[FBCode] Deactivate vmap monkey-patching in FBCode

[vmoens]

Stack from ghstack (oldest at bottom):

• [Refactor] Avoid lambda functions in core functionality #1136
• -> [FBCode] Deactivate vmap monkey-patching in FBCode #1135
• [Quality] Avoid torch.distributed imports at root #1134

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}12$. Worsened: $\large\color{#d91a1a}32$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	40.0950μs	18.1561μs	55.0778 KOps/s	60.5053 KOps/s	$\textbf{\color{#d91a1a}-8.97\%}$
test_plain_set_stack_nested	47.9300μs	18.2970μs	54.6538 KOps/s	58.4908 KOps/s	$\textbf{\color{#d91a1a}-6.56\%}$
test_plain_set_nested_inplace	62.0450μs	19.8470μs	50.3854 KOps/s	54.3557 KOps/s	$\textbf{\color{#d91a1a}-7.30\%}$
test_plain_set_stack_nested_inplace	58.6790μs	19.7280μs	50.6894 KOps/s	54.4226 KOps/s	$\textbf{\color{#d91a1a}-6.86\%}$
test_items	19.0350μs	4.2066μs	237.7213 KOps/s	237.9675 KOps/s	$\color{#d91a1a}-0.10\%$
test_items_nested	0.4490ms	0.4000ms	2.5001 KOps/s	2.4677 KOps/s	$\color{#35bf28}+1.31\%$
test_items_nested_locked	0.6933ms	0.3971ms	2.5180 KOps/s	2.4546 KOps/s	$\color{#35bf28}+2.58\%$
test_items_nested_leaf	0.1274ms	71.5707μs	13.9722 KOps/s	14.0344 KOps/s	$\color{#d91a1a}-0.44\%$
test_items_stack_nested	0.7934ms	0.4050ms	2.4690 KOps/s	2.4390 KOps/s	$\color{#35bf28}+1.23\%$
test_items_stack_nested_leaf	0.1423ms	74.0596μs	13.5026 KOps/s	13.5693 KOps/s	$\color{#d91a1a}-0.49\%$
test_items_stack_nested_locked	0.7594ms	0.4040ms	2.4752 KOps/s	2.4412 KOps/s	$\color{#35bf28}+1.39\%$
test_keys	21.3100μs	3.5839μs	279.0241 KOps/s	286.4257 KOps/s	$\color{#d91a1a}-2.58\%$
test_keys_nested	0.1849ms	0.1367ms	7.3131 KOps/s	7.4270 KOps/s	$\color{#d91a1a}-1.53\%$
test_keys_nested_locked	1.9252ms	0.1427ms	7.0093 KOps/s	7.0769 KOps/s	$\color{#d91a1a}-0.95\%$
test_keys_nested_leaf	0.1837ms	0.1173ms	8.5238 KOps/s	8.6237 KOps/s	$\color{#d91a1a}-1.16\%$
test_keys_stack_nested	0.2419ms	0.1348ms	7.4171 KOps/s	7.3975 KOps/s	$\color{#35bf28}+0.26\%$
test_keys_stack_nested_leaf	0.1887ms	0.1162ms	8.6042 KOps/s	8.6696 KOps/s	$\color{#d91a1a}-0.75\%$
test_keys_stack_nested_locked	0.2719ms	0.1417ms	7.0579 KOps/s	7.1251 KOps/s	$\color{#d91a1a}-0.94\%$
test_values	6.6464μs	1.0434μs	958.3796 KOps/s	960.4027 KOps/s	$\color{#d91a1a}-0.21\%$
test_values_nested	0.1308ms	56.1705μs	17.8029 KOps/s	17.5560 KOps/s	$\color{#35bf28}+1.41\%$
test_values_nested_locked	0.1112ms	55.4773μs	18.0254 KOps/s	18.3825 KOps/s	$\color{#d91a1a}-1.94\%$
test_values_nested_leaf	0.1140ms	60.1070μs	16.6370 KOps/s	17.0420 KOps/s	$\color{#d91a1a}-2.38\%$
test_values_stack_nested	0.1241ms	56.1365μs	17.8137 KOps/s	17.9689 KOps/s	$\color{#d91a1a}-0.86\%$
test_values_stack_nested_leaf	0.1166ms	59.5044μs	16.8055 KOps/s	16.7314 KOps/s	$\color{#35bf28}+0.44\%$
test_values_stack_nested_locked	0.1160ms	56.1697μs	17.8032 KOps/s	17.7655 KOps/s	$\color{#35bf28}+0.21\%$
test_membership	5.1287μs	0.7491μs	1.3349 MOps/s	1.3961 MOps/s	$\color{#d91a1a}-4.39\%$
test_membership_nested	25.2070μs	2.9409μs	340.0311 KOps/s	343.7777 KOps/s	$\color{#d91a1a}-1.09\%$
test_membership_nested_leaf	23.1130μs	2.9971μs	333.6506 KOps/s	341.9048 KOps/s	$\color{#d91a1a}-2.41\%$
test_membership_stacked_nested	28.3430μs	2.9310μs	341.1768 KOps/s	344.8048 KOps/s	$\color{#d91a1a}-1.05\%$
test_membership_stacked_nested_leaf	24.8970μs	2.9698μs	336.7282 KOps/s	346.6255 KOps/s	$\color{#d91a1a}-2.86\%$
test_membership_nested_last	45.8850μs	4.2793μs	233.6810 KOps/s	237.3931 KOps/s	$\color{#d91a1a}-1.56\%$
test_membership_nested_leaf_last	23.6140μs	4.2601μs	234.7341 KOps/s	236.0288 KOps/s	$\color{#d91a1a}-0.55\%$
test_membership_stacked_nested_last	43.3110μs	5.3710μs	186.1838 KOps/s	148.2057 KOps/s	$\textbf{\color{#35bf28}+25.63\%}$
test_membership_stacked_nested_leaf_last	28.1920μs	5.4790μs	182.5150 KOps/s	147.7408 KOps/s	$\textbf{\color{#35bf28}+23.54\%}$
test_nested_getleaf	50.2330μs	10.6923μs	93.5250 KOps/s	94.9583 KOps/s	$\color{#d91a1a}-1.51\%$
test_nested_get	46.0460μs	10.4215μs	95.9551 KOps/s	97.1296 KOps/s	$\color{#d91a1a}-1.21\%$
test_stacked_getleaf	37.2800μs	10.8344μs	92.2989 KOps/s	94.1207 KOps/s	$\color{#d91a1a}-1.94\%$
test_stacked_get	49.0020μs	10.1230μs	98.7848 KOps/s	96.4116 KOps/s	$\color{#35bf28}+2.46\%$
test_nested_getitemleaf	39.1400μs	11.0959μs	90.1233 KOps/s	90.0193 KOps/s	$\color{#35bf28}+0.12\%$
test_nested_getitem	34.4850μs	10.4642μs	95.5636 KOps/s	94.9361 KOps/s	$\color{#35bf28}+0.66\%$
test_stacked_getitemleaf	48.1500μs	11.1649μs	89.5661 KOps/s	91.1666 KOps/s	$\color{#d91a1a}-1.76\%$
test_stacked_getitem	29.7050μs	10.4149μs	96.0166 KOps/s	96.0517 KOps/s	$\color{#d91a1a}-0.04\%$
test_lock_nested	4.3484ms	0.4401ms	2.2723 KOps/s	2.2473 KOps/s	$\color{#35bf28}+1.11\%$
test_lock_stack_nested	0.6364ms	0.4058ms	2.4645 KOps/s	2.4489 KOps/s	$\color{#35bf28}+0.63\%$
test_unlock_nested	0.6596ms	0.3559ms	2.8099 KOps/s	2.7510 KOps/s	$\color{#35bf28}+2.14\%$
test_unlock_stack_nested	0.4037ms	0.3238ms	3.0884 KOps/s	3.0878 KOps/s	$\color{#35bf28}+0.02\%$
test_flatten_speed	0.1490ms	93.5367μs	10.6910 KOps/s	10.6642 KOps/s	$\color{#35bf28}+0.25\%$
test_unflatten_speed	0.8867ms	0.4965ms	2.0142 KOps/s	2.0465 KOps/s	$\color{#d91a1a}-1.57\%$
test_common_ops	4.4028ms	0.8046ms	1.2429 KOps/s	1.3542 KOps/s	$\textbf{\color{#d91a1a}-8.22\%}$
test_creation	12.3130μs	2.1168μs	472.4046 KOps/s	477.9785 KOps/s	$\color{#d91a1a}-1.17\%$
test_creation_empty	49.5820μs	12.4128μs	80.5619 KOps/s	114.5590 KOps/s	$\textbf{\color{#d91a1a}-29.68\%}$
test_creation_nested_1	40.1850μs	15.1161μs	66.1546 KOps/s	89.5032 KOps/s	$\textbf{\color{#d91a1a}-26.09\%}$
test_creation_nested_2	59.7820μs	19.3186μs	51.7635 KOps/s	63.7593 KOps/s	$\textbf{\color{#d91a1a}-18.81\%}$
test_clone	70.9920μs	12.9778μs	77.0546 KOps/s	75.7311 KOps/s	$\color{#35bf28}+1.75\%$
test_getitem[int]	1.2227ms	12.9602μs	77.1592 KOps/s	80.3344 KOps/s	$\color{#d91a1a}-3.95\%$
