[Feature] Plotting TensorDictSequential graphs

[vmoens]

Stack from ghstack (oldest at bottom):

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

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}29$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	39.8540μs	21.2707μs	47.0129 KOps/s	51.3039 KOps/s	$\textbf{\color{#d91a1a}-8.36\%}$
test_plain_set_stack_nested	78.1460μs	21.4907μs	46.5318 KOps/s	50.1711 KOps/s	$\textbf{\color{#d91a1a}-7.25\%}$
test_plain_set_nested_inplace	64.8010μs	23.3747μs	42.7813 KOps/s	45.9256 KOps/s	$\textbf{\color{#d91a1a}-6.85\%}$
test_plain_set_stack_nested_inplace	80.8110μs	23.2030μs	43.0979 KOps/s	46.3982 KOps/s	$\textbf{\color{#d91a1a}-7.11\%}$
test_items	42.1480μs	4.1195μs	242.7505 KOps/s	237.0291 KOps/s	$\color{#35bf28}+2.41\%$
test_items_nested	0.7660ms	0.4118ms	2.4285 KOps/s	2.4528 KOps/s	$\color{#d91a1a}-0.99\%$
test_items_nested_locked	0.5981ms	0.4128ms	2.4226 KOps/s	2.4644 KOps/s	$\color{#d91a1a}-1.69\%$
test_items_nested_leaf	0.1463ms	76.4647μs	13.0779 KOps/s	12.8727 KOps/s	$\color{#35bf28}+1.59\%$
test_items_stack_nested	0.6066ms	0.4138ms	2.4168 KOps/s	2.4293 KOps/s	$\color{#d91a1a}-0.51\%$
test_items_stack_nested_leaf	0.1508ms	79.3773μs	12.5981 KOps/s	12.4031 KOps/s	$\color{#35bf28}+1.57\%$
test_items_stack_nested_locked	0.5594ms	0.4144ms	2.4132 KOps/s	2.4338 KOps/s	$\color{#d91a1a}-0.84\%$
test_keys	41.2070μs	3.5498μs	281.7097 KOps/s	284.6820 KOps/s	$\color{#d91a1a}-1.04\%$
test_keys_nested	0.2766ms	0.1691ms	5.9121 KOps/s	6.0780 KOps/s	$\color{#d91a1a}-2.73\%$
test_keys_nested_locked	1.8430ms	0.1749ms	5.7177 KOps/s	5.8308 KOps/s	$\color{#d91a1a}-1.94\%$
test_keys_nested_leaf	0.2659ms	0.1469ms	6.8066 KOps/s	7.0099 KOps/s	$\color{#d91a1a}-2.90\%$
test_keys_stack_nested	0.2666ms	0.1658ms	6.0331 KOps/s	6.1684 KOps/s	$\color{#d91a1a}-2.19\%$
test_keys_stack_nested_leaf	0.2097ms	0.1452ms	6.8848 KOps/s	7.1286 KOps/s	$\color{#d91a1a}-3.42\%$
test_keys_stack_nested_locked	0.2854ms	0.1713ms	5.8377 KOps/s	5.9567 KOps/s	$\color{#d91a1a}-2.00\%$
test_values	9.1710μs	1.0528μs	949.8610 KOps/s	970.4251 KOps/s	$\color{#d91a1a}-2.12\%$
test_values_nested	0.1244ms	63.6554μs	15.7096 KOps/s	16.1737 KOps/s	$\color{#d91a1a}-2.87\%$
test_values_nested_locked	0.1215ms	63.3492μs	15.7855 KOps/s	16.0319 KOps/s	$\color{#d91a1a}-1.54\%$
test_values_nested_leaf	0.1288ms	72.5652μs	13.7807 KOps/s	13.0599 KOps/s	$\textbf{\color{#35bf28}+5.52\%}$
test_values_stack_nested	0.1248ms	64.1270μs	15.5940 KOps/s	15.9301 KOps/s	$\color{#d91a1a}-2.11\%$
test_values_stack_nested_leaf	0.1312ms	72.5077μs	13.7916 KOps/s	13.9169 KOps/s	$\color{#d91a1a}-0.90\%$
test_values_stack_nested_locked	0.1124ms	63.4441μs	15.7619 KOps/s	15.9349 KOps/s	$\color{#d91a1a}-1.09\%$
test_membership	26.2590μs	0.8926μs	1.1203 MOps/s	1.1707 MOps/s	$\color{#d91a1a}-4.31\%$
test_membership_nested	22.6620μs	2.9026μs	344.5241 KOps/s	343.8021 KOps/s	$\color{#35bf28}+0.21\%$
test_membership_nested_leaf	27.7520μs	2.9322μs	341.0425 KOps/s	343.3421 KOps/s	$\color{#d91a1a}-0.67\%$
test_membership_stacked_nested	41.2370μs	2.8913μs	345.8708 KOps/s	346.2548 KOps/s	$\color{#d91a1a}-0.11\%$
test_membership_stacked_nested_leaf	27.3510μs	2.8870μs	346.3806 KOps/s	340.8560 KOps/s	$\color{#35bf28}+1.62\%$
test_membership_nested_last	48.7610μs	4.3118μs	231.9240 KOps/s	231.3038 KOps/s	$\color{#35bf28}+0.27\%$
test_membership_nested_leaf_last	24.5250μs	4.3299μs	230.9534 KOps/s	227.2274 KOps/s	$\color{#35bf28}+1.64\%$
test_membership_stacked_nested_last	25.1970μs	4.2979μs	232.6722 KOps/s	194.5354 KOps/s	$\textbf{\color{#35bf28}+19.60\%}$
test_membership_stacked_nested_leaf_last	48.7510μs	4.3214μs	231.4073 KOps/s	196.2936 KOps/s	$\textbf{\color{#35bf28}+17.89\%}$
test_nested_getleaf	55.2540μs	10.6543μs	93.8592 KOps/s	91.6515 KOps/s	$\color{#35bf28}+2.41\%$
test_nested_get	35.3460μs	10.1510μs	98.5129 KOps/s	96.8730 KOps/s	$\color{#35bf28}+1.69\%$
test_stacked_getleaf	60.8830μs	10.5540μs	94.7505 KOps/s	92.9374 KOps/s	$\color{#35bf28}+1.95\%$
test_stacked_get	52.3170μs	10.0078μs	99.9219 KOps/s	97.3594 KOps/s	$\color{#35bf28}+2.63\%$
test_nested_getitemleaf	40.4460μs	11.2901μs	88.5733 KOps/s	88.5344 KOps/s	$\color{#35bf28}+0.04\%$
test_nested_getitem	59.2200μs	10.4409μs	95.7770 KOps/s	94.8550 KOps/s	$\color{#35bf28}+0.97\%$
test_stacked_getitemleaf	56.2250μs	11.3751μs	87.9113 KOps/s	88.3242 KOps/s	$\color{#d91a1a}-0.47\%$
test_stacked_getitem	31.9190μs	10.4718μs	95.4947 KOps/s	95.1541 KOps/s	$\color{#35bf28}+0.36\%$
test_lock_nested	4.6632ms	0.4664ms	2.1443 KOps/s	2.1824 KOps/s	$\color{#d91a1a}-1.74\%$
test_lock_stack_nested	0.8984ms	0.4286ms	2.3329 KOps/s	2.3686 KOps/s	$\color{#d91a1a}-1.50\%$
test_unlock_nested	0.6858ms	0.3752ms	2.6649 KOps/s	2.6657 KOps/s	$\color{#d91a1a}-0.03\%$
test_unlock_stack_nested	0.6557ms	0.3457ms	2.8930 KOps/s	2.9357 KOps/s	$\color{#d91a1a}-1.45\%$
test_flatten_speed	0.1963ms	0.1006ms	9.9395 KOps/s	9.8732 KOps/s	$\color{#35bf28}+0.67\%$
test_unflatten_speed	0.9261ms	0.5155ms	1.9398 KOps/s	1.9073 KOps/s	$\color{#35bf28}+1.70\%$
test_common_ops	1.6131ms	0.8140ms	1.2285 KOps/s	1.3105 KOps/s	$\textbf{\color{#d91a1a}-6.26\%}$
test_creation	22.5530μs	2.5047μs	399.2526 KOps/s	404.4421 KOps/s	$\color{#d91a1a}-1.28\%$
test_creation_empty	57.5280μs	12.8022μs	78.1114 KOps/s	105.0911 KOps/s	$\textbf{\color{#d91a1a}-25.67\%}$
test_creation_nested_1	56.4150μs	15.6849μs	63.7557 KOps/s	80.9666 KOps/s	$\textbf{\color{#d91a1a}-21.26\%}$
test_creation_nested_2	45.2440μs	20.5454μs	48.6728 KOps/s	58.7193 KOps/s	$\textbf{\color{#d91a1a}-17.11\%}$
test_clone	76.0620μs	13.7746μs	72.5973 KOps/s	74.0621 KOps/s	$\color{#d91a1a}-1.98\%$
test_getitem[int]	0.8363ms	12.9223μs	77.3855 KOps/s	77.7931 KOps/s	$\color{#d91a1a}-0.52\%$
test_getitem[slice_int]	0.1369ms	24.6341μs	40.5941 KOps/s	41.0457 KOps/s	$\color{#d91a1a}-1.10\%$
test_getitem[range]	0.2092ms	50.2676μs	19.8935 KOps/s	19.8858 KOps/s	$\color{#35bf28}+0.04\%$
