[BugFix] resilient _exclude_td_from_pytree

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] resilient _exclude_td_from_pytree #1038

[github-actions]

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

Total Benchmarks: 216. Improved: $\large\color{#35bf28}17$. Worsened: $\large\color{#d91a1a}12$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	47.8590μs	24.9025μs	40.1566 KOps/s	40.2328 KOps/s	$\color{#d91a1a}-0.19\%$
test_plain_set_stack_nested	56.4550μs	25.4234μs	39.3339 KOps/s	39.4117 KOps/s	$\color{#d91a1a}-0.20\%$
test_plain_set_nested_inplace	88.2250μs	27.7254μs	36.0680 KOps/s	36.5994 KOps/s	$\color{#d91a1a}-1.45\%$
test_plain_set_stack_nested_inplace	89.9470μs	27.8796μs	35.8686 KOps/s	36.8665 KOps/s	$\color{#d91a1a}-2.71\%$
test_items	23.7240μs	4.1947μs	238.3958 KOps/s	236.7938 KOps/s	$\color{#35bf28}+0.68\%$
test_items_nested	0.5428ms	0.3821ms	2.6172 KOps/s	2.6046 KOps/s	$\color{#35bf28}+0.48\%$
test_items_nested_locked	0.4815ms	0.3811ms	2.6240 KOps/s	2.6012 KOps/s	$\color{#35bf28}+0.88\%$
test_items_nested_leaf	0.1574ms	79.6355μs	12.5572 KOps/s	12.5738 KOps/s	$\color{#d91a1a}-0.13\%$
test_items_stack_nested	0.5263ms	0.3853ms	2.5956 KOps/s	2.5701 KOps/s	$\color{#35bf28}+0.99\%$
test_items_stack_nested_leaf	0.1628ms	80.6736μs	12.3956 KOps/s	12.1720 KOps/s	$\color{#35bf28}+1.84\%$
test_items_stack_nested_locked	0.5203ms	0.3806ms	2.6273 KOps/s	2.5887 KOps/s	$\color{#35bf28}+1.49\%$
test_keys	22.4520μs	3.4524μs	289.6503 KOps/s	282.8045 KOps/s	$\color{#35bf28}+2.42\%$
test_keys_nested	0.1822ms	0.1353ms	7.3931 KOps/s	7.4870 KOps/s	$\color{#d91a1a}-1.25\%$
test_keys_nested_locked	1.5988ms	0.1412ms	7.0801 KOps/s	7.1202 KOps/s	$\color{#d91a1a}-0.56\%$
test_keys_nested_leaf	0.2249ms	0.1189ms	8.4091 KOps/s	8.5241 KOps/s	$\color{#d91a1a}-1.35\%$
test_keys_stack_nested	0.2316ms	0.1346ms	7.4303 KOps/s	7.4326 KOps/s	$\color{#d91a1a}-0.03\%$
test_keys_stack_nested_leaf	0.2059ms	0.1186ms	8.4314 KOps/s	8.5496 KOps/s	$\color{#d91a1a}-1.38\%$
test_keys_stack_nested_locked	0.2615ms	0.1412ms	7.0838 KOps/s	7.1226 KOps/s	$\color{#d91a1a}-0.54\%$
test_values	5.7166μs	1.0312μs	969.7041 KOps/s	937.7524 KOps/s	$\color{#35bf28}+3.41\%$
test_values_nested	0.1648ms	93.1940μs	10.7303 KOps/s	10.6201 KOps/s	$\color{#35bf28}+1.04\%$
test_values_nested_locked	0.1678ms	92.8763μs	10.7670 KOps/s	10.6408 KOps/s	$\color{#35bf28}+1.19\%$
test_values_nested_leaf	0.1543ms	79.0898μs	12.6439 KOps/s	12.3814 KOps/s	$\color{#35bf28}+2.12\%$
test_values_stack_nested	0.1678ms	93.3006μs	10.7180 KOps/s	9.5891 KOps/s	$\textbf{\color{#35bf28}+11.77\%}$
test_values_stack_nested_leaf	0.1413ms	78.8265μs	12.6861 KOps/s	12.4319 KOps/s	$\color{#35bf28}+2.04\%$
test_values_stack_nested_locked	0.1803ms	92.9499μs	10.7585 KOps/s	10.5249 KOps/s	$\color{#35bf28}+2.22\%$
test_membership	1.8164μs	0.6987μs	1.4312 MOps/s	1.3017 MOps/s	$\textbf{\color{#35bf28}+9.95\%}$
test_membership_nested	28.0520μs	2.7105μs	368.9320 KOps/s	360.1632 KOps/s	$\color{#35bf28}+2.43\%$
test_membership_nested_leaf	33.7030μs	2.7364μs	365.4418 KOps/s	358.0169 KOps/s	$\color{#35bf28}+2.07\%$
test_membership_stacked_nested	30.0860μs	2.7056μs	369.6009 KOps/s	362.7811 KOps/s	$\color{#35bf28}+1.88\%$
test_membership_stacked_nested_leaf	25.2570μs	2.7046μs	369.7382 KOps/s	350.3620 KOps/s	$\textbf{\color{#35bf28}+5.53\%}$
test_membership_nested_last	25.3770μs	4.3399μs	230.4214 KOps/s	233.4599 KOps/s	$\color{#d91a1a}-1.30\%$
test_membership_nested_leaf_last	48.6920μs	4.2729μs	234.0307 KOps/s	234.1689 KOps/s	$\color{#d91a1a}-0.06\%$
test_membership_stacked_nested_last	30.7970μs	4.2344μs	236.1612 KOps/s	236.0254 KOps/s	$\color{#35bf28}+0.06\%$
test_membership_stacked_nested_leaf_last	41.8680μs	4.2320μs	236.2948 KOps/s	235.1342 KOps/s	$\color{#35bf28}+0.49\%$
test_nested_getleaf	38.8720μs	10.5054μs	95.1889 KOps/s	91.8571 KOps/s	$\color{#35bf28}+3.63\%$
test_nested_get	34.9350μs	10.0232μs	99.7686 KOps/s	96.9353 KOps/s	$\color{#35bf28}+2.92\%$
test_stacked_getleaf	40.6660μs	10.3789μs	96.3495 KOps/s	92.4905 KOps/s	$\color{#35bf28}+4.17\%$
test_stacked_get	29.5460μs	9.8807μs	101.2077 KOps/s	95.8552 KOps/s	$\textbf{\color{#35bf28}+5.58\%}$
test_nested_getitemleaf	40.7160μs	10.8012μs	92.5823 KOps/s	87.6254 KOps/s	$\textbf{\color{#35bf28}+5.66\%}$
test_nested_getitem	32.6220μs	10.1768μs	98.2630 KOps/s	93.4708 KOps/s	$\textbf{\color{#35bf28}+5.13\%}$
test_stacked_getitemleaf	40.3060μs	10.6660μs	93.7556 KOps/s	88.9655 KOps/s	$\textbf{\color{#35bf28}+5.38\%}$
test_stacked_getitem	34.1440μs	10.1615μs	98.4105 KOps/s	94.9043 KOps/s	$\color{#35bf28}+3.69\%$
test_lock_nested	84.7345ms	0.5853ms	1.7084 KOps/s	2.0280 KOps/s	$\textbf{\color{#d91a1a}-15.76\%}$
test_lock_stack_nested	0.7480ms	0.4706ms	2.1251 KOps/s	2.1192 KOps/s	$\color{#35bf28}+0.28\%$
test_unlock_nested	84.1843ms	0.5034ms	1.9864 KOps/s	2.4051 KOps/s	$\textbf{\color{#d91a1a}-17.41\%}$
test_unlock_stack_nested	0.6173ms	0.3857ms	2.5929 KOps/s	2.5741 KOps/s	$\color{#35bf28}+0.73\%$
test_flatten_speed	0.2292ms	99.7233μs	10.0277 KOps/s	9.9678 KOps/s	$\color{#35bf28}+0.60\%$
test_unflatten_speed	0.6603ms	0.5167ms	1.9352 KOps/s	1.9254 KOps/s	$\color{#35bf28}+0.51\%$
test_common_ops	6.6219ms	1.1453ms	873.0968 Ops/s	846.7849 Ops/s	$\color{#35bf28}+3.11\%$
test_creation	19.6970μs	2.1257μs	470.4402 KOps/s	489.7822 KOps/s	$\color{#d91a1a}-3.95\%$
test_creation_empty	56.9770μs	18.9991μs	52.6341 KOps/s	50.2218 KOps/s	$\color{#35bf28}+4.80\%$
test_creation_nested_1	69.2190μs	22.3221μs	44.7987 KOps/s	42.7459 KOps/s	$\color{#35bf28}+4.80\%$
test_creation_nested_2	81.9040μs	26.7397μs	37.3976 KOps/s	35.8504 KOps/s	$\color{#35bf28}+4.32\%$
test_clone	56.4460μs	16.8448μs	59.3655 KOps/s	60.0019 KOps/s	$\color{#d91a1a}-1.06\%$
