[BugFix] Pass type directly during reduction

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Pass type directly during reduction #1225
• [BugFix] Consolidate lazy stacks of non-tensors [EDIT] #1224

[github-actions]

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}13$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	73.3680μs	21.1358μs	47.3131 KOps/s	45.3598 KOps/s	$\color{#35bf28}+4.31\%$
test_plain_set_stack_nested	65.0730μs	21.5173μs	46.4743 KOps/s	45.3447 KOps/s	$\color{#35bf28}+2.49\%$
test_plain_set_nested_inplace	65.0220μs	23.5463μs	42.4695 KOps/s	42.1919 KOps/s	$\color{#35bf28}+0.66\%$
test_plain_set_stack_nested_inplace	53.0900μs	23.2974μs	42.9233 KOps/s	42.3885 KOps/s	$\color{#35bf28}+1.26\%$
test_items	41.1970μs	4.1558μs	240.6296 KOps/s	240.6172 KOps/s	$+0.01\%$
test_items_nested	0.6608ms	0.4022ms	2.4866 KOps/s	2.4608 KOps/s	$\color{#35bf28}+1.05\%$
test_items_nested_locked	0.5789ms	0.4008ms	2.4952 KOps/s	2.4487 KOps/s	$\color{#35bf28}+1.90\%$
test_items_nested_leaf	0.1560ms	75.8704μs	13.1804 KOps/s	12.9523 KOps/s	$\color{#35bf28}+1.76\%$
test_items_stack_nested	0.5842ms	0.4044ms	2.4726 KOps/s	2.4417 KOps/s	$\color{#35bf28}+1.27\%$
test_items_stack_nested_leaf	0.1450ms	78.7569μs	12.6973 KOps/s	12.4639 KOps/s	$\color{#35bf28}+1.87\%$
test_items_stack_nested_locked	0.7446ms	0.4029ms	2.4819 KOps/s	2.4368 KOps/s	$\color{#35bf28}+1.85\%$
test_keys	41.1180μs	3.5950μs	278.1604 KOps/s	288.1791 KOps/s	$\color{#d91a1a}-3.48\%$
test_keys_nested	0.2541ms	0.1632ms	6.1259 KOps/s	6.0375 KOps/s	$\color{#35bf28}+1.47\%$
test_keys_nested_locked	1.8175ms	0.1694ms	5.9030 KOps/s	5.7900 KOps/s	$\color{#35bf28}+1.95\%$
test_keys_nested_leaf	0.2237ms	0.1424ms	7.0208 KOps/s	6.9448 KOps/s	$\color{#35bf28}+1.09\%$
test_keys_stack_nested	0.2960ms	0.1622ms	6.1639 KOps/s	6.0936 KOps/s	$\color{#35bf28}+1.15\%$
test_keys_stack_nested_leaf	0.2349ms	0.1401ms	7.1384 KOps/s	7.1117 KOps/s	$\color{#35bf28}+0.38\%$
test_keys_stack_nested_locked	0.2592ms	0.1680ms	5.9510 KOps/s	5.8003 KOps/s	$\color{#35bf28}+2.60\%$
test_values	8.5080μs	1.0284μs	972.3776 KOps/s	912.5684 KOps/s	$\textbf{\color{#35bf28}+6.55\%}$
test_values_nested	0.1252ms	62.3406μs	16.0409 KOps/s	15.8741 KOps/s	$\color{#35bf28}+1.05\%$
test_values_nested_locked	0.1155ms	62.1109μs	16.1002 KOps/s	15.9515 KOps/s	$\color{#35bf28}+0.93\%$
test_values_nested_leaf	0.1236ms	70.5429μs	14.1758 KOps/s	13.6231 KOps/s	$\color{#35bf28}+4.06\%$
test_values_stack_nested	0.1128ms	63.3351μs	15.7890 KOps/s	15.4835 KOps/s	$\color{#35bf28}+1.97\%$
test_values_stack_nested_leaf	0.1199ms	70.3727μs	14.2101 KOps/s	14.0675 KOps/s	$\color{#35bf28}+1.01\%$
test_values_stack_nested_locked	0.1100ms	63.2012μs	15.8225 KOps/s	15.6124 KOps/s	$\color{#35bf28}+1.35\%$
test_membership	42.7300μs	0.8522μs	1.1734 MOps/s	1.1647 MOps/s	$\color{#35bf28}+0.75\%$
