[BugFix] Ensure grads and noned when needed

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Ensure grads and noned when needed #1069

[github-actions]

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

Total Benchmarks: 216. Improved: $\large\color{#35bf28}27$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	49.6130μs	21.4264μs	46.6714 KOps/s	46.0430 KOps/s	$\color{#35bf28}+1.36\%$
test_plain_set_stack_nested	60.1230μs	21.4382μs	46.6456 KOps/s	45.5783 KOps/s	$\color{#35bf28}+2.34\%$
test_plain_set_nested_inplace	61.1640μs	23.5136μs	42.5285 KOps/s	42.2147 KOps/s	$\color{#35bf28}+0.74\%$
test_plain_set_stack_nested_inplace	61.3350μs	23.5616μs	42.4420 KOps/s	42.0425 KOps/s	$\color{#35bf28}+0.95\%$
test_items	25.1470μs	4.1992μs	238.1395 KOps/s	240.2238 KOps/s	$\color{#d91a1a}-0.87\%$
test_items_nested	0.4065ms	0.3353ms	2.9824 KOps/s	2.9166 KOps/s	$\color{#35bf28}+2.26\%$
test_items_nested_locked	0.6329ms	0.3410ms	2.9328 KOps/s	2.9302 KOps/s	$\color{#35bf28}+0.09\%$
test_items_nested_leaf	0.1300ms	71.0052μs	14.0835 KOps/s	13.7194 KOps/s	$\color{#35bf28}+2.65\%$
test_items_stack_nested	0.6525ms	0.3407ms	2.9356 KOps/s	2.8923 KOps/s	$\color{#35bf28}+1.49\%$
test_items_stack_nested_leaf	0.1424ms	74.1460μs	13.4869 KOps/s	13.2453 KOps/s	$\color{#35bf28}+1.82\%$
test_items_stack_nested_locked	0.6212ms	0.3399ms	2.9423 KOps/s	2.9050 KOps/s	$\color{#35bf28}+1.28\%$
test_keys	17.7340μs	4.8409μs	206.5738 KOps/s	278.2824 KOps/s	$\textbf{\color{#d91a1a}-25.77\%}$
test_keys_nested	0.2003ms	0.1371ms	7.2965 KOps/s	7.0518 KOps/s	$\color{#35bf28}+3.47\%$
test_keys_nested_locked	0.6684ms	0.1413ms	7.0793 KOps/s	6.9443 KOps/s	$\color{#35bf28}+1.94\%$
test_keys_nested_leaf	0.2444ms	0.1142ms	8.7550 KOps/s	8.3382 KOps/s	$\color{#35bf28}+5.00\%$
test_keys_stack_nested	0.2517ms	0.1351ms	7.4026 KOps/s	7.1741 KOps/s	$\color{#35bf28}+3.19\%$
test_keys_stack_nested_leaf	0.2002ms	0.1154ms	8.6639 KOps/s	8.6160 KOps/s	$\color{#35bf28}+0.56\%$
test_keys_stack_nested_locked	0.2705ms	0.1403ms	7.1262 KOps/s	6.9075 KOps/s	$\color{#35bf28}+3.17\%$
test_values	5.2058μs	1.0354μs	965.7813 KOps/s	942.7813 KOps/s	$\color{#35bf28}+2.44\%$
test_values_nested	0.1091ms	55.1086μs	18.1460 KOps/s	17.3217 KOps/s	$\color{#35bf28}+4.76\%$
test_values_nested_locked	0.1111ms	54.6999μs	18.2816 KOps/s	17.4438 KOps/s	$\color{#35bf28}+4.80\%$
test_values_nested_leaf	0.1072ms	59.5297μs	16.7983 KOps/s	16.3596 KOps/s	$\color{#35bf28}+2.68\%$
test_values_stack_nested	0.1100ms	56.2781μs	17.7689 KOps/s	16.9999 KOps/s	$\color{#35bf28}+4.52\%$
test_values_stack_nested_leaf	0.1124ms	59.7659μs	16.7319 KOps/s	16.3294 KOps/s	$\color{#35bf28}+2.47\%$
test_values_stack_nested_locked	0.1075ms	55.9499μs	17.8731 KOps/s	17.1308 KOps/s	$\color{#35bf28}+4.33\%$
test_membership	5.6191μs	0.7299μs	1.3701 MOps/s	1.3123 MOps/s	$\color{#35bf28}+4.40\%$
test_membership_nested	33.5130μs	2.7396μs	365.0187 KOps/s	359.7195 KOps/s	$\color{#35bf28}+1.47\%$
test_membership_nested_leaf	23.8250μs	2.7362μs	365.4651 KOps/s	355.8225 KOps/s	$\color{#35bf28}+2.71\%$
test_membership_stacked_nested	22.1320μs	2.7166μs	368.1096 KOps/s	357.7841 KOps/s	$\color{#35bf28}+2.89\%$
test_membership_stacked_nested_leaf	15.4690μs	2.6897μs	371.7944 KOps/s	351.5829 KOps/s	$\textbf{\color{#35bf28}+5.75\%}$
test_membership_nested_last	19.9570μs	3.9851μs	250.9333 KOps/s	240.7254 KOps/s	$\color{#35bf28}+4.24\%$
test_membership_nested_leaf_last	37.3090μs	4.0172μs	248.9301 KOps/s	238.4439 KOps/s	$\color{#35bf28}+4.40\%$
test_membership_stacked_nested_last	27.1110μs	3.9664μs	252.1182 KOps/s	158.9135 KOps/s	$\textbf{\color{#35bf28}+58.65\%}$
test_membership_stacked_nested_leaf_last	35.8770μs	4.0215μs	248.6638 KOps/s	158.9444 KOps/s	$\textbf{\color{#35bf28}+56.45\%}$
test_nested_getleaf	36.2580μs	10.4859μs	95.3661 KOps/s	95.9031 KOps/s	$\color{#d91a1a}-0.56\%$
test_nested_get	28.1820μs	9.9769μs	100.2317 KOps/s	96.9187 KOps/s	$\color{#35bf28}+3.42\%$
test_stacked_getleaf	48.9320μs	10.4643μs	95.5628 KOps/s	95.0474 KOps/s	$\color{#35bf28}+0.54\%$
test_stacked_get	47.3420μs	10.0308μs	99.6933 KOps/s	99.2650 KOps/s	$\color{#35bf28}+0.43\%$
test_nested_getitemleaf	28.5730μs	10.8662μs	92.0287 KOps/s	88.4752 KOps/s	$\color{#35bf28}+4.02\%$
test_nested_getitem	33.6730μs	10.0651μs	99.3532 KOps/s	93.2774 KOps/s	$\textbf{\color{#35bf28}+6.51\%}$
test_stacked_getitemleaf	33.6930μs	10.9282μs	91.5067 KOps/s	89.8618 KOps/s	$\color{#35bf28}+1.83\%$
test_stacked_getitem	33.4330μs	10.2571μs	97.4937 KOps/s	95.5856 KOps/s	$\color{#35bf28}+2.00\%$
test_lock_nested	0.9772ms	0.4851ms	2.0612 KOps/s	2.0656 KOps/s	$\color{#d91a1a}-0.21\%$
test_lock_stack_nested	0.5613ms	0.4565ms	2.1904 KOps/s	2.2320 KOps/s	$\color{#d91a1a}-1.87\%$
test_unlock_nested	0.9912ms	0.4073ms	2.4555 KOps/s	2.4789 KOps/s	$\color{#d91a1a}-0.94\%$
test_unlock_stack_nested	0.6383ms	0.3785ms	2.6423 KOps/s	2.7296 KOps/s	$\color{#d91a1a}-3.20\%$
test_flatten_speed	0.1653ms	90.9944μs	10.9897 KOps/s	10.8559 KOps/s	$\color{#35bf28}+1.23\%$
test_unflatten_speed	0.8835ms	0.4717ms	2.1200 KOps/s	2.0901 KOps/s	$\color{#35bf28}+1.43\%$
test_common_ops	2.0218ms	1.1079ms	902.6039 Ops/s	865.2676 Ops/s	$\color{#35bf28}+4.31\%$
test_creation	16.7510μs	2.1111μs	473.6904 KOps/s	479.1918 KOps/s	$\color{#d91a1a}-1.15\%$
test_creation_empty	50.1330μs	18.0529μs	55.3927 KOps/s	53.2980 KOps/s	$\color{#35bf28}+3.93\%$
test_creation_nested_1	1.2426ms	21.2275μs	47.1087 KOps/s	45.2853 KOps/s	$\color{#35bf28}+4.03\%$
test_creation_nested_2	72.2050μs	25.5662μs	39.1142 KOps/s	38.2977 KOps/s	$\color{#35bf28}+2.13\%$
test_clone	48.4100μs	17.3422μs	57.6627 KOps/s	57.6336 KOps/s	$\color{#35bf28}+0.05\%$
test_getitem[int]	0.8518ms	16.4705μs	60.7147 KOps/s	58.3670 KOps/s	$\color{#35bf28}+4.02\%$
