[BugFix] Fix _foreach_copy_ for older versions of PT

[vmoens]

Stack from ghstack (oldest at bottom):

• -> [BugFix] Fix _foreach_copy_ for older versions of PT #1035

[github-actions]

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

Total Benchmarks: 216. Improved: $\large\color{#35bf28}34$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	52.8800μs	23.5445μs	42.4728 KOps/s	40.9754 KOps/s	$\color{#35bf28}+3.65\%$
test_plain_set_stack_nested	59.8920μs	23.6062μs	42.3617 KOps/s	39.8222 KOps/s	$\textbf{\color{#35bf28}+6.38\%}$
test_plain_set_nested_inplace	65.4130μs	25.5057μs	39.2070 KOps/s	36.5333 KOps/s	$\textbf{\color{#35bf28}+7.32\%}$
test_plain_set_stack_nested_inplace	90.0200μs	25.5773μs	39.0971 KOps/s	36.6362 KOps/s	$\textbf{\color{#35bf28}+6.72\%}$
test_items	34.0140μs	4.0986μs	243.9850 KOps/s	214.6165 KOps/s	$\textbf{\color{#35bf28}+13.68\%}$
test_items_nested	0.5990ms	0.3775ms	2.6490 KOps/s	2.6006 KOps/s	$\color{#35bf28}+1.86\%$
test_items_nested_locked	0.5826ms	0.3807ms	2.6271 KOps/s	2.5826 KOps/s	$\color{#35bf28}+1.72\%$
test_items_nested_leaf	0.1523ms	80.1237μs	12.4807 KOps/s	12.2991 KOps/s	$\color{#35bf28}+1.48\%$
test_items_stack_nested	0.5711ms	0.3845ms	2.6006 KOps/s	2.5736 KOps/s	$\color{#35bf28}+1.05\%$
test_items_stack_nested_leaf	0.1671ms	83.9236μs	11.9156 KOps/s	11.9296 KOps/s	$\color{#d91a1a}-0.12\%$
test_items_stack_nested_locked	0.6217ms	0.3841ms	2.6038 KOps/s	2.5926 KOps/s	$\color{#35bf28}+0.43\%$
test_keys	25.1670μs	3.6341μs	275.1734 KOps/s	275.3318 KOps/s	$\color{#d91a1a}-0.06\%$
test_keys_nested	0.2510ms	0.1349ms	7.4124 KOps/s	7.1638 KOps/s	$\color{#35bf28}+3.47\%$
test_keys_nested_locked	0.6918ms	0.1436ms	6.9620 KOps/s	6.9163 KOps/s	$\color{#35bf28}+0.66\%$
test_keys_nested_leaf	0.1948ms	0.1179ms	8.4834 KOps/s	8.1121 KOps/s	$\color{#35bf28}+4.58\%$
test_keys_stack_nested	0.2492ms	0.1363ms	7.3367 KOps/s	7.1401 KOps/s	$\color{#35bf28}+2.75\%$
test_keys_stack_nested_leaf	0.1868ms	0.1201ms	8.3280 KOps/s	8.0927 KOps/s	$\color{#35bf28}+2.91\%$
test_keys_stack_nested_locked	0.2611ms	0.1420ms	7.0406 KOps/s	6.8039 KOps/s	$\color{#35bf28}+3.48\%$
test_values	8.5760μs	1.0437μs	958.1477 KOps/s	945.6531 KOps/s	$\color{#35bf28}+1.32\%$
test_values_nested	0.1687ms	92.6996μs	10.7875 KOps/s	10.1541 KOps/s	$\textbf{\color{#35bf28}+6.24\%}$
test_values_nested_locked	0.1688ms	93.6565μs	10.6773 KOps/s	10.4638 KOps/s	$\color{#35bf28}+2.04\%$
test_values_nested_leaf	0.1464ms	78.8075μs	12.6891 KOps/s	12.2573 KOps/s	$\color{#35bf28}+3.52\%$
test_values_stack_nested	0.1636ms	93.1914μs	10.7306 KOps/s	10.5700 KOps/s	$\color{#35bf28}+1.52\%$
test_values_stack_nested_leaf	0.1616ms	78.3801μs	12.7583 KOps/s	12.2240 KOps/s	$\color{#35bf28}+4.37\%$
test_values_stack_nested_locked	0.1594ms	94.6361μs	10.5668 KOps/s	10.3204 KOps/s	$\color{#35bf28}+2.39\%$
test_membership	6.0499μs	0.7272μs	1.3751 MOps/s	1.1519 MOps/s	$\textbf{\color{#35bf28}+19.38\%}$
test_membership_nested	28.7540μs	2.7318μs	366.0596 KOps/s	367.3263 KOps/s	$\color{#d91a1a}-0.34\%$
test_membership_nested_leaf	28.3530μs	2.7204μs	367.5987 KOps/s	361.7287 KOps/s	$\color{#35bf28}+1.62\%$
test_membership_stacked_nested	26.3590μs	2.7440μs	364.4311 KOps/s	367.2817 KOps/s	$\color{#d91a1a}-0.78\%$
test_membership_stacked_nested_leaf	43.7820μs	2.7066μs	369.4620 KOps/s	363.5475 KOps/s	$\color{#35bf28}+1.63\%$
test_membership_nested_last	30.8980μs	4.2620μs	234.6307 KOps/s	238.6715 KOps/s	$\color{#d91a1a}-1.69\%$
test_membership_nested_leaf_last	50.3050μs	4.2124μs	237.3930 KOps/s	235.4393 KOps/s	$\color{#35bf28}+0.83\%$
test_membership_stacked_nested_last	45.6160μs	5.8279μs	171.5871 KOps/s	201.0910 KOps/s	$\textbf{\color{#d91a1a}-14.67\%}$
test_membership_stacked_nested_leaf_last	43.1210μs	5.8993μs	169.5125 KOps/s	198.3664 KOps/s	$\textbf{\color{#d91a1a}-14.55\%}$
test_nested_getleaf	55.2340μs	10.5302μs	94.9645 KOps/s	93.9151 KOps/s	$\color{#35bf28}+1.12\%$
test_nested_get	38.4920μs	10.0357μs	99.6439 KOps/s	99.3022 KOps/s	$\color{#35bf28}+0.34\%$
test_stacked_getleaf	50.5550μs	10.5022μs	95.2179 KOps/s	94.8649 KOps/s	$\color{#35bf28}+0.37\%$
test_stacked_get	35.7870μs	10.0645μs	99.3593 KOps/s	98.6879 KOps/s	$\color{#35bf28}+0.68\%$
test_nested_getitemleaf	32.4810μs	10.9496μs	91.3276 KOps/s	91.3077 KOps/s	$\color{#35bf28}+0.02\%$
test_nested_getitem	33.4330μs	10.3508μs	96.6107 KOps/s	96.7918 KOps/s	$\color{#d91a1a}-0.19\%$
test_stacked_getitemleaf	29.7450μs	10.9078μs	91.6777 KOps/s	89.9944 KOps/s	$\color{#35bf28}+1.87\%$
test_stacked_getitem	49.1230μs	10.2073μs	97.9693 KOps/s	96.8683 KOps/s	$\color{#35bf28}+1.14\%$
test_lock_nested	88.9962ms	0.6058ms	1.6508 KOps/s	1.9551 KOps/s	$\textbf{\color{#d91a1a}-15.56\%}$
test_lock_stack_nested	0.8422ms	0.4668ms	2.1420 KOps/s	2.0722 KOps/s	$\color{#35bf28}+3.37\%$
test_unlock_nested	90.6623ms	0.5207ms	1.9205 KOps/s	2.3270 KOps/s	$\textbf{\color{#d91a1a}-17.47\%}$
test_unlock_stack_nested	0.5251ms	0.3766ms	2.6554 KOps/s	2.5015 KOps/s	$\textbf{\color{#35bf28}+6.16\%}$
test_flatten_speed	0.3155ms	0.1031ms	9.6948 KOps/s	9.9331 KOps/s	$\color{#d91a1a}-2.40\%$
test_unflatten_speed	0.7335ms	0.5149ms	1.9422 KOps/s	1.9364 KOps/s	$\color{#35bf28}+0.30\%$
test_common_ops	3.8587ms	1.1047ms	905.2573 Ops/s	870.6569 Ops/s	$\color{#35bf28}+3.97\%$