test_membership_nested	0.1128ms	2.9906μs	334.3798 KOps/s	353.1233 KOps/s	$\textbf{\color{#d91a1a}-5.31\%}$
test_membership_nested_leaf	39.5540μs	2.9144μs	343.1227 KOps/s	350.8865 KOps/s	$\color{#d91a1a}-2.21\%$
test_membership_stacked_nested	16.4710μs	2.8775μs	347.5288 KOps/s	355.3770 KOps/s	$\color{#d91a1a}-2.21\%$
test_membership_stacked_nested_leaf	41.2780μs	2.9157μs	342.9715 KOps/s	355.6710 KOps/s	$\color{#d91a1a}-3.57\%$
test_membership_nested_last	31.5990μs	4.3774μs	228.4451 KOps/s	232.9084 KOps/s	$\color{#d91a1a}-1.92\%$
test_membership_nested_leaf_last	39.6140μs	4.3769μs	228.4730 KOps/s	224.8871 KOps/s	$\color{#35bf28}+1.59\%$
test_membership_stacked_nested_last	27.2420μs	4.3527μs	229.7434 KOps/s	110.7715 KOps/s	$\textbf{\color{#35bf28}+107.40\%}$
test_membership_stacked_nested_leaf_last	29.6350μs	4.3687μs	228.9014 KOps/s	110.9579 KOps/s	$\textbf{\color{#35bf28}+106.30\%}$
test_nested_getleaf	47.6100μs	10.7119μs	93.3540 KOps/s	93.9481 KOps/s	$\color{#d91a1a}-0.63\%$
test_nested_get	0.1862ms	10.1827μs	98.2059 KOps/s	99.1510 KOps/s	$\color{#d91a1a}-0.95\%$
test_stacked_getleaf	35.7670μs	10.5035μs	95.2060 KOps/s	94.3618 KOps/s	$\color{#35bf28}+0.89\%$
test_stacked_get	36.0280μs	10.0725μs	99.2799 KOps/s	98.7608 KOps/s	$\color{#35bf28}+0.53\%$
test_nested_getitemleaf	55.1640μs	11.3541μs	88.0736 KOps/s	88.8758 KOps/s	$\color{#d91a1a}-0.90\%$
test_nested_getitem	37.5310μs	10.6763μs	93.6650 KOps/s	92.7299 KOps/s	$\color{#35bf28}+1.01\%$
test_stacked_getitemleaf	41.7380μs	11.1099μs	90.0094 KOps/s	89.3541 KOps/s	$\color{#35bf28}+0.73\%$
test_stacked_getitem	54.9930μs	10.6815μs	93.6194 KOps/s	93.9636 KOps/s	$\color{#d91a1a}-0.37\%$
test_lock_nested	0.6442ms	0.4073ms	2.4549 KOps/s	2.4462 KOps/s	$\color{#35bf28}+0.36\%$
test_lock_stack_nested	0.6709ms	0.4125ms	2.4240 KOps/s	2.4174 KOps/s	$\color{#35bf28}+0.28\%$
test_unlock_nested	0.4276ms	0.3294ms	3.0360 KOps/s	2.9672 KOps/s	$\color{#35bf28}+2.32\%$
test_unlock_stack_nested	0.5217ms	0.3347ms	2.9876 KOps/s	2.9786 KOps/s	$\color{#35bf28}+0.30\%$
test_flatten_speed	0.1703ms	98.7328μs	10.1283 KOps/s	9.7420 KOps/s	$\color{#35bf28}+3.97\%$
test_unflatten_speed	0.9193ms	0.5181ms	1.9302 KOps/s	1.9349 KOps/s	$\color{#d91a1a}-0.24\%$
test_common_ops	4.7546ms	0.7912ms	1.2638 KOps/s	1.2356 KOps/s	$\color{#35bf28}+2.28\%$
test_creation	46.9780μs	2.6902μs	371.7235 KOps/s	394.5068 KOps/s	$\textbf{\color{#d91a1a}-5.78\%}$
test_creation_empty	41.2480μs	11.7597μs	85.0365 KOps/s	77.6438 KOps/s	$\textbf{\color{#35bf28}+9.52\%}$
test_creation_nested_1	46.3070μs	14.5521μs	68.7185 KOps/s	62.4425 KOps/s	$\textbf{\color{#35bf28}+10.05\%}$
test_creation_nested_2	74.4700μs	19.1495μs	52.2206 KOps/s	48.3439 KOps/s	$\textbf{\color{#35bf28}+8.02\%}$
test_clone	0.1020ms	13.5874μs	73.5978 KOps/s	73.1852 KOps/s	$\color{#35bf28}+0.56\%$
