[BugFix] Fix none ref in during reduction

[vmoens]

No description provided.

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}30$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	40.5860μs	16.6028μs	60.2309 KOps/s	53.7553 KOps/s	$\textbf{\color{#35bf28}+12.05\%}$
test_plain_set_stack_nested	41.6280μs	16.9107μs	59.1340 KOps/s	52.1779 KOps/s	$\textbf{\color{#35bf28}+13.33\%}$
test_plain_set_nested_inplace	42.7900μs	19.6220μs	50.9633 KOps/s	48.3609 KOps/s	$\textbf{\color{#35bf28}+5.38\%}$
test_plain_set_stack_nested_inplace	50.7650μs	19.4826μs	51.3279 KOps/s	48.8495 KOps/s	$\textbf{\color{#35bf28}+5.07\%}$
test_items	29.0040μs	4.1387μs	241.6231 KOps/s	233.7002 KOps/s	$\color{#35bf28}+3.39\%$
test_items_nested	0.7840ms	0.3455ms	2.8946 KOps/s	2.8790 KOps/s	$\color{#35bf28}+0.54\%$
test_items_nested_locked	0.5823ms	0.3346ms	2.9888 KOps/s	2.9018 KOps/s	$\color{#35bf28}+3.00\%$
test_items_nested_leaf	0.1313ms	70.5510μs	14.1742 KOps/s	13.9049 KOps/s	$\color{#35bf28}+1.94\%$
test_items_stack_nested	0.5288ms	0.3377ms	2.9616 KOps/s	2.8702 KOps/s	$\color{#35bf28}+3.19\%$
test_items_stack_nested_leaf	0.1323ms	72.5077μs	13.7916 KOps/s	13.2535 KOps/s	$\color{#35bf28}+4.06\%$
test_items_stack_nested_locked	0.5306ms	0.3403ms	2.9389 KOps/s	2.8661 KOps/s	$\color{#35bf28}+2.54\%$
test_keys	38.3640μs	3.5122μs	284.7197 KOps/s	259.3824 KOps/s	$\textbf{\color{#35bf28}+9.77\%}$
test_keys_nested	0.2813ms	0.1385ms	7.2222 KOps/s	7.2852 KOps/s	$\color{#d91a1a}-0.86\%$
test_keys_nested_locked	0.7737ms	0.1426ms	7.0123 KOps/s	6.9241 KOps/s	$\color{#35bf28}+1.27\%$
test_keys_nested_leaf	0.2212ms	0.1173ms	8.5285 KOps/s	8.4727 KOps/s	$\color{#35bf28}+0.66\%$
test_keys_stack_nested	0.2516ms	0.1388ms	7.2032 KOps/s	7.1652 KOps/s	$\color{#35bf28}+0.53\%$
test_keys_stack_nested_leaf	0.2170ms	0.1191ms	8.3986 KOps/s	8.3578 KOps/s	$\color{#35bf28}+0.49\%$
test_keys_stack_nested_locked	0.2656ms	0.1402ms	7.1307 KOps/s	6.8305 KOps/s	$\color{#35bf28}+4.40\%$
test_values	8.2352μs	1.0330μs	968.0761 KOps/s	947.8644 KOps/s	$\color{#35bf28}+2.13\%$
test_values_nested	0.1032ms	55.6941μs	17.9552 KOps/s	17.8669 KOps/s	$\color{#35bf28}+0.49\%$
test_values_nested_locked	0.1057ms	55.3644μs	18.0622 KOps/s	17.5489 KOps/s	$\color{#35bf28}+2.92\%$
test_values_nested_leaf	0.1314ms	59.6869μs	16.7541 KOps/s	16.5409 KOps/s	$\color{#35bf28}+1.29\%$
test_values_stack_nested	0.1040ms	56.1719μs	17.8025 KOps/s	17.6679 KOps/s	$\color{#35bf28}+0.76\%$
test_values_stack_nested_leaf	0.1494ms	60.3033μs	16.5828 KOps/s	16.2642 KOps/s	$\color{#35bf28}+1.96\%$
test_values_stack_nested_locked	0.1230ms	55.5893μs	17.9891 KOps/s	17.6467 KOps/s	$\color{#35bf28}+1.94\%$
test_membership	5.6577μs	0.7511μs	1.3313 MOps/s	1.3077 MOps/s	$\color{#35bf28}+1.81\%$
test_membership_nested	39.4240μs	2.7197μs	367.6917 KOps/s	342.6418 KOps/s	$\textbf{\color{#35bf28}+7.31\%}$
test_membership_nested_leaf	22.1610μs	2.7477μs	363.9362 KOps/s	361.6750 KOps/s	$\color{#35bf28}+0.63\%$
test_membership_stacked_nested	19.4860μs	2.6597μs	375.9813 KOps/s	361.9312 KOps/s	$\color{#35bf28}+3.88\%$
test_membership_stacked_nested_leaf	31.9790μs	2.6920μs	371.4742 KOps/s	360.4171 KOps/s	$\color{#35bf28}+3.07\%$
test_membership_nested_last	45.2040μs	4.0468μs	247.1072 KOps/s	241.1451 KOps/s	$\color{#35bf28}+2.47\%$
test_membership_nested_leaf_last	37.6500μs	4.1402μs	241.5346 KOps/s	242.7020 KOps/s	$\color{#d91a1a}-0.48\%$
test_membership_stacked_nested_last	23.7940μs	4.0126μs	249.2157 KOps/s	242.5953 KOps/s	$\color{#35bf28}+2.73\%$
test_membership_stacked_nested_leaf_last	23.3830μs	4.0348μs	247.8418 KOps/s	244.6537 KOps/s	$\color{#35bf28}+1.30\%$
test_nested_getleaf	54.8120μs	10.5863μs	94.4616 KOps/s	92.6252 KOps/s	$\color{#35bf28}+1.98\%$
test_nested_get	53.0390μs	10.0501μs	99.5011 KOps/s	96.9535 KOps/s	$\color{#35bf28}+2.63\%$
test_stacked_getleaf	43.9020μs	10.5810μs	94.5092 KOps/s	92.2690 KOps/s	$\color{#35bf28}+2.43\%$
test_stacked_get	49.7630μs	9.9297μs	100.7081 KOps/s	98.1584 KOps/s	$\color{#35bf28}+2.60\%$
test_nested_getitemleaf	37.7710μs	10.9045μs	91.7053 KOps/s	89.3317 KOps/s	$\color{#35bf28}+2.66\%$
test_nested_getitem	40.1150μs	10.1188μs	98.8257 KOps/s	95.4249 KOps/s	$\color{#35bf28}+3.56\%$
test_stacked_getitemleaf	53.6730μs	10.7711μs	92.8411 KOps/s	88.7462 KOps/s	$\color{#35bf28}+4.61\%$
test_stacked_getitem	39.6970μs	10.1451μs	98.5697 KOps/s	95.4030 KOps/s	$\color{#35bf28}+3.32\%$
test_lock_nested	1.1042ms	0.4397ms	2.2742 KOps/s	1.8329 KOps/s	$\textbf{\color{#35bf28}+24.08\%}$
test_lock_stack_nested	0.6488ms	0.4126ms	2.4238 KOps/s	2.4302 KOps/s	$\color{#d91a1a}-0.26\%$
test_unlock_nested	0.7724ms	0.3540ms	2.8249 KOps/s	2.7848 KOps/s	$\color{#35bf28}+1.44\%$
test_unlock_stack_nested	0.4408ms	0.3306ms	3.0250 KOps/s	3.0112 KOps/s	$\color{#35bf28}+0.46\%$
test_flatten_speed	0.1640ms	91.1870μs	10.9665 KOps/s	10.8077 KOps/s	$\color{#35bf28}+1.47\%$
test_unflatten_speed	0.5949ms	0.4648ms	2.1512 KOps/s	2.1184 KOps/s	$\color{#35bf28}+1.55\%$
test_common_ops	1.7149ms	0.7315ms	1.3671 KOps/s	1.2551 KOps/s	$\textbf{\color{#35bf28}+8.92\%}$
test_creation	23.0340μs	2.1938μs	455.8211 KOps/s	444.0820 KOps/s	$\color{#35bf28}+2.64\%$
test_creation_empty	36.7690μs	9.1195μs	109.6553 KOps/s	80.9392 KOps/s	$\textbf{\color{#35bf28}+35.48\%}$
test_creation_nested_1	38.7720μs	11.8683μs	84.2578 KOps/s	65.7993 KOps/s	$\textbf{\color{#35bf28}+28.05\%}$
test_creation_nested_2	70.1310μs	15.8927μs	62.9221 KOps/s	50.9088 KOps/s	$\textbf{\color{#35bf28}+23.60\%}$
test_clone	0.2079ms	13.4965μs	74.0935 KOps/s	78.3278 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_getitem[int]	1.5169ms	12.5779μs	79.5044 KOps/s	78.3388 KOps/s	$\color{#35bf28}+1.49\%$
test_getitem[slice_int]	0.1526ms	24.2907μs	41.1680 KOps/s	39.8830 KOps/s	$\color{#35bf28}+3.22\%$
test_getitem[range]	0.1995ms	50.2590μs	19.8970 KOps/s	20.3528 KOps/s	$\color{#d91a1a}-2.24\%$