test_getitem[slice_int]	0.1440ms	25.9150μs	38.5877 KOps/s	39.2007 KOps/s	$\color{#d91a1a}-1.56\%$
test_getitem[range]	0.1703ms	49.9480μs	20.0208 KOps/s	20.5343 KOps/s	$\color{#d91a1a}-2.50\%$
test_getitem[tuple]	0.1380ms	20.7580μs	48.1742 KOps/s	48.5521 KOps/s	$\color{#d91a1a}-0.78\%$
test_getitem[list]	0.1686ms	45.9274μs	21.7735 KOps/s	22.7433 KOps/s	$\color{#d91a1a}-4.26\%$
test_setitem_dim[int]	49.0710μs	25.3906μs	39.3846 KOps/s	38.6961 KOps/s	$\color{#35bf28}+1.78\%$
test_setitem_dim[slice_int]	83.9070μs	52.6949μs	18.9772 KOps/s	18.3351 KOps/s	$\color{#35bf28}+3.50\%$
test_setitem_dim[range]	0.1236ms	73.8429μs	13.5423 KOps/s	13.3669 KOps/s	$\color{#35bf28}+1.31\%$
test_setitem_dim[tuple]	85.2990μs	41.7861μs	23.9314 KOps/s	23.8106 KOps/s	$\color{#35bf28}+0.51\%$
test_setitem	64.1500μs	21.5100μs	46.4901 KOps/s	51.6479 KOps/s	$\textbf{\color{#d91a1a}-9.99\%}$
test_set	68.8780μs	20.9385μs	47.7589 KOps/s	54.4717 KOps/s	$\textbf{\color{#d91a1a}-12.32\%}$
test_set_shared	3.4300ms	0.1660ms	6.0252 KOps/s	5.9122 KOps/s	$\color{#35bf28}+1.91\%$
test_update	0.1200ms	24.6365μs	40.5901 KOps/s	50.2741 KOps/s	$\textbf{\color{#d91a1a}-19.26\%}$
test_update_nested	0.1016ms	35.0769μs	28.5088 KOps/s	32.5430 KOps/s	$\textbf{\color{#d91a1a}-12.40\%}$
test_update__nested	0.2581ms	31.8561μs	31.3912 KOps/s	29.6997 KOps/s	$\textbf{\color{#35bf28}+5.70\%}$
test_set_nested	67.6360μs	22.6394μs	44.1707 KOps/s	48.1480 KOps/s	$\textbf{\color{#d91a1a}-8.26\%}$
test_set_nested_new	77.7150μs	27.2909μs	36.6423 KOps/s	39.4211 KOps/s	$\textbf{\color{#d91a1a}-7.05\%}$
test_select	0.3940ms	43.4190μs	23.0314 KOps/s	23.8850 KOps/s	$\color{#d91a1a}-3.57\%$
test_select_nested	0.1260ms	59.4783μs	16.8129 KOps/s	16.8978 KOps/s	$\color{#d91a1a}-0.50\%$
test_exclude_nested	0.1534ms	78.5252μs	12.7348 KOps/s	12.7379 KOps/s	$\color{#d91a1a}-0.02\%$
test_empty[True]	0.5302ms	0.3783ms	2.6433 KOps/s	2.6339 KOps/s	$\color{#35bf28}+0.36\%$
test_empty[False]	8.0710μs	1.1866μs	842.7309 KOps/s	821.9498 KOps/s	$\color{#35bf28}+2.53\%$
test_unbind_speed	0.3830ms	0.2587ms	3.8653 KOps/s	3.7655 KOps/s	$\color{#35bf28}+2.65\%$
test_unbind_speed_stack0	0.4720ms	0.2526ms	3.9585 KOps/s	3.9315 KOps/s	$\color{#35bf28}+0.69\%$
test_unbind_speed_stack1	96.1509ms	0.7287ms	1.3723 KOps/s	1.6126 KOps/s	$\textbf{\color{#d91a1a}-14.90\%}$
test_split	1.8201ms	1.5897ms	629.0432 Ops/s	575.9940 Ops/s	$\textbf{\color{#35bf28}+9.21\%}$
test_chunk	88.1541ms	1.8664ms	535.7890 Ops/s	584.0087 Ops/s	$\textbf{\color{#d91a1a}-8.26\%}$
test_consolidate_njt[False-None]	8.4849ms	8.0001ms	124.9979 Ops/s	115.4930 Ops/s	$\textbf{\color{#35bf28}+8.23\%}$
test_creation[device0]	0.2157ms	90.2719μs	11.0776 KOps/s	10.6633 KOps/s	$\color{#35bf28}+3.89\%$
test_creation_from_tensor	3.0037ms	93.5804μs	10.6860 KOps/s	10.3297 KOps/s	$\color{#35bf28}+3.45\%$
test_add_one[memmap_tensor0]	0.1235ms	5.0583μs	197.6937 KOps/s	203.5954 KOps/s	$\color{#d91a1a}-2.90\%$
test_contiguous[memmap_tensor0]	10.1290μs	0.5188μs	1.9276 MOps/s	1.9652 MOps/s	$\color{#d91a1a}-1.91\%$
test_stack[memmap_tensor0]	41.6880μs	3.5324μs	283.0915 KOps/s	297.0641 KOps/s	$\color{#d91a1a}-4.70\%$
test_memmaptd_index	1.1196ms	0.2350ms	4.2557 KOps/s	4.2078 KOps/s	$\color{#35bf28}+1.14\%$
test_memmaptd_index_astensor	0.5882ms	0.3137ms	3.1880 KOps/s	3.1437 KOps/s	$\color{#35bf28}+1.41\%$
test_memmaptd_index_op	0.9360ms	0.6008ms	1.6644 KOps/s	1.8008 KOps/s	$\textbf{\color{#d91a1a}-7.57\%}$
test_serialize_model	0.1173s	0.1107s	9.0310 Ops/s	7.9109 Ops/s	$\textbf{\color{#35bf28}+14.16\%}$
test_serialize_model_pickle	0.4526s	0.3865s	2.5872 Ops/s	2.5805 Ops/s	$\color{#35bf28}+0.26\%$
test_serialize_weights	0.1950s	0.1245s	8.0310 Ops/s	9.0252 Ops/s	$\textbf{\color{#d91a1a}-11.02\%}$
test_serialize_weights_returnearly	0.1641s	0.1529s	6.5413 Ops/s	6.4453 Ops/s	$\color{#35bf28}+1.49\%$
test_serialize_weights_pickle	1.1521s	0.7017s	1.4250 Ops/s	2.4096 Ops/s	$\textbf{\color{#d91a1a}-40.86\%}$
test_serialize_weights_filesystem	0.1466s	0.1392s	7.1834 Ops/s	6.4309 Ops/s	$\textbf{\color{#35bf28}+11.70\%}$
test_serialize_model_filesystem	0.2345s	0.1561s	6.4048 Ops/s	6.8329 Ops/s	$\textbf{\color{#d91a1a}-6.27\%}$
test_reshape_pytree	86.9130μs	27.3943μs	36.5040 KOps/s	36.5000 KOps/s	$\color{#35bf28}+0.01\%$
test_reshape_td	65.7330μs	33.0697μs	30.2392 KOps/s	30.4772 KOps/s	$\color{#d91a1a}-0.78\%$
test_view_pytree	78.4570μs	26.9809μs	37.0633 KOps/s	36.4234 KOps/s	$\color{#35bf28}+1.76\%$
test_view_td	82.9250μs	38.9003μs	25.7068 KOps/s	24.7085 KOps/s	$\color{#35bf28}+4.04\%$
test_unbind_pytree	66.3440μs	30.1361μs	33.1828 KOps/s	32.8789 KOps/s	$\color{#35bf28}+0.92\%$
test_unbind_td	0.3565ms	38.4934μs	25.9785 KOps/s	25.7618 KOps/s	$\color{#35bf28}+0.84\%$
test_split_pytree	63.2380μs	30.3794μs	32.9170 KOps/s	32.8468 KOps/s	$\color{#35bf28}+0.21\%$
test_split_td	0.5635ms	45.8358μs	21.8170 KOps/s	22.3814 KOps/s	$\color{#d91a1a}-2.52\%$
test_add_pytree	77.9560μs	36.4891μs	27.4054 KOps/s	27.4418 KOps/s	$\color{#d91a1a}-0.13\%$
test_add_td	0.1176ms	56.5415μs	17.6861 KOps/s	19.3206 KOps/s	$\textbf{\color{#d91a1a}-8.46\%}$
test_compile_add_one_nested[tensordict-compile]	0.1694ms	62.2914μs	16.0536 KOps/s	15.9433 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_add_one_nested[tensordict-eager]	1.2493ms	0.1609ms	6.2154 KOps/s	5.2417 KOps/s	$\textbf{\color{#35bf28}+18.57\%}$
test_compile_add_one_nested[pytree-compile]	0.1076ms	46.0199μs	21.7297 KOps/s	21.4997 KOps/s	$\color{#35bf28}+1.07\%$
test_compile_add_one_nested[pytree-eager]	0.2173ms	0.1205ms	8.3012 KOps/s	8.2825 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_copy_nested[tensordict-compile]	59.2910μs	26.1546μs	38.2341 KOps/s	39.3765 KOps/s	$\color{#d91a1a}-2.90\%$
test_compile_copy_nested[tensordict-eager]	0.1162ms	53.5535μs	18.6729 KOps/s	18.7807 KOps/s	$\color{#d91a1a}-0.57\%$