test_creation	74.9210μs	2.0569μs	486.1687 KOps/s	478.9606 KOps/s	$\color{#35bf28}+1.50\%$
test_creation_empty	48.4810μs	17.6140μs	56.7731 KOps/s	51.7927 KOps/s	$\textbf{\color{#35bf28}+9.62\%}$
test_creation_nested_1	59.2110μs	21.2707μs	47.0130 KOps/s	43.9139 KOps/s	$\textbf{\color{#35bf28}+7.06\%}$
test_creation_nested_2	73.8290μs	25.5743μs	39.1018 KOps/s	37.4714 KOps/s	$\color{#35bf28}+4.35\%$
test_clone	0.1531ms	17.0243μs	58.7396 KOps/s	57.7807 KOps/s	$\color{#35bf28}+1.66\%$
test_getitem[int]	1.0153ms	16.3852μs	61.0307 KOps/s	57.9631 KOps/s	$\textbf{\color{#35bf28}+5.29\%}$
test_getitem[slice_int]	0.1350ms	29.7008μs	33.6692 KOps/s	31.4830 KOps/s	$\textbf{\color{#35bf28}+6.94\%}$
test_getitem[range]	0.5151ms	57.7821μs	17.3064 KOps/s	17.2950 KOps/s	$\color{#35bf28}+0.07\%$
test_getitem[tuple]	0.1611ms	25.1329μs	39.7885 KOps/s	38.2233 KOps/s	$\color{#35bf28}+4.09\%$
test_getitem[list]	0.1794ms	51.8156μs	19.2992 KOps/s	18.8411 KOps/s	$\color{#35bf28}+2.43\%$
test_setitem_dim[int]	68.2080μs	31.7322μs	31.5138 KOps/s	30.2722 KOps/s	$\color{#35bf28}+4.10\%$
test_setitem_dim[slice_int]	95.8810μs	58.7430μs	17.0233 KOps/s	15.8141 KOps/s	$\textbf{\color{#35bf28}+7.65\%}$
test_setitem_dim[range]	0.1299ms	82.3134μs	12.1487 KOps/s	12.0315 KOps/s	$\color{#35bf28}+0.97\%$
test_setitem_dim[tuple]	84.3780μs	47.6539μs	20.9846 KOps/s	20.1249 KOps/s	$\color{#35bf28}+4.27\%$
test_setitem	85.4610μs	29.4237μs	33.9862 KOps/s	33.2186 KOps/s	$\color{#35bf28}+2.31\%$
test_set	0.2562ms	28.8338μs	34.6815 KOps/s	33.8476 KOps/s	$\color{#35bf28}+2.46\%$
test_set_shared	3.1137ms	0.2151ms	4.6482 KOps/s	4.6240 KOps/s	$\color{#35bf28}+0.52\%$
test_update	0.2789ms	36.6901μs	27.2553 KOps/s	25.7331 KOps/s	$\textbf{\color{#35bf28}+5.92\%}$
test_update_nested	0.1719ms	47.2561μs	21.1613 KOps/s	20.0760 KOps/s	$\textbf{\color{#35bf28}+5.41\%}$
test_update__nested	0.9757ms	44.9865μs	22.2289 KOps/s	22.1386 KOps/s	$\color{#35bf28}+0.41\%$
test_set_nested	0.2562ms	32.7204μs	30.5619 KOps/s	31.2523 KOps/s	$\color{#d91a1a}-2.21\%$
test_set_nested_new	0.2499ms	36.9826μs	27.0397 KOps/s	27.2985 KOps/s	$\color{#d91a1a}-0.95\%$
test_select	0.2444ms	53.8883μs	18.5569 KOps/s	17.9964 KOps/s	$\color{#35bf28}+3.11\%$
test_select_nested	0.1175ms	59.8319μs	16.7135 KOps/s	16.6925 KOps/s	$\color{#35bf28}+0.13\%$
test_exclude_nested	0.1450ms	75.4418μs	13.2553 KOps/s	13.2511 KOps/s	$\color{#35bf28}+0.03\%$
test_empty[True]	0.4772ms	0.3534ms	2.8299 KOps/s	2.8191 KOps/s	$\color{#35bf28}+0.38\%$
test_empty[False]	6.6084μs	1.1885μs	841.3719 KOps/s	793.5746 KOps/s	$\textbf{\color{#35bf28}+6.02\%}$
test_unbind_speed	0.4077ms	0.3074ms	3.2526 KOps/s	3.2169 KOps/s	$\color{#35bf28}+1.11\%$
test_unbind_speed_stack0	0.4291ms	0.2912ms	3.4337 KOps/s	3.2457 KOps/s	$\textbf{\color{#35bf28}+5.79\%}$
test_unbind_speed_stack1	0.1012s	0.7958ms	1.2565 KOps/s	1.3309 KOps/s	$\textbf{\color{#d91a1a}-5.59\%}$
test_split	3.1298ms	1.9929ms	501.7789 Ops/s	435.0395 Ops/s	$\textbf{\color{#35bf28}+15.34\%}$
test_chunk	93.3572ms	2.1852ms	457.6277 Ops/s	436.1655 Ops/s	$\color{#35bf28}+4.92\%$
test_creation[device0]	0.2330ms	0.1144ms	8.7401 KOps/s	8.4894 KOps/s	$\color{#35bf28}+2.95\%$
test_creation_from_tensor	4.8757ms	0.1162ms	8.6078 KOps/s	8.3873 KOps/s	$\color{#35bf28}+2.63\%$
test_add_one[memmap_tensor0]	0.2311ms	6.9211μs	144.4856 KOps/s	134.1030 KOps/s	$\textbf{\color{#35bf28}+7.74\%}$
test_contiguous[memmap_tensor0]	28.1530μs	1.8648μs	536.2458 KOps/s	522.0150 KOps/s	$\color{#35bf28}+2.73\%$
test_stack[memmap_tensor0]	63.9500μs	5.4033μs	185.0729 KOps/s	172.5575 KOps/s	$\textbf{\color{#35bf28}+7.25\%}$
test_memmaptd_index	1.2009ms	0.4107ms	2.4348 KOps/s	2.3665 KOps/s	$\color{#35bf28}+2.89\%$
test_memmaptd_index_astensor	0.7984ms	0.5091ms	1.9642 KOps/s	1.9003 KOps/s	$\color{#35bf28}+3.36\%$
test_memmaptd_index_op	1.6331ms	1.0091ms	991.0137 Ops/s	925.6551 Ops/s	$\textbf{\color{#35bf28}+7.06\%}$
test_serialize_model	0.2160s	0.1290s	7.7490 Ops/s	8.3823 Ops/s	$\textbf{\color{#d91a1a}-7.56\%}$
test_serialize_model_pickle	0.4987s	0.4072s	2.4557 Ops/s	2.5857 Ops/s	$\textbf{\color{#d91a1a}-5.03\%}$
test_serialize_weights	0.1294s	0.1166s	8.5775 Ops/s	7.5196 Ops/s	$\textbf{\color{#35bf28}+14.07\%}$
test_serialize_weights_returnearly	0.2489s	0.1715s	5.8317 Ops/s	6.2534 Ops/s	$\textbf{\color{#d91a1a}-6.74\%}$
test_serialize_weights_pickle	0.5001s	0.4159s	2.4046 Ops/s	2.3993 Ops/s	$\color{#35bf28}+0.22\%$
test_serialize_weights_filesystem	0.1508s	0.1439s	6.9492 Ops/s	6.9915 Ops/s	$\color{#d91a1a}-0.61\%$
test_serialize_model_filesystem	0.1608s	0.1525s	6.5594 Ops/s	6.1106 Ops/s	$\textbf{\color{#35bf28}+7.34\%}$
test_reshape_pytree	0.1138ms	38.6713μs	25.8590 KOps/s	26.0610 KOps/s	$\color{#d91a1a}-0.78\%$
test_reshape_td	0.1022ms	48.7562μs	20.5102 KOps/s	20.7172 KOps/s	$\color{#d91a1a}-1.00\%$
test_view_pytree	0.1001ms	37.9878μs	26.3242 KOps/s	25.9792 KOps/s	$\color{#35bf28}+1.33\%$
test_view_td	0.1074ms	54.1056μs	18.4824 KOps/s	19.2610 KOps/s	$\color{#d91a1a}-4.04\%$
test_unbind_pytree	75.5420μs	35.3564μs	28.2834 KOps/s	27.3340 KOps/s	$\color{#35bf28}+3.47\%$
test_unbind_td	0.3266ms	46.8382μs	21.3501 KOps/s	21.6995 KOps/s	$\color{#d91a1a}-1.61\%$
test_split_pytree	81.2730μs	37.3479μs	26.7753 KOps/s	26.3193 KOps/s	$\color{#35bf28}+1.73\%$
test_split_td	0.2043ms	57.5127μs	17.3875 KOps/s	16.5817 KOps/s	$\color{#35bf28}+4.86\%$