test_getitem[tuple]	0.1298ms	20.4085μs	48.9992 KOps/s	49.5207 KOps/s	$\color{#d91a1a}-1.05\%$
test_getitem[list]	0.1840ms	46.7339μs	21.3977 KOps/s	22.1591 KOps/s	$\color{#d91a1a}-3.44\%$
test_setitem_dim[int]	57.1070μs	25.7408μs	38.8489 KOps/s	38.7876 KOps/s	$\color{#35bf28}+0.16\%$
test_setitem_dim[slice_int]	95.7890μs	52.6775μs	18.9834 KOps/s	19.1554 KOps/s	$\color{#d91a1a}-0.90\%$
test_setitem_dim[range]	0.1245ms	74.4414μs	13.4334 KOps/s	13.4159 KOps/s	$\color{#35bf28}+0.13\%$
test_setitem_dim[tuple]	0.1009ms	42.9957μs	23.2582 KOps/s	24.2896 KOps/s	$\color{#d91a1a}-4.25\%$
test_setitem	83.4960μs	21.2813μs	46.9896 KOps/s	51.4188 KOps/s	$\textbf{\color{#d91a1a}-8.61\%}$
test_set	71.0820μs	20.5775μs	48.5968 KOps/s	53.2569 KOps/s	$\textbf{\color{#d91a1a}-8.75\%}$
test_set_shared	3.2873ms	0.1752ms	5.7092 KOps/s	5.7990 KOps/s	$\color{#d91a1a}-1.55\%$
test_update	0.8085ms	23.9800μs	41.7015 KOps/s	49.1056 KOps/s	$\textbf{\color{#d91a1a}-15.08\%}$
test_update_nested	0.1571ms	34.3019μs	29.1529 KOps/s	31.0525 KOps/s	$\textbf{\color{#d91a1a}-6.12\%}$
test_update__nested	0.2341ms	34.7552μs	28.7727 KOps/s	28.0271 KOps/s	$\color{#35bf28}+2.66\%$
test_set_nested	84.4170μs	23.1946μs	43.1135 KOps/s	46.6390 KOps/s	$\textbf{\color{#d91a1a}-7.56\%}$
test_set_nested_new	95.6080μs	28.2170μs	35.4396 KOps/s	37.6005 KOps/s	$\textbf{\color{#d91a1a}-5.75\%}$
test_select	0.2219ms	44.5076μs	22.4681 KOps/s	23.6392 KOps/s	$\color{#d91a1a}-4.95\%$
test_select_nested	0.1283ms	62.2756μs	16.0577 KOps/s	16.0588 KOps/s	$-0.01\%$
test_exclude_nested	0.1515ms	81.8531μs	12.2170 KOps/s	12.2350 KOps/s	$\color{#d91a1a}-0.15\%$
test_empty[True]	0.7119ms	0.4224ms	2.3673 KOps/s	2.4200 KOps/s	$\color{#d91a1a}-2.18\%$
test_empty[False]	7.2260μs	1.3663μs	731.9137 KOps/s	722.7092 KOps/s	$\color{#35bf28}+1.27\%$
test_unbind_speed	0.4031ms	0.2745ms	3.6423 KOps/s	3.6876 KOps/s	$\color{#d91a1a}-1.23\%$
test_unbind_speed_stack0	0.3705ms	0.2671ms	3.7441 KOps/s	3.8145 KOps/s	$\color{#d91a1a}-1.84\%$
test_unbind_speed_stack1	0.1044s	0.7845ms	1.2747 KOps/s	1.3972 KOps/s	$\textbf{\color{#d91a1a}-8.76\%}$
test_split	93.9314ms	1.7538ms	570.1790 Ops/s	562.4531 Ops/s	$\color{#35bf28}+1.37\%$
test_chunk	1.8715ms	1.6098ms	621.2000 Ops/s	569.3409 Ops/s	$\textbf{\color{#35bf28}+9.11\%}$
test_consolidate_njt[False-None]	0.1028s	8.8192ms	113.3895 Ops/s	120.0041 Ops/s	$\textbf{\color{#d91a1a}-5.51\%}$
test_creation[device0]	0.2213ms	93.0396μs	10.7481 KOps/s	10.9474 KOps/s	$\color{#d91a1a}-1.82\%$
test_creation_from_tensor	3.7271ms	95.8824μs	10.4294 KOps/s	10.4599 KOps/s	$\color{#d91a1a}-0.29\%$
test_add_one[memmap_tensor0]	0.1343ms	5.0395μs	198.4333 KOps/s	194.2638 KOps/s	$\color{#35bf28}+2.15\%$
test_contiguous[memmap_tensor0]	18.3140μs	0.5305μs	1.8849 MOps/s	1.9502 MOps/s	$\color{#d91a1a}-3.35\%$
test_stack[memmap_tensor0]	31.3180μs	3.4570μs	289.2674 KOps/s	296.1339 KOps/s	$\color{#d91a1a}-2.32\%$
test_memmaptd_index	1.0322ms	0.2457ms	4.0701 KOps/s	4.1978 KOps/s	$\color{#d91a1a}-3.04\%$
test_memmaptd_index_astensor	0.5948ms	0.3349ms	2.9860 KOps/s	3.0809 KOps/s	$\color{#d91a1a}-3.08\%$
test_memmaptd_index_op	0.9681ms	0.6185ms	1.6169 KOps/s	1.7515 KOps/s	$\textbf{\color{#d91a1a}-7.69\%}$
test_serialize_model	0.1251s	0.1155s	8.6578 Ops/s	8.7976 Ops/s	$\color{#d91a1a}-1.59\%$
test_serialize_model_pickle	0.4612s	0.3910s	2.5578 Ops/s	2.5199 Ops/s	$\color{#35bf28}+1.50\%$
test_serialize_weights	0.1243s	0.1129s	8.8569 Ops/s	8.8145 Ops/s	$\color{#35bf28}+0.48\%$
test_serialize_weights_returnearly	0.2719s	0.1821s	5.4902 Ops/s	6.2615 Ops/s	$\textbf{\color{#d91a1a}-12.32\%}$
test_serialize_weights_pickle	0.6157s	0.4405s	2.2702 Ops/s	2.2623 Ops/s	$\color{#35bf28}+0.35\%$
test_serialize_weights_filesystem	0.1397s	0.1374s	7.2781 Ops/s	7.0871 Ops/s	$\color{#35bf28}+2.69\%$
test_serialize_model_filesystem	0.1593s	0.1460s	6.8479 Ops/s	6.1689 Ops/s	$\textbf{\color{#35bf28}+11.01\%}$
test_reshape_pytree	68.1870μs	26.6009μs	37.5927 KOps/s	38.3474 KOps/s	$\color{#d91a1a}-1.97\%$
test_reshape_td	68.2270μs	32.6734μs	30.6060 KOps/s	30.5358 KOps/s	$\color{#35bf28}+0.23\%$
test_view_pytree	69.6260μs	26.7072μs	37.4431 KOps/s	38.1725 KOps/s	$\color{#d91a1a}-1.91\%$
test_view_td	94.7970μs	38.3073μs	26.1047 KOps/s	25.6243 KOps/s	$\color{#35bf28}+1.87\%$
test_unbind_pytree	63.1880μs	29.7893μs	33.5691 KOps/s	33.5191 KOps/s	$\color{#35bf28}+0.15\%$
test_unbind_td	0.3183ms	40.9815μs	24.4013 KOps/s	24.9724 KOps/s	$\color{#d91a1a}-2.29\%$
test_split_pytree	0.1111ms	29.5880μs	33.7975 KOps/s	34.0418 KOps/s	$\color{#d91a1a}-0.72\%$
test_split_td	0.2044ms	45.2345μs	22.1070 KOps/s	21.9504 KOps/s	$\color{#35bf28}+0.71\%$
test_add_pytree	98.4370μs	36.2494μs	27.5866 KOps/s	27.5514 KOps/s	$\color{#35bf28}+0.13\%$
test_add_td	0.1361ms	59.9391μs	16.6836 KOps/s	18.1961 KOps/s	$\textbf{\color{#d91a1a}-8.31\%}$
test_compile_add_one_nested[tensordict-compile]	0.1292ms	62.1730μs	16.0841 KOps/s	15.8602 KOps/s	$\color{#35bf28}+1.41\%$
test_compile_add_one_nested[tensordict-eager]	1.5080ms	0.1704ms	5.8688 KOps/s	5.7482 KOps/s	$\color{#35bf28}+2.10\%$
test_compile_add_one_nested[pytree-compile]	0.1365ms	45.3738μs	22.0391 KOps/s	21.5228 KOps/s	$\color{#35bf28}+2.40\%$
test_compile_add_one_nested[pytree-eager]	0.2712ms	0.1215ms	8.2306 KOps/s	8.4330 KOps/s	$\color{#d91a1a}-2.40\%$
test_compile_copy_nested[tensordict-compile]	92.0010μs	26.3546μs	37.9440 KOps/s	38.9419 KOps/s	$\color{#d91a1a}-2.56\%$
test_compile_copy_nested[tensordict-eager]	0.1231ms	59.0386μs	16.9381 KOps/s	17.1328 KOps/s	$\color{#d91a1a}-1.14\%$
test_compile_copy_nested[pytree-compile]	0.1492ms	78.3206μs	12.7680 KOps/s	13.0864 KOps/s	$\color{#d91a1a}-2.43\%$
test_compile_copy_nested[pytree-eager]	0.1518ms	67.0857μs	14.9063 KOps/s	14.7939 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_add_one_flat[tensordict-compile]	0.2232ms	0.1061ms	9.4276 KOps/s	9.5787 KOps/s	$\color{#d91a1a}-1.58\%$
