[BugFix] Proper auto-batch size for unbatched tensors

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Proper auto-batch size for unbatched tensors #1213

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}4$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	57.1070μs	20.9949μs	47.6306 KOps/s	47.3509 KOps/s	$\color{#35bf28}+0.59\%$
test_plain_set_stack_nested	60.3330μs	21.3570μs	46.8230 KOps/s	47.2402 KOps/s	$\color{#d91a1a}-0.88\%$
test_plain_set_nested_inplace	52.1180μs	23.0887μs	43.3112 KOps/s	43.2040 KOps/s	$\color{#35bf28}+0.25\%$
test_plain_set_stack_nested_inplace	77.2350μs	23.0352μs	43.4119 KOps/s	43.2579 KOps/s	$\color{#35bf28}+0.36\%$
test_items	22.9730μs	4.2081μs	237.6365 KOps/s	239.3590 KOps/s	$\color{#d91a1a}-0.72\%$
test_items_nested	0.5575ms	0.4090ms	2.4450 KOps/s	2.4625 KOps/s	$\color{#d91a1a}-0.71\%$
test_items_nested_locked	0.5758ms	0.4086ms	2.4476 KOps/s	2.4693 KOps/s	$\color{#d91a1a}-0.88\%$
test_items_nested_leaf	0.1386ms	76.5577μs	13.0620 KOps/s	12.7316 KOps/s	$\color{#35bf28}+2.60\%$
test_items_stack_nested	0.5105ms	0.4141ms	2.4149 KOps/s	2.4587 KOps/s	$\color{#d91a1a}-1.78\%$
test_items_stack_nested_leaf	0.1516ms	76.7125μs	13.0357 KOps/s	12.6961 KOps/s	$\color{#35bf28}+2.67\%$
test_items_stack_nested_locked	0.5792ms	0.4117ms	2.4292 KOps/s	2.4623 KOps/s	$\color{#d91a1a}-1.34\%$
test_keys	23.6540μs	3.6379μs	274.8804 KOps/s	283.1849 KOps/s	$\color{#d91a1a}-2.93\%$
test_keys_nested	0.3841ms	0.1665ms	6.0049 KOps/s	6.0751 KOps/s	$\color{#d91a1a}-1.16\%$
test_keys_nested_locked	1.8287ms	0.1716ms	5.8274 KOps/s	5.8874 KOps/s	$\color{#d91a1a}-1.02\%$
test_keys_nested_leaf	0.2346ms	0.1442ms	6.9348 KOps/s	6.9952 KOps/s	$\color{#d91a1a}-0.86\%$
test_keys_stack_nested	0.2639ms	0.1639ms	6.1003 KOps/s	5.9270 KOps/s	$\color{#35bf28}+2.92\%$
test_keys_stack_nested_leaf	0.3663ms	0.1446ms	6.9142 KOps/s	7.0398 KOps/s	$\color{#d91a1a}-1.78\%$
test_keys_stack_nested_locked	0.2956ms	0.1706ms	5.8610 KOps/s	5.8567 KOps/s	$\color{#35bf28}+0.07\%$
test_values	5.8368μs	1.0471μs	955.0585 KOps/s	922.8459 KOps/s	$\color{#35bf28}+3.49\%$
test_values_nested	0.1703ms	64.6425μs	15.4697 KOps/s	16.0257 KOps/s	$\color{#d91a1a}-3.47\%$
test_values_nested_locked	0.1320ms	62.6129μs	15.9711 KOps/s	14.9438 KOps/s	$\textbf{\color{#35bf28}+6.87\%}$
test_values_nested_leaf	0.1665ms	71.1531μs	14.0542 KOps/s	13.9788 KOps/s	$\color{#35bf28}+0.54\%$
test_values_stack_nested	0.1238ms	62.7405μs	15.9387 KOps/s	16.0849 KOps/s	$\color{#d91a1a}-0.91\%$
test_values_stack_nested_leaf	0.1243ms	70.9114μs	14.1021 KOps/s	14.1068 KOps/s	$\color{#d91a1a}-0.03\%$
test_values_stack_nested_locked	0.1173ms	63.0481μs	15.8609 KOps/s	16.1650 KOps/s	$\color{#d91a1a}-1.88\%$
test_membership	2.3865μs	0.7362μs	1.3583 MOps/s	1.3925 MOps/s	$\color{#d91a1a}-2.46\%$
test_membership_nested	30.4670μs	2.9125μs	343.3439 KOps/s	342.2051 KOps/s	$\color{#35bf28}+0.33\%$
test_membership_nested_leaf	38.5090μs	2.9186μs	342.6277 KOps/s	348.2251 KOps/s	$\color{#d91a1a}-1.61\%$
test_membership_stacked_nested	28.9340μs	2.9232μs	342.0875 KOps/s	350.5871 KOps/s	$\color{#d91a1a}-2.42\%$
test_membership_stacked_nested_leaf	27.2710μs	2.9090μs	343.7634 KOps/s	344.2109 KOps/s	$\color{#d91a1a}-0.13\%$
test_membership_nested_last	56.7260μs	4.3027μs	232.4112 KOps/s	234.8880 KOps/s	$\color{#d91a1a}-1.05\%$
test_membership_nested_leaf_last	33.1120μs	4.3829μs	228.1578 KOps/s	234.5166 KOps/s	$\color{#d91a1a}-2.71\%$
test_membership_stacked_nested_last	34.2850μs	4.3258μs	231.1693 KOps/s	231.4955 KOps/s	$\color{#d91a1a}-0.14\%$
test_membership_stacked_nested_leaf_last	23.6840μs	4.2784μs	233.7318 KOps/s	231.0090 KOps/s	$\color{#35bf28}+1.18\%$
test_nested_getleaf	41.6080μs	10.5816μs	94.5040 KOps/s	93.0397 KOps/s	$\color{#35bf28}+1.57\%$
test_nested_get	55.4980μs	9.9369μs	100.6354 KOps/s	97.5632 KOps/s	$\color{#35bf28}+3.15\%$
test_stacked_getleaf	48.4700μs	10.4552μs	95.6464 KOps/s	94.5557 KOps/s	$\color{#35bf28}+1.15\%$
test_stacked_get	41.8590μs	9.7382μs	102.6881 KOps/s	98.6425 KOps/s	$\color{#35bf28}+4.10\%$
test_nested_getitemleaf	41.0070μs	11.2888μs	88.5836 KOps/s	89.2181 KOps/s	$\color{#d91a1a}-0.71\%$
test_nested_getitem	34.4240μs	10.6744μs	93.6822 KOps/s	92.3114 KOps/s	$\color{#35bf28}+1.48\%$
test_stacked_getitemleaf	61.2750μs	10.9861μs	91.0239 KOps/s	87.9566 KOps/s	$\color{#35bf28}+3.49\%$
test_stacked_getitem	36.8790μs	10.6455μs	93.9360 KOps/s	94.4582 KOps/s	$\color{#d91a1a}-0.55\%$
test_lock_nested	4.8462ms	0.4182ms	2.3909 KOps/s	2.4561 KOps/s	$\color{#d91a1a}-2.65\%$
test_lock_stack_nested	0.7249ms	0.4252ms	2.3518 KOps/s	2.3566 KOps/s	$\color{#d91a1a}-0.20\%$
test_unlock_nested	0.4830ms	0.3389ms	2.9510 KOps/s	2.9957 KOps/s	$\color{#d91a1a}-1.49\%$
test_unlock_stack_nested	0.4230ms	0.3453ms	2.8963 KOps/s	2.9095 KOps/s	$\color{#d91a1a}-0.46\%$
test_flatten_speed	0.1915ms	0.1003ms	9.9724 KOps/s	9.8341 KOps/s	$\color{#35bf28}+1.41\%$
test_unflatten_speed	0.6650ms	0.5175ms	1.9323 KOps/s	1.9069 KOps/s	$\color{#35bf28}+1.33\%$
test_common_ops	0.9938ms	0.8425ms	1.1870 KOps/s	1.1856 KOps/s	$\color{#35bf28}+0.12\%$
test_creation	43.4750μs	2.4700μs	404.8501 KOps/s	409.4829 KOps/s	$\color{#d91a1a}-1.13\%$
test_creation_empty	40.8170μs	13.1051μs	76.3062 KOps/s	77.1654 KOps/s	$\color{#d91a1a}-1.11\%$
test_creation_nested_1	49.1920μs	16.0730μs	62.2162 KOps/s	62.6387 KOps/s	$\color{#d91a1a}-0.67\%$
test_creation_nested_2	50.8950μs	20.4705μs	48.8507 KOps/s	48.5126 KOps/s	$\color{#35bf28}+0.70\%$
test_clone	70.9840μs	13.7898μs	72.5173 KOps/s	73.7750 KOps/s	$\color{#d91a1a}-1.70\%$
test_getitem[int]	0.8343ms	13.0429μs	76.6702 KOps/s	78.9796 KOps/s	$\color{#d91a1a}-2.92\%$
test_getitem[slice_int]	0.1421ms	24.6628μs	40.5468 KOps/s	41.2549 KOps/s	$\color{#d91a1a}-1.72\%$
test_getitem[range]	0.1899ms	49.7749μs	20.0905 KOps/s	19.4154 KOps/s	$\color{#35bf28}+3.48\%$
