[BugFix] smarter check in set_interaction_type

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] smarter check in set_interaction_type #1088

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}24$. Worsened: $\large\color{#d91a1a}16$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	35.8980μs	17.4099μs	57.4384 KOps/s	59.5026 KOps/s	$\color{#d91a1a}-3.47\%$
test_plain_set_stack_nested	51.4680μs	17.5791μs	56.8858 KOps/s	59.6125 KOps/s	$\color{#d91a1a}-4.57\%$
test_plain_set_nested_inplace	52.2180μs	19.3406μs	51.7047 KOps/s	53.2251 KOps/s	$\color{#d91a1a}-2.86\%$
test_plain_set_stack_nested_inplace	59.7520μs	19.3289μs	51.7360 KOps/s	52.6789 KOps/s	$\color{#d91a1a}-1.79\%$
test_items	30.1560μs	4.1204μs	242.6957 KOps/s	240.8965 KOps/s	$\color{#35bf28}+0.75\%$
test_items_nested	0.6654ms	0.3400ms	2.9409 KOps/s	2.8936 KOps/s	$\color{#35bf28}+1.64\%$
test_items_nested_locked	0.7330ms	0.3423ms	2.9218 KOps/s	2.9181 KOps/s	$\color{#35bf28}+0.13\%$
test_items_nested_leaf	0.1419ms	71.6045μs	13.9656 KOps/s	13.9872 KOps/s	$\color{#d91a1a}-0.15\%$
test_items_stack_nested	0.5569ms	0.3435ms	2.9114 KOps/s	2.8615 KOps/s	$\color{#35bf28}+1.74\%$
test_items_stack_nested_leaf	0.1566ms	73.8220μs	13.5461 KOps/s	13.6130 KOps/s	$\color{#d91a1a}-0.49\%$
test_items_stack_nested_locked	0.5542ms	0.3445ms	2.9027 KOps/s	2.7588 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_keys	32.5900μs	3.4914μs	286.4205 KOps/s	287.2921 KOps/s	$\color{#d91a1a}-0.30\%$
test_keys_nested	0.2594ms	0.1379ms	7.2520 KOps/s	7.3547 KOps/s	$\color{#d91a1a}-1.40\%$
test_keys_nested_locked	1.8391ms	0.1431ms	6.9885 KOps/s	7.0778 KOps/s	$\color{#d91a1a}-1.26\%$
test_keys_nested_leaf	0.2300ms	0.1182ms	8.4596 KOps/s	8.5553 KOps/s	$\color{#d91a1a}-1.12\%$
test_keys_stack_nested	0.2277ms	0.1382ms	7.2358 KOps/s	7.3932 KOps/s	$\color{#d91a1a}-2.13\%$
test_keys_stack_nested_leaf	0.2442ms	0.1182ms	8.4623 KOps/s	8.6740 KOps/s	$\color{#d91a1a}-2.44\%$
test_keys_stack_nested_locked	0.2705ms	0.1443ms	6.9316 KOps/s	7.0676 KOps/s	$\color{#d91a1a}-1.92\%$
test_values	5.6828μs	1.0276μs	973.1560 KOps/s	956.6345 KOps/s	$\color{#35bf28}+1.73\%$
test_values_nested	0.1246ms	54.7153μs	18.2764 KOps/s	17.5363 KOps/s	$\color{#35bf28}+4.22\%$
test_values_nested_locked	0.1091ms	54.5701μs	18.3251 KOps/s	17.6756 KOps/s	$\color{#35bf28}+3.67\%$
test_values_nested_leaf	0.1199ms	59.7767μs	16.7289 KOps/s	15.9658 KOps/s	$\color{#35bf28}+4.78\%$
test_values_stack_nested	0.1094ms	55.7614μs	17.9336 KOps/s	17.3180 KOps/s	$\color{#35bf28}+3.55\%$
test_values_stack_nested_leaf	0.1172ms	59.9909μs	16.6692 KOps/s	16.2058 KOps/s	$\color{#35bf28}+2.86\%$
test_values_stack_nested_locked	0.1073ms	55.9325μs	17.8787 KOps/s	17.3356 KOps/s	$\color{#35bf28}+3.13\%$
test_membership	5.2727μs	0.7247μs	1.3798 MOps/s	1.1426 MOps/s	$\textbf{\color{#35bf28}+20.76\%}$
test_membership_nested	23.1340μs	2.7181μs	367.8992 KOps/s	360.9438 KOps/s	$\color{#35bf28}+1.93\%$
test_membership_nested_leaf	39.7330μs	2.7443μs	364.3874 KOps/s	364.4274 KOps/s	$\color{#d91a1a}-0.01\%$
test_membership_stacked_nested	28.7140μs	2.7354μs	365.5795 KOps/s	363.9033 KOps/s	$\color{#35bf28}+0.46\%$
test_membership_stacked_nested_leaf	27.3530μs	2.7189μs	367.8025 KOps/s	350.3703 KOps/s	$\color{#35bf28}+4.98\%$
test_membership_nested_last	44.7360μs	3.9182μs	255.2189 KOps/s	245.3872 KOps/s	$\color{#35bf28}+4.01\%$
test_membership_nested_leaf_last	21.3790μs	3.9693μs	251.9321 KOps/s	242.8348 KOps/s	$\color{#35bf28}+3.75\%$
test_membership_stacked_nested_last	30.6070μs	3.9557μs	252.8016 KOps/s	173.4037 KOps/s	$\textbf{\color{#35bf28}+45.79\%}$
test_membership_stacked_nested_leaf_last	37.2990μs	3.9337μs	254.2151 KOps/s	172.3584 KOps/s	$\textbf{\color{#35bf28}+47.49\%}$
test_nested_getleaf	48.3100μs	10.8186μs	92.4336 KOps/s	91.3867 KOps/s	$\color{#35bf28}+1.15\%$
test_nested_get	40.7960μs	10.3190μs	96.9082 KOps/s	96.0105 KOps/s	$\color{#35bf28}+0.94\%$
test_stacked_getleaf	28.4420μs	10.7205μs	93.2792 KOps/s	92.2501 KOps/s	$\color{#35bf28}+1.12\%$
test_stacked_get	45.6050μs	10.0502μs	99.5005 KOps/s	96.4684 KOps/s	$\color{#35bf28}+3.14\%$
test_nested_getitemleaf	36.1670μs	11.2156μs	89.1618 KOps/s	88.2029 KOps/s	$\color{#35bf28}+1.09\%$
test_nested_getitem	37.8410μs	10.4592μs	95.6099 KOps/s	94.0497 KOps/s	$\color{#35bf28}+1.66\%$
test_stacked_getitemleaf	38.1510μs	11.1493μs	89.6920 KOps/s	88.3432 KOps/s	$\color{#35bf28}+1.53\%$
test_stacked_getitem	32.5110μs	10.4399μs	95.7862 KOps/s	93.6906 KOps/s	$\color{#35bf28}+2.24\%$
test_lock_nested	3.0507ms	0.4455ms	2.2445 KOps/s	1.8813 KOps/s	$\textbf{\color{#35bf28}+19.31\%}$
test_lock_stack_nested	0.5661ms	0.4083ms	2.4491 KOps/s	2.4981 KOps/s	$\color{#d91a1a}-1.96\%$
test_unlock_nested	0.6840ms	0.3604ms	2.7746 KOps/s	2.8090 KOps/s	$\color{#d91a1a}-1.23\%$
test_unlock_stack_nested	0.4368ms	0.3274ms	3.0545 KOps/s	3.0989 KOps/s	$\color{#d91a1a}-1.43\%$
test_flatten_speed	0.1902ms	91.5098μs	10.9278 KOps/s	10.9602 KOps/s	$\color{#d91a1a}-0.30\%$
test_unflatten_speed	0.9850ms	0.4760ms	2.1008 KOps/s	2.1080 KOps/s	$\color{#d91a1a}-0.34\%$
test_common_ops	1.4814ms	0.7468ms	1.3390 KOps/s	1.3874 KOps/s	$\color{#d91a1a}-3.49\%$
test_creation	24.6960μs	2.2234μs	449.7543 KOps/s	502.1704 KOps/s	$\textbf{\color{#d91a1a}-10.44\%}$
test_creation_empty	33.3030μs	10.4013μs	96.1415 KOps/s	117.4298 KOps/s	$\textbf{\color{#d91a1a}-18.13\%}$
test_creation_nested_1	39.0030μs	13.3121μs	75.1197 KOps/s	87.4421 KOps/s	$\textbf{\color{#d91a1a}-14.09\%}$
test_creation_nested_2	43.5620μs	17.7576μs	56.3139 KOps/s	64.3052 KOps/s	$\textbf{\color{#d91a1a}-12.43\%}$
test_clone	1.4160ms	12.8087μs	78.0720 KOps/s	75.0901 KOps/s	$\color{#35bf28}+3.97\%$
test_getitem[int]	1.0470ms	12.2020μs	81.9538 KOps/s	76.6822 KOps/s	$\textbf{\color{#35bf28}+6.87\%}$
test_getitem[slice_int]	0.1728ms	23.2889μs	42.9389 KOps/s	41.7165 KOps/s	$\color{#35bf28}+2.93\%$
