MOps/s	$\color{#35bf28}+0.74\%$
test_stack[memmap_tensor0]	31.9300μs	3.4643μs	288.6615 KOps/s	293.1973 KOps/s	$\color{#d91a1a}-1.55\%$
test_memmaptd_index	1.0733ms	0.2363ms	4.2325 KOps/s	4.1853 KOps/s	$\color{#35bf28}+1.13\%$
test_memmaptd_index_astensor	0.5488ms	0.3157ms	3.1678 KOps/s	3.1520 KOps/s	$\color{#35bf28}+0.50\%$
test_memmaptd_index_op	0.8837ms	0.5483ms	1.8239 KOps/s	1.6192 KOps/s	$\textbf{\color{#35bf28}+12.65\%}$
test_serialize_model	0.1181s	0.1124s	8.8943 Ops/s	8.6881 Ops/s	$\color{#35bf28}+2.37\%$
test_serialize_model_pickle	0.4520s	0.3934s	2.5417 Ops/s	2.5850 Ops/s	$\color{#d91a1a}-1.67\%$
test_serialize_weights	0.2065s	0.1271s	7.8680 Ops/s	8.8872 Ops/s	$\textbf{\color{#d91a1a}-11.47\%}$
test_serialize_weights_returnearly	0.1719s	0.1577s	6.3421 Ops/s	6.4941 Ops/s	$\color{#d91a1a}-2.34\%$
test_serialize_weights_pickle	0.4659s	0.4146s	2.4117 Ops/s	2.4798 Ops/s	$\color{#d91a1a}-2.75\%$
test_serialize_weights_filesystem	0.1453s	0.1387s	7.2096 Ops/s	7.0927 Ops/s	$\color{#35bf28}+1.65\%$
test_serialize_model_filesystem	0.2392s	0.1601s	6.2462 Ops/s	6.6927 Ops/s	$\textbf{\color{#d91a1a}-6.67\%}$
test_reshape_pytree	64.8910μs	27.2652μs	36.6768 KOps/s	38.1235 KOps/s	$\color{#d91a1a}-3.79\%$
test_reshape_td	76.7340μs	32.4022μs	30.8621 KOps/s	29.9498 KOps/s	$\color{#35bf28}+3.05\%$
test_view_pytree	87.9550μs	27.6066μs	36.2232 KOps/s	37.3980 KOps/s	$\color{#d91a1a}-3.14\%$
test_view_td	78.7670μs	37.4325μs	26.7147 KOps/s	25.8974 KOps/s	$\color{#35bf28}+3.16\%$
test_unbind_pytree	54.3320μs	30.1525μs	33.1648 KOps/s	33.8859 KOps/s	$\color{#d91a1a}-2.13\%$
test_unbind_td	0.3435ms	39.3315μs	25.4249 KOps/s	25.9276 KOps/s	$\color{#d91a1a}-1.94\%$
test_split_pytree	75.0310μs	29.8945μs	33.4509 KOps/s	34.0098 KOps/s	$\color{#d91a1a}-1.64\%$
test_split_td	0.2194ms	44.1159μs	22.6676 KOps/s	22.0495 KOps/s	$\color{#35bf28}+2.80\%$
test_add_pytree	90.9300μs	36.3265μs	27.5281 KOps/s	28.1071 KOps/s	$\color{#d91a1a}-2.06\%$
test_add_td	0.2732ms	53.3129μs	18.7572 KOps/s	16.6729 KOps/s	$\textbf{\color{#35bf28}+12.50\%}$
test_compile_add_one_nested[tensordict-compile]	0.1488ms	60.3154μs	16.5795 KOps/s	16.2463 KOps/s	$\color{#35bf28}+2.05\%$
test_compile_add_one_nested[tensordict-eager]	0.3437ms	0.1608ms	6.2204 KOps/s	6.1644 KOps/s	$\color{#35bf28}+0.91\%$
test_compile_add_one_nested[pytree-compile]	0.1038ms	45.4118μs	22.0207 KOps/s	21.9619 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_add_one_nested[pytree-eager]	0.2846ms	0.1198ms	8.3480 KOps/s	8.3505 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_copy_nested[tensordict-compile]	76.4330μs	25.7791μs	38.7911 KOps/s	38.8046 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_copy_nested[tensordict-eager]	0.1067ms	54.3266μs	18.4072 KOps/s	18.5866 KOps/s	$\color{#d91a1a}-0.97\%$
test_compile_copy_nested[pytree-compile]	0.1502ms	80.1200μs	12.4813 KOps/s	12.7033 KOps/s	$\color{#d91a1a}-1.75\%$
test_compile_copy_nested[pytree-eager]	0.1525ms	68.8711μs	14.5199 KOps/s	14.7802 KOps/s	$\color{#d91a1a}-1.76\%$
test_compile_add_one_flat[tensordict-compile]	0.2223ms	0.1031ms	9.6951 KOps/s	9.6427 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_add_one_flat[tensordict-eager]	0.4408ms	0.2015ms	4.9627 KOps/s	4.9154 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_add_one_flat[tensorclass-compile]	0.1059ms	43.3358μs	23.0756 KOps/s	22.8595 KOps/s	$\color{#35bf28}+0.95\%$
test_compile_add_one_flat[tensorclass-eager]	0.4923ms	61.2137μs	16.3362 KOps/s	16.1999 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_add_one_flat[pytree-compile]	0.1816ms	0.1020ms	9.8039 KOps/s	9.7949 KOps/s	$\color{#35bf28}+0.09\%$
test_compile_add_one_flat[pytree-eager]	0.3625ms	0.1990ms	5.0242 KOps/s	4.9687 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_add_self_flat[tensordict-eager]	0.3123ms	0.2142ms	4.6679 KOps/s	4.6525 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_add_self_flat[tensordict-compile]	0.2436ms	0.1059ms	9.4415 KOps/s	9.6054 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_add_self_flat[tensorclass-eager]	0.2060ms	55.3986μs	18.0510 KOps/s	18.4059 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_add_self_flat[tensorclass-compile]	0.1066ms	45.1777μs	22.1348 KOps/s	21.5140 KOps/s	$\color{#35bf28}+2.89\%$
test_compile_add_self_flat[pytree-eager]	0.6209ms	0.1592ms	6.2827 KOps/s	6.2794 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_add_self_flat[pytree-compile]	0.1830ms	0.1020ms	9.8055 KOps/s	9.7191 KOps/s	$\color{#35bf28}+0.89\%$
test_compile_copy_flat[tensordict-compile]	60.0030μs	20.9362μs	47.7642 KOps/s	47.1623 KOps/s	$\color{#35bf28}+1.28\%$