test_compile_copy_nested[pytree-compile]	0.2082ms	78.1066μs	12.8030 KOps/s	12.7728 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_copy_nested[pytree-eager]	0.1439ms	67.8634μs	14.7355 KOps/s	14.7885 KOps/s	$\color{#d91a1a}-0.36\%$
test_compile_add_one_flat[tensordict-compile]	0.2242ms	0.1038ms	9.6302 KOps/s	9.5214 KOps/s	$\color{#35bf28}+1.14\%$
test_compile_add_one_flat[tensordict-eager]	0.2859ms	0.2007ms	4.9836 KOps/s	4.9768 KOps/s	$\color{#35bf28}+0.14\%$
test_compile_add_one_flat[tensorclass-compile]	0.1080ms	45.5119μs	21.9723 KOps/s	22.6872 KOps/s	$\color{#d91a1a}-3.15\%$
test_compile_add_one_flat[tensorclass-eager]	0.3741ms	61.2658μs	16.3223 KOps/s	16.0289 KOps/s	$\color{#35bf28}+1.83\%$
test_compile_add_one_flat[pytree-compile]	0.2237ms	0.1040ms	9.6143 KOps/s	9.8193 KOps/s	$\color{#d91a1a}-2.09\%$
test_compile_add_one_flat[pytree-eager]	0.4301ms	0.2023ms	4.9424 KOps/s	4.9483 KOps/s	$\color{#d91a1a}-0.12\%$
test_compile_add_self_flat[tensordict-eager]	0.3128ms	0.2128ms	4.6986 KOps/s	4.6865 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_add_self_flat[tensordict-compile]	0.2274ms	0.1078ms	9.2792 KOps/s	9.3262 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_add_self_flat[tensorclass-eager]	0.1201ms	55.9357μs	17.8777 KOps/s	18.0415 KOps/s	$\color{#d91a1a}-0.91\%$
test_compile_add_self_flat[tensorclass-compile]	0.1315ms	47.3815μs	21.1053 KOps/s	21.8990 KOps/s	$\color{#d91a1a}-3.62\%$
test_compile_add_self_flat[pytree-eager]	0.5444ms	0.1594ms	6.2723 KOps/s	6.2276 KOps/s	$\color{#35bf28}+0.72\%$
test_compile_add_self_flat[pytree-compile]	0.1919ms	0.1039ms	9.6287 KOps/s	9.7219 KOps/s	$\color{#d91a1a}-0.96\%$
test_compile_copy_flat[tensordict-compile]	60.6630μs	21.4407μs	46.6402 KOps/s	48.6653 KOps/s	$\color{#d91a1a}-4.16\%$
test_compile_copy_flat[tensordict-eager]	0.1253ms	58.8319μs	16.9976 KOps/s	16.9973 KOps/s	$+0.00\%$
test_compile_copy_flat[pytree-compile]	0.2241ms	82.0959μs	12.1809 KOps/s	12.2964 KOps/s	$\color{#d91a1a}-0.94\%$
test_compile_copy_flat[pytree-eager]	0.1548ms	69.9135μs	14.3034 KOps/s	14.4762 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_assign_and_add[tensordict-compile]	0.4673ms	0.2103ms	4.7553 KOps/s	4.7778 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_assign_and_add[tensordict-eager]	1.4090ms	1.3127ms	761.7756 Ops/s	779.5152 Ops/s	$\color{#d91a1a}-2.28\%$
test_compile_assign_and_add[pytree-compile]	0.4101ms	0.2059ms	4.8572 KOps/s	4.8331 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_assign_and_add[pytree-eager]	1.6400ms	0.7816ms	1.2795 KOps/s	1.2991 KOps/s	$\color{#d91a1a}-1.51\%$
test_compile_assign_and_add_stack[compile]	0.5629ms	0.4565ms	2.1907 KOps/s	2.2013 KOps/s	$\color{#d91a1a}-0.48\%$
test_compile_assign_and_add_stack[eager]	4.1795ms	2.7644ms	361.7364 Ops/s	398.4745 Ops/s	$\textbf{\color{#d91a1a}-9.22\%}$
test_compile_indexing[tensor-tensordict-compile]	0.1404ms	36.5118μs	27.3884 KOps/s	27.0704 KOps/s	$\color{#35bf28}+1.17\%$
test_compile_indexing[tensor-tensordict-eager]	0.2264s	44.5252μs	22.4592 KOps/s	29.7378 KOps/s	$\textbf{\color{#d91a1a}-24.48\%}$
test_compile_indexing[tensor-tensorclass-compile]	71.4130μs	30.1541μs	33.1630 KOps/s	34.1343 KOps/s	$\color{#d91a1a}-2.85\%$
test_compile_indexing[tensor-tensorclass-eager]	86.6700μs	23.8406μs	41.9453 KOps/s	41.5049 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_indexing[tensor-pytree-compile]	70.7030μs	31.2041μs	32.0470 KOps/s	33.4954 KOps/s	$\color{#d91a1a}-4.32\%$
test_compile_indexing[tensor-pytree-eager]	69.3890μs	23.9175μs	41.8104 KOps/s	41.8937 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_indexing[slice-tensordict-compile]	0.1199ms	52.3702μs	19.0948 KOps/s	19.2551 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_indexing[slice-tensordict-eager]	0.4193ms	19.8990μs	50.2537 KOps/s	48.5492 KOps/s	$\color{#35bf28}+3.51\%$
test_compile_indexing[slice-tensorclass-compile]	98.5140μs	44.6596μs	22.3916 KOps/s	22.3296 KOps/s	$\color{#35bf28}+0.28\%$
test_compile_indexing[slice-tensorclass-eager]	65.8230μs	18.9063μs	52.8924 KOps/s	51.5444 KOps/s	$\color{#35bf28}+2.62\%$
test_compile_indexing[slice-pytree-compile]	0.1221ms	46.1307μs	21.6776 KOps/s	22.0592 KOps/s	$\color{#d91a1a}-1.73\%$
test_compile_indexing[slice-pytree-eager]	0.6231ms	19.0974μs	52.3632 KOps/s	50.8005 KOps/s	$\color{#35bf28}+3.08\%$
test_compile_indexing[int-tensordict-compile]	0.5254ms	53.7578μs	18.6019 KOps/s	18.9830 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_indexing[int-tensordict-eager]	0.8622ms	20.2209μs	49.4538 KOps/s	49.4769 KOps/s	$\color{#d91a1a}-0.05\%$
test_compile_indexing[int-tensorclass-compile]	0.1120ms	45.3906μs	22.0310 KOps/s	22.1814 KOps/s	$\color{#d91a1a}-0.68\%$
test_compile_indexing[int-tensorclass-eager]	57.4770μs	19.1150μs	52.3150 KOps/s	51.5077 KOps/s	$\color{#35bf28}+1.57\%$
test_compile_indexing[int-pytree-compile]	95.8390μs	44.9434μs	22.2502 KOps/s	22.1777 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_indexing[int-pytree-eager]	53.0290μs	19.0425μs	52.5142 KOps/s	51.9731 KOps/s	$\color{#35bf28}+1.04\%$
test_mod_add[eager]	99.9760μs	36.5498μs	27.3600 KOps/s	29.8468 KOps/s	$\textbf{\color{#d91a1a}-8.33\%}$
test_mod_add[compile]	92.0620μs	47.1802μs	21.1953 KOps/s	20.5422 KOps/s	$\color{#35bf28}+3.18\%$
test_mod_add[compile-overhead]	0.1220ms	46.6567μs	21.4331 KOps/s	20.8132 KOps/s	$\color{#35bf28}+2.98\%$
test_mod_wrap[eager]	0.4075ms	0.2277ms	4.3926 KOps/s	4.4113 KOps/s	$\color{#d91a1a}-0.42\%$
test_mod_wrap[compile]	0.4184ms	0.2085ms	4.7958 KOps/s	4.7743 KOps/s	$\color{#35bf28}+0.45\%$
test_mod_wrap[compile-overhead]	0.3769ms	0.2074ms	4.8213 KOps/s	4.7957 KOps/s	$\color{#35bf28}+0.53\%$
test_mod_wrap_and_backward[eager]	12.0482ms	10.6424ms	93.9639 Ops/s	85.0782 Ops/s	$\textbf{\color{#35bf28}+10.44\%}$
test_mod_wrap_and_backward[compile]	11.9395ms	10.5673ms	94.6314 Ops/s	83.2479 Ops/s	$\textbf{\color{#35bf28}+13.67\%}$
test_mod_wrap_and_backward[compile-overhead]	12.1289ms	10.5265ms	94.9979 Ops/s	80.9532 Ops/s	$\textbf{\color{#35bf28}+17.35\%}$