test_add_pytree	93.6870μs	43.9351μs	22.7609 KOps/s	22.0571 KOps/s	$\color{#35bf28}+3.19\%$
test_add_td	0.2280ms	85.2647μs	11.7282 KOps/s	11.4311 KOps/s	$\color{#35bf28}+2.60\%$
test_compile_add_one_nested[tensordict-compile]	0.1384ms	58.2395μs	17.1705 KOps/s	17.4359 KOps/s	$\color{#d91a1a}-1.52\%$
test_compile_add_one_nested[tensordict-eager]	0.4305ms	0.1980ms	5.0515 KOps/s	5.1825 KOps/s	$\color{#d91a1a}-2.53\%$
test_compile_add_one_nested[pytree-compile]	0.1512ms	57.1783μs	17.4891 KOps/s	17.8618 KOps/s	$\color{#d91a1a}-2.09\%$
test_compile_add_one_nested[pytree-eager]	0.2656ms	0.1376ms	7.2650 KOps/s	7.0667 KOps/s	$\color{#35bf28}+2.81\%$
test_compile_copy_nested[tensordict-compile]	61.0950μs	23.9636μs	41.7299 KOps/s	42.9859 KOps/s	$\color{#d91a1a}-2.92\%$
test_compile_copy_nested[tensordict-eager]	0.1554ms	75.2659μs	13.2862 KOps/s	13.3175 KOps/s	$\color{#d91a1a}-0.23\%$
test_compile_copy_nested[pytree-compile]	0.1479ms	74.2219μs	13.4731 KOps/s	13.4254 KOps/s	$\color{#35bf28}+0.36\%$
test_compile_copy_nested[pytree-eager]	0.1251ms	67.8824μs	14.7314 KOps/s	14.5859 KOps/s	$\color{#35bf28}+1.00\%$
test_compile_add_one_flat[tensordict-compile]	0.3685ms	0.1810ms	5.5242 KOps/s	5.5297 KOps/s	$\color{#d91a1a}-0.10\%$
test_compile_add_one_flat[tensordict-eager]	0.3907ms	0.2399ms	4.1692 KOps/s	4.1640 KOps/s	$\color{#35bf28}+0.13\%$
test_compile_add_one_flat[tensorclass-compile]	0.1016ms	47.4678μs	21.0669 KOps/s	21.2365 KOps/s	$\color{#d91a1a}-0.80\%$
test_compile_add_one_flat[tensorclass-eager]	0.4136ms	76.7435μs	13.0304 KOps/s	12.7776 KOps/s	$\color{#35bf28}+1.98\%$
test_compile_add_one_flat[pytree-compile]	0.2739ms	0.1734ms	5.7676 KOps/s	5.7290 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_add_one_flat[pytree-eager]	0.4689ms	0.2777ms	3.6014 KOps/s	3.4333 KOps/s	$\color{#35bf28}+4.90\%$
test_compile_add_self_flat[tensordict-eager]	0.4646ms	0.2757ms	3.6276 KOps/s	3.5951 KOps/s	$\color{#35bf28}+0.90\%$
test_compile_add_self_flat[tensordict-compile]	0.3781ms	0.1802ms	5.5508 KOps/s	5.5785 KOps/s	$\color{#d91a1a}-0.50\%$
test_compile_add_self_flat[tensorclass-eager]	0.1801ms	73.3073μs	13.6412 KOps/s	13.5762 KOps/s	$\color{#35bf28}+0.48\%$
test_compile_add_self_flat[tensorclass-compile]	0.1582ms	47.9641μs	20.8489 KOps/s	20.9809 KOps/s	$\color{#d91a1a}-0.63\%$
test_compile_add_self_flat[pytree-eager]	0.4585ms	0.2295ms	4.3577 KOps/s	4.3225 KOps/s	$\color{#35bf28}+0.81\%$
test_compile_add_self_flat[pytree-compile]	0.3797ms	0.1723ms	5.8035 KOps/s	5.6850 KOps/s	$\color{#35bf28}+2.08\%$
test_compile_copy_flat[tensordict-compile]	0.2596ms	0.1092ms	9.1582 KOps/s	8.8344 KOps/s	$\color{#35bf28}+3.66\%$
test_compile_copy_flat[tensordict-eager]	0.1939ms	78.2738μs	12.7757 KOps/s	11.9726 KOps/s	$\textbf{\color{#35bf28}+6.71\%}$
test_compile_copy_flat[pytree-compile]	0.1459ms	77.2340μs	12.9477 KOps/s	13.2093 KOps/s	$\color{#d91a1a}-1.98\%$
test_compile_copy_flat[pytree-eager]	0.1276ms	68.2885μs	14.6438 KOps/s	14.5260 KOps/s	$\color{#35bf28}+0.81\%$
test_compile_assign_and_add[tensordict-compile]	0.4013ms	0.1981ms	5.0473 KOps/s	5.1815 KOps/s	$\color{#d91a1a}-2.59\%$
test_compile_assign_and_add[tensordict-eager]	1.9826ms	1.7012ms	587.8192 Ops/s	567.8482 Ops/s	$\color{#35bf28}+3.52\%$
test_compile_assign_and_add[pytree-compile]	0.3985ms	0.1956ms	5.1125 KOps/s	5.2560 KOps/s	$\color{#d91a1a}-2.73\%$
test_compile_assign_and_add[pytree-eager]	1.7908ms	1.1071ms	903.2470 Ops/s	887.1507 Ops/s	$\color{#35bf28}+1.81\%$
test_compile_assign_and_add_stack[compile]	0.5352ms	0.4204ms	2.3789 KOps/s	2.3854 KOps/s	$\color{#d91a1a}-0.27\%$
test_compile_assign_and_add_stack[eager]	4.1010ms	3.8754ms	258.0404 Ops/s	244.4281 Ops/s	$\textbf{\color{#35bf28}+5.57\%}$
test_compile_indexing[tensor-tensordict-compile]	83.2570μs	34.2601μs	29.1885 KOps/s	29.7841 KOps/s	$\color{#d91a1a}-2.00\%$
test_compile_indexing[tensor-tensordict-eager]	0.6434ms	46.4505μs	21.5283 KOps/s	20.5776 KOps/s	$\color{#35bf28}+4.62\%$
test_compile_indexing[tensor-tensorclass-compile]	4.8755ms	29.9281μs	33.4134 KOps/s	34.2630 KOps/s	$\color{#d91a1a}-2.48\%$
test_compile_indexing[tensor-tensorclass-eager]	84.0080μs	27.9204μs	35.8161 KOps/s	35.2769 KOps/s	$\color{#35bf28}+1.53\%$
test_compile_indexing[tensor-pytree-compile]	81.1020μs	29.6591μs	33.7165 KOps/s	34.8974 KOps/s	$\color{#d91a1a}-3.38\%$
test_compile_indexing[tensor-pytree-eager]	0.1124ms	27.4074μs	36.4866 KOps/s	35.5108 KOps/s	$\color{#35bf28}+2.75\%$
test_compile_indexing[slice-tensordict-compile]	0.1410ms	72.9023μs	13.7170 KOps/s	13.6028 KOps/s	$\color{#35bf28}+0.84\%$
test_compile_indexing[slice-tensordict-eager]	0.5500ms	27.5079μs	36.3532 KOps/s	35.1045 KOps/s	$\color{#35bf28}+3.56\%$
test_compile_indexing[slice-tensorclass-compile]	0.1294ms	66.5336μs	15.0300 KOps/s	14.8106 KOps/s	$\color{#35bf28}+1.48\%$
test_compile_indexing[slice-tensorclass-eager]	61.8360μs	22.9128μs	43.6438 KOps/s	43.7330 KOps/s	$\color{#d91a1a}-0.20\%$
test_compile_indexing[slice-pytree-compile]	0.1533ms	66.6310μs	15.0080 KOps/s	14.9089 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_indexing[slice-pytree-eager]	81.9840μs	22.9614μs	43.5514 KOps/s	43.7475 KOps/s	$\color{#d91a1a}-0.45\%$
test_compile_indexing[int-tensordict-compile]	0.1411ms	72.3422μs	13.8232 KOps/s	13.5899 KOps/s	$\color{#35bf28}+1.72\%$