test_compile_add_one_flat[tensordict-eager]	0.3896ms	0.2230ms	4.4837 KOps/s	4.5613 KOps/s	$\color{#d91a1a}-1.70\%$
test_compile_add_one_flat[tensorclass-compile]	0.1007ms	45.1650μs	22.1410 KOps/s	22.5023 KOps/s	$\color{#d91a1a}-1.61\%$
test_compile_add_one_flat[tensorclass-eager]	0.4830ms	65.5846μs	15.2475 KOps/s	15.2131 KOps/s	$\color{#35bf28}+0.23\%$
test_compile_add_one_flat[pytree-compile]	0.2057ms	0.1046ms	9.5557 KOps/s	9.8612 KOps/s	$\color{#d91a1a}-3.10\%$
test_compile_add_one_flat[pytree-eager]	0.3116ms	0.2054ms	4.8684 KOps/s	4.9212 KOps/s	$\color{#d91a1a}-1.07\%$
test_compile_add_self_flat[tensordict-eager]	0.4058ms	0.2299ms	4.3494 KOps/s	4.2300 KOps/s	$\color{#35bf28}+2.82\%$
test_compile_add_self_flat[tensordict-compile]	0.1937ms	0.1076ms	9.2951 KOps/s	9.5830 KOps/s	$\color{#d91a1a}-3.00\%$
test_compile_add_self_flat[tensorclass-eager]	0.2014ms	60.0565μs	16.6510 KOps/s	16.9052 KOps/s	$\color{#d91a1a}-1.50\%$
test_compile_add_self_flat[tensorclass-compile]	0.1092ms	46.1325μs	21.6767 KOps/s	22.1237 KOps/s	$\color{#d91a1a}-2.02\%$
test_compile_add_self_flat[pytree-eager]	0.6015ms	0.1631ms	6.1294 KOps/s	6.2033 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_add_self_flat[pytree-compile]	0.2050ms	0.1043ms	9.5846 KOps/s	9.8154 KOps/s	$\color{#d91a1a}-2.35\%$
test_compile_copy_flat[tensordict-compile]	63.6590μs	21.2960μs	46.9571 KOps/s	47.4508 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_copy_flat[tensordict-eager]	0.1297ms	65.6229μs	15.2386 KOps/s	15.2541 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_copy_flat[pytree-compile]	0.1610ms	80.1088μs	12.4830 KOps/s	12.4598 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_copy_flat[pytree-eager]	0.1295ms	68.7149μs	14.5529 KOps/s	14.6231 KOps/s	$\color{#d91a1a}-0.48\%$
test_compile_assign_and_add[tensordict-compile]	0.3104ms	0.2082ms	4.8037 KOps/s	4.9149 KOps/s	$\color{#d91a1a}-2.26\%$
test_compile_assign_and_add[tensordict-eager]	2.0382ms	1.3026ms	767.6901 Ops/s	759.7948 Ops/s	$\color{#35bf28}+1.04\%$
test_compile_assign_and_add[pytree-compile]	0.2883ms	0.2058ms	4.8592 KOps/s	4.9546 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_assign_and_add[pytree-eager]	1.4637ms	0.7932ms	1.2607 KOps/s	1.2864 KOps/s	$\color{#d91a1a}-1.99\%$
test_compile_assign_and_add_stack[compile]	0.5593ms	0.4607ms	2.1704 KOps/s	2.2393 KOps/s	$\color{#d91a1a}-3.08\%$
test_compile_assign_and_add_stack[eager]	3.0135ms	2.7874ms	358.7535 Ops/s	376.9769 Ops/s	$\color{#d91a1a}-4.83\%$
test_compile_indexing[tensor-tensordict-compile]	84.3270μs	36.2639μs	27.5756 KOps/s	28.1681 KOps/s	$\color{#d91a1a}-2.10\%$
test_compile_indexing[tensor-tensordict-eager]	0.5487ms	34.8342μs	28.7074 KOps/s	29.6817 KOps/s	$\color{#d91a1a}-3.28\%$
test_compile_indexing[tensor-tensorclass-compile]	82.1230μs	29.1605μs	34.2930 KOps/s	34.2297 KOps/s	$\color{#35bf28}+0.18\%$
test_compile_indexing[tensor-tensorclass-eager]	72.9050μs	24.2192μs	41.2896 KOps/s	42.6383 KOps/s	$\color{#d91a1a}-3.16\%$
test_compile_indexing[tensor-pytree-compile]	87.1420μs	30.0355μs	33.2939 KOps/s	33.4503 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_indexing[tensor-pytree-eager]	65.1010μs	24.2029μs	41.3174 KOps/s	42.9683 KOps/s	$\color{#d91a1a}-3.84\%$
test_compile_indexing[slice-tensordict-compile]	0.1319ms	52.6470μs	18.9944 KOps/s	19.5880 KOps/s	$\color{#d91a1a}-3.03\%$
test_compile_indexing[slice-tensordict-eager]	0.5694ms	20.3781μs	49.0722 KOps/s	49.7532 KOps/s	$\color{#d91a1a}-1.37\%$
test_compile_indexing[slice-tensorclass-compile]	0.1008ms	45.0613μs	22.1920 KOps/s	22.9115 KOps/s	$\color{#d91a1a}-3.14\%$
test_compile_indexing[slice-tensorclass-eager]	96.4800μs	18.7322μs	53.3840 KOps/s	54.5255 KOps/s	$\color{#d91a1a}-2.09\%$
test_compile_indexing[slice-pytree-compile]	0.1018ms	45.5026μs	21.9768 KOps/s	22.5165 KOps/s	$\color{#d91a1a}-2.40\%$
test_compile_indexing[slice-pytree-eager]	54.7120μs	18.9824μs	52.6803 KOps/s	53.8890 KOps/s	$\color{#d91a1a}-2.24\%$
test_compile_indexing[int-tensordict-compile]	0.1107ms	53.9956μs	18.5200 KOps/s	19.1803 KOps/s	$\color{#d91a1a}-3.44\%$
test_compile_indexing[int-tensordict-eager]	1.0458ms	20.4271μs	48.9545 KOps/s	50.9235 KOps/s	$\color{#d91a1a}-3.87\%$
test_compile_indexing[int-tensorclass-compile]	0.1099ms	45.9124μs	21.7806 KOps/s	22.4933 KOps/s	$\color{#d91a1a}-3.17\%$
test_compile_indexing[int-tensorclass-eager]	78.3400μs	18.7705μs	53.2751 KOps/s	53.1059 KOps/s	$\color{#35bf28}+0.32\%$
test_compile_indexing[int-pytree-compile]	88.1250μs	45.5957μs	21.9319 KOps/s	22.4309 KOps/s	$\color{#d91a1a}-2.22\%$
test_compile_indexing[int-pytree-eager]	85.1990μs	19.2910μs	51.8376 KOps/s	54.1255 KOps/s	$\color{#d91a1a}-4.23\%$
test_mod_add[eager]	0.1192ms	35.2344μs	28.3813 KOps/s	29.2698 KOps/s	$\color{#d91a1a}-3.04\%$
test_mod_add[compile]	0.1369ms	48.8222μs	20.4825 KOps/s	20.4791 KOps/s	$\color{#35bf28}+0.02\%$
test_mod_add[compile-overhead]	0.1397ms	47.8877μs	20.8822 KOps/s	20.5729 KOps/s	$\color{#35bf28}+1.50\%$
test_mod_wrap[eager]	0.4745ms	0.2254ms	4.4375 KOps/s	4.3704 KOps/s	$\color{#35bf28}+1.53\%$
test_mod_wrap[compile]	0.2934ms	0.2107ms	4.7468 KOps/s	4.7816 KOps/s	$\color{#d91a1a}-0.73\%$
test_mod_wrap[compile-overhead]	0.3668ms	0.2073ms	4.8232 KOps/s	4.7675 KOps/s	$\color{#35bf28}+1.17\%$
test_mod_wrap_and_backward[eager]	13.1533ms	11.1401ms	89.7656 Ops/s	84.3023 Ops/s	$\textbf{\color{#35bf28}+6.48\%}$
test_mod_wrap_and_backward[compile]	12.5266ms	10.8146ms	92.4679 Ops/s	73.2740 Ops/s	$\textbf{\color{#35bf28}+26.19\%}$
test_mod_wrap_and_backward[compile-overhead]	12.2446ms	10.7905ms	92.6737 Ops/s	72.8180 Ops/s	$\textbf{\color{#35bf28}+27.27\%}$
test_seq_add[eager]	0.2419ms	0.1177ms	8.4935 KOps/s	8.3791 KOps/s	$\color{#35bf28}+1.36\%$
test_seq_add[compile]	0.1217ms	63.3450μs	15.7866 KOps/s	16.0727 KOps/s	$\color{#d91a1a}-1.78\%$