test_getitem[range]	0.1725ms	46.9888μs	21.2817 KOps/s	20.5196 KOps/s	$\color{#35bf28}+3.71\%$
test_getitem[tuple]	0.1458ms	19.1490μs	52.2222 KOps/s	48.9783 KOps/s	$\textbf{\color{#35bf28}+6.62\%}$
test_getitem[list]	0.1850ms	41.8030μs	23.9217 KOps/s	22.7269 KOps/s	$\textbf{\color{#35bf28}+5.26\%}$
test_setitem_dim[int]	53.9310μs	24.6717μs	40.5322 KOps/s	40.2229 KOps/s	$\color{#35bf28}+0.77\%$
test_setitem_dim[slice_int]	85.6900μs	50.4789μs	19.8103 KOps/s	19.8099 KOps/s	$+0.00\%$
test_setitem_dim[range]	0.1341ms	72.1190μs	13.8660 KOps/s	13.8601 KOps/s	$\color{#35bf28}+0.04\%$
test_setitem_dim[tuple]	93.6150μs	39.5271μs	25.2991 KOps/s	25.3483 KOps/s	$\color{#d91a1a}-0.19\%$
test_setitem	75.8110μs	19.7138μs	50.7258 KOps/s	53.0860 KOps/s	$\color{#d91a1a}-4.45\%$
test_set	82.1330μs	18.7210μs	53.4159 KOps/s	55.1527 KOps/s	$\color{#d91a1a}-3.15\%$
test_set_shared	5.0482ms	0.1670ms	5.9871 KOps/s	6.0644 KOps/s	$\color{#d91a1a}-1.28\%$
test_update	0.1092ms	20.7850μs	48.1115 KOps/s	50.8792 KOps/s	$\textbf{\color{#d91a1a}-5.44\%}$
test_update_nested	84.9490μs	29.4012μs	34.0122 KOps/s	34.2626 KOps/s	$\color{#d91a1a}-0.73\%$
test_update__nested	0.6534ms	31.1270μs	32.1265 KOps/s	29.6842 KOps/s	$\textbf{\color{#35bf28}+8.23\%}$
test_set_nested	75.2310μs	20.3620μs	49.1111 KOps/s	49.0407 KOps/s	$\color{#35bf28}+0.14\%$
test_set_nested_new	73.6270μs	24.8718μs	40.2062 KOps/s	40.3065 KOps/s	$\color{#d91a1a}-0.25\%$
test_select	0.1003ms	42.1021μs	23.7518 KOps/s	24.0179 KOps/s	$\color{#d91a1a}-1.11\%$
test_select_nested	0.1276ms	59.9147μs	16.6904 KOps/s	16.5380 KOps/s	$\color{#35bf28}+0.92\%$
test_exclude_nested	0.1522ms	74.5800μs	13.4084 KOps/s	13.2387 KOps/s	$\color{#35bf28}+1.28\%$
test_empty[True]	0.6609ms	0.3518ms	2.8429 KOps/s	2.8311 KOps/s	$\color{#35bf28}+0.42\%$
test_empty[False]	17.4502μs	1.2381μs	807.7113 KOps/s	822.5416 KOps/s	$\color{#d91a1a}-1.80\%$
test_unbind_speed	0.3726ms	0.2576ms	3.8823 KOps/s	3.8531 KOps/s	$\color{#35bf28}+0.76\%$
test_unbind_speed_stack0	0.5714ms	0.2533ms	3.9486 KOps/s	3.9940 KOps/s	$\color{#d91a1a}-1.14\%$
test_unbind_speed_stack1	98.8125ms	0.7465ms	1.3397 KOps/s	1.4832 KOps/s	$\textbf{\color{#d91a1a}-9.68\%}$
test_split	2.3996ms	1.5599ms	641.0665 Ops/s	581.4415 Ops/s	$\textbf{\color{#35bf28}+10.25\%}$
test_chunk	97.4279ms	1.8591ms	537.8844 Ops/s	576.7528 Ops/s	$\textbf{\color{#d91a1a}-6.74\%}$
test_consolidate_njt[False-None]	8.5297ms	8.0856ms	123.6766 Ops/s	123.5580 Ops/s	$\color{#35bf28}+0.10\%$
test_creation[device0]	3.3240ms	89.2132μs	11.2091 KOps/s	10.6211 KOps/s	$\textbf{\color{#35bf28}+5.54\%}$
test_creation_from_tensor	0.2216ms	91.7588μs	10.8981 KOps/s	10.5641 KOps/s	$\color{#35bf28}+3.16\%$
test_add_one[memmap_tensor0]	0.2054ms	4.7266μs	211.5698 KOps/s	210.8429 KOps/s	$\color{#35bf28}+0.34\%$
test_contiguous[memmap_tensor0]	16.8020μs	0.5285μs	1.8922 MOps/s	1.9099 MOps/s	$\color{#d91a1a}-0.93\%$
test_stack[memmap_tensor0]	39.0730μs	3.2637μs	306.3962 KOps/s	305.0429 KOps/s	$\color{#35bf28}+0.44\%$
test_memmaptd_index	0.9105ms	0.2319ms	4.3125 KOps/s	4.3231 KOps/s	$\color{#d91a1a}-0.25\%$
test_memmaptd_index_astensor	0.7928ms	0.3105ms	3.2204 KOps/s	3.1806 KOps/s	$\color{#35bf28}+1.25\%$
test_memmaptd_index_op	1.3215ms	0.5637ms	1.7739 KOps/s	1.8615 KOps/s	$\color{#d91a1a}-4.71\%$
test_serialize_model	0.1261s	0.1133s	8.8245 Ops/s	7.5610 Ops/s	$\textbf{\color{#35bf28}+16.71\%}$
test_serialize_model_pickle	0.4450s	0.3838s	2.6055 Ops/s	2.4440 Ops/s	$\textbf{\color{#35bf28}+6.61\%}$
test_serialize_weights	0.2223s	0.1284s	7.7852 Ops/s	8.7199 Ops/s	$\textbf{\color{#d91a1a}-10.72\%}$
test_serialize_weights_returnearly	0.1650s	0.1577s	6.3430 Ops/s	6.3438 Ops/s	$\color{#d91a1a}-0.01\%$
test_serialize_weights_pickle	0.5532s	0.4622s	2.1637 Ops/s	2.5521 Ops/s	$\textbf{\color{#d91a1a}-15.22\%}$
test_serialize_weights_filesystem	0.1501s	0.1384s	7.2250 Ops/s	6.4586 Ops/s	$\textbf{\color{#35bf28}+11.87\%}$
test_serialize_model_filesystem	0.2532s	0.1557s	6.4209 Ops/s	6.7851 Ops/s	$\textbf{\color{#d91a1a}-5.37\%}$
test_reshape_pytree	72.3950μs	26.5996μs	37.5946 KOps/s	36.1634 KOps/s	$\color{#35bf28}+3.96\%$
test_reshape_td	79.7890μs	31.7706μs	31.4757 KOps/s	28.4247 KOps/s	$\textbf{\color{#35bf28}+10.73\%}$
test_view_pytree	84.0570μs	27.0280μs	36.9986 KOps/s	36.9471 KOps/s	$\color{#35bf28}+0.14\%$
test_view_td	93.5140μs	36.4375μs	27.4443 KOps/s	25.6623 KOps/s	$\textbf{\color{#35bf28}+6.94\%}$
test_unbind_pytree	74.6890μs	29.7999μs	33.5571 KOps/s	33.2385 KOps/s	$\color{#35bf28}+0.96\%$
test_unbind_td	0.3582ms	37.8947μs	26.3889 KOps/s	25.4541 KOps/s	$\color{#35bf28}+3.67\%$
test_split_pytree	86.2530μs	30.0286μs	33.3016 KOps/s	33.7539 KOps/s	$\color{#d91a1a}-1.34\%$
test_split_td	0.5496ms	43.3783μs	23.0530 KOps/s	22.0282 KOps/s	$\color{#35bf28}+4.65\%$
test_add_pytree	0.1164ms	34.9718μs	28.5945 KOps/s	28.1110 KOps/s	$\color{#35bf28}+1.72\%$
test_add_td	0.1360ms	51.1460μs	19.5519 KOps/s	18.8706 KOps/s	$\color{#35bf28}+3.61\%$
test_compile_add_one_nested[tensordict-compile]	0.1431ms	60.9761μs	16.3999 KOps/s	16.2846 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_add_one_nested[tensordict-eager]	0.3061ms	0.1602ms	6.2419 KOps/s	6.0759 KOps/s	$\color{#35bf28}+2.73\%$
test_compile_add_one_nested[pytree-compile]	0.1118ms	46.0344μs	21.7229 KOps/s	22.2642 KOps/s	$\color{#d91a1a}-2.43\%$
test_compile_add_one_nested[pytree-eager]	0.2503ms	0.1168ms	8.5603 KOps/s	8.3252 KOps/s	$\color{#35bf28}+2.82\%$
test_compile_copy_nested[tensordict-compile]	72.4040μs	25.9649μs	38.5136 KOps/s	38.5554 KOps/s	$\color{#d91a1a}-0.11\%$
test_compile_copy_nested[tensordict-eager]	0.1152ms	52.8982μs	18.9042 KOps/s	18.1901 KOps/s	$\color{#35bf28}+3.93\%$
test_compile_copy_nested[pytree-compile]	0.1707ms	80.8384μs	12.3704 KOps/s	12.2207 KOps/s	$\color{#35bf28}+1.22\%$