test_seq_add[eager]	0.2496ms	0.1188ms	8.4163 KOps/s	8.8422 KOps/s	$\color{#d91a1a}-4.82\%$
test_seq_add[compile]	0.1129ms	62.5193μs	15.9951 KOps/s	16.2723 KOps/s	$\color{#d91a1a}-1.70\%$
test_seq_add[compile-overhead]	0.1153ms	59.2936μs	16.8652 KOps/s	16.9568 KOps/s	$\color{#d91a1a}-0.54\%$
test_seq_wrap[eager]	1.0130ms	0.4491ms	2.2267 KOps/s	2.2725 KOps/s	$\color{#d91a1a}-2.02\%$
test_seq_wrap[compile]	0.4160ms	0.2256ms	4.4332 KOps/s	4.3582 KOps/s	$\color{#35bf28}+1.72\%$
test_seq_wrap[compile-overhead]	0.4441ms	0.2279ms	4.3879 KOps/s	4.2612 KOps/s	$\color{#35bf28}+2.97\%$
test_func_call_runtime[False-eager]	0.7294ms	0.5394ms	1.8540 KOps/s	1.7745 KOps/s	$\color{#35bf28}+4.48\%$
test_func_call_runtime[False-compile]	0.8026ms	0.4277ms	2.3381 KOps/s	2.3553 KOps/s	$\color{#d91a1a}-0.73\%$
test_func_call_runtime[False-compile-overhead]	0.5171ms	0.4249ms	2.3534 KOps/s	2.3343 KOps/s	$\color{#35bf28}+0.82\%$
test_func_call_runtime[True-eager]	0.8358ms	0.7515ms	1.3307 KOps/s	1.2934 KOps/s	$\color{#35bf28}+2.89\%$
test_func_call_runtime[True-compile]	0.5449ms	0.4611ms	2.1688 KOps/s	2.1564 KOps/s	$\color{#35bf28}+0.57\%$
test_func_call_runtime[True-compile-overhead]	0.6180ms	0.4642ms	2.1541 KOps/s	2.1417 KOps/s	$\color{#35bf28}+0.58\%$
test_func_call_cm_runtime[False-eager]	0.8673ms	0.5378ms	1.8594 KOps/s	1.7995 KOps/s	$\color{#35bf28}+3.33\%$
test_func_call_cm_runtime[False-compile]	0.8027ms	0.4261ms	2.3467 KOps/s	2.3564 KOps/s	$\color{#d91a1a}-0.41\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5178ms	0.4258ms	2.3487 KOps/s	2.3461 KOps/s	$\color{#35bf28}+0.11\%$
test_func_call_cm_runtime[True-eager]	1.5164ms	0.8843ms	1.1308 KOps/s	1.1096 KOps/s	$\color{#35bf28}+1.91\%$
test_func_call_cm_runtime[True-compile]	0.6397ms	0.4878ms	2.0502 KOps/s	2.0565 KOps/s	$\color{#d91a1a}-0.31\%$
test_func_call_cm_runtime[True-compile-overhead]	0.6305ms	0.4869ms	2.0537 KOps/s	2.0518 KOps/s	$\color{#35bf28}+0.09\%$
test_vmap_func_call_cm_runtime[eager]	2.7805ms	1.9057ms	524.7528 Ops/s	530.8484 Ops/s	$\color{#d91a1a}-1.15\%$
test_vmap_func_call_cm_runtime[compile]	0.8410ms	0.5174ms	1.9328 KOps/s	1.9198 KOps/s	$\color{#35bf28}+0.68\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.0194ms	0.5166ms	1.9358 KOps/s	1.9137 KOps/s	$\color{#35bf28}+1.15\%$
test_distributed	0.2301ms	0.1263ms	7.9158 KOps/s	7.7633 KOps/s	$\color{#35bf28}+1.97\%$
test_tdmodule	55.2030μs	26.8713μs	37.2144 KOps/s	38.9645 KOps/s	$\color{#d91a1a}-4.49\%$
test_tdmodule_dispatch	80.8010μs	48.8253μs	20.4812 KOps/s	21.6029 KOps/s	$\textbf{\color{#d91a1a}-5.19\%}$
test_tdseq	54.6120μs	27.5986μs	36.2338 KOps/s	39.8497 KOps/s	$\textbf{\color{#d91a1a}-9.07\%}$
test_tdseq_dispatch	0.1009ms	54.0558μs	18.4994 KOps/s	20.6032 KOps/s	$\textbf{\color{#d91a1a}-10.21\%}$
test_instantiation_functorch	2.5221ms	1.5994ms	625.2510 Ops/s	630.4313 Ops/s	$\color{#d91a1a}-0.82\%$
test_exec_functorch	0.2923ms	0.1778ms	5.6255 KOps/s	5.4126 KOps/s	$\color{#35bf28}+3.93\%$
test_exec_functional_call	0.2959ms	0.1704ms	5.8676 KOps/s	5.5597 KOps/s	$\textbf{\color{#35bf28}+5.54\%}$
test_exec_td_decorator	0.4456ms	0.2279ms	4.3881 KOps/s	4.3016 KOps/s	$\color{#35bf28}+2.01\%$
test_vmap_mlp_speed_decorator[True-True]	1.2142ms	0.6608ms	1.5134 KOps/s	1.5163 KOps/s	$\color{#d91a1a}-0.19\%$
test_vmap_mlp_speed_decorator[True-False]	1.3286ms	0.6592ms	1.5171 KOps/s	1.4917 KOps/s	$\color{#35bf28}+1.70\%$
test_vmap_mlp_speed_decorator[False-True]	0.8527ms	0.5282ms	1.8932 KOps/s	1.8709 KOps/s	$\color{#35bf28}+1.20\%$
test_vmap_mlp_speed_decorator[False-False]	0.6606ms	0.5258ms	1.9018 KOps/s	1.8737 KOps/s	$\color{#35bf28}+1.50\%$
test_to_module_speed[True]	2.1453ms	1.2894ms	775.5308 Ops/s	781.4907 Ops/s	$\color{#d91a1a}-0.76\%$
test_to_module_speed[False]	1.7150ms	1.2572ms	795.4028 Ops/s	797.6166 Ops/s	$\color{#d91a1a}-0.28\%$
test_tc_init	99.6870μs	48.1987μs	20.7475 KOps/s	24.8047 KOps/s	$\textbf{\color{#d91a1a}-16.36\%}$
test_tc_init_nested	0.2155ms	97.3088μs	10.2766 KOps/s	12.3640 KOps/s	$\textbf{\color{#d91a1a}-16.88\%}$
test_tc_first_layer_tensor	20.6580μs	1.4945μs	669.1077 KOps/s	658.0302 KOps/s	$\color{#35bf28}+1.68\%$
test_tc_first_layer_nontensor	26.9900μs	4.6766μs	213.8287 KOps/s	213.2901 KOps/s	$\color{#35bf28}+0.25\%$
test_tc_second_layer_tensor	21.7510μs	2.7630μs	361.9195 KOps/s	360.3360 KOps/s	$\color{#35bf28}+0.44\%$
test_tc_second_layer_nontensor	35.3660μs	5.9067μs	169.3007 KOps/s	166.0555 KOps/s	$\color{#35bf28}+1.95\%$
test_unbind	0.2091s	12.5503ms	79.6796 Ops/s	84.2978 Ops/s	$\textbf{\color{#d91a1a}-5.48\%}$
test_full_like	14.3146ms	11.7589ms	85.0421 Ops/s	92.3558 Ops/s	$\textbf{\color{#d91a1a}-7.92\%}$
test_zeros_like	12.8306ms	8.1838ms	122.1921 Ops/s	140.6882 Ops/s	$\textbf{\color{#d91a1a}-13.15\%}$
test_ones_like	10.2930ms	7.5260ms	132.8726 Ops/s	137.2385 Ops/s	$\color{#d91a1a}-3.18\%$
test_clone	12.1790ms	8.7569ms	114.1950 Ops/s	112.4714 Ops/s	$\color{#35bf28}+1.53\%$
test_squeeze	61.3150μs	11.7270μs	85.2734 KOps/s	86.0497 KOps/s	$\color{#d91a1a}-0.90\%$
test_unsqueeze	0.2857ms	88.9371μs	11.2439 KOps/s	10.7371 KOps/s	$\color{#35bf28}+4.72\%$
test_split	0.3880ms	0.1998ms	5.0052 KOps/s	5.1936 KOps/s	$\color{#d91a1a}-3.63\%$
test_permute	0.3892ms	0.2039ms	4.9035 KOps/s	4.8592 KOps/s	$\color{#35bf28}+0.91\%$
test_stack	29.1480ms	24.6262ms	40.6071 Ops/s	40.9438 Ops/s	$\color{#d91a1a}-0.82\%$
test_cat	26.5040ms	24.3648ms	41.0427 Ops/s	40.4722 Ops/s	$\color{#35bf28}+1.41\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}6$. Worsened: $\large\color{#d91a1a}45$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1158ms	11.0786μs	90.2641 KOps/s	100.6182 KOps/s	$\textbf{\color{#d91a1a}-10.29\%}$
test_plain_set_stack_nested	39.4410μs	11.0311μs	90.6525 KOps/s	100.2648 KOps/s	$\textbf{\color{#d91a1a}-9.59\%}$
test_plain_set_nested_inplace	52.5410μs	11.9517μs	83.6700 KOps/s	92.1155 KOps/s	$\textbf{\color{#d91a1a}-9.17\%}$