test_seq_add[compile-overhead]	0.1502ms	61.7941μs	16.1828 KOps/s	16.6604 KOps/s	$\color{#d91a1a}-2.87\%$
test_seq_wrap[eager]	0.6464ms	0.4531ms	2.2070 KOps/s	2.2240 KOps/s	$\color{#d91a1a}-0.76\%$
test_seq_wrap[compile]	0.3575ms	0.2290ms	4.3669 KOps/s	4.2516 KOps/s	$\color{#35bf28}+2.71\%$
test_seq_wrap[compile-overhead]	0.4354ms	0.2287ms	4.3717 KOps/s	4.2561 KOps/s	$\color{#35bf28}+2.72\%$
test_func_call_runtime[False-eager]	0.9088ms	0.5404ms	1.8505 KOps/s	1.7545 KOps/s	$\textbf{\color{#35bf28}+5.47\%}$
test_func_call_runtime[False-compile]	0.5923ms	0.4295ms	2.3281 KOps/s	2.2991 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_runtime[False-compile-overhead]	0.5823ms	0.4265ms	2.3447 KOps/s	2.2857 KOps/s	$\color{#35bf28}+2.58\%$
test_func_call_runtime[True-eager]	1.2989ms	0.7578ms	1.3196 KOps/s	1.2619 KOps/s	$\color{#35bf28}+4.58\%$
test_func_call_runtime[True-compile]	0.6943ms	0.4678ms	2.1378 KOps/s	2.1026 KOps/s	$\color{#35bf28}+1.67\%$
test_func_call_runtime[True-compile-overhead]	0.7418ms	0.4706ms	2.1248 KOps/s	2.0960 KOps/s	$\color{#35bf28}+1.37\%$
test_func_call_cm_runtime[False-eager]	0.9406ms	0.5447ms	1.8358 KOps/s	1.7713 KOps/s	$\color{#35bf28}+3.64\%$
test_func_call_cm_runtime[False-compile]	0.8283ms	0.4309ms	2.3205 KOps/s	2.3049 KOps/s	$\color{#35bf28}+0.68\%$
test_func_call_cm_runtime[False-compile-overhead]	0.8290ms	0.4337ms	2.3059 KOps/s	2.3013 KOps/s	$\color{#35bf28}+0.20\%$
test_func_call_cm_runtime[True-eager]	1.5034ms	0.9083ms	1.1010 KOps/s	1.0757 KOps/s	$\color{#35bf28}+2.35\%$
test_func_call_cm_runtime[True-compile]	0.9447ms	0.4985ms	2.0061 KOps/s	1.9818 KOps/s	$\color{#35bf28}+1.22\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5949ms	0.4897ms	2.0419 KOps/s	2.0034 KOps/s	$\color{#35bf28}+1.92\%$
test_vmap_func_call_cm_runtime[eager]	3.1152ms	1.9351ms	516.7607 Ops/s	517.1543 Ops/s	$\color{#d91a1a}-0.08\%$
test_vmap_func_call_cm_runtime[compile]	0.9665ms	0.5233ms	1.9111 KOps/s	1.9116 KOps/s	$\color{#d91a1a}-0.03\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.1429ms	0.5331ms	1.8759 KOps/s	1.9074 KOps/s	$\color{#d91a1a}-1.66\%$
test_distributed	0.2883ms	0.1265ms	7.9082 KOps/s	7.7809 KOps/s	$\color{#35bf28}+1.64\%$
test_tdmodule	56.4860μs	27.3286μs	36.5917 KOps/s	39.7138 KOps/s	$\textbf{\color{#d91a1a}-7.86\%}$
test_tdmodule_dispatch	80.4210μs	50.2795μs	19.8888 KOps/s	22.1260 KOps/s	$\textbf{\color{#d91a1a}-10.11\%}$
test_tdseq	50.0030μs	30.6262μs	32.6518 KOps/s	35.4519 KOps/s	$\textbf{\color{#d91a1a}-7.90\%}$
test_tdseq_dispatch	0.1068ms	58.2976μs	17.1534 KOps/s	19.5199 KOps/s	$\textbf{\color{#d91a1a}-12.12\%}$
test_instantiation_functorch	2.3837ms	1.5493ms	645.4660 Ops/s	632.3847 Ops/s	$\color{#35bf28}+2.07\%$
test_exec_functorch	0.4128ms	0.1842ms	5.4294 KOps/s	5.4035 KOps/s	$\color{#35bf28}+0.48\%$
test_exec_functional_call	0.2878ms	0.1758ms	5.6874 KOps/s	5.5133 KOps/s	$\color{#35bf28}+3.16\%$
test_exec_td_decorator	0.5327ms	0.2350ms	4.2560 KOps/s	4.2081 KOps/s	$\color{#35bf28}+1.14\%$
test_vmap_mlp_speed_decorator[True-True]	1.1178ms	0.6672ms	1.4988 KOps/s	1.5186 KOps/s	$\color{#d91a1a}-1.31\%$
test_vmap_mlp_speed_decorator[True-False]	1.3201ms	0.6745ms	1.4826 KOps/s	1.5173 KOps/s	$\color{#d91a1a}-2.28\%$
test_vmap_mlp_speed_decorator[False-True]	0.7205ms	0.5321ms	1.8792 KOps/s	1.8582 KOps/s	$\color{#35bf28}+1.13\%$
test_vmap_mlp_speed_decorator[False-False]	0.9188ms	0.5351ms	1.8690 KOps/s	1.8578 KOps/s	$\color{#35bf28}+0.60\%$
test_to_module_speed[True]	1.6230ms	1.3353ms	748.8684 Ops/s	735.0132 Ops/s	$\color{#35bf28}+1.89\%$
test_to_module_speed[False]	1.6928ms	1.3001ms	769.1958 Ops/s	754.6603 Ops/s	$\color{#35bf28}+1.93\%$
test_tc_init	0.1082ms	51.5966μs	19.3811 KOps/s	22.0215 KOps/s	$\textbf{\color{#d91a1a}-11.99\%}$
test_tc_init_nested	0.2366ms	0.1023ms	9.7770 KOps/s	11.0303 KOps/s	$\textbf{\color{#d91a1a}-11.36\%}$
test_tc_first_layer_tensor	23.6440μs	1.5197μs	658.0127 KOps/s	668.0268 KOps/s	$\color{#d91a1a}-1.50\%$
test_tc_first_layer_nontensor	26.1890μs	4.6386μs	215.5822 KOps/s	210.5022 KOps/s	$\color{#35bf28}+2.41\%$
test_tc_second_layer_tensor	18.8150μs	2.7923μs	358.1335 KOps/s	353.9820 KOps/s	$\color{#35bf28}+1.17\%$
test_tc_second_layer_nontensor	29.8750μs	5.9690μs	167.5327 KOps/s	162.5379 KOps/s	$\color{#35bf28}+3.07\%$
test_unbind	0.2125s	13.4114ms	74.5635 Ops/s	70.7291 Ops/s	$\textbf{\color{#35bf28}+5.42\%}$
test_full_like	15.7327ms	11.4754ms	87.1427 Ops/s	146.0968 Ops/s	$\textbf{\color{#d91a1a}-40.35\%}$
test_zeros_like	13.1565ms	7.4908ms	133.4977 Ops/s	375.3555 Ops/s	$\textbf{\color{#d91a1a}-64.43\%}$
test_ones_like	15.1283ms	7.7716ms	128.6730 Ops/s	323.3975 Ops/s	$\textbf{\color{#d91a1a}-60.21\%}$
test_clone	14.3852ms	9.1749ms	108.9929 Ops/s	208.8154 Ops/s	$\textbf{\color{#d91a1a}-47.80\%}$
test_squeeze	80.0200μs	11.8159μs	84.6319 KOps/s	80.4650 KOps/s	$\textbf{\color{#35bf28}+5.18\%}$
test_unsqueeze	0.2916ms	92.6916μs	10.7885 KOps/s	10.4923 KOps/s	$\color{#35bf28}+2.82\%$
test_split	0.4100ms	0.1967ms	5.0849 KOps/s	4.9661 KOps/s	$\color{#35bf28}+2.39\%$
test_permute	0.3614ms	0.2064ms	4.8441 KOps/s	4.7651 KOps/s	$\color{#35bf28}+1.66\%$
test_stack	28.2167ms	24.5854ms	40.6746 Ops/s	41.6861 Ops/s	$\color{#d91a1a}-2.43\%$
test_cat	28.9677ms	24.4968ms	40.8216 Ops/s	42.1321 Ops/s	$\color{#d91a1a}-3.11\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}44$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	31.8300μs	11.5365μs	86.6814 KOps/s	75.9189 KOps/s	$\textbf{\color{#35bf28}+14.18\%}$
test_plain_set_stack_nested	35.2110μs	11.6256μs	86.0172 KOps/s	74.3207 KOps/s	$\textbf{\color{#35bf28}+15.74\%}$
test_plain_set_nested_inplace	41.7500μs	12.4259μs	80.4774 KOps/s	70.3751 KOps/s	$\textbf{\color{#35bf28}+14.35\%}$
test_plain_set_stack_nested_inplace	46.0010μs	12.4728μs	80.1745 KOps/s	70.2124 KOps/s	$\textbf{\color{#35bf28}+14.19\%}$
test_items	27.4700μs	2.8629μs	349.3013 KOps/s	344.7788 KOps/s	$\color{#35bf28}+1.31\%$