test_getitem[tuple]	0.1194ms	19.9847μs	50.0383 KOps/s	49.4537 KOps/s	$\color{#35bf28}+1.18\%$
test_getitem[list]	0.1538ms	45.4943μs	21.9808 KOps/s	21.7180 KOps/s	$\color{#35bf28}+1.21\%$
test_setitem_dim[int]	61.6460μs	25.4827μs	39.2423 KOps/s	38.3277 KOps/s	$\color{#35bf28}+2.39\%$
test_setitem_dim[slice_int]	0.1259ms	53.8677μs	18.5640 KOps/s	19.0695 KOps/s	$\color{#d91a1a}-2.65\%$
test_setitem_dim[range]	0.1420ms	80.5051μs	12.4216 KOps/s	12.6922 KOps/s	$\color{#d91a1a}-2.13\%$
test_setitem_dim[tuple]	75.1010μs	41.8603μs	23.8890 KOps/s	23.3218 KOps/s	$\color{#35bf28}+2.43\%$
test_setitem	71.2030μs	22.1641μs	45.1179 KOps/s	46.0977 KOps/s	$\color{#d91a1a}-2.13\%$
test_set	72.4060μs	21.4600μs	46.5983 KOps/s	47.3070 KOps/s	$\color{#d91a1a}-1.50\%$
test_set_shared	0.3999ms	0.1837ms	5.4430 KOps/s	5.4130 KOps/s	$\color{#35bf28}+0.55\%$
test_update	0.1166ms	25.3697μs	39.4170 KOps/s	40.4770 KOps/s	$\color{#d91a1a}-2.62\%$
test_update_nested	98.9350μs	35.6823μs	28.0251 KOps/s	28.2475 KOps/s	$\color{#d91a1a}-0.79\%$
test_update__nested	0.4387ms	34.4127μs	29.0591 KOps/s	29.4901 KOps/s	$\color{#d91a1a}-1.46\%$
test_set_nested	78.0360μs	23.6230μs	42.3315 KOps/s	43.5489 KOps/s	$\color{#d91a1a}-2.80\%$
test_set_nested_new	73.8290μs	28.1466μs	35.5283 KOps/s	35.5630 KOps/s	$\color{#d91a1a}-0.10\%$
test_select	0.1077ms	44.4116μs	22.5166 KOps/s	22.4954 KOps/s	$\color{#35bf28}+0.09\%$
test_select_nested	0.1373ms	63.6458μs	15.7120 KOps/s	15.8656 KOps/s	$\color{#d91a1a}-0.97\%$
test_exclude_nested	0.1685ms	81.0528μs	12.3376 KOps/s	12.3626 KOps/s	$\color{#d91a1a}-0.20\%$
test_empty[True]	0.5984ms	0.4117ms	2.4292 KOps/s	2.4616 KOps/s	$\color{#d91a1a}-1.31\%$
test_empty[False]	6.6300μs	1.3860μs	721.5146 KOps/s	726.4998 KOps/s	$\color{#d91a1a}-0.69\%$
test_unbind_speed	0.4627ms	0.2734ms	3.6581 KOps/s	3.7355 KOps/s	$\color{#d91a1a}-2.07\%$
test_unbind_speed_stack0	0.3748ms	0.2693ms	3.7128 KOps/s	3.7492 KOps/s	$\color{#d91a1a}-0.97\%$
test_unbind_speed_stack1	0.1023s	0.7435ms	1.3449 KOps/s	1.2426 KOps/s	$\textbf{\color{#35bf28}+8.24\%}$
test_split	0.1049s	1.7510ms	571.1145 Ops/s	562.8396 Ops/s	$\color{#35bf28}+1.47\%$
test_chunk	0.1134s	1.7863ms	559.8285 Ops/s	617.4815 Ops/s	$\textbf{\color{#d91a1a}-9.34\%}$
test_consolidate_njt[False-None]	8.6840ms	8.1931ms	122.0539 Ops/s	107.4551 Ops/s	$\textbf{\color{#35bf28}+13.59\%}$
test_creation[device0]	3.8754ms	93.9538μs	10.6435 KOps/s	10.5794 KOps/s	$\color{#35bf28}+0.61\%$
test_creation_from_tensor	0.2595ms	94.5022μs	10.5818 KOps/s	10.5076 KOps/s	$\color{#35bf28}+0.71\%$
test_add_one[memmap_tensor0]	83.9670μs	5.2581μs	190.1814 KOps/s	186.5720 KOps/s	$\color{#35bf28}+1.93\%$
test_contiguous[memmap_tensor0]	13.6150μs	0.5071μs	1.9722 MOps/s	1.8946 MOps/s	$\color{#35bf28}+4.10\%$
test_stack[memmap_tensor0]	30.0960μs	3.7983μs	263.2747 KOps/s	270.9095 KOps/s	$\color{#d91a1a}-2.82\%$
test_memmaptd_index	1.1440ms	0.2325ms	4.3011 KOps/s	4.3736 KOps/s	$\color{#d91a1a}-1.66\%$
test_memmaptd_index_astensor	0.4843ms	0.3173ms	3.1517 KOps/s	3.1679 KOps/s	$\color{#d91a1a}-0.51\%$
test_memmaptd_index_op	0.8870ms	0.6243ms	1.6018 KOps/s	1.5940 KOps/s	$\color{#35bf28}+0.49\%$
test_serialize_model	0.2210s	0.1329s	7.5232 Ops/s	8.6560 Ops/s	$\textbf{\color{#d91a1a}-13.09\%}$
test_serialize_model_pickle	0.4308s	0.3889s	2.5714 Ops/s	2.5372 Ops/s	$\color{#35bf28}+1.35\%$
test_serialize_weights	0.1285s	0.1135s	8.8115 Ops/s	8.4364 Ops/s	$\color{#35bf28}+4.45\%$
test_serialize_weights_returnearly	0.3594s	0.1876s	5.3313 Ops/s	5.6365 Ops/s	$\textbf{\color{#d91a1a}-5.41\%}$
test_serialize_weights_pickle	0.5355s	0.4207s	2.3771 Ops/s	2.5327 Ops/s	$\textbf{\color{#d91a1a}-6.15\%}$
test_serialize_weights_filesystem	0.2474s	0.1554s	6.4343 Ops/s	7.0182 Ops/s	$\textbf{\color{#d91a1a}-8.32\%}$
test_serialize_model_filesystem	0.1547s	0.1436s	6.9643 Ops/s	6.6335 Ops/s	$\color{#35bf28}+4.99\%$
test_reshape_pytree	72.0750μs	26.1560μs	38.2321 KOps/s	37.8392 KOps/s	$\color{#35bf28}+1.04\%$
test_reshape_td	79.8190μs	32.5921μs	30.6823 KOps/s	31.4498 KOps/s	$\color{#d91a1a}-2.44\%$
test_view_pytree	66.9150μs	25.8773μs	38.6439 KOps/s	38.6282 KOps/s	$\color{#35bf28}+0.04\%$
test_view_td	90.1580μs	38.4297μs	26.0215 KOps/s	26.2487 KOps/s	$\color{#d91a1a}-0.87\%$
test_unbind_pytree	68.6580μs	29.1583μs	34.2956 KOps/s	34.1627 KOps/s	$\color{#35bf28}+0.39\%$
test_unbind_td	0.3363ms	39.6499μs	25.2208 KOps/s	25.0255 KOps/s	$\color{#35bf28}+0.78\%$
test_split_pytree	89.1970μs	28.6044μs	34.9596 KOps/s	34.9636 KOps/s	$\color{#d91a1a}-0.01\%$
test_split_td	0.5907ms	45.0840μs	22.1808 KOps/s	22.3819 KOps/s	$\color{#d91a1a}-0.90\%$
test_add_pytree	97.3830μs	36.1285μs	27.6789 KOps/s	27.1092 KOps/s	$\color{#35bf28}+2.10\%$
test_add_td	0.1438ms	60.1986μs	16.6117 KOps/s	16.1146 KOps/s	$\color{#35bf28}+3.08\%$
test_compile_add_one_nested[tensordict-compile]	0.2063ms	67.8793μs	14.7320 KOps/s	14.9304 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_add_one_nested[tensordict-eager]	0.3061ms	0.1715ms	5.8321 KOps/s	5.6990 KOps/s	$\color{#35bf28}+2.34\%$
test_compile_add_one_nested[pytree-compile]	96.5510μs	46.4407μs	21.5328 KOps/s	21.5085 KOps/s	$\color{#35bf28}+0.11\%$
test_compile_add_one_nested[pytree-eager]	0.2004ms	0.1197ms	8.3571 KOps/s	8.2351 KOps/s	$\color{#35bf28}+1.48\%$
test_compile_copy_nested[tensordict-compile]	57.6280μs	28.4809μs	35.1113 KOps/s	36.3469 KOps/s	$\color{#d91a1a}-3.40\%$
test_compile_copy_nested[tensordict-eager]	0.1252ms	58.9449μs	16.9650 KOps/s	17.1440 KOps/s	$\color{#d91a1a}-1.04\%$
test_compile_copy_nested[pytree-compile]	0.1906ms	82.0468μs	12.1882 KOps/s	12.5092 KOps/s	$\color{#d91a1a}-2.57\%$
test_compile_copy_nested[pytree-eager]	0.1348ms	65.5904μs	15.2461 KOps/s	14.8708 KOps/s	$\color{#35bf28}+2.52\%$
test_compile_add_one_flat[tensordict-compile]	0.2462ms	0.1083ms	9.2296 KOps/s	8.9758 KOps/s	$\color{#35bf28}+2.83\%$