test_getitem[int]	0.7758ms	12.8211μs	77.9963 KOps/s	79.9601 KOps/s	$\color{#d91a1a}-2.46\%$
test_getitem[slice_int]	0.1324ms	23.9618μs	41.7331 KOps/s	42.5067 KOps/s	$\color{#d91a1a}-1.82\%$
test_getitem[range]	0.1598ms	49.7175μs	20.1136 KOps/s	19.9864 KOps/s	$\color{#35bf28}+0.64\%$
test_getitem[tuple]	0.1215ms	20.1111μs	49.7237 KOps/s	50.4114 KOps/s	$\color{#d91a1a}-1.36\%$
test_getitem[list]	0.1591ms	45.7504μs	21.8578 KOps/s	21.6441 KOps/s	$\color{#35bf28}+0.99\%$
test_setitem_dim[int]	51.2470μs	25.6388μs	39.0034 KOps/s	39.6802 KOps/s	$\color{#d91a1a}-1.71\%$
test_setitem_dim[slice_int]	97.7240μs	52.2575μs	19.1360 KOps/s	19.5637 KOps/s	$\color{#d91a1a}-2.19\%$
test_setitem_dim[range]	0.1658ms	76.7426μs	13.0306 KOps/s	12.7591 KOps/s	$\color{#35bf28}+2.13\%$
test_setitem_dim[tuple]	79.1690μs	41.6185μs	24.0278 KOps/s	24.6314 KOps/s	$\color{#d91a1a}-2.45\%$
test_setitem	0.1113ms	20.5314μs	48.7058 KOps/s	44.1266 KOps/s	$\textbf{\color{#35bf28}+10.38\%}$
test_set	0.1261ms	19.9291μs	50.1778 KOps/s	49.1116 KOps/s	$\color{#35bf28}+2.17\%$
test_set_shared	4.1007ms	0.1848ms	5.4127 KOps/s	5.5576 KOps/s	$\color{#d91a1a}-2.61\%$
test_update	0.1453ms	23.1352μs	43.2242 KOps/s	42.0405 KOps/s	$\color{#35bf28}+2.82\%$
test_update_nested	0.1495ms	34.0490μs	29.3695 KOps/s	29.7640 KOps/s	$\color{#d91a1a}-1.33\%$
test_update__nested	0.4420ms	34.4351μs	29.0401 KOps/s	30.0909 KOps/s	$\color{#d91a1a}-3.49\%$
test_set_nested	83.7580μs	22.2912μs	44.8608 KOps/s	44.9926 KOps/s	$\color{#d91a1a}-0.29\%$
test_set_nested_new	0.1335ms	27.1632μs	36.8145 KOps/s	37.6892 KOps/s	$\color{#d91a1a}-2.32\%$
test_select	0.1204ms	42.3684μs	23.6025 KOps/s	23.1762 KOps/s	$\color{#35bf28}+1.84\%$
test_select_nested	0.1255ms	62.8537μs	15.9100 KOps/s	16.0196 KOps/s	$\color{#d91a1a}-0.68\%$
test_exclude_nested	0.1356ms	82.0479μs	12.1880 KOps/s	12.2676 KOps/s	$\color{#d91a1a}-0.65\%$
test_empty[True]	0.5884ms	0.4053ms	2.4672 KOps/s	2.4744 KOps/s	$\color{#d91a1a}-0.29\%$
test_empty[False]	9.3820μs	1.4258μs	701.3465 KOps/s	721.2848 KOps/s	$\color{#d91a1a}-2.76\%$
test_unbind_speed	0.4107ms	0.2682ms	3.7291 KOps/s	3.7231 KOps/s	$\color{#35bf28}+0.16\%$
test_unbind_speed_stack0	0.6100ms	0.2645ms	3.7804 KOps/s	3.8210 KOps/s	$\color{#d91a1a}-1.06\%$
test_unbind_speed_stack1	0.1033s	0.7177ms	1.3933 KOps/s	1.3979 KOps/s	$\color{#d91a1a}-0.33\%$
test_split	97.1256ms	1.7119ms	584.1535 Ops/s	518.6044 Ops/s	$\textbf{\color{#35bf28}+12.64\%}$
test_chunk	0.1010s	1.7185ms	581.9057 Ops/s	633.7025 Ops/s	$\textbf{\color{#d91a1a}-8.17\%}$
test_consolidate_njt[False-None]	8.2517ms	8.0410ms	124.3630 Ops/s	110.9241 Ops/s	$\textbf{\color{#35bf28}+12.12\%}$
test_creation[device0]	0.2212ms	92.1591μs	10.8508 KOps/s	10.7421 KOps/s	$\color{#35bf28}+1.01\%$
test_creation_from_tensor	0.2273ms	93.9182μs	10.6476 KOps/s	10.5887 KOps/s	$\color{#35bf28}+0.56\%$