test_getitem[tuple]	0.1451ms	20.1744μs	49.5677 KOps/s	49.2376 KOps/s	$\color{#35bf28}+0.67\%$
test_getitem[list]	0.7351ms	45.5748μs	21.9420 KOps/s	22.6052 KOps/s	$\color{#d91a1a}-2.93\%$
test_setitem_dim[int]	62.8770μs	25.9778μs	38.4944 KOps/s	40.3748 KOps/s	$\color{#d91a1a}-4.66\%$
test_setitem_dim[slice_int]	89.8280μs	51.8508μs	19.2861 KOps/s	19.7872 KOps/s	$\color{#d91a1a}-2.53\%$
test_setitem_dim[range]	0.1609ms	77.2979μs	12.9370 KOps/s	13.5275 KOps/s	$\color{#d91a1a}-4.37\%$
test_setitem_dim[tuple]	84.0670μs	40.6031μs	24.6286 KOps/s	24.4949 KOps/s	$\color{#35bf28}+0.55\%$
test_setitem	0.2473ms	19.2595μs	51.9224 KOps/s	48.0652 KOps/s	$\textbf{\color{#35bf28}+8.03\%}$
test_set	89.8410μs	18.8050μs	53.1773 KOps/s	48.0593 KOps/s	$\textbf{\color{#35bf28}+10.65\%}$
test_set_shared	3.4724ms	0.1718ms	5.8202 KOps/s	5.9519 KOps/s	$\color{#d91a1a}-2.21\%$
test_update	0.1496ms	20.3668μs	49.0994 KOps/s	42.1357 KOps/s	$\textbf{\color{#35bf28}+16.53\%}$
test_update_nested	0.3243ms	29.3071μs	34.1215 KOps/s	30.3870 KOps/s	$\textbf{\color{#35bf28}+12.29\%}$
test_update__nested	0.3086ms	32.7985μs	30.4892 KOps/s	30.3736 KOps/s	$\color{#35bf28}+0.38\%$
test_set_nested	0.3670ms	20.7532μs	48.1853 KOps/s	44.1357 KOps/s	$\textbf{\color{#35bf28}+9.18\%}$
test_set_nested_new	0.2801ms	25.8718μs	38.6521 KOps/s	36.6964 KOps/s	$\textbf{\color{#35bf28}+5.33\%}$
test_select	0.2592ms	40.9560μs	24.4165 KOps/s	23.3633 KOps/s	$\color{#35bf28}+4.51\%$
test_select_nested	0.1129ms	59.6437μs	16.7662 KOps/s	16.5117 KOps/s	$\color{#35bf28}+1.54\%$
test_exclude_nested	0.1496ms	74.1716μs	13.4823 KOps/s	13.2318 KOps/s	$\color{#35bf28}+1.89\%$
test_empty[True]	0.5811ms	0.3507ms	2.8518 KOps/s	2.8196 KOps/s	$\color{#35bf28}+1.14\%$
test_empty[False]	13.7030μs	1.2175μs	821.3351 KOps/s	826.3690 KOps/s	$\color{#d91a1a}-0.61\%$
test_unbind_speed	0.3815ms	0.2638ms	3.7905 KOps/s	3.8338 KOps/s	$\color{#d91a1a}-1.13\%$
test_unbind_speed_stack0	0.4400ms	0.2595ms	3.8538 KOps/s	3.8984 KOps/s	$\color{#d91a1a}-1.14\%$
test_unbind_speed_stack1	0.1102s	0.7849ms	1.2740 KOps/s	1.4331 KOps/s	$\textbf{\color{#d91a1a}-11.10\%}$
test_split	0.1065s	1.7631ms	567.1898 Ops/s	562.7449 Ops/s	$\color{#35bf28}+0.79\%$
test_chunk	0.1076s	1.7672ms	565.8802 Ops/s	562.0015 Ops/s	$\color{#35bf28}+0.69\%$
test_consolidate_njt[False-None]	8.9206ms	8.2169ms	121.6997 Ops/s	122.9363 Ops/s	$\color{#d91a1a}-1.01\%$
test_creation[device0]	0.2759ms	92.1852μs	10.8477 KOps/s	11.1427 KOps/s	$\color{#d91a1a}-2.65\%$
test_creation_from_tensor	5.4089ms	97.2256μs	10.2854 KOps/s	10.8053 KOps/s	$\color{#d91a1a}-4.81\%$
test_add_one[memmap_tensor0]	0.1908ms	4.7824μs	209.0982 KOps/s	208.0007 KOps/s	$\color{#35bf28}+0.53\%$
test_contiguous[memmap_tensor0]	15.5290μs	0.5315μs	1.8816 MOps/s	1.9239 MOps/s	$\color{#d91a1a}-2.20\%$
test_stack[memmap_tensor0]	0.1222ms	3.4979μs	285.8855 KOps/s	296.7143 KOps/s	$\color{#d91a1a}-3.65\%$
test_memmaptd_index	1.0445ms	0.2391ms	4.1818 KOps/s	4.2335 KOps/s	$\color{#d91a1a}-1.22\%$
test_memmaptd_index_astensor	0.7245ms	0.3171ms	3.1536 KOps/s	3.1808 KOps/s	$\color{#d91a1a}-0.86\%$
test_memmaptd_index_op	1.0041ms	0.5543ms	1.8040 KOps/s	1.6625 KOps/s	$\textbf{\color{#35bf28}+8.51\%}$
test_serialize_model	0.1208s	0.1120s	8.9300 Ops/s	7.5691 Ops/s	$\textbf{\color{#35bf28}+17.98\%}$
test_serialize_model_pickle	0.4460s	0.3847s	2.5997 Ops/s	2.5016 Ops/s	$\color{#35bf28}+3.92\%$
test_serialize_weights	0.2066s	0.1297s	7.7082 Ops/s	9.0064 Ops/s	$\textbf{\color{#d91a1a}-14.42\%}$
test_serialize_weights_returnearly	0.1682s	0.1567s	6.3800 Ops/s	6.3863 Ops/s	$\color{#d91a1a}-0.10\%$
test_serialize_weights_pickle	0.5873s	0.4819s	2.0751 Ops/s	1.2125 Ops/s	$\textbf{\color{#35bf28}+71.15\%}$
test_serialize_weights_filesystem	0.1490s	0.1419s	7.0484 Ops/s	7.1350 Ops/s	$\color{#d91a1a}-1.21\%$
test_serialize_model_filesystem	0.1552s	0.1484s	6.7378 Ops/s	6.4872 Ops/s	$\color{#35bf28}+3.86\%$
test_reshape_pytree	98.7140μs	26.9704μs	37.0777 KOps/s	36.6808 KOps/s	$\color{#35bf28}+1.08\%$
test_reshape_td	69.3290μs	33.0621μs	30.2462 KOps/s	29.9157 KOps/s	$\color{#35bf28}+1.10\%$
test_view_pytree	0.1283ms	27.1488μs	36.8341 KOps/s	37.0390 KOps/s	$\color{#d91a1a}-0.55\%$
test_view_td	0.1007ms	38.1387μs	26.2201 KOps/s	26.5512 KOps/s	$\color{#d91a1a}-1.25\%$
test_unbind_pytree	79.8790μs	29.9415μs	33.3984 KOps/s	33.4545 KOps/s	$\color{#d91a1a}-0.17\%$
test_unbind_td	0.3353ms	39.1160μs	25.5650 KOps/s	26.0550 KOps/s	$\color{#d91a1a}-1.88\%$
test_split_pytree	78.1950μs	30.0056μs	33.3271 KOps/s	33.5186 KOps/s	$\color{#d91a1a}-0.57\%$
test_split_td	0.1050s	54.9622μs	18.1943 KOps/s	21.9165 KOps/s	$\textbf{\color{#d91a1a}-16.98\%}$
test_add_pytree	0.1013ms	35.4349μs	28.2208 KOps/s	28.3242 KOps/s	$\color{#d91a1a}-0.37\%$
test_add_td	0.1098ms	53.4344μs	18.7145 KOps/s	17.9044 KOps/s	$\color{#35bf28}+4.52\%$
test_compile_add_one_nested[tensordict-compile]	0.1322ms	62.1180μs	16.0984 KOps/s	15.9686 KOps/s	$\color{#35bf28}+0.81\%$
test_compile_add_one_nested[tensordict-eager]	0.3401ms	0.1591ms	6.2857 KOps/s	6.2324 KOps/s	$\color{#35bf28}+0.86\%$
test_compile_add_one_nested[pytree-compile]	0.1105ms	44.9685μs	22.2378 KOps/s	21.6161 KOps/s	$\color{#35bf28}+2.88\%$
test_compile_add_one_nested[pytree-eager]	0.2523ms	0.1194ms	8.3772 KOps/s	8.4801 KOps/s	$\color{#d91a1a}-1.21\%$
test_compile_copy_nested[tensordict-compile]	87.2330μs	25.0816μs	39.8698 KOps/s	38.6591 KOps/s	$\color{#35bf28}+3.13\%$
test_compile_copy_nested[tensordict-eager]	0.1743ms	54.1734μs	18.4592 KOps/s	18.3428 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_copy_nested[pytree-compile]	0.1552ms	79.2711μs	12.6149 KOps/s	12.4021 KOps/s	$\color{#35bf28}+1.72\%$
test_compile_copy_nested[pytree-eager]	0.1461ms	68.7421μs	14.5471 KOps/s	14.3592 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_add_one_flat[tensordict-compile]	0.2160ms	0.1047ms	9.5529 KOps/s	9.6325 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_add_one_flat[tensordict-eager]	0.3147ms	0.1974ms	5.0653 KOps/s	5.0380 KOps/s	$\color{#35bf28}+0.54\%$