test_compile_copy_flat[tensordict-eager]	0.1124ms	59.0581μs	16.9325 KOps/s	16.9396 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_copy_flat[pytree-compile]	0.1418ms	81.7655μs	12.2301 KOps/s	12.3292 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_copy_flat[pytree-eager]	0.1323ms	70.2385μs	14.2372 KOps/s	14.4709 KOps/s	$\color{#d91a1a}-1.62\%$
test_compile_assign_and_add[tensordict-compile]	0.2995ms	0.2055ms	4.8670 KOps/s	4.8424 KOps/s	$\color{#35bf28}+0.51\%$
test_compile_assign_and_add[tensordict-eager]	1.5391ms	1.2763ms	783.4846 Ops/s	765.9770 Ops/s	$\color{#35bf28}+2.29\%$
test_compile_assign_and_add[pytree-compile]	0.2857ms	0.1979ms	5.0523 KOps/s	4.9630 KOps/s	$\color{#35bf28}+1.80\%$
test_compile_assign_and_add[pytree-eager]	0.8771ms	0.7712ms	1.2966 KOps/s	1.2817 KOps/s	$\color{#35bf28}+1.16\%$
test_compile_assign_and_add_stack[compile]	0.6745ms	0.4525ms	2.2100 KOps/s	2.2220 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_assign_and_add_stack[eager]	2.6370ms	2.4929ms	401.1342 Ops/s	357.5765 Ops/s	$\textbf{\color{#35bf28}+12.18\%}$
test_compile_indexing[tensor-tensordict-compile]	98.8250μs	35.0684μs	28.5157 KOps/s	28.4918 KOps/s	$\color{#35bf28}+0.08\%$
test_compile_indexing[tensor-tensordict-eager]	0.7655ms	32.1253μs	31.1282 KOps/s	29.9918 KOps/s	$\color{#35bf28}+3.79\%$
test_compile_indexing[tensor-tensorclass-compile]	70.2610μs	28.5539μs	35.0215 KOps/s	33.9229 KOps/s	$\color{#35bf28}+3.24\%$
test_compile_indexing[tensor-tensorclass-eager]	83.2460μs	23.2697μs	42.9744 KOps/s	42.2841 KOps/s	$\color{#35bf28}+1.63\%$
test_compile_indexing[tensor-pytree-compile]	0.1106ms	29.4577μs	33.9469 KOps/s	32.7955 KOps/s	$\color{#35bf28}+3.51\%$
test_compile_indexing[tensor-pytree-eager]	82.8750μs	23.4250μs	42.6893 KOps/s	42.7229 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_indexing[slice-tensordict-compile]	0.1040ms	50.4709μs	19.8134 KOps/s	19.2385 KOps/s	$\color{#35bf28}+2.99\%$
test_compile_indexing[slice-tensordict-eager]	0.6274ms	19.3151μs	51.7729 KOps/s	48.7658 KOps/s	$\textbf{\color{#35bf28}+6.17\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1214ms	43.1482μs	23.1760 KOps/s	22.5123 KOps/s	$\color{#35bf28}+2.95\%$
test_compile_indexing[slice-tensorclass-eager]	76.7540μs	19.0447μs	52.5081 KOps/s	52.6217 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_indexing[slice-pytree-compile]	0.1123ms	44.5945μs	22.4243 KOps/s	21.9772 KOps/s	$\color{#35bf28}+2.03\%$
test_compile_indexing[slice-pytree-eager]	75.2910μs	18.9955μs	52.6441 KOps/s	53.0706 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_indexing[int-tensordict-compile]	0.1166ms	51.6805μs	19.3496 KOps/s	18.9878 KOps/s	$\color{#35bf28}+1.91\%$
test_compile_indexing[int-tensordict-eager]	0.9319ms	19.3456μs	51.6912 KOps/s	48.3662 KOps/s	$\textbf{\color{#35bf28}+6.87\%}$
test_compile_indexing[int-tensorclass-compile]	96.3400μs	44.2706μs	22.5884 KOps/s	21.8048 KOps/s	$\color{#35bf28}+3.59\%$
test_compile_indexing[int-tensorclass-eager]	74.5190μs	18.9616μs	52.7381 KOps/s	52.7342 KOps/s	$+0.01\%$
test_compile_indexing[int-pytree-compile]	0.1039ms	44.1085μs	22.6714 KOps/s	21.9540 KOps/s	$\color{#35bf28}+3.27\%$
test_compile_indexing[int-pytree-eager]	60.8740μs	19.0655μs	52.4508 KOps/s	52.7345 KOps/s	$\color{#d91a1a}-0.54\%$
test_mod_add[eager]	66.5150μs	24.5780μs	40.6869 KOps/s	37.7542 KOps/s	$\textbf{\color{#35bf28}+7.77\%}$
test_mod_add[compile]	0.1111ms	44.6790μs	22.3819 KOps/s	22.4632 KOps/s	$\color{#d91a1a}-0.36\%$
test_mod_add[compile-overhead]	0.1091ms	44.1369μs	22.6568 KOps/s	22.3884 KOps/s	$\color{#35bf28}+1.20\%$
test_mod_wrap[eager]	0.4351ms	0.2061ms	4.8518 KOps/s	4.6228 KOps/s	$\color{#35bf28}+4.96\%$
test_mod_wrap[compile]	1.4719ms	0.1977ms	5.0580 KOps/s	4.8875 KOps/s	$\color{#35bf28}+3.49\%$
test_mod_wrap[compile-overhead]	1.9208ms	0.1965ms	5.0880 KOps/s	4.9756 KOps/s	$\color{#35bf28}+2.26\%$
test_mod_wrap_and_backward[eager]	14.3906ms	11.6680ms	85.7047 Ops/s	87.2006 Ops/s	$\color{#d91a1a}-1.72\%$
test_mod_wrap_and_backward[compile]	16.2445ms	12.0987ms	82.6537 Ops/s	82.0028 Ops/s	$\color{#35bf28}+0.79\%$
test_mod_wrap_and_backward[compile-overhead]	15.3261ms	12.6963ms	78.7633 Ops/s	79.0572 Ops/s	$\color{#d91a1a}-0.37\%$
test_seq_add[eager]	0.2022ms	89.0669μs	11.2275 KOps/s	10.6461 KOps/s	$\textbf{\color{#35bf28}+5.46\%}$
test_seq_add[compile]	0.1407ms	58.9894μs	16.9522 KOps/s	16.6069 KOps/s	$\color{#35bf28}+2.08\%$
test_seq_add[compile-overhead]	0.1373ms	56.9145μs	17.5702 KOps/s	16.7242 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_seq_wrap[eager]	0.6011ms	0.3692ms	2.7082 KOps/s	2.5090 KOps/s	$\textbf{\color{#35bf28}+7.94\%}$
test_seq_wrap[compile]	0.3537ms	0.2221ms	4.5034 KOps/s	4.3734 KOps/s	$\color{#35bf28}+2.97\%$
test_seq_wrap[compile-overhead]	0.4214ms	0.2216ms	4.5134 KOps/s	4.3874 KOps/s	$\color{#35bf28}+2.87\%$
test_func_call_runtime[False-eager]	1.0573ms	0.5326ms	1.8775 KOps/s	1.8403 KOps/s	$\color{#35bf28}+2.02\%$