test_items_nested	0.4095ms	0.3617ms	2.7645 KOps/s	2.7675 KOps/s	$\color{#d91a1a}-0.11\%$
test_items_nested_locked	0.4270ms	0.3640ms	2.7475 KOps/s	2.7447 KOps/s	$\color{#35bf28}+0.10\%$
test_items_nested_leaf	93.6710μs	58.0115μs	17.2380 KOps/s	17.1098 KOps/s	$\color{#35bf28}+0.75\%$
test_items_stack_nested	0.4324ms	0.3599ms	2.7783 KOps/s	2.7751 KOps/s	$\color{#35bf28}+0.11\%$
test_items_stack_nested_leaf	85.2110μs	57.9536μs	17.2552 KOps/s	16.6792 KOps/s	$\color{#35bf28}+3.45\%$
test_items_stack_nested_locked	0.4274ms	0.3577ms	2.7955 KOps/s	2.7357 KOps/s	$\color{#35bf28}+2.19\%$
test_keys	24.3510μs	3.4305μs	291.4990 KOps/s	288.6302 KOps/s	$\color{#35bf28}+0.99\%$
test_keys_nested	0.1187ms	81.9190μs	12.2072 KOps/s	12.2632 KOps/s	$\color{#d91a1a}-0.46\%$
test_keys_nested_locked	0.7165ms	86.7991μs	11.5209 KOps/s	11.5212 KOps/s	$-0.00\%$
test_keys_nested_leaf	0.1082ms	72.4267μs	13.8071 KOps/s	13.8138 KOps/s	$\color{#d91a1a}-0.05\%$
test_keys_stack_nested	0.1204ms	81.6618μs	12.2456 KOps/s	12.2688 KOps/s	$\color{#d91a1a}-0.19\%$
test_keys_stack_nested_leaf	0.1080ms	72.8548μs	13.7259 KOps/s	13.6252 KOps/s	$\color{#35bf28}+0.74\%$
test_keys_stack_nested_locked	0.1281ms	87.9497μs	11.3701 KOps/s	11.5555 KOps/s	$\color{#d91a1a}-1.60\%$
test_values	4.0521μs	0.8485μs	1.1786 MOps/s	1.1781 MOps/s	$\color{#35bf28}+0.05\%$
test_values_nested	69.4810μs	34.8165μs	28.7220 KOps/s	29.2037 KOps/s	$\color{#d91a1a}-1.65\%$
test_values_nested_locked	65.1600μs	37.0682μs	26.9773 KOps/s	27.6593 KOps/s	$\color{#d91a1a}-2.47\%$
test_values_nested_leaf	0.1816ms	39.8477μs	25.0955 KOps/s	25.5484 KOps/s	$\color{#d91a1a}-1.77\%$
test_values_stack_nested	63.1210μs	34.9267μs	28.6314 KOps/s	29.1514 KOps/s	$\color{#d91a1a}-1.78\%$
test_values_stack_nested_leaf	72.7910μs	39.7728μs	25.1428 KOps/s	25.3813 KOps/s	$\color{#d91a1a}-0.94\%$
test_values_stack_nested_locked	65.3300μs	36.7667μs	27.1985 KOps/s	27.1914 KOps/s	$\color{#35bf28}+0.03\%$
test_membership	1.9165μs	0.5143μs	1.9442 MOps/s	1.9442 MOps/s	$+0.00\%$
test_membership_nested	17.1750μs	1.9552μs	511.4509 KOps/s	509.0174 KOps/s	$\color{#35bf28}+0.48\%$
test_membership_nested_leaf	16.2855μs	1.9833μs	504.2000 KOps/s	505.1205 KOps/s	$\color{#d91a1a}-0.18\%$
test_membership_stacked_nested	27.4400μs	1.9960μs	500.9954 KOps/s	481.3560 KOps/s	$\color{#35bf28}+4.08\%$
test_membership_stacked_nested_leaf	30.2100μs	2.0128μs	496.8158 KOps/s	479.8088 KOps/s	$\color{#35bf28}+3.54\%$
test_membership_nested_last	25.7600μs	2.9804μs	335.5287 KOps/s	330.7205 KOps/s	$\color{#35bf28}+1.45\%$
test_membership_nested_leaf_last	30.4400μs	3.0115μs	332.0569 KOps/s	331.4262 KOps/s	$\color{#35bf28}+0.19\%$
test_membership_stacked_nested_last	32.6600μs	3.0446μs	328.4475 KOps/s	121.3527 KOps/s	$\textbf{\color{#35bf28}+170.66\%}$
test_membership_stacked_nested_leaf_last	34.3100μs	3.0992μs	322.6592 KOps/s	120.4812 KOps/s	$\textbf{\color{#35bf28}+167.81\%}$
test_nested_getleaf	42.1400μs	6.1630μs	162.2594 KOps/s	161.1854 KOps/s	$\color{#35bf28}+0.67\%$
test_nested_get	29.9500μs	5.8099μs	172.1201 KOps/s	170.1450 KOps/s	$\color{#35bf28}+1.16\%$
test_stacked_getleaf	37.3810μs	6.1919μs	161.5026 KOps/s	162.7523 KOps/s	$\color{#d91a1a}-0.77\%$
test_stacked_get	40.2210μs	5.8276μs	171.5960 KOps/s	171.7953 KOps/s	$\color{#d91a1a}-0.12\%$
test_nested_getitemleaf	36.1610μs	6.2227μs	160.7025 KOps/s	158.8154 KOps/s	$\color{#35bf28}+1.19\%$
test_nested_getitem	34.4900μs	5.9531μs	167.9785 KOps/s	164.8474 KOps/s	$\color{#35bf28}+1.90\%$
test_stacked_getitemleaf	31.8500μs	6.3491μs	157.5019 KOps/s	158.6990 KOps/s	$\color{#d91a1a}-0.75\%$
test_stacked_getitem	38.0710μs	5.9430μs	168.2653 KOps/s	168.4355 KOps/s	$\color{#d91a1a}-0.10\%$
test_lock_nested	9.4422ms	0.3918ms	2.5522 KOps/s	2.6205 KOps/s	$\color{#d91a1a}-2.61\%$
test_lock_stack_nested	0.3968ms	0.3463ms	2.8879 KOps/s	2.9743 KOps/s	$\color{#d91a1a}-2.91\%$
test_unlock_nested	0.6158ms	0.3143ms	3.1814 KOps/s	3.2088 KOps/s	$\color{#d91a1a}-0.86\%$
test_unlock_stack_nested	0.3348ms	0.2828ms	3.5358 KOps/s	3.6343 KOps/s	$\color{#d91a1a}-2.71\%$
test_flatten_speed	0.1135ms	76.0581μs	13.1478 KOps/s	13.1336 KOps/s	$\color{#35bf28}+0.11\%$
test_unflatten_speed	0.3754ms	0.3223ms	3.1031 KOps/s	3.1277 KOps/s	$\color{#d91a1a}-0.79\%$
test_common_ops	1.6223ms	0.5737ms	1.7431 KOps/s	1.6138 KOps/s	$\textbf{\color{#35bf28}+8.01\%}$
test_creation	0.1773ms	1.7168μs	582.4943 KOps/s	584.8912 KOps/s	$\color{#d91a1a}-0.41\%$
test_creation_empty	36.2310μs	6.7709μs	147.6912 KOps/s	101.3763 KOps/s	$\textbf{\color{#35bf28}+45.69\%}$
test_creation_nested_1	34.4910μs	8.4004μs	119.0414 KOps/s	85.2224 KOps/s	$\textbf{\color{#35bf28}+39.68\%}$
test_creation_nested_2	33.7600μs	11.1247μs	89.8900 KOps/s	70.4545 KOps/s	$\textbf{\color{#35bf28}+27.59\%}$
test_clone	48.2910μs	9.7304μs	102.7704 KOps/s	102.1360 KOps/s	$\color{#35bf28}+0.62\%$
test_getitem[int]	1.6022ms	10.5777μs	94.5387 KOps/s	95.5557 KOps/s	$\color{#d91a1a}-1.06\%$
test_getitem[slice_int]	92.4989ms	29.0792μs	34.3889 KOps/s	49.5912 KOps/s	$\textbf{\color{#d91a1a}-30.66\%}$
test_getitem[range]	0.1261ms	35.9358μs	27.8274 KOps/s	28.1977 KOps/s	$\color{#d91a1a}-1.31\%$
test_getitem[tuple]	0.1034ms	17.6316μs	56.7163 KOps/s	55.2112 KOps/s	$\color{#35bf28}+2.73\%$
test_getitem[list]	0.2325ms	31.7809μs	31.4654 KOps/s	31.9408 KOps/s	$\color{#d91a1a}-1.49\%$
test_setitem_dim[int]	46.1200μs	17.6093μs	56.7881 KOps/s	56.4113 KOps/s	$\color{#35bf28}+0.67\%$
test_setitem_dim[slice_int]	59.1510μs	36.9392μs	27.0715 KOps/s	27.6661 KOps/s	$\color{#d91a1a}-2.15\%$
test_setitem_dim[range]	82.0810μs	50.5946μs	19.7650 KOps/s	20.1840 KOps/s	$\color{#d91a1a}-2.08\%$
test_setitem_dim[tuple]	52.6910μs	29.6849μs	33.6871 KOps/s	32.3427 KOps/s	$\color{#35bf28}+4.16\%$
test_setitem	92.9510μs	13.1494μs	76.0492 KOps/s	66.3968 KOps/s	$\textbf{\color{#35bf28}+14.54\%}$