test_getitem[int]	1.2183ms	16.7785μs	59.6000 KOps/s	60.9349 KOps/s	$\color{#d91a1a}-2.19\%$
test_getitem[slice_int]	0.1867ms	32.6784μs	30.6013 KOps/s	32.5435 KOps/s	$\textbf{\color{#d91a1a}-5.97\%}$
test_getitem[range]	0.1664ms	57.2898μs	17.4551 KOps/s	17.6573 KOps/s	$\color{#d91a1a}-1.15\%$
test_getitem[tuple]	0.1292ms	24.8273μs	40.2783 KOps/s	39.4964 KOps/s	$\color{#35bf28}+1.98\%$
test_getitem[list]	0.1822ms	53.2733μs	18.7711 KOps/s	19.1940 KOps/s	$\color{#d91a1a}-2.20\%$
test_setitem_dim[int]	85.1600μs	33.7832μs	29.6005 KOps/s	30.6127 KOps/s	$\color{#d91a1a}-3.31\%$
test_setitem_dim[slice_int]	0.1155ms	60.2938μs	16.5855 KOps/s	16.3409 KOps/s	$\color{#35bf28}+1.50\%$
test_setitem_dim[range]	0.1228ms	84.0006μs	11.9047 KOps/s	11.9423 KOps/s	$\color{#d91a1a}-0.31\%$
test_setitem_dim[tuple]	87.0430μs	49.9849μs	20.0060 KOps/s	20.0348 KOps/s	$\color{#d91a1a}-0.14\%$
test_setitem	0.1041ms	30.9390μs	32.3216 KOps/s	32.3022 KOps/s	$\color{#35bf28}+0.06\%$
test_set	73.6980μs	30.1152μs	33.2059 KOps/s	32.6935 KOps/s	$\color{#35bf28}+1.57\%$
test_set_shared	3.4884ms	0.2180ms	4.5869 KOps/s	4.6109 KOps/s	$\color{#d91a1a}-0.52\%$
test_update	88.0240μs	38.1768μs	26.1939 KOps/s	24.8336 KOps/s	$\textbf{\color{#35bf28}+5.48\%}$
test_update_nested	0.1204ms	48.8224μs	20.4824 KOps/s	19.6445 KOps/s	$\color{#35bf28}+4.27\%$
test_update__nested	0.9436ms	44.9286μs	22.2576 KOps/s	22.5212 KOps/s	$\color{#d91a1a}-1.17\%$
test_set_nested	79.8600μs	32.5582μs	30.7142 KOps/s	29.6633 KOps/s	$\color{#35bf28}+3.54\%$
test_set_nested_new	89.2770μs	38.1944μs	26.1819 KOps/s	25.6224 KOps/s	$\color{#35bf28}+2.18\%$
test_select	0.1480ms	57.1330μs	17.5030 KOps/s	17.6731 KOps/s	$\color{#d91a1a}-0.96\%$
test_select_nested	0.1312ms	58.7042μs	17.0345 KOps/s	16.6980 KOps/s	$\color{#35bf28}+2.02\%$
test_exclude_nested	0.1701ms	74.5447μs	13.4148 KOps/s	13.2256 KOps/s	$\color{#35bf28}+1.43\%$
test_empty[True]	0.5278ms	0.3522ms	2.8389 KOps/s	2.8633 KOps/s	$\color{#d91a1a}-0.85\%$
test_empty[False]	10.6200μs	1.2143μs	823.5515 KOps/s	721.0087 KOps/s	$\textbf{\color{#35bf28}+14.22\%}$
test_unbind_speed	0.4643ms	0.2984ms	3.3512 KOps/s	3.2952 KOps/s	$\color{#35bf28}+1.70\%$
test_unbind_speed_stack0	0.4719ms	0.2941ms	3.4004 KOps/s	3.4042 KOps/s	$\color{#d91a1a}-0.11\%$
test_unbind_speed_stack1	88.1382ms	0.8018ms	1.2472 KOps/s	1.3411 KOps/s	$\textbf{\color{#d91a1a}-7.00\%}$
test_split	2.2117ms	1.9942ms	501.4432 Ops/s	455.7540 Ops/s	$\textbf{\color{#35bf28}+10.02\%}$
test_chunk	88.5112ms	2.3400ms	427.3574 Ops/s	452.8576 Ops/s	$\textbf{\color{#d91a1a}-5.63\%}$
test_creation[device0]	0.2378ms	0.1157ms	8.6444 KOps/s	8.3383 KOps/s	$\color{#35bf28}+3.67\%$
test_creation_from_tensor	4.1101ms	0.1165ms	8.5855 KOps/s	8.5148 KOps/s	$\color{#35bf28}+0.83\%$
test_add_one[memmap_tensor0]	0.1844ms	7.2142μs	138.6165 KOps/s	137.7844 KOps/s	$\color{#35bf28}+0.60\%$
test_contiguous[memmap_tensor0]	20.2280μs	1.8887μs	529.4776 KOps/s	505.4972 KOps/s	$\color{#35bf28}+4.74\%$
test_stack[memmap_tensor0]	44.5740μs	5.5934μs	178.7828 KOps/s	175.9485 KOps/s	$\color{#35bf28}+1.61\%$
test_memmaptd_index	1.0411ms	0.4046ms	2.4715 KOps/s	2.4529 KOps/s	$\color{#35bf28}+0.76\%$
test_memmaptd_index_astensor	0.8031ms	0.5036ms	1.9855 KOps/s	1.9754 KOps/s	$\color{#35bf28}+0.51\%$
test_memmaptd_index_op	1.6889ms	1.0517ms	950.8416 Ops/s	927.3544 Ops/s	$\color{#35bf28}+2.53\%$
test_serialize_model	0.1257s	0.1180s	8.4716 Ops/s	8.4951 Ops/s	$\color{#d91a1a}-0.28\%$
test_serialize_model_pickle	0.4799s	0.3945s	2.5351 Ops/s	2.4992 Ops/s	$\color{#35bf28}+1.44\%$
test_serialize_weights	0.1276s	0.1160s	8.6221 Ops/s	7.6408 Ops/s	$\textbf{\color{#35bf28}+12.84\%}$
test_serialize_weights_returnearly	0.2468s	0.1714s	5.8359 Ops/s	6.3096 Ops/s	$\textbf{\color{#d91a1a}-7.51\%}$
test_serialize_weights_pickle	0.4638s	0.4063s	2.4611 Ops/s	2.3054 Ops/s	$\textbf{\color{#35bf28}+6.75\%}$
test_serialize_weights_filesystem	0.1552s	0.1426s	7.0112 Ops/s	6.9886 Ops/s	$\color{#35bf28}+0.32\%$
test_serialize_model_filesystem	0.1597s	0.1512s	6.6118 Ops/s	6.5406 Ops/s	$\color{#35bf28}+1.09\%$
test_reshape_pytree	80.9210μs	38.8267μs	25.7555 KOps/s	25.7157 KOps/s	$\color{#35bf28}+0.15\%$
test_reshape_td	0.1057ms	46.9208μs	21.3125 KOps/s	21.5592 KOps/s	$\color{#d91a1a}-1.14\%$
test_view_pytree	83.6370μs	38.6088μs	25.9008 KOps/s	25.8212 KOps/s	$\color{#35bf28}+0.31\%$
test_view_td	0.1228ms	51.8958μs	19.2694 KOps/s	19.3438 KOps/s	$\color{#d91a1a}-0.38\%$
test_unbind_pytree	79.9700μs	35.1158μs	28.4772 KOps/s	27.7581 KOps/s	$\color{#35bf28}+2.59\%$
test_unbind_td	0.2960ms	44.2590μs	22.5943 KOps/s	22.6074 KOps/s	$\color{#d91a1a}-0.06\%$
test_split_pytree	92.2230μs	37.6964μs	26.5277 KOps/s	25.9729 KOps/s	$\color{#35bf28}+2.14\%$
test_split_td	89.8265ms	68.9549μs	14.5022 KOps/s	14.9003 KOps/s	$\color{#d91a1a}-2.67\%$
test_add_pytree	0.1594ms	45.1077μs	22.1692 KOps/s	22.4952 KOps/s	$\color{#d91a1a}-1.45\%$
test_add_td	0.1783ms	88.1836μs	11.3400 KOps/s	10.8403 KOps/s	$\color{#35bf28}+4.61\%$
test_compile_add_one_nested[tensordict-compile]	0.1242ms	58.3411μs	17.1406 KOps/s	16.7433 KOps/s	$\color{#35bf28}+2.37\%$
test_compile_add_one_nested[tensordict-eager]	0.2791ms	0.1945ms	5.1411 KOps/s	5.1915 KOps/s	$\color{#d91a1a}-0.97\%$
test_compile_add_one_nested[pytree-compile]	0.1242ms	57.1694μs	17.4919 KOps/s	17.4729 KOps/s	$\color{#35bf28}+0.11\%$
test_compile_add_one_nested[pytree-eager]	0.2541ms	0.1406ms	7.1103 KOps/s	7.0967 KOps/s	$\color{#35bf28}+0.19\%$
test_compile_copy_nested[tensordict-compile]	76.3730μs	23.0316μs	43.4186 KOps/s	41.9433 KOps/s	$\color{#35bf28}+3.52\%$
test_compile_copy_nested[tensordict-eager]	0.1364ms	73.9945μs	13.5145 KOps/s	13.5204 KOps/s	$\color{#d91a1a}-0.04\%$
