[Feature,Refactor] Refactor from_dict, add from_any, from_dataclass

[vmoens]

Stack from ghstack (oldest at bottom):

• [BugFix] select_out_keys for Prob sequential #1103
• -> [Feature,Refactor] Refactor from_dict, add from_any, from_dataclass #1102

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}48$. Worsened: $\large\color{#d91a1a}10$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	34.1030μs	17.2870μs	57.8470 KOps/s	56.3746 KOps/s	$\color{#35bf28}+2.61\%$
test_plain_set_stack_nested	65.0210μs	17.4660μs	57.2542 KOps/s	56.0036 KOps/s	$\color{#35bf28}+2.23\%$
test_plain_set_nested_inplace	55.3240μs	18.9520μs	52.7649 KOps/s	51.5807 KOps/s	$\color{#35bf28}+2.30\%$
test_plain_set_stack_nested_inplace	72.4350μs	18.8674μs	53.0016 KOps/s	52.2718 KOps/s	$\color{#35bf28}+1.40\%$
test_items	21.6310μs	4.1198μs	242.7306 KOps/s	243.4579 KOps/s	$\color{#d91a1a}-0.30\%$
test_items_nested	0.4708ms	0.3333ms	3.0002 KOps/s	2.8415 KOps/s	$\textbf{\color{#35bf28}+5.58\%}$
test_items_nested_locked	0.6368ms	0.3329ms	3.0037 KOps/s	2.8708 KOps/s	$\color{#35bf28}+4.63\%$
test_items_nested_leaf	0.1356ms	71.4812μs	13.9897 KOps/s	14.0393 KOps/s	$\color{#d91a1a}-0.35\%$
test_items_stack_nested	0.4881ms	0.3383ms	2.9557 KOps/s	2.8419 KOps/s	$\color{#35bf28}+4.00\%$
test_items_stack_nested_leaf	0.1323ms	75.5267μs	13.2403 KOps/s	13.4843 KOps/s	$\color{#d91a1a}-1.81\%$
test_items_stack_nested_locked	0.4903ms	0.3332ms	3.0015 KOps/s	2.7671 KOps/s	$\textbf{\color{#35bf28}+8.47\%}$
test_keys	22.6730μs	3.8586μs	259.1585 KOps/s	282.0440 KOps/s	$\textbf{\color{#d91a1a}-8.11\%}$
test_keys_nested	0.2286ms	0.1345ms	7.4353 KOps/s	7.3717 KOps/s	$\color{#35bf28}+0.86\%$
test_keys_nested_locked	0.7490ms	0.1394ms	7.1740 KOps/s	7.1005 KOps/s	$\color{#35bf28}+1.03\%$
test_keys_nested_leaf	1.8942ms	0.1165ms	8.5854 KOps/s	8.5536 KOps/s	$\color{#35bf28}+0.37\%$
test_keys_stack_nested	0.2239ms	0.1343ms	7.4446 KOps/s	7.1735 KOps/s	$\color{#35bf28}+3.78\%$
test_keys_stack_nested_leaf	0.1927ms	0.1142ms	8.7581 KOps/s	8.5811 KOps/s	$\color{#35bf28}+2.06\%$
test_keys_stack_nested_locked	0.2301ms	0.1391ms	7.1912 KOps/s	7.0244 KOps/s	$\color{#35bf28}+2.37\%$
test_values	6.7000μs	1.0260μs	974.6449 KOps/s	963.5079 KOps/s	$\color{#35bf28}+1.16\%$
test_values_nested	0.1001ms	54.5122μs	18.3445 KOps/s	18.0208 KOps/s	$\color{#35bf28}+1.80\%$
test_values_nested_locked	0.1201ms	55.2840μs	18.0884 KOps/s	18.1233 KOps/s	$\color{#d91a1a}-0.19\%$
test_values_nested_leaf	0.1062ms	59.8788μs	16.7004 KOps/s	16.6361 KOps/s	$\color{#35bf28}+0.39\%$
test_values_stack_nested	0.1053ms	56.7540μs	17.6199 KOps/s	16.8062 KOps/s	$\color{#35bf28}+4.84\%$
test_values_stack_nested_leaf	0.1097ms	59.0373μs	16.9385 KOps/s	16.3261 KOps/s	$\color{#35bf28}+3.75\%$
test_values_stack_nested_locked	0.1089ms	57.0036μs	17.5427 KOps/s	17.6031 KOps/s	$\color{#d91a1a}-0.34\%$
test_membership	6.5280μs	0.7469μs	1.3388 MOps/s	1.1018 MOps/s	$\textbf{\color{#35bf28}+21.51\%}$
test_membership_nested	43.1400μs	2.8189μs	354.7493 KOps/s	365.0361 KOps/s	$\color{#d91a1a}-2.82\%$
test_membership_nested_leaf	23.5540μs	2.8578μs	349.9173 KOps/s	357.4106 KOps/s	$\color{#d91a1a}-2.10\%$
test_membership_stacked_nested	43.9210μs	2.7756μs	360.2785 KOps/s	359.4974 KOps/s	$\color{#35bf28}+0.22\%$
test_membership_stacked_nested_leaf	23.0830μs	2.8034μs	356.7145 KOps/s	361.4756 KOps/s	$\color{#d91a1a}-1.32\%$
test_membership_nested_last	46.9570μs	4.0779μs	245.2232 KOps/s	246.9645 KOps/s	$\color{#d91a1a}-0.71\%$
test_membership_nested_leaf_last	35.4450μs	4.0621μs	246.1788 KOps/s	245.3356 KOps/s	$\color{#35bf28}+0.34\%$
test_membership_stacked_nested_last	0.1060ms	7.5913μs	131.7301 KOps/s	195.0668 KOps/s	$\textbf{\color{#d91a1a}-32.47\%}$
test_membership_stacked_nested_leaf_last	75.7500μs	7.5953μs	131.6600 KOps/s	193.5114 KOps/s	$\textbf{\color{#d91a1a}-31.96\%}$
test_nested_getleaf	50.7450μs	10.4127μs	96.0366 KOps/s	94.7349 KOps/s	$\color{#35bf28}+1.37\%$
test_nested_get	51.1850μs	9.9014μs	100.9954 KOps/s	98.9429 KOps/s	$\color{#35bf28}+2.07\%$
test_stacked_getleaf	35.3960μs	10.4981μs	95.2551 KOps/s	94.9434 KOps/s	$\color{#35bf28}+0.33\%$
test_stacked_get	50.6590μs	9.7251μs	102.8271 KOps/s	99.0751 KOps/s	$\color{#35bf28}+3.79\%$
test_nested_getitemleaf	36.5880μs	10.9133μs	91.6311 KOps/s	90.2462 KOps/s	$\color{#35bf28}+1.53\%$
test_nested_getitem	54.7520μs	10.2976μs	97.1103 KOps/s	96.7406 KOps/s	$\color{#35bf28}+0.38\%$
test_stacked_getitemleaf	39.8240μs	10.7964μs	92.6232 KOps/s	90.2104 KOps/s	$\color{#35bf28}+2.67\%$
test_stacked_getitem	48.9610μs	10.0986μs	99.0236 KOps/s	97.9131 KOps/s	$\color{#35bf28}+1.13\%$
test_lock_nested	1.9152ms	0.4342ms	2.3031 KOps/s	1.8559 KOps/s	$\textbf{\color{#35bf28}+24.09\%}$
test_lock_stack_nested	0.6106ms	0.4028ms	2.4829 KOps/s	2.4118 KOps/s	$\color{#35bf28}+2.95\%$
test_unlock_nested	0.6661ms	0.3530ms	2.8328 KOps/s	2.7649 KOps/s	$\color{#35bf28}+2.46\%$
test_unlock_stack_nested	0.4987ms	0.3203ms	3.1218 KOps/s	3.0073 KOps/s	$\color{#35bf28}+3.81\%$
test_flatten_speed	0.2071ms	93.1737μs	10.7326 KOps/s	10.6823 KOps/s	$\color{#35bf28}+0.47\%$
test_unflatten_speed	0.8802ms	0.4786ms	2.0893 KOps/s	2.0879 KOps/s	$\color{#35bf28}+0.07\%$
test_common_ops	6.0371ms	0.7966ms	1.2553 KOps/s	1.3015 KOps/s	$\color{#d91a1a}-3.55\%$
test_creation	17.0620μs	2.0621μs	484.9500 KOps/s	480.0849 KOps/s	$\color{#35bf28}+1.01\%$
test_creation_empty	36.3880μs	10.3491μs	96.6264 KOps/s	92.9795 KOps/s	$\color{#35bf28}+3.92\%$
test_creation_nested_1	42.6490μs	13.0626μs	76.5543 KOps/s	72.8668 KOps/s	$\textbf{\color{#35bf28}+5.06\%}$
test_creation_nested_2	48.3000μs	17.6919μs	56.5232 KOps/s	56.8051 KOps/s	$\color{#d91a1a}-0.50\%$
test_clone	45.2840μs	13.1196μs	76.2218 KOps/s	72.8570 KOps/s	$\color{#35bf28}+4.62\%$
test_getitem[int]	0.9921ms	12.3675μs	80.8573 KOps/s	76.5172 KOps/s	$\textbf{\color{#35bf28}+5.67\%}$