test_set	93.9010μs	12.7397μs	78.4949 KOps/s	68.3891 KOps/s	$\textbf{\color{#35bf28}+14.78\%}$
test_set_shared	1.5139ms	0.1494ms	6.6950 KOps/s	6.8248 KOps/s	$\color{#d91a1a}-1.90\%$
test_update	0.4843ms	15.2095μs	65.7484 KOps/s	54.3369 KOps/s	$\textbf{\color{#35bf28}+21.00\%}$
test_update_nested	92.1510μs	20.1265μs	49.6857 KOps/s	40.5151 KOps/s	$\textbf{\color{#35bf28}+22.64\%}$
test_update__nested	1.0157ms	24.0740μs	41.5385 KOps/s	42.0905 KOps/s	$\color{#d91a1a}-1.31\%$
test_set_nested	83.1210μs	14.2391μs	70.2291 KOps/s	63.6181 KOps/s	$\textbf{\color{#35bf28}+10.39\%}$
test_set_nested_new	91.6810μs	16.2771μs	61.4360 KOps/s	54.3200 KOps/s	$\textbf{\color{#35bf28}+13.10\%}$
test_select	0.1092ms	29.4089μs	34.0033 KOps/s	33.0807 KOps/s	$\color{#35bf28}+2.79\%$
test_select_nested	81.0710μs	43.2063μs	23.1448 KOps/s	22.7806 KOps/s	$\color{#35bf28}+1.60\%$
test_exclude_nested	86.3410μs	61.3504μs	16.2998 KOps/s	15.9524 KOps/s	$\color{#35bf28}+2.18\%$
test_empty[True]	0.3214ms	0.2858ms	3.4995 KOps/s	3.4407 KOps/s	$\color{#35bf28}+1.71\%$
test_empty[False]	3.2920μs	0.8285μs	1.2070 MOps/s	1.2151 MOps/s	$\color{#d91a1a}-0.67\%$
test_to	84.1110μs	55.3234μs	18.0755 KOps/s	18.2756 KOps/s	$\color{#d91a1a}-1.09\%$
test_to_nonblocking	91.5510μs	47.5690μs	21.0221 KOps/s	21.4671 KOps/s	$\color{#d91a1a}-2.07\%$
test_unbind_speed	0.7763ms	0.2349ms	4.2572 KOps/s	4.2369 KOps/s	$\color{#35bf28}+0.48\%$
test_unbind_speed_stack0	0.3148ms	0.2364ms	4.2307 KOps/s	4.3035 KOps/s	$\color{#d91a1a}-1.69\%$
test_unbind_speed_stack1	92.6779ms	0.6645ms	1.5048 KOps/s	1.5159 KOps/s	$\color{#d91a1a}-0.73\%$
test_split	93.6078ms	1.5816ms	632.2899 Ops/s	589.2031 Ops/s	$\textbf{\color{#35bf28}+7.31\%}$
test_chunk	96.3926ms	1.7336ms	576.8188 Ops/s	699.9203 Ops/s	$\textbf{\color{#d91a1a}-17.59\%}$
test_consolidate[False-None]	2.8109ms	2.6995ms	370.4403 Ops/s	338.3843 Ops/s	$\textbf{\color{#35bf28}+9.47\%}$
test_consolidate[default-None]	1.7771ms	1.6580ms	603.1487 Ops/s	603.4071 Ops/s	$\color{#d91a1a}-0.04\%$
test_consolidate[reduce-overhead-None]	1.7980ms	1.6932ms	590.6104 Ops/s	587.6905 Ops/s	$\color{#35bf28}+0.50\%$
test_consolidate_njt[False-None]	6.6954ms	6.4060ms	156.1047 Ops/s	113.0592 Ops/s	$\textbf{\color{#35bf28}+38.07\%}$
test_to[False-False-None]	1.8061ms	1.7096ms	584.9229 Ops/s	577.9310 Ops/s	$\color{#35bf28}+1.21\%$
test_to[True-False-None]	1.5967ms	1.3213ms	756.8447 Ops/s	761.9630 Ops/s	$\color{#d91a1a}-0.67\%$
test_to[within-False-None]	4.3529ms	4.1931ms	238.4851 Ops/s	241.2899 Ops/s	$\color{#d91a1a}-1.16\%$
test_to[True-default-None]	5.7313ms	5.2969ms	188.7903 Ops/s	185.9608 Ops/s	$\color{#35bf28}+1.52\%$
test_to_njt[False-False-None]	7.1372ms	6.8652ms	145.6619 Ops/s	138.6315 Ops/s	$\textbf{\color{#35bf28}+5.07\%}$
test_to_njt[True-False-None]	5.6790ms	5.4797ms	182.4925 Ops/s	170.2447 Ops/s	$\textbf{\color{#35bf28}+7.19\%}$
test_to_njt[within-False-None]	12.2037ms	11.9097ms	83.9654 Ops/s	78.1977 Ops/s	$\textbf{\color{#35bf28}+7.38\%}$
test_creation[device0]	0.4837ms	78.0010μs	12.8203 KOps/s	12.3361 KOps/s	$\color{#35bf28}+3.93\%$
test_creation_from_tensor	0.5471ms	81.9047μs	12.2093 KOps/s	11.7471 KOps/s	$\color{#35bf28}+3.93\%$
test_add_one[memmap_tensor0]	0.4109ms	6.1298μs	163.1375 KOps/s	169.3449 KOps/s	$\color{#d91a1a}-3.67\%$
test_contiguous[memmap_tensor0]	1.9975μs	0.4024μs	2.4852 MOps/s	2.4746 MOps/s	$\color{#35bf28}+0.43\%$
test_stack[memmap_tensor0]	23.7300μs	4.2948μs	232.8403 KOps/s	235.9435 KOps/s	$\color{#d91a1a}-1.32\%$
test_memmaptd_index	1.7216ms	0.2458ms	4.0685 KOps/s	3.8500 KOps/s	$\textbf{\color{#35bf28}+5.68\%}$
test_memmaptd_index_astensor	0.8196ms	0.3078ms	3.2487 KOps/s	3.2516 KOps/s	$\color{#d91a1a}-0.09\%$
test_memmaptd_index_op	0.9686ms	0.5343ms	1.8715 KOps/s	1.6995 KOps/s	$\textbf{\color{#35bf28}+10.12\%}$
test_serialize_model	0.1311s	0.1297s	7.7107 Ops/s	7.6475 Ops/s	$\color{#35bf28}+0.83\%$
test_serialize_model_pickle	1.3789s	1.2215s	0.8187 Ops/s	0.8408 Ops/s	$\color{#d91a1a}-2.63\%$
test_serialize_weights	0.4359s	0.1739s	5.7517 Ops/s	7.6982 Ops/s	$\textbf{\color{#d91a1a}-25.28\%}$
test_serialize_weights_returnearly	0.3406s	53.8358ms	18.5750 Ops/s	14.9864 Ops/s	$\textbf{\color{#35bf28}+23.95\%}$
test_serialize_weights_pickle	1.3663s	1.2177s	0.8212 Ops/s	0.8229 Ops/s	$\color{#d91a1a}-0.21\%$
test_reshape_pytree	54.7400μs	21.6667μs	46.1538 KOps/s	45.3976 KOps/s	$\color{#35bf28}+1.67\%$
test_reshape_td	56.7200μs	25.5240μs	39.1788 KOps/s	36.5623 KOps/s	$\textbf{\color{#35bf28}+7.16\%}$
test_view_pytree	58.1900μs	21.4968μs	46.5185 KOps/s	45.9099 KOps/s	$\color{#35bf28}+1.33\%$
test_view_td	56.9910μs	29.2620μs	34.1741 KOps/s	32.2195 KOps/s	$\textbf{\color{#35bf28}+6.07\%}$
test_unbind_pytree	56.7410μs	28.0456μs	35.6563 KOps/s	36.4826 KOps/s	$\color{#d91a1a}-2.27\%$
test_unbind_td	0.7468ms	36.7736μs	27.1934 KOps/s	28.1247 KOps/s	$\color{#d91a1a}-3.31\%$
test_split_pytree	56.8310μs	30.4582μs	32.8319 KOps/s	33.8506 KOps/s	$\color{#d91a1a}-3.01\%$
test_split_td	0.9148ms	38.6956μs	25.8427 KOps/s	26.3790 KOps/s	$\color{#d91a1a}-2.03\%$
test_add_pytree	71.0710μs	33.2015μs	30.1191 KOps/s	30.6251 KOps/s	$\color{#d91a1a}-1.65\%$
test_add_td	0.1006ms	47.4132μs	21.0912 KOps/s	18.6816 KOps/s	$\textbf{\color{#35bf28}+12.90\%}$
test_compile_add_one_nested[tensordict-compile]	0.1682ms	0.1170ms	8.5474 KOps/s	8.0917 KOps/s	$\textbf{\color{#35bf28}+5.63\%}$
test_compile_add_one_nested[tensordict-eager]	0.2182ms	0.1300ms	7.6906 KOps/s	7.6596 KOps/s	$\color{#35bf28}+0.41\%$
test_compile_add_one_nested[pytree-compile]	0.2048ms	93.4081μs	10.7057 KOps/s	10.5897 KOps/s	$\color{#35bf28}+1.10\%$
test_compile_add_one_nested[pytree-eager]	1.3786ms	0.1469ms	6.8090 KOps/s	6.6431 KOps/s	$\color{#35bf28}+2.50\%$
test_compile_copy_nested[tensordict-compile]	64.8610μs	23.0413μs	43.4003 KOps/s	44.9489 KOps/s	$\color{#d91a1a}-3.45\%$
test_compile_copy_nested[tensordict-eager]	61.4610μs	29.2733μs	34.1609 KOps/s	34.0204 KOps/s	$\color{#35bf28}+0.41\%$