test_compile_add_one_flat[tensorclass-compile]	0.1196ms	44.7599μs	22.3414 KOps/s	22.1011 KOps/s	$\color{#35bf28}+1.09\%$
test_compile_add_one_flat[tensorclass-eager]	0.4627ms	61.3270μs	16.3060 KOps/s	16.4315 KOps/s	$\color{#d91a1a}-0.76\%$
test_compile_add_one_flat[pytree-compile]	0.1755ms	0.1019ms	9.8129 KOps/s	9.8705 KOps/s	$\color{#d91a1a}-0.58\%$
test_compile_add_one_flat[pytree-eager]	0.6295ms	0.2086ms	4.7943 KOps/s	4.9039 KOps/s	$\color{#d91a1a}-2.24\%$
test_compile_add_self_flat[tensordict-eager]	0.3350ms	0.2072ms	4.8252 KOps/s	4.7913 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_add_self_flat[tensordict-compile]	0.1794ms	0.1042ms	9.5952 KOps/s	9.6249 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_add_self_flat[tensorclass-eager]	0.1313ms	55.6266μs	17.9770 KOps/s	18.3573 KOps/s	$\color{#d91a1a}-2.07\%$
test_compile_add_self_flat[tensorclass-compile]	0.1002ms	46.9173μs	21.3141 KOps/s	21.8843 KOps/s	$\color{#d91a1a}-2.61\%$
test_compile_add_self_flat[pytree-eager]	0.5880ms	0.1607ms	6.2224 KOps/s	6.0662 KOps/s	$\color{#35bf28}+2.57\%$
test_compile_add_self_flat[pytree-compile]	0.2176ms	0.1036ms	9.6498 KOps/s	9.7859 KOps/s	$\color{#d91a1a}-1.39\%$
test_compile_copy_flat[tensordict-compile]	76.3230μs	21.0250μs	47.5625 KOps/s	45.6213 KOps/s	$\color{#35bf28}+4.25\%$
test_compile_copy_flat[tensordict-eager]	0.1432ms	61.0686μs	16.3750 KOps/s	17.0485 KOps/s	$\color{#d91a1a}-3.95\%$
test_compile_copy_flat[pytree-compile]	0.1958ms	82.6631μs	12.0973 KOps/s	12.2146 KOps/s	$\color{#d91a1a}-0.96\%$
test_compile_copy_flat[pytree-eager]	0.1383ms	70.6213μs	14.1600 KOps/s	14.0778 KOps/s	$\color{#35bf28}+0.58\%$
test_compile_assign_and_add[tensordict-compile]	0.3105ms	0.2092ms	4.7806 KOps/s	4.8749 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_assign_and_add[tensordict-eager]	2.0820ms	1.2493ms	800.4385 Ops/s	784.0718 Ops/s	$\color{#35bf28}+2.09\%$
test_compile_assign_and_add[pytree-compile]	0.2942ms	0.2000ms	5.0003 KOps/s	5.0855 KOps/s	$\color{#d91a1a}-1.68\%$
test_compile_assign_and_add[pytree-eager]	1.3419ms	0.7789ms	1.2838 KOps/s	1.2994 KOps/s	$\color{#d91a1a}-1.20\%$
test_compile_assign_and_add_stack[compile]	0.6562ms	0.4472ms	2.2361 KOps/s	2.2483 KOps/s	$\color{#d91a1a}-0.54\%$
test_compile_assign_and_add_stack[eager]	2.6961ms	2.4547ms	407.3824 Ops/s	376.7274 Ops/s	$\textbf{\color{#35bf28}+8.14\%}$
test_compile_indexing[tensor-tensordict-compile]	97.3120μs	35.4999μs	28.1691 KOps/s	28.1006 KOps/s	$\color{#35bf28}+0.24\%$
test_compile_indexing[tensor-tensordict-eager]	0.7444ms	32.9399μs	30.3583 KOps/s	29.9132 KOps/s	$\color{#35bf28}+1.49\%$
test_compile_indexing[tensor-tensorclass-compile]	71.1120μs	29.2299μs	34.2115 KOps/s	34.2921 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_indexing[tensor-tensorclass-eager]	91.3610μs	23.1939μs	43.1148 KOps/s	42.1229 KOps/s	$\color{#35bf28}+2.35\%$
test_compile_indexing[tensor-pytree-compile]	94.8170μs	30.0721μs	33.2535 KOps/s	33.6535 KOps/s	$\color{#d91a1a}-1.19\%$
test_compile_indexing[tensor-pytree-eager]	65.6020μs	23.0430μs	43.3971 KOps/s	42.6397 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_indexing[slice-tensordict-compile]	96.6410μs	51.5221μs	19.4091 KOps/s	19.1733 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_indexing[slice-tensordict-eager]	0.5921ms	20.0117μs	49.9707 KOps/s	48.3471 KOps/s	$\color{#35bf28}+3.36\%$
test_compile_indexing[slice-tensorclass-compile]	0.1115ms	44.3826μs	22.5313 KOps/s	22.4940 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_indexing[slice-tensorclass-eager]	56.0650μs	18.9236μs	52.8440 KOps/s	52.1942 KOps/s	$\color{#35bf28}+1.24\%$
test_compile_indexing[slice-pytree-compile]	0.1155ms	45.2602μs	22.0945 KOps/s	22.1374 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_indexing[slice-pytree-eager]	64.7520μs	18.8473μs	53.0580 KOps/s	50.1821 KOps/s	$\textbf{\color{#35bf28}+5.73\%}$
test_compile_indexing[int-tensordict-compile]	0.1483ms	53.2903μs	18.7651 KOps/s	18.9924 KOps/s	$\color{#d91a1a}-1.20\%$
test_compile_indexing[int-tensordict-eager]	0.9621ms	19.5857μs	51.0576 KOps/s	48.4864 KOps/s	$\textbf{\color{#35bf28}+5.30\%}$
test_compile_indexing[int-tensorclass-compile]	0.1023ms	45.4255μs	22.0141 KOps/s	22.0339 KOps/s	$\color{#d91a1a}-0.09\%$
test_compile_indexing[int-tensorclass-eager]	98.3610μs	18.9433μs	52.7890 KOps/s	52.0633 KOps/s	$\color{#35bf28}+1.39\%$
test_compile_indexing[int-pytree-compile]	0.1327ms	45.9880μs	21.7448 KOps/s	22.1914 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_indexing[int-pytree-eager]	56.4360μs	19.0099μs	52.6041 KOps/s	53.0945 KOps/s	$\color{#d91a1a}-0.92\%$
test_mod_add[eager]	68.8490μs	25.6399μs	39.0016 KOps/s	37.2745 KOps/s	$\color{#35bf28}+4.63\%$
test_mod_add[compile]	95.8390μs	44.8970μs	22.2732 KOps/s	22.3373 KOps/s	$\color{#d91a1a}-0.29\%$
test_mod_add[compile-overhead]	97.7130μs	45.2000μs	22.1239 KOps/s	21.7871 KOps/s	$\color{#35bf28}+1.55\%$
test_mod_wrap[eager]	0.3986ms	0.2149ms	4.6542 KOps/s	4.6612 KOps/s	$\color{#d91a1a}-0.15\%$
test_mod_wrap[compile]	1.5723ms	0.2046ms	4.8873 KOps/s	4.9964 KOps/s	$\color{#d91a1a}-2.18\%$
test_mod_wrap[compile-overhead]	1.7609ms	0.2035ms	4.9130 KOps/s	5.0191 KOps/s	$\color{#d91a1a}-2.11\%$
test_mod_wrap_and_backward[eager]	11.9754ms	10.6536ms	93.8646 Ops/s	87.4241 Ops/s	$\textbf{\color{#35bf28}+7.37\%}$
test_mod_wrap_and_backward[compile]	11.6896ms	10.5638ms	94.6632 Ops/s	84.1309 Ops/s	$\textbf{\color{#35bf28}+12.52\%}$
test_mod_wrap_and_backward[compile-overhead]	11.4314ms	10.4063ms	96.0960 Ops/s	80.0586 Ops/s	$\textbf{\color{#35bf28}+20.03\%}$
test_seq_add[eager]	0.1953ms	91.2079μs	10.9640 KOps/s	10.7331 KOps/s	$\color{#35bf28}+2.15\%$