test_compile_copy_nested[pytree-eager]	0.1389ms	68.5397μs	14.5901 KOps/s	14.5089 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_add_one_flat[tensordict-compile]	0.2399ms	0.1039ms	9.6215 KOps/s	9.5147 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_add_one_flat[tensordict-eager]	0.3261ms	0.1984ms	5.0412 KOps/s	4.9961 KOps/s	$\color{#35bf28}+0.90\%$
test_compile_add_one_flat[tensorclass-compile]	0.1120ms	45.3264μs	22.0622 KOps/s	22.9908 KOps/s	$\color{#d91a1a}-4.04\%$
test_compile_add_one_flat[tensorclass-eager]	0.3560ms	61.2062μs	16.3382 KOps/s	16.1087 KOps/s	$\color{#35bf28}+1.42\%$
test_compile_add_one_flat[pytree-compile]	0.1561ms	0.1027ms	9.7417 KOps/s	9.8213 KOps/s	$\color{#d91a1a}-0.81\%$
test_compile_add_one_flat[pytree-eager]	0.3450ms	0.1977ms	5.0586 KOps/s	4.9116 KOps/s	$\color{#35bf28}+2.99\%$
test_compile_add_self_flat[tensordict-eager]	0.4523ms	0.2090ms	4.7836 KOps/s	4.6776 KOps/s	$\color{#35bf28}+2.27\%$
test_compile_add_self_flat[tensordict-compile]	0.2061ms	0.1094ms	9.1427 KOps/s	9.4914 KOps/s	$\color{#d91a1a}-3.67\%$
test_compile_add_self_flat[tensorclass-eager]	1.1265ms	54.2871μs	18.4206 KOps/s	18.3427 KOps/s	$\color{#35bf28}+0.42\%$
test_compile_add_self_flat[tensorclass-compile]	0.1342ms	47.2044μs	21.1845 KOps/s	22.1154 KOps/s	$\color{#d91a1a}-4.21\%$
test_compile_add_self_flat[pytree-eager]	0.5619ms	0.1568ms	6.3774 KOps/s	6.1957 KOps/s	$\color{#35bf28}+2.93\%$
test_compile_add_self_flat[pytree-compile]	0.1809ms	0.1031ms	9.6953 KOps/s	9.7046 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_copy_flat[tensordict-compile]	50.3340μs	21.0136μs	47.5883 KOps/s	48.6919 KOps/s	$\color{#d91a1a}-2.27\%$
test_compile_copy_flat[tensordict-eager]	0.1258ms	58.7718μs	17.0150 KOps/s	16.7179 KOps/s	$\color{#35bf28}+1.78\%$
test_compile_copy_flat[pytree-compile]	0.1781ms	80.9412μs	12.3547 KOps/s	11.7140 KOps/s	$\textbf{\color{#35bf28}+5.47\%}$
test_compile_copy_flat[pytree-eager]	0.1549ms	68.8372μs	14.5270 KOps/s	13.9109 KOps/s	$\color{#35bf28}+4.43\%$
test_compile_assign_and_add[tensordict-compile]	0.3029ms	0.2069ms	4.8322 KOps/s	4.9000 KOps/s	$\color{#d91a1a}-1.38\%$
test_compile_assign_and_add[tensordict-eager]	2.0244ms	1.2755ms	784.0297 Ops/s	778.3010 Ops/s	$\color{#35bf28}+0.74\%$
test_compile_assign_and_add[pytree-compile]	0.3636ms	0.1998ms	5.0040 KOps/s	4.9632 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_assign_and_add[pytree-eager]	0.9519ms	0.7768ms	1.2874 KOps/s	1.2674 KOps/s	$\color{#35bf28}+1.58\%$
test_compile_assign_and_add_stack[compile]	0.5952ms	0.4572ms	2.1872 KOps/s	2.2444 KOps/s	$\color{#d91a1a}-2.55\%$
test_compile_assign_and_add_stack[eager]	3.9381ms	2.5073ms	398.8396 Ops/s	416.6871 Ops/s	$\color{#d91a1a}-4.28\%$
test_compile_indexing[tensor-tensordict-compile]	0.1171ms	35.2161μs	28.3961 KOps/s	29.8570 KOps/s	$\color{#d91a1a}-4.89\%$
test_compile_indexing[tensor-tensordict-eager]	0.4474ms	31.0716μs	32.1838 KOps/s	30.2805 KOps/s	$\textbf{\color{#35bf28}+6.29\%}$
test_compile_indexing[tensor-tensorclass-compile]	72.6550μs	28.9036μs	34.5977 KOps/s	35.0973 KOps/s	$\color{#d91a1a}-1.42\%$
test_compile_indexing[tensor-tensorclass-eager]	63.9800μs	23.0559μs	43.3728 KOps/s	43.1566 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_indexing[tensor-pytree-compile]	76.7740μs	29.5986μs	33.7854 KOps/s	34.3171 KOps/s	$\color{#d91a1a}-1.55\%$
test_compile_indexing[tensor-pytree-eager]	71.9340μs	22.9704μs	43.5342 KOps/s	43.6304 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_indexing[slice-tensordict-compile]	0.1313ms	50.8184μs	19.6779 KOps/s	20.0025 KOps/s	$\color{#d91a1a}-1.62\%$
test_compile_indexing[slice-tensordict-eager]	0.4278ms	18.7060μs	53.4587 KOps/s	48.1827 KOps/s	$\textbf{\color{#35bf28}+10.95\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1043ms	44.0033μs	22.7256 KOps/s	22.9140 KOps/s	$\color{#d91a1a}-0.82\%$
test_compile_indexing[slice-tensorclass-eager]	56.5960μs	18.7048μs	53.4621 KOps/s	51.9357 KOps/s	$\color{#35bf28}+2.94\%$
test_compile_indexing[slice-pytree-compile]	0.1153ms	44.6741μs	22.3844 KOps/s	22.5715 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_indexing[slice-pytree-eager]	55.4830μs	18.8395μs	53.0800 KOps/s	52.2337 KOps/s	$\color{#35bf28}+1.62\%$
test_compile_indexing[int-tensordict-compile]	0.1152ms	51.9602μs	19.2455 KOps/s	19.4042 KOps/s	$\color{#d91a1a}-0.82\%$
test_compile_indexing[int-tensordict-eager]	0.8443ms	18.6217μs	53.7007 KOps/s	48.4827 KOps/s	$\textbf{\color{#35bf28}+10.76\%}$
test_compile_indexing[int-tensorclass-compile]	0.1034ms	44.8743μs	22.2845 KOps/s	22.7496 KOps/s	$\color{#d91a1a}-2.04\%$
test_compile_indexing[int-tensorclass-eager]	53.8900μs	18.7509μs	53.3307 KOps/s	52.7792 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_indexing[int-pytree-compile]	0.1030ms	45.0894μs	22.1782 KOps/s	22.7636 KOps/s	$\color{#d91a1a}-2.57\%$
test_compile_indexing[int-pytree-eager]	61.3340μs	18.9183μs	52.8588 KOps/s	53.0530 KOps/s	$\color{#d91a1a}-0.37\%$
test_mod_add[eager]	87.0750μs	25.3163μs	39.5003 KOps/s	42.3380 KOps/s	$\textbf{\color{#d91a1a}-6.70\%}$
test_mod_add[compile]	0.1055ms	44.6326μs	22.4051 KOps/s	23.1459 KOps/s	$\color{#d91a1a}-3.20\%$
test_mod_add[compile-overhead]	0.1067ms	45.2310μs	22.1087 KOps/s	22.6220 KOps/s	$\color{#d91a1a}-2.27\%$
test_mod_wrap[eager]	0.3573ms	0.2110ms	4.7387 KOps/s	4.7893 KOps/s	$\color{#d91a1a}-1.06\%$
test_mod_wrap[compile]	1.2516ms	0.2036ms	4.9124 KOps/s	4.9648 KOps/s	$\color{#d91a1a}-1.05\%$
test_mod_wrap[compile-overhead]	1.2395ms	0.2026ms	4.9362 KOps/s	4.9758 KOps/s	$\color{#d91a1a}-0.80\%$
test_mod_wrap_and_backward[eager]	15.2617ms	10.9357ms	91.4435 Ops/s	81.0193 Ops/s	$\textbf{\color{#35bf28}+12.87\%}$
test_mod_wrap_and_backward[compile]	12.2738ms	11.2457ms	88.9226 Ops/s	70.9064 Ops/s	$\textbf{\color{#35bf28}+25.41\%}$
test_mod_wrap_and_backward[compile-overhead]	19.5706ms	12.1533ms	82.2818 Ops/s	79.4250 Ops/s	$\color{#35bf28}+3.60\%$