test_add_one[memmap_tensor0]	86.3230μs	4.8856μs	204.6824 KOps/s	201.4157 KOps/s	$\color{#35bf28}+1.62\%$
test_contiguous[memmap_tensor0]	7.8840μs	0.5048μs	1.9809 MOps/s	1.9491 MOps/s	$\color{#35bf28}+1.63\%$
test_stack[memmap_tensor0]	50.1840μs	3.3511μs	298.4089 KOps/s	298.4609 KOps/s	$\color{#d91a1a}-0.02\%$
test_memmaptd_index	0.3193ms	0.2319ms	4.3116 KOps/s	4.4980 KOps/s	$\color{#d91a1a}-4.14\%$
test_memmaptd_index_astensor	0.7355ms	0.3176ms	3.1485 KOps/s	3.2425 KOps/s	$\color{#d91a1a}-2.90\%$
test_memmaptd_index_op	0.7790ms	0.5785ms	1.7287 KOps/s	1.7067 KOps/s	$\color{#35bf28}+1.29\%$
test_serialize_model	0.2092s	0.1293s	7.7313 Ops/s	8.5487 Ops/s	$\textbf{\color{#d91a1a}-9.56\%}$
test_serialize_model_pickle	0.5023s	0.4074s	2.4545 Ops/s	2.5190 Ops/s	$\color{#d91a1a}-2.56\%$
test_serialize_weights	0.1234s	0.1149s	8.7030 Ops/s	8.2882 Ops/s	$\textbf{\color{#35bf28}+5.00\%}$
test_serialize_weights_returnearly	0.1817s	0.1589s	6.2923 Ops/s	6.0984 Ops/s	$\color{#35bf28}+3.18\%$
test_serialize_weights_pickle	1.0098s	0.7333s	1.3637 Ops/s	2.3827 Ops/s	$\textbf{\color{#d91a1a}-42.77\%}$
test_serialize_weights_filesystem	0.2361s	0.1550s	6.4497 Ops/s	6.2814 Ops/s	$\color{#35bf28}+2.68\%$
test_serialize_model_filesystem	0.1523s	0.1419s	7.0487 Ops/s	6.5652 Ops/s	$\textbf{\color{#35bf28}+7.36\%}$
test_reshape_pytree	78.7000μs	26.2062μs	38.1589 KOps/s	37.7948 KOps/s	$\color{#35bf28}+0.96\%$
test_reshape_td	81.7040μs	32.3632μs	30.8993 KOps/s	30.3990 KOps/s	$\color{#35bf28}+1.65\%$
test_view_pytree	58.0190μs	25.9782μs	38.4938 KOps/s	38.1041 KOps/s	$\color{#35bf28}+1.02\%$
test_view_td	97.9140μs	40.2615μs	24.8377 KOps/s	24.9388 KOps/s	$\color{#d91a1a}-0.41\%$
test_unbind_pytree	72.5560μs	29.7452μs	33.6188 KOps/s	34.2928 KOps/s	$\color{#d91a1a}-1.97\%$
test_unbind_td	0.3660ms	39.8114μs	25.1184 KOps/s	25.1713 KOps/s	$\color{#d91a1a}-0.21\%$
test_split_pytree	70.4020μs	28.8539μs	34.6574 KOps/s	34.7229 KOps/s	$\color{#d91a1a}-0.19\%$
test_split_td	0.5477ms	44.6268μs	22.4081 KOps/s	22.6094 KOps/s	$\color{#d91a1a}-0.89\%$
test_add_pytree	0.1011ms	35.1322μs	28.4639 KOps/s	28.0532 KOps/s	$\color{#35bf28}+1.46\%$
test_add_td	0.1710ms	55.9808μs	17.8633 KOps/s	16.8212 KOps/s	$\textbf{\color{#35bf28}+6.19\%}$
test_compile_add_one_nested[tensordict-compile]	0.1306ms	66.4415μs	15.0508 KOps/s	15.1489 KOps/s	$\color{#d91a1a}-0.65\%$
test_compile_add_one_nested[tensordict-eager]	1.4046ms	0.1703ms	5.8725 KOps/s	5.8281 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_add_one_nested[pytree-compile]	87.8250μs	45.6847μs	21.8892 KOps/s	21.7597 KOps/s	$\color{#35bf28}+0.59\%$
test_compile_add_one_nested[pytree-eager]	0.2019ms	0.1177ms	8.4984 KOps/s	8.4135 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_copy_nested[tensordict-compile]	82.6350μs	27.9906μs	35.7263 KOps/s	36.2986 KOps/s	$\color{#d91a1a}-1.58\%$