test_func_call_runtime[False-compile]	0.5562ms	0.4210ms	2.3751 KOps/s	2.3353 KOps/s	$\color{#35bf28}+1.70\%$
test_func_call_runtime[False-compile-overhead]	0.7945ms	0.4239ms	2.3593 KOps/s	2.3221 KOps/s	$\color{#35bf28}+1.60\%$
test_func_call_runtime[True-eager]	0.9901ms	0.7455ms	1.3413 KOps/s	1.3060 KOps/s	$\color{#35bf28}+2.70\%$
test_func_call_runtime[True-compile]	0.6132ms	0.4602ms	2.1731 KOps/s	2.1227 KOps/s	$\color{#35bf28}+2.37\%$
test_func_call_runtime[True-compile-overhead]	0.7222ms	0.4650ms	2.1507 KOps/s	2.1196 KOps/s	$\color{#35bf28}+1.47\%$
test_func_call_cm_runtime[False-eager]	0.9274ms	0.5335ms	1.8745 KOps/s	1.8067 KOps/s	$\color{#35bf28}+3.75\%$
test_func_call_cm_runtime[False-compile]	0.6555ms	0.4214ms	2.3728 KOps/s	2.3326 KOps/s	$\color{#35bf28}+1.72\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5273ms	0.4196ms	2.3831 KOps/s	2.3499 KOps/s	$\color{#35bf28}+1.41\%$
test_func_call_cm_runtime[True-eager]	1.1634ms	0.8678ms	1.1524 KOps/s	1.1073 KOps/s	$\color{#35bf28}+4.07\%$
test_func_call_cm_runtime[True-compile]	0.6533ms	0.4824ms	2.0730 KOps/s	2.0277 KOps/s	$\color{#35bf28}+2.23\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6177ms	0.4834ms	2.0687 KOps/s	2.0400 KOps/s	$\color{#35bf28}+1.41\%$
test_vmap_func_call_cm_runtime[eager]	3.0141ms	1.8442ms	542.2359 Ops/s	526.4900 Ops/s	$\color{#35bf28}+2.99\%$
test_vmap_func_call_cm_runtime[compile]	0.7079ms	0.5142ms	1.9450 KOps/s	1.9206 KOps/s	$\color{#35bf28}+1.27\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6070ms	0.5124ms	1.9514 KOps/s	1.9218 KOps/s	$\color{#35bf28}+1.54\%$
test_distributed	0.2861ms	0.1246ms	8.0267 KOps/s	7.7851 KOps/s	$\color{#35bf28}+3.10\%$
test_tdmodule	35.5760μs	17.3963μs	57.4834 KOps/s	48.4014 KOps/s	$\textbf{\color{#35bf28}+18.76\%}$
test_tdmodule_dispatch	72.6450μs	37.9839μs	26.3270 KOps/s	23.0477 KOps/s	$\textbf{\color{#35bf28}+14.23\%}$
test_tdseq	36.9790μs	19.9784μs	50.0540 KOps/s	44.3422 KOps/s	$\textbf{\color{#35bf28}+12.88\%}$
test_tdseq_dispatch	72.5460μs	42.9136μs	23.3026 KOps/s	20.7969 KOps/s	$\textbf{\color{#35bf28}+12.05\%}$
test_instantiation_functorch	1.9945ms	1.5362ms	650.9698 Ops/s	647.8214 Ops/s	$\color{#35bf28}+0.49\%$
test_exec_functorch	0.3296ms	0.1731ms	5.7772 KOps/s	5.4734 KOps/s	$\textbf{\color{#35bf28}+5.55\%}$
test_exec_functional_call	0.2857ms	0.1699ms	5.8850 KOps/s	5.6760 KOps/s	$\color{#35bf28}+3.68\%$
test_exec_td_decorator	0.4664ms	0.2253ms	4.4394 KOps/s	4.3215 KOps/s	$\color{#35bf28}+2.73\%$
test_vmap_mlp_speed_decorator[True-True]	0.8589ms	0.6114ms	1.6355 KOps/s	1.5654 KOps/s	$\color{#35bf28}+4.48\%$
test_vmap_mlp_speed_decorator[True-False]	0.9989ms	0.6141ms	1.6283 KOps/s	1.5644 KOps/s	$\color{#35bf28}+4.08\%$
test_vmap_mlp_speed_decorator[False-True]	1.6090ms	0.5068ms	1.9731 KOps/s	1.8679 KOps/s	$\textbf{\color{#35bf28}+5.63\%}$
test_vmap_mlp_speed_decorator[False-False]	0.7241ms	0.5041ms	1.9836 KOps/s	1.9173 KOps/s	$\color{#35bf28}+3.46\%$
test_to_module_speed[True]	2.0459ms	1.3000ms	769.2147 Ops/s	782.2826 Ops/s	$\color{#d91a1a}-1.67\%$
test_to_module_speed[False]	1.9058ms	1.2648ms	790.6157 Ops/s	802.6015 Ops/s	$\color{#d91a1a}-1.49\%$
test_tc_init	75.0700μs	42.5555μs	23.4987 KOps/s	21.3937 KOps/s	$\textbf{\color{#35bf28}+9.84\%}$
test_tc_init_nested	0.1505ms	85.6742μs	11.6721 KOps/s	10.6271 KOps/s	$\textbf{\color{#35bf28}+9.83\%}$
test_tc_first_layer_tensor	15.4590μs	1.5902μs	628.8522 KOps/s	666.3685 KOps/s	$\textbf{\color{#d91a1a}-5.63\%}$
test_tc_first_layer_nontensor	50.9850μs	4.9803μs	200.7912 KOps/s	211.8039 KOps/s	$\textbf{\color{#d91a1a}-5.20\%}$
test_tc_second_layer_tensor	41.8100μs	2.8266μs	353.7782 KOps/s	353.9634 KOps/s	$\color{#d91a1a}-0.05\%$
test_tc_second_layer_nontensor	25.6370μs	6.3053μs	158.5964 KOps/s	164.0535 KOps/s	$\color{#d91a1a}-3.33\%$
test_unbind	0.2238s	12.0792ms	82.7870 Ops/s	77.0111 Ops/s	$\textbf{\color{#35bf28}+7.50\%}$
test_full_like	7.9579ms	6.8527ms	145.9270 Ops/s	84.9757 Ops/s	$\textbf{\color{#35bf28}+71.73\%}$
test_zeros_like	3.0932ms	2.6493ms	377.4578 Ops/s	137.7689 Ops/s	$\textbf{\color{#35bf28}+173.98\%}$
test_ones_like	3.4083ms	3.1090ms	321.6421 Ops/s	136.0185 Ops/s	$\textbf{\color{#35bf28}+136.47\%}$
test_clone	5.6757ms	4.7365ms	211.1250 Ops/s	113.1374 Ops/s	$\textbf{\color{#35bf28}+86.61\%}$
test_squeeze	55.6340μs	11.6587μs	85.7729 KOps/s	84.0014 KOps/s	$\color{#35bf28}+2.11\%$
test_unsqueeze	0.1684ms	87.6482μs	11.4092 KOps/s	11.1576 KOps/s	$\color{#35bf28}+2.26\%$
test_split	0.5071ms	0.1896ms	5.2733 KOps/s	5.4065 KOps/s	$\color{#d91a1a}-2.46\%$
test_permute	0.3114ms	0.2149ms	4.6526 KOps/s	4.5056 KOps/s	$\color{#35bf28}+3.26\%$
test_stack	27.9885ms	23.8889ms	41.8604 Ops/s	42.2108 Ops/s	$\color{#d91a1a}-0.83\%$