test_seq_add[eager]	0.2233ms	91.0127μs	10.9875 KOps/s	11.3638 KOps/s	$\color{#d91a1a}-3.31\%$
test_seq_add[compile]	0.1179ms	60.0787μs	16.6448 KOps/s	17.0088 KOps/s	$\color{#d91a1a}-2.14\%$
test_seq_add[compile-overhead]	0.1293ms	58.1478μs	17.1975 KOps/s	17.2803 KOps/s	$\color{#d91a1a}-0.48\%$
test_seq_wrap[eager]	0.4903ms	0.3755ms	2.6632 KOps/s	2.6550 KOps/s	$\color{#35bf28}+0.31\%$
test_seq_wrap[compile]	0.3730ms	0.2235ms	4.4744 KOps/s	4.4726 KOps/s	$\color{#35bf28}+0.04\%$
test_seq_wrap[compile-overhead]	0.3403ms	0.2229ms	4.4868 KOps/s	4.5349 KOps/s	$\color{#d91a1a}-1.06\%$
test_func_call_runtime[False-eager]	0.9464ms	0.5355ms	1.8674 KOps/s	1.7853 KOps/s	$\color{#35bf28}+4.60\%$
test_func_call_runtime[False-compile]	0.4984ms	0.4150ms	2.4095 KOps/s	2.3990 KOps/s	$\color{#35bf28}+0.44\%$
test_func_call_runtime[False-compile-overhead]	0.5425ms	0.4170ms	2.3979 KOps/s	2.3887 KOps/s	$\color{#35bf28}+0.38\%$
test_func_call_runtime[True-eager]	0.8786ms	0.7416ms	1.3484 KOps/s	1.3183 KOps/s	$\color{#35bf28}+2.28\%$
test_func_call_runtime[True-compile]	0.8126ms	0.4613ms	2.1677 KOps/s	2.1744 KOps/s	$\color{#d91a1a}-0.31\%$
test_func_call_runtime[True-compile-overhead]	0.5800ms	0.4630ms	2.1598 KOps/s	2.1677 KOps/s	$\color{#d91a1a}-0.36\%$
test_func_call_cm_runtime[False-eager]	0.8765ms	0.5345ms	1.8709 KOps/s	1.8525 KOps/s	$\color{#35bf28}+0.99\%$
test_func_call_cm_runtime[False-compile]	0.6834ms	0.4159ms	2.4044 KOps/s	2.3559 KOps/s	$\color{#35bf28}+2.06\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5499ms	0.4188ms	2.3876 KOps/s	2.3643 KOps/s	$\color{#35bf28}+0.99\%$
test_func_call_cm_runtime[True-eager]	1.9608ms	0.8996ms	1.1117 KOps/s	1.1221 KOps/s	$\color{#d91a1a}-0.93\%$
test_func_call_cm_runtime[True-compile]	0.6240ms	0.4840ms	2.0659 KOps/s	2.0600 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[True-compile-overhead]	0.8676ms	0.4867ms	2.0548 KOps/s	2.0703 KOps/s	$\color{#d91a1a}-0.75\%$
test_vmap_func_call_cm_runtime[eager]	3.3883ms	1.8580ms	538.2103 Ops/s	536.2717 Ops/s	$\color{#35bf28}+0.36\%$
test_vmap_func_call_cm_runtime[compile]	0.8815ms	0.5098ms	1.9615 KOps/s	1.9600 KOps/s	$\color{#35bf28}+0.08\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.6193ms	0.5088ms	1.9655 KOps/s	1.9635 KOps/s	$\color{#35bf28}+0.10\%$
test_distributed	0.2950ms	0.1254ms	7.9754 KOps/s	7.7688 KOps/s	$\color{#35bf28}+2.66\%$
test_tdmodule	50.9350μs	18.4057μs	54.3310 KOps/s	56.9155 KOps/s	$\color{#d91a1a}-4.54\%$
test_tdmodule_dispatch	65.0310μs	36.1947μs	27.6283 KOps/s	29.0057 KOps/s	$\color{#d91a1a}-4.75\%$
test_tdseq	46.5970μs	20.5039μs	48.7712 KOps/s	51.3487 KOps/s	$\textbf{\color{#d91a1a}-5.02\%}$
test_tdseq_dispatch	61.7850μs	40.3769μs	24.7667 KOps/s	25.9198 KOps/s	$\color{#d91a1a}-4.45\%$
test_instantiation_functorch	2.1923ms	1.5491ms	645.5543 Ops/s	640.3584 Ops/s	$\color{#35bf28}+0.81\%$
test_exec_functorch	0.3050ms	0.1793ms	5.5777 KOps/s	5.6244 KOps/s	$\color{#d91a1a}-0.83\%$
test_exec_functional_call	0.2604ms	0.1707ms	5.8582 KOps/s	5.8937 KOps/s	$\color{#d91a1a}-0.60\%$
test_exec_td_decorator	0.5191ms	0.2260ms	4.4250 KOps/s	4.2813 KOps/s	$\color{#35bf28}+3.36\%$
test_vmap_mlp_speed_decorator[True-True]	1.0019ms	0.6348ms	1.5754 KOps/s	1.6126 KOps/s	$\color{#d91a1a}-2.31\%$
test_vmap_mlp_speed_decorator[True-False]	0.9653ms	0.6329ms	1.5799 KOps/s	1.6163 KOps/s	$\color{#d91a1a}-2.25\%$
test_vmap_mlp_speed_decorator[False-True]	0.8380ms	0.5200ms	1.9231 KOps/s	1.9318 KOps/s	$\color{#d91a1a}-0.45\%$
test_vmap_mlp_speed_decorator[False-False]	0.6595ms	0.5172ms	1.9335 KOps/s	1.9312 KOps/s	$\color{#35bf28}+0.12\%$
test_to_module_speed[True]	1.5102ms	1.3014ms	768.4017 Ops/s	755.4232 Ops/s	$\color{#35bf28}+1.72\%$
test_to_module_speed[False]	1.6533ms	1.2791ms	781.8266 Ops/s	792.1184 Ops/s	$\color{#d91a1a}-1.30\%$
test_tc_init	72.7460μs	44.2578μs	22.5949 KOps/s	21.9265 KOps/s	$\color{#35bf28}+3.05\%$
test_tc_init_nested	0.1821ms	91.8115μs	10.8919 KOps/s	10.9410 KOps/s	$\color{#d91a1a}-0.45\%$
test_tc_first_layer_tensor	18.3850μs	1.5711μs	636.4942 KOps/s	631.0430 KOps/s	$\color{#35bf28}+0.86\%$
test_tc_first_layer_nontensor	24.3750μs	4.6636μs	214.4251 KOps/s	204.7612 KOps/s	$\color{#35bf28}+4.72\%$
test_tc_second_layer_tensor	30.0360μs	2.8133μs	355.4499 KOps/s	339.1885 KOps/s	$\color{#35bf28}+4.79\%$
test_tc_second_layer_nontensor	27.6720μs	6.0259μs	165.9505 KOps/s	157.6228 KOps/s	$\textbf{\color{#35bf28}+5.28\%}$
test_unbind	0.2251s	12.3892ms	80.7156 Ops/s	77.5525 Ops/s	$\color{#35bf28}+4.08\%$
test_full_like	16.6616ms	12.1302ms	82.4388 Ops/s	138.9978 Ops/s	$\textbf{\color{#d91a1a}-40.69\%}$
test_zeros_like	12.0422ms	7.0696ms	141.4505 Ops/s	362.8178 Ops/s	$\textbf{\color{#d91a1a}-61.01\%}$
test_ones_like	13.6793ms	7.3462ms	136.1245 Ops/s	305.5424 Ops/s	$\textbf{\color{#d91a1a}-55.45\%}$
test_clone	12.0543ms	8.9079ms	112.2593 Ops/s	197.8508 Ops/s	$\textbf{\color{#d91a1a}-43.26\%}$
test_squeeze	65.6720μs	11.9402μs	83.7503 KOps/s	83.8878 KOps/s	$\color{#d91a1a}-0.16\%$
test_unsqueeze	0.1637ms	87.7094μs	11.4013 KOps/s	11.3718 KOps/s	$\color{#35bf28}+0.26\%$
test_split	0.4832ms	0.1862ms	5.3703 KOps/s	5.3333 KOps/s	$\color{#35bf28}+0.69\%$
test_permute	0.3596ms	0.2108ms	4.7427 KOps/s	4.4723 KOps/s	$\textbf{\color{#35bf28}+6.05\%}$
test_stack	27.1289ms	24.8636ms	40.2194 Ops/s	40.1610 Ops/s	$\color{#35bf28}+0.15\%$
test_cat	27.2721ms	24.7036ms	40.4799 Ops/s	40.6633 Ops/s	$\color{#d91a1a}-0.45\%$

[github-actions]

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

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

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	33.7500μs	10.7623μs	92.9172 KOps/s	94.3459 KOps/s	$\color{#d91a1a}-1.51\%$
test_plain_set_stack_nested	36.2410μs	10.8752μs	91.9524 KOps/s	94.1488 KOps/s	$\color{#d91a1a}-2.33\%$
test_plain_set_nested_inplace	41.6300μs	11.6720μs	85.6750 KOps/s	88.1718 KOps/s	$\color{#d91a1a}-2.83\%$