test_compile_add_one_flat[tensordict-eager]	0.3691ms	0.2179ms	4.5894 KOps/s	4.5901 KOps/s	$\color{#d91a1a}-0.01\%$
test_compile_add_one_flat[tensorclass-compile]	0.1117ms	48.2675μs	20.7179 KOps/s	20.9094 KOps/s	$\color{#d91a1a}-0.92\%$
test_compile_add_one_flat[tensorclass-eager]	0.1908ms	69.2666μs	14.4370 KOps/s	14.6627 KOps/s	$\color{#d91a1a}-1.54\%$
test_compile_add_one_flat[pytree-compile]	0.1869ms	0.1021ms	9.7897 KOps/s	9.8070 KOps/s	$\color{#d91a1a}-0.18\%$
test_compile_add_one_flat[pytree-eager]	0.3275ms	0.2051ms	4.8746 KOps/s	4.8958 KOps/s	$\color{#d91a1a}-0.43\%$
test_compile_add_self_flat[tensordict-eager]	0.3713ms	0.2333ms	4.2864 KOps/s	4.2783 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_add_self_flat[tensordict-compile]	0.2150ms	0.1115ms	8.9652 KOps/s	9.0390 KOps/s	$\color{#d91a1a}-0.82\%$
test_compile_add_self_flat[tensorclass-eager]	0.1421ms	64.1538μs	15.5875 KOps/s	16.1087 KOps/s	$\color{#d91a1a}-3.24\%$
test_compile_add_self_flat[tensorclass-compile]	0.1299ms	48.6225μs	20.5666 KOps/s	20.4847 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_add_self_flat[pytree-eager]	0.3123ms	0.1583ms	6.3153 KOps/s	6.2684 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_add_self_flat[pytree-compile]	0.2258ms	0.1008ms	9.9237 KOps/s	9.6689 KOps/s	$\color{#35bf28}+2.63\%$
test_compile_copy_flat[tensordict-compile]	50.3750μs	21.5255μs	46.4566 KOps/s	47.3016 KOps/s	$\color{#d91a1a}-1.79\%$
test_compile_copy_flat[tensordict-eager]	0.1341ms	67.4743μs	14.8205 KOps/s	14.9753 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_copy_flat[pytree-compile]	0.1602ms	81.5523μs	12.2621 KOps/s	12.2914 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_copy_flat[pytree-eager]	0.1292ms	67.0616μs	14.9117 KOps/s	14.5980 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_assign_and_add[tensordict-compile]	0.2983ms	0.2167ms	4.6141 KOps/s	4.6269 KOps/s	$\color{#d91a1a}-0.28\%$
test_compile_assign_and_add[tensordict-eager]	1.7631ms	1.4079ms	710.2999 Ops/s	716.8569 Ops/s	$\color{#d91a1a}-0.91\%$
test_compile_assign_and_add[pytree-compile]	0.3482ms	0.2116ms	4.7265 KOps/s	4.7208 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_assign_and_add[pytree-eager]	1.0078ms	0.8374ms	1.1942 KOps/s	1.1968 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_assign_and_add_stack[compile]	1.0273ms	0.4866ms	2.0552 KOps/s	2.1716 KOps/s	$\textbf{\color{#d91a1a}-5.36\%}$
test_compile_assign_and_add_stack[eager]	3.0436ms	2.7773ms	360.0634 Ops/s	352.8001 Ops/s	$\color{#35bf28}+2.06\%$
test_compile_indexing[tensor-tensordict-compile]	97.2920μs	39.1050μs	25.5722 KOps/s	26.0728 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_indexing[tensor-tensordict-eager]	0.5628ms	33.0731μs	30.2360 KOps/s	29.7562 KOps/s	$\color{#35bf28}+1.61\%$
test_compile_indexing[tensor-tensorclass-compile]	75.3510μs	31.5108μs	31.7351 KOps/s	32.9986 KOps/s	$\color{#d91a1a}-3.83\%$
test_compile_indexing[tensor-tensorclass-eager]	82.7150μs	22.9159μs	43.6378 KOps/s	44.0026 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_indexing[tensor-pytree-compile]	84.7490μs	32.4947μs	30.7742 KOps/s	31.9310 KOps/s	$\color{#d91a1a}-3.62\%$
test_compile_indexing[tensor-pytree-eager]	88.6660μs	22.8668μs	43.7315 KOps/s	43.4447 KOps/s	$\color{#35bf28}+0.66\%$
test_compile_indexing[slice-tensordict-compile]	0.1319ms	54.1294μs	18.4743 KOps/s	18.3707 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_indexing[slice-tensordict-eager]	0.3799ms	19.7023μs	50.7555 KOps/s	48.7564 KOps/s	$\color{#35bf28}+4.10\%$
test_compile_indexing[slice-tensorclass-compile]	0.1157ms	46.7913μs	21.3715 KOps/s	21.0231 KOps/s	$\color{#35bf28}+1.66\%$
test_compile_indexing[slice-tensorclass-eager]	77.2240μs	18.2580μs	54.7706 KOps/s	52.8804 KOps/s	$\color{#35bf28}+3.57\%$
test_compile_indexing[slice-pytree-compile]	0.1072ms	47.1439μs	21.2117 KOps/s	20.7126 KOps/s	$\color{#35bf28}+2.41\%$
test_compile_indexing[slice-pytree-eager]	59.2110μs	18.5488μs	53.9118 KOps/s	52.9155 KOps/s	$\color{#35bf28}+1.88\%$
test_compile_indexing[int-tensordict-compile]	0.1293ms	55.7975μs	17.9219 KOps/s	17.8321 KOps/s	$\color{#35bf28}+0.50\%$
test_compile_indexing[int-tensordict-eager]	1.0816ms	19.8290μs	50.4312 KOps/s	50.3308 KOps/s	$\color{#35bf28}+0.20\%$
test_compile_indexing[int-tensorclass-compile]	0.1145ms	47.4613μs	21.0698 KOps/s	20.8894 KOps/s	$\color{#35bf28}+0.86\%$
test_compile_indexing[int-tensorclass-eager]	67.9480μs	18.3744μs	54.4234 KOps/s	52.8649 KOps/s	$\color{#35bf28}+2.95\%$
test_compile_indexing[int-pytree-compile]	0.1141ms	46.6165μs	21.4516 KOps/s	20.6856 KOps/s	$\color{#35bf28}+3.70\%$
test_compile_indexing[int-pytree-eager]	56.9260μs	18.2841μs	54.6923 KOps/s	53.6871 KOps/s	$\color{#35bf28}+1.87\%$
test_mod_add[eager]	88.7560μs	36.9388μs	27.0718 KOps/s	27.1944 KOps/s	$\color{#d91a1a}-0.45\%$
test_mod_add[compile]	0.1245ms	66.9673μs	14.9327 KOps/s	15.1110 KOps/s	$\color{#d91a1a}-1.18\%$
test_mod_add[compile-overhead]	0.1259ms	65.5926μs	15.2456 KOps/s	15.1043 KOps/s	$\color{#35bf28}+0.94\%$
test_mod_wrap[eager]	0.3875ms	0.2300ms	4.3477 KOps/s	4.2952 KOps/s	$\color{#35bf28}+1.22\%$
test_mod_wrap[compile]	2.0261ms	0.2345ms	4.2650 KOps/s	4.2648 KOps/s	$+0.00\%$
test_mod_wrap[compile-overhead]	0.4214ms	0.2308ms	4.3325 KOps/s	4.2686 KOps/s	$\color{#35bf28}+1.50\%$
test_mod_wrap_and_backward[eager]	16.8172ms	12.9565ms	77.1812 Ops/s	73.9185 Ops/s	$\color{#35bf28}+4.41\%$
test_mod_wrap_and_backward[compile]	14.5086ms	11.9268ms	83.8448 Ops/s	86.4909 Ops/s	$\color{#d91a1a}-3.06\%$
test_mod_wrap_and_backward[compile-overhead]	17.4350ms	11.6809ms	85.6101 Ops/s	89.5651 Ops/s	$\color{#d91a1a}-4.42\%$
test_seq_add[eager]	0.2433ms	0.1215ms	8.2297 KOps/s	7.9756 KOps/s	$\color{#35bf28}+3.19\%$
test_seq_add[compile]	0.1493ms	78.6100μs	12.7210 KOps/s	12.7766 KOps/s	$\color{#d91a1a}-0.43\%$
test_seq_add[compile-overhead]	0.1502ms	78.0352μs	12.8147 KOps/s	13.2193 KOps/s	$\color{#d91a1a}-3.06\%$