test_cat	30.5291ms	23.8859ms	41.8656 Ops/s	42.4604 Ops/s	$\color{#d91a1a}-1.40\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	24.9000μs	10.2185μs	97.8621 KOps/s	99.8951 KOps/s	$\color{#d91a1a}-2.04\%$
test_plain_set_stack_nested	30.2710μs	10.2269μs	97.7814 KOps/s	98.8616 KOps/s	$\color{#d91a1a}-1.09\%$
test_plain_set_nested_inplace	0.3949ms	11.0741μs	90.3004 KOps/s	92.2874 KOps/s	$\color{#d91a1a}-2.15\%$
test_plain_set_stack_nested_inplace	41.7700μs	11.0442μs	90.5452 KOps/s	92.1674 KOps/s	$\color{#d91a1a}-1.76\%$
test_items	74.2810μs	2.8383μs	352.3218 KOps/s	345.3352 KOps/s	$\color{#35bf28}+2.02\%$
test_items_nested	0.7101ms	0.3291ms	3.0383 KOps/s	3.0755 KOps/s	$\color{#d91a1a}-1.21\%$
test_items_nested_locked	0.7062ms	0.3301ms	3.0290 KOps/s	3.0583 KOps/s	$\color{#d91a1a}-0.96\%$
test_items_nested_leaf	93.3420μs	58.7198μs	17.0300 KOps/s	17.1854 KOps/s	$\color{#d91a1a}-0.90\%$
test_items_stack_nested	0.7351ms	0.3309ms	3.0221 KOps/s	3.0474 KOps/s	$\color{#d91a1a}-0.83\%$
test_items_stack_nested_leaf	0.4474ms	59.0903μs	16.9233 KOps/s	16.7987 KOps/s	$\color{#35bf28}+0.74\%$
test_items_stack_nested_locked	0.7140ms	0.3323ms	3.0093 KOps/s	3.0345 KOps/s	$\color{#d91a1a}-0.83\%$
test_keys	0.3865ms	3.4952μs	286.1068 KOps/s	290.6916 KOps/s	$\color{#d91a1a}-1.58\%$
test_keys_nested	0.4502ms	70.8941μs	14.1055 KOps/s	14.1286 KOps/s	$\color{#d91a1a}-0.16\%$
test_keys_nested_locked	0.7464ms	75.4628μs	13.2516 KOps/s	13.0399 KOps/s	$\color{#35bf28}+1.62\%$
test_keys_nested_leaf	0.4366ms	61.8838μs	16.1593 KOps/s	16.1348 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_stack_nested	0.4407ms	71.0290μs	14.0787 KOps/s	13.9995 KOps/s	$\color{#35bf28}+0.57\%$
test_keys_stack_nested_leaf	0.4364ms	63.2446μs	15.8116 KOps/s	15.8739 KOps/s	$\color{#d91a1a}-0.39\%$
test_keys_stack_nested_locked	0.4542ms	77.0413μs	12.9800 KOps/s	13.0067 KOps/s	$\color{#d91a1a}-0.20\%$
test_values	5.9233μs	0.8652μs	1.1558 MOps/s	1.1845 MOps/s	$\color{#d91a1a}-2.42\%$
test_values_nested	0.4088ms	31.2737μs	31.9757 KOps/s	32.2357 KOps/s	$\color{#d91a1a}-0.81\%$
test_values_nested_locked	0.4110ms	32.9815μs	30.3200 KOps/s	30.6111 KOps/s	$\color{#d91a1a}-0.95\%$
test_values_nested_leaf	55.5510μs	33.6748μs	29.6958 KOps/s	29.8338 KOps/s	$\color{#d91a1a}-0.46\%$
test_values_stack_nested	0.4250ms	31.7777μs	31.4686 KOps/s	31.5805 KOps/s	$\color{#d91a1a}-0.35\%$
test_values_stack_nested_leaf	0.4288ms	34.6167μs	28.8877 KOps/s	29.1986 KOps/s	$\color{#d91a1a}-1.06\%$
test_values_stack_nested_locked	0.4156ms	33.5914μs	29.7695 KOps/s	30.0355 KOps/s	$\color{#d91a1a}-0.89\%$
test_membership	19.3129μs	0.5079μs	1.9690 MOps/s	1.9688 MOps/s	$\color{#35bf28}+0.01\%$
test_membership_nested	25.5610μs	1.9743μs	506.5162 KOps/s	506.6315 KOps/s	$\color{#d91a1a}-0.02\%$
test_membership_nested_leaf	0.2027ms	1.9306μs	517.9686 KOps/s	513.8766 KOps/s	$\color{#35bf28}+0.80\%$
test_membership_stacked_nested	27.8600μs	1.9978μs	500.5627 KOps/s	499.4050 KOps/s	$\color{#35bf28}+0.23\%$
test_membership_stacked_nested_leaf	0.3954ms	2.0047μs	498.8250 KOps/s	499.4223 KOps/s	$\color{#d91a1a}-0.12\%$
test_membership_nested_last	47.4910μs	2.8470μs	351.2430 KOps/s	347.5757 KOps/s	$\color{#35bf28}+1.06\%$
test_membership_nested_leaf_last	0.4332ms	2.8682μs	348.6485 KOps/s	350.0377 KOps/s	$\color{#d91a1a}-0.40\%$
test_membership_stacked_nested_last	28.4910μs	3.5084μs	285.0296 KOps/s	231.2359 KOps/s	$\textbf{\color{#35bf28}+23.26\%}$
test_membership_stacked_nested_leaf_last	33.8200μs	3.5051μs	285.3016 KOps/s	231.3885 KOps/s	$\textbf{\color{#35bf28}+23.30\%}$
test_nested_getleaf	0.3903ms	6.0361μs	165.6698 KOps/s	166.3404 KOps/s	$\color{#d91a1a}-0.40\%$
test_nested_get	0.3859ms	5.7258μs	174.6468 KOps/s	175.4355 KOps/s	$\color{#d91a1a}-0.45\%$
test_stacked_getleaf	39.9210μs	6.0259μs	165.9516 KOps/s	166.7570 KOps/s	$\color{#d91a1a}-0.48\%$
test_stacked_get	0.3928ms	5.7051μs	175.2814 KOps/s	175.3185 KOps/s	$\color{#d91a1a}-0.02\%$
test_nested_getitemleaf	26.7210μs	6.1133μs	163.5782 KOps/s	164.8903 KOps/s	$\color{#d91a1a}-0.80\%$
test_nested_getitem	25.8510μs	5.8038μs	172.3009 KOps/s	173.2926 KOps/s	$\color{#d91a1a}-0.57\%$
test_stacked_getitemleaf	38.8110μs	6.0740μs	164.6362 KOps/s	166.0280 KOps/s	$\color{#d91a1a}-0.84\%$
test_stacked_getitem	0.3895ms	5.8021μs	172.3506 KOps/s	173.9821 KOps/s	$\color{#d91a1a}-0.94\%$
test_lock_nested	9.3928ms	0.3751ms	2.6660 KOps/s	2.6592 KOps/s	$\color{#35bf28}+0.26\%$
test_lock_stack_nested	0.3973ms	0.3373ms	2.9651 KOps/s	2.9400 KOps/s	$\color{#35bf28}+0.85\%$
test_unlock_nested	0.6448ms	0.3085ms	3.2413 KOps/s	3.2281 KOps/s	$\color{#35bf28}+0.41\%$
test_unlock_stack_nested	0.3290ms	0.2737ms	3.6542 KOps/s	3.6251 KOps/s	$\color{#35bf28}+0.80\%$
test_flatten_speed	0.4640ms	72.3100μs	13.8294 KOps/s	13.5877 KOps/s	$\color{#35bf28}+1.78\%$