test_compile_copy_nested[pytree-compile]	0.1612ms	75.2433μs	13.2902 KOps/s	13.2070 KOps/s	$\color{#35bf28}+0.63\%$
test_compile_copy_nested[pytree-eager]	0.1372ms	67.7897μs	14.7515 KOps/s	14.5365 KOps/s	$\color{#35bf28}+1.48\%$
test_compile_add_one_flat[tensordict-compile]	0.3130ms	0.1816ms	5.5077 KOps/s	5.4791 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_add_one_flat[tensordict-eager]	1.0700ms	0.2424ms	4.1258 KOps/s	4.1133 KOps/s	$\color{#35bf28}+0.30\%$
test_compile_add_one_flat[tensorclass-compile]	0.1136ms	47.7843μs	20.9274 KOps/s	20.5138 KOps/s	$\color{#35bf28}+2.02\%$
test_compile_add_one_flat[tensorclass-eager]	0.4334ms	76.0826μs	13.1436 KOps/s	13.0515 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_add_one_flat[pytree-compile]	0.2529ms	0.1752ms	5.7091 KOps/s	5.7699 KOps/s	$\color{#d91a1a}-1.05\%$
test_compile_add_one_flat[pytree-eager]	0.4440ms	0.2881ms	3.4715 KOps/s	3.5182 KOps/s	$\color{#d91a1a}-1.33\%$
test_compile_add_self_flat[tensordict-eager]	0.3772ms	0.2737ms	3.6532 KOps/s	3.6622 KOps/s	$\color{#d91a1a}-0.25\%$
test_compile_add_self_flat[tensordict-compile]	0.3585ms	0.1852ms	5.3991 KOps/s	5.5376 KOps/s	$\color{#d91a1a}-2.50\%$
test_compile_add_self_flat[tensorclass-eager]	0.9248ms	75.1730μs	13.3026 KOps/s	13.5118 KOps/s	$\color{#d91a1a}-1.55\%$
test_compile_add_self_flat[tensorclass-compile]	0.1217ms	48.0763μs	20.8003 KOps/s	20.3033 KOps/s	$\color{#35bf28}+2.45\%$
test_compile_add_self_flat[pytree-eager]	0.3388ms	0.2319ms	4.3126 KOps/s	4.3552 KOps/s	$\color{#d91a1a}-0.98\%$
test_compile_add_self_flat[pytree-compile]	0.3197ms	0.1771ms	5.6472 KOps/s	5.6579 KOps/s	$\color{#d91a1a}-0.19\%$
test_compile_copy_flat[tensordict-compile]	0.2063ms	0.1127ms	8.8692 KOps/s	9.0484 KOps/s	$\color{#d91a1a}-1.98\%$
test_compile_copy_flat[tensordict-eager]	0.1541ms	79.9095μs	12.5142 KOps/s	12.8533 KOps/s	$\color{#d91a1a}-2.64\%$
test_compile_copy_flat[pytree-compile]	0.1392ms	76.4839μs	13.0746 KOps/s	12.5282 KOps/s	$\color{#35bf28}+4.36\%$
test_compile_copy_flat[pytree-eager]	0.1542ms	68.4037μs	14.6191 KOps/s	14.1098 KOps/s	$\color{#35bf28}+3.61\%$
test_compile_assign_and_add[tensordict-compile]	0.3593ms	0.1923ms	5.2013 KOps/s	5.2408 KOps/s	$\color{#d91a1a}-0.75\%$
test_compile_assign_and_add[tensordict-eager]	2.7401ms	1.6843ms	593.7020 Ops/s	563.1431 Ops/s	$\textbf{\color{#35bf28}+5.43\%}$
test_compile_assign_and_add[pytree-compile]	0.2464ms	0.1898ms	5.2694 KOps/s	5.1743 KOps/s	$\color{#35bf28}+1.84\%$
test_compile_assign_and_add[pytree-eager]	1.3212ms	1.0797ms	926.2034 Ops/s	906.0108 Ops/s	$\color{#35bf28}+2.23\%$
test_compile_assign_and_add_stack[compile]	0.7593ms	0.4131ms	2.4209 KOps/s	2.4561 KOps/s	$\color{#d91a1a}-1.43\%$
test_compile_assign_and_add_stack[eager]	6.0357ms	4.0423ms	247.3860 Ops/s	233.8505 Ops/s	$\textbf{\color{#35bf28}+5.79\%}$
test_compile_indexing[tensor-tensordict-compile]	89.4980μs	33.6915μs	29.6810 KOps/s	28.9459 KOps/s	$\color{#35bf28}+2.54\%$
test_compile_indexing[tensor-tensordict-eager]	1.0385ms	47.9238μs	20.8665 KOps/s	20.7659 KOps/s	$\color{#35bf28}+0.48\%$
test_compile_indexing[tensor-tensorclass-compile]	72.4350μs	30.0759μs	33.2492 KOps/s	32.2785 KOps/s	$\color{#35bf28}+3.01\%$
test_compile_indexing[tensor-tensorclass-eager]	77.9960μs	28.9538μs	34.5378 KOps/s	34.3578 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_indexing[tensor-pytree-compile]	78.3770μs	29.7954μs	33.5622 KOps/s	32.8341 KOps/s	$\color{#35bf28}+2.22\%$
test_compile_indexing[tensor-pytree-eager]	78.9080μs	29.0014μs	34.4811 KOps/s	34.5129 KOps/s	$\color{#d91a1a}-0.09\%$
test_compile_indexing[slice-tensordict-compile]	0.1704ms	74.0510μs	13.5042 KOps/s	13.2776 KOps/s	$\color{#35bf28}+1.71\%$
test_compile_indexing[slice-tensordict-eager]	0.5560ms	27.4642μs	36.4111 KOps/s	36.8001 KOps/s	$\color{#d91a1a}-1.06\%$
test_compile_indexing[slice-tensorclass-compile]	0.1495ms	68.5955μs	14.5782 KOps/s	14.3800 KOps/s	$\color{#35bf28}+1.38\%$
test_compile_indexing[slice-tensorclass-eager]	79.9620μs	23.2499μs	43.0109 KOps/s	42.6981 KOps/s	$\color{#35bf28}+0.73\%$
test_compile_indexing[slice-pytree-compile]	0.1393ms	68.1877μs	14.6654 KOps/s	14.4760 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_indexing[slice-pytree-eager]	67.7460μs	23.5081μs	42.5386 KOps/s	42.3670 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_indexing[int-tensordict-compile]	0.1704ms	73.4805μs	13.6091 KOps/s	13.3997 KOps/s	$\color{#35bf28}+1.56\%$
test_compile_indexing[int-tensordict-eager]	0.8180ms	27.6123μs	36.2158 KOps/s	37.1174 KOps/s	$\color{#d91a1a}-2.43\%$
test_compile_indexing[int-tensorclass-compile]	0.1464ms	68.5861μs	14.5802 KOps/s	14.4345 KOps/s	$\color{#35bf28}+1.01\%$
test_compile_indexing[int-tensorclass-eager]	56.2350μs	23.1873μs	43.1271 KOps/s	43.0572 KOps/s	$\color{#35bf28}+0.16\%$
test_compile_indexing[int-pytree-compile]	0.4411ms	68.5328μs	14.5916 KOps/s	14.5958 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_indexing[int-pytree-eager]	86.1010μs	23.0814μs	43.3249 KOps/s	43.1787 KOps/s	$\color{#35bf28}+0.34\%$
test_mod_add[eager]	70.7330μs	26.4841μs	37.7585 KOps/s	37.7909 KOps/s	$\color{#d91a1a}-0.09\%$
test_mod_add[compile]	98.6050μs	37.3366μs	26.7834 KOps/s	25.9009 KOps/s	$\color{#35bf28}+3.41\%$
test_mod_add[compile-overhead]	0.1061ms	37.7893μs	26.4625 KOps/s	25.9674 KOps/s	$\color{#35bf28}+1.91\%$
test_mod_wrap[eager]	0.3534ms	0.2068ms	4.8355 KOps/s	4.8068 KOps/s	$\color{#35bf28}+0.60\%$
test_mod_wrap[compile]	0.3164ms	0.2279ms	4.3872 KOps/s	4.3053 KOps/s	$\color{#35bf28}+1.90\%$
test_mod_wrap[compile-overhead]	0.3118ms	0.2284ms	4.3781 KOps/s	4.3093 KOps/s	$\color{#35bf28}+1.60\%$
test_mod_wrap_and_backward[eager]	12.8119ms	11.1678ms	89.5430 Ops/s	92.7169 Ops/s	$\color{#d91a1a}-3.42\%$