test_compile_indexing[int-tensordict-eager]	0.9131ms	27.5132μs	36.3461 KOps/s	35.2534 KOps/s	$\color{#35bf28}+3.10\%$
test_compile_indexing[int-tensorclass-compile]	0.1428ms	66.3083μs	15.0811 KOps/s	14.7630 KOps/s	$\color{#35bf28}+2.15\%$
test_compile_indexing[int-tensorclass-eager]	91.8920μs	22.8708μs	43.7239 KOps/s	43.3926 KOps/s	$\color{#35bf28}+0.76\%$
test_compile_indexing[int-pytree-compile]	0.1432ms	66.0903μs	15.1308 KOps/s	14.4321 KOps/s	$\color{#35bf28}+4.84\%$
test_compile_indexing[int-pytree-eager]	0.1133ms	22.7839μs	43.8906 KOps/s	43.7835 KOps/s	$\color{#35bf28}+0.24\%$
test_mod_add[eager]	0.1054ms	24.1882μs	41.3425 KOps/s	39.3816 KOps/s	$\color{#35bf28}+4.98\%$
test_mod_add[compile]	0.1073ms	37.4242μs	26.7207 KOps/s	26.3768 KOps/s	$\color{#35bf28}+1.30\%$
test_mod_add[compile-overhead]	86.5130μs	37.7638μs	26.4804 KOps/s	27.0643 KOps/s	$\color{#d91a1a}-2.16\%$
test_mod_wrap[eager]	0.3347ms	0.2080ms	4.8068 KOps/s	4.7259 KOps/s	$\color{#35bf28}+1.71\%$
test_mod_wrap[compile]	0.3753ms	0.2325ms	4.3011 KOps/s	4.2919 KOps/s	$\color{#35bf28}+0.22\%$
test_mod_wrap[compile-overhead]	0.3345ms	0.2305ms	4.3386 KOps/s	4.3312 KOps/s	$\color{#35bf28}+0.17\%$
test_mod_wrap_and_backward[eager]	12.3936ms	10.7155ms	93.3230 Ops/s	86.4881 Ops/s	$\textbf{\color{#35bf28}+7.90\%}$
test_mod_wrap_and_backward[compile]	12.5517ms	10.6348ms	94.0309 Ops/s	80.7215 Ops/s	$\textbf{\color{#35bf28}+16.49\%}$
test_mod_wrap_and_backward[compile-overhead]	11.5605ms	10.6760ms	93.6678 Ops/s	78.2329 Ops/s	$\textbf{\color{#35bf28}+19.73\%}$
test_seq_add[eager]	0.2769ms	89.8437μs	11.1304 KOps/s	10.8707 KOps/s	$\color{#35bf28}+2.39\%$
test_seq_add[compile]	0.1865ms	63.3898μs	15.7754 KOps/s	15.5066 KOps/s	$\color{#35bf28}+1.73\%$
test_seq_add[compile-overhead]	0.1615ms	61.8316μs	16.1730 KOps/s	15.9605 KOps/s	$\color{#35bf28}+1.33\%$
test_seq_wrap[eager]	1.0108ms	0.3797ms	2.6338 KOps/s	2.5352 KOps/s	$\color{#35bf28}+3.89\%$
test_seq_wrap[compile]	0.4119ms	0.2664ms	3.7537 KOps/s	3.6813 KOps/s	$\color{#35bf28}+1.97\%$
test_seq_wrap[compile-overhead]	0.5166ms	0.2698ms	3.7067 KOps/s	3.6683 KOps/s	$\color{#35bf28}+1.05\%$
test_func_call_runtime[False-eager]	0.8943ms	0.5263ms	1.9000 KOps/s	1.8735 KOps/s	$\color{#35bf28}+1.41\%$
test_func_call_runtime[False-compile]	0.6168ms	0.4982ms	2.0072 KOps/s	1.9695 KOps/s	$\color{#35bf28}+1.91\%$
test_func_call_runtime[False-compile-overhead]	0.6372ms	0.4991ms	2.0035 KOps/s	1.9400 KOps/s	$\color{#35bf28}+3.28\%$
test_func_call_runtime[True-eager]	0.8771ms	0.7375ms	1.3560 KOps/s	1.3336 KOps/s	$\color{#35bf28}+1.68\%$
test_func_call_runtime[True-compile]	1.0889ms	0.5175ms	1.9322 KOps/s	1.9371 KOps/s	$\color{#d91a1a}-0.25\%$
test_func_call_runtime[True-compile-overhead]	0.6270ms	0.5155ms	1.9398 KOps/s	1.9274 KOps/s	$\color{#35bf28}+0.64\%$
test_func_call_cm_runtime[False-eager]	0.8172ms	0.5275ms	1.8958 KOps/s	1.8904 KOps/s	$\color{#35bf28}+0.29\%$
test_func_call_cm_runtime[False-compile]	1.0801ms	0.5035ms	1.9860 KOps/s	1.9620 KOps/s	$\color{#35bf28}+1.22\%$
test_func_call_cm_runtime[False-compile-overhead]	0.5823ms	0.5017ms	1.9930 KOps/s	1.9579 KOps/s	$\color{#35bf28}+1.80\%$
test_func_call_cm_runtime[True-eager]	1.2377ms	0.9093ms	1.0997 KOps/s	1.0860 KOps/s	$\color{#35bf28}+1.26\%$
test_func_call_cm_runtime[True-compile]	0.9060ms	0.7497ms	1.3338 KOps/s	1.3168 KOps/s	$\color{#35bf28}+1.29\%$
test_func_call_cm_runtime[True-compile-overhead]	1.0720ms	0.7589ms	1.3177 KOps/s	1.2953 KOps/s	$\color{#35bf28}+1.73\%$
test_vmap_func_call_cm_runtime[eager]	3.5418ms	1.9432ms	514.6263 Ops/s	504.8087 Ops/s	$\color{#35bf28}+1.94\%$
test_vmap_func_call_cm_runtime[compile]	2.9311ms	2.0044ms	498.9140 Ops/s	489.9636 Ops/s	$\color{#35bf28}+1.83\%$
test_vmap_func_call_cm_runtime[compile-overhead]	2.6846ms	1.9943ms	501.4372 Ops/s	488.2073 Ops/s	$\color{#35bf28}+2.71\%$
test_distributed	0.2797ms	0.1276ms	7.8364 KOps/s	7.6358 KOps/s	$\color{#35bf28}+2.63\%$
test_tdmodule	31.2190μs	17.5192μs	57.0804 KOps/s	51.4164 KOps/s	$\textbf{\color{#35bf28}+11.02\%}$
test_tdmodule_dispatch	56.3960μs	34.8917μs	28.6601 KOps/s	26.7990 KOps/s	$\textbf{\color{#35bf28}+6.94\%}$
test_tdseq	39.3440μs	20.2508μs	49.3808 KOps/s	45.8632 KOps/s	$\textbf{\color{#35bf28}+7.67\%}$
test_tdseq_dispatch	80.9230μs	40.2942μs	24.8175 KOps/s	23.4576 KOps/s	$\textbf{\color{#35bf28}+5.80\%}$
test_instantiation_functorch	1.6797ms	1.5385ms	649.9991 Ops/s	624.5841 Ops/s	$\color{#35bf28}+4.07\%$
test_exec_functorch	0.3189ms	0.1816ms	5.5058 KOps/s	5.3165 KOps/s	$\color{#35bf28}+3.56\%$
test_exec_functional_call	0.3976ms	0.1727ms	5.7919 KOps/s	5.5966 KOps/s	$\color{#35bf28}+3.49\%$
test_exec_td_decorator	0.5318ms	0.2340ms	4.2740 KOps/s	4.1758 KOps/s	$\color{#35bf28}+2.35\%$
test_vmap_mlp_speed_decorator[True-True]	0.9642ms	0.6475ms	1.5445 KOps/s	1.5305 KOps/s	$\color{#35bf28}+0.91\%$
test_vmap_mlp_speed_decorator[True-False]	0.9376ms	0.6467ms	1.5463 KOps/s	1.5106 KOps/s	$\color{#35bf28}+2.37\%$
test_vmap_mlp_speed_decorator[False-True]	0.7186ms	0.5330ms	1.8761 KOps/s	1.8348 KOps/s	$\color{#35bf28}+2.25\%$
test_vmap_mlp_speed_decorator[False-False]	0.8697ms	0.5353ms	1.8679 KOps/s	1.8528 KOps/s	$\color{#35bf28}+0.82\%$
test_to_module_speed[True]	2.3320ms	1.4107ms	708.8656 Ops/s	699.8337 Ops/s	$\color{#35bf28}+1.29\%$