test_unflatten_speed	0.6762ms	0.2955ms	3.3844 KOps/s	3.3509 KOps/s	$\color{#35bf28}+1.00\%$
test_common_ops	1.6761ms	0.5613ms	1.7817 KOps/s	1.8160 KOps/s	$\color{#d91a1a}-1.89\%$
test_creation	33.8600μs	1.4763μs	677.3602 KOps/s	682.5516 KOps/s	$\color{#d91a1a}-0.76\%$
test_creation_empty	40.1710μs	6.5879μs	151.7929 KOps/s	160.0070 KOps/s	$\textbf{\color{#d91a1a}-5.13\%}$
test_creation_nested_1	0.3963ms	8.0421μs	124.3451 KOps/s	129.3631 KOps/s	$\color{#d91a1a}-3.88\%$
test_creation_nested_2	84.1310μs	10.5816μs	94.5034 KOps/s	96.9561 KOps/s	$\color{#d91a1a}-2.53\%$
test_clone	88.3320μs	9.9505μs	100.4978 KOps/s	99.6382 KOps/s	$\color{#35bf28}+0.86\%$
test_getitem[int]	1.5197ms	10.8732μs	91.9694 KOps/s	91.4444 KOps/s	$\color{#35bf28}+0.57\%$
test_getitem[slice_int]	0.1113ms	20.6523μs	48.4207 KOps/s	47.9618 KOps/s	$\color{#35bf28}+0.96\%$
test_getitem[range]	0.1236ms	36.5872μs	27.3319 KOps/s	26.7619 KOps/s	$\color{#35bf28}+2.13\%$
test_getitem[tuple]	0.1041ms	17.9482μs	55.7160 KOps/s	54.5644 KOps/s	$\color{#35bf28}+2.11\%$
test_getitem[list]	0.4324ms	31.7688μs	31.4774 KOps/s	31.2033 KOps/s	$\color{#35bf28}+0.88\%$
test_setitem_dim[int]	37.4000μs	17.6347μs	56.7064 KOps/s	55.2355 KOps/s	$\color{#35bf28}+2.66\%$
test_setitem_dim[slice_int]	62.6020μs	35.5216μs	28.1519 KOps/s	27.6235 KOps/s	$\color{#35bf28}+1.91\%$
test_setitem_dim[range]	80.1410μs	50.8185μs	19.6779 KOps/s	19.0809 KOps/s	$\color{#35bf28}+3.13\%$
test_setitem_dim[tuple]	61.8810μs	30.6026μs	32.6770 KOps/s	31.9785 KOps/s	$\color{#35bf28}+2.18\%$
test_setitem	0.4071ms	13.6807μs	73.0958 KOps/s	74.1500 KOps/s	$\color{#d91a1a}-1.42\%$
test_set	93.3120μs	13.0004μs	76.9209 KOps/s	77.5412 KOps/s	$\color{#d91a1a}-0.80\%$
test_set_shared	1.6054ms	0.1444ms	6.9275 KOps/s	6.9053 KOps/s	$\color{#35bf28}+0.32\%$
test_update	1.0541ms	15.2045μs	65.7699 KOps/s	66.7690 KOps/s	$\color{#d91a1a}-1.50\%$
test_update_nested	0.4230ms	19.8275μs	50.4350 KOps/s	51.9448 KOps/s	$\color{#d91a1a}-2.91\%$
test_update__nested	0.1331ms	23.2068μs	43.0909 KOps/s	42.6409 KOps/s	$\color{#35bf28}+1.06\%$
test_set_nested	81.3820μs	13.5533μs	73.7830 KOps/s	71.0719 KOps/s	$\color{#35bf28}+3.81\%$
test_set_nested_new	0.4121ms	16.4887μs	60.6474 KOps/s	61.7090 KOps/s	$\color{#d91a1a}-1.72\%$
test_select	57.8520μs	28.0837μs	35.6079 KOps/s	35.6813 KOps/s	$\color{#d91a1a}-0.21\%$
test_select_nested	0.4287ms	41.8887μs	23.8728 KOps/s	24.7229 KOps/s	$\color{#d91a1a}-3.44\%$
test_exclude_nested	0.4542ms	59.9718μs	16.6745 KOps/s	16.8744 KOps/s	$\color{#d91a1a}-1.18\%$
test_empty[True]	0.6505ms	0.2613ms	3.8263 KOps/s	3.8585 KOps/s	$\color{#d91a1a}-0.83\%$
test_empty[False]	39.3118μs	0.7483μs	1.3364 MOps/s	1.3495 MOps/s	$\color{#d91a1a}-0.97\%$
test_to	88.5910μs	54.7869μs	18.2526 KOps/s	17.5277 KOps/s	$\color{#35bf28}+4.14\%$
test_to_nonblocking	90.2320μs	45.3883μs	22.0321 KOps/s	21.4768 KOps/s	$\color{#35bf28}+2.59\%$
test_unbind_speed	1.7809ms	0.2335ms	4.2830 KOps/s	4.1263 KOps/s	$\color{#35bf28}+3.80\%$
test_unbind_speed_stack0	0.6360ms	0.2327ms	4.2966 KOps/s	4.1527 KOps/s	$\color{#35bf28}+3.47\%$
test_unbind_speed_stack1	94.7583ms	0.6458ms	1.5486 KOps/s	1.5432 KOps/s	$\color{#35bf28}+0.35\%$
test_split	95.3389ms	1.5837ms	631.4239 Ops/s	567.8458 Ops/s	$\textbf{\color{#35bf28}+11.20\%}$
test_chunk	95.4440ms	1.5776ms	633.8732 Ops/s	674.3434 Ops/s	$\textbf{\color{#d91a1a}-6.00\%}$
test_consolidate[False-None]	98.1111ms	2.8219ms	354.3666 Ops/s	350.7518 Ops/s	$\color{#35bf28}+1.03\%$
test_consolidate[default-None]	1.7834ms	1.6764ms	596.5155 Ops/s	586.9378 Ops/s	$\color{#35bf28}+1.63\%$
test_consolidate[reduce-overhead-None]	1.8348ms	1.7245ms	579.8783 Ops/s	573.8165 Ops/s	$\color{#35bf28}+1.06\%$
test_consolidate_njt[False-None]	6.8084ms	6.5420ms	152.8580 Ops/s	155.5383 Ops/s	$\color{#d91a1a}-1.72\%$
test_to[False-False-None]	1.7566ms	1.6588ms	602.8392 Ops/s	590.8750 Ops/s	$\color{#35bf28}+2.02\%$
test_to[True-False-None]	1.5925ms	1.3305ms	751.5818 Ops/s	747.1782 Ops/s	$\color{#35bf28}+0.59\%$
test_to[within-False-None]	4.2605ms	3.9954ms	250.2888 Ops/s	247.0419 Ops/s	$\color{#35bf28}+1.31\%$
test_to[True-default-None]	5.6670ms	5.3286ms	187.6666 Ops/s	187.6875 Ops/s	$\color{#d91a1a}-0.01\%$
test_to_njt[False-False-None]	7.6602ms	7.0580ms	141.6840 Ops/s	140.1569 Ops/s	$\color{#35bf28}+1.09\%$
test_to_njt[True-False-None]	5.8904ms	5.5113ms	181.4465 Ops/s	180.4268 Ops/s	$\color{#35bf28}+0.57\%$
test_to_njt[within-False-None]	12.9582ms	12.2383ms	81.7108 Ops/s	81.9406 Ops/s	$\color{#d91a1a}-0.28\%$
test_creation[device0]	0.4617ms	78.0845μs	12.8066 KOps/s	12.7442 KOps/s	$\color{#35bf28}+0.49\%$
test_creation_from_tensor	0.5667ms	81.7621μs	12.2306 KOps/s	11.8324 KOps/s	$\color{#35bf28}+3.37\%$
test_add_one[memmap_tensor0]	0.4694ms	6.3447μs	157.6116 KOps/s	153.1751 KOps/s	$\color{#35bf28}+2.90\%$