test_mod_wrap_and_backward[compile]	15.6033ms	12.1785ms	82.1117 Ops/s	92.6137 Ops/s	$\textbf{\color{#d91a1a}-11.34\%}$
test_mod_wrap_and_backward[compile-overhead]	14.6949ms	12.2775ms	81.4500 Ops/s	92.6182 Ops/s	$\textbf{\color{#d91a1a}-12.06\%}$
test_seq_add[eager]	0.1739ms	95.7190μs	10.4472 KOps/s	10.5615 KOps/s	$\color{#d91a1a}-1.08\%$
test_seq_add[compile]	0.1428ms	64.1545μs	15.5874 KOps/s	15.4522 KOps/s	$\color{#35bf28}+0.87\%$
test_seq_add[compile-overhead]	0.1290ms	62.3782μs	16.0312 KOps/s	15.6807 KOps/s	$\color{#35bf28}+2.24\%$
test_seq_wrap[eager]	0.7352ms	0.3893ms	2.5688 KOps/s	2.4606 KOps/s	$\color{#35bf28}+4.40\%$
test_seq_wrap[compile]	0.4866ms	0.2699ms	3.7053 KOps/s	3.6662 KOps/s	$\color{#35bf28}+1.07\%$
test_seq_wrap[compile-overhead]	0.4112ms	0.2703ms	3.6996 KOps/s	3.6868 KOps/s	$\color{#35bf28}+0.35\%$
test_func_call_runtime[False-eager]	0.8463ms	0.5130ms	1.9494 KOps/s	1.9161 KOps/s	$\color{#35bf28}+1.74\%$
test_func_call_runtime[False-compile]	0.6115ms	0.4968ms	2.0129 KOps/s	1.9857 KOps/s	$\color{#35bf28}+1.37\%$
test_func_call_runtime[False-compile-overhead]	1.0378ms	0.5020ms	1.9922 KOps/s	2.0088 KOps/s	$\color{#d91a1a}-0.83\%$
test_func_call_runtime[True-eager]	0.8752ms	0.7365ms	1.3578 KOps/s	1.3636 KOps/s	$\color{#d91a1a}-0.42\%$
test_func_call_runtime[True-compile]	0.8691ms	0.5193ms	1.9256 KOps/s	1.8709 KOps/s	$\color{#35bf28}+2.92\%$
test_func_call_runtime[True-compile-overhead]	0.8858ms	0.5137ms	1.9466 KOps/s	1.9652 KOps/s	$\color{#d91a1a}-0.95\%$
test_func_call_cm_runtime[False-eager]	1.1536ms	0.5105ms	1.9590 KOps/s	1.9495 KOps/s	$\color{#35bf28}+0.49\%$
test_func_call_cm_runtime[False-compile]	1.0388ms	0.5092ms	1.9638 KOps/s	1.9835 KOps/s	$\color{#d91a1a}-0.99\%$
test_func_call_cm_runtime[False-compile-overhead]	0.6275ms	0.4995ms	2.0020 KOps/s	1.9805 KOps/s	$\color{#35bf28}+1.08\%$
test_func_call_cm_runtime[True-eager]	1.0937ms	0.8841ms	1.1312 KOps/s	1.1333 KOps/s	$\color{#d91a1a}-0.19\%$
test_func_call_cm_runtime[True-compile]	1.0171ms	0.7303ms	1.3693 KOps/s	1.3594 KOps/s	$\color{#35bf28}+0.73\%$
test_func_call_cm_runtime[True-compile-overhead]	0.8722ms	0.7276ms	1.3743 KOps/s	1.3502 KOps/s	$\color{#35bf28}+1.79\%$
test_vmap_func_call_cm_runtime[eager]	2.8229ms	1.8973ms	527.0728 Ops/s	520.2379 Ops/s	$\color{#35bf28}+1.31\%$
test_vmap_func_call_cm_runtime[compile]	2.5925ms	1.9235ms	519.8859 Ops/s	507.6825 Ops/s	$\color{#35bf28}+2.40\%$
test_vmap_func_call_cm_runtime[compile-overhead]	2.8626ms	1.9297ms	518.2069 Ops/s	507.9333 Ops/s	$\color{#35bf28}+2.02\%$
test_distributed	0.2771ms	0.1277ms	7.8292 KOps/s	7.6850 KOps/s	$\color{#35bf28}+1.88\%$
test_tdmodule	46.8070μs	19.3110μs	51.7839 KOps/s	53.2886 KOps/s	$\color{#d91a1a}-2.82\%$
test_tdmodule_dispatch	76.8740μs	37.7396μs	26.4974 KOps/s	25.5024 KOps/s	$\color{#35bf28}+3.90\%$
test_tdseq	37.1400μs	21.0097μs	47.5970 KOps/s	44.0399 KOps/s	$\textbf{\color{#35bf28}+8.08\%}$
test_tdseq_dispatch	67.2260μs	42.7828μs	23.3739 KOps/s	22.0312 KOps/s	$\textbf{\color{#35bf28}+6.09\%}$
test_instantiation_functorch	1.8423ms	1.5883ms	629.6023 Ops/s	627.7135 Ops/s	$\color{#35bf28}+0.30\%$
test_exec_functorch	0.4096ms	0.1855ms	5.3894 KOps/s	5.4128 KOps/s	$\color{#d91a1a}-0.43\%$
test_exec_functional_call	0.2844ms	0.1725ms	5.7964 KOps/s	5.7631 KOps/s	$\color{#35bf28}+0.58\%$
test_exec_td_decorator	0.4734ms	0.2339ms	4.2747 KOps/s	4.2396 KOps/s	$\color{#35bf28}+0.83\%$
test_vmap_mlp_speed_decorator[True-True]	0.9960ms	0.6387ms	1.5656 KOps/s	1.5376 KOps/s	$\color{#35bf28}+1.83\%$
test_vmap_mlp_speed_decorator[True-False]	1.0174ms	0.6488ms	1.5414 KOps/s	1.5442 KOps/s	$\color{#d91a1a}-0.18\%$
test_vmap_mlp_speed_decorator[False-True]	0.9311ms	0.5220ms	1.9156 KOps/s	1.8758 KOps/s	$\color{#35bf28}+2.12\%$
test_vmap_mlp_speed_decorator[False-False]	0.7828ms	0.5219ms	1.9162 KOps/s	1.8483 KOps/s	$\color{#35bf28}+3.68\%$
test_to_module_speed[True]	1.9533ms	1.4075ms	710.4713 Ops/s	699.6878 Ops/s	$\color{#35bf28}+1.54\%$
test_to_module_speed[False]	1.4813ms	1.3737ms	727.9630 Ops/s	724.0173 Ops/s	$\color{#35bf28}+0.54\%$
test_tc_init	80.6110μs	47.4070μs	21.0939 KOps/s	20.2308 KOps/s	$\color{#35bf28}+4.27\%$
test_tc_init_nested	0.1645ms	95.5965μs	10.4606 KOps/s	10.1395 KOps/s	$\color{#35bf28}+3.17\%$
test_tc_first_layer_tensor	19.8170μs	1.5622μs	640.1326 KOps/s	656.1862 KOps/s	$\color{#d91a1a}-2.45\%$
test_tc_first_layer_nontensor	41.9290μs	4.7240μs	211.6866 KOps/s	213.7321 KOps/s	$\color{#d91a1a}-0.96\%$
test_tc_second_layer_tensor	30.1860μs	2.8603μs	349.6149 KOps/s	355.6859 KOps/s	$\color{#d91a1a}-1.71\%$
test_tc_second_layer_nontensor	47.9800μs	6.1201μs	163.3952 KOps/s	164.6178 KOps/s	$\color{#d91a1a}-0.74\%$
test_unbind	7.7313ms	7.4730ms	133.8145 Ops/s	75.2138 Ops/s	$\textbf{\color{#35bf28}+77.91\%}$
test_full_like	17.9703ms	11.4354ms	87.4475 Ops/s	140.6018 Ops/s	$\textbf{\color{#d91a1a}-37.80\%}$
test_zeros_like	12.4020ms	7.6559ms	130.6185 Ops/s	361.4289 Ops/s	$\textbf{\color{#d91a1a}-63.86\%}$
test_ones_like	14.4237ms	7.6713ms	130.3552 Ops/s	167.2743 Ops/s	$\textbf{\color{#d91a1a}-22.07\%}$
test_clone	17.8084ms	9.2173ms	108.4919 Ops/s	127.4136 Ops/s	$\textbf{\color{#d91a1a}-14.85\%}$
test_squeeze	48.2500μs	12.8555μs	77.7876 KOps/s	80.8392 KOps/s	$\color{#d91a1a}-3.77\%$
test_unsqueeze	0.1871ms	92.0212μs	10.8671 KOps/s	10.7993 KOps/s	$\color{#35bf28}+0.63\%$
test_split	0.4278ms	0.1923ms	5.2014 KOps/s	5.0656 KOps/s	$\color{#35bf28}+2.68\%$
test_permute	0.4216ms	0.2214ms	4.5157 KOps/s	4.5189 KOps/s	$\color{#d91a1a}-0.07\%$
test_stack	27.8654ms	25.1260ms	39.7994 Ops/s	38.9083 Ops/s	$\color{#35bf28}+2.29\%$
test_cat	29.6801ms	25.3070ms	39.5147 Ops/s	38.6441 Ops/s	$\color{#35bf28}+2.25\%$