test_plain_set_stack_nested_inplace	36.9000μs	11.9240μs	83.8643 KOps/s	92.1665 KOps/s	$\textbf{\color{#d91a1a}-9.01\%}$
test_items	36.2710μs	2.9090μs	343.7555 KOps/s	342.9961 KOps/s	$\color{#35bf28}+0.22\%$
test_items_nested	0.3868ms	0.3550ms	2.8170 KOps/s	2.8591 KOps/s	$\color{#d91a1a}-1.47\%$
test_items_nested_locked	0.5088ms	0.3575ms	2.7968 KOps/s	2.8167 KOps/s	$\color{#d91a1a}-0.71\%$
test_items_nested_leaf	94.9720μs	58.7503μs	17.0212 KOps/s	17.1740 KOps/s	$\color{#d91a1a}-0.89\%$
test_items_stack_nested	0.3932ms	0.3541ms	2.8241 KOps/s	2.8397 KOps/s	$\color{#d91a1a}-0.55\%$
test_items_stack_nested_leaf	92.9120μs	58.2887μs	17.1560 KOps/s	17.0940 KOps/s	$\color{#35bf28}+0.36\%$
test_items_stack_nested_locked	0.5284ms	0.3574ms	2.7983 KOps/s	2.8218 KOps/s	$\color{#d91a1a}-0.83\%$
test_keys	0.1812ms	3.4886μs	286.6495 KOps/s	289.2708 KOps/s	$\color{#d91a1a}-0.91\%$
test_keys_nested	0.2433ms	69.5048μs	14.3875 KOps/s	14.2619 KOps/s	$\color{#35bf28}+0.88\%$
test_keys_nested_locked	0.8872ms	75.8019μs	13.1923 KOps/s	13.2085 KOps/s	$\color{#d91a1a}-0.12\%$
test_keys_nested_leaf	93.5220μs	61.5573μs	16.2450 KOps/s	16.2891 KOps/s	$\color{#d91a1a}-0.27\%$
test_keys_stack_nested	0.1029ms	70.4171μs	14.2011 KOps/s	14.0734 KOps/s	$\color{#35bf28}+0.91\%$
test_keys_stack_nested_leaf	0.1358ms	61.6380μs	16.2238 KOps/s	16.1322 KOps/s	$\color{#35bf28}+0.57\%$
test_keys_stack_nested_locked	0.1215ms	75.6306μs	13.2222 KOps/s	13.0660 KOps/s	$\color{#35bf28}+1.20\%$
test_values	5.6585μs	0.8435μs	1.1855 MOps/s	1.1870 MOps/s	$\color{#d91a1a}-0.12\%$
test_values_nested	0.1447ms	31.2624μs	31.9873 KOps/s	32.1996 KOps/s	$\color{#d91a1a}-0.66\%$
test_values_nested_locked	66.2510μs	33.0857μs	30.2246 KOps/s	30.4706 KOps/s	$\color{#d91a1a}-0.81\%$
test_values_nested_leaf	0.1598ms	33.6406μs	29.7260 KOps/s	29.7692 KOps/s	$\color{#d91a1a}-0.15\%$
test_values_stack_nested	0.1811ms	31.4562μs	31.7903 KOps/s	31.8373 KOps/s	$\color{#d91a1a}-0.15\%$
test_values_stack_nested_leaf	0.1006ms	33.9553μs	29.4505 KOps/s	29.5747 KOps/s	$\color{#d91a1a}-0.42\%$
test_values_stack_nested_locked	61.4120μs	33.1593μs	30.1575 KOps/s	30.0827 KOps/s	$\color{#35bf28}+0.25\%$
test_membership	2.0810μs	0.5101μs	1.9602 MOps/s	1.9663 MOps/s	$\color{#d91a1a}-0.31\%$
test_membership_nested	37.1155μs	2.0409μs	489.9712 KOps/s	492.4674 KOps/s	$\color{#d91a1a}-0.51\%$
test_membership_nested_leaf	17.9650μs	2.0658μs	484.0844 KOps/s	494.9287 KOps/s	$\color{#d91a1a}-2.19\%$
test_membership_stacked_nested	0.1480ms	2.1382μs	467.6756 KOps/s	475.6107 KOps/s	$\color{#d91a1a}-1.67\%$
test_membership_stacked_nested_leaf	32.1710μs	2.1190μs	471.9170 KOps/s	471.3529 KOps/s	$\color{#35bf28}+0.12\%$
test_membership_nested_last	42.3010μs	2.9317μs	341.0974 KOps/s	339.0168 KOps/s	$\color{#35bf28}+0.61\%$
test_membership_nested_leaf_last	33.7610μs	2.9853μs	334.9727 KOps/s	335.1325 KOps/s	$\color{#d91a1a}-0.05\%$
test_membership_stacked_nested_last	28.1300μs	3.0111μs	332.1012 KOps/s	336.4362 KOps/s	$\color{#d91a1a}-1.29\%$
test_membership_stacked_nested_leaf_last	29.1010μs	2.9623μs	337.5730 KOps/s	340.1518 KOps/s	$\color{#d91a1a}-0.76\%$
test_nested_getleaf	69.5320μs	6.1125μs	163.5985 KOps/s	162.4525 KOps/s	$\color{#35bf28}+0.71\%$
test_nested_get	28.3210μs	5.8698μs	170.3650 KOps/s	172.0342 KOps/s	$\color{#d91a1a}-0.97\%$
test_stacked_getleaf	37.5510μs	6.1652μs	162.2013 KOps/s	163.1330 KOps/s	$\color{#d91a1a}-0.57\%$
test_stacked_get	32.5810μs	5.8593μs	170.6691 KOps/s	171.5939 KOps/s	$\color{#d91a1a}-0.54\%$
test_nested_getitemleaf	35.2110μs	6.2808μs	159.2163 KOps/s	160.2169 KOps/s	$\color{#d91a1a}-0.62\%$
test_nested_getitem	29.6800μs	5.9711μs	167.4736 KOps/s	169.5126 KOps/s	$\color{#d91a1a}-1.20\%$
test_stacked_getitemleaf	58.7610μs	6.2576μs	159.8063 KOps/s	160.2491 KOps/s	$\color{#d91a1a}-0.28\%$
test_stacked_getitem	66.6220μs	5.9190μs	168.9461 KOps/s	169.5119 KOps/s	$\color{#d91a1a}-0.33\%$
test_lock_nested	9.1979ms	0.3782ms	2.6444 KOps/s	2.7079 KOps/s	$\color{#d91a1a}-2.34\%$
test_lock_stack_nested	0.4647ms	0.3367ms	2.9698 KOps/s	3.0020 KOps/s	$\color{#d91a1a}-1.08\%$
test_unlock_nested	0.7575ms	0.3101ms	3.2243 KOps/s	3.2974 KOps/s	$\color{#d91a1a}-2.22\%$
test_unlock_stack_nested	0.4425ms	0.2766ms	3.6156 KOps/s	3.6759 KOps/s	$\color{#d91a1a}-1.64\%$
test_flatten_speed	0.2582ms	75.2230μs	13.2938 KOps/s	13.3987 KOps/s	$\color{#d91a1a}-0.78\%$
test_unflatten_speed	0.5364ms	0.3132ms	3.1924 KOps/s	3.2709 KOps/s	$\color{#d91a1a}-2.40\%$
test_common_ops	1.7862ms	0.6146ms	1.6270 KOps/s	1.8041 KOps/s	$\textbf{\color{#d91a1a}-9.82\%}$
test_creation	0.1149ms	1.4789μs	676.1619 KOps/s	675.4237 KOps/s	$\color{#35bf28}+0.11\%$
test_creation_empty	32.8210μs	8.3112μs	120.3194 KOps/s	165.1243 KOps/s	$\textbf{\color{#d91a1a}-27.13\%}$
test_creation_nested_1	1.7595ms	9.7785μs	102.2655 KOps/s	131.4359 KOps/s	$\textbf{\color{#d91a1a}-22.19\%}$
test_creation_nested_2	37.4910μs	12.5219μs	79.8600 KOps/s	99.6338 KOps/s	$\textbf{\color{#d91a1a}-19.85\%}$
test_clone	0.1060ms	11.0072μs	90.8492 KOps/s	96.2708 KOps/s	$\textbf{\color{#d91a1a}-5.63\%}$
test_getitem[int]	1.2943ms	10.7736μs	92.8199 KOps/s	94.8147 KOps/s	$\color{#d91a1a}-2.10\%$
test_getitem[slice_int]	94.3627ms	30.1494μs	33.1682 KOps/s	49.6787 KOps/s	$\textbf{\color{#d91a1a}-33.23\%}$
test_getitem[range]	0.1624ms	36.6095μs	27.3153 KOps/s	27.6115 KOps/s	$\color{#d91a1a}-1.07\%$
test_getitem[tuple]	0.1056ms	18.5402μs	53.9369 KOps/s	56.2152 KOps/s	$\color{#d91a1a}-4.05\%$
test_getitem[list]	0.2541ms	34.9370μs	28.6229 KOps/s	31.3866 KOps/s	$\textbf{\color{#d91a1a}-8.81\%}$
test_setitem_dim[int]	40.9810μs	19.7504μs	50.6318 KOps/s	54.9696 KOps/s	$\textbf{\color{#d91a1a}-7.89\%}$
test_setitem_dim[slice_int]	63.7420μs	40.3749μs	24.7679 KOps/s	26.8818 KOps/s	$\textbf{\color{#d91a1a}-7.86\%}$
test_setitem_dim[range]	0.1593ms	56.9013μs	17.5743 KOps/s	19.2895 KOps/s	$\textbf{\color{#d91a1a}-8.89\%}$
test_setitem_dim[tuple]	73.4010μs	34.2807μs	29.1710 KOps/s	31.8396 KOps/s	$\textbf{\color{#d91a1a}-8.38\%}$