test_seq_add[compile]	0.1139ms	59.4130μs	16.8313 KOps/s	16.3234 KOps/s	$\color{#35bf28}+3.11\%$
test_seq_add[compile-overhead]	0.1312ms	58.7557μs	17.0196 KOps/s	16.9908 KOps/s	$\color{#35bf28}+0.17\%$
test_seq_wrap[eager]	0.5934ms	0.3800ms	2.6313 KOps/s	2.5431 KOps/s	$\color{#35bf28}+3.47\%$
test_seq_wrap[compile]	0.2982ms	0.2261ms	4.4230 KOps/s	4.4349 KOps/s	$\color{#d91a1a}-0.27\%$
test_seq_wrap[compile-overhead]	0.4016ms	0.2250ms	4.4446 KOps/s	4.5474 KOps/s	$\color{#d91a1a}-2.26\%$
test_func_call_runtime[False-eager]	0.8344ms	0.5541ms	1.8048 KOps/s	1.8832 KOps/s	$\color{#d91a1a}-4.16\%$
test_func_call_runtime[False-compile]	0.5102ms	0.4263ms	2.3457 KOps/s	2.3590 KOps/s	$\color{#d91a1a}-0.56\%$
test_func_call_runtime[False-compile-overhead]	0.5203ms	0.4271ms	2.3416 KOps/s	2.3686 KOps/s	$\color{#d91a1a}-1.14\%$
test_func_call_runtime[True-eager]	1.0061ms	0.7638ms	1.3093 KOps/s	1.3387 KOps/s	$\color{#d91a1a}-2.20\%$
test_func_call_runtime[True-compile]	0.5560ms	0.4654ms	2.1486 KOps/s	2.1563 KOps/s	$\color{#d91a1a}-0.36\%$
test_func_call_runtime[True-compile-overhead]	0.9313ms	0.4669ms	2.1419 KOps/s	2.1772 KOps/s	$\color{#d91a1a}-1.62\%$
test_func_call_cm_runtime[False-eager]	1.2457ms	0.5586ms	1.7903 KOps/s	1.9071 KOps/s	$\textbf{\color{#d91a1a}-6.12\%}$
test_func_call_cm_runtime[False-compile]	0.5457ms	0.4253ms	2.3515 KOps/s	2.3475 KOps/s	$\color{#35bf28}+0.17\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7090ms	0.4270ms	2.3417 KOps/s	2.3676 KOps/s	$\color{#d91a1a}-1.10\%$
test_func_call_cm_runtime[True-eager]	1.0118ms	0.9017ms	1.1090 KOps/s	1.1281 KOps/s	$\color{#d91a1a}-1.69\%$
test_func_call_cm_runtime[True-compile]	0.5972ms	0.4927ms	2.0295 KOps/s	2.0473 KOps/s	$\color{#d91a1a}-0.87\%$
test_func_call_cm_runtime[True-compile-overhead]	1.0060ms	0.4926ms	2.0301 KOps/s	2.0505 KOps/s	$\color{#d91a1a}-0.99\%$
test_vmap_func_call_cm_runtime[eager]	2.4107ms	1.8650ms	536.1878 Ops/s	537.1660 Ops/s	$\color{#d91a1a}-0.18\%$
test_vmap_func_call_cm_runtime[compile]	0.5969ms	0.5130ms	1.9492 KOps/s	1.9244 KOps/s	$\color{#35bf28}+1.29\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.0170ms	0.5156ms	1.9396 KOps/s	1.9484 KOps/s	$\color{#d91a1a}-0.45\%$
test_distributed	0.2462ms	0.1255ms	7.9681 KOps/s	7.8601 KOps/s	$\color{#35bf28}+1.37\%$
test_tdmodule	74.4990μs	17.7425μs	56.3618 KOps/s	52.5846 KOps/s	$\textbf{\color{#35bf28}+7.18\%}$
test_tdmodule_dispatch	55.6440μs	35.4775μs	28.1869 KOps/s	25.9671 KOps/s	$\textbf{\color{#35bf28}+8.55\%}$
test_tdseq	39.5540μs	20.0409μs	49.8981 KOps/s	46.0242 KOps/s	$\textbf{\color{#35bf28}+8.42\%}$
test_tdseq_dispatch	62.7280μs	39.2157μs	25.5000 KOps/s	23.0534 KOps/s	$\textbf{\color{#35bf28}+10.61\%}$
test_instantiation_functorch	1.6644ms	1.5334ms	652.1652 Ops/s	656.0623 Ops/s	$\color{#d91a1a}-0.59\%$
test_exec_functorch	0.3103ms	0.1787ms	5.5948 KOps/s	5.6156 KOps/s	$\color{#d91a1a}-0.37\%$
test_exec_functional_call	0.3473ms	0.1741ms	5.7447 KOps/s	5.9257 KOps/s	$\color{#d91a1a}-3.06\%$
test_exec_td_decorator	0.4703ms	0.2295ms	4.3567 KOps/s	4.5367 KOps/s	$\color{#d91a1a}-3.97\%$
test_vmap_mlp_speed_decorator[True-True]	0.9312ms	0.6386ms	1.5660 KOps/s	1.5932 KOps/s	$\color{#d91a1a}-1.71\%$
test_vmap_mlp_speed_decorator[True-False]	1.0748ms	0.6463ms	1.5473 KOps/s	1.5882 KOps/s	$\color{#d91a1a}-2.57\%$
test_vmap_mlp_speed_decorator[False-True]	0.7340ms	0.5214ms	1.9178 KOps/s	1.9206 KOps/s	$\color{#d91a1a}-0.15\%$
test_vmap_mlp_speed_decorator[False-False]	0.7594ms	0.5223ms	1.9147 KOps/s	1.9457 KOps/s	$\color{#d91a1a}-1.60\%$
test_to_module_speed[True]	1.4470ms	1.2875ms	776.6968 Ops/s	771.5516 Ops/s	$\color{#35bf28}+0.67\%$
test_to_module_speed[False]	2.0236ms	1.2643ms	790.9454 Ops/s	783.6167 Ops/s	$\color{#35bf28}+0.94\%$
test_tc_init	84.3170μs	43.5355μs	22.9697 KOps/s	22.3524 KOps/s	$\color{#35bf28}+2.76\%$
test_tc_init_nested	0.1624ms	87.5218μs	11.4257 KOps/s	11.1046 KOps/s	$\color{#35bf28}+2.89\%$
test_tc_first_layer_tensor	21.4200μs	1.5085μs	662.8960 KOps/s	653.3894 KOps/s	$\color{#35bf28}+1.45\%$
test_tc_first_layer_nontensor	23.1230μs	4.7193μs	211.8957 KOps/s	206.9079 KOps/s	$\color{#35bf28}+2.41\%$
test_tc_second_layer_tensor	24.1350μs	2.8696μs	348.4774 KOps/s	358.8628 KOps/s	$\color{#d91a1a}-2.89\%$
test_tc_second_layer_nontensor	33.6630μs	6.0998μs	163.9392 KOps/s	165.2651 KOps/s	$\color{#d91a1a}-0.80\%$
test_unbind	0.2116s	14.3915ms	69.4856 Ops/s	84.4013 Ops/s	$\textbf{\color{#d91a1a}-17.67\%}$
test_full_like	7.2962ms	6.7293ms	148.6028 Ops/s	148.2838 Ops/s	$\color{#35bf28}+0.22\%$
test_zeros_like	2.9334ms	2.5997ms	384.6617 Ops/s	382.7859 Ops/s	$\color{#35bf28}+0.49\%$
test_ones_like	3.3179ms	3.0283ms	330.2194 Ops/s	330.2229 Ops/s	$-0.00\%$
test_clone	5.0356ms	4.7391ms	211.0120 Ops/s	212.5965 Ops/s	$\color{#d91a1a}-0.75\%$
test_squeeze	59.2310μs	11.8470μs	84.4095 KOps/s	83.0342 KOps/s	$\color{#35bf28}+1.66\%$
test_unsqueeze	0.1751ms	89.1064μs	11.2225 KOps/s	10.9357 KOps/s	$\color{#35bf28}+2.62\%$
test_split	0.5208ms	0.1893ms	5.2838 KOps/s	5.2321 KOps/s	$\color{#35bf28}+0.99\%$
test_permute	0.3607ms	0.2112ms	4.7358 KOps/s	4.6029 KOps/s	$\color{#35bf28}+2.89\%$
test_stack	28.4973ms	24.1040ms	41.4869 Ops/s	40.3889 Ops/s	$\color{#35bf28}+2.72\%$
test_cat	27.6769ms	24.0081ms	41.6525 Ops/s	40.7324 Ops/s	$\color{#35bf28}+2.26\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}11$. Worsened: $\large\color{#d91a1a}24$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	29.4500μs	10.3215μs	96.8848 KOps/s	96.7693 KOps/s	$\color{#35bf28}+0.12\%$
test_plain_set_stack_nested	30.9000μs	10.4484μs	95.7082 KOps/s	95.4751 KOps/s	$\color{#35bf28}+0.24\%$
test_plain_set_nested_inplace	44.6600μs	11.3086μs	88.4283 KOps/s	88.3781 KOps/s	$\color{#35bf28}+0.06\%$
test_plain_set_stack_nested_inplace	35.0510μs	11.1781μs	89.4605 KOps/s	87.6616 KOps/s	$\color{#35bf28}+2.05\%$
test_items	38.2110μs	2.8646μs	349.0830 KOps/s	347.1388 KOps/s	$\color{#35bf28}+0.56\%$