test_seq_wrap[eager]	0.7818ms	0.4619ms	2.1650 KOps/s	2.1327 KOps/s	$\color{#35bf28}+1.51\%$
test_seq_wrap[compile]	0.4558ms	0.2492ms	4.0123 KOps/s	4.0443 KOps/s	$\color{#d91a1a}-0.79\%$
test_seq_wrap[compile-overhead]	0.4348ms	0.2480ms	4.0322 KOps/s	4.0739 KOps/s	$\color{#d91a1a}-1.02\%$
test_func_call_runtime[False-eager]	0.8320ms	0.5502ms	1.8177 KOps/s	1.7868 KOps/s	$\color{#35bf28}+1.73\%$
test_func_call_runtime[False-compile]	0.5913ms	0.4526ms	2.2097 KOps/s	2.2420 KOps/s	$\color{#d91a1a}-1.44\%$
test_func_call_runtime[False-compile-overhead]	0.6270ms	0.4502ms	2.2213 KOps/s	2.2183 KOps/s	$\color{#35bf28}+0.14\%$
test_func_call_runtime[True-eager]	0.8982ms	0.7712ms	1.2967 KOps/s	1.2984 KOps/s	$\color{#d91a1a}-0.13\%$
test_func_call_runtime[True-compile]	0.9898ms	0.4706ms	2.1250 KOps/s	2.1545 KOps/s	$\color{#d91a1a}-1.37\%$
test_func_call_runtime[True-compile-overhead]	1.0208ms	0.4715ms	2.1209 KOps/s	2.1124 KOps/s	$\color{#35bf28}+0.41\%$
test_func_call_cm_runtime[False-eager]	0.9346ms	0.5546ms	1.8030 KOps/s	1.8008 KOps/s	$\color{#35bf28}+0.12\%$
test_func_call_cm_runtime[False-compile]	0.5457ms	0.4463ms	2.2406 KOps/s	2.2309 KOps/s	$\color{#35bf28}+0.44\%$
test_func_call_cm_runtime[False-compile-overhead]	0.7330ms	0.4439ms	2.2527 KOps/s	2.2178 KOps/s	$\color{#35bf28}+1.57\%$
test_func_call_cm_runtime[True-eager]	1.8527ms	0.9126ms	1.0958 KOps/s	1.0828 KOps/s	$\color{#35bf28}+1.20\%$
test_func_call_cm_runtime[True-compile]	1.6044ms	0.8175ms	1.2233 KOps/s	1.2249 KOps/s	$\color{#d91a1a}-0.13\%$
test_func_call_cm_runtime[True-compile-overhead]	1.1140ms	0.8090ms	1.2360 KOps/s	1.2103 KOps/s	$\color{#35bf28}+2.13\%$
test_vmap_func_call_cm_runtime[eager]	2.5402ms	1.9272ms	518.8781 Ops/s	512.7897 Ops/s	$\color{#35bf28}+1.19\%$
test_vmap_func_call_cm_runtime[compile]	0.7495ms	0.5415ms	1.8469 KOps/s	1.8291 KOps/s	$\color{#35bf28}+0.97\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.0955ms	0.5479ms	1.8252 KOps/s	1.8254 KOps/s	$\color{#d91a1a}-0.01\%$
test_distributed	1.3204ms	0.1286ms	7.7742 KOps/s	7.7052 KOps/s	$\color{#35bf28}+0.89\%$
test_tdmodule	84.7290μs	28.1792μs	35.4872 KOps/s	35.6846 KOps/s	$\color{#d91a1a}-0.55\%$
test_tdmodule_dispatch	76.6330μs	51.0532μs	19.5874 KOps/s	19.6791 KOps/s	$\color{#d91a1a}-0.47\%$
test_tdseq	56.0250μs	29.7308μs	33.6352 KOps/s	32.8777 KOps/s	$\color{#35bf28}+2.30\%$
test_tdseq_dispatch	74.6490μs	55.4863μs	18.0225 KOps/s	17.6827 KOps/s	$\color{#35bf28}+1.92\%$
test_instantiation_functorch	1.6083ms	1.5146ms	660.2616 Ops/s	651.0226 Ops/s	$\color{#35bf28}+1.42\%$
test_exec_functorch	0.3273ms	0.1836ms	5.4461 KOps/s	5.5475 KOps/s	$\color{#d91a1a}-1.83\%$
test_exec_functional_call	0.3382ms	0.1755ms	5.6990 KOps/s	5.6055 KOps/s	$\color{#35bf28}+1.67\%$
test_exec_td_decorator	0.5138ms	0.2358ms	4.2409 KOps/s	4.2233 KOps/s	$\color{#35bf28}+0.42\%$
test_vmap_mlp_speed_decorator[True-True]	1.2539ms	0.6811ms	1.4682 KOps/s	1.4890 KOps/s	$\color{#d91a1a}-1.40\%$
test_vmap_mlp_speed_decorator[True-False]	0.9278ms	0.6758ms	1.4797 KOps/s	1.4783 KOps/s	$\color{#35bf28}+0.10\%$
test_vmap_mlp_speed_decorator[False-True]	0.8688ms	0.5437ms	1.8393 KOps/s	1.8497 KOps/s	$\color{#d91a1a}-0.56\%$
test_vmap_mlp_speed_decorator[False-False]	0.7534ms	0.5437ms	1.8392 KOps/s	1.8445 KOps/s	$\color{#d91a1a}-0.29\%$
test_to_module_speed[True]	2.1325ms	1.3274ms	753.3513 Ops/s	759.0852 Ops/s	$\color{#d91a1a}-0.76\%$
test_to_module_speed[False]	1.8217ms	1.2878ms	776.5105 Ops/s	773.7262 Ops/s	$\color{#35bf28}+0.36\%$
test_tc_init	95.7390μs	49.0205μs	20.3996 KOps/s	20.3739 KOps/s	$\color{#35bf28}+0.13\%$
test_tc_init_nested	0.1901ms	99.3319μs	10.0673 KOps/s	10.0808 KOps/s	$\color{#d91a1a}-0.13\%$
test_tc_first_layer_tensor	28.6630μs	1.6113μs	620.6139 KOps/s	665.5847 KOps/s	$\textbf{\color{#d91a1a}-6.76\%}$
test_tc_first_layer_nontensor	23.4940μs	4.7046μs	212.5577 KOps/s	218.9070 KOps/s	$\color{#d91a1a}-2.90\%$
test_tc_second_layer_tensor	34.5340μs	2.8837μs	346.7736 KOps/s	362.4403 KOps/s	$\color{#d91a1a}-4.32\%$
test_tc_second_layer_nontensor	42.7500μs	5.9779μs	167.2833 KOps/s	168.5727 KOps/s	$\color{#d91a1a}-0.76\%$
test_unbind	0.2541s	14.3446ms	69.7129 Ops/s	75.7509 Ops/s	$\textbf{\color{#d91a1a}-7.97\%}$
test_full_like	9.7960ms	7.7712ms	128.6809 Ops/s	117.8524 Ops/s	$\textbf{\color{#35bf28}+9.19\%}$
test_zeros_like	4.5833ms	3.0174ms	331.4110 Ops/s	367.2735 Ops/s	$\textbf{\color{#d91a1a}-9.76\%}$
test_ones_like	3.8613ms	3.3535ms	298.1966 Ops/s	319.0202 Ops/s	$\textbf{\color{#d91a1a}-6.53\%}$
test_clone	6.5847ms	5.4588ms	183.1911 Ops/s	196.2277 Ops/s	$\textbf{\color{#d91a1a}-6.64\%}$
test_squeeze	93.1650μs	12.7253μs	78.5837 KOps/s	80.3976 KOps/s	$\color{#d91a1a}-2.26\%$
test_unsqueeze	0.1637ms	93.1776μs	10.7322 KOps/s	10.9223 KOps/s	$\color{#d91a1a}-1.74\%$
test_split	0.3431ms	0.1972ms	5.0709 KOps/s	5.0861 KOps/s	$\color{#d91a1a}-0.30\%$
test_permute	0.3500ms	0.2021ms	4.9484 KOps/s	4.9232 KOps/s	$\color{#35bf28}+0.51\%$
test_stack	27.9861ms	25.5588ms	39.1254 Ops/s	38.6171 Ops/s	$\color{#35bf28}+1.32\%$
test_cat	28.5121ms	25.6800ms	38.9409 Ops/s	38.7201 Ops/s	$\color{#35bf28}+0.57\%$

[github-actions]

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

Total Benchmarks: 229. Improved: $\large\color{#35bf28}30$. Worsened: $\large\color{#d91a1a}17$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.2011ms	11.2552μs	88.8477 KOps/s	77.0871 KOps/s	$\textbf{\color{#35bf28}+15.26\%}$
test_plain_set_stack_nested	33.3710μs	11.3554μs	88.0640 KOps/s	76.7191 KOps/s	$\textbf{\color{#35bf28}+14.79\%}$
test_plain_set_nested_inplace	0.1771ms	12.2472μs	81.6515 KOps/s	70.6160 KOps/s	$\textbf{\color{#35bf28}+15.63\%}$
test_plain_set_stack_nested_inplace	41.0810μs	12.3024μs	81.2848 KOps/s	71.4602 KOps/s	$\textbf{\color{#35bf28}+13.75\%}$
test_items	71.9710μs	2.8772μs	347.5600 KOps/s	345.3599 KOps/s	$\color{#35bf28}+0.64\%$