test_plain_set_stack_nested_inplace	36.7010μs	11.6223μs	86.0413 KOps/s	87.2667 KOps/s	$\color{#d91a1a}-1.40\%$
test_items	34.2500μs	2.8892μs	346.1217 KOps/s	343.4461 KOps/s	$\color{#35bf28}+0.78\%$
test_items_nested	0.3850ms	0.3227ms	3.0989 KOps/s	3.1243 KOps/s	$\color{#d91a1a}-0.81\%$
test_items_nested_locked	0.3932ms	0.3266ms	3.0618 KOps/s	3.1268 KOps/s	$\color{#d91a1a}-2.08\%$
test_items_nested_leaf	90.8420μs	59.3650μs	16.8449 KOps/s	17.0574 KOps/s	$\color{#d91a1a}-1.25\%$
test_items_stack_nested	0.5001ms	0.3253ms	3.0737 KOps/s	3.1446 KOps/s	$\color{#d91a1a}-2.25\%$
test_items_stack_nested_leaf	78.5520μs	58.8136μs	17.0029 KOps/s	16.8485 KOps/s	$\color{#35bf28}+0.92\%$
test_items_stack_nested_locked	0.3829ms	0.3234ms	3.0924 KOps/s	3.1065 KOps/s	$\color{#d91a1a}-0.45\%$
test_keys	38.6710μs	3.5111μs	284.8080 KOps/s	285.9702 KOps/s	$\color{#d91a1a}-0.41\%$
test_keys_nested	0.1023ms	71.1768μs	14.0495 KOps/s	14.1962 KOps/s	$\color{#d91a1a}-1.03\%$
test_keys_nested_locked	0.6883ms	76.4162μs	13.0862 KOps/s	13.1516 KOps/s	$\color{#d91a1a}-0.50\%$
test_keys_nested_leaf	0.1080ms	62.4521μs	16.0123 KOps/s	16.1736 KOps/s	$\color{#d91a1a}-1.00\%$
test_keys_stack_nested	0.1000ms	71.1642μs	14.0520 KOps/s	14.1424 KOps/s	$\color{#d91a1a}-0.64\%$
test_keys_stack_nested_leaf	96.0720μs	61.7908μs	16.1836 KOps/s	16.1973 KOps/s	$\color{#d91a1a}-0.08\%$
test_keys_stack_nested_locked	0.1350ms	76.0533μs	13.1487 KOps/s	13.2588 KOps/s	$\color{#d91a1a}-0.83\%$
test_values	4.8818μs	0.8570μs	1.1669 MOps/s	1.1697 MOps/s	$\color{#d91a1a}-0.24\%$
test_values_nested	58.8810μs	31.3523μs	31.8956 KOps/s	31.9798 KOps/s	$\color{#d91a1a}-0.26\%$
test_values_nested_locked	66.1110μs	33.0194μs	30.2852 KOps/s	30.5057 KOps/s	$\color{#d91a1a}-0.72\%$
test_values_nested_leaf	70.7110μs	34.0637μs	29.3568 KOps/s	29.6686 KOps/s	$\color{#d91a1a}-1.05\%$
test_values_stack_nested	76.4520μs	31.6860μs	31.5596 KOps/s	31.4709 KOps/s	$\color{#35bf28}+0.28\%$
test_values_stack_nested_leaf	81.4220μs	34.2122μs	29.2293 KOps/s	29.1831 KOps/s	$\color{#35bf28}+0.16\%$
test_values_stack_nested_locked	71.1510μs	33.0992μs	30.2123 KOps/s	30.1312 KOps/s	$\color{#35bf28}+0.27\%$
test_membership	2.4145μs	0.5295μs	1.8885 MOps/s	1.9652 MOps/s	$\color{#d91a1a}-3.90\%$
test_membership_nested	24.4555μs	1.8643μs	536.3816 KOps/s	506.8249 KOps/s	$\textbf{\color{#35bf28}+5.83\%}$
test_membership_nested_leaf	11.7937μs	1.8556μs	538.9002 KOps/s	522.9324 KOps/s	$\color{#35bf28}+3.05\%$
test_membership_stacked_nested	35.6010μs	1.9440μs	514.4162 KOps/s	493.9891 KOps/s	$\color{#35bf28}+4.14\%$
test_membership_stacked_nested_leaf	25.6700μs	1.9552μs	511.4659 KOps/s	493.9102 KOps/s	$\color{#35bf28}+3.55\%$
test_membership_nested_last	45.3910μs	2.7871μs	358.7929 KOps/s	350.4614 KOps/s	$\color{#35bf28}+2.38\%$
test_membership_nested_leaf_last	26.7300μs	2.7962μs	357.6262 KOps/s	353.3583 KOps/s	$\color{#35bf28}+1.21\%$
test_membership_stacked_nested_last	29.9710μs	2.8018μs	356.9130 KOps/s	126.4670 KOps/s	$\textbf{\color{#35bf28}+182.22\%}$
test_membership_stacked_nested_leaf_last	41.9210μs	2.8017μs	356.9322 KOps/s	126.8591 KOps/s	$\textbf{\color{#35bf28}+181.36\%}$
test_nested_getleaf	49.9510μs	6.0254μs	165.9641 KOps/s	164.8005 KOps/s	$\color{#35bf28}+0.71\%$
test_nested_get	27.5400μs	5.7537μs	173.8011 KOps/s	175.6067 KOps/s	$\color{#d91a1a}-1.03\%$
test_stacked_getleaf	42.0500μs	6.0572μs	165.0926 KOps/s	166.3710 KOps/s	$\color{#d91a1a}-0.77\%$
test_stacked_get	36.2800μs	5.7496μs	173.9236 KOps/s	174.8010 KOps/s	$\color{#d91a1a}-0.50\%$
test_nested_getitemleaf	40.4310μs	6.1289μs	163.1626 KOps/s	163.2880 KOps/s	$\color{#d91a1a}-0.08\%$
test_nested_getitem	29.1100μs	5.8552μs	170.7894 KOps/s	171.9674 KOps/s	$\color{#d91a1a}-0.69\%$
test_stacked_getitemleaf	36.0710μs	6.0750μs	164.6087 KOps/s	164.2231 KOps/s	$\color{#35bf28}+0.23\%$
test_stacked_getitem	34.1510μs	5.8472μs	171.0216 KOps/s	173.1542 KOps/s	$\color{#d91a1a}-1.23\%$
test_lock_nested	9.6395ms	0.3803ms	2.6298 KOps/s	2.6612 KOps/s	$\color{#d91a1a}-1.18\%$
test_lock_stack_nested	0.4431ms	0.3404ms	2.9377 KOps/s	2.9813 KOps/s	$\color{#d91a1a}-1.46\%$
test_unlock_nested	0.6603ms	0.3130ms	3.1953 KOps/s	3.2356 KOps/s	$\color{#d91a1a}-1.25\%$
test_unlock_stack_nested	0.3399ms	0.2818ms	3.5484 KOps/s	3.6744 KOps/s	$\color{#d91a1a}-3.43\%$
test_flatten_speed	0.1068ms	73.5638μs	13.5937 KOps/s	13.7696 KOps/s	$\color{#d91a1a}-1.28\%$
test_unflatten_speed	0.3500ms	0.2954ms	3.3852 KOps/s	3.3745 KOps/s	$\color{#35bf28}+0.32\%$
test_common_ops	1.7912ms	0.6009ms	1.6641 KOps/s	1.6789 KOps/s	$\color{#d91a1a}-0.88\%$
test_creation	0.1025ms	1.4889μs	671.6513 KOps/s	672.4776 KOps/s	$\color{#d91a1a}-0.12\%$
test_creation_empty	30.1100μs	7.6515μs	130.6941 KOps/s	136.7355 KOps/s	$\color{#d91a1a}-4.42\%$
test_creation_nested_1	31.7800μs	9.1503μs	109.2860 KOps/s	113.6074 KOps/s	$\color{#d91a1a}-3.80\%$
test_creation_nested_2	39.2510μs	11.6677μs	85.7064 KOps/s	88.3408 KOps/s	$\color{#d91a1a}-2.98\%$
test_clone	59.1810μs	11.2016μs	89.2731 KOps/s	89.8928 KOps/s	$\color{#d91a1a}-0.69\%$
test_getitem[int]	93.5310ms	16.4671μs	60.7271 KOps/s	91.1164 KOps/s	$\textbf{\color{#d91a1a}-33.35\%}$
test_getitem[slice_int]	0.1043ms	21.4566μs	46.6057 KOps/s	46.8429 KOps/s	$\color{#d91a1a}-0.51\%$
test_getitem[range]	0.1377ms	38.7654μs	25.7962 KOps/s	25.7847 KOps/s	$\color{#35bf28}+0.04\%$
test_getitem[tuple]	0.1221ms	18.6106μs	53.7329 KOps/s	53.4467 KOps/s	$\color{#35bf28}+0.54\%$
test_getitem[list]	0.2125ms	34.5301μs	28.9602 KOps/s	28.9516 KOps/s	$\color{#35bf28}+0.03\%$
test_setitem_dim[int]	42.8810μs	19.9163μs	50.2101 KOps/s	49.8760 KOps/s	$\color{#35bf28}+0.67\%$
test_setitem_dim[slice_int]	61.0610μs	38.5563μs	25.9361 KOps/s	25.7876 KOps/s	$\color{#35bf28}+0.58\%$
test_setitem_dim[range]	86.6220μs	54.9463μs	18.1996 KOps/s	18.6757 KOps/s	$\color{#d91a1a}-2.55\%$
test_setitem_dim[tuple]	52.3710μs	32.4482μs	30.8183 KOps/s	30.8353 KOps/s	$\color{#d91a1a}-0.05\%$