test_getitem[slice_int]	0.1361ms	23.2593μs	42.9936 KOps/s	38.9394 KOps/s	$\textbf{\color{#35bf28}+10.41\%}$
test_getitem[range]	0.1744ms	48.4228μs	20.6514 KOps/s	20.5514 KOps/s	$\color{#35bf28}+0.49\%$
test_getitem[tuple]	0.1383ms	19.4972μs	51.2895 KOps/s	48.4659 KOps/s	$\textbf{\color{#35bf28}+5.83\%}$
test_getitem[list]	0.1961ms	43.6847μs	22.8913 KOps/s	22.5987 KOps/s	$\color{#35bf28}+1.29\%$
test_setitem_dim[int]	57.7580μs	24.7905μs	40.3380 KOps/s	39.2229 KOps/s	$\color{#35bf28}+2.84\%$
test_setitem_dim[slice_int]	86.1200μs	50.1646μs	19.9344 KOps/s	19.4117 KOps/s	$\color{#35bf28}+2.69\%$
test_setitem_dim[range]	0.1170ms	72.6845μs	13.7581 KOps/s	13.5223 KOps/s	$\color{#35bf28}+1.74\%$
test_setitem_dim[tuple]	74.9800μs	40.4840μs	24.7011 KOps/s	24.4349 KOps/s	$\color{#35bf28}+1.09\%$
test_setitem	56.7160μs	19.6712μs	50.8356 KOps/s	47.7425 KOps/s	$\textbf{\color{#35bf28}+6.48\%}$
test_set	0.1642ms	18.9497μs	52.7712 KOps/s	48.2414 KOps/s	$\textbf{\color{#35bf28}+9.39\%}$
test_set_shared	1.3216ms	0.1691ms	5.9151 KOps/s	5.8857 KOps/s	$\color{#35bf28}+0.50\%$
test_update	0.2022ms	20.8645μs	47.9282 KOps/s	43.6360 KOps/s	$\textbf{\color{#35bf28}+9.84\%}$
test_update_nested	0.2165ms	31.3960μs	31.8511 KOps/s	30.2551 KOps/s	$\textbf{\color{#35bf28}+5.28\%}$
test_update__nested	0.6793ms	32.3547μs	30.9074 KOps/s	29.7656 KOps/s	$\color{#35bf28}+3.84\%$
test_set_nested	0.1417ms	20.8131μs	48.0466 KOps/s	44.8777 KOps/s	$\textbf{\color{#35bf28}+7.06\%}$
test_set_nested_new	0.2446ms	25.5363μs	39.1600 KOps/s	37.6342 KOps/s	$\color{#35bf28}+4.05\%$
test_select	0.2688ms	41.1061μs	24.3273 KOps/s	23.1139 KOps/s	$\textbf{\color{#35bf28}+5.25\%}$
test_select_nested	0.1276ms	59.6169μs	16.7738 KOps/s	16.4294 KOps/s	$\color{#35bf28}+2.10\%$
test_exclude_nested	0.1358ms	73.8328μs	13.5441 KOps/s	13.2114 KOps/s	$\color{#35bf28}+2.52\%$
test_empty[True]	0.5639ms	0.3470ms	2.8820 KOps/s	2.8365 KOps/s	$\color{#35bf28}+1.60\%$
test_empty[False]	10.6897μs	1.1889μs	841.0938 KOps/s	802.0429 KOps/s	$\color{#35bf28}+4.87\%$
test_unbind_speed	0.4586ms	0.2568ms	3.8939 KOps/s	3.7382 KOps/s	$\color{#35bf28}+4.17\%$
test_unbind_speed_stack0	0.4941ms	0.2518ms	3.9714 KOps/s	3.8478 KOps/s	$\color{#35bf28}+3.21\%$
test_unbind_speed_stack1	0.1020s	0.7463ms	1.3400 KOps/s	1.4110 KOps/s	$\textbf{\color{#d91a1a}-5.03\%}$
test_split	2.4748ms	1.5662ms	638.5084 Ops/s	565.7705 Ops/s	$\textbf{\color{#35bf28}+12.86\%}$
test_chunk	0.1020s	1.8673ms	535.5227 Ops/s	566.1277 Ops/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_consolidate_njt[False-None]	9.6366ms	8.1948ms	122.0289 Ops/s	120.3956 Ops/s	$\color{#35bf28}+1.36\%$
test_creation[device0]	3.9614ms	92.2221μs	10.8434 KOps/s	10.7099 KOps/s	$\color{#35bf28}+1.25\%$
test_creation_from_tensor	0.2554ms	94.9987μs	10.5265 KOps/s	10.3980 KOps/s	$\color{#35bf28}+1.24\%$
test_add_one[memmap_tensor0]	0.2270ms	4.6118μs	216.8329 KOps/s	194.9969 KOps/s	$\textbf{\color{#35bf28}+11.20\%}$
test_contiguous[memmap_tensor0]	8.7260μs	0.5118μs	1.9541 MOps/s	1.9214 MOps/s	$\color{#35bf28}+1.70\%$
test_stack[memmap_tensor0]	55.3730μs	3.3043μs	302.6331 KOps/s	284.3044 KOps/s	$\textbf{\color{#35bf28}+6.45\%}$
test_memmaptd_index	0.4455ms	0.2277ms	4.3915 KOps/s	4.0766 KOps/s	$\textbf{\color{#35bf28}+7.73\%}$
test_memmaptd_index_astensor	0.5367ms	0.3045ms	3.2839 KOps/s	3.0965 KOps/s	$\textbf{\color{#35bf28}+6.05\%}$
test_memmaptd_index_op	1.4144ms	0.5639ms	1.7732 KOps/s	1.6572 KOps/s	$\textbf{\color{#35bf28}+7.00\%}$
test_serialize_model	0.1203s	0.1177s	8.4928 Ops/s	7.6414 Ops/s	$\textbf{\color{#35bf28}+11.14\%}$
test_serialize_model_pickle	0.4954s	0.3914s	2.5546 Ops/s	2.5950 Ops/s	$\color{#d91a1a}-1.56\%$
test_serialize_weights	0.2182s	0.1301s	7.6863 Ops/s	8.9022 Ops/s	$\textbf{\color{#d91a1a}-13.66\%}$
test_serialize_weights_returnearly	0.1717s	0.1588s	6.2973 Ops/s	6.3515 Ops/s	$\color{#d91a1a}-0.85\%$
test_serialize_weights_pickle	1.1507s	0.7034s	1.4217 Ops/s	2.5563 Ops/s	$\textbf{\color{#d91a1a}-44.38\%}$
test_serialize_weights_filesystem	0.1470s	0.1411s	7.0872 Ops/s	6.4512 Ops/s	$\textbf{\color{#35bf28}+9.86\%}$
test_serialize_model_filesystem	0.2491s	0.1533s	6.5214 Ops/s	6.7168 Ops/s	$\color{#d91a1a}-2.91\%$
test_reshape_pytree	60.5230μs	26.6092μs	37.5810 KOps/s	36.5519 KOps/s	$\color{#35bf28}+2.82\%$
test_reshape_td	87.9540μs	31.9148μs	31.3334 KOps/s	29.2872 KOps/s	$\textbf{\color{#35bf28}+6.99\%}$
test_view_pytree	57.3060μs	26.4279μs	37.8387 KOps/s	37.2346 KOps/s	$\color{#35bf28}+1.62\%$
test_view_td	77.3240μs	36.9628μs	27.0542 KOps/s	24.9384 KOps/s	$\textbf{\color{#35bf28}+8.48\%}$
test_unbind_pytree	79.5680μs	29.5646μs	33.8243 KOps/s	33.0724 KOps/s	$\color{#35bf28}+2.27\%$
test_unbind_td	0.3192ms	37.7963μs	26.4576 KOps/s	25.4117 KOps/s	$\color{#35bf28}+4.12\%$
test_split_pytree	73.3970μs	29.2527μs	34.1849 KOps/s	33.0969 KOps/s	$\color{#35bf28}+3.29\%$
test_split_td	0.5412ms	43.3068μs	23.0911 KOps/s	21.1785 KOps/s	$\textbf{\color{#35bf28}+9.03\%}$
test_add_pytree	75.4010μs	34.9125μs	28.6431 KOps/s	27.0223 KOps/s	$\textbf{\color{#35bf28}+6.00\%}$
test_add_td	0.1317ms	55.0160μs	18.1765 KOps/s	17.2024 KOps/s	$\textbf{\color{#35bf28}+5.66\%}$
test_compile_add_one_nested[tensordict-compile]	0.1557ms	61.7519μs	16.1938 KOps/s	16.3039 KOps/s	$\color{#d91a1a}-0.68\%$
test_compile_add_one_nested[tensordict-eager]	1.4120ms	0.1580ms	6.3281 KOps/s	6.2149 KOps/s	$\color{#35bf28}+1.82\%$
test_compile_add_one_nested[pytree-compile]	0.1082ms	44.7869μs	22.3279 KOps/s	21.6575 KOps/s	$\color{#35bf28}+3.10\%$
test_compile_add_one_nested[pytree-eager]	0.2241ms	0.1174ms	8.5206 KOps/s	8.2820 KOps/s	$\color{#35bf28}+2.88\%$
test_compile_copy_nested[tensordict-compile]	73.0360μs	25.4129μs	39.3500 KOps/s	38.5797 KOps/s	$\color{#35bf28}+2.00\%$
test_compile_copy_nested[tensordict-eager]	0.1225ms	54.8542μs	18.2301 KOps/s	18.6301 KOps/s	$\color{#d91a1a}-2.15\%$
test_compile_copy_nested[pytree-compile]	0.1498ms	79.2385μs	12.6201 KOps/s	12.5375 KOps/s	$\color{#35bf28}+0.66\%$