test_getitem[slice_int]	0.1374ms	30.3036μs	32.9994 KOps/s	31.4157 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_getitem[range]	0.1753ms	58.4010μs	17.1230 KOps/s	16.3407 KOps/s	$\color{#35bf28}+4.79\%$
test_getitem[tuple]	0.1314ms	24.7842μs	40.3484 KOps/s	39.8414 KOps/s	$\color{#35bf28}+1.27\%$
test_getitem[list]	0.1607ms	53.6984μs	18.6225 KOps/s	17.9551 KOps/s	$\color{#35bf28}+3.72\%$
test_setitem_dim[int]	65.7530μs	32.4588μs	30.8083 KOps/s	30.2550 KOps/s	$\color{#35bf28}+1.83\%$
test_setitem_dim[slice_int]	87.1230μs	60.2406μs	16.6001 KOps/s	15.6774 KOps/s	$\textbf{\color{#35bf28}+5.89\%}$
test_setitem_dim[range]	0.1485ms	83.1816μs	12.0219 KOps/s	11.6316 KOps/s	$\color{#35bf28}+3.36\%$
test_setitem_dim[tuple]	93.4950μs	47.9060μs	20.8742 KOps/s	19.8161 KOps/s	$\textbf{\color{#35bf28}+5.34\%}$
test_setitem	0.1057ms	30.0711μs	33.2545 KOps/s	32.7496 KOps/s	$\color{#35bf28}+1.54\%$
test_set	0.1550ms	29.4654μs	33.9381 KOps/s	33.8503 KOps/s	$\color{#35bf28}+0.26\%$
test_set_shared	4.6396ms	0.2172ms	4.6030 KOps/s	4.6285 KOps/s	$\color{#d91a1a}-0.55\%$
test_update	0.4188ms	37.9177μs	26.3729 KOps/s	26.4824 KOps/s	$\color{#d91a1a}-0.41\%$
test_update_nested	0.1988ms	46.4509μs	21.5281 KOps/s	20.3288 KOps/s	$\textbf{\color{#35bf28}+5.90\%}$
test_update__nested	0.9772ms	41.5548μs	24.0646 KOps/s	24.0927 KOps/s	$\color{#d91a1a}-0.12\%$
test_set_nested	0.1054ms	32.0966μs	31.1560 KOps/s	29.5728 KOps/s	$\textbf{\color{#35bf28}+5.35\%}$
test_set_nested_new	0.1009ms	36.7691μs	27.1968 KOps/s	26.3619 KOps/s	$\color{#35bf28}+3.17\%$
test_select	0.1233ms	53.9404μs	18.5390 KOps/s	17.2096 KOps/s	$\textbf{\color{#35bf28}+7.72\%}$
test_select_nested	0.1180ms	58.1263μs	17.2039 KOps/s	16.7919 KOps/s	$\color{#35bf28}+2.45\%$
test_exclude_nested	0.5464ms	77.8463μs	12.8458 KOps/s	13.3917 KOps/s	$\color{#d91a1a}-4.08\%$
test_empty[True]	0.4581ms	0.3491ms	2.8646 KOps/s	2.8263 KOps/s	$\color{#35bf28}+1.36\%$
test_empty[False]	9.2450μs	1.2207μs	819.2289 KOps/s	749.1616 KOps/s	$\textbf{\color{#35bf28}+9.35\%}$
test_unbind_speed	0.3584ms	0.3047ms	3.2822 KOps/s	3.3607 KOps/s	$\color{#d91a1a}-2.34\%$
test_unbind_speed_stack0	0.9042ms	0.2977ms	3.3588 KOps/s	3.4455 KOps/s	$\color{#d91a1a}-2.52\%$
test_unbind_speed_stack1	98.8955ms	0.8115ms	1.2323 KOps/s	1.3729 KOps/s	$\textbf{\color{#d91a1a}-10.24\%}$
test_split	2.4152ms	1.9673ms	508.3194 Ops/s	507.1358 Ops/s	$\color{#35bf28}+0.23\%$
test_chunk	0.1055s	2.1924ms	456.1262 Ops/s	420.2582 Ops/s	$\textbf{\color{#35bf28}+8.53\%}$
test_creation[device0]	0.5197ms	0.1162ms	8.6035 KOps/s	8.4894 KOps/s	$\color{#35bf28}+1.34\%$
test_creation_from_tensor	3.3715ms	0.1190ms	8.4047 KOps/s	8.3343 KOps/s	$\color{#35bf28}+0.84\%$
test_add_one[memmap_tensor0]	0.2638ms	7.4802μs	133.6855 KOps/s	137.8046 KOps/s	$\color{#d91a1a}-2.99\%$
test_contiguous[memmap_tensor0]	18.6950μs	1.8623μs	536.9591 KOps/s	516.9724 KOps/s	$\color{#35bf28}+3.87\%$
test_stack[memmap_tensor0]	81.0520μs	5.3121μs	188.2481 KOps/s	175.4203 KOps/s	$\textbf{\color{#35bf28}+7.31\%}$
test_memmaptd_index	1.0931ms	0.3875ms	2.5803 KOps/s	2.5127 KOps/s	$\color{#35bf28}+2.69\%$
test_memmaptd_index_astensor	1.1106ms	0.4714ms	2.1214 KOps/s	2.0662 KOps/s	$\color{#35bf28}+2.67\%$
test_memmaptd_index_op	1.3613ms	1.0071ms	992.9887 Ops/s	974.0518 Ops/s	$\color{#35bf28}+1.94\%$
test_serialize_model	0.1257s	0.1159s	8.6251 Ops/s	8.4671 Ops/s	$\color{#35bf28}+1.87\%$
test_serialize_model_pickle	0.4356s	0.3883s	2.5756 Ops/s	2.5038 Ops/s	$\color{#35bf28}+2.87\%$
test_serialize_weights	0.1156s	0.1126s	8.8802 Ops/s	7.6255 Ops/s	$\textbf{\color{#35bf28}+16.45\%}$
test_serialize_weights_returnearly	0.2670s	0.1743s	5.7388 Ops/s	6.3599 Ops/s	$\textbf{\color{#d91a1a}-9.77\%}$
test_serialize_weights_pickle	0.5175s	0.4292s	2.3299 Ops/s	2.5267 Ops/s	$\textbf{\color{#d91a1a}-7.79\%}$
test_serialize_weights_filesystem	0.2369s	0.1542s	6.4852 Ops/s	6.9500 Ops/s	$\textbf{\color{#d91a1a}-6.69\%}$
test_serialize_model_filesystem	0.1609s	0.1438s	6.9523 Ops/s	6.7945 Ops/s	$\color{#35bf28}+2.32\%$
test_reshape_pytree	0.1165ms	39.2502μs	25.4776 KOps/s	24.4193 KOps/s	$\color{#35bf28}+4.33\%$
test_reshape_td	0.1175ms	48.0432μs	20.8146 KOps/s	21.3112 KOps/s	$\color{#d91a1a}-2.33\%$
test_view_pytree	87.7140μs	38.7729μs	25.7912 KOps/s	24.9740 KOps/s	$\color{#35bf28}+3.27\%$
test_view_td	0.1143ms	54.2698μs	18.4265 KOps/s	18.9347 KOps/s	$\color{#d91a1a}-2.68\%$
test_unbind_pytree	74.1490μs	35.1557μs	28.4449 KOps/s	27.2432 KOps/s	$\color{#35bf28}+4.41\%$
test_unbind_td	0.3073ms	45.0138μs	22.2154 KOps/s	22.2173 KOps/s	$-0.01\%$
test_split_pytree	0.1151ms	38.0296μs	26.2953 KOps/s	25.6700 KOps/s	$\color{#35bf28}+2.44\%$
test_split_td	0.1919ms	56.9988μs	17.5442 KOps/s	17.1918 KOps/s	$\color{#35bf28}+2.05\%$
test_add_pytree	0.1466ms	47.2111μs	21.1815 KOps/s	21.5570 KOps/s	$\color{#d91a1a}-1.74\%$
test_add_td	0.1706ms	86.3540μs	11.5802 KOps/s	11.6684 KOps/s	$\color{#d91a1a}-0.76\%$
test_compile_add_one_nested[tensordict-compile]	0.1523ms	70.0422μs	14.2771 KOps/s	13.8260 KOps/s	$\color{#35bf28}+3.26\%$
test_compile_add_one_nested[tensordict-eager]	0.3548ms	0.1877ms	5.3284 KOps/s	5.3839 KOps/s	$\color{#d91a1a}-1.03\%$
test_compile_add_one_nested[pytree-compile]	0.1197ms	53.3437μs	18.7464 KOps/s	17.8109 KOps/s	$\textbf{\color{#35bf28}+5.25\%}$
test_compile_add_one_nested[pytree-eager]	0.2276ms	0.1454ms	6.8758 KOps/s	6.7806 KOps/s	$\color{#35bf28}+1.40\%$
test_compile_copy_nested[tensordict-compile]	83.9970μs	25.6571μs	38.9756 KOps/s	37.9497 KOps/s	$\color{#35bf28}+2.70\%$
test_compile_copy_nested[tensordict-eager]	0.1196ms	70.3331μs	14.2181 KOps/s	14.2758 KOps/s	$\color{#d91a1a}-0.40\%$
test_compile_copy_nested[pytree-compile]	0.1367ms	78.0182μs	12.8175 KOps/s	12.4027 KOps/s	$\color{#35bf28}+3.34\%$