test_setitem	89.8820μs	15.5334μs	64.3774 KOps/s	63.9287 KOps/s	$\color{#35bf28}+0.70\%$
test_set	93.2420μs	14.8652μs	67.2710 KOps/s	67.4911 KOps/s	$\color{#d91a1a}-0.33\%$
test_set_shared	1.6582ms	0.1482ms	6.7494 KOps/s	6.7278 KOps/s	$\color{#35bf28}+0.32\%$
test_update	53.0210μs	17.8934μs	55.8864 KOps/s	56.6615 KOps/s	$\color{#d91a1a}-1.37\%$
test_update_nested	0.2090ms	22.9991μs	43.4800 KOps/s	45.4355 KOps/s	$\color{#d91a1a}-4.30\%$
test_update__nested	0.1298ms	24.8517μs	40.2387 KOps/s	39.6895 KOps/s	$\color{#35bf28}+1.38\%$
test_set_nested	75.9120μs	15.9931μs	62.5270 KOps/s	61.1379 KOps/s	$\color{#35bf28}+2.27\%$
test_set_nested_new	84.9220μs	18.3412μs	54.5220 KOps/s	54.0751 KOps/s	$\color{#35bf28}+0.83\%$
test_select	85.6320μs	30.6303μs	32.6474 KOps/s	31.2942 KOps/s	$\color{#35bf28}+4.32\%$
test_select_nested	70.6120μs	42.0171μs	23.7998 KOps/s	23.7503 KOps/s	$\color{#35bf28}+0.21\%$
test_exclude_nested	0.1076ms	59.5642μs	16.7886 KOps/s	16.8401 KOps/s	$\color{#d91a1a}-0.31\%$
test_empty[True]	0.3288ms	0.2578ms	3.8785 KOps/s	3.9434 KOps/s	$\color{#d91a1a}-1.64\%$
test_empty[False]	3.9251μs	0.7580μs	1.3192 MOps/s	1.3421 MOps/s	$\color{#d91a1a}-1.71\%$
test_to	87.4620μs	56.4897μs	17.7023 KOps/s	17.8787 KOps/s	$\color{#d91a1a}-0.99\%$
test_to_nonblocking	96.0820μs	46.7373μs	21.3962 KOps/s	20.9780 KOps/s	$\color{#35bf28}+1.99\%$
test_unbind_speed	1.5702ms	0.2390ms	4.1849 KOps/s	4.2262 KOps/s	$\color{#d91a1a}-0.98\%$
test_unbind_speed_stack0	0.3103ms	0.2391ms	4.1819 KOps/s	4.3364 KOps/s	$\color{#d91a1a}-3.56\%$
test_unbind_speed_stack1	93.8259ms	0.6557ms	1.5250 KOps/s	1.6902 KOps/s	$\textbf{\color{#d91a1a}-9.78\%}$
test_split	96.7470ms	1.7718ms	564.3994 Ops/s	605.8706 Ops/s	$\textbf{\color{#d91a1a}-6.84\%}$
test_chunk	97.9202ms	1.6416ms	609.1538 Ops/s	609.2520 Ops/s	$\color{#d91a1a}-0.02\%$
test_consolidate[False-None]	2.7656ms	2.6445ms	378.1372 Ops/s	347.5363 Ops/s	$\textbf{\color{#35bf28}+8.81\%}$
test_consolidate[default-None]	1.8503ms	1.7537ms	570.2139 Ops/s	571.6644 Ops/s	$\color{#d91a1a}-0.25\%$
test_consolidate[reduce-overhead-None]	1.9233ms	1.7998ms	555.6267 Ops/s	554.5658 Ops/s	$\color{#35bf28}+0.19\%$
test_consolidate_njt[False-None]	6.9521ms	6.7030ms	149.1868 Ops/s	152.0320 Ops/s	$\color{#d91a1a}-1.87\%$
test_to[False-False-None]	1.8172ms	1.7098ms	584.8551 Ops/s	575.9916 Ops/s	$\color{#35bf28}+1.54\%$
test_to[True-False-None]	1.5669ms	1.3300ms	751.9054 Ops/s	743.7323 Ops/s	$\color{#35bf28}+1.10\%$
test_to[within-False-None]	4.2369ms	4.1041ms	243.6601 Ops/s	244.8456 Ops/s	$\color{#d91a1a}-0.48\%$
test_to[True-default-None]	5.7436ms	5.3412ms	187.2244 Ops/s	186.6455 Ops/s	$\color{#35bf28}+0.31\%$
test_to_njt[False-False-None]	7.2876ms	7.0243ms	142.3636 Ops/s	141.9494 Ops/s	$\color{#35bf28}+0.29\%$
test_to_njt[True-False-None]	5.8436ms	5.6123ms	178.1794 Ops/s	178.1171 Ops/s	$\color{#35bf28}+0.04\%$
test_to_njt[within-False-None]	12.7395ms	12.3442ms	81.0095 Ops/s	82.7041 Ops/s	$\color{#d91a1a}-2.05\%$
test_creation[device0]	0.4694ms	80.2694μs	12.4581 KOps/s	12.2100 KOps/s	$\color{#35bf28}+2.03\%$
test_creation_from_tensor	0.5360ms	84.9980μs	11.7650 KOps/s	11.3722 KOps/s	$\color{#35bf28}+3.45\%$
test_add_one[memmap_tensor0]	0.3331ms	7.2827μs	137.3110 KOps/s	134.5047 KOps/s	$\color{#35bf28}+2.09\%$
test_contiguous[memmap_tensor0]	2.0200μs	0.4153μs	2.4081 MOps/s	2.3796 MOps/s	$\color{#35bf28}+1.20\%$
test_stack[memmap_tensor0]	42.1610μs	4.8438μs	206.4492 KOps/s	210.5523 KOps/s	$\color{#d91a1a}-1.95\%$
test_memmaptd_index	1.9716ms	0.2590ms	3.8608 KOps/s	3.8935 KOps/s	$\color{#d91a1a}-0.84\%$
test_memmaptd_index_astensor	0.6046ms	0.3176ms	3.1484 KOps/s	3.1779 KOps/s	$\color{#d91a1a}-0.93\%$
test_memmaptd_index_op	1.0307ms	0.6048ms	1.6534 KOps/s	1.6371 KOps/s	$\color{#35bf28}+1.00\%$
test_serialize_model	0.1326s	0.1317s	7.5909 Ops/s	7.5175 Ops/s	$\color{#35bf28}+0.98\%$
test_serialize_model_pickle	1.3465s	1.2186s	0.8206 Ops/s	0.8413 Ops/s	$\color{#d91a1a}-2.46\%$
test_serialize_weights	0.1331s	0.1314s	7.6111 Ops/s	7.6193 Ops/s	$\color{#d91a1a}-0.11\%$
test_serialize_weights_returnearly	0.3269s	55.0275ms	18.1727 Ops/s	12.5207 Ops/s	$\textbf{\color{#35bf28}+45.14\%}$
test_serialize_weights_pickle	1.3737s	1.1994s	0.8337 Ops/s	0.8198 Ops/s	$\color{#35bf28}+1.70\%$
test_reshape_pytree	66.1010μs	22.7706μs	43.9163 KOps/s	43.6830 KOps/s	$\color{#35bf28}+0.53\%$
test_reshape_td	55.1910μs	26.9526μs	37.1022 KOps/s	37.3473 KOps/s	$\color{#d91a1a}-0.66\%$
test_view_pytree	52.3110μs	22.4515μs	44.5405 KOps/s	44.5742 KOps/s	$\color{#d91a1a}-0.08\%$
test_view_td	70.1110μs	30.9671μs	32.2923 KOps/s	31.7645 KOps/s	$\color{#35bf28}+1.66\%$
test_unbind_pytree	53.9310μs	28.5524μs	35.0234 KOps/s	35.0781 KOps/s	$\color{#d91a1a}-0.16\%$
test_unbind_td	0.5969ms	36.7331μs	27.2234 KOps/s	27.7694 KOps/s	$\color{#d91a1a}-1.97\%$
test_split_pytree	75.4020μs	30.8515μs	32.4133 KOps/s	32.2672 KOps/s	$\color{#35bf28}+0.45\%$
test_split_td	0.5558ms	39.5271μs	25.2991 KOps/s	24.8802 KOps/s	$\color{#35bf28}+1.68\%$
test_add_pytree	72.0210μs	36.0855μs	27.7120 KOps/s	27.2830 KOps/s	$\color{#35bf28}+1.57\%$
test_add_td	84.1620μs	48.3141μs	20.6979 KOps/s	20.9309 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_add_one_nested[tensordict-compile]	0.2508ms	0.1221ms	8.1911 KOps/s	8.0022 KOps/s	$\color{#35bf28}+2.36\%$
test_compile_add_one_nested[tensordict-eager]	0.2145ms	0.1257ms	7.9536 KOps/s	7.6805 KOps/s	$\color{#35bf28}+3.56\%$
test_compile_add_one_nested[pytree-compile]	0.1577ms	0.1003ms	9.9728 KOps/s	10.0860 KOps/s	$\color{#d91a1a}-1.12\%$
test_compile_add_one_nested[pytree-eager]	1.4397ms	0.1541ms	6.4883 KOps/s	6.3393 KOps/s	$\color{#35bf28}+2.35\%$
test_compile_copy_nested[tensordict-compile]	89.8220μs	25.7582μs	38.8226 KOps/s	43.4929 KOps/s	$\textbf{\color{#d91a1a}-10.74\%}$