test_to_module_speed[False]	1.9965ms	1.3653ms	732.4277 Ops/s	720.0742 Ops/s	$\color{#35bf28}+1.72\%$
test_tc_init	92.8450μs	45.4230μs	22.0153 KOps/s	20.6985 KOps/s	$\textbf{\color{#35bf28}+6.36\%}$
test_tc_init_nested	0.1589ms	90.6594μs	11.0303 KOps/s	10.5114 KOps/s	$\color{#35bf28}+4.94\%$
test_tc_first_layer_tensor	15.7890μs	1.5974μs	626.0218 KOps/s	615.9985 KOps/s	$\color{#35bf28}+1.63\%$
test_tc_first_layer_nontensor	23.2740μs	4.7056μs	212.5119 KOps/s	205.6748 KOps/s	$\color{#35bf28}+3.32\%$
test_tc_second_layer_tensor	40.6570μs	2.8598μs	349.6726 KOps/s	340.0885 KOps/s	$\color{#35bf28}+2.82\%$
test_tc_second_layer_nontensor	52.0780μs	6.0911μs	164.1742 KOps/s	162.5731 KOps/s	$\color{#35bf28}+0.98\%$
test_unbind	0.4750s	13.5066ms	74.0380 Ops/s	75.1760 Ops/s	$\color{#d91a1a}-1.51\%$
test_full_like	8.5786ms	7.4698ms	133.8720 Ops/s	141.3096 Ops/s	$\textbf{\color{#d91a1a}-5.26\%}$
test_zeros_like	3.3634ms	2.7703ms	360.9698 Ops/s	363.3751 Ops/s	$\color{#d91a1a}-0.66\%$
test_ones_like	4.1743ms	3.2060ms	311.9197 Ops/s	144.7759 Ops/s	$\textbf{\color{#35bf28}+115.45\%}$
test_clone	5.4791ms	4.9492ms	202.0528 Ops/s	116.7959 Ops/s	$\textbf{\color{#35bf28}+73.00\%}$
test_squeeze	64.9130μs	12.3877μs	80.7253 KOps/s	80.8775 KOps/s	$\color{#d91a1a}-0.19\%$
test_unsqueeze	0.3571ms	94.5528μs	10.5761 KOps/s	10.6916 KOps/s	$\color{#d91a1a}-1.08\%$
test_split	0.3912ms	0.1927ms	5.1894 KOps/s	4.9619 KOps/s	$\color{#35bf28}+4.58\%$
test_permute	0.4011ms	0.2236ms	4.4718 KOps/s	4.4253 KOps/s	$\color{#35bf28}+1.05\%$
test_stack	31.1722ms	24.6929ms	40.4975 Ops/s	40.7536 Ops/s	$\color{#d91a1a}-0.63\%$
test_cat	29.0726ms	24.5817ms	40.6806 Ops/s	41.6093 Ops/s	$\color{#d91a1a}-2.23\%$

[github-actions]

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

Total Benchmarks: 218. Improved: $\large\color{#35bf28}21$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results

Name	Max	Mean	Ops	Ops on Repo HEAD	Change
test_plain_set_nested	0.1476ms	16.4937μs	60.6294 KOps/s	57.1575 KOps/s	$\textbf{\color{#35bf28}+6.07\%}$
test_plain_set_stack_nested	37.2900μs	16.5780μs	60.3209 KOps/s	57.0105 KOps/s	$\textbf{\color{#35bf28}+5.81\%}$
test_plain_set_nested_inplace	47.6200μs	17.6462μs	56.6695 KOps/s	53.8312 KOps/s	$\textbf{\color{#35bf28}+5.27\%}$
test_plain_set_stack_nested_inplace	50.3110μs	17.5804μs	56.8814 KOps/s	53.3699 KOps/s	$\textbf{\color{#35bf28}+6.58\%}$
test_items	21.4800μs	2.8458μs	351.3926 KOps/s	348.0382 KOps/s	$\color{#35bf28}+0.96\%$
test_items_nested	0.3729ms	0.3382ms	2.9566 KOps/s	2.9438 KOps/s	$\color{#35bf28}+0.44\%$
test_items_nested_locked	0.3833ms	0.3396ms	2.9449 KOps/s	2.9674 KOps/s	$\color{#d91a1a}-0.76\%$
test_items_nested_leaf	91.7120μs	62.1810μs	16.0821 KOps/s	15.9196 KOps/s	$\color{#35bf28}+1.02\%$
test_items_stack_nested	0.4219ms	0.3398ms	2.9425 KOps/s	2.9236 KOps/s	$\color{#35bf28}+0.65\%$
test_items_stack_nested_leaf	91.3120μs	62.8614μs	15.9080 KOps/s	15.5087 KOps/s	$\color{#35bf28}+2.57\%$
test_items_stack_nested_locked	0.3737ms	0.3434ms	2.9119 KOps/s	2.9123 KOps/s	$\color{#d91a1a}-0.01\%$
test_keys	25.1410μs	3.3987μs	294.2317 KOps/s	289.6323 KOps/s	$\color{#35bf28}+1.59\%$
test_keys_nested	98.8520μs	71.0264μs	14.0793 KOps/s	13.9689 KOps/s	$\color{#35bf28}+0.79\%$
test_keys_nested_locked	2.5988ms	76.9714μs	12.9918 KOps/s	12.9071 KOps/s	$\color{#35bf28}+0.66\%$
test_keys_nested_leaf	94.6110μs	61.9635μs	16.1385 KOps/s	16.1277 KOps/s	$\color{#35bf28}+0.07\%$
test_keys_stack_nested	96.6310μs	70.6640μs	14.1515 KOps/s	14.2790 KOps/s	$\color{#d91a1a}-0.89\%$
test_keys_stack_nested_leaf	88.4610μs	61.2269μs	16.3327 KOps/s	16.2387 KOps/s	$\color{#35bf28}+0.58\%$
test_keys_stack_nested_locked	0.1077ms	77.1728μs	12.9579 KOps/s	13.0060 KOps/s	$\color{#d91a1a}-0.37\%$
test_values	4.1467μs	0.8362μs	1.1958 MOps/s	1.1568 MOps/s	$\color{#35bf28}+3.38\%$
test_values_nested	78.7720μs	48.5228μs	20.6089 KOps/s	20.4514 KOps/s	$\color{#35bf28}+0.77\%$
test_values_nested_locked	84.2320μs	49.8793μs	20.0484 KOps/s	19.8734 KOps/s	$\color{#35bf28}+0.88\%$
test_values_nested_leaf	69.3210μs	42.6768μs	23.4319 KOps/s	23.3726 KOps/s	$\color{#35bf28}+0.25\%$
test_values_stack_nested	86.1010μs	49.0237μs	20.3983 KOps/s	20.0809 KOps/s	$\color{#35bf28}+1.58\%$
test_values_stack_nested_leaf	66.5710μs	43.1082μs	23.1974 KOps/s	23.1062 KOps/s	$\color{#35bf28}+0.39\%$
test_values_stack_nested_locked	80.2520μs	50.3434μs	19.8636 KOps/s	19.3857 KOps/s	$\color{#35bf28}+2.47\%$
test_membership	1.7145μs	0.5007μs	1.9971 MOps/s	1.9645 MOps/s	$\color{#35bf28}+1.66\%$
test_membership_nested	16.5005μs	1.8925μs	528.3984 KOps/s	507.3444 KOps/s	$\color{#35bf28}+4.15\%$
test_membership_nested_leaf	14.9600μs	1.8971μs	527.1171 KOps/s	538.5882 KOps/s	$\color{#d91a1a}-2.13\%$
test_membership_stacked_nested	36.2510μs	1.9316μs	517.6935 KOps/s	520.1716 KOps/s	$\color{#d91a1a}-0.48\%$
test_membership_stacked_nested_leaf	27.4900μs	1.9624μs	509.5718 KOps/s	518.6846 KOps/s	$\color{#d91a1a}-1.76\%$
test_membership_nested_last	36.1300μs	2.9862μs	334.8743 KOps/s	336.2173 KOps/s	$\color{#d91a1a}-0.40\%$
test_membership_nested_leaf_last	23.8100μs	3.0051μs	332.7637 KOps/s	328.6513 KOps/s	$\color{#35bf28}+1.25\%$
test_membership_stacked_nested_last	26.6900μs	3.0007μs	333.2570 KOps/s	160.5686 KOps/s	$\textbf{\color{#35bf28}+107.55\%}$