test_setitem	0.1566ms	15.9568μs	62.6692 KOps/s	73.0623 KOps/s	$\textbf{\color{#d91a1a}-14.22\%}$
test_set	0.1516ms	15.5112μs	64.4694 KOps/s	75.6511 KOps/s	$\textbf{\color{#d91a1a}-14.78\%}$
test_set_shared	1.5257ms	0.1471ms	6.8004 KOps/s	6.8964 KOps/s	$\color{#d91a1a}-1.39\%$
test_update	0.3608ms	18.3622μs	54.4596 KOps/s	66.4583 KOps/s	$\textbf{\color{#d91a1a}-18.05\%}$
test_update_nested	0.1394ms	23.3547μs	42.8180 KOps/s	49.7052 KOps/s	$\textbf{\color{#d91a1a}-13.86\%}$
test_update__nested	1.0986ms	24.8512μs	40.2396 KOps/s	41.5833 KOps/s	$\color{#d91a1a}-3.23\%$
test_set_nested	0.1142ms	16.6491μs	60.0631 KOps/s	68.7662 KOps/s	$\textbf{\color{#d91a1a}-12.66\%}$
test_set_nested_new	0.1622ms	18.9506μs	52.7688 KOps/s	60.1863 KOps/s	$\textbf{\color{#d91a1a}-12.32\%}$
test_select	0.1399ms	31.6094μs	31.6361 KOps/s	34.2043 KOps/s	$\textbf{\color{#d91a1a}-7.51\%}$
test_select_nested	78.9720μs	42.4660μs	23.5482 KOps/s	23.8535 KOps/s	$\color{#d91a1a}-1.28\%$
test_exclude_nested	0.3731ms	62.6698μs	15.9566 KOps/s	16.2605 KOps/s	$\color{#d91a1a}-1.87\%$
test_empty[True]	0.3392ms	0.2789ms	3.5856 KOps/s	3.6370 KOps/s	$\color{#d91a1a}-1.41\%$
test_empty[False]	3.6161μs	0.7559μs	1.3229 MOps/s	1.3599 MOps/s	$\color{#d91a1a}-2.72\%$
test_to	85.5020μs	53.4486μs	18.7096 KOps/s	18.5422 KOps/s	$\color{#35bf28}+0.90\%$
test_to_nonblocking	0.1981ms	45.9317μs	21.7714 KOps/s	21.9947 KOps/s	$\color{#d91a1a}-1.02\%$
test_unbind_speed	0.2610ms	0.2306ms	4.3367 KOps/s	4.3968 KOps/s	$\color{#d91a1a}-1.37\%$
test_unbind_speed_stack0	0.3429ms	0.2346ms	4.2624 KOps/s	4.3402 KOps/s	$\color{#d91a1a}-1.79\%$
test_unbind_speed_stack1	93.6029ms	0.6515ms	1.5349 KOps/s	1.5510 KOps/s	$\color{#d91a1a}-1.04\%$
test_split	95.1344ms	1.7192ms	581.6725 Ops/s	592.9489 Ops/s	$\color{#d91a1a}-1.90\%$
test_chunk	1.5610ms	1.4470ms	691.0719 Ops/s	633.1177 Ops/s	$\textbf{\color{#35bf28}+9.15\%}$
test_consolidate[False-None]	97.5196ms	2.8346ms	352.7839 Ops/s	384.8231 Ops/s	$\textbf{\color{#d91a1a}-8.33\%}$
test_consolidate[default-None]	1.8197ms	1.6753ms	596.9143 Ops/s	607.9256 Ops/s	$\color{#d91a1a}-1.81\%$
test_consolidate[reduce-overhead-None]	1.9615ms	1.7067ms	585.9161 Ops/s	601.3920 Ops/s	$\color{#d91a1a}-2.57\%$
test_consolidate_njt[False-None]	6.6486ms	6.4360ms	155.3748 Ops/s	154.0013 Ops/s	$\color{#35bf28}+0.89\%$
test_to[False-False-None]	1.8606ms	1.6415ms	609.1873 Ops/s	611.4548 Ops/s	$\color{#d91a1a}-0.37\%$
test_to[True-False-None]	1.5305ms	1.2742ms	784.8116 Ops/s	813.5842 Ops/s	$\color{#d91a1a}-3.54\%$
test_to[within-False-None]	4.2501ms	3.9975ms	250.1591 Ops/s	254.4441 Ops/s	$\color{#d91a1a}-1.68\%$
test_to[True-default-None]	5.4032ms	5.2604ms	190.1014 Ops/s	191.5795 Ops/s	$\color{#d91a1a}-0.77\%$
test_to_njt[False-False-None]	7.3286ms	7.1066ms	140.7135 Ops/s	143.9985 Ops/s	$\color{#d91a1a}-2.28\%$
test_to_njt[True-False-None]	5.9748ms	5.7448ms	174.0690 Ops/s	179.8099 Ops/s	$\color{#d91a1a}-3.19\%$
test_to_njt[within-False-None]	12.2321ms	12.0004ms	83.3307 Ops/s	83.0075 Ops/s	$\color{#35bf28}+0.39\%$
test_creation[device0]	0.4420ms	83.0846μs	12.0359 KOps/s	12.4186 KOps/s	$\color{#d91a1a}-3.08\%$
test_creation_from_tensor	0.6111ms	85.4006μs	11.7095 KOps/s	11.7798 KOps/s	$\color{#d91a1a}-0.60\%$
test_add_one[memmap_tensor0]	0.3249ms	6.8847μs	145.2492 KOps/s	152.4422 KOps/s	$\color{#d91a1a}-4.72\%$
test_contiguous[memmap_tensor0]	1.8205μs	0.4000μs	2.5000 MOps/s	2.4461 MOps/s	$\color{#35bf28}+2.20\%$
test_stack[memmap_tensor0]	42.1010μs	4.4198μs	226.2558 KOps/s	228.0005 KOps/s	$\color{#d91a1a}-0.77\%$
test_memmaptd_index	1.8237ms	0.2454ms	4.0754 KOps/s	4.0688 KOps/s	$\color{#35bf28}+0.16\%$
test_memmaptd_index_astensor	0.6047ms	0.3024ms	3.3072 KOps/s	3.2974 KOps/s	$\color{#35bf28}+0.30\%$
test_memmaptd_index_op	1.0121ms	0.5854ms	1.7083 KOps/s	1.8082 KOps/s	$\textbf{\color{#d91a1a}-5.52\%}$
test_serialize_model	0.4149s	0.1712s	5.8424 Ops/s	7.6392 Ops/s	$\textbf{\color{#d91a1a}-23.52\%}$
test_serialize_model_pickle	1.3504s	1.2171s	0.8216 Ops/s	0.8239 Ops/s	$\color{#d91a1a}-0.28\%$
test_serialize_weights	0.1302s	0.1292s	7.7390 Ops/s	7.7231 Ops/s	$\color{#35bf28}+0.20\%$
test_serialize_weights_returnearly	0.3178s	52.8066ms	18.9370 Ops/s	23.5690 Ops/s	$\textbf{\color{#d91a1a}-19.65\%}$
test_serialize_weights_pickle	1.3670s	1.2246s	0.8166 Ops/s	0.8196 Ops/s	$\color{#d91a1a}-0.37\%$
test_reshape_pytree	0.1659ms	22.0507μs	45.3500 KOps/s	44.7411 KOps/s	$\color{#35bf28}+1.36\%$
test_reshape_td	0.1550ms	26.5026μs	37.7322 KOps/s	38.0033 KOps/s	$\color{#d91a1a}-0.71\%$
test_view_pytree	0.1476ms	21.7448μs	45.9880 KOps/s	44.9435 KOps/s	$\color{#35bf28}+2.32\%$
test_view_td	0.1176ms	30.1658μs	33.1501 KOps/s	33.3866 KOps/s	$\color{#d91a1a}-0.71\%$
test_unbind_pytree	0.1448ms	28.0275μs	35.6793 KOps/s	35.7743 KOps/s	$\color{#d91a1a}-0.27\%$
test_unbind_td	0.7867ms	36.0986μs	27.7019 KOps/s	27.9938 KOps/s	$\color{#d91a1a}-1.04\%$
test_split_pytree	0.1549ms	29.8576μs	33.4923 KOps/s	33.0619 KOps/s	$\color{#35bf28}+1.30\%$
test_split_td	0.9944ms	38.4470μs	26.0099 KOps/s	25.7446 KOps/s	$\color{#35bf28}+1.03\%$
test_add_pytree	0.1380ms	34.8385μs	28.7039 KOps/s	29.0986 KOps/s	$\color{#d91a1a}-1.36\%$
test_add_td	0.2269ms	52.0380μs	19.2167 KOps/s	22.9815 KOps/s	$\textbf{\color{#d91a1a}-16.38\%}$
test_compile_add_one_nested[tensordict-compile]	0.2598ms	0.1211ms	8.2576 KOps/s	7.6516 KOps/s	$\textbf{\color{#35bf28}+7.92\%}$
test_compile_add_one_nested[tensordict-eager]	0.2763ms	0.1267ms	7.8952 KOps/s	8.0138 KOps/s	$\color{#d91a1a}-1.48\%$
test_compile_add_one_nested[pytree-compile]	0.2460ms	95.9315μs	10.4241 KOps/s	10.3261 KOps/s	$\color{#35bf28}+0.95\%$
test_compile_add_one_nested[pytree-eager]	0.3036ms	0.1486ms	6.7282 KOps/s	6.8882 KOps/s	$\color{#d91a1a}-2.32\%$
test_compile_copy_nested[tensordict-compile]	0.1656ms	24.2520μs	41.2337 KOps/s	44.2945 KOps/s	$\textbf{\color{#d91a1a}-6.91\%}$