[github-actions]

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

Total Benchmarks: 218. Improved: $\large\color{#35bf28}9$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1483ms	18.0638μs	55.3593 KOps/s	56.4384 KOps/s	$\color{#d91a1a}-1.91\%$
test_plain_set_stack_nested	43.1810μs	17.9559μs	55.6920 KOps/s	56.3291 KOps/s	$\color{#d91a1a}-1.13\%$
test_plain_set_nested_inplace	46.4310μs	19.2979μs	51.8191 KOps/s	52.6664 KOps/s	$\color{#d91a1a}-1.61\%$
test_plain_set_stack_nested_inplace	48.4310μs	19.3663μs	51.6360 KOps/s	52.3707 KOps/s	$\color{#d91a1a}-1.40\%$
test_items	26.7700μs	2.9039μs	344.3629 KOps/s	342.8539 KOps/s	$\color{#35bf28}+0.44\%$
test_items_nested	0.3844ms	0.3410ms	2.9327 KOps/s	2.9340 KOps/s	$\color{#d91a1a}-0.04\%$
test_items_nested_locked	0.3703ms	0.3463ms	2.8876 KOps/s	2.9404 KOps/s	$\color{#d91a1a}-1.79\%$
test_items_nested_leaf	91.8820μs	63.4437μs	15.7620 KOps/s	15.9183 KOps/s	$\color{#d91a1a}-0.98\%$
test_items_stack_nested	0.3941ms	0.3413ms	2.9296 KOps/s	2.9426 KOps/s	$\color{#d91a1a}-0.44\%$
test_items_stack_nested_leaf	96.3220μs	64.3499μs	15.5401 KOps/s	15.5431 KOps/s	$\color{#d91a1a}-0.02\%$
test_items_stack_nested_locked	0.4267ms	0.3426ms	2.9191 KOps/s	2.9379 KOps/s	$\color{#d91a1a}-0.64\%$
test_keys	24.2810μs	3.4670μs	288.4376 KOps/s	292.5903 KOps/s	$\color{#d91a1a}-1.42\%$
test_keys_nested	0.1103ms	70.9496μs	14.0945 KOps/s	14.1196 KOps/s	$\color{#d91a1a}-0.18\%$
test_keys_nested_locked	2.2955ms	76.2961μs	13.1068 KOps/s	12.9725 KOps/s	$\color{#35bf28}+1.04\%$
test_keys_nested_leaf	93.9120μs	61.6013μs	16.2334 KOps/s	16.2090 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_stack_nested	0.1046ms	71.4201μs	14.0017 KOps/s	13.8287 KOps/s	$\color{#35bf28}+1.25\%$
test_keys_stack_nested_leaf	96.6520μs	62.9852μs	15.8767 KOps/s	15.6656 KOps/s	$\color{#35bf28}+1.35\%$
test_keys_stack_nested_locked	0.1209ms	77.2678μs	12.9420 KOps/s	12.8116 KOps/s	$\color{#35bf28}+1.02\%$
test_values	5.3017μs	0.8405μs	1.1897 MOps/s	1.1889 MOps/s	$\color{#35bf28}+0.06\%$
test_values_nested	88.6620μs	49.0464μs	20.3889 KOps/s	20.3251 KOps/s	$\color{#35bf28}+0.31\%$
test_values_nested_locked	85.7310μs	50.6971μs	19.7250 KOps/s	19.2980 KOps/s	$\color{#35bf28}+2.21\%$
test_values_nested_leaf	83.7610μs	42.7692μs	23.3813 KOps/s	23.3987 KOps/s	$\color{#d91a1a}-0.07\%$
test_values_stack_nested	80.5320μs	49.8860μs	20.0457 KOps/s	19.9321 KOps/s	$\color{#35bf28}+0.57\%$
test_values_stack_nested_leaf	81.3820μs	43.4116μs	23.0353 KOps/s	22.6656 KOps/s	$\color{#35bf28}+1.63\%$
test_values_stack_nested_locked	85.3710μs	51.3638μs	19.4690 KOps/s	19.0426 KOps/s	$\color{#35bf28}+2.24\%$
test_membership	1.8181μs	0.5016μs	1.9937 MOps/s	1.9811 MOps/s	$\color{#35bf28}+0.64\%$
test_membership_nested	15.8405μs	1.9264μs	519.1120 KOps/s	528.1899 KOps/s	$\color{#d91a1a}-1.72\%$
test_membership_nested_leaf	23.3105μs	1.9000μs	526.3089 KOps/s	525.1248 KOps/s	$\color{#35bf28}+0.23\%$
test_membership_stacked_nested	27.5610μs	2.0157μs	496.1019 KOps/s	518.7334 KOps/s	$\color{#d91a1a}-4.36\%$
test_membership_stacked_nested_leaf	30.1310μs	2.0069μs	498.2852 KOps/s	514.5171 KOps/s	$\color{#d91a1a}-3.15\%$
test_membership_nested_last	82.4230μs	2.9976μs	333.6052 KOps/s	335.3111 KOps/s	$\color{#d91a1a}-0.51\%$
test_membership_nested_leaf_last	26.9010μs	3.0794μs	324.7434 KOps/s	336.1542 KOps/s	$\color{#d91a1a}-3.39\%$
test_membership_stacked_nested_last	33.2710μs	3.1007μs	322.5100 KOps/s	337.1262 KOps/s	$\color{#d91a1a}-4.34\%$
test_membership_stacked_nested_leaf_last	29.8510μs	3.0372μs	329.2500 KOps/s	337.6696 KOps/s	$\color{#d91a1a}-2.49\%$
test_nested_getleaf	31.4310μs	6.0319μs	165.7840 KOps/s	164.1809 KOps/s	$\color{#35bf28}+0.98\%$
test_nested_get	34.6210μs	5.6968μs	175.5386 KOps/s	171.2446 KOps/s	$\color{#35bf28}+2.51\%$
test_stacked_getleaf	32.9910μs	6.0009μs	166.6429 KOps/s	167.2311 KOps/s	$\color{#d91a1a}-0.35\%$
test_stacked_get	35.3210μs	5.6379μs	177.3708 KOps/s	175.4768 KOps/s	$\color{#35bf28}+1.08\%$
test_nested_getitemleaf	31.2410μs	6.1630μs	162.2579 KOps/s	162.7275 KOps/s	$\color{#d91a1a}-0.29\%$
test_nested_getitem	28.5810μs	5.7762μs	173.1250 KOps/s	171.7771 KOps/s	$\color{#35bf28}+0.78\%$
test_stacked_getitemleaf	39.6100μs	6.0820μs	164.4202 KOps/s	165.9381 KOps/s	$\color{#d91a1a}-0.91\%$
test_stacked_getitem	33.8000μs	5.7734μs	173.2078 KOps/s	176.9318 KOps/s	$\color{#d91a1a}-2.10\%$
test_lock_nested	4.8905ms	0.4332ms	2.3086 KOps/s	2.3675 KOps/s	$\color{#d91a1a}-2.49\%$
test_lock_stack_nested	0.4290ms	0.3955ms	2.5283 KOps/s	2.5763 KOps/s	$\color{#d91a1a}-1.86\%$
test_unlock_nested	0.7579ms	0.3670ms	2.7247 KOps/s	2.7813 KOps/s	$\color{#d91a1a}-2.04\%$
test_unlock_stack_nested	0.3782ms	0.3335ms	2.9986 KOps/s	3.0637 KOps/s	$\color{#d91a1a}-2.12\%$
test_flatten_speed	0.1061ms	77.1517μs	12.9615 KOps/s	12.9426 KOps/s	$\color{#35bf28}+0.15\%$
test_unflatten_speed	0.3806ms	0.3276ms	3.0523 KOps/s	3.0821 KOps/s	$\color{#d91a1a}-0.97\%$
test_common_ops	1.6404ms	1.3112ms	762.6332 Ops/s	779.5054 Ops/s	$\color{#d91a1a}-2.16\%$
test_creation	24.8200μs	1.4988μs	667.1844 KOps/s	684.7589 KOps/s	$\color{#d91a1a}-2.57\%$
test_creation_empty	62.9610μs	17.9553μs	55.6939 KOps/s	57.5964 KOps/s	$\color{#d91a1a}-3.30\%$
test_creation_nested_1	53.2300μs	19.5961μs	51.0306 KOps/s	51.3939 KOps/s	$\color{#d91a1a}-0.71\%$
test_creation_nested_2	1.2094ms	23.1743μs	43.1513 KOps/s	45.7836 KOps/s	$\textbf{\color{#d91a1a}-5.75\%}$
test_clone	62.4210μs	29.0315μs	34.4454 KOps/s	35.0890 KOps/s	$\color{#d91a1a}-1.83\%$
test_getitem[int]	1.1914ms	15.7246μs	63.5948 KOps/s	65.0397 KOps/s	$\color{#d91a1a}-2.22\%$
test_getitem[slice_int]	91.7979ms	37.7283μs	26.5053 KOps/s	37.5638 KOps/s	$\textbf{\color{#d91a1a}-29.44\%}$