test_contiguous[memmap_tensor0]	1.9665μs	0.4157μs	2.4053 MOps/s	2.3486 MOps/s	$\color{#35bf28}+2.42\%$
test_stack[memmap_tensor0]	35.5110μs	4.5852μs	218.0914 KOps/s	215.5248 KOps/s	$\color{#35bf28}+1.19\%$
test_memmaptd_index	1.7844ms	0.2482ms	4.0297 KOps/s	3.9250 KOps/s	$\color{#35bf28}+2.67\%$
test_memmaptd_index_astensor	0.8853ms	0.3055ms	3.2728 KOps/s	3.2502 KOps/s	$\color{#35bf28}+0.69\%$
test_memmaptd_index_op	0.9588ms	0.5472ms	1.8273 KOps/s	1.8170 KOps/s	$\color{#35bf28}+0.57\%$
test_serialize_model	0.1322s	0.1306s	7.6543 Ops/s	7.6396 Ops/s	$\color{#35bf28}+0.19\%$
test_serialize_model_pickle	1.3797s	1.1991s	0.8340 Ops/s	0.8445 Ops/s	$\color{#d91a1a}-1.24\%$
test_serialize_weights	0.1306s	0.1301s	7.6836 Ops/s	7.6723 Ops/s	$\color{#35bf28}+0.15\%$
test_serialize_weights_returnearly	0.3228s	53.8958ms	18.5543 Ops/s	23.5251 Ops/s	$\textbf{\color{#d91a1a}-21.13\%}$
test_serialize_weights_pickle	1.3682s	1.1932s	0.8381 Ops/s	0.8230 Ops/s	$\color{#35bf28}+1.83\%$
test_reshape_pytree	62.9410μs	22.3376μs	44.7675 KOps/s	44.1189 KOps/s	$\color{#35bf28}+1.47\%$
test_reshape_td	62.5520μs	26.8642μs	37.2242 KOps/s	36.9726 KOps/s	$\color{#35bf28}+0.68\%$
test_view_pytree	53.6410μs	22.1568μs	45.1329 KOps/s	44.4421 KOps/s	$\color{#35bf28}+1.55\%$
test_view_td	89.3220μs	29.4015μs	34.0118 KOps/s	33.0333 KOps/s	$\color{#35bf28}+2.96\%$
test_unbind_pytree	60.6410μs	27.7432μs	36.0448 KOps/s	34.9338 KOps/s	$\color{#35bf28}+3.18\%$
test_unbind_td	0.5577ms	35.2627μs	28.3586 KOps/s	27.5407 KOps/s	$\color{#35bf28}+2.97\%$
test_split_pytree	76.6020μs	30.4570μs	32.8332 KOps/s	32.6477 KOps/s	$\color{#35bf28}+0.57\%$
test_split_td	0.5569ms	38.9367μs	25.6827 KOps/s	24.9543 KOps/s	$\color{#35bf28}+2.92\%$
test_add_pytree	73.8620μs	33.1152μs	30.1976 KOps/s	30.4476 KOps/s	$\color{#d91a1a}-0.82\%$
test_add_td	0.1055ms	44.4141μs	22.5154 KOps/s	22.3352 KOps/s	$\color{#35bf28}+0.81\%$
test_compile_add_one_nested[tensordict-compile]	0.1737ms	0.1197ms	8.3576 KOps/s	8.0476 KOps/s	$\color{#35bf28}+3.85\%$
test_compile_add_one_nested[tensordict-eager]	0.2214ms	0.1261ms	7.9293 KOps/s	7.9945 KOps/s	$\color{#d91a1a}-0.82\%$
test_compile_add_one_nested[pytree-compile]	0.1444ms	97.3848μs	10.2685 KOps/s	9.8976 KOps/s	$\color{#35bf28}+3.75\%$
test_compile_add_one_nested[pytree-eager]	1.1704ms	0.1491ms	6.7088 KOps/s	6.7250 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_copy_nested[tensordict-compile]	0.1797ms	23.6160μs	42.3442 KOps/s	42.6471 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_copy_nested[tensordict-eager]	54.9420μs	27.3439μs	36.5712 KOps/s	36.9298 KOps/s	$\color{#d91a1a}-0.97\%$
test_compile_copy_nested[pytree-compile]	0.4830ms	64.6656μs	15.4642 KOps/s	15.2640 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_copy_nested[pytree-eager]	84.5920μs	49.6364μs	20.1465 KOps/s	19.7163 KOps/s	$\color{#35bf28}+2.18\%$
test_compile_add_one_flat[tensordict-compile]	0.1832ms	0.1434ms	6.9712 KOps/s	6.9072 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_add_one_flat[tensordict-eager]	0.2997ms	0.2080ms	4.8072 KOps/s	4.8244 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_add_one_flat[tensorclass-compile]	0.1398ms	99.6294μs	10.0372 KOps/s	10.1922 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_add_one_flat[tensorclass-eager]	0.1096ms	51.0645μs	19.5831 KOps/s	18.5727 KOps/s	$\textbf{\color{#35bf28}+5.44\%}$
test_compile_add_one_flat[pytree-compile]	0.1838ms	0.1385ms	7.2180 KOps/s	7.2348 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_add_one_flat[pytree-eager]	0.5811ms	0.4773ms	2.0953 KOps/s	2.1022 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_add_self_flat[tensordict-eager]	0.3753ms	0.2483ms	4.0273 KOps/s	3.9892 KOps/s	$\color{#35bf28}+0.95\%$
test_compile_add_self_flat[tensordict-compile]	0.1828ms	0.1451ms	6.8935 KOps/s	6.8504 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_add_self_flat[tensorclass-eager]	0.1425ms	60.8580μs	16.4317 KOps/s	15.8667 KOps/s	$\color{#35bf28}+3.56\%$
test_compile_add_self_flat[tensorclass-compile]	0.1573ms	0.1061ms	9.4241 KOps/s	9.7680 KOps/s	$\color{#d91a1a}-3.52\%$
test_compile_add_self_flat[pytree-eager]	0.4857ms	0.4078ms	2.4524 KOps/s	2.4941 KOps/s	$\color{#d91a1a}-1.67\%$
test_compile_add_self_flat[pytree-compile]	0.1902ms	0.1420ms	7.0409 KOps/s	7.3062 KOps/s	$\color{#d91a1a}-3.63\%$
test_compile_copy_flat[tensordict-compile]	55.2710μs	19.1236μs	52.2913 KOps/s	54.9694 KOps/s	$\color{#d91a1a}-4.87\%$
test_compile_copy_flat[tensordict-eager]	60.8110μs	27.6706μs	36.1394 KOps/s	36.8125 KOps/s	$\color{#d91a1a}-1.83\%$
test_compile_copy_flat[pytree-compile]	0.1144ms	70.9462μs	14.0952 KOps/s	14.2260 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_copy_flat[pytree-eager]	88.9420μs	51.8942μs	19.2700 KOps/s	19.4381 KOps/s	$\color{#d91a1a}-0.87\%$