test_items_nested	0.4222ms	0.3690ms	2.7100 KOps/s	2.7281 KOps/s	$\color{#d91a1a}-0.67\%$
test_items_nested_locked	0.4187ms	0.3647ms	2.7416 KOps/s	2.7519 KOps/s	$\color{#d91a1a}-0.37\%$
test_items_nested_leaf	80.4410μs	58.9167μs	16.9731 KOps/s	16.9430 KOps/s	$\color{#35bf28}+0.18\%$
test_items_stack_nested	0.4360ms	0.3654ms	2.7364 KOps/s	2.7506 KOps/s	$\color{#d91a1a}-0.52\%$
test_items_stack_nested_leaf	92.6720μs	60.2808μs	16.5890 KOps/s	16.5154 KOps/s	$\color{#35bf28}+0.45\%$
test_items_stack_nested_locked	0.4150ms	0.3714ms	2.6926 KOps/s	2.7261 KOps/s	$\color{#d91a1a}-1.23\%$
test_keys	29.5210μs	3.5009μs	285.6403 KOps/s	284.1845 KOps/s	$\color{#35bf28}+0.51\%$
test_keys_nested	0.1179ms	88.6969μs	11.2743 KOps/s	11.4113 KOps/s	$\color{#d91a1a}-1.20\%$
test_keys_nested_locked	0.7495ms	94.3335μs	10.6007 KOps/s	10.6893 KOps/s	$\color{#d91a1a}-0.83\%$
test_keys_nested_leaf	0.1060ms	78.3669μs	12.7605 KOps/s	12.5692 KOps/s	$\color{#35bf28}+1.52\%$
test_keys_stack_nested	0.1178ms	89.0011μs	11.2358 KOps/s	11.1211 KOps/s	$\color{#35bf28}+1.03\%$
test_keys_stack_nested_leaf	0.1062ms	79.7269μs	12.5428 KOps/s	12.4132 KOps/s	$\color{#35bf28}+1.04\%$
test_keys_stack_nested_locked	0.1294ms	95.4871μs	10.4726 KOps/s	10.4282 KOps/s	$\color{#35bf28}+0.43\%$
test_values	5.6383μs	0.8630μs	1.1588 MOps/s	1.1763 MOps/s	$\color{#d91a1a}-1.49\%$
test_values_nested	67.3020μs	37.5960μs	26.5986 KOps/s	26.1389 KOps/s	$\color{#35bf28}+1.76\%$
test_values_nested_locked	0.1037ms	39.1845μs	25.5203 KOps/s	24.6713 KOps/s	$\color{#35bf28}+3.44\%$
test_values_nested_leaf	68.9610μs	42.3267μs	23.6258 KOps/s	23.1690 KOps/s	$\color{#35bf28}+1.97\%$
test_values_stack_nested	80.3810μs	38.8513μs	25.7392 KOps/s	25.5797 KOps/s	$\color{#35bf28}+0.62\%$
test_values_stack_nested_leaf	0.1163ms	43.1716μs	23.1634 KOps/s	22.9729 KOps/s	$\color{#35bf28}+0.83\%$
test_values_stack_nested_locked	72.6110μs	40.0906μs	24.9435 KOps/s	24.6235 KOps/s	$\color{#35bf28}+1.30\%$
test_membership	1.7211μs	0.5486μs	1.8228 MOps/s	1.8196 MOps/s	$\color{#35bf28}+0.18\%$
test_membership_nested	48.2910μs	2.1072μs	474.5746 KOps/s	489.3804 KOps/s	$\color{#d91a1a}-3.03\%$
test_membership_nested_leaf	13.8705μs	2.0603μs	485.3695 KOps/s	489.1017 KOps/s	$\color{#d91a1a}-0.76\%$
test_membership_stacked_nested	68.4810μs	2.1183μs	472.0857 KOps/s	467.7627 KOps/s	$\color{#35bf28}+0.92\%$
test_membership_stacked_nested_leaf	29.3600μs	2.1341μs	468.5753 KOps/s	468.1121 KOps/s	$\color{#35bf28}+0.10\%$
test_membership_nested_last	34.0700μs	3.1650μs	315.9512 KOps/s	310.5232 KOps/s	$\color{#35bf28}+1.75\%$
test_membership_nested_leaf_last	26.8200μs	3.1573μs	316.7243 KOps/s	311.9119 KOps/s	$\color{#35bf28}+1.54\%$
test_membership_stacked_nested_last	0.1387ms	5.4840μs	182.3488 KOps/s	312.9915 KOps/s	$\textbf{\color{#d91a1a}-41.74\%}$
test_membership_stacked_nested_leaf_last	34.2010μs	5.4773μs	182.5712 KOps/s	315.3807 KOps/s	$\textbf{\color{#d91a1a}-42.11\%}$
test_nested_getleaf	35.6600μs	6.1719μs	162.0236 KOps/s	163.3556 KOps/s	$\color{#d91a1a}-0.82\%$
test_nested_get	35.1610μs	5.8316μs	171.4786 KOps/s	172.5176 KOps/s	$\color{#d91a1a}-0.60\%$
test_stacked_getleaf	27.6210μs	6.1680μs	162.1260 KOps/s	161.6157 KOps/s	$\color{#35bf28}+0.32\%$
test_stacked_get	32.9910μs	5.8756μs	170.1953 KOps/s	171.3769 KOps/s	$\color{#d91a1a}-0.69\%$
test_nested_getitemleaf	82.7410μs	6.5020μs	153.7988 KOps/s	156.4412 KOps/s	$\color{#d91a1a}-1.69\%$
test_nested_getitem	35.2900μs	6.1881μs	161.6015 KOps/s	163.2347 KOps/s	$\color{#d91a1a}-1.00\%$
test_stacked_getitemleaf	32.4310μs	6.4040μs	156.1532 KOps/s	155.6216 KOps/s	$\color{#35bf28}+0.34\%$
test_stacked_getitem	32.8010μs	6.1166μs	163.4905 KOps/s	164.8941 KOps/s	$\color{#d91a1a}-0.85\%$
test_lock_nested	10.4080ms	0.3526ms	2.8359 KOps/s	2.8966 KOps/s	$\color{#d91a1a}-2.10\%$
test_lock_stack_nested	0.4712ms	0.3406ms	2.9357 KOps/s	2.9370 KOps/s	$\color{#d91a1a}-0.04\%$
test_unlock_nested	0.6474ms	0.2831ms	3.5327 KOps/s	3.6299 KOps/s	$\color{#d91a1a}-2.68\%$
test_unlock_stack_nested	0.4464ms	0.2786ms	3.5890 KOps/s	3.6013 KOps/s	$\color{#d91a1a}-0.34\%$
test_flatten_speed	0.1149ms	75.7192μs	13.2067 KOps/s	13.2617 KOps/s	$\color{#d91a1a}-0.42\%$
test_unflatten_speed	0.5182ms	0.3213ms	3.1123 KOps/s	3.0614 KOps/s	$\color{#35bf28}+1.66\%$
test_common_ops	0.8439ms	0.5826ms	1.7165 KOps/s	1.5605 KOps/s	$\textbf{\color{#35bf28}+9.99\%}$
test_creation	0.1218ms	1.7845μs	560.3859 KOps/s	560.3042 KOps/s	$\color{#35bf28}+0.01\%$
test_creation_empty	33.4000μs	6.4870μs	154.1545 KOps/s	102.1742 KOps/s	$\textbf{\color{#35bf28}+50.87\%}$
test_creation_nested_1	37.9710μs	8.1518μs	122.6718 KOps/s	87.3806 KOps/s	$\textbf{\color{#35bf28}+40.39\%}$
test_creation_nested_2	40.4810μs	10.9476μs	91.3441 KOps/s	69.9926 KOps/s	$\textbf{\color{#35bf28}+30.51\%}$
test_clone	45.8200μs	11.0915μs	90.1594 KOps/s	93.4204 KOps/s	$\color{#d91a1a}-3.49\%$
test_getitem[int]	1.2545ms	11.1251μs	89.8867 KOps/s	94.5777 KOps/s	$\color{#d91a1a}-4.96\%$
test_getitem[slice_int]	0.1456ms	21.7651μs	45.9451 KOps/s	48.8945 KOps/s	$\textbf{\color{#d91a1a}-6.03\%}$
test_getitem[range]	0.2103ms	40.0821μs	24.9488 KOps/s	26.8865 KOps/s	$\textbf{\color{#d91a1a}-7.21\%}$
test_getitem[tuple]	0.1087ms	18.6193μs	53.7076 KOps/s	55.8988 KOps/s	$\color{#d91a1a}-3.92\%$
test_getitem[list]	0.1438ms	34.2434μs	29.2027 KOps/s	30.4922 KOps/s	$\color{#d91a1a}-4.23\%$
test_setitem_dim[int]	0.1405ms	20.2527μs	49.3761 KOps/s	49.7825 KOps/s	$\color{#d91a1a}-0.82\%$
test_setitem_dim[slice_int]	65.2010μs	38.9765μs	25.6565 KOps/s	25.9890 KOps/s	$\color{#d91a1a}-1.28\%$
test_setitem_dim[range]	95.8420μs	55.5598μs	17.9986 KOps/s	18.8378 KOps/s	$\color{#d91a1a}-4.45\%$
test_setitem_dim[tuple]	67.8310μs	33.1859μs	30.1333 KOps/s	30.7438 KOps/s	$\color{#d91a1a}-1.99\%$
test_setitem	0.1347ms	14.5559μs	68.7009 KOps/s	63.3757 KOps/s	$\textbf{\color{#35bf28}+8.40\%}$