test_compile_copy_nested[tensordict-eager]	58.7210μs	26.9453μs	37.1122 KOps/s	36.3689 KOps/s	$\color{#35bf28}+2.04\%$
test_compile_copy_nested[pytree-compile]	99.9220μs	64.6933μs	15.4576 KOps/s	15.2020 KOps/s	$\color{#35bf28}+1.68\%$
test_compile_copy_nested[pytree-eager]	76.7810μs	49.7177μs	20.1136 KOps/s	19.8415 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_add_one_flat[tensordict-compile]	0.1891ms	0.1458ms	6.8606 KOps/s	6.7870 KOps/s	$\color{#35bf28}+1.08\%$
test_compile_add_one_flat[tensordict-eager]	0.3209ms	0.2090ms	4.7846 KOps/s	4.7279 KOps/s	$\color{#35bf28}+1.20\%$
test_compile_add_one_flat[tensorclass-compile]	0.1543ms	0.1013ms	9.8677 KOps/s	9.8911 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_add_one_flat[tensorclass-eager]	0.1096ms	50.8540μs	19.6641 KOps/s	19.1639 KOps/s	$\color{#35bf28}+2.61\%$
test_compile_add_one_flat[pytree-compile]	0.1824ms	0.1400ms	7.1444 KOps/s	7.1959 KOps/s	$\color{#d91a1a}-0.72\%$
test_compile_add_one_flat[pytree-eager]	0.7114ms	0.5025ms	1.9900 KOps/s	1.9678 KOps/s	$\color{#35bf28}+1.13\%$
test_compile_add_self_flat[tensordict-eager]	0.4087ms	0.2474ms	4.0414 KOps/s	3.9382 KOps/s	$\color{#35bf28}+2.62\%$
test_compile_add_self_flat[tensordict-compile]	0.2045ms	0.1476ms	6.7736 KOps/s	6.9139 KOps/s	$\color{#d91a1a}-2.03\%$
test_compile_add_self_flat[tensorclass-eager]	0.1459ms	62.3067μs	16.0496 KOps/s	16.1479 KOps/s	$\color{#d91a1a}-0.61\%$
test_compile_add_self_flat[tensorclass-compile]	0.1594ms	0.1011ms	9.8934 KOps/s	9.8562 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_add_self_flat[pytree-eager]	0.5525ms	0.4248ms	2.3540 KOps/s	2.3319 KOps/s	$\color{#35bf28}+0.95\%$
test_compile_add_self_flat[pytree-compile]	0.1912ms	0.1392ms	7.1848 KOps/s	7.2919 KOps/s	$\color{#d91a1a}-1.47\%$
test_compile_copy_flat[tensordict-compile]	53.8110μs	20.1780μs	49.5588 KOps/s	54.1490 KOps/s	$\textbf{\color{#d91a1a}-8.48\%}$
test_compile_copy_flat[tensordict-eager]	55.9910μs	27.0975μs	36.9037 KOps/s	37.2673 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_copy_flat[pytree-compile]	0.2515ms	71.2109μs	14.0428 KOps/s	14.1742 KOps/s	$\color{#d91a1a}-0.93\%$
test_compile_copy_flat[pytree-eager]	85.2410μs	51.6491μs	19.3614 KOps/s	19.2454 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_assign_and_add[tensordict-compile]	1.6599ms	0.4056ms	2.4655 KOps/s	2.1732 KOps/s	$\textbf{\color{#35bf28}+13.45\%}$
test_compile_assign_and_add[tensordict-eager]	2.7789ms	2.6282ms	380.4955 Ops/s	373.7844 Ops/s	$\color{#35bf28}+1.80\%$
test_compile_assign_and_add[pytree-compile]	1.6445ms	0.4461ms	2.2416 KOps/s	2.1622 KOps/s	$\color{#35bf28}+3.67\%$
test_compile_assign_and_add[pytree-eager]	2.8831ms	2.7447ms	364.3346 Ops/s	358.9467 Ops/s	$\color{#35bf28}+1.50\%$
test_compile_indexing[tensor-tensordict-compile]	0.2509ms	0.1191ms	8.3978 KOps/s	8.3493 KOps/s	$\color{#35bf28}+0.58\%$
test_compile_indexing[tensor-tensordict-eager]	0.5916ms	81.5861μs	12.2570 KOps/s	11.3601 KOps/s	$\textbf{\color{#35bf28}+7.89\%}$
test_compile_indexing[tensor-tensorclass-compile]	0.1836ms	0.1103ms	9.0668 KOps/s	8.7364 KOps/s	$\color{#35bf28}+3.78\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1092ms	69.8144μs	14.3237 KOps/s	13.8874 KOps/s	$\color{#35bf28}+3.14\%$
test_compile_indexing[tensor-pytree-compile]	0.1825ms	0.1094ms	9.1393 KOps/s	8.7772 KOps/s	$\color{#35bf28}+4.13\%$
test_compile_indexing[tensor-pytree-eager]	0.1292ms	70.4171μs	14.2011 KOps/s	14.0270 KOps/s	$\color{#35bf28}+1.24\%$
test_compile_indexing[slice-tensordict-compile]	0.1535ms	0.1031ms	9.6972 KOps/s	9.7970 KOps/s	$\color{#d91a1a}-1.02\%$
test_compile_indexing[slice-tensordict-eager]	0.1655ms	17.0428μs	58.6759 KOps/s	53.5134 KOps/s	$\textbf{\color{#35bf28}+9.65\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1345ms	98.9034μs	10.1109 KOps/s	10.1978 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_indexing[slice-tensorclass-eager]	47.7610μs	16.5459μs	60.4381 KOps/s	60.7227 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_indexing[slice-pytree-compile]	0.1485ms	99.1287μs	10.0879 KOps/s	10.0549 KOps/s	$\color{#35bf28}+0.33\%$
test_compile_indexing[slice-pytree-eager]	44.4710μs	16.2565μs	61.5138 KOps/s	61.0925 KOps/s	$\color{#35bf28}+0.69\%$
test_compile_indexing[int-tensordict-compile]	0.1675ms	0.1029ms	9.7151 KOps/s	9.6516 KOps/s	$\color{#35bf28}+0.66\%$
test_compile_indexing[int-tensordict-eager]	0.5590ms	17.3865μs	57.5158 KOps/s	55.5459 KOps/s	$\color{#35bf28}+3.55\%$
test_compile_indexing[int-tensorclass-compile]	0.1400ms	99.5370μs	10.0465 KOps/s	10.0543 KOps/s	$\color{#d91a1a}-0.08\%$
test_compile_indexing[int-tensorclass-eager]	53.7210μs	16.3263μs	61.2510 KOps/s	60.6490 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_indexing[int-pytree-compile]	0.1463ms	99.3950μs	10.0609 KOps/s	10.1683 KOps/s	$\color{#d91a1a}-1.06\%$
test_compile_indexing[int-pytree-eager]	0.1629ms	16.2237μs	61.6382 KOps/s	61.3353 KOps/s	$\color{#35bf28}+0.49\%$
test_mod_add[eager]	73.4610μs	32.3178μs	30.9427 KOps/s	29.1566 KOps/s	$\textbf{\color{#35bf28}+6.13\%}$
test_mod_add[compile]	0.1419ms	78.2576μs	12.7783 KOps/s	12.4485 KOps/s	$\color{#35bf28}+2.65\%$
test_mod_add[compile-overhead]	0.3213ms	0.1680ms	5.9524 KOps/s	5.3197 KOps/s	$\textbf{\color{#35bf28}+11.89\%}$
test_mod_wrap[eager]	0.3275ms	0.2498ms	4.0024 KOps/s	3.8193 KOps/s	$\color{#35bf28}+4.79\%$
test_mod_wrap[compile]	1.5766ms	0.2884ms	3.4672 KOps/s	3.4021 KOps/s	$\color{#35bf28}+1.92\%$
test_mod_wrap[compile-overhead]	6.9695ms	3.6735ms	272.2188 Ops/s	265.3132 Ops/s	$\color{#35bf28}+2.60\%$
test_mod_wrap_and_backward[eager]	1.6886ms	1.4050ms	711.7633 Ops/s	673.9979 Ops/s	$\textbf{\color{#35bf28}+5.60\%}$
test_mod_wrap_and_backward[compile]	1.5902ms	1.2889ms	775.8538 Ops/s	709.8176 Ops/s	$\textbf{\color{#35bf28}+9.30\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3768ms	0.9335ms	1.0712 KOps/s	952.2861 Ops/s	$\textbf{\color{#35bf28}+12.49\%}$