test_compile_copy_nested[tensordict-eager]	0.1159ms	58.7293μs	17.0273 KOps/s	17.0511 KOps/s	$\color{#d91a1a}-0.14\%$
test_compile_copy_nested[pytree-compile]	0.1709ms	78.5992μs	12.7228 KOps/s	12.5507 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_copy_nested[pytree-eager]	0.1312ms	66.1031μs	15.1279 KOps/s	14.8380 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_add_one_flat[tensordict-compile]	0.1987ms	0.1083ms	9.2339 KOps/s	9.2857 KOps/s	$\color{#d91a1a}-0.56\%$
test_compile_add_one_flat[tensordict-eager]	0.4016ms	0.2139ms	4.6743 KOps/s	4.6126 KOps/s	$\color{#35bf28}+1.34\%$
test_compile_add_one_flat[tensorclass-compile]	0.1830ms	47.0642μs	21.2476 KOps/s	21.1639 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_add_one_flat[tensorclass-eager]	0.1410ms	65.1917μs	15.3394 KOps/s	14.9966 KOps/s	$\color{#35bf28}+2.29\%$
test_compile_add_one_flat[pytree-compile]	0.2439ms	0.1029ms	9.7227 KOps/s	9.9226 KOps/s	$\color{#d91a1a}-2.01\%$
test_compile_add_one_flat[pytree-eager]	0.4268ms	0.2029ms	4.9289 KOps/s	4.9291 KOps/s	$-0.00\%$
test_compile_add_self_flat[tensordict-eager]	0.4313ms	0.2320ms	4.3102 KOps/s	4.2746 KOps/s	$\color{#35bf28}+0.83\%$
test_compile_add_self_flat[tensordict-compile]	0.4054ms	0.1151ms	8.6851 KOps/s	9.2672 KOps/s	$\textbf{\color{#d91a1a}-6.28\%}$
test_compile_add_self_flat[tensorclass-eager]	0.2084ms	61.5999μs	16.2338 KOps/s	16.0743 KOps/s	$\color{#35bf28}+0.99\%$
test_compile_add_self_flat[tensorclass-compile]	0.1627ms	47.3931μs	21.1001 KOps/s	20.6972 KOps/s	$\color{#35bf28}+1.95\%$
test_compile_add_self_flat[pytree-eager]	0.3433ms	0.1572ms	6.3595 KOps/s	6.3959 KOps/s	$\color{#d91a1a}-0.57\%$
test_compile_add_self_flat[pytree-compile]	0.4360ms	0.1020ms	9.8027 KOps/s	9.8645 KOps/s	$\color{#d91a1a}-0.63\%$
test_compile_copy_flat[tensordict-compile]	73.9690μs	21.1314μs	47.3229 KOps/s	47.7182 KOps/s	$\color{#d91a1a}-0.83\%$
test_compile_copy_flat[tensordict-eager]	0.1462ms	67.8538μs	14.7376 KOps/s	14.4543 KOps/s	$\color{#35bf28}+1.96\%$
test_compile_copy_flat[pytree-compile]	0.1526ms	83.9161μs	11.9167 KOps/s	12.2173 KOps/s	$\color{#d91a1a}-2.46\%$
test_compile_copy_flat[pytree-eager]	0.1529ms	68.3169μs	14.6377 KOps/s	15.1121 KOps/s	$\color{#d91a1a}-3.14\%$
test_compile_assign_and_add[tensordict-compile]	0.3028ms	0.2196ms	4.5541 KOps/s	4.6308 KOps/s	$\color{#d91a1a}-1.66\%$
test_compile_assign_and_add[tensordict-eager]	2.6519ms	1.3622ms	734.1234 Ops/s	732.7073 Ops/s	$\color{#35bf28}+0.19\%$
test_compile_assign_and_add[pytree-compile]	0.2839ms	0.2160ms	4.6293 KOps/s	4.6876 KOps/s	$\color{#d91a1a}-1.25\%$
test_compile_assign_and_add[pytree-eager]	1.2274ms	0.8200ms	1.2196 KOps/s	1.2271 KOps/s	$\color{#d91a1a}-0.62\%$
test_compile_assign_and_add_stack[compile]	0.6458ms	0.4662ms	2.1452 KOps/s	2.1847 KOps/s	$\color{#d91a1a}-1.81\%$
test_compile_assign_and_add_stack[eager]	2.9150ms	2.6282ms	380.4955 Ops/s	358.1458 Ops/s	$\textbf{\color{#35bf28}+6.24\%}$