test_compile_copy_nested[pytree-eager]	0.1538ms	68.1158μs	14.6809 KOps/s	14.6091 KOps/s	$\color{#35bf28}+0.49\%$
test_compile_add_one_flat[tensordict-compile]	0.2018ms	0.1038ms	9.6378 KOps/s	9.3719 KOps/s	$\color{#35bf28}+2.84\%$
test_compile_add_one_flat[tensordict-eager]	0.3213ms	0.1936ms	5.1648 KOps/s	4.9857 KOps/s	$\color{#35bf28}+3.59\%$
test_compile_add_one_flat[tensorclass-compile]	88.7450μs	44.1390μs	22.6557 KOps/s	22.3760 KOps/s	$\color{#35bf28}+1.25\%$
test_compile_add_one_flat[tensorclass-eager]	0.4773ms	60.8717μs	16.4280 KOps/s	16.3307 KOps/s	$\color{#35bf28}+0.60\%$
test_compile_add_one_flat[pytree-compile]	0.1870ms	0.1017ms	9.8339 KOps/s	9.6050 KOps/s	$\color{#35bf28}+2.38\%$
test_compile_add_one_flat[pytree-eager]	0.3715ms	0.2018ms	4.9559 KOps/s	4.8608 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_add_self_flat[tensordict-eager]	0.3409ms	0.2046ms	4.8865 KOps/s	4.7165 KOps/s	$\color{#35bf28}+3.61\%$
test_compile_add_self_flat[tensordict-compile]	0.2032ms	0.1055ms	9.4763 KOps/s	9.3426 KOps/s	$\color{#35bf28}+1.43\%$
test_compile_add_self_flat[tensorclass-eager]	0.2444ms	57.0981μs	17.5137 KOps/s	18.3048 KOps/s	$\color{#d91a1a}-4.32\%$
test_compile_add_self_flat[tensorclass-compile]	0.1232ms	46.2514μs	21.6210 KOps/s	20.9764 KOps/s	$\color{#35bf28}+3.07\%$
test_compile_add_self_flat[pytree-eager]	1.6657ms	0.1605ms	6.2288 KOps/s	6.2372 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_add_self_flat[pytree-compile]	0.1837ms	0.1017ms	9.8299 KOps/s	9.5112 KOps/s	$\color{#35bf28}+3.35\%$
test_compile_copy_flat[tensordict-compile]	59.0600μs	20.9775μs	47.6702 KOps/s	46.9939 KOps/s	$\color{#35bf28}+1.44\%$
test_compile_copy_flat[tensordict-eager]	0.1342ms	57.7505μs	17.3159 KOps/s	16.7974 KOps/s	$\color{#35bf28}+3.09\%$
test_compile_copy_flat[pytree-compile]	0.1754ms	82.6690μs	12.0964 KOps/s	12.1435 KOps/s	$\color{#d91a1a}-0.39\%$
test_compile_copy_flat[pytree-eager]	0.1361ms	68.3195μs	14.6371 KOps/s	14.3555 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_assign_and_add[tensordict-compile]	0.7159ms	0.2051ms	4.8755 KOps/s	4.6262 KOps/s	$\textbf{\color{#35bf28}+5.39\%}$
test_compile_assign_and_add[tensordict-eager]	1.4894ms	1.2318ms	811.7940 Ops/s	771.6898 Ops/s	$\textbf{\color{#35bf28}+5.20\%}$
test_compile_assign_and_add[pytree-compile]	0.3275ms	0.1987ms	5.0330 KOps/s	4.7738 KOps/s	$\textbf{\color{#35bf28}+5.43\%}$
test_compile_assign_and_add[pytree-eager]	1.5917ms	0.7686ms	1.3011 KOps/s	1.2668 KOps/s	$\color{#35bf28}+2.71\%$
test_compile_assign_and_add_stack[compile]	0.5306ms	0.4500ms	2.2224 KOps/s	2.1390 KOps/s	$\color{#35bf28}+3.90\%$
test_compile_assign_and_add_stack[eager]	2.6093ms	2.4827ms	402.7908 Ops/s	384.7244 Ops/s	$\color{#35bf28}+4.70\%$
test_compile_indexing[tensor-tensordict-compile]	83.9460μs	35.3448μs	28.2927 KOps/s	27.9305 KOps/s	$\color{#35bf28}+1.30\%$
test_compile_indexing[tensor-tensordict-eager]	0.5600ms	31.6103μs	31.6353 KOps/s	29.4471 KOps/s	$\textbf{\color{#35bf28}+7.43\%}$
test_compile_indexing[tensor-tensorclass-compile]	80.3900μs	28.3866μs	35.2279 KOps/s	34.2757 KOps/s	$\color{#35bf28}+2.78\%$
test_compile_indexing[tensor-tensorclass-eager]	62.2760μs	22.7733μs	43.9111 KOps/s	43.2975 KOps/s	$\color{#35bf28}+1.42\%$
test_compile_indexing[tensor-pytree-compile]	72.0240μs	29.0890μs	34.3773 KOps/s	33.4281 KOps/s	$\color{#35bf28}+2.84\%$
test_compile_indexing[tensor-pytree-eager]	62.2260μs	22.7869μs	43.8849 KOps/s	43.1447 KOps/s	$\color{#35bf28}+1.72\%$
test_compile_indexing[slice-tensordict-compile]	0.1083ms	51.1309μs	19.5576 KOps/s	19.0059 KOps/s	$\color{#35bf28}+2.90\%$
test_compile_indexing[slice-tensordict-eager]	0.3474ms	18.7119μs	53.4419 KOps/s	48.1986 KOps/s	$\textbf{\color{#35bf28}+10.88\%}$
test_compile_indexing[slice-tensorclass-compile]	0.1494ms	44.3547μs	22.5455 KOps/s	22.0423 KOps/s	$\color{#35bf28}+2.28\%$
test_compile_indexing[slice-tensorclass-eager]	59.4100μs	18.6061μs	53.7458 KOps/s	52.1190 KOps/s	$\color{#35bf28}+3.12\%$
test_compile_indexing[slice-pytree-compile]	0.1092ms	45.0308μs	22.2070 KOps/s	21.6272 KOps/s	$\color{#35bf28}+2.68\%$
test_compile_indexing[slice-pytree-eager]	94.1740μs	18.7034μs	53.4663 KOps/s	52.3221 KOps/s	$\color{#35bf28}+2.19\%$
test_compile_indexing[int-tensordict-compile]	0.1206ms	51.5705μs	19.3909 KOps/s	18.5757 KOps/s	$\color{#35bf28}+4.39\%$
test_compile_indexing[int-tensordict-eager]	0.9293ms	19.1896μs	52.1117 KOps/s	47.7547 KOps/s	$\textbf{\color{#35bf28}+9.12\%}$
test_compile_indexing[int-tensorclass-compile]	0.1071ms	44.4743μs	22.4849 KOps/s	21.5909 KOps/s	$\color{#35bf28}+4.14\%$
test_compile_indexing[int-tensorclass-eager]	70.6110μs	18.6178μs	53.7120 KOps/s	51.5279 KOps/s	$\color{#35bf28}+4.24\%$
test_compile_indexing[int-pytree-compile]	0.1022ms	43.8393μs	22.8106 KOps/s	21.4974 KOps/s	$\textbf{\color{#35bf28}+6.11\%}$
test_compile_indexing[int-pytree-eager]	0.4314ms	18.5020μs	54.0481 KOps/s	52.6353 KOps/s	$\color{#35bf28}+2.68\%$
test_mod_add[eager]	70.7420μs	25.1104μs	39.8241 KOps/s	37.1479 KOps/s	$\textbf{\color{#35bf28}+7.20\%}$
test_mod_add[compile]	89.2470μs	43.5611μs	22.9562 KOps/s	22.0727 KOps/s	$\color{#35bf28}+4.00\%$
test_mod_add[compile-overhead]	93.7950μs	42.9986μs	23.2565 KOps/s	21.2648 KOps/s	$\textbf{\color{#35bf28}+9.37\%}$
test_mod_wrap[eager]	0.3077ms	0.2090ms	4.7838 KOps/s	4.6559 KOps/s	$\color{#35bf28}+2.75\%$
test_mod_wrap[compile]	1.8893ms	0.2004ms	4.9911 KOps/s	4.7532 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_mod_wrap[compile-overhead]	1.7310ms	0.1971ms	5.0746 KOps/s	4.6126 KOps/s	$\textbf{\color{#35bf28}+10.01\%}$
test_mod_wrap_and_backward[eager]	14.0737ms	11.0926ms	90.1503 Ops/s	93.3450 Ops/s	$\color{#d91a1a}-3.42\%$
test_mod_wrap_and_backward[compile]	15.7313ms	12.4913ms	80.0554 Ops/s	87.2356 Ops/s	$\textbf{\color{#d91a1a}-8.23\%}$