test_membership_stacked_nested_leaf_last	28.8800μs	2.9598μs	337.8610 KOps/s	160.9732 KOps/s	$\textbf{\color{#35bf28}+109.89\%}$
test_nested_getleaf	35.3710μs	6.0377μs	165.6267 KOps/s	162.5076 KOps/s	$\color{#35bf28}+1.92\%$
test_nested_get	37.0910μs	5.6267μs	177.7237 KOps/s	172.4554 KOps/s	$\color{#35bf28}+3.05\%$
test_stacked_getleaf	37.6200μs	5.9844μs	167.1001 KOps/s	165.2412 KOps/s	$\color{#35bf28}+1.13\%$
test_stacked_get	24.8700μs	5.5846μs	179.0636 KOps/s	174.5539 KOps/s	$\color{#35bf28}+2.58\%$
test_nested_getitemleaf	32.6710μs	6.0719μs	164.6935 KOps/s	162.5136 KOps/s	$\color{#35bf28}+1.34\%$
test_nested_getitem	32.0100μs	5.6400μs	177.3046 KOps/s	175.4921 KOps/s	$\color{#35bf28}+1.03\%$
test_stacked_getitemleaf	44.8910μs	6.1537μs	162.5048 KOps/s	164.3897 KOps/s	$\color{#d91a1a}-1.15\%$
test_stacked_getitem	0.9800ms	5.6126μs	178.1703 KOps/s	174.5915 KOps/s	$\color{#35bf28}+2.05\%$
test_lock_nested	7.0441ms	0.4327ms	2.3111 KOps/s	2.3030 KOps/s	$\color{#35bf28}+0.35\%$
test_lock_stack_nested	0.4730ms	0.3918ms	2.5520 KOps/s	2.5780 KOps/s	$\color{#d91a1a}-1.01\%$
test_unlock_nested	0.7924ms	0.3642ms	2.7455 KOps/s	2.7061 KOps/s	$\color{#35bf28}+1.46\%$
test_unlock_stack_nested	0.3784ms	0.3306ms	3.0252 KOps/s	3.0724 KOps/s	$\color{#d91a1a}-1.54\%$
test_flatten_speed	0.1557ms	75.9714μs	13.1628 KOps/s	12.9415 KOps/s	$\color{#35bf28}+1.71\%$
test_unflatten_speed	0.3625ms	0.3201ms	3.1241 KOps/s	3.1505 KOps/s	$\color{#d91a1a}-0.84\%$
test_common_ops	1.5908ms	1.2767ms	783.2405 Ops/s	776.0090 Ops/s	$\color{#35bf28}+0.93\%$
test_creation	27.4910μs	1.4597μs	685.0593 KOps/s	674.2716 KOps/s	$\color{#35bf28}+1.60\%$
test_creation_empty	45.6810μs	15.0762μs	66.3298 KOps/s	58.9033 KOps/s	$\textbf{\color{#35bf28}+12.61\%}$
test_creation_nested_1	64.6010μs	16.9982μs	58.8298 KOps/s	53.6792 KOps/s	$\textbf{\color{#35bf28}+9.60\%}$
test_creation_nested_2	43.7710μs	19.4175μs	51.5000 KOps/s	47.7138 KOps/s	$\textbf{\color{#35bf28}+7.94\%}$
test_clone	71.7410μs	29.0431μs	34.4316 KOps/s	34.3150 KOps/s	$\color{#35bf28}+0.34\%$
test_getitem[int]	1.3447ms	16.1603μs	61.8802 KOps/s	60.9231 KOps/s	$\color{#35bf28}+1.57\%$
test_getitem[slice_int]	0.1269ms	28.0133μs	35.6974 KOps/s	35.5897 KOps/s	$\color{#35bf28}+0.30\%$
test_getitem[range]	0.2281ms	0.1121ms	8.9207 KOps/s	8.8508 KOps/s	$\color{#35bf28}+0.79\%$
test_getitem[tuple]	0.1197ms	23.8852μs	41.8670 KOps/s	41.1253 KOps/s	$\color{#35bf28}+1.80\%$
test_getitem[list]	0.2046ms	0.1010ms	9.8967 KOps/s	9.9256 KOps/s	$\color{#d91a1a}-0.29\%$
test_setitem_dim[int]	77.7110μs	45.3270μs	22.0619 KOps/s	21.9718 KOps/s	$\color{#35bf28}+0.41\%$
test_setitem_dim[slice_int]	93.9810μs	67.7299μs	14.7645 KOps/s	14.8387 KOps/s	$\color{#d91a1a}-0.50\%$
test_setitem_dim[range]	0.1633ms	0.1290ms	7.7518 KOps/s	7.7545 KOps/s	$\color{#d91a1a}-0.03\%$
test_setitem_dim[tuple]	84.7410μs	61.5486μs	16.2473 KOps/s	16.3998 KOps/s	$\color{#d91a1a}-0.93\%$
test_setitem	92.3520μs	42.7086μs	23.4145 KOps/s	23.2248 KOps/s	$\color{#35bf28}+0.82\%$
test_set	72.0310μs	41.6464μs	24.0117 KOps/s	23.8555 KOps/s	$\color{#35bf28}+0.65\%$
test_set_shared	0.3533ms	55.0887μs	18.1526 KOps/s	18.2615 KOps/s	$\color{#d91a1a}-0.60\%$
test_update	98.4020μs	50.7435μs	19.7069 KOps/s	19.3192 KOps/s	$\color{#35bf28}+2.01\%$
test_update_nested	0.1115ms	58.1351μs	17.2013 KOps/s	16.8590 KOps/s	$\color{#35bf28}+2.03\%$
test_update__nested	0.4233ms	65.1936μs	15.3389 KOps/s	16.2805 KOps/s	$\textbf{\color{#d91a1a}-5.78\%}$
test_set_nested	95.7720μs	44.8888μs	22.2773 KOps/s	22.6925 KOps/s	$\color{#d91a1a}-1.83\%$
test_set_nested_new	85.7610μs	49.8259μs	20.0699 KOps/s	20.8435 KOps/s	$\color{#d91a1a}-3.71\%$
test_select	0.1047ms	60.5045μs	16.5277 KOps/s	16.2479 KOps/s	$\color{#35bf28}+1.72\%$
test_select_nested	80.8420μs	41.7226μs	23.9678 KOps/s	24.2177 KOps/s	$\color{#d91a1a}-1.03\%$
test_exclude_nested	90.2710μs	59.1456μs	16.9074 KOps/s	17.1021 KOps/s	$\color{#d91a1a}-1.14\%$
test_empty[True]	0.3050ms	0.2587ms	3.8648 KOps/s	3.8964 KOps/s	$\color{#d91a1a}-0.81\%$
test_empty[False]	5.0691μs	0.7433μs	1.3453 MOps/s	1.3508 MOps/s	$\color{#d91a1a}-0.41\%$
test_to	57.1500μs	26.5358μs	37.6850 KOps/s	37.4427 KOps/s	$\color{#35bf28}+0.65\%$
test_to_nonblocking	69.1610μs	25.2376μs	39.6235 KOps/s	39.0636 KOps/s	$\color{#35bf28}+1.43\%$
test_unbind_speed	1.4633ms	0.2802ms	3.5687 KOps/s	3.5774 KOps/s	$\color{#d91a1a}-0.24\%$
test_unbind_speed_stack0	0.3580ms	0.2731ms	3.6614 KOps/s	3.6276 KOps/s	$\color{#35bf28}+0.93\%$
test_unbind_speed_stack1	91.7594ms	0.7075ms	1.4135 KOps/s	1.4244 KOps/s	$\color{#d91a1a}-0.77\%$
test_split	93.6574ms	2.1555ms	463.9246 Ops/s	450.5751 Ops/s	$\color{#35bf28}+2.96\%$
test_chunk	93.3817ms	2.1499ms	465.1467 Ops/s	447.0926 Ops/s	$\color{#35bf28}+4.04\%$
test_creation[device0]	0.3474ms	0.1294ms	7.7261 KOps/s	7.8266 KOps/s	$\color{#d91a1a}-1.28\%$
test_creation_from_tensor	0.3601ms	0.1363ms	7.3394 KOps/s	7.5515 KOps/s	$\color{#d91a1a}-2.81\%$
test_add_one[memmap_tensor0]	0.2336ms	9.0939μs	109.9633 KOps/s	113.4235 KOps/s	$\color{#d91a1a}-3.05\%$
test_contiguous[memmap_tensor0]	30.9700μs	2.1924μs	456.1139 KOps/s	455.4627 KOps/s	$\color{#35bf28}+0.14\%$
test_stack[memmap_tensor0]	42.8600μs	6.6754μs	149.8036 KOps/s	141.2164 KOps/s	$\textbf{\color{#35bf28}+6.08\%}$