test_compile_copy_nested[pytree-eager]	0.1230ms	67.5378μs	14.8065 KOps/s	14.5954 KOps/s	$\color{#35bf28}+1.45\%$
test_compile_add_one_flat[tensordict-compile]	0.1910ms	0.1130ms	8.8502 KOps/s	8.4252 KOps/s	$\textbf{\color{#35bf28}+5.04\%}$
test_compile_add_one_flat[tensordict-eager]	0.7595ms	0.2101ms	4.7597 KOps/s	4.6530 KOps/s	$\color{#35bf28}+2.29\%$
test_compile_add_one_flat[tensorclass-compile]	0.1269ms	52.2822μs	19.1270 KOps/s	17.4504 KOps/s	$\textbf{\color{#35bf28}+9.61\%}$
test_compile_add_one_flat[tensorclass-eager]	0.4547ms	71.3767μs	14.0102 KOps/s	14.1363 KOps/s	$\color{#d91a1a}-0.89\%$
test_compile_add_one_flat[pytree-compile]	0.6306ms	0.1136ms	8.8061 KOps/s	8.6940 KOps/s	$\color{#35bf28}+1.29\%$
test_compile_add_one_flat[pytree-eager]	0.6002ms	0.2995ms	3.3391 KOps/s	3.2430 KOps/s	$\color{#35bf28}+2.96\%$
test_compile_add_self_flat[tensordict-eager]	0.3872ms	0.2176ms	4.5947 KOps/s	4.5223 KOps/s	$\color{#35bf28}+1.60\%$
test_compile_add_self_flat[tensordict-compile]	0.2125ms	0.1148ms	8.7085 KOps/s	8.5921 KOps/s	$\color{#35bf28}+1.35\%$
test_compile_add_self_flat[tensorclass-eager]	0.3152ms	63.0258μs	15.8665 KOps/s	15.8816 KOps/s	$\color{#d91a1a}-0.09\%$
test_compile_add_self_flat[tensorclass-compile]	0.1233ms	53.8319μs	18.5763 KOps/s	17.8261 KOps/s	$\color{#35bf28}+4.21\%$
test_compile_add_self_flat[pytree-eager]	0.6478ms	0.2454ms	4.0749 KOps/s	4.0201 KOps/s	$\color{#35bf28}+1.36\%$
test_compile_add_self_flat[pytree-compile]	0.1890ms	0.1118ms	8.9424 KOps/s	8.7276 KOps/s	$\color{#35bf28}+2.46\%$
test_compile_copy_flat[tensordict-compile]	57.4870μs	21.6270μs	46.2385 KOps/s	47.2645 KOps/s	$\color{#d91a1a}-2.17\%$
test_compile_copy_flat[tensordict-eager]	0.1316ms	60.0472μs	16.6536 KOps/s	16.7233 KOps/s	$\color{#d91a1a}-0.42\%$
test_compile_copy_flat[pytree-compile]	0.1632ms	79.4468μs	12.5870 KOps/s	11.3378 KOps/s	$\textbf{\color{#35bf28}+11.02\%}$
test_compile_copy_flat[pytree-eager]	0.1270ms	69.6111μs	14.3655 KOps/s	14.1133 KOps/s	$\color{#35bf28}+1.79\%$
test_compile_assign_and_add[tensordict-compile]	0.4545ms	0.2191ms	4.5631 KOps/s	4.5311 KOps/s	$\color{#35bf28}+0.71\%$
test_compile_assign_and_add[tensordict-eager]	2.8607ms	1.7673ms	565.8406 Ops/s	561.3136 Ops/s	$\color{#35bf28}+0.81\%$
test_compile_assign_and_add[pytree-compile]	0.2954ms	0.2110ms	4.7384 KOps/s	4.6418 KOps/s	$\color{#35bf28}+2.08\%$
test_compile_assign_and_add[pytree-eager]	1.9785ms	1.1743ms	851.5532 Ops/s	838.4011 Ops/s	$\color{#35bf28}+1.57\%$
test_compile_assign_and_add_stack[compile]	0.5726ms	0.4648ms	2.1514 KOps/s	2.1203 KOps/s	$\color{#35bf28}+1.47\%$
test_compile_assign_and_add_stack[eager]	4.2133ms	3.9498ms	253.1745 Ops/s	247.6838 Ops/s	$\color{#35bf28}+2.22\%$
test_compile_indexing[tensor-tensordict-compile]	0.1092ms	44.4576μs	22.4934 KOps/s	21.9444 KOps/s	$\color{#35bf28}+2.50\%$
test_compile_indexing[tensor-tensordict-eager]	0.5112ms	50.5057μs	19.7997 KOps/s	19.4237 KOps/s	$\color{#35bf28}+1.94\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1072ms	36.8802μs	27.1148 KOps/s	25.6426 KOps/s	$\textbf{\color{#35bf28}+5.74\%}$
test_compile_indexing[tensor-tensorclass-eager]	78.4670μs	28.8343μs	34.6809 KOps/s	33.7988 KOps/s	$\color{#35bf28}+2.61\%$
test_compile_indexing[tensor-pytree-compile]	0.1151ms	37.2440μs	26.8500 KOps/s	24.9589 KOps/s	$\textbf{\color{#35bf28}+7.58\%}$
test_compile_indexing[tensor-pytree-eager]	0.1011ms	28.8898μs	34.6143 KOps/s	33.4919 KOps/s	$\color{#35bf28}+3.35\%$
test_compile_indexing[slice-tensordict-compile]	0.1424ms	77.9583μs	12.8274 KOps/s	12.3422 KOps/s	$\color{#35bf28}+3.93\%$
test_compile_indexing[slice-tensordict-eager]	0.5571ms	29.1719μs	34.2795 KOps/s	34.8753 KOps/s	$\color{#d91a1a}-1.71\%$
test_compile_indexing[slice-tensorclass-compile]	0.5569ms	73.0235μs	13.6942 KOps/s	13.7989 KOps/s	$\color{#d91a1a}-0.76\%$
test_compile_indexing[slice-tensorclass-eager]	80.9520μs	23.4144μs	42.7088 KOps/s	42.2024 KOps/s	$\color{#35bf28}+1.20\%$
test_compile_indexing[slice-pytree-compile]	0.1570ms	71.6312μs	13.9604 KOps/s	13.5407 KOps/s	$\color{#35bf28}+3.10\%$
test_compile_indexing[slice-pytree-eager]	73.5170μs	23.1745μs	43.1508 KOps/s	42.1678 KOps/s	$\color{#35bf28}+2.33\%$
test_compile_indexing[int-tensordict-compile]	0.2259ms	80.6972μs	12.3920 KOps/s	12.2007 KOps/s	$\color{#35bf28}+1.57\%$
test_compile_indexing[int-tensordict-eager]	1.1007ms	29.3952μs	34.0192 KOps/s	34.6861 KOps/s	$\color{#d91a1a}-1.92\%$
test_compile_indexing[int-tensorclass-compile]	0.1617ms	71.4076μs	14.0041 KOps/s	13.7303 KOps/s	$\color{#35bf28}+1.99\%$
test_compile_indexing[int-tensorclass-eager]	68.2280μs	23.2690μs	42.9755 KOps/s	41.8765 KOps/s	$\color{#35bf28}+2.62\%$
test_compile_indexing[int-pytree-compile]	0.1598ms	71.0495μs	14.0747 KOps/s	13.7412 KOps/s	$\color{#35bf28}+2.43\%$
test_compile_indexing[int-pytree-eager]	87.2130μs	23.5938μs	42.3841 KOps/s	42.2316 KOps/s	$\color{#35bf28}+0.36\%$
test_mod_add[eager]	85.0690μs	26.7797μs	37.3418 KOps/s	36.8857 KOps/s	$\color{#35bf28}+1.24\%$
test_mod_add[compile]	0.1215ms	44.4078μs	22.5186 KOps/s	21.1640 KOps/s	$\textbf{\color{#35bf28}+6.40\%}$
test_mod_add[compile-overhead]	0.1181ms	44.5136μs	22.4650 KOps/s	21.6689 KOps/s	$\color{#35bf28}+3.67\%$
test_mod_wrap[eager]	0.4209ms	0.2200ms	4.5459 KOps/s	4.5140 KOps/s	$\color{#35bf28}+0.71\%$
test_mod_wrap[compile]	1.6748ms	0.2036ms	4.9110 KOps/s	4.8314 KOps/s	$\color{#35bf28}+1.65\%$
test_mod_wrap[compile-overhead]	1.6821ms	0.2021ms	4.9475 KOps/s	4.8427 KOps/s	$\color{#35bf28}+2.16\%$
test_mod_wrap_and_backward[eager]	18.7637ms	13.1051ms	76.3064 Ops/s	81.8971 Ops/s	$\textbf{\color{#d91a1a}-6.83\%}$
test_mod_wrap_and_backward[compile]	17.6042ms	12.1402ms	82.3713 Ops/s	74.8248 Ops/s	$\textbf{\color{#35bf28}+10.09\%}$