test_items_nested	0.3737ms	0.3188ms	3.1370 KOps/s	3.1547 KOps/s	$\color{#d91a1a}-0.56\%$
test_items_nested_locked	0.3771ms	0.3229ms	3.0971 KOps/s	3.1494 KOps/s	$\color{#d91a1a}-1.66\%$
test_items_nested_leaf	82.5720μs	58.3501μs	17.1379 KOps/s	17.3074 KOps/s	$\color{#d91a1a}-0.98\%$
test_items_stack_nested	0.3779ms	0.3247ms	3.0794 KOps/s	3.1516 KOps/s	$\color{#d91a1a}-2.29\%$
test_items_stack_nested_leaf	86.4510μs	59.5448μs	16.7941 KOps/s	17.3206 KOps/s	$\color{#d91a1a}-3.04\%$
test_items_stack_nested_locked	0.3943ms	0.3239ms	3.0875 KOps/s	3.1325 KOps/s	$\color{#d91a1a}-1.44\%$
test_keys	26.1800μs	3.4626μs	288.8037 KOps/s	290.7804 KOps/s	$\color{#d91a1a}-0.68\%$
test_keys_nested	0.1043ms	70.3305μs	14.2186 KOps/s	14.3637 KOps/s	$\color{#d91a1a}-1.01\%$
test_keys_nested_locked	0.8190ms	75.4502μs	13.2538 KOps/s	13.2516 KOps/s	$\color{#35bf28}+0.02\%$
test_keys_nested_leaf	92.2520μs	61.8556μs	16.1667 KOps/s	16.4087 KOps/s	$\color{#d91a1a}-1.47\%$
test_keys_stack_nested	0.1115ms	70.4725μs	14.1899 KOps/s	14.2395 KOps/s	$\color{#d91a1a}-0.35\%$
test_keys_stack_nested_leaf	96.9720μs	62.2749μs	16.0578 KOps/s	16.6544 KOps/s	$\color{#d91a1a}-3.58\%$
test_keys_stack_nested_locked	0.1201ms	76.0401μs	13.1510 KOps/s	13.2114 KOps/s	$\color{#d91a1a}-0.46\%$
test_values	6.4867μs	0.8530μs	1.1723 MOps/s	1.1675 MOps/s	$\color{#35bf28}+0.41\%$
test_values_nested	64.8710μs	31.1950μs	32.0564 KOps/s	32.3432 KOps/s	$\color{#d91a1a}-0.89\%$
test_values_nested_locked	67.0720μs	32.7935μs	30.4939 KOps/s	30.6440 KOps/s	$\color{#d91a1a}-0.49\%$
test_values_nested_leaf	79.5120μs	33.7350μs	29.6428 KOps/s	30.0357 KOps/s	$\color{#d91a1a}-1.31\%$
test_values_stack_nested	65.2020μs	31.7709μs	31.4754 KOps/s	32.2054 KOps/s	$\color{#d91a1a}-2.27\%$
test_values_stack_nested_leaf	71.9120μs	34.0256μs	29.3896 KOps/s	29.9466 KOps/s	$\color{#d91a1a}-1.86\%$
test_values_stack_nested_locked	62.7710μs	33.6976μs	29.6757 KOps/s	30.4965 KOps/s	$\color{#d91a1a}-2.69\%$
test_membership	4.7585μs	0.5035μs	1.9862 MOps/s	1.9573 MOps/s	$\color{#35bf28}+1.48\%$
test_membership_nested	19.0255μs	1.9084μs	524.0011 KOps/s	511.1840 KOps/s	$\color{#35bf28}+2.51\%$
test_membership_nested_leaf	17.7355μs	1.9104μs	523.4389 KOps/s	531.6737 KOps/s	$\color{#d91a1a}-1.55\%$
test_membership_stacked_nested	38.9210μs	1.9793μs	505.2379 KOps/s	493.7857 KOps/s	$\color{#35bf28}+2.32\%$
test_membership_stacked_nested_leaf	36.7900μs	1.9539μs	511.7996 KOps/s	492.3153 KOps/s	$\color{#35bf28}+3.96\%$
test_membership_nested_last	47.1810μs	2.8159μs	355.1249 KOps/s	356.8732 KOps/s	$\color{#d91a1a}-0.49\%$
test_membership_nested_leaf_last	33.5510μs	2.8272μs	353.7110 KOps/s	355.4504 KOps/s	$\color{#d91a1a}-0.49\%$
test_membership_stacked_nested_last	32.1110μs	2.7894μs	358.5042 KOps/s	355.3804 KOps/s	$\color{#35bf28}+0.88\%$
test_membership_stacked_nested_leaf_last	41.2100μs	2.8141μs	355.3558 KOps/s	352.4309 KOps/s	$\color{#35bf28}+0.83\%$
test_nested_getleaf	38.4710μs	5.9825μs	167.1533 KOps/s	165.8154 KOps/s	$\color{#35bf28}+0.81\%$
test_nested_get	40.6510μs	5.6728μs	176.2799 KOps/s	175.5891 KOps/s	$\color{#35bf28}+0.39\%$
test_stacked_getleaf	39.4100μs	6.0191μs	166.1391 KOps/s	167.6686 KOps/s	$\color{#d91a1a}-0.91\%$
test_stacked_get	35.7010μs	5.7278μs	174.5866 KOps/s	176.7040 KOps/s	$\color{#d91a1a}-1.20\%$
test_nested_getitemleaf	38.1710μs	6.0715μs	164.7038 KOps/s	164.7211 KOps/s	$\color{#d91a1a}-0.01\%$
test_nested_getitem	41.0700μs	5.8030μs	172.3241 KOps/s	174.0449 KOps/s	$\color{#d91a1a}-0.99\%$
test_stacked_getitemleaf	38.2300μs	6.0850μs	164.3384 KOps/s	165.3812 KOps/s	$\color{#d91a1a}-0.63\%$
test_stacked_getitem	35.6800μs	5.7779μs	173.0728 KOps/s	174.1964 KOps/s	$\color{#d91a1a}-0.64\%$
test_lock_nested	0.7018ms	0.3655ms	2.7358 KOps/s	2.7322 KOps/s	$\color{#35bf28}+0.13\%$
test_lock_stack_nested	0.3900ms	0.3351ms	2.9841 KOps/s	2.9944 KOps/s	$\color{#d91a1a}-0.34\%$
test_unlock_nested	0.6042ms	0.3060ms	3.2682 KOps/s	3.2897 KOps/s	$\color{#d91a1a}-0.65\%$
test_unlock_stack_nested	0.3254ms	0.2730ms	3.6632 KOps/s	3.6646 KOps/s	$\color{#d91a1a}-0.04\%$
test_flatten_speed	0.1116ms	72.5417μs	13.7852 KOps/s	13.8544 KOps/s	$\color{#d91a1a}-0.50\%$
test_unflatten_speed	0.3414ms	0.2911ms	3.4353 KOps/s	3.4710 KOps/s	$\color{#d91a1a}-1.03\%$
test_common_ops	1.5838ms	0.5776ms	1.7314 KOps/s	1.7646 KOps/s	$\color{#d91a1a}-1.88\%$
test_creation	0.1022ms	1.4636μs	683.2597 KOps/s	677.2743 KOps/s	$\color{#35bf28}+0.88\%$
test_creation_empty	40.3410μs	6.8215μs	146.5954 KOps/s	143.0655 KOps/s	$\color{#35bf28}+2.47\%$
test_creation_nested_1	38.8400μs	8.3976μs	119.0818 KOps/s	118.6975 KOps/s	$\color{#35bf28}+0.32\%$
test_creation_nested_2	46.9900μs	10.8138μs	92.4748 KOps/s	92.6646 KOps/s	$\color{#d91a1a}-0.20\%$
test_clone	33.4210μs	11.0232μs	90.7181 KOps/s	99.2873 KOps/s	$\textbf{\color{#d91a1a}-8.63\%}$
test_getitem[int]	1.6650ms	10.6739μs	93.6868 KOps/s	95.5162 KOps/s	$\color{#d91a1a}-1.92\%$
test_getitem[slice_int]	0.1435ms	20.8140μs	48.0446 KOps/s	50.7924 KOps/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_getitem[range]	0.1412ms	37.4027μs	26.7360 KOps/s	27.5672 KOps/s	$\color{#d91a1a}-3.02\%$
test_getitem[tuple]	0.1087ms	18.1370μs	55.1360 KOps/s	56.6831 KOps/s	$\color{#d91a1a}-2.73\%$
test_getitem[list]	0.1540ms	33.1334μs	30.1810 KOps/s	31.2008 KOps/s	$\color{#d91a1a}-3.27\%$
test_setitem_dim[int]	44.3300μs	18.9891μs	52.6617 KOps/s	55.5214 KOps/s	$\textbf{\color{#d91a1a}-5.15\%}$
test_setitem_dim[slice_int]	61.0910μs	38.5539μs	25.9377 KOps/s	27.4547 KOps/s	$\textbf{\color{#d91a1a}-5.53\%}$
test_setitem_dim[range]	77.9920μs	53.0691μs	18.8434 KOps/s	19.1450 KOps/s	$\color{#d91a1a}-1.58\%$
test_setitem_dim[tuple]	51.6110μs	30.7532μs	32.5169 KOps/s	32.0078 KOps/s	$\color{#35bf28}+1.59\%$
test_setitem	0.1239ms	15.1332μs	66.0801 KOps/s	72.2250 KOps/s	$\textbf{\color{#d91a1a}-8.51\%}$