test_mod_wrap_and_backward[compile-overhead]	13.8482ms	11.8640ms	84.2883 Ops/s	93.8170 Ops/s	$\textbf{\color{#d91a1a}-10.16\%}$
test_seq_add[eager]	0.2176ms	90.2009μs	11.0864 KOps/s	10.7605 KOps/s	$\color{#35bf28}+3.03\%$
test_seq_add[compile]	0.1128ms	57.6025μs	17.3604 KOps/s	16.5197 KOps/s	$\textbf{\color{#35bf28}+5.09\%}$
test_seq_add[compile-overhead]	0.1413ms	57.1612μs	17.4944 KOps/s	16.5699 KOps/s	$\textbf{\color{#35bf28}+5.58\%}$
test_seq_wrap[eager]	0.5687ms	0.3848ms	2.5985 KOps/s	2.5110 KOps/s	$\color{#35bf28}+3.48\%$
test_seq_wrap[compile]	0.3591ms	0.2215ms	4.5142 KOps/s	4.2936 KOps/s	$\textbf{\color{#35bf28}+5.14\%}$
test_seq_wrap[compile-overhead]	0.4073ms	0.2219ms	4.5073 KOps/s	4.3496 KOps/s	$\color{#35bf28}+3.63\%$
test_func_call_runtime[False-eager]	0.7679ms	0.5324ms	1.8784 KOps/s	1.8023 KOps/s	$\color{#35bf28}+4.22\%$
test_func_call_runtime[False-compile]	0.5082ms	0.4166ms	2.4002 KOps/s	2.2897 KOps/s	$\color{#35bf28}+4.82\%$
test_func_call_runtime[False-compile-overhead]	0.9115ms	0.4190ms	2.3864 KOps/s	2.2869 KOps/s	$\color{#35bf28}+4.35\%$
test_func_call_runtime[True-eager]	1.1993ms	0.7445ms	1.3432 KOps/s	1.2912 KOps/s	$\color{#35bf28}+4.03\%$
test_func_call_runtime[True-compile]	1.0872ms	0.4601ms	2.1737 KOps/s	2.1088 KOps/s	$\color{#35bf28}+3.08\%$
test_func_call_runtime[True-compile-overhead]	0.5948ms	0.4522ms	2.2112 KOps/s	2.0806 KOps/s	$\textbf{\color{#35bf28}+6.27\%}$
test_func_call_cm_runtime[False-eager]	0.9855ms	0.5276ms	1.8953 KOps/s	1.8013 KOps/s	$\textbf{\color{#35bf28}+5.21\%}$
test_func_call_cm_runtime[False-compile]	0.5194ms	0.4149ms	2.4105 KOps/s	2.2642 KOps/s	$\textbf{\color{#35bf28}+6.46\%}$
test_func_call_cm_runtime[False-compile-overhead]	0.5536ms	0.4142ms	2.4145 KOps/s	2.2759 KOps/s	$\textbf{\color{#35bf28}+6.09\%}$
test_func_call_cm_runtime[True-eager]	1.3418ms	0.8693ms	1.1504 KOps/s	1.1087 KOps/s	$\color{#35bf28}+3.76\%$
test_func_call_cm_runtime[True-compile]	0.6017ms	0.4782ms	2.0912 KOps/s	1.9835 KOps/s	$\textbf{\color{#35bf28}+5.43\%}$
test_func_call_cm_runtime[True-compile-overhead]	0.5790ms	0.4795ms	2.0855 KOps/s	1.9910 KOps/s	$\color{#35bf28}+4.75\%$
test_vmap_func_call_cm_runtime[eager]	2.6558ms	1.8684ms	535.2033 Ops/s	516.8055 Ops/s	$\color{#35bf28}+3.56\%$
test_vmap_func_call_cm_runtime[compile]	1.0098ms	0.5075ms	1.9703 KOps/s	1.8930 KOps/s	$\color{#35bf28}+4.08\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.9966ms	0.5083ms	1.9675 KOps/s	1.8880 KOps/s	$\color{#35bf28}+4.21\%$
test_distributed	0.2546ms	0.1258ms	7.9504 KOps/s	7.8288 KOps/s	$\color{#35bf28}+1.55\%$
test_tdmodule	33.9740μs	18.3381μs	54.5312 KOps/s	53.9625 KOps/s	$\color{#35bf28}+1.05\%$
test_tdmodule_dispatch	82.0930μs	39.4660μs	25.3382 KOps/s	27.5962 KOps/s	$\textbf{\color{#d91a1a}-8.18\%}$
test_tdseq	42.3890μs	21.0052μs	47.6073 KOps/s	47.2447 KOps/s	$\color{#35bf28}+0.77\%$
test_tdseq_dispatch	78.8970μs	43.2497μs	23.1215 KOps/s	20.6295 KOps/s	$\textbf{\color{#35bf28}+12.08\%}$
test_instantiation_functorch	1.7781ms	1.4981ms	667.5203 Ops/s	647.6180 Ops/s	$\color{#35bf28}+3.07\%$
test_exec_functorch	0.3968ms	0.1791ms	5.5822 KOps/s	5.3712 KOps/s	$\color{#35bf28}+3.93\%$
test_exec_functional_call	0.4471ms	0.1727ms	5.7906 KOps/s	5.6065 KOps/s	$\color{#35bf28}+3.28\%$
test_exec_td_decorator	0.4808ms	0.2241ms	4.4622 KOps/s	4.2987 KOps/s	$\color{#35bf28}+3.80\%$
test_vmap_mlp_speed_decorator[True-True]	0.9224ms	0.6383ms	1.5666 KOps/s	1.5432 KOps/s	$\color{#35bf28}+1.52\%$
test_vmap_mlp_speed_decorator[True-False]	0.9513ms	0.6377ms	1.5681 KOps/s	1.5391 KOps/s	$\color{#35bf28}+1.89\%$
test_vmap_mlp_speed_decorator[False-True]	0.7763ms	0.5262ms	1.9006 KOps/s	1.8658 KOps/s	$\color{#35bf28}+1.86\%$
test_vmap_mlp_speed_decorator[False-False]	0.8411ms	0.5229ms	1.9124 KOps/s	1.8672 KOps/s	$\color{#35bf28}+2.42\%$
test_to_module_speed[True]	2.0473ms	1.3008ms	768.7777 Ops/s	784.4814 Ops/s	$\color{#d91a1a}-2.00\%$
test_to_module_speed[False]	1.9752ms	1.2552ms	796.6598 Ops/s	796.4047 Ops/s	$\color{#35bf28}+0.03\%$
test_tc_init	81.1920μs	45.6105μs	21.9248 KOps/s	21.8243 KOps/s	$\color{#35bf28}+0.46\%$
test_tc_init_nested	0.1603ms	89.9266μs	11.1202 KOps/s	10.7845 KOps/s	$\color{#35bf28}+3.11\%$
test_tc_first_layer_tensor	17.7030μs	1.5086μs	662.8518 KOps/s	665.6418 KOps/s	$\color{#d91a1a}-0.42\%$
test_tc_first_layer_nontensor	44.4630μs	4.6173μs	216.5784 KOps/s	208.1628 KOps/s	$\color{#35bf28}+4.04\%$
test_tc_second_layer_tensor	20.7280μs	2.7547μs	363.0162 KOps/s	356.6495 KOps/s	$\color{#35bf28}+1.79\%$
test_tc_second_layer_nontensor	46.9370μs	5.9589μs	167.8157 KOps/s	162.0685 KOps/s	$\color{#35bf28}+3.55\%$
test_unbind	0.2155s	11.2444ms	88.9331 Ops/s	84.4262 Ops/s	$\textbf{\color{#35bf28}+5.34\%}$
test_full_like	17.8120ms	11.7449ms	85.1431 Ops/s	88.4085 Ops/s	$\color{#d91a1a}-3.69\%$
test_zeros_like	14.8970ms	7.4934ms	133.4510 Ops/s	136.3769 Ops/s	$\color{#d91a1a}-2.15\%$
test_ones_like	9.6305ms	7.7323ms	129.3274 Ops/s	133.8260 Ops/s	$\color{#d91a1a}-3.36\%$
test_clone	12.4458ms	9.1106ms	109.7623 Ops/s	108.5650 Ops/s	$\color{#35bf28}+1.10\%$
test_squeeze	56.6780μs	11.8531μs	84.3664 KOps/s	83.2392 KOps/s	$\color{#35bf28}+1.35\%$
test_unsqueeze	0.1786ms	89.1765μs	11.2137 KOps/s	11.2476 KOps/s	$\color{#d91a1a}-0.30\%$
test_split	0.5000ms	0.1870ms	5.3465 KOps/s	5.1563 KOps/s	$\color{#35bf28}+3.69\%$
test_permute	0.7479ms	0.2235ms	4.4748 KOps/s	4.5954 KOps/s	$\color{#d91a1a}-2.63\%$
test_stack	27.2679ms	23.5010ms	42.5514 Ops/s	40.6041 Ops/s	$\color{#35bf28}+4.80\%$
test_cat	25.8057ms	23.4635ms	42.6193 Ops/s	40.7704 Ops/s	$\color{#35bf28}+4.53\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}23$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	34.6300μs	10.8122μs	92.4877 KOps/s	92.5624 KOps/s	$\color{#d91a1a}-0.08\%$
test_plain_set_stack_nested	34.8400μs	10.8999μs	91.7439 KOps/s	91.9987 KOps/s	$\color{#d91a1a}-0.28\%$
test_plain_set_nested_inplace	41.9300μs	11.7965μs	84.7708 KOps/s	85.1144 KOps/s	$\color{#d91a1a}-0.40\%$