test_compile_copy_nested[tensordict-eager]	0.1575ms	26.3295μs	37.9803 KOps/s	37.3403 KOps/s	$\color{#35bf28}+1.71\%$
test_compile_copy_nested[pytree-compile]	0.4409ms	64.2791μs	15.5572 KOps/s	15.3652 KOps/s	$\color{#35bf28}+1.25\%$
test_compile_copy_nested[pytree-eager]	93.6420μs	49.6398μs	20.1451 KOps/s	20.0310 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_add_one_flat[tensordict-compile]	0.5505ms	0.1431ms	6.9872 KOps/s	7.1525 KOps/s	$\color{#d91a1a}-2.31\%$
test_compile_add_one_flat[tensordict-eager]	0.6383ms	0.2110ms	4.7399 KOps/s	4.7941 KOps/s	$\color{#d91a1a}-1.13\%$
test_compile_add_one_flat[tensorclass-compile]	0.2578ms	97.7025μs	10.2352 KOps/s	10.1008 KOps/s	$\color{#35bf28}+1.33\%$
test_compile_add_one_flat[tensorclass-eager]	0.2082ms	52.1853μs	19.1625 KOps/s	19.2634 KOps/s	$\color{#d91a1a}-0.52\%$
test_compile_add_one_flat[pytree-compile]	0.5200ms	0.1349ms	7.4109 KOps/s	7.4575 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_add_one_flat[pytree-eager]	0.8454ms	0.4766ms	2.0982 KOps/s	2.1390 KOps/s	$\color{#d91a1a}-1.91\%$
test_compile_add_self_flat[tensordict-eager]	0.4272ms	0.2521ms	3.9672 KOps/s	4.0527 KOps/s	$\color{#d91a1a}-2.11\%$
test_compile_add_self_flat[tensordict-compile]	0.3063ms	0.1500ms	6.6684 KOps/s	7.0646 KOps/s	$\textbf{\color{#d91a1a}-5.61\%}$
test_compile_add_self_flat[tensorclass-eager]	0.2532ms	62.7456μs	15.9374 KOps/s	16.4594 KOps/s	$\color{#d91a1a}-3.17\%$
test_compile_add_self_flat[tensorclass-compile]	0.2790ms	0.1009ms	9.9101 KOps/s	10.3736 KOps/s	$\color{#d91a1a}-4.47\%$
test_compile_add_self_flat[pytree-eager]	0.7656ms	0.3985ms	2.5093 KOps/s	2.5504 KOps/s	$\color{#d91a1a}-1.61\%$
test_compile_add_self_flat[pytree-compile]	0.2744ms	0.1356ms	7.3752 KOps/s	7.4951 KOps/s	$\color{#d91a1a}-1.60\%$
test_compile_copy_flat[tensordict-compile]	0.2128ms	21.5743μs	46.3514 KOps/s	54.9703 KOps/s	$\textbf{\color{#d91a1a}-15.68\%}$
test_compile_copy_flat[tensordict-eager]	97.5820μs	26.2686μs	38.0683 KOps/s	37.4626 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_copy_flat[pytree-compile]	0.1826ms	69.0604μs	14.4801 KOps/s	14.4462 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_copy_flat[pytree-eager]	0.1046ms	51.3201μs	19.4855 KOps/s	19.4447 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_assign_and_add[tensordict-compile]	1.6011ms	0.3887ms	2.5728 KOps/s	2.2799 KOps/s	$\textbf{\color{#35bf28}+12.85\%}$
test_compile_assign_and_add[tensordict-eager]	2.7954ms	2.5678ms	389.4420 Ops/s	379.8061 Ops/s	$\color{#35bf28}+2.54\%$
test_compile_assign_and_add[pytree-compile]	1.5802ms	0.4337ms	2.3059 KOps/s	2.3072 KOps/s	$\color{#d91a1a}-0.06\%$
test_compile_assign_and_add[pytree-eager]	2.8259ms	2.5573ms	391.0323 Ops/s	397.3517 Ops/s	$\color{#d91a1a}-1.59\%$
test_compile_indexing[tensor-tensordict-compile]	0.3701ms	0.1129ms	8.8573 KOps/s	8.8037 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_indexing[tensor-tensordict-eager]	0.5870ms	77.9194μs	12.8338 KOps/s	13.0526 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5352ms	0.1061ms	9.4268 KOps/s	9.7782 KOps/s	$\color{#d91a1a}-3.59\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2573ms	71.2560μs	14.0339 KOps/s	15.1333 KOps/s	$\textbf{\color{#d91a1a}-7.26\%}$
test_compile_indexing[tensor-pytree-compile]	0.2891ms	0.1111ms	8.9970 KOps/s	9.6938 KOps/s	$\textbf{\color{#d91a1a}-7.19\%}$
test_compile_indexing[tensor-pytree-eager]	0.2609ms	70.6369μs	14.1569 KOps/s	14.9575 KOps/s	$\textbf{\color{#d91a1a}-5.35\%}$
test_compile_indexing[slice-tensordict-compile]	0.2506ms	99.8020μs	10.0198 KOps/s	10.1596 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_indexing[slice-tensordict-eager]	0.1573ms	17.0621μs	58.6095 KOps/s	58.3262 KOps/s	$\color{#35bf28}+0.49\%$
test_compile_indexing[slice-tensorclass-compile]	0.2571ms	95.3688μs	10.4856 KOps/s	10.6571 KOps/s	$\color{#d91a1a}-1.61\%$
test_compile_indexing[slice-tensorclass-eager]	0.1505ms	15.8286μs	63.1769 KOps/s	63.4140 KOps/s	$\color{#d91a1a}-0.37\%$
test_compile_indexing[slice-pytree-compile]	0.3020ms	95.5092μs	10.4702 KOps/s	10.5472 KOps/s	$\color{#d91a1a}-0.73\%$
test_compile_indexing[slice-pytree-eager]	0.2159ms	15.8129μs	63.2395 KOps/s	63.3615 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_indexing[int-tensordict-compile]	0.2832ms	0.1016ms	9.8395 KOps/s	10.1296 KOps/s	$\color{#d91a1a}-2.86\%$
test_compile_indexing[int-tensordict-eager]	0.6503ms	16.8118μs	59.4819 KOps/s	57.7602 KOps/s	$\color{#35bf28}+2.98\%$
test_compile_indexing[int-tensorclass-compile]	0.2992ms	0.1008ms	9.9226 KOps/s	10.5514 KOps/s	$\textbf{\color{#d91a1a}-5.96\%}$
test_compile_indexing[int-tensorclass-eager]	0.1535ms	15.6218μs	64.0130 KOps/s	63.1865 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_indexing[int-pytree-compile]	0.2954ms	95.3312μs	10.4897 KOps/s	10.5850 KOps/s	$\color{#d91a1a}-0.90\%$
test_compile_indexing[int-pytree-eager]	0.1356ms	15.7392μs	63.5357 KOps/s	63.6314 KOps/s	$\color{#d91a1a}-0.15\%$
test_mod_add[eager]	0.1885ms	37.6430μs	26.5653 KOps/s	28.4673 KOps/s	$\textbf{\color{#d91a1a}-6.68\%}$
test_mod_add[compile]	0.3781ms	79.3550μs	12.6016 KOps/s	12.5653 KOps/s	$\color{#35bf28}+0.29\%$
test_mod_add[compile-overhead]	0.3187ms	0.1650ms	6.0596 KOps/s	5.7915 KOps/s	$\color{#35bf28}+4.63\%$
test_mod_wrap[eager]	0.3904ms	0.2437ms	4.1034 KOps/s	4.1456 KOps/s	$\color{#d91a1a}-1.02\%$
test_mod_wrap[compile]	0.4256ms	0.2773ms	3.6063 KOps/s	3.5473 KOps/s	$\color{#35bf28}+1.66\%$
test_mod_wrap[compile-overhead]	7.1049ms	3.6844ms	271.4173 Ops/s	263.2948 Ops/s	$\color{#35bf28}+3.08\%$
test_mod_wrap_and_backward[eager]	1.5426ms	1.3494ms	741.0492 Ops/s	696.7550 Ops/s	$\textbf{\color{#35bf28}+6.36\%}$
test_mod_wrap_and_backward[compile]	1.4587ms	1.2627ms	791.9411 Ops/s	737.7342 Ops/s	$\textbf{\color{#35bf28}+7.35\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3821ms	0.9178ms	1.0895 KOps/s	951.4185 Ops/s	$\textbf{\color{#35bf28}+14.52\%}$