test_getitem[range]	0.1525ms	0.1082ms	9.2403 KOps/s	9.0429 KOps/s	$\color{#35bf28}+2.18\%$
test_getitem[tuple]	0.1196ms	23.6326μs	42.3145 KOps/s	43.1940 KOps/s	$\color{#d91a1a}-2.04\%$
test_getitem[list]	0.1958ms	99.4508μs	10.0552 KOps/s	10.0924 KOps/s	$\color{#d91a1a}-0.37\%$
test_setitem_dim[int]	85.9420μs	44.6245μs	22.4092 KOps/s	22.3615 KOps/s	$\color{#35bf28}+0.21\%$
test_setitem_dim[slice_int]	92.8410μs	66.1024μs	15.1280 KOps/s	15.1605 KOps/s	$\color{#d91a1a}-0.21\%$
test_setitem_dim[range]	0.1775ms	0.1271ms	7.8698 KOps/s	7.8958 KOps/s	$\color{#d91a1a}-0.33\%$
test_setitem_dim[tuple]	98.7210μs	60.3583μs	16.5677 KOps/s	16.5362 KOps/s	$\color{#35bf28}+0.19\%$
test_setitem	86.5720μs	43.7005μs	22.8831 KOps/s	23.4001 KOps/s	$\color{#d91a1a}-2.21\%$
test_set	88.8020μs	42.2651μs	23.6602 KOps/s	24.2706 KOps/s	$\color{#d91a1a}-2.51\%$
test_set_shared	0.3550ms	53.7504μs	18.6045 KOps/s	18.6582 KOps/s	$\color{#d91a1a}-0.29\%$
test_update	93.7220μs	53.4180μs	18.7203 KOps/s	19.3506 KOps/s	$\color{#d91a1a}-3.26\%$
test_update_nested	0.1083ms	60.5287μs	16.5211 KOps/s	16.9179 KOps/s	$\color{#d91a1a}-2.35\%$
test_update__nested	0.4234ms	66.3846μs	15.0637 KOps/s	16.4703 KOps/s	$\textbf{\color{#d91a1a}-8.54\%}$
test_set_nested	80.9620μs	45.0522μs	22.1965 KOps/s	22.3654 KOps/s	$\color{#d91a1a}-0.76\%$
test_set_nested_new	90.3720μs	49.2572μs	20.3016 KOps/s	20.8345 KOps/s	$\color{#d91a1a}-2.56\%$
test_select	0.1033ms	61.9745μs	16.1357 KOps/s	16.0731 KOps/s	$\color{#35bf28}+0.39\%$
test_select_nested	0.2048ms	41.6282μs	24.0222 KOps/s	23.7256 KOps/s	$\color{#35bf28}+1.25\%$
test_exclude_nested	0.1055ms	59.2315μs	16.8829 KOps/s	16.7479 KOps/s	$\color{#35bf28}+0.81\%$
test_empty[True]	0.2944ms	0.2589ms	3.8625 KOps/s	3.8529 KOps/s	$\color{#35bf28}+0.25\%$
test_empty[False]	3.0370μs	0.7381μs	1.3549 MOps/s	1.3588 MOps/s	$\color{#d91a1a}-0.29\%$
test_to	54.7610μs	26.8923μs	37.1853 KOps/s	37.2858 KOps/s	$\color{#d91a1a}-0.27\%$
test_to_nonblocking	66.8810μs	25.7167μs	38.8853 KOps/s	38.5008 KOps/s	$\color{#35bf28}+1.00\%$
test_unbind_speed	1.0681ms	0.2756ms	3.6282 KOps/s	3.6618 KOps/s	$\color{#d91a1a}-0.92\%$
test_unbind_speed_stack0	0.3259ms	0.2781ms	3.5957 KOps/s	3.6308 KOps/s	$\color{#d91a1a}-0.97\%$
test_unbind_speed_stack1	91.2836ms	0.7101ms	1.4082 KOps/s	1.4036 KOps/s	$\color{#35bf28}+0.33\%$
test_split	92.6550ms	2.1243ms	470.7334 Ops/s	473.1984 Ops/s	$\color{#d91a1a}-0.52\%$
test_chunk	94.7604ms	2.1241ms	470.7960 Ops/s	471.7039 Ops/s	$\color{#d91a1a}-0.19\%$
test_creation[device0]	0.3820ms	0.1266ms	7.8996 KOps/s	7.9174 KOps/s	$\color{#d91a1a}-0.22\%$
test_creation_from_tensor	0.4225ms	0.1289ms	7.7560 KOps/s	7.7836 KOps/s	$\color{#d91a1a}-0.35\%$
test_add_one[memmap_tensor0]	0.1425ms	9.0299μs	110.7432 KOps/s	116.5694 KOps/s	$\color{#d91a1a}-5.00\%$
test_contiguous[memmap_tensor0]	30.7710μs	2.1336μs	468.6900 KOps/s	466.7331 KOps/s	$\color{#35bf28}+0.42\%$
test_stack[memmap_tensor0]	36.1510μs	6.2260μs	160.6163 KOps/s	157.4833 KOps/s	$\color{#35bf28}+1.99\%$
test_memmaptd_index	1.0559ms	0.4189ms	2.3870 KOps/s	2.3758 KOps/s	$\color{#35bf28}+0.47\%$
test_memmaptd_index_astensor	0.7417ms	0.4896ms	2.0423 KOps/s	2.0267 KOps/s	$\color{#35bf28}+0.77\%$
test_memmaptd_index_op	1.4813ms	1.0797ms	926.2164 Ops/s	957.9826 Ops/s	$\color{#d91a1a}-3.32\%$
test_serialize_model	0.1311s	0.1298s	7.7054 Ops/s	7.6687 Ops/s	$\color{#35bf28}+0.48\%$
test_serialize_model_pickle	1.3520s	1.2128s	0.8246 Ops/s	0.8233 Ops/s	$\color{#35bf28}+0.15\%$
test_serialize_weights	0.2225s	0.1426s	7.0116 Ops/s	6.9884 Ops/s	$\color{#35bf28}+0.33\%$
test_serialize_weights_returnearly	0.2318s	56.1929ms	17.7959 Ops/s	17.8920 Ops/s	$\color{#d91a1a}-0.54\%$
test_serialize_weights_pickle	1.3818s	1.2212s	0.8189 Ops/s	0.8381 Ops/s	$\color{#d91a1a}-2.29\%$
test_reshape_pytree	79.5010μs	37.3009μs	26.8090 KOps/s	28.0037 KOps/s	$\color{#d91a1a}-4.27\%$
test_reshape_td	81.3820μs	42.0655μs	23.7725 KOps/s	25.0742 KOps/s	$\textbf{\color{#d91a1a}-5.19\%}$
test_view_pytree	72.7720μs	37.1674μs	26.9053 KOps/s	28.3346 KOps/s	$\textbf{\color{#d91a1a}-5.04\%}$
test_view_td	82.0110μs	49.2171μs	20.3181 KOps/s	21.6453 KOps/s	$\textbf{\color{#d91a1a}-6.13\%}$
test_unbind_pytree	64.4810μs	33.6881μs	29.6841 KOps/s	29.4779 KOps/s	$\color{#35bf28}+0.70\%$
test_unbind_td	0.4217ms	43.3084μs	23.0902 KOps/s	23.6345 KOps/s	$\color{#d91a1a}-2.30\%$
test_split_pytree	83.6620μs	46.0459μs	21.7175 KOps/s	21.4592 KOps/s	$\color{#35bf28}+1.20\%$
test_split_td	0.7109ms	58.4007μs	17.1231 KOps/s	16.0120 KOps/s	$\textbf{\color{#35bf28}+6.94\%}$
test_add_pytree	0.1043ms	62.1235μs	16.0970 KOps/s	17.6801 KOps/s	$\textbf{\color{#d91a1a}-8.95\%}$
test_add_td	0.1533ms	0.1051ms	9.5174 KOps/s	10.6454 KOps/s	$\textbf{\color{#d91a1a}-10.60\%}$
test_compile_add_one_nested[tensordict-compile]	0.2372ms	0.1597ms	6.2605 KOps/s	6.3133 KOps/s	$\color{#d91a1a}-0.84\%$
test_compile_add_one_nested[tensordict-eager]	0.5666ms	0.1623ms	6.1621 KOps/s	6.2314 KOps/s	$\color{#d91a1a}-1.11\%$
test_compile_add_one_nested[pytree-compile]	0.5688ms	0.1522ms	6.5704 KOps/s	6.5319 KOps/s	$\color{#35bf28}+0.59\%$
test_compile_add_one_nested[pytree-eager]	0.5975ms	0.1864ms	5.3641 KOps/s	5.4787 KOps/s	$\color{#d91a1a}-2.09\%$
test_compile_copy_nested[tensordict-compile]	73.7120μs	21.1682μs	47.2408 KOps/s	47.1951 KOps/s	$\color{#35bf28}+0.10\%$
test_compile_copy_nested[tensordict-eager]	93.0920μs	48.4568μs	20.6369 KOps/s	20.3624 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_copy_nested[pytree-compile]	0.4139ms	65.0099μs	15.3823 KOps/s	15.6214 KOps/s	$\color{#d91a1a}-1.53\%$
test_compile_copy_nested[pytree-eager]	97.2210μs	49.8361μs	20.0658 KOps/s	20.1228 KOps/s	$\color{#d91a1a}-0.28\%$