test_mod_wrap_and_backward[compile-overhead]	13.1888ms	10.7032ms	93.4299 Ops/s	87.4501 Ops/s	$\textbf{\color{#35bf28}+6.84\%}$
test_seq_add[eager]	0.1638ms	90.5444μs	11.0443 KOps/s	10.3746 KOps/s	$\textbf{\color{#35bf28}+6.46\%}$
test_seq_add[compile]	0.1182ms	59.3116μs	16.8601 KOps/s	16.6813 KOps/s	$\color{#35bf28}+1.07\%$
test_seq_add[compile-overhead]	0.1198ms	58.2017μs	17.1816 KOps/s	17.0095 KOps/s	$\color{#35bf28}+1.01\%$
test_seq_wrap[eager]	0.7278ms	0.4017ms	2.4893 KOps/s	2.4767 KOps/s	$\color{#35bf28}+0.51\%$
test_seq_wrap[compile]	0.3647ms	0.2268ms	4.4095 KOps/s	4.4148 KOps/s	$\color{#d91a1a}-0.12\%$
test_seq_wrap[compile-overhead]	0.4283ms	0.2261ms	4.4230 KOps/s	4.4593 KOps/s	$\color{#d91a1a}-0.81\%$
test_func_call_runtime[False-eager]	1.2254ms	0.5549ms	1.8021 KOps/s	1.7828 KOps/s	$\color{#35bf28}+1.08\%$
test_func_call_runtime[False-compile]	0.7822ms	0.4247ms	2.3547 KOps/s	2.3310 KOps/s	$\color{#35bf28}+1.01\%$
test_func_call_runtime[False-compile-overhead]	0.5228ms	0.4260ms	2.3476 KOps/s	2.2045 KOps/s	$\textbf{\color{#35bf28}+6.49\%}$
test_func_call_runtime[True-eager]	0.9246ms	0.7709ms	1.2972 KOps/s	1.3022 KOps/s	$\color{#d91a1a}-0.38\%$
test_func_call_runtime[True-compile]	0.6935ms	0.4625ms	2.1623 KOps/s	2.1097 KOps/s	$\color{#35bf28}+2.49\%$
test_func_call_runtime[True-compile-overhead]	0.6157ms	0.4670ms	2.1414 KOps/s	2.1302 KOps/s	$\color{#35bf28}+0.52\%$
test_func_call_cm_runtime[False-eager]	0.9651ms	0.5566ms	1.7966 KOps/s	1.8013 KOps/s	$\color{#d91a1a}-0.26\%$
test_func_call_cm_runtime[False-compile]	0.8147ms	0.4261ms	2.3469 KOps/s	2.3156 KOps/s	$\color{#35bf28}+1.35\%$
test_func_call_cm_runtime[False-compile-overhead]	0.8769ms	0.4243ms	2.3570 KOps/s	2.3353 KOps/s	$\color{#35bf28}+0.93\%$
test_func_call_cm_runtime[True-eager]	1.3964ms	0.9084ms	1.1008 KOps/s	1.1030 KOps/s	$\color{#d91a1a}-0.19\%$
test_func_call_cm_runtime[True-compile]	1.0011ms	0.4892ms	2.0443 KOps/s	2.0132 KOps/s	$\color{#35bf28}+1.55\%$
test_func_call_cm_runtime[True-compile-overhead]	1.0177ms	0.4940ms	2.0244 KOps/s	1.9634 KOps/s	$\color{#35bf28}+3.11\%$
test_vmap_func_call_cm_runtime[eager]	2.5140ms	1.8629ms	536.8002 Ops/s	525.5669 Ops/s	$\color{#35bf28}+2.14\%$
test_vmap_func_call_cm_runtime[compile]	0.8839ms	0.5153ms	1.9405 KOps/s	1.9178 KOps/s	$\color{#35bf28}+1.18\%$
test_vmap_func_call_cm_runtime[compile-overhead]	1.6189ms	0.5170ms	1.9342 KOps/s	1.9153 KOps/s	$\color{#35bf28}+0.99\%$
test_distributed	0.2825ms	0.1233ms	8.1096 KOps/s	7.8963 KOps/s	$\color{#35bf28}+2.70\%$
test_tdmodule	34.2540μs	18.7153μs	53.4321 KOps/s	51.5716 KOps/s	$\color{#35bf28}+3.61\%$
test_tdmodule_dispatch	62.4470μs	36.5667μs	27.3473 KOps/s	26.3815 KOps/s	$\color{#35bf28}+3.66\%$
test_tdseq	47.6890μs	21.3079μs	46.9309 KOps/s	45.4890 KOps/s	$\color{#35bf28}+3.17\%$
test_tdseq_dispatch	67.6170μs	40.8210μs	24.4972 KOps/s	23.1888 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_instantiation_functorch	2.4139ms	1.5329ms	652.3370 Ops/s	649.6572 Ops/s	$\color{#35bf28}+0.41\%$
test_exec_functorch	0.3363ms	0.1795ms	5.5700 KOps/s	5.4672 KOps/s	$\color{#35bf28}+1.88\%$
test_exec_functional_call	0.2873ms	0.1752ms	5.7092 KOps/s	5.5880 KOps/s	$\color{#35bf28}+2.17\%$
test_exec_td_decorator	0.5023ms	0.2304ms	4.3397 KOps/s	4.3127 KOps/s	$\color{#35bf28}+0.63\%$
test_vmap_mlp_speed_decorator[True-True]	0.8587ms	0.6277ms	1.5930 KOps/s	1.5635 KOps/s	$\color{#35bf28}+1.89\%$
test_vmap_mlp_speed_decorator[True-False]	0.9204ms	0.6262ms	1.5969 KOps/s	1.5541 KOps/s	$\color{#35bf28}+2.75\%$
test_vmap_mlp_speed_decorator[False-True]	0.6986ms	0.5134ms	1.9478 KOps/s	1.8767 KOps/s	$\color{#35bf28}+3.79\%$
test_vmap_mlp_speed_decorator[False-False]	0.7983ms	0.5134ms	1.9477 KOps/s	1.9055 KOps/s	$\color{#35bf28}+2.21\%$
test_to_module_speed[True]	1.3820ms	1.2839ms	778.9051 Ops/s	784.6799 Ops/s	$\color{#d91a1a}-0.74\%$
test_to_module_speed[False]	1.6255ms	1.2676ms	788.9102 Ops/s	795.8454 Ops/s	$\color{#d91a1a}-0.87\%$
test_tc_init	98.3140μs	45.8439μs	21.8132 KOps/s	21.6767 KOps/s	$\color{#35bf28}+0.63\%$
test_tc_init_nested	0.1985ms	90.2572μs	11.0794 KOps/s	10.9272 KOps/s	$\color{#35bf28}+1.39\%$
test_tc_first_layer_tensor	26.0980μs	1.5335μs	652.1134 KOps/s	657.9800 KOps/s	$\color{#d91a1a}-0.89\%$
test_tc_first_layer_nontensor	43.9930μs	4.7536μs	210.3669 KOps/s	201.8592 KOps/s	$\color{#35bf28}+4.21\%$
test_tc_second_layer_tensor	18.4550μs	2.8354μs	352.6895 KOps/s	363.7948 KOps/s	$\color{#d91a1a}-3.05\%$
test_tc_second_layer_nontensor	24.5770μs	6.0851μs	164.3359 KOps/s	159.4528 KOps/s	$\color{#35bf28}+3.06\%$
test_unbind	0.2080s	12.1825ms	82.0846 Ops/s	76.4124 Ops/s	$\textbf{\color{#35bf28}+7.42\%}$
test_full_like	7.8346ms	6.7981ms	147.0993 Ops/s	143.9310 Ops/s	$\color{#35bf28}+2.20\%$
test_zeros_like	3.0139ms	2.6110ms	382.9892 Ops/s	367.2204 Ops/s	$\color{#35bf28}+4.29\%$
test_ones_like	3.3971ms	3.0680ms	325.9440 Ops/s	320.4610 Ops/s	$\color{#35bf28}+1.71\%$
test_clone	5.3023ms	4.8360ms	206.7840 Ops/s	204.9076 Ops/s	$\color{#35bf28}+0.92\%$
test_squeeze	62.4660μs	12.0414μs	83.0471 KOps/s	86.9250 KOps/s	$\color{#d91a1a}-4.46\%$
test_unsqueeze	0.2105ms	88.4146μs	11.3104 KOps/s	11.5021 KOps/s	$\color{#d91a1a}-1.67\%$
test_split	0.4854ms	0.1841ms	5.4305 KOps/s	5.2653 KOps/s	$\color{#35bf28}+3.14\%$
test_permute	0.3362ms	0.2135ms	4.6842 KOps/s	4.6497 KOps/s	$\color{#35bf28}+0.74\%$
test_stack	26.8154ms	24.2457ms	41.2444 Ops/s	39.9090 Ops/s	$\color{#35bf28}+3.35\%$
test_cat	29.1559ms	24.0184ms	41.6347 Ops/s	40.1004 Ops/s	$\color{#35bf28}+3.83\%$

[github-actions]

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

Total Benchmarks: 222. Improved: $\large\color{#35bf28}31$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	45.3910μs	14.8827μs	67.1923 KOps/s	63.2061 KOps/s	$\textbf{\color{#35bf28}+6.31\%}$