test_set	55.7510μs	14.0511μs	71.1687 KOps/s	65.1160 KOps/s	$\textbf{\color{#35bf28}+9.30\%}$
test_set_shared	0.5789ms	0.1613ms	6.2003 KOps/s	6.1507 KOps/s	$\color{#35bf28}+0.81\%$
test_update	0.2158ms	15.9326μs	62.7643 KOps/s	52.6241 KOps/s	$\textbf{\color{#35bf28}+19.27\%}$
test_update_nested	58.3110μs	21.2485μs	47.0620 KOps/s	40.9407 KOps/s	$\textbf{\color{#35bf28}+14.95\%}$
test_update__nested	0.5766ms	27.7576μs	36.0262 KOps/s	38.8164 KOps/s	$\textbf{\color{#d91a1a}-7.19\%}$
test_set_nested	68.2310μs	15.5968μs	64.1157 KOps/s	60.5052 KOps/s	$\textbf{\color{#35bf28}+5.97\%}$
test_set_nested_new	53.1510μs	17.6255μs	56.7361 KOps/s	52.7113 KOps/s	$\textbf{\color{#35bf28}+7.64\%}$
test_select	70.3610μs	29.0523μs	34.4207 KOps/s	32.3890 KOps/s	$\textbf{\color{#35bf28}+6.27\%}$
test_select_nested	67.0110μs	44.2838μs	22.5816 KOps/s	22.7780 KOps/s	$\color{#d91a1a}-0.86\%$
test_exclude_nested	0.1316ms	63.5439μs	15.7372 KOps/s	15.9850 KOps/s	$\color{#d91a1a}-1.55\%$
test_empty[True]	0.4427ms	0.2951ms	3.3889 KOps/s	3.3572 KOps/s	$\color{#35bf28}+0.94\%$
test_empty[False]	3.7740μs	0.8381μs	1.1932 MOps/s	1.1931 MOps/s	$\color{#35bf28}+0.01\%$
test_to	87.2310μs	58.6451μs	17.0517 KOps/s	17.4923 KOps/s	$\color{#d91a1a}-2.52\%$
test_to_nonblocking	0.1952ms	47.6951μs	20.9665 KOps/s	20.6021 KOps/s	$\color{#35bf28}+1.77\%$
test_unbind_speed	0.2773ms	0.2460ms	4.0651 KOps/s	4.1549 KOps/s	$\color{#d91a1a}-2.16\%$
test_unbind_speed_stack0	0.2724ms	0.2379ms	4.2029 KOps/s	4.1898 KOps/s	$\color{#35bf28}+0.31\%$
test_unbind_speed_stack1	0.1007s	0.7384ms	1.3542 KOps/s	1.3555 KOps/s	$\color{#d91a1a}-0.09\%$
test_split	0.1118s	1.6586ms	602.9126 Ops/s	621.1709 Ops/s	$\color{#d91a1a}-2.94\%$
test_chunk	0.1053s	1.6377ms	610.6090 Ops/s	632.0932 Ops/s	$\color{#d91a1a}-3.40\%$
test_consolidate[False-None]	3.4773ms	2.6947ms	371.0938 Ops/s	372.2742 Ops/s	$\color{#d91a1a}-0.32\%$
test_consolidate[default-None]	2.3837ms	1.7372ms	575.6464 Ops/s	586.6357 Ops/s	$\color{#d91a1a}-1.87\%$
test_consolidate[reduce-overhead-None]	1.9216ms	1.7696ms	565.0908 Ops/s	577.4533 Ops/s	$\color{#d91a1a}-2.14\%$
test_consolidate_njt[False-None]	6.8647ms	6.5208ms	153.3564 Ops/s	151.1578 Ops/s	$\color{#35bf28}+1.45\%$
test_to[False-False-None]	0.3275s	2.2882ms	437.0327 Ops/s	564.0187 Ops/s	$\textbf{\color{#d91a1a}-22.51\%}$
test_to[True-False-None]	1.5590ms	1.3432ms	744.5070 Ops/s	749.8854 Ops/s	$\color{#d91a1a}-0.72\%$
test_to[within-False-None]	4.4843ms	4.1512ms	240.8939 Ops/s	241.1613 Ops/s	$\color{#d91a1a}-0.11\%$
test_to[True-default-None]	6.0979ms	5.3005ms	188.6632 Ops/s	192.1335 Ops/s	$\color{#d91a1a}-1.81\%$
test_to_njt[False-False-None]	7.3839ms	6.9788ms	143.2905 Ops/s	141.6569 Ops/s	$\color{#35bf28}+1.15\%$
test_to_njt[True-False-None]	5.8444ms	5.4792ms	182.5079 Ops/s	178.3952 Ops/s	$\color{#35bf28}+2.31\%$
test_to_njt[within-False-None]	13.0869ms	12.1463ms	82.3295 Ops/s	80.7192 Ops/s	$\color{#35bf28}+1.99\%$
test_creation[device0]	0.4649ms	81.1177μs	12.3278 KOps/s	11.7668 KOps/s	$\color{#35bf28}+4.77\%$
test_creation_from_tensor	0.5508ms	85.8724μs	11.6452 KOps/s	11.2785 KOps/s	$\color{#35bf28}+3.25\%$
test_add_one[memmap_tensor0]	0.4111ms	7.1451μs	139.9565 KOps/s	145.3637 KOps/s	$\color{#d91a1a}-3.72\%$
test_contiguous[memmap_tensor0]	9.4507μs	0.4165μs	2.4007 MOps/s	2.3650 MOps/s	$\color{#35bf28}+1.51\%$
test_stack[memmap_tensor0]	0.1845ms	4.7125μs	212.2022 KOps/s	230.1414 KOps/s	$\textbf{\color{#d91a1a}-7.79\%}$
test_memmaptd_index	1.7888ms	0.2532ms	3.9495 KOps/s	4.1624 KOps/s	$\textbf{\color{#d91a1a}-5.11\%}$
test_memmaptd_index_astensor	0.4502ms	0.3117ms	3.2086 KOps/s	3.3268 KOps/s	$\color{#d91a1a}-3.55\%$
test_memmaptd_index_op	0.7600ms	0.5703ms	1.7533 KOps/s	1.6908 KOps/s	$\color{#35bf28}+3.70\%$
test_serialize_model	0.1319s	0.1306s	7.6555 Ops/s	7.6188 Ops/s	$\color{#35bf28}+0.48\%$
test_serialize_model_pickle	2.8583s	1.8101s	0.5525 Ops/s	0.8242 Ops/s	$\textbf{\color{#d91a1a}-32.97\%}$
test_serialize_weights	0.1313s	0.1301s	7.6881 Ops/s	7.6647 Ops/s	$\color{#35bf28}+0.31\%$
test_serialize_weights_returnearly	0.4055s	66.4271ms	15.0541 Ops/s	22.8270 Ops/s	$\textbf{\color{#d91a1a}-34.05\%}$
test_serialize_weights_pickle	1.5857s	1.2800s	0.7813 Ops/s	0.8174 Ops/s	$\color{#d91a1a}-4.43\%$
test_reshape_pytree	57.3110μs	22.4552μs	44.5332 KOps/s	44.9605 KOps/s	$\color{#d91a1a}-0.95\%$
test_reshape_td	0.1859ms	26.7394μs	37.3980 KOps/s	36.5625 KOps/s	$\color{#35bf28}+2.29\%$
test_view_pytree	0.1822ms	22.2641μs	44.9154 KOps/s	45.8537 KOps/s	$\color{#d91a1a}-2.05\%$
test_view_td	0.1933ms	31.3700μs	31.8776 KOps/s	30.7310 KOps/s	$\color{#35bf28}+3.73\%$
test_unbind_pytree	0.1680ms	28.2910μs	35.3469 KOps/s	36.0203 KOps/s	$\color{#d91a1a}-1.87\%$
test_unbind_td	0.6312ms	36.4684μs	27.4210 KOps/s	27.5293 KOps/s	$\color{#d91a1a}-0.39\%$
test_split_pytree	0.2253ms	29.6928μs	33.6782 KOps/s	34.1080 KOps/s	$\color{#d91a1a}-1.26\%$
test_split_td	0.7176ms	38.5058μs	25.9701 KOps/s	14.5238 KOps/s	$\textbf{\color{#35bf28}+78.81\%}$
test_add_pytree	0.2281ms	35.2957μs	28.3321 KOps/s	28.6584 KOps/s	$\color{#d91a1a}-1.14\%$
test_add_td	0.2412ms	44.9560μs	22.2440 KOps/s	19.7658 KOps/s	$\textbf{\color{#35bf28}+12.54\%}$
test_compile_add_one_nested[tensordict-compile]	0.3060ms	0.1222ms	8.1827 KOps/s	7.5951 KOps/s	$\textbf{\color{#35bf28}+7.74\%}$
test_compile_add_one_nested[tensordict-eager]	0.5260ms	0.1309ms	7.6406 KOps/s	7.4545 KOps/s	$\color{#35bf28}+2.50\%$
test_compile_add_one_nested[pytree-compile]	0.2388ms	95.7200μs	10.4471 KOps/s	9.9692 KOps/s	$\color{#35bf28}+4.79\%$
test_compile_add_one_nested[pytree-eager]	1.0230ms	0.1496ms	6.6827 KOps/s	6.2055 KOps/s	$\textbf{\color{#35bf28}+7.69\%}$
test_compile_copy_nested[tensordict-compile]	0.4494ms	23.9008μs	41.8396 KOps/s	44.2021 KOps/s	$\textbf{\color{#d91a1a}-5.34\%}$