test_compile_indexing[tensor-tensordict-compile]	93.3660μs	37.8243μs	26.4381 KOps/s	25.8872 KOps/s	$\color{#35bf28}+2.13\%$
test_compile_indexing[tensor-tensordict-eager]	0.5544ms	33.0900μs	30.2206 KOps/s	30.5639 KOps/s	$\color{#d91a1a}-1.12\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1174ms	30.7676μs	32.5017 KOps/s	32.8333 KOps/s	$\color{#d91a1a}-1.01\%$
test_compile_indexing[tensor-tensorclass-eager]	79.2100μs	23.2979μs	42.9223 KOps/s	43.3388 KOps/s	$\color{#d91a1a}-0.96\%$
test_compile_indexing[tensor-pytree-compile]	0.1189ms	31.4903μs	31.7558 KOps/s	31.4984 KOps/s	$\color{#35bf28}+0.82\%$
test_compile_indexing[tensor-pytree-eager]	0.1373ms	24.1744μs	41.3661 KOps/s	43.4761 KOps/s	$\color{#d91a1a}-4.85\%$
test_compile_indexing[slice-tensordict-compile]	0.1261ms	54.2853μs	18.4212 KOps/s	18.5150 KOps/s	$\color{#d91a1a}-0.51\%$
test_compile_indexing[slice-tensordict-eager]	0.3307ms	20.3382μs	49.1687 KOps/s	50.7323 KOps/s	$\color{#d91a1a}-3.08\%$
test_compile_indexing[slice-tensorclass-compile]	0.1061ms	47.1833μs	21.1939 KOps/s	21.5121 KOps/s	$\color{#d91a1a}-1.48\%$
test_compile_indexing[slice-tensorclass-eager]	63.9500μs	18.4151μs	54.3033 KOps/s	54.4553 KOps/s	$\color{#d91a1a}-0.28\%$
test_compile_indexing[slice-pytree-compile]	0.1267ms	47.1081μs	21.2278 KOps/s	20.7580 KOps/s	$\color{#35bf28}+2.26\%$
test_compile_indexing[slice-pytree-eager]	83.8880μs	18.3140μs	54.6031 KOps/s	53.6559 KOps/s	$\color{#35bf28}+1.77\%$
test_compile_indexing[int-tensordict-compile]	0.1055ms	54.8565μs	18.2294 KOps/s	18.2699 KOps/s	$\color{#d91a1a}-0.22\%$
test_compile_indexing[int-tensordict-eager]	0.8771ms	19.8710μs	50.3246 KOps/s	51.8446 KOps/s	$\color{#d91a1a}-2.93\%$
test_compile_indexing[int-tensorclass-compile]	0.1075ms	47.0439μs	21.2567 KOps/s	21.7539 KOps/s	$\color{#d91a1a}-2.29\%$
test_compile_indexing[int-tensorclass-eager]	47.9400μs	18.2035μs	54.9345 KOps/s	55.3269 KOps/s	$\color{#d91a1a}-0.71\%$
test_compile_indexing[int-pytree-compile]	0.1358ms	47.3930μs	21.1002 KOps/s	21.8220 KOps/s	$\color{#d91a1a}-3.31\%$
test_compile_indexing[int-pytree-eager]	74.7810μs	18.3269μs	54.5646 KOps/s	55.1995 KOps/s	$\color{#d91a1a}-1.15\%$
test_mod_add[eager]	80.9430μs	34.6324μs	28.8747 KOps/s	27.7389 KOps/s	$\color{#35bf28}+4.09\%$
test_mod_add[compile]	0.1332ms	66.4537μs	15.0481 KOps/s	15.9242 KOps/s	$\textbf{\color{#d91a1a}-5.50\%}$
test_mod_add[compile-overhead]	0.1636ms	66.5532μs	15.0256 KOps/s	15.6896 KOps/s	$\color{#d91a1a}-4.23\%$
test_mod_wrap[eager]	0.3912ms	0.2255ms	4.4350 KOps/s	4.5273 KOps/s	$\color{#d91a1a}-2.04\%$
test_mod_wrap[compile]	2.3750ms	0.2343ms	4.2682 KOps/s	4.3994 KOps/s	$\color{#d91a1a}-2.98\%$
test_mod_wrap[compile-overhead]	0.4083ms	0.2264ms	4.4172 KOps/s	4.4274 KOps/s	$\color{#d91a1a}-0.23\%$