test_plain_set_stack_nested	0.1188ms	15.1261μs	66.1110 KOps/s	62.3097 KOps/s	$\textbf{\color{#35bf28}+6.10\%}$
test_plain_set_nested_inplace	60.5710μs	15.8973μs	62.9038 KOps/s	59.4723 KOps/s	$\textbf{\color{#35bf28}+5.77\%}$
test_plain_set_stack_nested_inplace	80.2110μs	15.8711μs	63.0075 KOps/s	59.1612 KOps/s	$\textbf{\color{#35bf28}+6.50\%}$
test_items	22.9300μs	2.9212μs	342.3247 KOps/s	341.9150 KOps/s	$\color{#35bf28}+0.12\%$
test_items_nested	0.4480ms	0.3190ms	3.1349 KOps/s	3.1615 KOps/s	$\color{#d91a1a}-0.84\%$
test_items_nested_locked	0.4144ms	0.3217ms	3.1086 KOps/s	3.1491 KOps/s	$\color{#d91a1a}-1.29\%$
test_items_nested_leaf	88.9710μs	58.3364μs	17.1420 KOps/s	17.1112 KOps/s	$\color{#35bf28}+0.18\%$
test_items_stack_nested	0.3831ms	0.3256ms	3.0717 KOps/s	3.1757 KOps/s	$\color{#d91a1a}-3.28\%$
test_items_stack_nested_leaf	88.9020μs	59.3970μs	16.8359 KOps/s	17.0379 KOps/s	$\color{#d91a1a}-1.19\%$
test_items_stack_nested_locked	0.4122ms	0.3266ms	3.0617 KOps/s	3.1594 KOps/s	$\color{#d91a1a}-3.09\%$
test_keys	0.6847ms	3.4859μs	286.8707 KOps/s	288.5850 KOps/s	$\color{#d91a1a}-0.59\%$
test_keys_nested	0.1002ms	71.9301μs	13.9024 KOps/s	13.8355 KOps/s	$\color{#35bf28}+0.48\%$
test_keys_nested_locked	2.7462ms	77.5944μs	12.8875 KOps/s	12.8478 KOps/s	$\color{#35bf28}+0.31\%$
test_keys_nested_leaf	91.5710μs	63.2571μs	15.8085 KOps/s	15.7421 KOps/s	$\color{#35bf28}+0.42\%$
test_keys_stack_nested	0.1201ms	72.6514μs	13.7644 KOps/s	13.6552 KOps/s	$\color{#35bf28}+0.80\%$
test_keys_stack_nested_leaf	91.4020μs	63.6645μs	15.7074 KOps/s	15.6833 KOps/s	$\color{#35bf28}+0.15\%$
test_keys_stack_nested_locked	0.2483ms	77.4961μs	12.9039 KOps/s	12.8106 KOps/s	$\color{#35bf28}+0.73\%$
test_values	32.2523μs	0.8591μs	1.1640 MOps/s	1.1553 MOps/s	$\color{#35bf28}+0.75\%$
test_values_nested	57.9910μs	30.9086μs	32.3534 KOps/s	32.2493 KOps/s	$\color{#35bf28}+0.32\%$
test_values_nested_locked	65.3410μs	32.5035μs	30.7659 KOps/s	30.5315 KOps/s	$\color{#35bf28}+0.77\%$
test_values_nested_leaf	61.9610μs	33.5559μs	29.8010 KOps/s	29.7061 KOps/s	$\color{#35bf28}+0.32\%$
test_values_stack_nested	63.8910μs	31.4634μs	31.7830 KOps/s	31.9635 KOps/s	$\color{#d91a1a}-0.56\%$
test_values_stack_nested_leaf	55.5210μs	34.0709μs	29.3506 KOps/s	29.5765 KOps/s	$\color{#d91a1a}-0.76\%$
test_values_stack_nested_locked	59.9710μs	32.6876μs	30.5926 KOps/s	30.3586 KOps/s	$\color{#35bf28}+0.77\%$
test_membership	1.6331μs	0.5053μs	1.9790 MOps/s	1.9640 MOps/s	$\color{#35bf28}+0.76\%$
test_membership_nested	15.8855μs	1.9422μs	514.8729 KOps/s	532.5396 KOps/s	$\color{#d91a1a}-3.32\%$
test_membership_nested_leaf	13.8700μs	1.9351μs	516.7654 KOps/s	531.4385 KOps/s	$\color{#d91a1a}-2.76\%$
test_membership_stacked_nested	28.5510μs	2.0208μs	494.8518 KOps/s	515.1788 KOps/s	$\color{#d91a1a}-3.95\%$
test_membership_stacked_nested_leaf	0.1912ms	2.0177μs	495.6232 KOps/s	521.6396 KOps/s	$\color{#d91a1a}-4.99\%$
test_membership_nested_last	23.2210μs	2.8269μs	353.7411 KOps/s	355.3344 KOps/s	$\color{#d91a1a}-0.45\%$
test_membership_nested_leaf_last	33.1810μs	2.8690μs	348.5521 KOps/s	356.7887 KOps/s	$\color{#d91a1a}-2.31\%$
test_membership_stacked_nested_last	0.1997ms	2.8886μs	346.1929 KOps/s	358.2392 KOps/s	$\color{#d91a1a}-3.36\%$
test_membership_stacked_nested_leaf_last	28.5100μs	2.8447μs	351.5342 KOps/s	358.8436 KOps/s	$\color{#d91a1a}-2.04\%$
test_nested_getleaf	26.2100μs	5.9999μs	166.6688 KOps/s	167.2306 KOps/s	$\color{#d91a1a}-0.34\%$
test_nested_get	33.8510μs	5.6845μs	175.9158 KOps/s	176.7904 KOps/s	$\color{#d91a1a}-0.49\%$
test_stacked_getleaf	30.8810μs	5.9709μs	167.4781 KOps/s	166.9726 KOps/s	$\color{#35bf28}+0.30\%$
test_stacked_get	31.1300μs	5.6780μs	176.1181 KOps/s	176.2536 KOps/s	$\color{#d91a1a}-0.08\%$
test_nested_getitemleaf	30.6010μs	6.0791μs	164.4979 KOps/s	165.1749 KOps/s	$\color{#d91a1a}-0.41\%$
test_nested_getitem	35.1310μs	5.7838μs	172.8974 KOps/s	174.1798 KOps/s	$\color{#d91a1a}-0.74\%$
test_stacked_getitemleaf	26.2310μs	6.1089μs	163.6950 KOps/s	164.6724 KOps/s	$\color{#d91a1a}-0.59\%$
test_stacked_getitem	32.2700μs	5.7904μs	172.6984 KOps/s	172.9578 KOps/s	$\color{#d91a1a}-0.15\%$
test_lock_nested	5.4375ms	0.4261ms	2.3468 KOps/s	2.3218 KOps/s	$\color{#35bf28}+1.07\%$
test_lock_stack_nested	0.5375ms	0.3930ms	2.5447 KOps/s	2.4841 KOps/s	$\color{#35bf28}+2.44\%$
test_unlock_nested	0.7615ms	0.3624ms	2.7596 KOps/s	2.6749 KOps/s	$\color{#35bf28}+3.17\%$
test_unlock_stack_nested	0.4249ms	0.3327ms	3.0053 KOps/s	2.9243 KOps/s	$\color{#35bf28}+2.77\%$
test_flatten_speed	0.1335ms	74.0817μs	13.4986 KOps/s	13.4666 KOps/s	$\color{#35bf28}+0.24\%$
test_unflatten_speed	0.3314ms	0.2923ms	3.4208 KOps/s	3.4058 KOps/s	$\color{#35bf28}+0.44\%$
test_common_ops	1.5689ms	1.2352ms	809.5987 Ops/s	777.0831 Ops/s	$\color{#35bf28}+4.18\%$
test_creation	23.5010μs	1.5893μs	629.2000 KOps/s	628.1802 KOps/s	$\color{#35bf28}+0.16\%$
test_creation_empty	56.5310μs	16.2965μs	61.3628 KOps/s	54.2265 KOps/s	$\textbf{\color{#35bf28}+13.16\%}$
test_creation_nested_1	53.0710μs	18.0808μs	55.3073 KOps/s	49.6555 KOps/s	$\textbf{\color{#35bf28}+11.38\%}$
test_creation_nested_2	47.6110μs	20.6102μs	48.5198 KOps/s	44.0294 KOps/s	$\textbf{\color{#35bf28}+10.20\%}$
test_clone	0.1832ms	28.6559μs	34.8968 KOps/s	34.5448 KOps/s	$\color{#35bf28}+1.02\%$
test_getitem[int]	1.2449ms	16.5248μs	60.5152 KOps/s	59.7923 KOps/s	$\color{#35bf28}+1.21\%$
test_getitem[slice_int]	0.1342ms	29.0648μs	34.4059 KOps/s	34.0083 KOps/s	$\color{#35bf28}+1.17\%$
test_getitem[range]	0.2399ms	0.1130ms	8.8500 KOps/s	8.9038 KOps/s	$\color{#d91a1a}-0.60\%$
test_getitem[tuple]	0.1507ms	25.3655μs	39.4236 KOps/s	39.9591 KOps/s	$\color{#d91a1a}-1.34\%$