test_plain_set_stack_nested_inplace	35.9410μs	11.7547μs	85.0724 KOps/s	85.3195 KOps/s	$\color{#d91a1a}-0.29\%$
test_items	23.5910μs	2.9018μs	344.6192 KOps/s	341.0243 KOps/s	$\color{#35bf28}+1.05\%$
test_items_nested	0.3740ms	0.3200ms	3.1251 KOps/s	3.0954 KOps/s	$\color{#35bf28}+0.96\%$
test_items_nested_locked	0.3669ms	0.3212ms	3.1136 KOps/s	3.0740 KOps/s	$\color{#35bf28}+1.29\%$
test_items_nested_leaf	94.4110μs	59.1904μs	16.8946 KOps/s	16.8243 KOps/s	$\color{#35bf28}+0.42\%$
test_items_stack_nested	0.3564ms	0.3192ms	3.1325 KOps/s	3.1124 KOps/s	$\color{#35bf28}+0.64\%$
test_items_stack_nested_leaf	88.4410μs	60.0535μs	16.6518 KOps/s	16.8351 KOps/s	$\color{#d91a1a}-1.09\%$
test_items_stack_nested_locked	0.3505ms	0.3217ms	3.1090 KOps/s	3.0907 KOps/s	$\color{#35bf28}+0.59\%$
test_keys	27.1310μs	3.4818μs	287.2097 KOps/s	288.9295 KOps/s	$\color{#d91a1a}-0.60\%$
test_keys_nested	0.1030ms	71.7735μs	13.9327 KOps/s	13.9113 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_nested_locked	0.8214ms	76.8135μs	13.0185 KOps/s	12.8403 KOps/s	$\color{#35bf28}+1.39\%$
test_keys_nested_leaf	0.1022ms	63.1604μs	15.8327 KOps/s	15.7897 KOps/s	$\color{#35bf28}+0.27\%$
test_keys_stack_nested	0.1079ms	72.3021μs	13.8309 KOps/s	13.9348 KOps/s	$\color{#d91a1a}-0.75\%$
test_keys_stack_nested_leaf	0.1006ms	63.3790μs	15.7781 KOps/s	15.8189 KOps/s	$\color{#d91a1a}-0.26\%$
test_keys_stack_nested_locked	0.1132ms	77.5082μs	12.9019 KOps/s	12.9138 KOps/s	$\color{#d91a1a}-0.09\%$
test_values	5.5502μs	0.8458μs	1.1823 MOps/s	1.1826 MOps/s	$\color{#d91a1a}-0.03\%$
test_values_nested	59.4400μs	31.6212μs	31.6243 KOps/s	31.3617 KOps/s	$\color{#35bf28}+0.84\%$
test_values_nested_locked	54.8500μs	32.7165μs	30.5656 KOps/s	30.5257 KOps/s	$\color{#35bf28}+0.13\%$
test_values_nested_leaf	61.9510μs	34.1449μs	29.2869 KOps/s	29.1836 KOps/s	$\color{#35bf28}+0.35\%$
test_values_stack_nested	59.9310μs	32.2501μs	31.0077 KOps/s	31.1229 KOps/s	$\color{#d91a1a}-0.37\%$
test_values_stack_nested_leaf	57.1410μs	34.6093μs	28.8940 KOps/s	29.1327 KOps/s	$\color{#d91a1a}-0.82\%$
test_values_stack_nested_locked	60.5210μs	32.9488μs	30.3501 KOps/s	30.2173 KOps/s	$\color{#35bf28}+0.44\%$
test_membership	2.2190μs	0.5307μs	1.8842 MOps/s	1.8799 MOps/s	$\color{#35bf28}+0.23\%$
test_membership_nested	22.6905μs	1.9072μs	524.3300 KOps/s	521.9835 KOps/s	$\color{#35bf28}+0.45\%$
test_membership_nested_leaf	15.4250μs	1.9292μs	518.3443 KOps/s	519.5267 KOps/s	$\color{#d91a1a}-0.23\%$
test_membership_stacked_nested	24.7510μs	1.9546μs	511.6089 KOps/s	496.9881 KOps/s	$\color{#35bf28}+2.94\%$
test_membership_stacked_nested_leaf	34.5810μs	1.9503μs	512.7419 KOps/s	499.8167 KOps/s	$\color{#35bf28}+2.59\%$
test_membership_nested_last	26.8500μs	2.8612μs	349.5076 KOps/s	348.1006 KOps/s	$\color{#35bf28}+0.40\%$
test_membership_nested_leaf_last	31.8510μs	2.9065μs	344.0592 KOps/s	349.9412 KOps/s	$\color{#d91a1a}-1.68\%$
test_membership_stacked_nested_last	33.4810μs	5.8603μs	170.6384 KOps/s	343.8574 KOps/s	$\textbf{\color{#d91a1a}-50.38\%}$
test_membership_stacked_nested_leaf_last	26.4410μs	5.8145μs	171.9824 KOps/s	348.3986 KOps/s	$\textbf{\color{#d91a1a}-50.64\%}$
test_nested_getleaf	36.0910μs	6.3228μs	158.1590 KOps/s	157.3946 KOps/s	$\color{#35bf28}+0.49\%$
test_nested_get	27.0410μs	6.0503μs	165.2799 KOps/s	164.1901 KOps/s	$\color{#35bf28}+0.66\%$
test_stacked_getleaf	32.0410μs	6.4036μs	156.1626 KOps/s	157.0537 KOps/s	$\color{#d91a1a}-0.57\%$
test_stacked_get	31.0300μs	6.0323μs	165.7730 KOps/s	165.7925 KOps/s	$\color{#d91a1a}-0.01\%$
test_nested_getitemleaf	26.6410μs	6.5056μs	153.7141 KOps/s	154.8499 KOps/s	$\color{#d91a1a}-0.73\%$
test_nested_getitem	36.2510μs	6.1113μs	163.6325 KOps/s	162.7739 KOps/s	$\color{#35bf28}+0.53\%$
test_stacked_getitemleaf	31.7810μs	6.4678μs	154.6124 KOps/s	153.1855 KOps/s	$\color{#35bf28}+0.93\%$
test_stacked_getitem	26.9810μs	6.1527μs	162.5312 KOps/s	163.2015 KOps/s	$\color{#d91a1a}-0.41\%$
test_lock_nested	0.8147ms	0.3728ms	2.6826 KOps/s	2.6183 KOps/s	$\color{#35bf28}+2.45\%$
test_lock_stack_nested	0.4229ms	0.3438ms	2.9086 KOps/s	2.8971 KOps/s	$\color{#35bf28}+0.39\%$
test_unlock_nested	0.6296ms	0.3074ms	3.2527 KOps/s	3.2436 KOps/s	$\color{#35bf28}+0.28\%$
test_unlock_stack_nested	0.3093ms	0.2746ms	3.6420 KOps/s	3.6024 KOps/s	$\color{#35bf28}+1.10\%$
test_flatten_speed	0.1205ms	74.0250μs	13.5090 KOps/s	13.5285 KOps/s	$\color{#d91a1a}-0.14\%$
test_unflatten_speed	0.3306ms	0.2982ms	3.3533 KOps/s	3.3020 KOps/s	$\color{#35bf28}+1.55\%$
test_common_ops	1.5759ms	0.5823ms	1.7173 KOps/s	1.7002 KOps/s	$\color{#35bf28}+1.01\%$
test_creation	0.1878ms	1.5047μs	664.6024 KOps/s	673.3490 KOps/s	$\color{#d91a1a}-1.30\%$
test_creation_empty	33.0310μs	7.4053μs	135.0381 KOps/s	137.4276 KOps/s	$\color{#d91a1a}-1.74\%$
test_creation_nested_1	41.2810μs	8.9882μs	111.2573 KOps/s	111.7356 KOps/s	$\color{#d91a1a}-0.43\%$
test_creation_nested_2	44.0100μs	11.3955μs	87.7542 KOps/s	88.2602 KOps/s	$\color{#d91a1a}-0.57\%$
test_clone	0.1335ms	10.3615μs	96.5113 KOps/s	90.7138 KOps/s	$\textbf{\color{#35bf28}+6.39\%}$
test_getitem[int]	1.6779ms	10.4611μs	95.5920 KOps/s	90.7461 KOps/s	$\textbf{\color{#35bf28}+5.34\%}$
test_getitem[slice_int]	0.1172ms	20.3675μs	49.0978 KOps/s	46.5288 KOps/s	$\textbf{\color{#35bf28}+5.52\%}$
test_getitem[range]	0.1492ms	38.2396μs	26.1509 KOps/s	25.5049 KOps/s	$\color{#35bf28}+2.53\%$
test_getitem[tuple]	0.1141ms	17.9595μs	55.6808 KOps/s	54.0315 KOps/s	$\color{#35bf28}+3.05\%$
test_getitem[list]	0.1432ms	33.5663μs	29.7918 KOps/s	29.3804 KOps/s	$\color{#35bf28}+1.40\%$
test_setitem_dim[int]	53.8610μs	18.4730μs	54.1331 KOps/s	50.3596 KOps/s	$\textbf{\color{#35bf28}+7.49\%}$
test_setitem_dim[slice_int]	58.5910μs	37.6769μs	26.5415 KOps/s	25.8187 KOps/s	$\color{#35bf28}+2.80\%$
test_setitem_dim[range]	87.8210μs	53.9517μs	18.5351 KOps/s	18.6205 KOps/s	$\color{#d91a1a}-0.46\%$