test_seq_add[eager]	0.2945ms	0.1172ms	8.5344 KOps/s	9.2682 KOps/s	$\textbf{\color{#d91a1a}-7.92\%}$
test_seq_add[compile]	0.2693ms	92.3448μs	10.8290 KOps/s	10.8471 KOps/s	$\color{#d91a1a}-0.17\%$
test_seq_add[compile-overhead]	0.3161ms	0.1340ms	7.4642 KOps/s	7.8452 KOps/s	$\color{#d91a1a}-4.86\%$
test_seq_wrap[eager]	0.6240ms	0.4290ms	2.3309 KOps/s	2.4714 KOps/s	$\textbf{\color{#d91a1a}-5.69\%}$
test_seq_wrap[compile]	0.5205ms	0.3157ms	3.1681 KOps/s	3.3774 KOps/s	$\textbf{\color{#d91a1a}-6.20\%}$
test_seq_wrap[compile-overhead]	0.3667ms	0.2218ms	4.5088 KOps/s	4.4900 KOps/s	$\color{#35bf28}+0.42\%$
test_func_call_runtime[False-eager]	0.8758ms	0.7269ms	1.3757 KOps/s	1.3902 KOps/s	$\color{#d91a1a}-1.04\%$
test_func_call_runtime[False-compile]	0.9708ms	0.7568ms	1.3213 KOps/s	1.3461 KOps/s	$\color{#d91a1a}-1.84\%$
test_func_call_runtime[False-compile-overhead]	0.5046ms	0.3632ms	2.7536 KOps/s	2.7889 KOps/s	$\color{#d91a1a}-1.27\%$
test_func_call_runtime[True-eager]	1.0554ms	0.8848ms	1.1302 KOps/s	1.1232 KOps/s	$\color{#35bf28}+0.62\%$
test_func_call_runtime[True-compile]	0.9287ms	0.7536ms	1.3270 KOps/s	1.3305 KOps/s	$\color{#d91a1a}-0.27\%$
test_func_call_runtime[True-compile-overhead]	0.5726ms	0.3831ms	2.6100 KOps/s	2.6423 KOps/s	$\color{#d91a1a}-1.22\%$
test_func_call_cm_runtime[False-eager]	0.8740ms	0.7217ms	1.3855 KOps/s	1.3867 KOps/s	$\color{#d91a1a}-0.08\%$
test_func_call_cm_runtime[False-compile]	0.8999ms	0.7399ms	1.3515 KOps/s	1.3578 KOps/s	$\color{#d91a1a}-0.46\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5052ms	0.3602ms	2.7764 KOps/s	2.7564 KOps/s	$\color{#35bf28}+0.72\%$
test_func_call_cm_runtime[True-eager]	1.1510ms	0.9814ms	1.0190 KOps/s	1.0111 KOps/s	$\color{#35bf28}+0.79\%$
test_func_call_cm_runtime[True-compile]	0.9335ms	0.7846ms	1.2746 KOps/s	1.2833 KOps/s	$\color{#d91a1a}-0.68\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5495ms	0.4036ms	2.4775 KOps/s	2.4578 KOps/s	$\color{#35bf28}+0.80\%$
test_vmap_func_call_cm_runtime[eager]	2.5213ms	2.0448ms	489.0493 Ops/s	483.7951 Ops/s	$\color{#35bf28}+1.09\%$
test_vmap_func_call_cm_runtime[compile]	0.9862ms	0.8055ms	1.2415 KOps/s	1.2546 KOps/s	$\color{#d91a1a}-1.04\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5625ms	0.4061ms	2.4624 KOps/s	2.4553 KOps/s	$\color{#35bf28}+0.29\%$
test_distributed	4.2462ms	0.2906ms	3.4416 KOps/s	8.1819 KOps/s	$\textbf{\color{#d91a1a}-57.94\%}$
test_tdmodule	64.9010μs	19.2100μs	52.0562 KOps/s	53.3287 KOps/s	$\color{#d91a1a}-2.39\%$
test_tdmodule_dispatch	0.1595ms	34.7893μs	28.7444 KOps/s	30.8069 KOps/s	$\textbf{\color{#d91a1a}-6.69\%}$
test_tdseq	37.9410μs	18.6636μs	53.5804 KOps/s	57.1845 KOps/s	$\textbf{\color{#d91a1a}-6.30\%}$
test_tdseq_dispatch	0.1127ms	36.3180μs	27.5345 KOps/s	29.6530 KOps/s	$\textbf{\color{#d91a1a}-7.14\%}$
test_instantiation_functorch	1.7315ms	1.5320ms	652.7337 Ops/s	641.2741 Ops/s	$\color{#35bf28}+1.79\%$
test_exec_functorch	0.2853ms	0.1464ms	6.8290 KOps/s	7.0397 KOps/s	$\color{#d91a1a}-2.99\%$
test_exec_functional_call	0.2423ms	0.1390ms	7.1947 KOps/s	7.4337 KOps/s	$\color{#d91a1a}-3.22\%$
test_exec_td_decorator	0.3738ms	0.1858ms	5.3829 KOps/s	5.5439 KOps/s	$\color{#d91a1a}-2.90\%$
test_vmap_mlp_speed_decorator[True-True]	0.8397ms	0.6737ms	1.4844 KOps/s	1.4478 KOps/s	$\color{#35bf28}+2.53\%$
test_vmap_mlp_speed_decorator[True-False]	0.8469ms	0.6717ms	1.4887 KOps/s	1.4616 KOps/s	$\color{#35bf28}+1.86\%$
test_vmap_mlp_speed_decorator[False-True]	0.7489ms	0.5828ms	1.7158 KOps/s	1.7028 KOps/s	$\color{#35bf28}+0.76\%$
test_vmap_mlp_speed_decorator[False-False]	0.7843ms	0.5905ms	1.6936 KOps/s	1.6974 KOps/s	$\color{#d91a1a}-0.23\%$
test_vmap_transformer_speed_decorator[True-True]	19.1030ms	18.8868ms	52.9470 Ops/s	52.9716 Ops/s	$\color{#d91a1a}-0.05\%$
test_vmap_transformer_speed_decorator[True-False]	19.7197ms	19.0164ms	52.5863 Ops/s	52.9226 Ops/s	$\color{#d91a1a}-0.64\%$
test_vmap_transformer_speed_decorator[False-True]	19.0397ms	18.8330ms	53.0982 Ops/s	53.3530 Ops/s	$\color{#d91a1a}-0.48\%$
test_vmap_transformer_speed_decorator[False-False]	18.9847ms	18.7790ms	53.2509 Ops/s	53.3424 Ops/s	$\color{#d91a1a}-0.17\%$
test_to_module_speed[True]	1.0543ms	0.9441ms	1.0592 KOps/s	1.0546 KOps/s	$\color{#35bf28}+0.43\%$
test_to_module_speed[False]	1.3027ms	0.9329ms	1.0719 KOps/s	1.0720 KOps/s	$\color{#d91a1a}-0.01\%$
test_tc_init	0.1190ms	36.2921μs	27.5542 KOps/s	29.5447 KOps/s	$\textbf{\color{#d91a1a}-6.74\%}$
test_tc_init_nested	0.1544ms	73.3899μs	13.6259 KOps/s	14.9013 KOps/s	$\textbf{\color{#d91a1a}-8.56\%}$
test_tc_first_layer_tensor	4.7286μs	0.7027μs	1.4231 MOps/s	1.4649 MOps/s	$\color{#d91a1a}-2.85\%$
test_tc_first_layer_nontensor	26.0010μs	2.3052μs	433.8070 KOps/s	430.6062 KOps/s	$\color{#35bf28}+0.74\%$
test_tc_second_layer_tensor	16.8955μs	1.4377μs	695.5535 KOps/s	714.6280 KOps/s	$\color{#d91a1a}-2.67\%$
test_tc_second_layer_nontensor	65.7120μs	3.0455μs	328.3540 KOps/s	326.8768 KOps/s	$\color{#35bf28}+0.45\%$
test_unbind	0.2209s	9.9300ms	100.7050 Ops/s	151.1369 Ops/s	$\textbf{\color{#d91a1a}-33.37\%}$
test_full_like	11.4193ms	9.5652ms	104.5454 Ops/s	105.2023 Ops/s	$\color{#d91a1a}-0.62\%$
test_zeros_like	5.3477ms	4.3647ms	229.1102 Ops/s	231.6464 Ops/s	$\color{#d91a1a}-1.09\%$
test_ones_like	4.9352ms	4.3703ms	228.8187 Ops/s	229.2322 Ops/s	$\color{#d91a1a}-0.18\%$
test_clone	11.9437ms	9.3713ms	106.7085 Ops/s	106.6670 Ops/s	$\color{#35bf28}+0.04\%$
test_squeeze	0.1491ms	9.2081μs	108.6001 KOps/s	108.4736 KOps/s	$\color{#35bf28}+0.12\%$
test_unsqueeze	0.2115ms	71.9543μs	13.8977 KOps/s	14.4467 KOps/s	$\color{#d91a1a}-3.80\%$
test_split	0.2979ms	0.1577ms	6.3399 KOps/s	6.3373 KOps/s	$\color{#35bf28}+0.04\%$
test_permute	0.3030ms	0.1790ms	5.5857 KOps/s	5.7446 KOps/s	$\color{#d91a1a}-2.77\%$
test_stack	51.5214ms	51.0742ms	19.5794 Ops/s	19.4460 Ops/s	$\color{#35bf28}+0.69\%$
test_cat	51.6775ms	50.9127ms	19.6415 Ops/s	19.4921 Ops/s	$\color{#35bf28}+0.77\%$