test_getitem[list]	0.2710ms	0.1005ms	9.9489 KOps/s	9.9172 KOps/s	$\color{#35bf28}+0.32\%$
test_setitem_dim[int]	68.7510μs	43.8002μs	22.8309 KOps/s	22.9551 KOps/s	$\color{#d91a1a}-0.54\%$
test_setitem_dim[slice_int]	0.2086ms	66.3746μs	15.0660 KOps/s	15.0765 KOps/s	$\color{#d91a1a}-0.07\%$
test_setitem_dim[range]	0.2467ms	0.1273ms	7.8531 KOps/s	7.8574 KOps/s	$\color{#d91a1a}-0.06\%$
test_setitem_dim[tuple]	0.1999ms	59.7974μs	16.7231 KOps/s	16.7032 KOps/s	$\color{#35bf28}+0.12\%$
test_setitem	0.1932ms	41.8945μs	23.8695 KOps/s	23.5264 KOps/s	$\color{#35bf28}+1.46\%$
test_set	0.2196ms	40.3789μs	24.7654 KOps/s	23.7098 KOps/s	$\color{#35bf28}+4.45\%$
test_set_shared	0.4014ms	50.2158μs	19.9141 KOps/s	19.6094 KOps/s	$\color{#35bf28}+1.55\%$
test_update	0.2261ms	49.1518μs	20.3451 KOps/s	18.9404 KOps/s	$\textbf{\color{#35bf28}+7.42\%}$
test_update_nested	0.2339ms	56.8174μs	17.6002 KOps/s	16.8676 KOps/s	$\color{#35bf28}+4.34\%$
test_update__nested	0.1991ms	58.4123μs	17.1197 KOps/s	16.2464 KOps/s	$\textbf{\color{#35bf28}+5.38\%}$
test_set_nested	0.2290ms	42.5436μs	23.5053 KOps/s	22.0009 KOps/s	$\textbf{\color{#35bf28}+6.84\%}$
test_set_nested_new	0.2038ms	46.5038μs	21.5036 KOps/s	20.6735 KOps/s	$\color{#35bf28}+4.02\%$
test_select	0.2069ms	60.0347μs	16.6570 KOps/s	16.2701 KOps/s	$\color{#35bf28}+2.38\%$
test_select_nested	70.9110μs	42.5966μs	23.4760 KOps/s	22.8503 KOps/s	$\color{#35bf28}+2.74\%$
test_exclude_nested	97.0710μs	60.1861μs	16.6151 KOps/s	16.4659 KOps/s	$\color{#35bf28}+0.91\%$
test_empty[True]	0.3378ms	0.2614ms	3.8253 KOps/s	3.8942 KOps/s	$\color{#d91a1a}-1.77\%$
test_empty[False]	2.9760μs	0.8275μs	1.2084 MOps/s	1.1964 MOps/s	$\color{#35bf28}+1.01\%$
test_to	0.1517ms	51.0070μs	19.6052 KOps/s	18.8272 KOps/s	$\color{#35bf28}+4.13\%$
test_to_nonblocking	83.2610μs	49.1861μs	20.3309 KOps/s	19.3611 KOps/s	$\textbf{\color{#35bf28}+5.01\%}$
test_unbind_speed	1.2034ms	0.2776ms	3.6019 KOps/s	3.4613 KOps/s	$\color{#35bf28}+4.06\%$
test_unbind_speed_stack0	0.3764ms	0.2763ms	3.6190 KOps/s	3.5306 KOps/s	$\color{#35bf28}+2.50\%$
test_unbind_speed_stack1	97.3206ms	0.7064ms	1.4156 KOps/s	1.3679 KOps/s	$\color{#35bf28}+3.49\%$
test_split	0.1002s	2.2768ms	439.2089 Ops/s	433.9057 Ops/s	$\color{#35bf28}+1.22\%$
test_chunk	0.1022s	2.2984ms	435.0909 Ops/s	430.5495 Ops/s	$\color{#35bf28}+1.05\%$
test_to[False]	6.2445ms	5.9792ms	167.2460 Ops/s	157.6223 Ops/s	$\textbf{\color{#35bf28}+6.11\%}$
test_to[True]	4.8230ms	4.4319ms	225.6348 Ops/s	220.7713 Ops/s	$\color{#35bf28}+2.20\%$
test_to_njt[False]	0.3596s	0.2754s	3.6306 Ops/s	3.5710 Ops/s	$\color{#35bf28}+1.67\%$
test_to_njt[True]	0.3735s	0.2878s	3.4744 Ops/s	3.7192 Ops/s	$\textbf{\color{#d91a1a}-6.58\%}$
test_creation[device0]	0.4214ms	0.1289ms	7.7598 KOps/s	7.7932 KOps/s	$\color{#d91a1a}-0.43\%$
test_creation_from_tensor	0.4662ms	0.1301ms	7.6875 KOps/s	7.6663 KOps/s	$\color{#35bf28}+0.28\%$
test_add_one[memmap_tensor0]	0.1382ms	8.8341μs	113.1978 KOps/s	111.5715 KOps/s	$\color{#35bf28}+1.46\%$
test_contiguous[memmap_tensor0]	35.0410μs	2.1896μs	456.7103 KOps/s	440.6131 KOps/s	$\color{#35bf28}+3.65\%$
test_stack[memmap_tensor0]	0.1517ms	6.9770μs	143.3284 KOps/s	137.8855 KOps/s	$\color{#35bf28}+3.95\%$
test_memmaptd_index	1.0569ms	0.4381ms	2.2828 KOps/s	2.2593 KOps/s	$\color{#35bf28}+1.04\%$
test_memmaptd_index_astensor	0.8567ms	0.4933ms	2.0273 KOps/s	1.9877 KOps/s	$\color{#35bf28}+1.99\%$
test_memmaptd_index_op	1.4155ms	1.0239ms	976.6815 Ops/s	909.3106 Ops/s	$\textbf{\color{#35bf28}+7.41\%}$
test_serialize_model	0.1322s	0.1307s	7.6497 Ops/s	6.8789 Ops/s	$\textbf{\color{#35bf28}+11.20\%}$
test_serialize_model_pickle	1.3604s	1.2132s	0.8243 Ops/s	0.8233 Ops/s	$\color{#35bf28}+0.12\%$
test_serialize_weights	0.1308s	0.1302s	7.6793 Ops/s	7.6795 Ops/s	$-0.00\%$
test_serialize_weights_returnearly	0.2239s	55.8344ms	17.9101 Ops/s	17.8015 Ops/s	$\color{#35bf28}+0.61\%$
test_serialize_weights_pickle	1.3457s	1.2128s	0.8245 Ops/s	0.8191 Ops/s	$\color{#35bf28}+0.66\%$
test_reshape_pytree	0.1564ms	36.4005μs	27.4721 KOps/s	26.7927 KOps/s	$\color{#35bf28}+2.54\%$
test_reshape_td	73.6510μs	41.4370μs	24.1330 KOps/s	23.8291 KOps/s	$\color{#35bf28}+1.28\%$
test_view_pytree	0.1703ms	36.4133μs	27.4625 KOps/s	27.0844 KOps/s	$\color{#35bf28}+1.40\%$
test_view_td	0.1727ms	46.3442μs	21.5777 KOps/s	21.4751 KOps/s	$\color{#35bf28}+0.48\%$
test_unbind_pytree	0.1359ms	35.4885μs	28.1782 KOps/s	27.8468 KOps/s	$\color{#35bf28}+1.19\%$
test_unbind_td	0.5523ms	42.6241μs	23.4609 KOps/s	21.9740 KOps/s	$\textbf{\color{#35bf28}+6.77\%}$
test_split_pytree	0.5121ms	47.2680μs	21.1560 KOps/s	21.4317 KOps/s	$\color{#d91a1a}-1.29\%$
test_split_td	0.2733ms	59.1310μs	16.9116 KOps/s	14.4819 KOps/s	$\textbf{\color{#35bf28}+16.78\%}$
test_add_pytree	0.2093ms	59.2161μs	16.8873 KOps/s	16.6475 KOps/s	$\color{#35bf28}+1.44\%$
test_add_td	0.2438ms	89.3914μs	11.1868 KOps/s	10.1921 KOps/s	$\textbf{\color{#35bf28}+9.76\%}$
test_compile_add_one_nested[tensordict-compile]	0.3098ms	0.1634ms	6.1183 KOps/s	5.9152 KOps/s	$\color{#35bf28}+3.43\%$
test_compile_add_one_nested[tensordict-eager]	0.5504ms	0.1527ms	6.5476 KOps/s	6.3960 KOps/s	$\color{#35bf28}+2.37\%$
test_compile_add_one_nested[pytree-compile]	0.5701ms	0.1582ms	6.3226 KOps/s	6.1999 KOps/s	$\color{#35bf28}+1.98\%$
test_compile_add_one_nested[pytree-eager]	0.5876ms	0.1830ms	5.4638 KOps/s	5.4085 KOps/s	$\color{#35bf28}+1.02\%$
test_compile_copy_nested[tensordict-compile]	0.4304ms	21.8075μs	45.8557 KOps/s	45.3592 KOps/s	$\color{#35bf28}+1.09\%$
test_compile_copy_nested[tensordict-eager]	0.4710ms	45.6090μs	21.9255 KOps/s	21.9116 KOps/s	$\color{#35bf28}+0.06\%$
test_compile_copy_nested[pytree-compile]	0.4669ms	68.5084μs	14.5967 KOps/s	14.7691 KOps/s	$\color{#d91a1a}-1.17\%$