test_compile_add_one_flat[tensordict-compile]	0.4082ms	0.3134ms	3.1906 KOps/s	3.1910 KOps/s	$\color{#d91a1a}-0.01\%$
test_compile_add_one_flat[tensordict-eager]	0.3515ms	0.2323ms	4.3052 KOps/s	4.2792 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_add_one_flat[tensorclass-compile]	0.2092ms	0.1256ms	7.9634 KOps/s	7.9525 KOps/s	$\color{#35bf28}+0.14\%$
test_compile_add_one_flat[tensorclass-eager]	0.1270ms	65.3582μs	15.3003 KOps/s	15.3143 KOps/s	$\color{#d91a1a}-0.09\%$
test_compile_add_one_flat[pytree-compile]	0.3861ms	0.3210ms	3.1152 KOps/s	3.1258 KOps/s	$\color{#d91a1a}-0.34\%$
test_compile_add_one_flat[pytree-eager]	1.0086ms	0.6313ms	1.5841 KOps/s	1.6416 KOps/s	$\color{#d91a1a}-3.50\%$
test_compile_add_self_flat[tensordict-eager]	0.6768ms	0.2837ms	3.5252 KOps/s	3.5211 KOps/s	$\color{#35bf28}+0.12\%$
test_compile_add_self_flat[tensordict-compile]	0.7371ms	0.3145ms	3.1796 KOps/s	3.1920 KOps/s	$\color{#d91a1a}-0.39\%$
test_compile_add_self_flat[tensorclass-eager]	0.1589ms	77.1733μs	12.9578 KOps/s	13.0189 KOps/s	$\color{#d91a1a}-0.47\%$
test_compile_add_self_flat[tensorclass-compile]	0.1716ms	0.1261ms	7.9289 KOps/s	7.9253 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_add_self_flat[pytree-eager]	0.9306ms	0.5272ms	1.8968 KOps/s	1.9488 KOps/s	$\color{#d91a1a}-2.67\%$
test_compile_add_self_flat[pytree-compile]	0.4604ms	0.3214ms	3.1116 KOps/s	3.1063 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_copy_flat[tensordict-compile]	63.3710μs	18.8214μs	53.1310 KOps/s	51.8708 KOps/s	$\color{#35bf28}+2.43\%$
test_compile_copy_flat[tensordict-eager]	86.6210μs	38.1913μs	26.1840 KOps/s	26.0002 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_copy_flat[pytree-compile]	0.1202ms	70.5995μs	14.1644 KOps/s	14.3671 KOps/s	$\color{#d91a1a}-1.41\%$
test_compile_copy_flat[pytree-eager]	0.1065ms	52.1177μs	19.1873 KOps/s	19.3521 KOps/s	$\color{#d91a1a}-0.85\%$
test_compile_assign_and_add[tensordict-compile]	2.3720ms	0.8192ms	1.2206 KOps/s	1.1335 KOps/s	$\textbf{\color{#35bf28}+7.68\%}$
test_compile_assign_and_add[tensordict-eager]	3.5613ms	3.2035ms	312.1579 Ops/s	310.7484 Ops/s	$\color{#35bf28}+0.45\%$
test_compile_assign_and_add[pytree-compile]	2.4265ms	0.8335ms	1.1998 KOps/s	1.1188 KOps/s	$\textbf{\color{#35bf28}+7.24\%}$
test_compile_assign_and_add[pytree-eager]	3.4542ms	3.2175ms	310.8048 Ops/s	319.5724 Ops/s	$\color{#d91a1a}-2.74\%$
test_compile_indexing[tensor-tensordict-compile]	0.2143ms	0.1160ms	8.6242 KOps/s	8.5717 KOps/s	$\color{#35bf28}+0.61\%$
test_compile_indexing[tensor-tensordict-eager]	0.1878ms	61.5092μs	16.2577 KOps/s	16.2976 KOps/s	$\color{#d91a1a}-0.24\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1594ms	0.1111ms	8.9986 KOps/s	8.9943 KOps/s	$\color{#35bf28}+0.05\%$
test_compile_indexing[tensor-tensorclass-eager]	0.1660ms	43.6985μs	22.8841 KOps/s	22.9767 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_indexing[tensor-pytree-compile]	0.1786ms	0.1122ms	8.9155 KOps/s	8.8773 KOps/s	$\color{#35bf28}+0.43\%$
test_compile_indexing[tensor-pytree-eager]	95.6220μs	43.5732μs	22.9499 KOps/s	23.2376 KOps/s	$\color{#d91a1a}-1.24\%$
test_compile_indexing[slice-tensordict-compile]	0.2050ms	0.1424ms	7.0207 KOps/s	7.0094 KOps/s	$\color{#35bf28}+0.16\%$
test_compile_indexing[slice-tensordict-eager]	0.1540ms	23.9177μs	41.8100 KOps/s	40.7004 KOps/s	$\color{#35bf28}+2.73\%$
test_compile_indexing[slice-tensorclass-compile]	0.1796ms	0.1361ms	7.3490 KOps/s	7.2640 KOps/s	$\color{#35bf28}+1.17\%$
test_compile_indexing[slice-tensorclass-eager]	58.5010μs	20.3556μs	49.1266 KOps/s	49.2384 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_indexing[slice-pytree-compile]	0.1766ms	0.1370ms	7.2979 KOps/s	7.1966 KOps/s	$\color{#35bf28}+1.41\%$
test_compile_indexing[slice-pytree-eager]	56.4210μs	20.4526μs	48.8936 KOps/s	49.6139 KOps/s	$\color{#d91a1a}-1.45\%$
test_compile_indexing[int-tensordict-compile]	0.2337ms	0.1427ms	7.0090 KOps/s	6.9638 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_indexing[int-tensordict-eager]	0.4912ms	23.5979μs	42.3767 KOps/s	41.4399 KOps/s	$\color{#35bf28}+2.26\%$
test_compile_indexing[int-tensorclass-compile]	0.2088ms	0.1379ms	7.2501 KOps/s	7.2392 KOps/s	$\color{#35bf28}+0.15\%$
test_compile_indexing[int-tensorclass-eager]	54.0510μs	20.6576μs	48.4084 KOps/s	49.4356 KOps/s	$\color{#d91a1a}-2.08\%$
test_compile_indexing[int-pytree-compile]	0.1763ms	0.1378ms	7.2589 KOps/s	7.2299 KOps/s	$\color{#35bf28}+0.40\%$
test_compile_indexing[int-pytree-eager]	52.8500μs	20.6513μs	48.4232 KOps/s	49.5870 KOps/s	$\color{#d91a1a}-2.35\%$
test_mod_add[eager]	76.7610μs	32.8487μs	30.4426 KOps/s	30.4290 KOps/s	$\color{#35bf28}+0.04\%$
test_mod_add[compile]	0.1200ms	78.5084μs	12.7375 KOps/s	12.4441 KOps/s	$\color{#35bf28}+2.36\%$
test_mod_add[compile-overhead]	0.3019ms	0.1496ms	6.6848 KOps/s	6.2572 KOps/s	$\textbf{\color{#35bf28}+6.83\%}$
test_mod_wrap[eager]	0.3285ms	0.2428ms	4.1192 KOps/s	4.1048 KOps/s	$\color{#35bf28}+0.35\%$
test_mod_wrap[compile]	1.3413ms	0.2910ms	3.4366 KOps/s	3.3347 KOps/s	$\color{#35bf28}+3.06\%$
test_mod_wrap[compile-overhead]	7.6277ms	4.0815ms	245.0076 Ops/s	254.8212 Ops/s	$\color{#d91a1a}-3.85\%$
test_mod_wrap_and_backward[eager]	1.5299ms	1.3767ms	726.3633 Ops/s	693.9433 Ops/s	$\color{#35bf28}+4.67\%$
test_mod_wrap_and_backward[compile]	1.5614ms	1.3288ms	752.5729 Ops/s	701.2918 Ops/s	$\textbf{\color{#35bf28}+7.31\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3301ms	0.9069ms	1.1026 KOps/s	1.0015 KOps/s	$\textbf{\color{#35bf28}+10.09\%}$
test_seq_add[eager]	0.1522ms	0.1055ms	9.4788 KOps/s	9.9569 KOps/s	$\color{#d91a1a}-4.80\%$
test_seq_add[compile]	0.1352ms	88.3021μs	11.3248 KOps/s	11.2235 KOps/s	$\color{#35bf28}+0.90\%$