test_compile_copy_nested[pytree-compile]	0.3612ms	64.4536μs	15.5150 KOps/s	15.3358 KOps/s	$\color{#35bf28}+1.17\%$
test_compile_copy_nested[pytree-eager]	85.3410μs	48.5270μs	20.6071 KOps/s	20.1437 KOps/s	$\color{#35bf28}+2.30\%$
test_compile_add_one_flat[tensordict-compile]	0.1982ms	0.1400ms	7.1407 KOps/s	7.1824 KOps/s	$\color{#d91a1a}-0.58\%$
test_compile_add_one_flat[tensordict-eager]	0.3052ms	0.2149ms	4.6533 KOps/s	4.7124 KOps/s	$\color{#d91a1a}-1.25\%$
test_compile_add_one_flat[tensorclass-compile]	0.1552ms	96.0676μs	10.4093 KOps/s	10.3099 KOps/s	$\color{#35bf28}+0.96\%$
test_compile_add_one_flat[tensorclass-eager]	0.1078ms	52.3240μs	19.1117 KOps/s	18.3835 KOps/s	$\color{#35bf28}+3.96\%$
test_compile_add_one_flat[pytree-compile]	0.2564ms	0.1347ms	7.4262 KOps/s	7.1213 KOps/s	$\color{#35bf28}+4.28\%$
test_compile_add_one_flat[pytree-eager]	0.5365ms	0.4694ms	2.1302 KOps/s	2.0750 KOps/s	$\color{#35bf28}+2.66\%$
test_compile_add_self_flat[tensordict-eager]	0.3774ms	0.2595ms	3.8541 KOps/s	3.9019 KOps/s	$\color{#d91a1a}-1.22\%$
test_compile_add_self_flat[tensordict-compile]	0.2362ms	0.1441ms	6.9406 KOps/s	7.2237 KOps/s	$\color{#d91a1a}-3.92\%$
test_compile_add_self_flat[tensorclass-eager]	0.1430ms	64.7464μs	15.4449 KOps/s	15.5796 KOps/s	$\color{#d91a1a}-0.86\%$
test_compile_add_self_flat[tensorclass-compile]	0.1330ms	97.0894μs	10.2998 KOps/s	9.9554 KOps/s	$\color{#35bf28}+3.46\%$
test_compile_add_self_flat[pytree-eager]	0.4649ms	0.4024ms	2.4853 KOps/s	2.4354 KOps/s	$\color{#35bf28}+2.05\%$
test_compile_add_self_flat[pytree-compile]	0.1800ms	0.1327ms	7.5365 KOps/s	7.4752 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_copy_flat[tensordict-compile]	49.9210μs	19.4632μs	51.3791 KOps/s	51.6743 KOps/s	$\color{#d91a1a}-0.57\%$
test_compile_copy_flat[tensordict-eager]	80.3010μs	32.1188μs	31.1344 KOps/s	32.4944 KOps/s	$\color{#d91a1a}-4.19\%$
test_compile_copy_flat[pytree-compile]	0.2262ms	70.7335μs	14.1376 KOps/s	13.9958 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_copy_flat[pytree-eager]	85.8110μs	51.9387μs	19.2535 KOps/s	19.4508 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_assign_and_add[tensordict-compile]	1.6157ms	0.3876ms	2.5802 KOps/s	2.2549 KOps/s	$\textbf{\color{#35bf28}+14.43\%}$
test_compile_assign_and_add[tensordict-eager]	2.9101ms	2.6001ms	384.6048 Ops/s	378.9625 Ops/s	$\color{#35bf28}+1.49\%$
test_compile_assign_and_add[pytree-compile]	1.5788ms	0.3782ms	2.6442 KOps/s	2.2329 KOps/s	$\textbf{\color{#35bf28}+18.42\%}$
test_compile_assign_and_add[pytree-eager]	2.6646ms	2.5559ms	391.2584 Ops/s	382.8438 Ops/s	$\color{#35bf28}+2.20\%$
test_compile_indexing[tensor-tensordict-compile]	0.6791ms	0.1100ms	9.0912 KOps/s	8.9042 KOps/s	$\color{#35bf28}+2.10\%$
test_compile_indexing[tensor-tensordict-eager]	0.5653ms	75.0006μs	13.3332 KOps/s	12.7763 KOps/s	$\color{#35bf28}+4.36\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5097ms	0.1036ms	9.6489 KOps/s	9.6685 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1108ms	65.6694μs	15.2278 KOps/s	14.9550 KOps/s	$\color{#35bf28}+1.82\%$
test_compile_indexing[tensor-pytree-compile]	1.0791ms	0.1045ms	9.5669 KOps/s	9.0355 KOps/s	$\textbf{\color{#35bf28}+5.88\%}$
test_compile_indexing[tensor-pytree-eager]	0.4539ms	65.7063μs	15.2192 KOps/s	13.9731 KOps/s	$\textbf{\color{#35bf28}+8.92\%}$
test_compile_indexing[slice-tensordict-compile]	0.1616ms	98.4602μs	10.1564 KOps/s	10.0564 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_indexing[slice-tensordict-eager]	0.4068ms	16.9414μs	59.0270 KOps/s	58.5370 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_indexing[slice-tensorclass-compile]	0.4963ms	93.9852μs	10.6400 KOps/s	10.4675 KOps/s	$\color{#35bf28}+1.65\%$
test_compile_indexing[slice-tensorclass-eager]	49.0310μs	15.9671μs	62.6288 KOps/s	62.0597 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_indexing[slice-pytree-compile]	0.4938ms	98.4108μs	10.1615 KOps/s	10.4718 KOps/s	$\color{#d91a1a}-2.96\%$
test_compile_indexing[slice-pytree-eager]	0.3972ms	16.3005μs	61.3477 KOps/s	63.8963 KOps/s	$\color{#d91a1a}-3.99\%$
test_compile_indexing[int-tensordict-compile]	0.4895ms	0.1016ms	9.8413 KOps/s	9.9525 KOps/s	$\color{#d91a1a}-1.12\%$
test_compile_indexing[int-tensordict-eager]	0.5797ms	16.7535μs	59.6889 KOps/s	58.9920 KOps/s	$\color{#35bf28}+1.18\%$
test_compile_indexing[int-tensorclass-compile]	0.4961ms	97.5043μs	10.2560 KOps/s	10.4564 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_indexing[int-tensorclass-eager]	0.1604ms	16.7265μs	59.7853 KOps/s	63.3300 KOps/s	$\textbf{\color{#d91a1a}-5.60\%}$
test_compile_indexing[int-pytree-compile]	0.4942ms	97.7994μs	10.2250 KOps/s	10.4737 KOps/s	$\color{#d91a1a}-2.37\%$
test_compile_indexing[int-pytree-eager]	0.4032ms	17.4841μs	57.1948 KOps/s	63.9671 KOps/s	$\textbf{\color{#d91a1a}-10.59\%}$
test_mod_add[eager]	0.4330ms	37.4718μs	26.6867 KOps/s	26.4948 KOps/s	$\color{#35bf28}+0.72\%$
test_mod_add[compile]	0.1420ms	82.6933μs	12.0929 KOps/s	12.3534 KOps/s	$\color{#d91a1a}-2.11\%$
test_mod_add[compile-overhead]	0.3234ms	0.1634ms	6.1183 KOps/s	5.7425 KOps/s	$\textbf{\color{#35bf28}+6.54\%}$
test_mod_wrap[eager]	0.6521ms	0.2368ms	4.2232 KOps/s	3.8336 KOps/s	$\textbf{\color{#35bf28}+10.16\%}$
test_mod_wrap[compile]	0.3654ms	0.2767ms	3.6144 KOps/s	3.5349 KOps/s	$\color{#35bf28}+2.25\%$
test_mod_wrap[compile-overhead]	7.3685ms	3.7757ms	264.8516 Ops/s	273.1945 Ops/s	$\color{#d91a1a}-3.05\%$
test_mod_wrap_and_backward[eager]	1.4185ms	1.3058ms	765.7954 Ops/s	700.7490 Ops/s	$\textbf{\color{#35bf28}+9.28\%}$
test_mod_wrap_and_backward[compile]	1.3500ms	1.2473ms	801.7395 Ops/s	793.8073 Ops/s	$\color{#35bf28}+1.00\%$
test_mod_wrap_and_backward[compile-overhead]	1.3538ms	0.9209ms	1.0859 KOps/s	1.0796 KOps/s	$\color{#35bf28}+0.58\%$
test_seq_add[eager]	0.1774ms	0.1130ms	8.8478 KOps/s	8.2173 KOps/s	$\textbf{\color{#35bf28}+7.67\%}$