test_mod_wrap_and_backward[eager]	19.4238ms	13.4911ms	74.1231 Ops/s	74.9291 Ops/s	$\color{#d91a1a}-1.08\%$
test_mod_wrap_and_backward[compile]	13.1499ms	11.3408ms	88.1771 Ops/s	87.7428 Ops/s	$\color{#35bf28}+0.49\%$
test_mod_wrap_and_backward[compile-overhead]	12.0926ms	11.1710ms	89.5179 Ops/s	89.8320 Ops/s	$\color{#d91a1a}-0.35\%$
test_seq_add[eager]	0.2210ms	0.1191ms	8.3970 KOps/s	8.4611 KOps/s	$\color{#d91a1a}-0.76\%$
test_seq_add[compile]	0.1997ms	80.4299μs	12.4332 KOps/s	13.4032 KOps/s	$\textbf{\color{#d91a1a}-7.24\%}$
test_seq_add[compile-overhead]	0.1428ms	76.9224μs	13.0001 KOps/s	13.4634 KOps/s	$\color{#d91a1a}-3.44\%$
test_seq_wrap[eager]	0.8222ms	0.4460ms	2.2421 KOps/s	2.2462 KOps/s	$\color{#d91a1a}-0.18\%$
test_seq_wrap[compile]	0.4481ms	0.2487ms	4.0217 KOps/s	4.1618 KOps/s	$\color{#d91a1a}-3.37\%$
test_seq_wrap[compile-overhead]	0.3313ms	0.2501ms	3.9984 KOps/s	4.2099 KOps/s	$\textbf{\color{#d91a1a}-5.02\%}$
test_func_call_runtime[False-eager]	0.7561ms	0.5420ms	1.8452 KOps/s	1.8494 KOps/s	$\color{#d91a1a}-0.23\%$
test_func_call_runtime[False-compile]	0.5698ms	0.4505ms	2.2198 KOps/s	2.2500 KOps/s	$\color{#d91a1a}-1.34\%$
test_func_call_runtime[False-compile-overhead]	0.8287ms	0.4512ms	2.2164 KOps/s	2.2920 KOps/s	$\color{#d91a1a}-3.30\%$
test_func_call_runtime[True-eager]	0.9000ms	0.7619ms	1.3124 KOps/s	1.3582 KOps/s	$\color{#d91a1a}-3.37\%$
test_func_call_runtime[True-compile]	0.6256ms	0.4699ms	2.1282 KOps/s	2.1494 KOps/s	$\color{#d91a1a}-0.99\%$
test_func_call_runtime[True-compile-overhead]	0.6160ms	0.4699ms	2.1283 KOps/s	2.1657 KOps/s	$\color{#d91a1a}-1.73\%$
test_func_call_cm_runtime[False-eager]	0.8265ms	0.5479ms	1.8253 KOps/s	1.8730 KOps/s	$\color{#d91a1a}-2.55\%$
test_func_call_cm_runtime[False-compile]	0.8214ms	0.4521ms	2.2119 KOps/s	2.2626 KOps/s	$\color{#d91a1a}-2.24\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5781ms	0.4474ms	2.2354 KOps/s	2.2326 KOps/s	$\color{#35bf28}+0.13\%$
test_func_call_cm_runtime[True-eager]	1.5846ms	0.8975ms	1.1142 KOps/s	1.1183 KOps/s	$\color{#d91a1a}-0.37\%$
test_func_call_cm_runtime[True-compile]	1.0849ms	0.7906ms	1.2649 KOps/s	1.2566 KOps/s	$\color{#35bf28}+0.66\%$
test_func_call_cm_runtime[True-compile-overhead]	1.4683ms	0.8007ms	1.2489 KOps/s	1.2425 KOps/s	$\color{#35bf28}+0.51\%$
test_vmap_func_call_cm_runtime[eager]	3.1397ms	1.9428ms	514.7152 Ops/s	521.3969 Ops/s	$\color{#d91a1a}-1.28\%$
test_vmap_func_call_cm_runtime[compile]	0.6900ms	0.5462ms	1.8310 KOps/s	1.8434 KOps/s	$\color{#d91a1a}-0.67\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.7357ms	0.5500ms	1.8181 KOps/s	1.8362 KOps/s	$\color{#d91a1a}-0.98\%$
test_distributed	0.2565ms	0.1264ms	7.9120 KOps/s	7.8942 KOps/s	$\color{#35bf28}+0.23\%$
test_tdmodule	0.1180ms	27.3747μs	36.5300 KOps/s	36.8614 KOps/s	$\color{#d91a1a}-0.90\%$
test_tdmodule_dispatch	77.4650μs	50.4470μs	19.8228 KOps/s	20.2867 KOps/s	$\color{#d91a1a}-2.29\%$