test_memmaptd_index	1.3590ms	0.4246ms	2.3550 KOps/s	2.2560 KOps/s	$\color{#35bf28}+4.39\%$
test_memmaptd_index_astensor	0.7461ms	0.5022ms	1.9911 KOps/s	1.9301 KOps/s	$\color{#35bf28}+3.16\%$
test_memmaptd_index_op	1.4521ms	1.0404ms	961.1486 Ops/s	928.6160 Ops/s	$\color{#35bf28}+3.50\%$
test_serialize_model	0.1322s	0.1301s	7.6879 Ops/s	7.6955 Ops/s	$\color{#d91a1a}-0.10\%$
test_serialize_model_pickle	1.3477s	1.2132s	0.8243 Ops/s	0.8239 Ops/s	$\color{#35bf28}+0.04\%$
test_serialize_weights	0.2245s	0.1430s	6.9906 Ops/s	7.0059 Ops/s	$\color{#d91a1a}-0.22\%$
test_serialize_weights_returnearly	0.2140s	56.9957ms	17.5452 Ops/s	17.7991 Ops/s	$\color{#d91a1a}-1.43\%$
test_serialize_weights_pickle	1.3721s	1.2183s	0.8208 Ops/s	0.8220 Ops/s	$\color{#d91a1a}-0.14\%$
test_reshape_pytree	76.9010μs	35.2999μs	28.3287 KOps/s	28.1059 KOps/s	$\color{#35bf28}+0.79\%$
test_reshape_td	81.9710μs	41.0699μs	24.3487 KOps/s	23.4307 KOps/s	$\color{#35bf28}+3.92\%$
test_view_pytree	66.5510μs	34.7618μs	28.7672 KOps/s	28.3663 KOps/s	$\color{#35bf28}+1.41\%$
test_view_td	81.9410μs	45.4914μs	21.9822 KOps/s	21.2266 KOps/s	$\color{#35bf28}+3.56\%$
test_unbind_pytree	69.7810μs	33.5423μs	29.8131 KOps/s	29.4293 KOps/s	$\color{#35bf28}+1.30\%$
test_unbind_td	0.5296ms	42.0449μs	23.7841 KOps/s	23.3351 KOps/s	$\color{#35bf28}+1.92\%$
test_split_pytree	0.1034ms	46.3063μs	21.5953 KOps/s	21.7514 KOps/s	$\color{#d91a1a}-0.72\%$
test_split_td	93.7510ms	64.3986μs	15.5283 KOps/s	17.4197 KOps/s	$\textbf{\color{#d91a1a}-10.86\%}$
test_add_pytree	0.1031ms	56.9738μs	17.5519 KOps/s	17.6258 KOps/s	$\color{#d91a1a}-0.42\%$
test_add_td	0.1666ms	96.4142μs	10.3719 KOps/s	10.3460 KOps/s	$\color{#35bf28}+0.25\%$
test_compile_add_one_nested[tensordict-compile]	0.2112ms	0.1603ms	6.2369 KOps/s	6.1721 KOps/s	$\color{#35bf28}+1.05\%$
test_compile_add_one_nested[tensordict-eager]	0.2866ms	0.1644ms	6.0825 KOps/s	6.0360 KOps/s	$\color{#35bf28}+0.77\%$
test_compile_add_one_nested[pytree-compile]	0.1824ms	0.1442ms	6.9364 KOps/s	6.9177 KOps/s	$\color{#35bf28}+0.27\%$
test_compile_add_one_nested[pytree-eager]	0.2361ms	0.1850ms	5.4060 KOps/s	5.4390 KOps/s	$\color{#d91a1a}-0.61\%$
test_compile_copy_nested[tensordict-compile]	64.7210μs	21.7761μs	45.9219 KOps/s	45.4771 KOps/s	$\color{#35bf28}+0.98\%$
test_compile_copy_nested[tensordict-eager]	81.4710μs	49.3786μs	20.2517 KOps/s	20.1006 KOps/s	$\color{#35bf28}+0.75\%$
test_compile_copy_nested[pytree-compile]	0.2258ms	65.4696μs	15.2743 KOps/s	15.4375 KOps/s	$\color{#d91a1a}-1.06\%$
test_compile_copy_nested[pytree-eager]	92.8120μs	49.4949μs	20.2041 KOps/s	19.9807 KOps/s	$\color{#35bf28}+1.12\%$
test_compile_add_one_flat[tensordict-compile]	0.3832ms	0.3214ms	3.1112 KOps/s	3.0904 KOps/s	$\color{#35bf28}+0.67\%$
test_compile_add_one_flat[tensordict-eager]	0.3919ms	0.2335ms	4.2821 KOps/s	4.1641 KOps/s	$\color{#35bf28}+2.84\%$
test_compile_add_one_flat[tensorclass-compile]	0.1766ms	0.1271ms	7.8702 KOps/s	7.7663 KOps/s	$\color{#35bf28}+1.34\%$
test_compile_add_one_flat[tensorclass-eager]	0.1233ms	65.3770μs	15.2959 KOps/s	14.2221 KOps/s	$\textbf{\color{#35bf28}+7.55\%}$
test_compile_add_one_flat[pytree-compile]	0.4589ms	0.3231ms	3.0953 KOps/s	3.1273 KOps/s	$\color{#d91a1a}-1.02\%$
test_compile_add_one_flat[pytree-eager]	0.7966ms	0.6631ms	1.5080 KOps/s	1.5979 KOps/s	$\textbf{\color{#d91a1a}-5.62\%}$
test_compile_add_self_flat[tensordict-eager]	0.4284ms	0.2853ms	3.5048 KOps/s	3.4478 KOps/s	$\color{#35bf28}+1.65\%$
test_compile_add_self_flat[tensordict-compile]	0.4377ms	0.3272ms	3.0559 KOps/s	3.0825 KOps/s	$\color{#d91a1a}-0.86\%$
test_compile_add_self_flat[tensorclass-eager]	0.1883ms	78.4725μs	12.7433 KOps/s	12.6609 KOps/s	$\color{#35bf28}+0.65\%$
test_compile_add_self_flat[tensorclass-compile]	0.2012ms	0.1353ms	7.3920 KOps/s	7.5996 KOps/s	$\color{#d91a1a}-2.73\%$
test_compile_add_self_flat[pytree-eager]	0.7211ms	0.5420ms	1.8450 KOps/s	1.9084 KOps/s	$\color{#d91a1a}-3.32\%$
test_compile_add_self_flat[pytree-compile]	0.6153ms	0.3279ms	3.0496 KOps/s	3.1362 KOps/s	$\color{#d91a1a}-2.76\%$
test_compile_copy_flat[tensordict-compile]	0.1109ms	20.2313μs	49.4284 KOps/s	49.4949 KOps/s	$\color{#d91a1a}-0.13\%$
test_compile_copy_flat[tensordict-eager]	0.1343ms	37.9104μs	26.3780 KOps/s	25.3332 KOps/s	$\color{#35bf28}+4.12\%$
test_compile_copy_flat[pytree-compile]	0.1525ms	69.9791μs	14.2900 KOps/s	14.3131 KOps/s	$\color{#d91a1a}-0.16\%$
test_compile_copy_flat[pytree-eager]	0.1393ms	51.5883μs	19.3842 KOps/s	19.3669 KOps/s	$\color{#35bf28}+0.09\%$
test_compile_assign_and_add[tensordict-compile]	2.3505ms	0.7776ms	1.2861 KOps/s	1.1132 KOps/s	$\textbf{\color{#35bf28}+15.52\%}$
test_compile_assign_and_add[tensordict-eager]	3.5642ms	3.3344ms	299.9068 Ops/s	309.9770 Ops/s	$\color{#d91a1a}-3.25\%$
test_compile_assign_and_add[pytree-compile]	2.3142ms	0.8312ms	1.2030 KOps/s	1.1320 KOps/s	$\textbf{\color{#35bf28}+6.28\%}$
test_compile_assign_and_add[pytree-eager]	3.5630ms	3.2595ms	306.7947 Ops/s	311.4700 Ops/s	$\color{#d91a1a}-1.50\%$
test_compile_indexing[tensor-tensordict-compile]	0.1766ms	0.1081ms	9.2486 KOps/s	9.0168 KOps/s	$\color{#35bf28}+2.57\%$
test_compile_indexing[tensor-tensordict-eager]	0.1975ms	61.9772μs	16.1350 KOps/s	15.5495 KOps/s	$\color{#35bf28}+3.77\%$
test_compile_indexing[tensor-tensorclass-compile]	0.1921ms	0.1024ms	9.7682 KOps/s	9.7515 KOps/s	$\color{#35bf28}+0.17\%$