test_compile_copy_nested[pytree-eager]	0.4440ms	52.6213μs	19.0037 KOps/s	19.3001 KOps/s	$\color{#d91a1a}-1.54\%$
test_compile_add_one_flat[tensordict-compile]	0.4529ms	0.3184ms	3.1411 KOps/s	3.1147 KOps/s	$\color{#35bf28}+0.85\%$
test_compile_add_one_flat[tensordict-eager]	0.6074ms	0.2147ms	4.6584 KOps/s	4.6036 KOps/s	$\color{#35bf28}+1.19\%$
test_compile_add_one_flat[tensorclass-compile]	0.2836ms	0.1362ms	7.3443 KOps/s	7.2187 KOps/s	$\color{#35bf28}+1.74\%$
test_compile_add_one_flat[tensorclass-eager]	0.4667ms	62.9575μs	15.8837 KOps/s	15.7623 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_add_one_flat[pytree-compile]	0.4607ms	0.3252ms	3.0749 KOps/s	3.0350 KOps/s	$\color{#35bf28}+1.32\%$
test_compile_add_one_flat[pytree-eager]	1.0318ms	0.6500ms	1.5385 KOps/s	1.6019 KOps/s	$\color{#d91a1a}-3.96\%$
test_compile_add_self_flat[tensordict-eager]	0.6605ms	0.2553ms	3.9164 KOps/s	3.8592 KOps/s	$\color{#35bf28}+1.48\%$
test_compile_add_self_flat[tensordict-compile]	0.5183ms	0.3217ms	3.1086 KOps/s	3.0925 KOps/s	$\color{#35bf28}+0.52\%$
test_compile_add_self_flat[tensorclass-eager]	0.4922ms	73.3706μs	13.6294 KOps/s	14.1336 KOps/s	$\color{#d91a1a}-3.57\%$
test_compile_add_self_flat[tensorclass-compile]	0.5313ms	0.1362ms	7.3412 KOps/s	7.4573 KOps/s	$\color{#d91a1a}-1.56\%$
test_compile_add_self_flat[pytree-eager]	0.9240ms	0.5133ms	1.9482 KOps/s	1.9203 KOps/s	$\color{#35bf28}+1.45\%$
test_compile_add_self_flat[pytree-compile]	0.4776ms	0.3262ms	3.0658 KOps/s	3.0281 KOps/s	$\color{#35bf28}+1.24\%$
test_compile_copy_flat[tensordict-compile]	0.4206ms	18.3225μs	54.5776 KOps/s	54.3835 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_copy_flat[tensordict-eager]	0.4083ms	28.7589μs	34.7718 KOps/s	34.4127 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_copy_flat[pytree-compile]	0.2406ms	69.7395μs	14.3391 KOps/s	14.1845 KOps/s	$\color{#35bf28}+1.09\%$
test_compile_copy_flat[pytree-eager]	0.4553ms	52.1770μs	19.1655 KOps/s	19.1614 KOps/s	$\color{#35bf28}+0.02\%$
test_compile_assign_and_add[tensordict-compile]	2.3991ms	0.8302ms	1.2046 KOps/s	1.0996 KOps/s	$\textbf{\color{#35bf28}+9.55\%}$
test_compile_assign_and_add[tensordict-eager]	3.7344ms	3.3650ms	297.1794 Ops/s	309.5542 Ops/s	$\color{#d91a1a}-4.00\%$
test_compile_assign_and_add[pytree-compile]	2.4014ms	0.8461ms	1.1819 KOps/s	1.0788 KOps/s	$\textbf{\color{#35bf28}+9.56\%}$
test_compile_assign_and_add[pytree-eager]	3.6678ms	3.2047ms	312.0444 Ops/s	306.9339 Ops/s	$\color{#35bf28}+1.67\%$
test_compile_indexing[tensor-tensordict-compile]	0.3045ms	0.1244ms	8.0413 KOps/s	8.0439 KOps/s	$\color{#d91a1a}-0.03\%$
test_compile_indexing[tensor-tensordict-eager]	0.4594ms	62.2380μs	16.0674 KOps/s	16.2546 KOps/s	$\color{#d91a1a}-1.15\%$
test_compile_indexing[tensor-tensorclass-compile]	0.2688ms	0.1235ms	8.0973 KOps/s	8.4906 KOps/s	$\color{#d91a1a}-4.63\%$
test_compile_indexing[tensor-tensorclass-eager]	0.2255ms	47.1888μs	21.1915 KOps/s	23.4059 KOps/s	$\textbf{\color{#d91a1a}-9.46\%}$
test_compile_indexing[tensor-pytree-compile]	0.2803ms	0.1231ms	8.1213 KOps/s	8.4178 KOps/s	$\color{#d91a1a}-3.52\%$
test_compile_indexing[tensor-pytree-eager]	0.4351ms	45.1487μs	22.1490 KOps/s	23.5004 KOps/s	$\textbf{\color{#d91a1a}-5.75\%}$
test_compile_indexing[slice-tensordict-compile]	0.5608ms	0.1542ms	6.4850 KOps/s	6.5438 KOps/s	$\color{#d91a1a}-0.90\%$
test_compile_indexing[slice-tensordict-eager]	0.4180ms	26.8091μs	37.3007 KOps/s	37.1050 KOps/s	$\color{#35bf28}+0.53\%$
test_compile_indexing[slice-tensorclass-compile]	0.2960ms	0.1449ms	6.9009 KOps/s	6.8299 KOps/s	$\color{#35bf28}+1.04\%$
test_compile_indexing[slice-tensorclass-eager]	0.4257ms	21.7490μs	45.9792 KOps/s	45.3865 KOps/s	$\color{#35bf28}+1.31\%$
test_compile_indexing[slice-pytree-compile]	0.5658ms	0.1458ms	6.8584 KOps/s	6.8147 KOps/s	$\color{#35bf28}+0.64\%$
test_compile_indexing[slice-pytree-eager]	0.4097ms	25.1111μs	39.8230 KOps/s	45.0698 KOps/s	$\textbf{\color{#d91a1a}-11.64\%}$
test_compile_indexing[int-tensordict-compile]	0.5449ms	0.1527ms	6.5495 KOps/s	6.5138 KOps/s	$\color{#35bf28}+0.55\%$
test_compile_indexing[int-tensordict-eager]	0.4630ms	26.6045μs	37.5876 KOps/s	36.7807 KOps/s	$\color{#35bf28}+2.19\%$
test_compile_indexing[int-tensorclass-compile]	0.5335ms	0.1461ms	6.8449 KOps/s	6.8245 KOps/s	$\color{#35bf28}+0.30\%$
test_compile_indexing[int-tensorclass-eager]	0.4103ms	21.6698μs	46.1471 KOps/s	45.7231 KOps/s	$\color{#35bf28}+0.93\%$
test_compile_indexing[int-pytree-compile]	0.5426ms	0.1460ms	6.8507 KOps/s	6.8249 KOps/s	$\color{#35bf28}+0.38\%$
test_compile_indexing[int-pytree-eager]	0.1480ms	21.5909μs	46.3158 KOps/s	45.7294 KOps/s	$\color{#35bf28}+1.28\%$
test_mod_add[eager]	0.2173ms	33.4338μs	29.9099 KOps/s	29.1580 KOps/s	$\color{#35bf28}+2.58\%$
test_mod_add[compile]	0.2397ms	77.8976μs	12.8374 KOps/s	12.8783 KOps/s	$\color{#d91a1a}-0.32\%$
test_mod_add[compile-overhead]	0.3102ms	0.1540ms	6.4954 KOps/s	5.6300 KOps/s	$\textbf{\color{#35bf28}+15.37\%}$
test_mod_wrap[eager]	0.4464ms	0.2537ms	3.9409 KOps/s	4.1129 KOps/s	$\color{#d91a1a}-4.18\%$
test_mod_wrap[compile]	0.4338ms	0.2834ms	3.5286 KOps/s	3.5027 KOps/s	$\color{#35bf28}+0.74\%$
test_mod_wrap[compile-overhead]	7.7923ms	4.1204ms	242.6965 Ops/s	243.7769 Ops/s	$\color{#d91a1a}-0.44\%$
test_mod_wrap_and_backward[eager]	1.6594ms	1.3365ms	748.2189 Ops/s	705.5210 Ops/s	$\textbf{\color{#35bf28}+6.05\%}$
test_mod_wrap_and_backward[compile]	1.5125ms	1.2576ms	795.1582 Ops/s	719.7293 Ops/s	$\textbf{\color{#35bf28}+10.48\%}$
test_mod_wrap_and_backward[compile-overhead]	1.3438ms	0.9116ms	1.0970 KOps/s	955.8758 Ops/s	$\textbf{\color{#35bf28}+14.76\%}$
test_seq_add[eager]	0.2458ms	97.9066μs	10.2138 KOps/s	9.8975 KOps/s	$\color{#35bf28}+3.20\%$
test_seq_add[compile]	0.2380ms	87.1892μs	11.4693 KOps/s	11.3944 KOps/s	$\color{#35bf28}+0.66\%$