test_setitem_dim[tuple]	57.4210μs	31.6267μs	31.6188 KOps/s	30.6865 KOps/s	$\color{#35bf28}+3.04\%$
test_setitem	0.1349ms	14.4711μs	69.1033 KOps/s	65.1007 KOps/s	$\textbf{\color{#35bf28}+6.15\%}$
test_set	0.1550ms	14.1215μs	70.8141 KOps/s	67.4960 KOps/s	$\color{#35bf28}+4.92\%$
test_set_shared	1.6281ms	0.1479ms	6.7613 KOps/s	6.7773 KOps/s	$\color{#d91a1a}-0.24\%$
test_update	0.4457ms	16.5647μs	60.3694 KOps/s	57.6184 KOps/s	$\color{#35bf28}+4.77\%$
test_update_nested	0.1280ms	21.4207μs	46.6838 KOps/s	45.7586 KOps/s	$\color{#35bf28}+2.02\%$
test_update__nested	1.2636ms	24.8657μs	40.2161 KOps/s	39.6401 KOps/s	$\color{#35bf28}+1.45\%$
test_set_nested	0.1302ms	15.3715μs	65.0553 KOps/s	62.5005 KOps/s	$\color{#35bf28}+4.09\%$
test_set_nested_new	0.1249ms	17.6343μs	56.7078 KOps/s	53.7903 KOps/s	$\textbf{\color{#35bf28}+5.42\%}$
test_select	0.1331ms	29.0222μs	34.4564 KOps/s	33.2391 KOps/s	$\color{#35bf28}+3.66\%$
test_select_nested	70.6510μs	42.7005μs	23.4189 KOps/s	22.9885 KOps/s	$\color{#35bf28}+1.87\%$
test_exclude_nested	88.5110μs	59.7424μs	16.7385 KOps/s	16.7147 KOps/s	$\color{#35bf28}+0.14\%$
test_empty[True]	0.2959ms	0.2591ms	3.8600 KOps/s	3.8179 KOps/s	$\color{#35bf28}+1.10\%$
test_empty[False]	3.1651μs	0.7443μs	1.3435 MOps/s	1.3617 MOps/s	$\color{#d91a1a}-1.34\%$
test_to	84.6310μs	56.3023μs	17.7613 KOps/s	18.0894 KOps/s	$\color{#d91a1a}-1.81\%$
test_to_nonblocking	1.0336ms	46.0479μs	21.7165 KOps/s	21.3551 KOps/s	$\color{#35bf28}+1.69\%$
test_unbind_speed	0.2870ms	0.2329ms	4.2944 KOps/s	4.2045 KOps/s	$\color{#35bf28}+2.14\%$
test_unbind_speed_stack0	0.2837ms	0.2293ms	4.3615 KOps/s	4.2537 KOps/s	$\color{#35bf28}+2.54\%$
test_unbind_speed_stack1	95.8547ms	0.6507ms	1.5369 KOps/s	1.5115 KOps/s	$\color{#35bf28}+1.68\%$
test_split	97.8094ms	1.5711ms	636.4985 Ops/s	613.3760 Ops/s	$\color{#35bf28}+3.77\%$
test_chunk	99.9002ms	1.7080ms	585.4820 Ops/s	561.5600 Ops/s	$\color{#35bf28}+4.26\%$
test_consolidate[False-None]	3.1838ms	2.6416ms	378.5603 Ops/s	385.0178 Ops/s	$\color{#d91a1a}-1.68\%$
test_consolidate[default-None]	1.7295ms	1.6772ms	596.2204 Ops/s	592.9537 Ops/s	$\color{#35bf28}+0.55\%$
test_consolidate[reduce-overhead-None]	1.7909ms	1.7132ms	583.6981 Ops/s	580.6406 Ops/s	$\color{#35bf28}+0.53\%$
test_consolidate_njt[False-None]	6.9814ms	6.6824ms	149.6476 Ops/s	148.7582 Ops/s	$\color{#35bf28}+0.60\%$
test_to[False-False-None]	1.7947ms	1.6760ms	596.6628 Ops/s	607.8715 Ops/s	$\color{#d91a1a}-1.84\%$
test_to[True-False-None]	1.4977ms	1.2654ms	790.2646 Ops/s	758.3347 Ops/s	$\color{#35bf28}+4.21\%$
test_to[within-False-None]	4.1305ms	4.0101ms	249.3676 Ops/s	179.2276 Ops/s	$\textbf{\color{#35bf28}+39.13\%}$
test_to[True-default-None]	5.4057ms	5.1420ms	194.4758 Ops/s	183.7736 Ops/s	$\textbf{\color{#35bf28}+5.82\%}$
test_to_njt[False-False-None]	7.1189ms	7.0431ms	141.9834 Ops/s	137.0739 Ops/s	$\color{#35bf28}+3.58\%$
test_to_njt[True-False-None]	6.0424ms	5.5869ms	178.9917 Ops/s	173.2458 Ops/s	$\color{#35bf28}+3.32\%$
test_to_njt[within-False-None]	12.4466ms	12.3602ms	80.9051 Ops/s	77.2759 Ops/s	$\color{#35bf28}+4.70\%$
test_creation[device0]	0.5366ms	79.8593μs	12.5220 KOps/s	11.8364 KOps/s	$\textbf{\color{#35bf28}+5.79\%}$
test_creation_from_tensor	0.5243ms	83.0774μs	12.0370 KOps/s	11.5125 KOps/s	$\color{#35bf28}+4.56\%$
test_add_one[memmap_tensor0]	0.3906ms	6.7304μs	148.5797 KOps/s	138.1429 KOps/s	$\textbf{\color{#35bf28}+7.56\%}$
test_contiguous[memmap_tensor0]	2.4481μs	0.4297μs	2.3271 MOps/s	2.4378 MOps/s	$\color{#d91a1a}-4.54\%$
test_stack[memmap_tensor0]	38.6500μs	4.3631μs	229.1966 KOps/s	214.8246 KOps/s	$\textbf{\color{#35bf28}+6.69\%}$
test_memmaptd_index	1.7622ms	0.2496ms	4.0061 KOps/s	3.9161 KOps/s	$\color{#35bf28}+2.30\%$
test_memmaptd_index_astensor	0.6016ms	0.3104ms	3.2217 KOps/s	3.2011 KOps/s	$\color{#35bf28}+0.64\%$
test_memmaptd_index_op	0.3049s	0.7517ms	1.3304 KOps/s	1.6556 KOps/s	$\textbf{\color{#d91a1a}-19.64\%}$
test_serialize_model	0.1322s	0.1315s	7.6049 Ops/s	7.6315 Ops/s	$\color{#d91a1a}-0.35\%$
test_serialize_model_pickle	1.3779s	1.1921s	0.8388 Ops/s	0.8253 Ops/s	$\color{#35bf28}+1.64\%$
test_serialize_weights	0.1308s	0.1297s	7.7074 Ops/s	7.7061 Ops/s	$\color{#35bf28}+0.02\%$
test_serialize_weights_returnearly	0.3096s	53.3619ms	18.7400 Ops/s	15.0601 Ops/s	$\textbf{\color{#35bf28}+24.43\%}$
test_serialize_weights_pickle	1.3763s	1.1906s	0.8399 Ops/s	0.8439 Ops/s	$\color{#d91a1a}-0.47\%$
test_reshape_pytree	55.9710μs	23.0459μs	43.3916 KOps/s	43.5660 KOps/s	$\color{#d91a1a}-0.40\%$
test_reshape_td	57.5110μs	27.2235μs	36.7330 KOps/s	36.5968 KOps/s	$\color{#35bf28}+0.37\%$
test_view_pytree	56.7100μs	23.1538μs	43.1895 KOps/s	43.7558 KOps/s	$\color{#d91a1a}-1.29\%$
test_view_td	65.1610μs	30.4041μs	32.8903 KOps/s	33.4658 KOps/s	$\color{#d91a1a}-1.72\%$
test_unbind_pytree	61.2710μs	29.1559μs	34.2984 KOps/s	34.8368 KOps/s	$\color{#d91a1a}-1.55\%$
test_unbind_td	0.7422ms	37.2165μs	26.8698 KOps/s	27.5057 KOps/s	$\color{#d91a1a}-2.31\%$
test_split_pytree	65.3410μs	30.7847μs	32.4836 KOps/s	32.2996 KOps/s	$\color{#35bf28}+0.57\%$
test_split_td	0.9543ms	40.8518μs	24.4787 KOps/s	24.5576 KOps/s	$\color{#d91a1a}-0.32\%$
test_add_pytree	72.2610μs	34.8921μs	28.6598 KOps/s	27.5463 KOps/s	$\color{#35bf28}+4.04\%$
test_add_td	97.1620μs	48.0080μs	20.8299 KOps/s	20.5762 KOps/s	$\color{#35bf28}+1.23\%$
test_compile_add_one_nested[tensordict-compile]	0.1735ms	0.1233ms	8.1117 KOps/s	7.9478 KOps/s	$\color{#35bf28}+2.06\%$
test_compile_add_one_nested[tensordict-eager]	0.2201ms	0.1266ms	7.9014 KOps/s	7.7347 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_add_one_nested[pytree-compile]	0.1395ms	97.9494μs	10.2094 KOps/s	9.9313 KOps/s	$\color{#35bf28}+2.80\%$
test_compile_add_one_nested[pytree-eager]	1.2886ms	0.1501ms	6.6608 KOps/s	6.6369 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_copy_nested[tensordict-compile]	0.1621ms	29.7963μs	33.5612 KOps/s	42.3799 KOps/s	$\textbf{\color{#d91a1a}-20.81\%}$