test_compile_copy_nested[tensordict-eager]	0.4211ms	29.1539μs	34.3007 KOps/s	33.6662 KOps/s	$\color{#35bf28}+1.88\%$
test_compile_copy_nested[pytree-compile]	0.4755ms	64.8089μs	15.4300 KOps/s	15.0188 KOps/s	$\color{#35bf28}+2.74\%$
test_compile_copy_nested[pytree-eager]	0.4469ms	49.6476μs	20.1420 KOps/s	20.0541 KOps/s	$\color{#35bf28}+0.44\%$
test_compile_add_one_flat[tensordict-compile]	0.2893ms	0.1422ms	7.0341 KOps/s	6.9944 KOps/s	$\color{#35bf28}+0.57\%$
test_compile_add_one_flat[tensordict-eager]	0.6391ms	0.2202ms	4.5416 KOps/s	4.5661 KOps/s	$\color{#d91a1a}-0.53\%$
test_compile_add_one_flat[tensorclass-compile]	0.2465ms	0.1008ms	9.9196 KOps/s	9.9044 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_add_one_flat[tensorclass-eager]	0.4829ms	58.3638μs	17.1339 KOps/s	17.6842 KOps/s	$\color{#d91a1a}-3.11\%$
test_compile_add_one_flat[pytree-compile]	0.2774ms	0.1369ms	7.3047 KOps/s	7.3916 KOps/s	$\color{#d91a1a}-1.18\%$
test_compile_add_one_flat[pytree-eager]	0.8673ms	0.4864ms	2.0561 KOps/s	2.0742 KOps/s	$\color{#d91a1a}-0.88\%$
test_compile_add_self_flat[tensordict-eager]	0.6946ms	0.2654ms	3.7685 KOps/s	3.7448 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_add_self_flat[tensordict-compile]	0.5425ms	0.1436ms	6.9641 KOps/s	7.0496 KOps/s	$\color{#d91a1a}-1.21\%$
test_compile_add_self_flat[tensorclass-eager]	0.2133ms	70.8439μs	14.1155 KOps/s	14.3900 KOps/s	$\color{#d91a1a}-1.91\%$
test_compile_add_self_flat[tensorclass-compile]	0.2523ms	0.1010ms	9.8983 KOps/s	10.1864 KOps/s	$\color{#d91a1a}-2.83\%$
test_compile_add_self_flat[pytree-eager]	0.5753ms	0.4058ms	2.4645 KOps/s	2.4641 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_add_self_flat[pytree-compile]	0.5277ms	0.1345ms	7.4336 KOps/s	7.3800 KOps/s	$\color{#35bf28}+0.73\%$
test_compile_copy_flat[tensordict-compile]	0.4247ms	18.4640μs	54.1595 KOps/s	57.6213 KOps/s	$\textbf{\color{#d91a1a}-6.01\%}$
test_compile_copy_flat[tensordict-eager]	0.4255ms	31.1080μs	32.1461 KOps/s	32.3713 KOps/s	$\color{#d91a1a}-0.70\%$
test_compile_copy_flat[pytree-compile]	0.4625ms	69.8037μs	14.3259 KOps/s	14.2144 KOps/s	$\color{#35bf28}+0.78\%$
test_compile_copy_flat[pytree-eager]	0.1138ms	50.8184μs	19.6779 KOps/s	19.5502 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_assign_and_add[tensordict-compile]	1.6620ms	0.4495ms	2.2245 KOps/s	2.1505 KOps/s	$\color{#35bf28}+3.44\%$
test_compile_assign_and_add[tensordict-eager]	2.9207ms	2.6442ms	378.1917 Ops/s	386.0796 Ops/s	$\color{#d91a1a}-2.04\%$
test_compile_assign_and_add[pytree-compile]	1.6256ms	0.4371ms	2.2878 KOps/s	2.2305 KOps/s	$\color{#35bf28}+2.57\%$
test_compile_assign_and_add[pytree-eager]	2.8806ms	2.6560ms	376.4991 Ops/s	380.4933 Ops/s	$\color{#d91a1a}-1.05\%$
test_compile_indexing[tensor-tensordict-compile]	0.3206ms	0.1165ms	8.5804 KOps/s	8.3750 KOps/s	$\color{#35bf28}+2.45\%$
test_compile_indexing[tensor-tensordict-eager]	0.5970ms	80.2600μs	12.4595 KOps/s	12.4888 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_indexing[tensor-tensorclass-compile]	0.6887ms	0.1106ms	9.0387 KOps/s	9.3306 KOps/s	$\color{#d91a1a}-3.13\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2354ms	67.8891μs	14.7299 KOps/s	14.5064 KOps/s	$\color{#35bf28}+1.54\%$
test_compile_indexing[tensor-pytree-compile]	0.2937ms	0.1144ms	8.7426 KOps/s	9.0348 KOps/s	$\color{#d91a1a}-3.23\%$
test_compile_indexing[tensor-pytree-eager]	0.2575ms	67.5284μs	14.8086 KOps/s	13.8068 KOps/s	$\textbf{\color{#35bf28}+7.26\%}$
test_compile_indexing[slice-tensordict-compile]	0.2800ms	0.1062ms	9.4148 KOps/s	9.7720 KOps/s	$\color{#d91a1a}-3.66\%$
test_compile_indexing[slice-tensordict-eager]	0.1541ms	17.2351μs	58.0213 KOps/s	58.3665 KOps/s	$\color{#d91a1a}-0.59\%$
test_compile_indexing[slice-tensorclass-compile]	0.2489ms	98.8125μs	10.1202 KOps/s	10.3344 KOps/s	$\color{#d91a1a}-2.07\%$
test_compile_indexing[slice-tensorclass-eager]	0.1463ms	15.8985μs	62.8989 KOps/s	63.2284 KOps/s	$\color{#d91a1a}-0.52\%$
test_compile_indexing[slice-pytree-compile]	0.2550ms	0.1010ms	9.8994 KOps/s	10.2527 KOps/s	$\color{#d91a1a}-3.45\%$
test_compile_indexing[slice-pytree-eager]	0.1406ms	16.0302μs	62.3823 KOps/s	64.1626 KOps/s	$\color{#d91a1a}-2.77\%$
test_compile_indexing[int-tensordict-compile]	0.2839ms	0.1067ms	9.3717 KOps/s	9.5850 KOps/s	$\color{#d91a1a}-2.22\%$
test_compile_indexing[int-tensordict-eager]	0.6634ms	17.0171μs	58.7645 KOps/s	58.7605 KOps/s	$+0.01\%$
test_compile_indexing[int-tensorclass-compile]	0.2526ms	0.1016ms	9.8415 KOps/s	10.1995 KOps/s	$\color{#d91a1a}-3.51\%$
test_compile_indexing[int-tensorclass-eager]	0.1535ms	15.9291μs	62.7783 KOps/s	63.9376 KOps/s	$\color{#d91a1a}-1.81\%$
test_compile_indexing[int-pytree-compile]	0.2493ms	0.1022ms	9.7846 KOps/s	10.1830 KOps/s	$\color{#d91a1a}-3.91\%$
test_compile_indexing[int-pytree-eager]	0.1629ms	15.9632μs	62.6441 KOps/s	63.1737 KOps/s	$\color{#d91a1a}-0.84\%$
test_mod_add[eager]	0.2221ms	38.1758μs	26.1946 KOps/s	24.8495 KOps/s	$\textbf{\color{#35bf28}+5.41\%}$
test_mod_add[compile]	0.2271ms	80.6128μs	12.4050 KOps/s	11.6049 KOps/s	$\textbf{\color{#35bf28}+6.89\%}$
test_mod_add[compile-overhead]	0.3505ms	0.1751ms	5.7104 KOps/s	5.5694 KOps/s	$\color{#35bf28}+2.53\%$
test_mod_wrap[eager]	0.4135ms	0.2615ms	3.8236 KOps/s	3.8435 KOps/s	$\color{#d91a1a}-0.52\%$
test_mod_wrap[compile]	0.4764ms	0.2990ms	3.3448 KOps/s	3.4371 KOps/s	$\color{#d91a1a}-2.68\%$
test_mod_wrap[compile-overhead]	7.5653ms	3.9959ms	250.2547 Ops/s	399.0243 Ops/s	$\textbf{\color{#d91a1a}-37.28\%}$
test_mod_wrap_and_backward[eager]	1.6253ms	1.4492ms	690.0491 Ops/s	678.6411 Ops/s	$\color{#35bf28}+1.68\%$
test_mod_wrap_and_backward[compile]	1.5754ms	1.3861ms	721.4400 Ops/s	724.6930 Ops/s	$\color{#d91a1a}-0.45\%$
test_mod_wrap_and_backward[compile-overhead]	1.7148ms	1.0784ms	927.2951 Ops/s	941.1596 Ops/s	$\color{#d91a1a}-1.47\%$
test_seq_add[eager]	0.2973ms	0.1147ms	8.7183 KOps/s	8.2161 KOps/s	$\textbf{\color{#35bf28}+6.11\%}$