test_seq_add[eager]	0.1666ms	0.1040ms	9.6120 KOps/s	9.9522 KOps/s	$\color{#d91a1a}-3.42\%$
test_seq_add[compile]	0.5616ms	92.5478μs	10.8052 KOps/s	11.1444 KOps/s	$\color{#d91a1a}-3.04\%$
test_seq_add[compile-overhead]	0.1955ms	0.1340ms	7.4641 KOps/s	7.6642 KOps/s	$\color{#d91a1a}-2.61\%$
test_seq_wrap[eager]	0.8132ms	0.4043ms	2.4736 KOps/s	2.5700 KOps/s	$\color{#d91a1a}-3.75\%$
test_seq_wrap[compile]	0.4015ms	0.3051ms	3.2777 KOps/s	3.2272 KOps/s	$\color{#35bf28}+1.56\%$
test_seq_wrap[compile-overhead]	0.6640ms	0.2271ms	4.4027 KOps/s	4.3856 KOps/s	$\color{#35bf28}+0.39\%$
test_func_call_runtime[False-eager]	1.1891ms	0.7628ms	1.3110 KOps/s	1.2813 KOps/s	$\color{#35bf28}+2.31\%$
test_func_call_runtime[False-compile]	1.2136ms	0.7635ms	1.3098 KOps/s	1.3064 KOps/s	$\color{#35bf28}+0.26\%$
test_func_call_runtime[False-compile-overhead]	0.4646ms	0.3726ms	2.6836 KOps/s	2.6796 KOps/s	$\color{#35bf28}+0.15\%$
test_func_call_runtime[True-eager]	1.3812ms	0.9279ms	1.0777 KOps/s	1.0587 KOps/s	$\color{#35bf28}+1.79\%$
test_func_call_runtime[True-compile]	1.2411ms	0.7846ms	1.2746 KOps/s	1.2651 KOps/s	$\color{#35bf28}+0.75\%$
test_func_call_runtime[True-compile-overhead]	0.8285ms	0.3918ms	2.5526 KOps/s	2.5445 KOps/s	$\color{#35bf28}+0.32\%$
test_func_call_cm_runtime[False-eager]	1.2021ms	0.7611ms	1.3139 KOps/s	1.2845 KOps/s	$\color{#35bf28}+2.29\%$
test_func_call_cm_runtime[False-compile]	1.2322ms	0.7855ms	1.2731 KOps/s	1.2952 KOps/s	$\color{#d91a1a}-1.70\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4295ms	0.3722ms	2.6869 KOps/s	2.6835 KOps/s	$\color{#35bf28}+0.13\%$
test_func_call_cm_runtime[True-eager]	1.1865ms	1.0147ms	985.5607 Ops/s	959.9797 Ops/s	$\color{#35bf28}+2.66\%$
test_func_call_cm_runtime[True-compile]	1.2869ms	0.8108ms	1.2333 KOps/s	1.2264 KOps/s	$\color{#35bf28}+0.56\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5390ms	0.4254ms	2.3507 KOps/s	2.3747 KOps/s	$\color{#d91a1a}-1.01\%$
test_vmap_func_call_cm_runtime[eager]	2.6534ms	2.1621ms	462.5229 Ops/s	467.0256 Ops/s	$\color{#d91a1a}-0.96\%$
test_vmap_func_call_cm_runtime[compile]	1.3475ms	0.8428ms	1.1866 KOps/s	1.1883 KOps/s	$\color{#d91a1a}-0.15\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4858ms	0.4220ms	2.3697 KOps/s	2.3526 KOps/s	$\color{#35bf28}+0.73\%$
test_distributed	3.0622ms	0.1799ms	5.5581 KOps/s	8.4847 KOps/s	$\textbf{\color{#d91a1a}-34.49\%}$
test_tdmodule	0.1282ms	13.8087μs	72.4179 KOps/s	69.6237 KOps/s	$\color{#35bf28}+4.01\%$
test_tdmodule_dispatch	47.1010μs	27.1238μs	36.8681 KOps/s	35.4598 KOps/s	$\color{#35bf28}+3.97\%$
test_tdseq	33.9600μs	15.1551μs	65.9845 KOps/s	63.4530 KOps/s	$\color{#35bf28}+3.99\%$
test_tdseq_dispatch	52.5220μs	30.1610μs	33.1554 KOps/s	31.9017 KOps/s	$\color{#35bf28}+3.93\%$
test_instantiation_functorch	1.7651ms	1.5590ms	641.4203 Ops/s	616.5528 Ops/s	$\color{#35bf28}+4.03\%$
test_exec_functorch	0.2492ms	0.1510ms	6.6209 KOps/s	6.6035 KOps/s	$\color{#35bf28}+0.26\%$
test_exec_functional_call	0.1939ms	0.1425ms	7.0173 KOps/s	6.8553 KOps/s	$\color{#35bf28}+2.36\%$
test_exec_td_decorator	0.3760ms	0.1863ms	5.3686 KOps/s	5.2104 KOps/s	$\color{#35bf28}+3.04\%$
test_vmap_mlp_speed_decorator[True-True]	1.1269ms	0.6850ms	1.4598 KOps/s	1.4503 KOps/s	$\color{#35bf28}+0.66\%$
test_vmap_mlp_speed_decorator[True-False]	1.0993ms	0.6837ms	1.4626 KOps/s	1.4464 KOps/s	$\color{#35bf28}+1.12\%$
test_vmap_mlp_speed_decorator[False-True]	1.0784ms	0.6058ms	1.6506 KOps/s	1.5869 KOps/s	$\color{#35bf28}+4.02\%$
test_vmap_mlp_speed_decorator[False-False]	1.0319ms	0.6035ms	1.6570 KOps/s	1.5886 KOps/s	$\color{#35bf28}+4.31\%$
test_vmap_transformer_speed_decorator[True-True]	20.5019ms	19.5734ms	51.0896 Ops/s	50.5036 Ops/s	$\color{#35bf28}+1.16\%$
test_vmap_transformer_speed_decorator[True-False]	20.0139ms	19.6333ms	50.9339 Ops/s	50.6989 Ops/s	$\color{#35bf28}+0.46\%$
test_vmap_transformer_speed_decorator[False-True]	19.8487ms	19.4924ms	51.3020 Ops/s	50.7647 Ops/s	$\color{#35bf28}+1.06\%$
test_vmap_transformer_speed_decorator[False-False]	20.0441ms	19.4771ms	51.3425 Ops/s	50.6562 Ops/s	$\color{#35bf28}+1.35\%$
test_to_module_speed[True]	2.2927ms	0.9400ms	1.0639 KOps/s	1.0635 KOps/s	$\color{#35bf28}+0.04\%$
test_to_module_speed[False]	1.0583ms	0.9215ms	1.0852 KOps/s	1.0855 KOps/s	$\color{#d91a1a}-0.02\%$
test_tc_init	72.8610μs	35.9081μs	27.8489 KOps/s	27.5689 KOps/s	$\color{#35bf28}+1.02\%$
test_tc_init_nested	0.1254ms	70.3635μs	14.2119 KOps/s	14.2986 KOps/s	$\color{#d91a1a}-0.61\%$
test_tc_first_layer_tensor	13.6360μs	0.7007μs	1.4271 MOps/s	1.4132 MOps/s	$\color{#35bf28}+0.98\%$
test_tc_first_layer_nontensor	91.3620μs	2.2828μs	438.0667 KOps/s	432.7547 KOps/s	$\color{#35bf28}+1.23\%$
test_tc_second_layer_tensor	97.3667μs	1.4289μs	699.8313 KOps/s	688.3363 KOps/s	$\color{#35bf28}+1.67\%$
test_tc_second_layer_nontensor	28.6210μs	3.0622μs	326.5575 KOps/s	332.3500 KOps/s	$\color{#d91a1a}-1.74\%$
test_unbind	6.9423ms	6.6896ms	149.4860 Ops/s	150.7109 Ops/s	$\color{#d91a1a}-0.81\%$
test_full_like	12.1765ms	9.1553ms	109.2259 Ops/s	107.0234 Ops/s	$\color{#35bf28}+2.06\%$
test_zeros_like	5.8828ms	4.2183ms	237.0643 Ops/s	114.2577 Ops/s	$\textbf{\color{#35bf28}+107.48\%}$
test_ones_like	4.4575ms	4.2324ms	236.2753 Ops/s	231.3651 Ops/s	$\color{#35bf28}+2.12\%$
test_clone	11.3526ms	9.0458ms	110.5485 Ops/s	157.1430 Ops/s	$\textbf{\color{#d91a1a}-29.65\%}$
test_squeeze	0.4395ms	9.2157μs	108.5105 KOps/s	107.7182 KOps/s	$\color{#35bf28}+0.74\%$
test_unsqueeze	0.1183ms	69.7808μs	14.3306 KOps/s	13.1858 KOps/s	$\textbf{\color{#35bf28}+8.68\%}$
test_split	0.5771ms	0.1572ms	6.3604 KOps/s	6.2373 KOps/s	$\color{#35bf28}+1.97\%$
test_permute	0.2550ms	0.1839ms	5.4367 KOps/s	5.1887 KOps/s	$\color{#35bf28}+4.78\%$
test_stack	51.2410ms	50.7933ms	19.6876 Ops/s	19.5819 Ops/s	$\color{#35bf28}+0.54\%$
test_cat	50.8395ms	50.4707ms	19.8135 Ops/s	19.7042 Ops/s	$\color{#35bf28}+0.55\%$