test_compile_copy_nested[tensordict-eager]	65.6810μs	27.5494μs	36.2984 KOps/s	36.7671 KOps/s	$\color{#d91a1a}-1.27\%$
test_compile_copy_nested[pytree-compile]	0.2938ms	66.0390μs	15.1426 KOps/s	15.1893 KOps/s	$\color{#d91a1a}-0.31\%$
test_compile_copy_nested[pytree-eager]	88.7610μs	51.5865μs	19.3849 KOps/s	19.6963 KOps/s	$\color{#d91a1a}-1.58\%$
test_compile_add_one_flat[tensordict-compile]	0.1897ms	0.1440ms	6.9456 KOps/s	7.0135 KOps/s	$\color{#d91a1a}-0.97\%$
test_compile_add_one_flat[tensordict-eager]	0.3009ms	0.2106ms	4.7490 KOps/s	4.8261 KOps/s	$\color{#d91a1a}-1.60\%$
test_compile_add_one_flat[tensorclass-compile]	0.1358ms	99.5475μs	10.0455 KOps/s	10.1312 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_add_one_flat[tensorclass-eager]	0.1087ms	51.6660μs	19.3551 KOps/s	18.8968 KOps/s	$\color{#35bf28}+2.43\%$
test_compile_add_one_flat[pytree-compile]	0.1861ms	0.1381ms	7.2415 KOps/s	7.2877 KOps/s	$\color{#d91a1a}-0.63\%$
test_compile_add_one_flat[pytree-eager]	0.5304ms	0.4788ms	2.0886 KOps/s	2.0569 KOps/s	$\color{#35bf28}+1.54\%$
test_compile_add_self_flat[tensordict-eager]	0.3684ms	0.2511ms	3.9826 KOps/s	4.0108 KOps/s	$\color{#d91a1a}-0.70\%$
test_compile_add_self_flat[tensordict-compile]	0.2066ms	0.1448ms	6.9071 KOps/s	6.9432 KOps/s	$\color{#d91a1a}-0.52\%$
test_compile_add_self_flat[tensorclass-eager]	0.1418ms	62.2211μs	16.0717 KOps/s	15.1530 KOps/s	$\textbf{\color{#35bf28}+6.06\%}$
test_compile_add_self_flat[tensorclass-compile]	0.1417ms	99.8203μs	10.0180 KOps/s	10.1239 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_add_self_flat[pytree-eager]	0.4492ms	0.4035ms	2.4780 KOps/s	2.4285 KOps/s	$\color{#35bf28}+2.04\%$
test_compile_add_self_flat[pytree-compile]	0.1861ms	0.1348ms	7.4180 KOps/s	7.2575 KOps/s	$\color{#35bf28}+2.21\%$
test_compile_copy_flat[tensordict-compile]	50.2000μs	19.2531μs	51.9398 KOps/s	51.3543 KOps/s	$\color{#35bf28}+1.14\%$
test_compile_copy_flat[tensordict-eager]	66.8310μs	28.0226μs	35.6855 KOps/s	37.4314 KOps/s	$\color{#d91a1a}-4.66\%$
test_compile_copy_flat[pytree-compile]	0.1033ms	71.2474μs	14.0356 KOps/s	14.1827 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_copy_flat[pytree-eager]	80.7110μs	52.5279μs	19.0375 KOps/s	19.4100 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_assign_and_add[tensordict-compile]	1.6504ms	0.3978ms	2.5141 KOps/s	2.2112 KOps/s	$\textbf{\color{#35bf28}+13.70\%}$
test_compile_assign_and_add[tensordict-eager]	2.7571ms	2.6006ms	384.5334 Ops/s	374.0492 Ops/s	$\color{#35bf28}+2.80\%$
test_compile_assign_and_add[pytree-compile]	1.6179ms	0.4391ms	2.2775 KOps/s	2.2486 KOps/s	$\color{#35bf28}+1.29\%$
test_compile_assign_and_add[pytree-eager]	2.8856ms	2.6485ms	377.5705 Ops/s	382.9704 Ops/s	$\color{#d91a1a}-1.41\%$
test_compile_indexing[tensor-tensordict-compile]	0.1828ms	0.1183ms	8.4548 KOps/s	8.6960 KOps/s	$\color{#d91a1a}-2.77\%$
test_compile_indexing[tensor-tensordict-eager]	0.5646ms	83.1746μs	12.0229 KOps/s	12.4444 KOps/s	$\color{#d91a1a}-3.39\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1725ms	0.1071ms	9.3371 KOps/s	9.4070 KOps/s	$\color{#d91a1a}-0.74\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1082ms	67.2436μs	14.8713 KOps/s	14.2975 KOps/s	$\color{#35bf28}+4.01\%$
test_compile_indexing[tensor-pytree-compile]	0.1664ms	0.1119ms	8.9352 KOps/s	9.3025 KOps/s	$\color{#d91a1a}-3.95\%$
test_compile_indexing[tensor-pytree-eager]	0.1264ms	70.4240μs	14.1997 KOps/s	14.5207 KOps/s	$\color{#d91a1a}-2.21\%$
test_compile_indexing[slice-tensordict-compile]	0.1478ms	0.1033ms	9.6787 KOps/s	9.4940 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_indexing[slice-tensordict-eager]	0.1483ms	16.9867μs	58.8698 KOps/s	55.1549 KOps/s	$\textbf{\color{#35bf28}+6.74\%}$
test_compile_indexing[slice-tensorclass-compile]	0.2073ms	97.5830μs	10.2477 KOps/s	10.2094 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_indexing[slice-tensorclass-eager]	52.4700μs	16.2219μs	61.6450 KOps/s	60.1562 KOps/s	$\color{#35bf28}+2.47\%$
test_compile_indexing[slice-pytree-compile]	0.1497ms	98.3406μs	10.1687 KOps/s	10.1471 KOps/s	$\color{#35bf28}+0.21\%$
test_compile_indexing[slice-pytree-eager]	51.9610μs	16.0168μs	62.4344 KOps/s	60.3092 KOps/s	$\color{#35bf28}+3.52\%$
test_compile_indexing[int-tensordict-compile]	0.1471ms	0.1064ms	9.3948 KOps/s	9.7294 KOps/s	$\color{#d91a1a}-3.44\%$
test_compile_indexing[int-tensordict-eager]	0.5986ms	16.7641μs	59.6512 KOps/s	55.3629 KOps/s	$\textbf{\color{#35bf28}+7.75\%}$
test_compile_indexing[int-tensorclass-compile]	0.1543ms	0.1031ms	9.7001 KOps/s	10.1662 KOps/s	$\color{#d91a1a}-4.58\%$
test_compile_indexing[int-tensorclass-eager]	47.8210μs	16.1230μs	62.0233 KOps/s	60.3895 KOps/s	$\color{#35bf28}+2.71\%$
test_compile_indexing[int-pytree-compile]	0.1410ms	97.6983μs	10.2356 KOps/s	10.1645 KOps/s	$\color{#35bf28}+0.70\%$
test_compile_indexing[int-pytree-eager]	57.4110μs	16.0389μs	62.3486 KOps/s	60.8096 KOps/s	$\color{#35bf28}+2.53\%$
test_mod_add[eager]	71.0510μs	31.5734μs	31.6722 KOps/s	31.4615 KOps/s	$\color{#35bf28}+0.67\%$
test_mod_add[compile]	0.1253ms	79.4239μs	12.5907 KOps/s	12.6180 KOps/s	$\color{#d91a1a}-0.22\%$
test_mod_add[compile-overhead]	0.3349ms	0.1710ms	5.8479 KOps/s	5.7436 KOps/s	$\color{#35bf28}+1.82\%$
test_mod_wrap[eager]	0.3225ms	0.2412ms	4.1456 KOps/s	4.0749 KOps/s	$\color{#35bf28}+1.73\%$
test_mod_wrap[compile]	1.6065ms	0.2862ms	3.4939 KOps/s	3.4010 KOps/s	$\color{#35bf28}+2.73\%$
test_mod_wrap[compile-overhead]	6.9801ms	3.7171ms	269.0247 Ops/s	267.6315 Ops/s	$\color{#35bf28}+0.52\%$
test_mod_wrap_and_backward[eager]	1.6607ms	1.4773ms	676.8950 Ops/s	703.2336 Ops/s	$\color{#d91a1a}-3.75\%$
test_mod_wrap_and_backward[compile]	1.3981ms	1.2744ms	784.7023 Ops/s	725.8378 Ops/s	$\textbf{\color{#35bf28}+8.11\%}$
test_mod_wrap_and_backward[compile-overhead]	1.4062ms	0.9319ms	1.0731 KOps/s	964.5683 Ops/s	$\textbf{\color{#35bf28}+11.25\%}$