test_seq_add[compile-overhead]	0.1940ms	0.1221ms	8.1923 KOps/s	8.0865 KOps/s	$\color{#35bf28}+1.31\%$
test_seq_wrap[eager]	0.6791ms	0.3866ms	2.5868 KOps/s	2.5188 KOps/s	$\color{#35bf28}+2.70\%$
test_seq_wrap[compile]	0.4011ms	0.3083ms	3.2433 KOps/s	3.1933 KOps/s	$\color{#35bf28}+1.56\%$
test_seq_wrap[compile-overhead]	0.2918ms	0.2156ms	4.6386 KOps/s	4.5845 KOps/s	$\color{#35bf28}+1.18\%$
test_func_call_runtime[False-eager]	0.8173ms	0.7365ms	1.3578 KOps/s	1.3193 KOps/s	$\color{#35bf28}+2.92\%$
test_func_call_runtime[False-compile]	0.8254ms	0.7726ms	1.2944 KOps/s	1.2911 KOps/s	$\color{#35bf28}+0.25\%$
test_func_call_runtime[False-compile-overhead]	0.4045ms	0.3510ms	2.8492 KOps/s	2.8143 KOps/s	$\color{#35bf28}+1.24\%$
test_func_call_runtime[True-eager]	1.0224ms	0.8933ms	1.1195 KOps/s	1.0935 KOps/s	$\color{#35bf28}+2.38\%$
test_func_call_runtime[True-compile]	0.9183ms	0.7931ms	1.2609 KOps/s	1.2451 KOps/s	$\color{#35bf28}+1.27\%$
test_func_call_runtime[True-compile-overhead]	0.4530ms	0.3740ms	2.6739 KOps/s	2.6671 KOps/s	$\color{#35bf28}+0.25\%$
test_func_call_cm_runtime[False-eager]	0.8034ms	0.7314ms	1.3673 KOps/s	1.2667 KOps/s	$\textbf{\color{#35bf28}+7.94\%}$
test_func_call_cm_runtime[False-compile]	0.8967ms	0.7770ms	1.2870 KOps/s	1.2602 KOps/s	$\color{#35bf28}+2.13\%$
test_func_call_cm_runtime[False-compile-overhead]	0.4539ms	0.3547ms	2.8192 KOps/s	2.8110 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[True-eager]	1.1744ms	1.0056ms	994.4637 Ops/s	986.7863 Ops/s	$\color{#35bf28}+0.78\%$
test_func_call_cm_runtime[True-compile]	0.9289ms	0.8202ms	1.2192 KOps/s	1.1698 KOps/s	$\color{#35bf28}+4.23\%$
test_func_call_cm_runtime[True-compile-overhead]	0.4530ms	0.4005ms	2.4968 KOps/s	2.4834 KOps/s	$\color{#35bf28}+0.54\%$
test_vmap_func_call_cm_runtime[eager]	2.5819ms	2.1064ms	474.7479 Ops/s	473.1135 Ops/s	$\color{#35bf28}+0.35\%$
test_vmap_func_call_cm_runtime[compile]	0.9035ms	0.8394ms	1.1913 KOps/s	1.1192 KOps/s	$\textbf{\color{#35bf28}+6.44\%}$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4567ms	0.4032ms	2.4802 KOps/s	2.4718 KOps/s	$\color{#35bf28}+0.34\%$
test_distributed	5.5627ms	0.3233ms	3.0927 KOps/s	8.8801 KOps/s	$\textbf{\color{#d91a1a}-65.17\%}$
test_tdmodule	0.1332ms	16.6154μs	60.1852 KOps/s	63.6959 KOps/s	$\textbf{\color{#d91a1a}-5.51\%}$
test_tdmodule_dispatch	68.7710μs	32.9397μs	30.3585 KOps/s	32.0603 KOps/s	$\textbf{\color{#d91a1a}-5.31\%}$
test_tdseq	40.6810μs	17.5152μs	57.0932 KOps/s	59.7826 KOps/s	$\color{#d91a1a}-4.50\%$
test_tdseq_dispatch	59.1010μs	35.1283μs	28.4671 KOps/s	29.2913 KOps/s	$\color{#d91a1a}-2.81\%$
test_instantiation_functorch	2.0049ms	1.8400ms	543.4930 Ops/s	538.4400 Ops/s	$\color{#35bf28}+0.94\%$
test_exec_functorch	0.2968ms	0.2024ms	4.9409 KOps/s	4.7798 KOps/s	$\color{#35bf28}+3.37\%$
test_exec_functional_call	0.2656ms	0.2055ms	4.8673 KOps/s	4.8302 KOps/s	$\color{#35bf28}+0.77\%$
test_exec_td_decorator	0.4403ms	0.2576ms	3.8826 KOps/s	3.8574 KOps/s	$\color{#35bf28}+0.65\%$
test_vmap_mlp_speed_decorator[True-True]	0.8057ms	0.6903ms	1.4487 KOps/s	1.4508 KOps/s	$\color{#d91a1a}-0.15\%$
test_vmap_mlp_speed_decorator[True-False]	0.8347ms	0.6981ms	1.4325 KOps/s	1.4519 KOps/s	$\color{#d91a1a}-1.33\%$
test_vmap_mlp_speed_decorator[False-True]	0.7401ms	0.6064ms	1.6491 KOps/s	1.6580 KOps/s	$\color{#d91a1a}-0.54\%$
test_vmap_mlp_speed_decorator[False-False]	0.6989ms	0.6182ms	1.6177 KOps/s	1.6515 KOps/s	$\color{#d91a1a}-2.05\%$
test_vmap_transformer_speed_decorator[True-True]	20.3144ms	19.7261ms	50.6941 Ops/s	50.7759 Ops/s	$\color{#d91a1a}-0.16\%$
test_vmap_transformer_speed_decorator[True-False]	20.4404ms	19.7449ms	50.6459 Ops/s	50.5623 Ops/s	$\color{#35bf28}+0.17\%$
test_vmap_transformer_speed_decorator[False-True]	19.7108ms	19.5606ms	51.1232 Ops/s	50.9117 Ops/s	$\color{#35bf28}+0.42\%$
test_vmap_transformer_speed_decorator[False-False]	20.3778ms	19.7071ms	50.7431 Ops/s	50.6922 Ops/s	$\color{#35bf28}+0.10\%$
test_to_module_speed[True]	1.3897ms	0.9996ms	1.0004 KOps/s	992.5223 Ops/s	$\color{#35bf28}+0.79\%$
test_to_module_speed[False]	1.4220ms	0.9699ms	1.0310 KOps/s	1.0179 KOps/s	$\color{#35bf28}+1.29\%$
test_tc_init	67.0410μs	39.1855μs	25.5196 KOps/s	26.3361 KOps/s	$\color{#d91a1a}-3.10\%$
test_tc_init_nested	0.1152ms	78.5995μs	12.7227 KOps/s	13.0666 KOps/s	$\color{#d91a1a}-2.63\%$
test_tc_first_layer_tensor	4.6901μs	0.6729μs	1.4860 MOps/s	1.4843 MOps/s	$\color{#35bf28}+0.12\%$
test_tc_first_layer_nontensor	18.4800μs	2.2583μs	442.8018 KOps/s	448.1177 KOps/s	$\color{#d91a1a}-1.19\%$
test_tc_second_layer_tensor	10.2127μs	1.3633μs	733.5161 KOps/s	732.5715 KOps/s	$\color{#35bf28}+0.13\%$
test_tc_second_layer_nontensor	32.8300μs	2.9307μs	341.2157 KOps/s	336.5638 KOps/s	$\color{#35bf28}+1.38\%$
test_unbind	0.1885s	9.4041ms	106.3370 Ops/s	93.9671 Ops/s	$\textbf{\color{#35bf28}+13.16\%}$
test_full_like	0.6553ms	0.5738ms	1.7429 KOps/s	1.7390 KOps/s	$\color{#35bf28}+0.22\%$
test_zeros_like	0.2866ms	0.1979ms	5.0536 KOps/s	5.0513 KOps/s	$\color{#35bf28}+0.05\%$
test_ones_like	0.2333ms	0.1977ms	5.0571 KOps/s	5.0554 KOps/s	$\color{#35bf28}+0.03\%$
test_clone	0.4488ms	0.4147ms	2.4114 KOps/s	2.4119 KOps/s	$\color{#d91a1a}-0.02\%$
test_squeeze	35.3510μs	9.7996μs	102.0455 KOps/s	102.3003 KOps/s	$\color{#d91a1a}-0.25\%$
test_unsqueeze	0.2157ms	75.4801μs	13.2485 KOps/s	13.2780 KOps/s	$\color{#d91a1a}-0.22\%$
test_split	0.3801ms	0.1531ms	6.5308 KOps/s	6.3988 KOps/s	$\color{#35bf28}+2.06\%$
test_permute	0.2645ms	0.1804ms	5.5419 KOps/s	5.5729 KOps/s	$\color{#d91a1a}-0.56\%$
test_stack	1.2640ms	0.8689ms	1.1509 KOps/s	1.1495 KOps/s	$\color{#35bf28}+0.12\%$
test_cat	1.2669ms	1.2310ms	812.3335 Ops/s	812.1075 Ops/s	$\color{#35bf28}+0.03\%$