test_compile_assign_and_add[tensordict-compile]	1.6491ms	0.3962ms	2.5238 KOps/s	2.1943 KOps/s	$\textbf{\color{#35bf28}+15.02\%}$
test_compile_assign_and_add[tensordict-eager]	2.6443ms	2.5523ms	391.8054 Ops/s	381.4130 Ops/s	$\color{#35bf28}+2.72\%$
test_compile_assign_and_add[pytree-compile]	1.6309ms	0.4408ms	2.2686 KOps/s	2.2390 KOps/s	$\color{#35bf28}+1.32\%$
test_compile_assign_and_add[pytree-eager]	2.7205ms	2.5967ms	385.0979 Ops/s	379.8208 Ops/s	$\color{#35bf28}+1.39\%$
test_compile_indexing[tensor-tensordict-compile]	0.6271ms	0.1160ms	8.6214 KOps/s	8.9151 KOps/s	$\color{#d91a1a}-3.29\%$
test_compile_indexing[tensor-tensordict-eager]	0.5476ms	76.5633μs	13.0611 KOps/s	12.7135 KOps/s	$\color{#35bf28}+2.73\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5409ms	0.1035ms	9.6577 KOps/s	9.3460 KOps/s	$\color{#35bf28}+3.34\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1137ms	65.4952μs	15.2683 KOps/s	14.1592 KOps/s	$\textbf{\color{#35bf28}+7.83\%}$
test_compile_indexing[tensor-pytree-compile]	0.1398ms	0.1038ms	9.6377 KOps/s	9.1632 KOps/s	$\textbf{\color{#35bf28}+5.18\%}$
test_compile_indexing[tensor-pytree-eager]	0.1141ms	65.1066μs	15.3594 KOps/s	14.2206 KOps/s	$\textbf{\color{#35bf28}+8.01\%}$
test_compile_indexing[slice-tensordict-compile]	0.1504ms	0.1008ms	9.9208 KOps/s	9.8801 KOps/s	$\color{#35bf28}+0.41\%$
test_compile_indexing[slice-tensordict-eager]	0.1448ms	19.5642μs	51.1137 KOps/s	52.0531 KOps/s	$\color{#d91a1a}-1.80\%$
test_compile_indexing[slice-tensorclass-compile]	0.1391ms	95.7443μs	10.4445 KOps/s	10.0052 KOps/s	$\color{#35bf28}+4.39\%$
test_compile_indexing[slice-tensorclass-eager]	52.2510μs	16.1260μs	62.0115 KOps/s	62.8203 KOps/s	$\color{#d91a1a}-1.29\%$
test_compile_indexing[slice-pytree-compile]	0.1605ms	96.7136μs	10.3398 KOps/s	10.2677 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_indexing[slice-pytree-eager]	44.6910μs	15.8795μs	62.9743 KOps/s	62.7996 KOps/s	$\color{#35bf28}+0.28\%$
test_compile_indexing[int-tensordict-compile]	0.1509ms	0.1023ms	9.7761 KOps/s	9.8521 KOps/s	$\color{#d91a1a}-0.77\%$
test_compile_indexing[int-tensordict-eager]	0.5688ms	17.4325μs	57.3640 KOps/s	55.7023 KOps/s	$\color{#35bf28}+2.98\%$
test_compile_indexing[int-tensorclass-compile]	0.1428ms	96.5957μs	10.3524 KOps/s	10.2783 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_indexing[int-tensorclass-eager]	64.9710μs	15.7013μs	63.6891 KOps/s	62.4153 KOps/s	$\color{#35bf28}+2.04\%$
test_compile_indexing[int-pytree-compile]	0.1418ms	96.8216μs	10.3283 KOps/s	10.3294 KOps/s	$\color{#d91a1a}-0.01\%$
test_compile_indexing[int-pytree-eager]	42.1910μs	15.7666μs	63.4250 KOps/s	61.8278 KOps/s	$\color{#35bf28}+2.58\%$
test_mod_add[eager]	71.5620μs	30.6067μs	32.6726 KOps/s	32.6797 KOps/s	$\color{#d91a1a}-0.02\%$
test_mod_add[compile]	0.3972ms	77.9663μs	12.8261 KOps/s	12.5901 KOps/s	$\color{#35bf28}+1.87\%$
test_mod_add[compile-overhead]	0.3290ms	0.1658ms	6.0327 KOps/s	5.7260 KOps/s	$\textbf{\color{#35bf28}+5.36\%}$
test_mod_wrap[eager]	0.3152ms	0.2346ms	4.2627 KOps/s	3.9397 KOps/s	$\textbf{\color{#35bf28}+8.20\%}$
test_mod_wrap[compile]	1.6401ms	0.2973ms	3.3641 KOps/s	3.4737 KOps/s	$\color{#d91a1a}-3.16\%$
test_mod_wrap[compile-overhead]	7.2001ms	3.8544ms	259.4464 Ops/s	260.1657 Ops/s	$\color{#d91a1a}-0.28\%$
test_mod_wrap_and_backward[eager]	1.4223ms	1.3021ms	767.9900 Ops/s	700.9003 Ops/s	$\textbf{\color{#35bf28}+9.57\%}$
test_mod_wrap_and_backward[compile]	1.3627ms	1.2609ms	793.0529 Ops/s	720.7158 Ops/s	$\textbf{\color{#35bf28}+10.04\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3676ms	0.9268ms	1.0790 KOps/s	919.4822 Ops/s	$\textbf{\color{#35bf28}+17.34\%}$
test_seq_add[eager]	0.1615ms	98.8447μs	10.1169 KOps/s	10.5066 KOps/s	$\color{#d91a1a}-3.71\%$
test_seq_add[compile]	0.1525ms	90.7578μs	11.0183 KOps/s	11.5198 KOps/s	$\color{#d91a1a}-4.35\%$
test_seq_add[compile-overhead]	0.1881ms	0.1282ms	7.8009 KOps/s	7.7042 KOps/s	$\color{#35bf28}+1.26\%$
test_seq_wrap[eager]	0.4712ms	0.3832ms	2.6099 KOps/s	2.6614 KOps/s	$\color{#d91a1a}-1.94\%$
test_seq_wrap[compile]	1.5992ms	0.3075ms	3.2519 KOps/s	3.2681 KOps/s	$\color{#d91a1a}-0.50\%$
test_seq_wrap[compile-overhead]	0.2771ms	0.2287ms	4.3730 KOps/s	4.4322 KOps/s	$\color{#d91a1a}-1.33\%$
test_func_call_runtime[False-eager]	0.8259ms	0.7449ms	1.3425 KOps/s	1.3442 KOps/s	$\color{#d91a1a}-0.13\%$
test_func_call_runtime[False-compile]	0.8353ms	0.7511ms	1.3314 KOps/s	1.3034 KOps/s	$\color{#35bf28}+2.15\%$
test_func_call_runtime[False-compile-overhead]	0.4594ms	0.3637ms	2.7494 KOps/s	2.7271 KOps/s	$\color{#35bf28}+0.82\%$
test_func_call_runtime[True-eager]	0.9953ms	0.8751ms	1.1427 KOps/s	1.1175 KOps/s	$\color{#35bf28}+2.26\%$
test_func_call_runtime[True-compile]	0.8533ms	0.7757ms	1.2891 KOps/s	1.2804 KOps/s	$\color{#35bf28}+0.68\%$