test_seq_add[compile-overhead]	0.2740ms	0.1257ms	7.9563 KOps/s	7.7745 KOps/s	$\color{#35bf28}+2.34\%$
test_seq_wrap[eager]	0.5166ms	0.3756ms	2.6627 KOps/s	2.5319 KOps/s	$\textbf{\color{#35bf28}+5.17\%}$
test_seq_wrap[compile]	0.4500ms	0.3008ms	3.3248 KOps/s	3.2818 KOps/s	$\color{#35bf28}+1.31\%$
test_seq_wrap[compile-overhead]	0.3734ms	0.2226ms	4.4915 KOps/s	4.4314 KOps/s	$\color{#35bf28}+1.36\%$
test_func_call_runtime[False-eager]	0.8984ms	0.7251ms	1.3792 KOps/s	1.2802 KOps/s	$\textbf{\color{#35bf28}+7.73\%}$
test_func_call_runtime[False-compile]	0.9765ms	0.7573ms	1.3204 KOps/s	1.2909 KOps/s	$\color{#35bf28}+2.29\%$
test_func_call_runtime[False-compile-overhead]	0.5438ms	0.3649ms	2.7403 KOps/s	2.7358 KOps/s	$\color{#35bf28}+0.16\%$
test_func_call_runtime[True-eager]	1.0699ms	0.8826ms	1.1330 KOps/s	1.1198 KOps/s	$\color{#35bf28}+1.18\%$
test_func_call_runtime[True-compile]	0.9377ms	0.7799ms	1.2823 KOps/s	1.2773 KOps/s	$\color{#35bf28}+0.39\%$
test_func_call_runtime[True-compile-overhead]	0.5759ms	0.3825ms	2.6141 KOps/s	2.6055 KOps/s	$\color{#35bf28}+0.33\%$
test_func_call_cm_runtime[False-eager]	0.8678ms	0.7230ms	1.3831 KOps/s	1.3188 KOps/s	$\color{#35bf28}+4.87\%$
test_func_call_cm_runtime[False-compile]	0.9123ms	0.7590ms	1.3175 KOps/s	1.2954 KOps/s	$\color{#35bf28}+1.71\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5045ms	0.3639ms	2.7478 KOps/s	2.7318 KOps/s	$\color{#35bf28}+0.58\%$
test_func_call_cm_runtime[True-eager]	1.1897ms	0.9938ms	1.0062 KOps/s	1.0035 KOps/s	$\color{#35bf28}+0.27\%$
test_func_call_cm_runtime[True-compile]	0.9863ms	0.8077ms	1.2380 KOps/s	1.2180 KOps/s	$\color{#35bf28}+1.64\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5667ms	0.4093ms	2.4434 KOps/s	2.4200 KOps/s	$\color{#35bf28}+0.96\%$
test_vmap_func_call_cm_runtime[eager]	2.6249ms	2.0356ms	491.2613 Ops/s	489.1744 Ops/s	$\color{#35bf28}+0.43\%$
test_vmap_func_call_cm_runtime[compile]	1.0223ms	0.8263ms	1.2101 KOps/s	1.2035 KOps/s	$\color{#35bf28}+0.55\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.5308ms	0.4163ms	2.4019 KOps/s	2.4138 KOps/s	$\color{#d91a1a}-0.49\%$
test_distributed	6.4762ms	0.1791ms	5.5831 KOps/s	8.7140 KOps/s	$\textbf{\color{#d91a1a}-35.93\%}$
test_tdmodule	45.6110μs	14.8708μs	67.2459 KOps/s	60.1476 KOps/s	$\textbf{\color{#35bf28}+11.80\%}$
test_tdmodule_dispatch	62.0510μs	29.6577μs	33.7181 KOps/s	31.3906 KOps/s	$\textbf{\color{#35bf28}+7.41\%}$
test_tdseq	35.3800μs	16.3612μs	61.1201 KOps/s	56.8305 KOps/s	$\textbf{\color{#35bf28}+7.55\%}$
test_tdseq_dispatch	60.7020μs	32.6996μs	30.5814 KOps/s	28.3619 KOps/s	$\textbf{\color{#35bf28}+7.83\%}$
test_instantiation_functorch	2.0698ms	1.8945ms	527.8395 Ops/s	520.0393 Ops/s	$\color{#35bf28}+1.50\%$
test_exec_functorch	0.3547ms	0.2093ms	4.7781 KOps/s	4.6817 KOps/s	$\color{#35bf28}+2.06\%$
test_exec_functional_call	0.3963ms	0.2077ms	4.8143 KOps/s	4.6846 KOps/s	$\color{#35bf28}+2.77\%$
test_exec_td_decorator	0.4463ms	0.2543ms	3.9326 KOps/s	3.8467 KOps/s	$\color{#35bf28}+2.23\%$
test_vmap_mlp_speed_decorator[True-True]	0.8523ms	0.6628ms	1.5086 KOps/s	1.5021 KOps/s	$\color{#35bf28}+0.44\%$
test_vmap_mlp_speed_decorator[True-False]	0.8608ms	0.6610ms	1.5129 KOps/s	1.5079 KOps/s	$\color{#35bf28}+0.33\%$
test_vmap_mlp_speed_decorator[False-True]	0.7589ms	0.5783ms	1.7291 KOps/s	1.7279 KOps/s	$\color{#35bf28}+0.07\%$
test_vmap_mlp_speed_decorator[False-False]	0.7810ms	0.5821ms	1.7178 KOps/s	1.7266 KOps/s	$\color{#d91a1a}-0.51\%$
test_vmap_transformer_speed_decorator[True-True]	19.4123ms	19.0125ms	52.5971 Ops/s	52.6883 Ops/s	$\color{#d91a1a}-0.17\%$
test_vmap_transformer_speed_decorator[True-False]	19.2397ms	19.0099ms	52.6042 Ops/s	52.5981 Ops/s	$\color{#35bf28}+0.01\%$
test_vmap_transformer_speed_decorator[False-True]	19.1731ms	18.8795ms	52.9676 Ops/s	53.1178 Ops/s	$\color{#d91a1a}-0.28\%$
test_vmap_transformer_speed_decorator[False-False]	19.2046ms	18.8750ms	52.9802 Ops/s	52.6553 Ops/s	$\color{#35bf28}+0.62\%$
test_to_module_speed[True]	1.4757ms	0.9570ms	1.0449 KOps/s	1.0457 KOps/s	$\color{#d91a1a}-0.07\%$
test_to_module_speed[False]	1.3281ms	0.9462ms	1.0569 KOps/s	1.0639 KOps/s	$\color{#d91a1a}-0.65\%$
test_tc_init	71.2210μs	34.2584μs	29.1900 KOps/s	25.8693 KOps/s	$\textbf{\color{#35bf28}+12.84\%}$
test_tc_init_nested	0.2539ms	69.6717μs	14.3530 KOps/s	12.5843 KOps/s	$\textbf{\color{#35bf28}+14.05\%}$
test_tc_first_layer_tensor	27.2861μs	0.7179μs	1.3929 MOps/s	1.3959 MOps/s	$\color{#d91a1a}-0.22\%$
test_tc_first_layer_nontensor	18.9310μs	2.4453μs	408.9441 KOps/s	436.9755 KOps/s	$\textbf{\color{#d91a1a}-6.41\%}$
test_tc_second_layer_tensor	9.1527μs	1.4537μs	687.9199 KOps/s	683.2213 KOps/s	$\color{#35bf28}+0.69\%$
test_tc_second_layer_nontensor	41.4810μs	3.1917μs	313.3079 KOps/s	322.5501 KOps/s	$\color{#d91a1a}-2.87\%$
test_unbind	0.2043s	11.1398ms	89.7684 Ops/s	88.3935 Ops/s	$\color{#35bf28}+1.56\%$
test_full_like	0.7905ms	0.5763ms	1.7353 KOps/s	1.7388 KOps/s	$\color{#d91a1a}-0.20\%$
test_zeros_like	0.3624ms	0.1982ms	5.0450 KOps/s	5.0438 KOps/s	$\color{#35bf28}+0.02\%$
test_ones_like	0.3469ms	0.1981ms	5.0483 KOps/s	5.0468 KOps/s	$\color{#35bf28}+0.03\%$
test_clone	0.5550ms	0.4153ms	2.4080 KOps/s	2.4067 KOps/s	$\color{#35bf28}+0.06\%$
test_squeeze	0.1228ms	9.4100μs	106.2700 KOps/s	106.0284 KOps/s	$\color{#35bf28}+0.23\%$
test_unsqueeze	0.2168ms	72.3942μs	13.8133 KOps/s	13.5006 KOps/s	$\color{#35bf28}+2.32\%$
test_split	0.4019ms	0.1647ms	6.0712 KOps/s	5.9016 KOps/s	$\color{#35bf28}+2.87\%$
test_permute	0.2822ms	0.1754ms	5.7007 KOps/s	5.6767 KOps/s	$\color{#35bf28}+0.42\%$
test_stack	1.2757ms	0.8697ms	1.1498 KOps/s	1.2025 KOps/s	$\color{#d91a1a}-4.38\%$
test_cat	1.3439ms	1.2317ms	811.8582 Ops/s	811.9529 Ops/s	$\color{#d91a1a}-0.01\%$