test_seq_add[eager]	0.2057ms	96.4828μs	10.3645 KOps/s	9.5215 KOps/s	$\textbf{\color{#35bf28}+8.85\%}$
test_seq_add[compile]	0.1571ms	88.3307μs	11.3211 KOps/s	11.0874 KOps/s	$\color{#35bf28}+2.11\%$
test_seq_add[compile-overhead]	0.1954ms	0.1363ms	7.3376 KOps/s	7.7300 KOps/s	$\textbf{\color{#d91a1a}-5.08\%}$
test_seq_wrap[eager]	0.5504ms	0.3799ms	2.6323 KOps/s	2.4561 KOps/s	$\textbf{\color{#35bf28}+7.17\%}$
test_seq_wrap[compile]	0.3487ms	0.3007ms	3.3256 KOps/s	3.2602 KOps/s	$\color{#35bf28}+2.01\%$
test_seq_wrap[compile-overhead]	0.3155ms	0.2334ms	4.2841 KOps/s	4.4278 KOps/s	$\color{#d91a1a}-3.24\%$
test_func_call_runtime[False-eager]	0.9107ms	0.7838ms	1.2758 KOps/s	1.2540 KOps/s	$\color{#35bf28}+1.74\%$
test_func_call_runtime[False-compile]	0.9991ms	0.7827ms	1.2777 KOps/s	1.3206 KOps/s	$\color{#d91a1a}-3.25\%$
test_func_call_runtime[False-compile-overhead]	0.4130ms	0.3676ms	2.7203 KOps/s	2.7255 KOps/s	$\color{#d91a1a}-0.19\%$
test_func_call_runtime[True-eager]	1.0352ms	0.8997ms	1.1114 KOps/s	1.1059 KOps/s	$\color{#35bf28}+0.50\%$
test_func_call_runtime[True-compile]	0.8413ms	0.7656ms	1.3062 KOps/s	1.2952 KOps/s	$\color{#35bf28}+0.85\%$
test_func_call_runtime[True-compile-overhead]	0.4353ms	0.3898ms	2.5656 KOps/s	2.5879 KOps/s	$\color{#d91a1a}-0.86\%$
test_func_call_cm_runtime[False-eager]	0.8118ms	0.7368ms	1.3573 KOps/s	1.3002 KOps/s	$\color{#35bf28}+4.39\%$
test_func_call_cm_runtime[False-compile]	1.1188ms	0.7608ms	1.3143 KOps/s	1.3175 KOps/s	$\color{#d91a1a}-0.24\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4268ms	0.3702ms	2.7011 KOps/s	2.7064 KOps/s	$\color{#d91a1a}-0.20\%$
test_func_call_cm_runtime[True-eager]	1.0997ms	0.9951ms	1.0049 KOps/s	986.5199 Ops/s	$\color{#35bf28}+1.87\%$
test_func_call_cm_runtime[True-compile]	0.8862ms	0.7961ms	1.2561 KOps/s	1.2324 KOps/s	$\color{#35bf28}+1.92\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4657ms	0.4171ms	2.3974 KOps/s	2.4077 KOps/s	$\color{#d91a1a}-0.42\%$
test_vmap_func_call_cm_runtime[eager]	2.5906ms	2.0842ms	479.8072 Ops/s	477.8236 Ops/s	$\color{#35bf28}+0.42\%$
test_vmap_func_call_cm_runtime[compile]	0.8809ms	0.8084ms	1.2370 KOps/s	1.1899 KOps/s	$\color{#35bf28}+3.95\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4568ms	0.4153ms	2.4080 KOps/s	2.4096 KOps/s	$\color{#d91a1a}-0.06\%$
test_distributed	2.8720ms	0.2227ms	4.4910 KOps/s	8.8720 KOps/s	$\textbf{\color{#d91a1a}-49.38\%}$
test_tdmodule	26.7300μs	14.1270μs	70.7865 KOps/s	71.2420 KOps/s	$\color{#d91a1a}-0.64\%$
test_tdmodule_dispatch	0.3043ms	28.6272μs	34.9318 KOps/s	37.1927 KOps/s	$\textbf{\color{#d91a1a}-6.08\%}$
test_tdseq	33.3800μs	15.3617μs	65.0971 KOps/s	65.2183 KOps/s	$\color{#d91a1a}-0.19\%$
test_tdseq_dispatch	50.5610μs	31.4624μs	31.7839 KOps/s	33.3841 KOps/s	$\color{#d91a1a}-4.79\%$
test_instantiation_functorch	1.6509ms	1.5730ms	635.7269 Ops/s	635.9201 Ops/s	$\color{#d91a1a}-0.03\%$
test_exec_functorch	0.1825ms	0.1463ms	6.8368 KOps/s	6.5779 KOps/s	$\color{#35bf28}+3.94\%$
test_exec_functional_call	0.1890ms	0.1409ms	7.0992 KOps/s	6.8698 KOps/s	$\color{#35bf28}+3.34\%$
test_exec_td_decorator	0.3757ms	0.1862ms	5.3714 KOps/s	5.2355 KOps/s	$\color{#35bf28}+2.60\%$
test_vmap_mlp_speed_decorator[True-True]	0.7974ms	0.6742ms	1.4831 KOps/s	1.4828 KOps/s	$\color{#35bf28}+0.02\%$
test_vmap_mlp_speed_decorator[True-False]	0.8726ms	0.6725ms	1.4871 KOps/s	1.4836 KOps/s	$\color{#35bf28}+0.23\%$
test_vmap_mlp_speed_decorator[False-True]	0.6976ms	0.5911ms	1.6919 KOps/s	1.6790 KOps/s	$\color{#35bf28}+0.77\%$
test_vmap_mlp_speed_decorator[False-False]	0.6998ms	0.5936ms	1.6846 KOps/s	1.6783 KOps/s	$\color{#35bf28}+0.37\%$
test_vmap_transformer_speed_decorator[True-True]	19.1964ms	19.1202ms	52.3006 Ops/s	52.2491 Ops/s	$\color{#35bf28}+0.10\%$
test_vmap_transformer_speed_decorator[True-False]	20.8021ms	19.2066ms	52.0653 Ops/s	52.1819 Ops/s	$\color{#d91a1a}-0.22\%$
test_vmap_transformer_speed_decorator[False-True]	19.1409ms	19.0703ms	52.4377 Ops/s	52.3623 Ops/s	$\color{#35bf28}+0.14\%$
test_vmap_transformer_speed_decorator[False-False]	19.1013ms	19.0176ms	52.5830 Ops/s	52.0649 Ops/s	$\color{#35bf28}+0.99\%$
test_to_module_speed[True]	1.0553ms	0.9448ms	1.0584 KOps/s	1.0560 KOps/s	$\color{#35bf28}+0.23\%$
test_to_module_speed[False]	1.3755ms	0.9399ms	1.0640 KOps/s	1.0791 KOps/s	$\color{#d91a1a}-1.40\%$
test_tc_init	61.4610μs	33.5841μs	29.7760 KOps/s	27.8993 KOps/s	$\textbf{\color{#35bf28}+6.73\%}$
test_tc_init_nested	0.1036ms	71.8143μs	13.9248 KOps/s	13.8094 KOps/s	$\color{#35bf28}+0.84\%$
test_tc_first_layer_tensor	10.5316μs	0.7515μs	1.3306 MOps/s	1.3338 MOps/s	$\color{#d91a1a}-0.24\%$
test_tc_first_layer_nontensor	24.6210μs	2.4136μs	414.3217 KOps/s	407.7908 KOps/s	$\color{#35bf28}+1.60\%$
test_tc_second_layer_tensor	11.6133μs	1.5479μs	646.0347 KOps/s	653.3888 KOps/s	$\color{#d91a1a}-1.13\%$
test_tc_second_layer_nontensor	0.1317ms	3.2176μs	310.7884 KOps/s	310.7626 KOps/s	$+0.01\%$
test_unbind	0.2252s	9.8167ms	101.8669 Ops/s	98.3818 Ops/s	$\color{#35bf28}+3.54\%$
test_full_like	10.1271ms	9.4765ms	105.5242 Ops/s	103.4456 Ops/s	$\color{#35bf28}+2.01\%$
test_zeros_like	9.5422ms	7.2916ms	137.1433 Ops/s	136.6968 Ops/s	$\color{#35bf28}+0.33\%$
test_ones_like	4.9472ms	4.1746ms	239.5428 Ops/s	227.4156 Ops/s	$\textbf{\color{#35bf28}+5.33\%}$
test_clone	7.7082ms	6.5736ms	152.1230 Ops/s	144.1428 Ops/s	$\textbf{\color{#35bf28}+5.54\%}$
test_squeeze	59.5300μs	9.6850μs	103.2526 KOps/s	103.2331 KOps/s	$\color{#35bf28}+0.02\%$
test_unsqueeze	0.1191ms	71.1493μs	14.0549 KOps/s	14.0219 KOps/s	$\color{#35bf28}+0.24\%$
test_split	0.4173ms	0.1553ms	6.4409 KOps/s	6.2813 KOps/s	$\color{#35bf28}+2.54\%$
test_permute	0.2243ms	0.1749ms	5.7184 KOps/s	5.6075 KOps/s	$\color{#35bf28}+1.98\%$
test_stack	53.7709ms	51.5892ms	19.3839 Ops/s	19.2069 Ops/s	$\color{#35bf28}+0.92\%$
test_cat	52.0024ms	51.2311ms	19.5194 Ops/s	19.3734 Ops/s	$\color{#35bf28}+0.75\%$