test_set	0.1238ms	14.7563μs	67.7677 KOps/s	74.5405 KOps/s	$\textbf{\color{#d91a1a}-9.09\%}$
test_set_shared	1.5373ms	0.1467ms	6.8171 KOps/s	6.8735 KOps/s	$\color{#d91a1a}-0.82\%$
test_update	0.3736ms	16.8890μs	59.2103 KOps/s	63.4912 KOps/s	$\textbf{\color{#d91a1a}-6.74\%}$
test_update_nested	0.1248ms	21.5265μs	46.4544 KOps/s	49.2330 KOps/s	$\textbf{\color{#d91a1a}-5.64\%}$
test_update__nested	1.1198ms	25.3720μs	39.4135 KOps/s	42.2399 KOps/s	$\textbf{\color{#d91a1a}-6.69\%}$
test_set_nested	0.1193ms	15.8301μs	63.1706 KOps/s	68.5737 KOps/s	$\textbf{\color{#d91a1a}-7.88\%}$
test_set_nested_new	0.1212ms	18.0669μs	55.3497 KOps/s	60.0782 KOps/s	$\textbf{\color{#d91a1a}-7.87\%}$
test_select	0.1330ms	29.9509μs	33.3880 KOps/s	34.7783 KOps/s	$\color{#d91a1a}-4.00\%$
test_select_nested	71.4810μs	42.1907μs	23.7019 KOps/s	23.9876 KOps/s	$\color{#d91a1a}-1.19\%$
test_exclude_nested	0.5829ms	59.5149μs	16.8025 KOps/s	16.7149 KOps/s	$\color{#35bf28}+0.52\%$
test_empty[True]	0.3075ms	0.2574ms	3.8848 KOps/s	3.8900 KOps/s	$\color{#d91a1a}-0.14\%$
test_empty[False]	3.3950μs	0.7536μs	1.3269 MOps/s	1.3343 MOps/s	$\color{#d91a1a}-0.55\%$
test_to	83.2210μs	54.9283μs	18.2056 KOps/s	18.8192 KOps/s	$\color{#d91a1a}-3.26\%$
test_to_nonblocking	94.4210μs	45.8725μs	21.7996 KOps/s	21.9137 KOps/s	$\color{#d91a1a}-0.52\%$
test_unbind_speed	0.8719ms	0.2292ms	4.3638 KOps/s	4.3799 KOps/s	$\color{#d91a1a}-0.37\%$
test_unbind_speed_stack0	0.3008ms	0.2332ms	4.2890 KOps/s	4.4171 KOps/s	$\color{#d91a1a}-2.90\%$
test_unbind_speed_stack1	93.5861ms	0.6537ms	1.5297 KOps/s	1.5420 KOps/s	$\color{#d91a1a}-0.80\%$
test_split	95.7630ms	1.7556ms	569.5998 Ops/s	654.4073 Ops/s	$\textbf{\color{#d91a1a}-12.96\%}$
test_chunk	1.6537ms	1.4755ms	677.7286 Ops/s	598.3911 Ops/s	$\textbf{\color{#35bf28}+13.26\%}$
test_consolidate[False-None]	97.8991ms	2.9024ms	344.5451 Ops/s	390.1698 Ops/s	$\textbf{\color{#d91a1a}-11.69\%}$
test_consolidate[default-None]	1.7602ms	1.6567ms	603.6210 Ops/s	614.4838 Ops/s	$\color{#d91a1a}-1.77\%$
test_consolidate[reduce-overhead-None]	1.7602ms	1.6865ms	592.9302 Ops/s	595.0022 Ops/s	$\color{#d91a1a}-0.35\%$
test_consolidate_njt[False-None]	6.6463ms	6.4524ms	154.9814 Ops/s	155.8723 Ops/s	$\color{#d91a1a}-0.57\%$
test_to[False-False-None]	1.7806ms	1.6799ms	595.2836 Ops/s	591.5503 Ops/s	$\color{#35bf28}+0.63\%$
test_to[True-False-None]	1.4843ms	1.2656ms	790.1085 Ops/s	802.1462 Ops/s	$\color{#d91a1a}-1.50\%$
test_to[within-False-None]	4.2239ms	3.9530ms	252.9723 Ops/s	254.5127 Ops/s	$\color{#d91a1a}-0.61\%$
test_to[True-default-None]	5.3619ms	5.0909ms	196.4308 Ops/s	198.3677 Ops/s	$\color{#d91a1a}-0.98\%$
test_to_njt[False-False-None]	7.1207ms	6.8842ms	145.2610 Ops/s	144.0818 Ops/s	$\color{#35bf28}+0.82\%$
test_to_njt[True-False-None]	5.9195ms	5.4672ms	182.9084 Ops/s	183.7201 Ops/s	$\color{#d91a1a}-0.44\%$
test_to_njt[within-False-None]	12.2629ms	11.9930ms	83.3820 Ops/s	82.8748 Ops/s	$\color{#35bf28}+0.61\%$
test_creation[device0]	0.3708ms	78.9031μs	12.6738 KOps/s	12.8656 KOps/s	$\color{#d91a1a}-1.49\%$
test_creation_from_tensor	0.6203ms	82.0786μs	12.1834 KOps/s	12.0997 KOps/s	$\color{#35bf28}+0.69\%$
test_add_one[memmap_tensor0]	0.4098ms	6.8163μs	146.7075 KOps/s	153.6255 KOps/s	$\color{#d91a1a}-4.50\%$
test_contiguous[memmap_tensor0]	1.8156μs	0.3972μs	2.5173 MOps/s	2.5043 MOps/s	$\color{#35bf28}+0.52\%$
test_stack[memmap_tensor0]	44.7210μs	4.3771μs	228.4606 KOps/s	228.6125 KOps/s	$\color{#d91a1a}-0.07\%$
test_memmaptd_index	1.6500ms	0.2444ms	4.0921 KOps/s	4.0432 KOps/s	$\color{#35bf28}+1.21\%$
test_memmaptd_index_astensor	0.5764ms	0.2987ms	3.3473 KOps/s	3.2823 KOps/s	$\color{#35bf28}+1.98\%$
test_memmaptd_index_op	1.0368ms	0.5647ms	1.7708 KOps/s	1.7842 KOps/s	$\color{#d91a1a}-0.75\%$
test_serialize_model	0.1322s	0.1311s	7.6279 Ops/s	7.6131 Ops/s	$\color{#35bf28}+0.19\%$
test_serialize_model_pickle	1.3763s	1.1914s	0.8394 Ops/s	0.8238 Ops/s	$\color{#35bf28}+1.89\%$
test_serialize_weights	0.4065s	0.1696s	5.8953 Ops/s	7.7073 Ops/s	$\textbf{\color{#d91a1a}-23.51\%}$
test_serialize_weights_returnearly	0.3344s	52.4488ms	19.0662 Ops/s	15.1855 Ops/s	$\textbf{\color{#35bf28}+25.55\%}$
test_serialize_weights_pickle	1.3470s	1.2212s	0.8189 Ops/s	0.8231 Ops/s	$\color{#d91a1a}-0.51\%$
test_reshape_pytree	49.5110μs	22.1468μs	45.1533 KOps/s	42.1101 KOps/s	$\textbf{\color{#35bf28}+7.23\%}$
test_reshape_td	71.9010μs	26.5254μs	37.6997 KOps/s	37.5256 KOps/s	$\color{#35bf28}+0.46\%$
test_view_pytree	74.0220μs	22.1466μs	45.1536 KOps/s	45.6736 KOps/s	$\color{#d91a1a}-1.14\%$
test_view_td	62.5010μs	31.1527μs	32.1000 KOps/s	33.1663 KOps/s	$\color{#d91a1a}-3.21\%$
test_unbind_pytree	0.1091ms	28.2560μs	35.3907 KOps/s	35.9409 KOps/s	$\color{#d91a1a}-1.53\%$
test_unbind_td	0.7831ms	35.9994μs	27.7782 KOps/s	28.4712 KOps/s	$\color{#d91a1a}-2.43\%$
test_split_pytree	74.8910μs	30.5871μs	32.6935 KOps/s	30.5791 KOps/s	$\textbf{\color{#35bf28}+6.91\%}$
test_split_td	0.7650ms	38.8590μs	25.7341 KOps/s	25.8796 KOps/s	$\color{#d91a1a}-0.56\%$
test_add_pytree	71.7120μs	35.5365μs	28.1401 KOps/s	29.7981 KOps/s	$\textbf{\color{#d91a1a}-5.56\%}$
test_add_td	0.1937ms	48.4486μs	20.6404 KOps/s	21.3099 KOps/s	$\color{#d91a1a}-3.14\%$
test_compile_add_one_nested[tensordict-compile]	0.1797ms	0.1245ms	8.0295 KOps/s	8.2042 KOps/s	$\color{#d91a1a}-2.13\%$
test_compile_add_one_nested[tensordict-eager]	0.5089ms	0.1252ms	7.9873 KOps/s	7.9664 KOps/s	$\color{#35bf28}+0.26\%$
test_compile_add_one_nested[pytree-compile]	0.1630ms	95.4988μs	10.4713 KOps/s	10.6329 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_add_one_nested[pytree-eager]	0.5438ms	0.1503ms	6.6525 KOps/s	6.6501 KOps/s	$\color{#35bf28}+0.04\%$
test_compile_copy_nested[tensordict-compile]	64.8510μs	27.1965μs	36.7694 KOps/s	42.3143 KOps/s	$\textbf{\color{#d91a1a}-13.10\%}$
test_compile_copy_nested[tensordict-eager]	0.4068ms	26.4112μs	37.8627 KOps/s	36.9423 KOps/s	$\color{#35bf28}+2.49\%$