test_seq_add[compile]	0.1592ms	86.4349μs	11.5694 KOps/s	11.2402 KOps/s	$\color{#35bf28}+2.93\%$
test_seq_add[compile-overhead]	0.1668ms	0.1288ms	7.7634 KOps/s	7.7997 KOps/s	$\color{#d91a1a}-0.46\%$
test_seq_wrap[eager]	0.4756ms	0.4196ms	2.3830 KOps/s	2.3703 KOps/s	$\color{#35bf28}+0.53\%$
test_seq_wrap[compile]	0.3819ms	0.3018ms	3.3131 KOps/s	3.3467 KOps/s	$\color{#d91a1a}-1.00\%$
test_seq_wrap[compile-overhead]	0.2715ms	0.2210ms	4.5242 KOps/s	4.4827 KOps/s	$\color{#35bf28}+0.93\%$
test_func_call_runtime[False-eager]	0.9344ms	0.7591ms	1.3174 KOps/s	1.4089 KOps/s	$\textbf{\color{#d91a1a}-6.49\%}$
test_func_call_runtime[False-compile]	0.8215ms	0.7276ms	1.3743 KOps/s	1.3530 KOps/s	$\color{#35bf28}+1.58\%$
test_func_call_runtime[False-compile-overhead]	0.4200ms	0.3591ms	2.7848 KOps/s	2.7764 KOps/s	$\color{#35bf28}+0.31\%$
test_func_call_runtime[True-eager]	0.9393ms	0.8614ms	1.1609 KOps/s	1.1210 KOps/s	$\color{#35bf28}+3.57\%$
test_func_call_runtime[True-compile]	0.8178ms	0.7590ms	1.3175 KOps/s	1.3213 KOps/s	$\color{#d91a1a}-0.29\%$
test_func_call_runtime[True-compile-overhead]	0.4736ms	0.3811ms	2.6243 KOps/s	2.6352 KOps/s	$\color{#d91a1a}-0.41\%$
test_func_call_cm_runtime[False-eager]	0.8436ms	0.7429ms	1.3460 KOps/s	1.3988 KOps/s	$\color{#d91a1a}-3.77\%$
test_func_call_cm_runtime[False-compile]	1.0780ms	0.7363ms	1.3582 KOps/s	1.3306 KOps/s	$\color{#35bf28}+2.07\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4258ms	0.3627ms	2.7574 KOps/s	2.7736 KOps/s	$\color{#d91a1a}-0.58\%$
test_func_call_cm_runtime[True-eager]	1.0831ms	0.9622ms	1.0393 KOps/s	1.0177 KOps/s	$\color{#35bf28}+2.12\%$
test_func_call_cm_runtime[True-compile]	0.9136ms	0.7786ms	1.2844 KOps/s	1.2662 KOps/s	$\color{#35bf28}+1.44\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4597ms	0.4068ms	2.4583 KOps/s	2.4411 KOps/s	$\color{#35bf28}+0.71\%$
test_vmap_func_call_cm_runtime[eager]	2.4707ms	2.0003ms	499.9154 Ops/s	492.0025 Ops/s	$\color{#35bf28}+1.61\%$
test_vmap_func_call_cm_runtime[compile]	0.9018ms	0.7952ms	1.2575 KOps/s	1.2487 KOps/s	$\color{#35bf28}+0.71\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4515ms	0.4075ms	2.4539 KOps/s	2.4356 KOps/s	$\color{#35bf28}+0.75\%$
test_distributed	6.5148ms	0.1607ms	6.2230 KOps/s	8.5706 KOps/s	$\textbf{\color{#d91a1a}-27.39\%}$
test_tdmodule	72.0410μs	18.3128μs	54.6066 KOps/s	48.1304 KOps/s	$\textbf{\color{#35bf28}+13.46\%}$
test_tdmodule_dispatch	54.2810μs	33.6386μs	29.7278 KOps/s	26.7061 KOps/s	$\textbf{\color{#35bf28}+11.31\%}$
test_tdseq	39.9900μs	19.9352μs	50.1625 KOps/s	45.4784 KOps/s	$\textbf{\color{#35bf28}+10.30\%}$
test_tdseq_dispatch	58.5810μs	36.5098μs	27.3899 KOps/s	24.4418 KOps/s	$\textbf{\color{#35bf28}+12.06\%}$
test_instantiation_functorch	1.8502ms	1.5228ms	656.6687 Ops/s	654.5387 Ops/s	$\color{#35bf28}+0.33\%$
test_exec_functorch	0.5768ms	0.1436ms	6.9631 KOps/s	6.9628 KOps/s	$+0.00\%$
test_exec_functional_call	0.1777ms	0.1334ms	7.4975 KOps/s	7.7472 KOps/s	$\color{#d91a1a}-3.22\%$
test_exec_td_decorator	0.5926ms	0.1811ms	5.5231 KOps/s	5.6462 KOps/s	$\color{#d91a1a}-2.18\%$
test_vmap_mlp_speed_decorator[True-True]	1.0924ms	0.6642ms	1.5055 KOps/s	1.4951 KOps/s	$\color{#35bf28}+0.70\%$
test_vmap_mlp_speed_decorator[True-False]	1.0549ms	0.6581ms	1.5195 KOps/s	1.4957 KOps/s	$\color{#35bf28}+1.59\%$
test_vmap_mlp_speed_decorator[False-True]	0.9990ms	0.5974ms	1.6739 KOps/s	1.7441 KOps/s	$\color{#d91a1a}-4.02\%$
test_vmap_mlp_speed_decorator[False-False]	0.9940ms	0.5975ms	1.6738 KOps/s	1.7333 KOps/s	$\color{#d91a1a}-3.43\%$
test_vmap_transformer_speed_decorator[True-True]	19.1701ms	18.5447ms	53.9237 Ops/s	53.9059 Ops/s	$\color{#35bf28}+0.03\%$
test_vmap_transformer_speed_decorator[True-False]	19.2156ms	18.5828ms	53.8133 Ops/s	53.7399 Ops/s	$\color{#35bf28}+0.14\%$
test_vmap_transformer_speed_decorator[False-True]	19.0022ms	18.4290ms	54.2622 Ops/s	54.2508 Ops/s	$\color{#35bf28}+0.02\%$
test_vmap_transformer_speed_decorator[False-False]	19.4612ms	18.4217ms	54.2839 Ops/s	54.1243 Ops/s	$\color{#35bf28}+0.30\%$
test_to_module_speed[True]	1.4318ms	0.9759ms	1.0247 KOps/s	1.0315 KOps/s	$\color{#d91a1a}-0.66\%$
test_to_module_speed[False]	1.3527ms	0.9545ms	1.0476 KOps/s	1.0378 KOps/s	$\color{#35bf28}+0.95\%$
test_tc_init	66.4300μs	37.2552μs	26.8419 KOps/s	25.2303 KOps/s	$\textbf{\color{#35bf28}+6.39\%}$
test_tc_init_nested	0.2195ms	75.9624μs	13.1644 KOps/s	12.0801 KOps/s	$\textbf{\color{#35bf28}+8.98\%}$
test_tc_first_layer_tensor	54.5050μs	0.6952μs	1.4384 MOps/s	1.4140 MOps/s	$\color{#35bf28}+1.73\%$
test_tc_first_layer_nontensor	24.0210μs	2.3328μs	428.6686 KOps/s	431.2735 KOps/s	$\color{#d91a1a}-0.60\%$
test_tc_second_layer_tensor	96.1088μs	1.4068μs	710.8156 KOps/s	716.6233 KOps/s	$\color{#d91a1a}-0.81\%$
test_tc_second_layer_nontensor	26.0210μs	3.0323μs	329.7790 KOps/s	329.4427 KOps/s	$\color{#35bf28}+0.10\%$
test_unbind	0.2202s	10.1671ms	98.3568 Ops/s	144.6915 Ops/s	$\textbf{\color{#d91a1a}-32.02\%}$
test_full_like	9.8872ms	9.2192ms	108.4694 Ops/s	106.3454 Ops/s	$\color{#35bf28}+2.00\%$
test_zeros_like	5.2110ms	4.3129ms	231.8607 Ops/s	115.2723 Ops/s	$\textbf{\color{#35bf28}+101.14\%}$
test_ones_like	9.3240ms	7.1934ms	139.0157 Ops/s	232.0149 Ops/s	$\textbf{\color{#d91a1a}-40.08\%}$
test_clone	11.6714ms	9.0661ms	110.3010 Ops/s	158.6598 Ops/s	$\textbf{\color{#d91a1a}-30.48\%}$
test_squeeze	60.4110μs	9.3888μs	106.5095 KOps/s	107.1121 KOps/s	$\color{#d91a1a}-0.56\%$
test_unsqueeze	0.1187ms	69.9956μs	14.2866 KOps/s	14.0659 KOps/s	$\color{#35bf28}+1.57\%$
test_split	0.5462ms	0.1538ms	6.5011 KOps/s	6.3491 KOps/s	$\color{#35bf28}+2.39\%$
test_permute	0.3162ms	0.1780ms	5.6179 KOps/s	5.5258 KOps/s	$\color{#35bf28}+1.67\%$
test_stack	51.1102ms	50.8321ms	19.6726 Ops/s	18.8417 Ops/s	$\color{#35bf28}+4.41\%$
test_cat	50.9168ms	50.6069ms	19.7601 Ops/s	19.0072 Ops/s	$\color{#35bf28}+3.96\%$