test_func_call_runtime[True-compile-overhead]	0.4527ms	0.3846ms	2.6004 KOps/s	2.5976 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[False-eager]	0.8286ms	0.7702ms	1.2983 KOps/s	1.3134 KOps/s	$\color{#d91a1a}-1.15\%$
test_func_call_cm_runtime[False-compile]	0.8417ms	0.7662ms	1.3051 KOps/s	1.3124 KOps/s	$\color{#d91a1a}-0.56\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4151ms	0.3674ms	2.7221 KOps/s	2.7197 KOps/s	$\color{#35bf28}+0.09\%$
test_func_call_cm_runtime[True-eager]	1.1099ms	0.9777ms	1.0228 KOps/s	1.0018 KOps/s	$\color{#35bf28}+2.10\%$
test_func_call_cm_runtime[True-compile]	0.8980ms	0.8047ms	1.2427 KOps/s	1.2369 KOps/s	$\color{#35bf28}+0.47\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4655ms	0.4114ms	2.4310 KOps/s	2.4178 KOps/s	$\color{#35bf28}+0.55\%$
test_vmap_func_call_cm_runtime[eager]	2.4617ms	1.9964ms	500.8988 Ops/s	494.6911 Ops/s	$\color{#35bf28}+1.25\%$
test_vmap_func_call_cm_runtime[compile]	1.0666ms	0.8142ms	1.2282 KOps/s	1.2099 KOps/s	$\color{#35bf28}+1.52\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4832ms	0.4151ms	2.4089 KOps/s	2.3860 KOps/s	$\color{#35bf28}+0.96\%$
test_distributed	0.8640ms	0.1216ms	8.2270 KOps/s	8.4857 KOps/s	$\color{#d91a1a}-3.05\%$
test_tdmodule	33.6400μs	13.0699μs	76.5115 KOps/s	77.2737 KOps/s	$\color{#d91a1a}-0.99\%$
test_tdmodule_dispatch	73.4910μs	27.5384μs	36.3129 KOps/s	37.0403 KOps/s	$\color{#d91a1a}-1.96\%$
test_tdseq	33.4410μs	14.9524μs	66.8787 KOps/s	68.6415 KOps/s	$\color{#d91a1a}-2.57\%$
test_tdseq_dispatch	49.9910μs	30.4010μs	32.8937 KOps/s	32.8142 KOps/s	$\color{#35bf28}+0.24\%$
test_instantiation_functorch	1.6103ms	1.5267ms	655.0110 Ops/s	645.8354 Ops/s	$\color{#35bf28}+1.42\%$
test_exec_functorch	0.1775ms	0.1403ms	7.1284 KOps/s	7.1714 KOps/s	$\color{#d91a1a}-0.60\%$
test_exec_functional_call	0.1691ms	0.1349ms	7.4136 KOps/s	7.4408 KOps/s	$\color{#d91a1a}-0.36\%$
test_exec_td_decorator	0.3632ms	0.1788ms	5.5936 KOps/s	5.6127 KOps/s	$\color{#d91a1a}-0.34\%$
test_vmap_mlp_speed_decorator[True-True]	0.8135ms	0.6602ms	1.5146 KOps/s	1.5335 KOps/s	$\color{#d91a1a}-1.23\%$
test_vmap_mlp_speed_decorator[True-False]	0.7476ms	0.6546ms	1.5276 KOps/s	1.5277 KOps/s	$-0.01\%$
test_vmap_mlp_speed_decorator[False-True]	0.7184ms	0.5798ms	1.7248 KOps/s	1.7407 KOps/s	$\color{#d91a1a}-0.91\%$
test_vmap_mlp_speed_decorator[False-False]	0.6740ms	0.5694ms	1.7562 KOps/s	1.6654 KOps/s	$\textbf{\color{#35bf28}+5.45\%}$
test_vmap_transformer_speed_decorator[True-True]	18.8838ms	18.4867ms	54.0928 Ops/s	53.5011 Ops/s	$\color{#35bf28}+1.11\%$
test_vmap_transformer_speed_decorator[True-False]	18.6916ms	18.5389ms	53.9406 Ops/s	53.2530 Ops/s	$\color{#35bf28}+1.29\%$
test_vmap_transformer_speed_decorator[False-True]	18.6133ms	18.4368ms	54.2393 Ops/s	53.7467 Ops/s	$\color{#35bf28}+0.92\%$
test_vmap_transformer_speed_decorator[False-False]	18.4496ms	18.3809ms	54.4043 Ops/s	53.8097 Ops/s	$\color{#35bf28}+1.10\%$
test_to_module_speed[True]	1.0223ms	0.9315ms	1.0736 KOps/s	1.0746 KOps/s	$\color{#d91a1a}-0.09\%$
test_to_module_speed[False]	1.3655ms	0.9188ms	1.0884 KOps/s	1.1007 KOps/s	$\color{#d91a1a}-1.12\%$
test_tc_init	55.7210μs	34.9298μs	28.6289 KOps/s	30.9157 KOps/s	$\textbf{\color{#d91a1a}-7.40\%}$
test_tc_init_nested	0.1055ms	72.2170μs	13.8472 KOps/s	15.0129 KOps/s	$\textbf{\color{#d91a1a}-7.76\%}$
test_tc_first_layer_tensor	4.2101μs	0.6980μs	1.4326 MOps/s	1.4818 MOps/s	$\color{#d91a1a}-3.32\%$
test_tc_first_layer_nontensor	23.8410μs	2.3325μs	428.7259 KOps/s	431.1607 KOps/s	$\color{#d91a1a}-0.56\%$
test_tc_second_layer_tensor	20.5955μs	1.4274μs	700.5569 KOps/s	702.5504 KOps/s	$\color{#d91a1a}-0.28\%$
test_tc_second_layer_nontensor	28.8000μs	3.0868μs	323.9630 KOps/s	327.3838 KOps/s	$\color{#d91a1a}-1.04\%$
test_unbind	0.2296s	9.9171ms	100.8364 Ops/s	151.7349 Ops/s	$\textbf{\color{#d91a1a}-33.54\%}$
test_full_like	11.1007ms	9.1147ms	109.7127 Ops/s	107.8302 Ops/s	$\color{#35bf28}+1.75\%$
test_zeros_like	9.2540ms	7.2991ms	137.0039 Ops/s	114.5410 Ops/s	$\textbf{\color{#35bf28}+19.61\%}$
test_ones_like	5.2729ms	4.3202ms	231.4723 Ops/s	232.8777 Ops/s	$\color{#d91a1a}-0.60\%$
test_clone	6.6418ms	6.3072ms	158.5481 Ops/s	156.9472 Ops/s	$\color{#35bf28}+1.02\%$
test_squeeze	59.2010μs	9.5359μs	104.8674 KOps/s	106.6462 KOps/s	$\color{#d91a1a}-1.67\%$
test_unsqueeze	0.1220ms	75.1097μs	13.3139 KOps/s	14.1450 KOps/s	$\textbf{\color{#d91a1a}-5.88\%}$
test_split	0.3880ms	0.1552ms	6.4424 KOps/s	6.1936 KOps/s	$\color{#35bf28}+4.02\%$
test_permute	0.2190ms	0.1766ms	5.6617 KOps/s	5.6105 KOps/s	$\color{#35bf28}+0.91\%$
test_stack	53.2250ms	50.7994ms	19.6853 Ops/s	19.7705 Ops/s	$\color{#d91a1a}-0.43\%$
test_cat	50.7669ms	50.4545ms	19.8198 Ops/s	19.9084 Ops/s	$\color{#d91a1a}-0.44\%$