test_tdseq	50.5050μs	29.4125μs	33.9992 KOps/s	34.3308 KOps/s	$\color{#d91a1a}-0.97\%$
test_tdseq_dispatch	94.8680μs	54.1311μs	18.4737 KOps/s	18.0873 KOps/s	$\color{#35bf28}+2.14\%$
test_instantiation_functorch	2.3338ms	1.5267ms	655.0289 Ops/s	654.4930 Ops/s	$\color{#35bf28}+0.08\%$
test_exec_functorch	0.3209ms	0.1754ms	5.7015 KOps/s	5.7577 KOps/s	$\color{#d91a1a}-0.98\%$
test_exec_functional_call	0.3480ms	0.1714ms	5.8341 KOps/s	5.9462 KOps/s	$\color{#d91a1a}-1.88\%$
test_exec_td_decorator	0.5495ms	0.2350ms	4.2547 KOps/s	4.3809 KOps/s	$\color{#d91a1a}-2.88\%$
test_vmap_mlp_speed_decorator[True-True]	0.8612ms	0.6597ms	1.5159 KOps/s	1.4930 KOps/s	$\color{#35bf28}+1.53\%$
test_vmap_mlp_speed_decorator[True-False]	0.8635ms	0.6591ms	1.5173 KOps/s	1.5241 KOps/s	$\color{#d91a1a}-0.45\%$
test_vmap_mlp_speed_decorator[False-True]	0.8636ms	0.5330ms	1.8762 KOps/s	1.8856 KOps/s	$\color{#d91a1a}-0.50\%$
test_vmap_mlp_speed_decorator[False-False]	0.7554ms	0.5342ms	1.8720 KOps/s	1.8903 KOps/s	$\color{#d91a1a}-0.97\%$
test_to_module_speed[True]	1.9207ms	1.3410ms	745.6889 Ops/s	756.7163 Ops/s	$\color{#d91a1a}-1.46\%$
test_to_module_speed[False]	1.8672ms	1.3070ms	765.1127 Ops/s	761.4716 Ops/s	$\color{#35bf28}+0.48\%$
test_tc_init	0.1032ms	48.6976μs	20.5349 KOps/s	20.4788 KOps/s	$\color{#35bf28}+0.27\%$
test_tc_init_nested	0.1682ms	94.2443μs	10.6107 KOps/s	10.3317 KOps/s	$\color{#35bf28}+2.70\%$
test_tc_first_layer_tensor	27.3310μs	1.5374μs	650.4461 KOps/s	667.5568 KOps/s	$\color{#d91a1a}-2.56\%$
test_tc_first_layer_nontensor	55.2040μs	4.6578μs	214.6950 KOps/s	208.9653 KOps/s	$\color{#35bf28}+2.74\%$
test_tc_second_layer_tensor	29.8450μs	2.8476μs	351.1668 KOps/s	355.4703 KOps/s	$\color{#d91a1a}-1.21\%$
test_tc_second_layer_nontensor	53.8090μs	6.0841μs	164.3627 KOps/s	159.6604 KOps/s	$\color{#35bf28}+2.95\%$
test_unbind	0.2461s	13.8782ms	72.0552 Ops/s	61.5748 Ops/s	$\textbf{\color{#35bf28}+17.02\%}$
test_full_like	9.8828ms	8.9548ms	111.6722 Ops/s	132.8169 Ops/s	$\textbf{\color{#d91a1a}-15.92\%}$
test_zeros_like	6.1890ms	3.2957ms	303.4220 Ops/s	366.5040 Ops/s	$\textbf{\color{#d91a1a}-17.21\%}$
test_ones_like	4.7759ms	3.5759ms	279.6519 Ops/s	301.7850 Ops/s	$\textbf{\color{#d91a1a}-7.33\%}$
test_clone	10.9172ms	6.8794ms	145.3619 Ops/s	197.5532 Ops/s	$\textbf{\color{#d91a1a}-26.42\%}$
test_squeeze	83.1470μs	12.7846μs	78.2190 KOps/s	78.7710 KOps/s	$\color{#d91a1a}-0.70\%$
test_unsqueeze	0.3757ms	96.2666μs	10.3878 KOps/s	10.9264 KOps/s	$\color{#d91a1a}-4.93\%$
test_split	0.3958ms	0.1971ms	5.0731 KOps/s	5.2796 KOps/s	$\color{#d91a1a}-3.91\%$
test_permute	0.3257ms	0.2086ms	4.7948 KOps/s	4.9780 KOps/s	$\color{#d91a1a}-3.68\%$
test_stack	29.9063ms	24.9307ms	40.1112 Ops/s	39.0256 Ops/s	$\color{#35bf28}+2.78\%$
test_cat	29.7383ms	25.2118ms	39.6640 Ops/s	38.3236 Ops/s	$\color{#35bf28}+3.50\%$