test_compile_copy_nested[pytree-compile]	0.1553ms	64.4018μs	15.5275 KOps/s	15.2574 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_copy_nested[pytree-eager]	94.2520μs	49.2376μs	20.3097 KOps/s	20.1671 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_add_one_flat[tensordict-compile]	0.1820ms	0.1420ms	7.0427 KOps/s	7.0240 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_add_one_flat[tensordict-eager]	0.3222ms	0.2094ms	4.7745 KOps/s	4.8283 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_add_one_flat[tensorclass-compile]	0.1542ms	96.8657μs	10.3236 KOps/s	10.3492 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_one_flat[tensorclass-eager]	0.4345ms	51.6844μs	19.3482 KOps/s	19.6156 KOps/s	$\color{#d91a1a}-1.36\%$
test_compile_add_one_flat[pytree-compile]	0.1863ms	0.1349ms	7.4134 KOps/s	7.3794 KOps/s	$\color{#35bf28}+0.46\%$
test_compile_add_one_flat[pytree-eager]	0.8597ms	0.4809ms	2.0793 KOps/s	2.0873 KOps/s	$\color{#d91a1a}-0.38\%$
test_compile_add_self_flat[tensordict-eager]	0.6622ms	0.2491ms	4.0142 KOps/s	4.0260 KOps/s	$\color{#d91a1a}-0.29\%$
test_compile_add_self_flat[tensordict-compile]	0.2055ms	0.1443ms	6.9305 KOps/s	6.9686 KOps/s	$\color{#d91a1a}-0.55\%$
test_compile_add_self_flat[tensorclass-eager]	0.1492ms	61.5633μs	16.2434 KOps/s	16.2955 KOps/s	$\color{#d91a1a}-0.32\%$
test_compile_add_self_flat[tensorclass-compile]	0.1491ms	98.0459μs	10.1993 KOps/s	10.2326 KOps/s	$\color{#d91a1a}-0.33\%$
test_compile_add_self_flat[pytree-eager]	0.8093ms	0.4093ms	2.4431 KOps/s	2.4747 KOps/s	$\color{#d91a1a}-1.28\%$
test_compile_add_self_flat[pytree-compile]	0.1862ms	0.1357ms	7.3684 KOps/s	7.4920 KOps/s	$\color{#d91a1a}-1.65\%$
test_compile_copy_flat[tensordict-compile]	0.4069ms	19.8336μs	50.4195 KOps/s	54.5498 KOps/s	$\textbf{\color{#d91a1a}-7.57\%}$
test_compile_copy_flat[tensordict-eager]	0.4253ms	27.2058μs	36.7568 KOps/s	36.9175 KOps/s	$\color{#d91a1a}-0.44\%$
test_compile_copy_flat[pytree-compile]	0.4356ms	69.2194μs	14.4468 KOps/s	14.1178 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_copy_flat[pytree-eager]	0.4310ms	51.3721μs	19.4658 KOps/s	19.2500 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_assign_and_add[tensordict-compile]	1.6708ms	0.3996ms	2.5028 KOps/s	2.2310 KOps/s	$\textbf{\color{#35bf28}+12.18\%}$
test_compile_assign_and_add[tensordict-eager]	3.0683ms	2.6481ms	377.6348 Ops/s	392.7003 Ops/s	$\color{#d91a1a}-3.84\%$
test_compile_assign_and_add[pytree-compile]	1.6135ms	0.3845ms	2.6007 KOps/s	2.2623 KOps/s	$\textbf{\color{#35bf28}+14.96\%}$
test_compile_assign_and_add[pytree-eager]	3.0107ms	2.7133ms	368.5493 Ops/s	383.0025 Ops/s	$\color{#d91a1a}-3.77\%$
test_compile_indexing[tensor-tensordict-compile]	0.5338ms	0.1201ms	8.3263 KOps/s	8.8274 KOps/s	$\textbf{\color{#d91a1a}-5.68\%}$
test_compile_indexing[tensor-tensordict-eager]	0.5740ms	82.2440μs	12.1589 KOps/s	11.8452 KOps/s	$\color{#35bf28}+2.65\%$
test_compile_indexing[tensor-tensorclass-compile]	0.5677ms	0.1066ms	9.3794 KOps/s	9.0559 KOps/s	$\color{#35bf28}+3.57\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1182ms	68.3020μs	14.6409 KOps/s	13.9250 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_compile_indexing[tensor-pytree-compile]	0.1701ms	0.1123ms	8.9031 KOps/s	9.0357 KOps/s	$\color{#d91a1a}-1.47\%$
test_compile_indexing[tensor-pytree-eager]	0.1501ms	72.2577μs	13.8394 KOps/s	13.9795 KOps/s	$\color{#d91a1a}-1.00\%$
test_compile_indexing[slice-tensordict-compile]	0.1395ms	0.1002ms	9.9842 KOps/s	9.9869 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_indexing[slice-tensordict-eager]	0.1450ms	17.2051μs	58.1224 KOps/s	57.9483 KOps/s	$\color{#35bf28}+0.30\%$
test_compile_indexing[slice-tensorclass-compile]	0.1807ms	94.2269μs	10.6127 KOps/s	10.1447 KOps/s	$\color{#35bf28}+4.61\%$
test_compile_indexing[slice-tensorclass-eager]	49.4610μs	15.8170μs	63.2233 KOps/s	64.4646 KOps/s	$\color{#d91a1a}-1.93\%$
test_compile_indexing[slice-pytree-compile]	0.1440ms	94.7152μs	10.5580 KOps/s	10.5185 KOps/s	$\color{#35bf28}+0.37\%$
test_compile_indexing[slice-pytree-eager]	54.2910μs	15.8717μs	63.0052 KOps/s	64.5898 KOps/s	$\color{#d91a1a}-2.45\%$
test_compile_indexing[int-tensordict-compile]	0.1982ms	0.1028ms	9.7279 KOps/s	9.9248 KOps/s	$\color{#d91a1a}-1.98\%$
test_compile_indexing[int-tensordict-eager]	0.5871ms	16.9881μs	58.8649 KOps/s	59.2851 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_indexing[int-tensorclass-compile]	0.1523ms	96.6223μs	10.3496 KOps/s	10.4903 KOps/s	$\color{#d91a1a}-1.34\%$
test_compile_indexing[int-tensorclass-eager]	45.1710μs	15.8798μs	62.9730 KOps/s	64.5141 KOps/s	$\color{#d91a1a}-2.39\%$
test_compile_indexing[int-pytree-compile]	0.2087ms	95.1747μs	10.5070 KOps/s	10.4836 KOps/s	$\color{#35bf28}+0.22\%$
test_compile_indexing[int-pytree-eager]	55.9610μs	15.8025μs	63.2811 KOps/s	63.7004 KOps/s	$\color{#d91a1a}-0.66\%$
test_mod_add[eager]	79.6210μs	32.4238μs	30.8416 KOps/s	32.7675 KOps/s	$\textbf{\color{#d91a1a}-5.88\%}$
test_mod_add[compile]	0.3472ms	76.1309μs	13.1353 KOps/s	13.2256 KOps/s	$\color{#d91a1a}-0.68\%$
test_mod_add[compile-overhead]	0.3158ms	0.1695ms	5.9001 KOps/s	5.6275 KOps/s	$\color{#35bf28}+4.84\%$
test_mod_wrap[eager]	0.3243ms	0.2417ms	4.1373 KOps/s	4.1462 KOps/s	$\color{#d91a1a}-0.21\%$
test_mod_wrap[compile]	1.5727ms	0.2792ms	3.5817 KOps/s	3.4880 KOps/s	$\color{#35bf28}+2.69\%$
test_mod_wrap[compile-overhead]	7.2131ms	3.7614ms	265.8607 Ops/s	265.5484 Ops/s	$\color{#35bf28}+0.12\%$
test_mod_wrap_and_backward[eager]	1.4814ms	1.3519ms	739.7215 Ops/s	696.5994 Ops/s	$\textbf{\color{#35bf28}+6.19\%}$
test_mod_wrap_and_backward[compile]	1.3551ms	1.2545ms	797.1305 Ops/s	735.5250 Ops/s	$\textbf{\color{#35bf28}+8.38\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3548ms	0.9002ms	1.1109 KOps/s	982.7674 Ops/s	$\textbf{\color{#35bf28}+13.04\%}$