test_seq_add[compile]	0.2913ms	88.7198μs	11.2714 KOps/s	11.2664 KOps/s	$\color{#35bf28}+0.04\%$
test_seq_add[compile-overhead]	0.2869ms	0.1308ms	7.6478 KOps/s	7.6793 KOps/s	$\color{#d91a1a}-0.41\%$
test_seq_wrap[eager]	0.6042ms	0.4248ms	2.3541 KOps/s	2.2585 KOps/s	$\color{#35bf28}+4.23\%$
test_seq_wrap[compile]	0.4815ms	0.3025ms	3.3060 KOps/s	3.2861 KOps/s	$\color{#35bf28}+0.60\%$
test_seq_wrap[compile-overhead]	0.3770ms	0.2269ms	4.4079 KOps/s	4.3877 KOps/s	$\color{#35bf28}+0.46\%$
test_func_call_runtime[False-eager]	0.9221ms	0.7436ms	1.3449 KOps/s	1.2808 KOps/s	$\textbf{\color{#35bf28}+5.00\%}$
test_func_call_runtime[False-compile]	0.9796ms	0.7388ms	1.3535 KOps/s	1.3293 KOps/s	$\color{#35bf28}+1.82\%$
test_func_call_runtime[False-compile-overhead]	0.5102ms	0.3681ms	2.7164 KOps/s	2.7055 KOps/s	$\color{#35bf28}+0.40\%$
test_func_call_runtime[True-eager]	1.1172ms	0.9089ms	1.1002 KOps/s	1.0826 KOps/s	$\color{#35bf28}+1.62\%$
test_func_call_runtime[True-compile]	0.9302ms	0.7575ms	1.3201 KOps/s	1.2793 KOps/s	$\color{#35bf28}+3.19\%$
test_func_call_runtime[True-compile-overhead]	0.5221ms	0.3875ms	2.5807 KOps/s	2.5554 KOps/s	$\color{#35bf28}+0.99\%$
test_func_call_cm_runtime[False-eager]	0.9371ms	0.7439ms	1.3443 KOps/s	1.3196 KOps/s	$\color{#35bf28}+1.87\%$
test_func_call_cm_runtime[False-compile]	0.9482ms	0.7534ms	1.3273 KOps/s	1.3281 KOps/s	$\color{#d91a1a}-0.06\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5135ms	0.3682ms	2.7161 KOps/s	2.6848 KOps/s	$\color{#35bf28}+1.17\%$
test_func_call_cm_runtime[True-eager]	1.1880ms	1.0161ms	984.2020 Ops/s	969.8706 Ops/s	$\color{#35bf28}+1.48\%$
test_func_call_cm_runtime[True-compile]	1.1955ms	1.0031ms	996.9571 Ops/s	985.9188 Ops/s	$\color{#35bf28}+1.12\%$
test_func_call_cm_runtime[True-compile-overhead]	1.2155ms	1.0076ms	992.4786 Ops/s	980.7352 Ops/s	$\color{#35bf28}+1.20\%$
test_vmap_func_call_cm_runtime[eager]	2.5281ms	2.1159ms	472.6122 Ops/s	466.6312 Ops/s	$\color{#35bf28}+1.28\%$
test_vmap_func_call_cm_runtime[compile]	0.9594ms	0.8087ms	1.2366 KOps/s	1.2024 KOps/s	$\color{#35bf28}+2.84\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5864ms	0.4193ms	2.3849 KOps/s	2.3467 KOps/s	$\color{#35bf28}+1.63\%$
test_distributed	4.0403ms	0.1895ms	5.2783 KOps/s	7.9738 KOps/s	$\textbf{\color{#d91a1a}-33.80\%}$
test_tdmodule	94.1120μs	19.1043μs	52.3442 KOps/s	44.1991 KOps/s	$\textbf{\color{#35bf28}+18.43\%}$
test_tdmodule_dispatch	67.5310μs	32.9359μs	30.3620 KOps/s	25.0918 KOps/s	$\textbf{\color{#35bf28}+21.00\%}$
test_tdseq	40.9310μs	19.5448μs	51.1644 KOps/s	44.8294 KOps/s	$\textbf{\color{#35bf28}+14.13\%}$
test_tdseq_dispatch	56.8610μs	36.2680μs	27.5725 KOps/s	24.5453 KOps/s	$\textbf{\color{#35bf28}+12.33\%}$
test_instantiation_functorch	1.6901ms	1.5752ms	634.8533 Ops/s	633.0756 Ops/s	$\color{#35bf28}+0.28\%$
test_exec_functorch	0.2527ms	0.1477ms	6.7686 KOps/s	6.9467 KOps/s	$\color{#d91a1a}-2.56\%$
test_exec_functional_call	0.2855ms	0.1413ms	7.0763 KOps/s	7.2294 KOps/s	$\color{#d91a1a}-2.12\%$
test_exec_td_decorator	0.3754ms	0.1912ms	5.2307 KOps/s	5.3309 KOps/s	$\color{#d91a1a}-1.88\%$
test_vmap_mlp_speed_decorator[True-True]	0.8888ms	0.6885ms	1.4524 KOps/s	1.4296 KOps/s	$\color{#35bf28}+1.59\%$
test_vmap_mlp_speed_decorator[True-False]	0.8513ms	0.6896ms	1.4501 KOps/s	1.4315 KOps/s	$\color{#35bf28}+1.30\%$
test_vmap_mlp_speed_decorator[False-True]	0.7661ms	0.6034ms	1.6573 KOps/s	1.6173 KOps/s	$\color{#35bf28}+2.48\%$
test_vmap_mlp_speed_decorator[False-False]	0.7960ms	0.6025ms	1.6598 KOps/s	1.5877 KOps/s	$\color{#35bf28}+4.54\%$
test_vmap_transformer_speed_decorator[True-True]	20.2696ms	19.3711ms	51.6233 Ops/s	51.3085 Ops/s	$\color{#35bf28}+0.61\%$
test_vmap_transformer_speed_decorator[True-False]	19.7340ms	19.3579ms	51.6586 Ops/s	51.5398 Ops/s	$\color{#35bf28}+0.23\%$
test_vmap_transformer_speed_decorator[False-True]	19.6152ms	19.2123ms	52.0501 Ops/s	51.6562 Ops/s	$\color{#35bf28}+0.76\%$
test_vmap_transformer_speed_decorator[False-False]	19.5396ms	19.2416ms	51.9707 Ops/s	51.8442 Ops/s	$\color{#35bf28}+0.24\%$
test_to_module_speed[True]	1.5002ms	0.9587ms	1.0431 KOps/s	1.0255 KOps/s	$\color{#35bf28}+1.72\%$
test_to_module_speed[False]	1.0591ms	0.9393ms	1.0646 KOps/s	1.0561 KOps/s	$\color{#35bf28}+0.80\%$
test_tc_init	82.3820μs	34.1019μs	29.3239 KOps/s	27.2712 KOps/s	$\textbf{\color{#35bf28}+7.53\%}$
test_tc_init_nested	0.1118ms	67.1599μs	14.8898 KOps/s	13.9452 KOps/s	$\textbf{\color{#35bf28}+6.77\%}$
test_tc_first_layer_tensor	23.0800μs	0.8375μs	1.1941 MOps/s	1.1913 MOps/s	$\color{#35bf28}+0.23\%$
test_tc_first_layer_nontensor	27.2300μs	2.2658μs	441.3531 KOps/s	437.0780 KOps/s	$\color{#35bf28}+0.98\%$
test_tc_second_layer_tensor	24.8300μs	1.5397μs	649.4580 KOps/s	698.7636 KOps/s	$\textbf{\color{#d91a1a}-7.06\%}$
test_tc_second_layer_nontensor	47.5510μs	2.9765μs	335.9617 KOps/s	328.8358 KOps/s	$\color{#35bf28}+2.17\%$
test_unbind	0.2449s	13.0896ms	76.3964 Ops/s	140.2193 Ops/s	$\textbf{\color{#d91a1a}-45.52\%}$
test_full_like	12.0507ms	10.7174ms	93.3061 Ops/s	96.2215 Ops/s	$\color{#d91a1a}-3.03\%$
test_zeros_like	10.1103ms	7.5683ms	132.1308 Ops/s	216.3332 Ops/s	$\textbf{\color{#d91a1a}-38.92\%}$
test_ones_like	5.5755ms	4.7552ms	210.2949 Ops/s	211.9273 Ops/s	$\color{#d91a1a}-0.77\%$
test_clone	9.1430ms	7.6361ms	130.9566 Ops/s	131.3813 Ops/s	$\color{#d91a1a}-0.32\%$
test_squeeze	0.1603ms	9.7011μs	103.0809 KOps/s	103.4029 KOps/s	$\color{#d91a1a}-0.31\%$
test_unsqueeze	0.2142ms	71.8271μs	13.9223 KOps/s	13.8785 KOps/s	$\color{#35bf28}+0.32\%$
test_split	0.3871ms	0.1593ms	6.2777 KOps/s	6.2864 KOps/s	$\color{#d91a1a}-0.14\%$
test_permute	0.3326ms	0.1852ms	5.4002 KOps/s	5.6124 KOps/s	$\color{#d91a1a}-3.78\%$
test_stack	54.2895ms	53.0875ms	18.8368 Ops/s	18.7559 Ops/s	$\color{#35bf28}+0.43\%$
test_cat	53.6724ms	52.8348ms	18.9269 Ops/s	18.9065 Ops/s	$\color{#35bf28}+0.11\%$