test_seq_add[eager]	0.1687ms	98.6532μs	10.1365 KOps/s	10.4979 KOps/s	$\color{#d91a1a}-3.44\%$
test_seq_add[compile]	0.1360ms	88.2197μs	11.3353 KOps/s	11.4541 KOps/s	$\color{#d91a1a}-1.04\%$
test_seq_add[compile-overhead]	0.1816ms	0.1334ms	7.4989 KOps/s	7.9180 KOps/s	$\textbf{\color{#d91a1a}-5.29\%}$
test_seq_wrap[eager]	0.4615ms	0.3958ms	2.5263 KOps/s	2.5030 KOps/s	$\color{#35bf28}+0.93\%$
test_seq_wrap[compile]	0.3900ms	0.3094ms	3.2317 KOps/s	3.3427 KOps/s	$\color{#d91a1a}-3.32\%$
test_seq_wrap[compile-overhead]	0.2771ms	0.2200ms	4.5444 KOps/s	4.5199 KOps/s	$\color{#35bf28}+0.54\%$
test_func_call_runtime[False-eager]	0.8926ms	0.7837ms	1.2760 KOps/s	1.3567 KOps/s	$\textbf{\color{#d91a1a}-5.95\%}$
test_func_call_runtime[False-compile]	0.8391ms	0.7394ms	1.3525 KOps/s	1.3561 KOps/s	$\color{#d91a1a}-0.26\%$
test_func_call_runtime[False-compile-overhead]	0.4353ms	0.3567ms	2.8038 KOps/s	2.7916 KOps/s	$\color{#35bf28}+0.44\%$
test_func_call_runtime[True-eager]	0.9891ms	0.8923ms	1.1208 KOps/s	1.1057 KOps/s	$\color{#35bf28}+1.36\%$
test_func_call_runtime[True-compile]	0.8473ms	0.7536ms	1.3270 KOps/s	1.3208 KOps/s	$\color{#35bf28}+0.47\%$
test_func_call_runtime[True-compile-overhead]	0.4415ms	0.3777ms	2.6476 KOps/s	2.6534 KOps/s	$\color{#d91a1a}-0.22\%$
test_func_call_cm_runtime[False-eager]	0.8685ms	0.7316ms	1.3669 KOps/s	1.3571 KOps/s	$\color{#35bf28}+0.72\%$
test_func_call_cm_runtime[False-compile]	0.8325ms	0.7406ms	1.3503 KOps/s	1.3522 KOps/s	$\color{#d91a1a}-0.14\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4083ms	0.3575ms	2.7970 KOps/s	2.7718 KOps/s	$\color{#35bf28}+0.91\%$
test_func_call_cm_runtime[True-eager]	1.1323ms	0.9965ms	1.0035 KOps/s	992.7250 Ops/s	$\color{#35bf28}+1.08\%$
test_func_call_cm_runtime[True-compile]	0.8536ms	0.7883ms	1.2685 KOps/s	1.2759 KOps/s	$\color{#d91a1a}-0.58\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4751ms	0.4024ms	2.4854 KOps/s	2.4621 KOps/s	$\color{#35bf28}+0.95\%$
test_vmap_func_call_cm_runtime[eager]	2.5425ms	2.0698ms	483.1315 Ops/s	480.5992 Ops/s	$\color{#35bf28}+0.53\%$
test_vmap_func_call_cm_runtime[compile]	0.8948ms	0.8021ms	1.2468 KOps/s	1.2577 KOps/s	$\color{#d91a1a}-0.87\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5083ms	0.4102ms	2.4380 KOps/s	2.4539 KOps/s	$\color{#d91a1a}-0.65\%$
test_distributed	1.7731ms	0.2168ms	4.6132 KOps/s	8.7636 KOps/s	$\textbf{\color{#d91a1a}-47.36\%}$
test_tdmodule	0.1176ms	13.5849μs	73.6110 KOps/s	70.2065 KOps/s	$\color{#35bf28}+4.85\%$
test_tdmodule_dispatch	54.5310μs	26.7683μs	37.3577 KOps/s	37.1569 KOps/s	$\color{#35bf28}+0.54\%$
test_tdseq	33.6010μs	15.3507μs	65.1436 KOps/s	68.5052 KOps/s	$\color{#d91a1a}-4.91\%$
test_tdseq_dispatch	53.1410μs	29.9925μs	33.3417 KOps/s	34.3472 KOps/s	$\color{#d91a1a}-2.93\%$
test_instantiation_functorch	1.6494ms	1.5454ms	647.0663 Ops/s	652.3996 Ops/s	$\color{#d91a1a}-0.82\%$
test_exec_functorch	0.2086ms	0.1480ms	6.7573 KOps/s	7.1028 KOps/s	$\color{#d91a1a}-4.86\%$
test_exec_functional_call	0.1886ms	0.1395ms	7.1686 KOps/s	7.4112 KOps/s	$\color{#d91a1a}-3.27\%$
test_exec_td_decorator	0.3710ms	0.1848ms	5.4098 KOps/s	5.5665 KOps/s	$\color{#d91a1a}-2.81\%$
test_vmap_mlp_speed_decorator[True-True]	0.7477ms	0.6774ms	1.4762 KOps/s	1.4268 KOps/s	$\color{#35bf28}+3.46\%$
test_vmap_mlp_speed_decorator[True-False]	0.7938ms	0.6871ms	1.4555 KOps/s	1.4272 KOps/s	$\color{#35bf28}+1.98\%$
test_vmap_mlp_speed_decorator[False-True]	0.7017ms	0.5927ms	1.6872 KOps/s	1.6163 KOps/s	$\color{#35bf28}+4.39\%$
test_vmap_mlp_speed_decorator[False-False]	0.7339ms	0.6070ms	1.6474 KOps/s	1.6207 KOps/s	$\color{#35bf28}+1.64\%$
test_vmap_transformer_speed_decorator[True-True]	19.2643ms	19.1598ms	52.1927 Ops/s	52.1995 Ops/s	$\color{#d91a1a}-0.01\%$
test_vmap_transformer_speed_decorator[True-False]	19.9084ms	19.2026ms	52.0762 Ops/s	52.1781 Ops/s	$\color{#d91a1a}-0.20\%$
test_vmap_transformer_speed_decorator[False-True]	19.7217ms	19.1612ms	52.1888 Ops/s	52.5919 Ops/s	$\color{#d91a1a}-0.77\%$
test_vmap_transformer_speed_decorator[False-False]	19.6729ms	19.1025ms	52.3491 Ops/s	52.2363 Ops/s	$\color{#35bf28}+0.22\%$
test_to_module_speed[True]	1.0642ms	0.9374ms	1.0668 KOps/s	1.0584 KOps/s	$\color{#35bf28}+0.79\%$
test_to_module_speed[False]	1.3039ms	0.9201ms	1.0869 KOps/s	1.0721 KOps/s	$\color{#35bf28}+1.37\%$
test_tc_init	59.8110μs	33.6181μs	29.7459 KOps/s	28.5435 KOps/s	$\color{#35bf28}+4.21\%$
test_tc_init_nested	0.1668ms	68.9871μs	14.4955 KOps/s	14.1561 KOps/s	$\color{#35bf28}+2.40\%$
test_tc_first_layer_tensor	5.0344μs	0.6974μs	1.4339 MOps/s	1.4296 MOps/s	$\color{#35bf28}+0.30\%$
test_tc_first_layer_nontensor	25.5100μs	2.3038μs	434.0706 KOps/s	436.6491 KOps/s	$\color{#d91a1a}-0.59\%$
test_tc_second_layer_tensor	30.2880μs	1.3970μs	715.8370 KOps/s	701.1565 KOps/s	$\color{#35bf28}+2.09\%$
test_tc_second_layer_nontensor	28.9000μs	3.0160μs	331.5666 KOps/s	330.1466 KOps/s	$\color{#35bf28}+0.43\%$
test_unbind	0.2234s	9.9805ms	100.1953 Ops/s	152.3695 Ops/s	$\textbf{\color{#d91a1a}-34.24\%}$
test_full_like	9.3906ms	9.0635ms	110.3328 Ops/s	109.3104 Ops/s	$\color{#35bf28}+0.94\%$
test_zeros_like	5.5592ms	4.3065ms	232.2051 Ops/s	137.6844 Ops/s	$\textbf{\color{#35bf28}+68.65\%}$
test_ones_like	4.9869ms	4.3131ms	231.8515 Ops/s	232.0197 Ops/s	$\color{#d91a1a}-0.07\%$
test_clone	6.7051ms	6.2586ms	159.7811 Ops/s	159.2780 Ops/s	$\color{#35bf28}+0.32\%$
test_squeeze	60.1110μs	9.2417μs	108.2047 KOps/s	109.0928 KOps/s	$\color{#d91a1a}-0.81\%$
test_unsqueeze	0.1215ms	70.4829μs	14.1878 KOps/s	13.8679 KOps/s	$\color{#35bf28}+2.31\%$
test_split	0.3574ms	0.1551ms	6.4466 KOps/s	6.3860 KOps/s	$\color{#35bf28}+0.95\%$
test_permute	0.2318ms	0.1718ms	5.8199 KOps/s	5.7316 KOps/s	$\color{#35bf28}+1.54\%$
test_stack	53.5104ms	53.2133ms	18.7923 Ops/s	19.9793 Ops/s	$\textbf{\color{#d91a1a}-5.94\%}$
test_cat	53.3456ms	51.7514ms	19.3232 Ops/s	20.0298 Ops/s	$\color{#d91a1a}-3.53\%$