test_compile_indexing[tensor-tensorclass-eager]	98.7120μs	43.7683μs	22.8476 KOps/s	22.6404 KOps/s	$\color{#35bf28}+0.92\%$
test_compile_indexing[tensor-pytree-compile]	0.1782ms	0.1033ms	9.6847 KOps/s	9.6102 KOps/s	$\color{#35bf28}+0.78\%$
test_compile_indexing[tensor-pytree-eager]	0.1112ms	45.1800μs	22.1337 KOps/s	22.1053 KOps/s	$\color{#35bf28}+0.13\%$
test_compile_indexing[slice-tensordict-compile]	0.1914ms	0.1378ms	7.2548 KOps/s	7.2143 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_indexing[slice-tensordict-eager]	0.1548ms	24.8740μs	40.2027 KOps/s	39.6582 KOps/s	$\color{#35bf28}+1.37\%$
test_compile_indexing[slice-tensorclass-compile]	0.1845ms	0.1363ms	7.3392 KOps/s	7.6196 KOps/s	$\color{#d91a1a}-3.68\%$
test_compile_indexing[slice-tensorclass-eager]	77.9810μs	20.6584μs	48.4065 KOps/s	47.8995 KOps/s	$\color{#35bf28}+1.06\%$
test_compile_indexing[slice-pytree-compile]	0.2090ms	0.1315ms	7.6044 KOps/s	7.5623 KOps/s	$\color{#35bf28}+0.56\%$
test_compile_indexing[slice-pytree-eager]	61.0710μs	20.4194μs	48.9731 KOps/s	47.6644 KOps/s	$\color{#35bf28}+2.75\%$
test_compile_indexing[int-tensordict-compile]	0.2054ms	0.1410ms	7.0935 KOps/s	7.0558 KOps/s	$\color{#35bf28}+0.53\%$
test_compile_indexing[int-tensordict-eager]	0.5054ms	24.5594μs	40.7176 KOps/s	39.4309 KOps/s	$\color{#35bf28}+3.26\%$
test_compile_indexing[int-tensorclass-compile]	0.1999ms	0.1372ms	7.2884 KOps/s	7.5599 KOps/s	$\color{#d91a1a}-3.59\%$
test_compile_indexing[int-tensorclass-eager]	70.0910μs	21.4300μs	46.6636 KOps/s	47.7780 KOps/s	$\color{#d91a1a}-2.33\%$
test_compile_indexing[int-pytree-compile]	0.2253ms	0.1401ms	7.1386 KOps/s	7.5694 KOps/s	$\textbf{\color{#d91a1a}-5.69\%}$
test_compile_indexing[int-pytree-eager]	65.1710μs	21.0618μs	47.4792 KOps/s	47.3638 KOps/s	$\color{#35bf28}+0.24\%$
test_mod_add[eager]	83.9410μs	33.8109μs	29.5763 KOps/s	29.8813 KOps/s	$\color{#d91a1a}-1.02\%$
test_mod_add[compile]	0.3053ms	73.2673μs	13.6487 KOps/s	13.4030 KOps/s	$\color{#35bf28}+1.83\%$
test_mod_add[compile-overhead]	0.2685ms	0.1354ms	7.3841 KOps/s	7.1167 KOps/s	$\color{#35bf28}+3.76\%$
test_mod_wrap[eager]	0.4463ms	0.2412ms	4.1452 KOps/s	4.0216 KOps/s	$\color{#35bf28}+3.07\%$
test_mod_wrap[compile]	1.4139ms	0.3022ms	3.3095 KOps/s	3.3143 KOps/s	$\color{#d91a1a}-0.15\%$
test_mod_wrap[compile-overhead]	7.3649ms	3.9822ms	251.1184 Ops/s	252.0569 Ops/s	$\color{#d91a1a}-0.37\%$
test_mod_wrap_and_backward[eager]	2.0268ms	1.4318ms	698.4413 Ops/s	687.0644 Ops/s	$\color{#35bf28}+1.66\%$
test_mod_wrap_and_backward[compile]	1.7242ms	1.4171ms	705.6732 Ops/s	691.6834 Ops/s	$\color{#35bf28}+2.02\%$
test_mod_wrap_and_backward[compile-overhead]	1.7533ms	1.0582ms	944.9720 Ops/s	973.7104 Ops/s	$\color{#d91a1a}-2.95\%$
test_seq_add[eager]	0.1583ms	0.1070ms	9.3453 KOps/s	9.2932 KOps/s	$\color{#35bf28}+0.56\%$
test_seq_add[compile]	0.1558ms	86.1136μs	11.6126 KOps/s	11.5218 KOps/s	$\color{#35bf28}+0.79\%$
test_seq_add[compile-overhead]	0.2180ms	0.1188ms	8.4155 KOps/s	8.6730 KOps/s	$\color{#d91a1a}-2.97\%$
test_seq_wrap[eager]	0.4521ms	0.3819ms	2.6185 KOps/s	2.4201 KOps/s	$\textbf{\color{#35bf28}+8.20\%}$
test_seq_wrap[compile]	0.4373ms	0.3140ms	3.1851 KOps/s	3.0768 KOps/s	$\color{#35bf28}+3.52\%$
test_seq_wrap[compile-overhead]	0.2752ms	0.2190ms	4.5671 KOps/s	4.4660 KOps/s	$\color{#35bf28}+2.26\%$
test_func_call_runtime[False-eager]	1.2684ms	0.7970ms	1.2547 KOps/s	1.2393 KOps/s	$\color{#35bf28}+1.25\%$
test_func_call_runtime[False-compile]	0.9849ms	0.7937ms	1.2600 KOps/s	1.2175 KOps/s	$\color{#35bf28}+3.48\%$
test_func_call_runtime[False-compile-overhead]	0.4179ms	0.3605ms	2.7736 KOps/s	2.7324 KOps/s	$\color{#35bf28}+1.51\%$
test_func_call_runtime[True-eager]	0.9941ms	0.9092ms	1.0999 KOps/s	1.0679 KOps/s	$\color{#35bf28}+3.00\%$
test_func_call_runtime[True-compile]	0.8969ms	0.8160ms	1.2255 KOps/s	1.1920 KOps/s	$\color{#35bf28}+2.81\%$
test_func_call_runtime[True-compile-overhead]	0.4657ms	0.3834ms	2.6079 KOps/s	2.5721 KOps/s	$\color{#35bf28}+1.39\%$
test_func_call_cm_runtime[False-eager]	0.7914ms	0.7377ms	1.3555 KOps/s	1.2405 KOps/s	$\textbf{\color{#35bf28}+9.27\%}$
test_func_call_cm_runtime[False-compile]	1.2303ms	0.7948ms	1.2581 KOps/s	1.1909 KOps/s	$\textbf{\color{#35bf28}+5.64\%}$
test_func_call_cm_runtime[False-compile-overhead]	0.4058ms	0.3613ms	2.7677 KOps/s	2.7388 KOps/s	$\color{#35bf28}+1.05\%$
test_func_call_cm_runtime[True-eager]	1.1324ms	1.0216ms	978.9032 Ops/s	972.3170 Ops/s	$\color{#35bf28}+0.68\%$
test_func_call_cm_runtime[True-compile]	0.8967ms	0.8409ms	1.1892 KOps/s	1.1620 KOps/s	$\color{#35bf28}+2.34\%$
test_func_call_cm_runtime[True-compile-overhead]	0.5159ms	0.4080ms	2.4511 KOps/s	2.4219 KOps/s	$\color{#35bf28}+1.20\%$
test_vmap_func_call_cm_runtime[eager]	2.7968ms	2.1052ms	475.0093 Ops/s	474.2261 Ops/s	$\color{#35bf28}+0.17\%$
test_vmap_func_call_cm_runtime[compile]	1.3419ms	0.8655ms	1.1554 KOps/s	1.1463 KOps/s	$\color{#35bf28}+0.79\%$
test_vmap_func_call_cm_runtime[compile-overhead]	0.4759ms	0.4084ms	2.4487 KOps/s	2.4139 KOps/s	$\color{#35bf28}+1.44\%$
test_distributed	2.4353ms	0.1257ms	7.9528 KOps/s	8.4203 KOps/s	$\textbf{\color{#d91a1a}-5.55\%}$
test_tdmodule	37.1610μs	14.5963μs	68.5107 KOps/s	60.5615 KOps/s	$\textbf{\color{#35bf28}+13.13\%}$
test_tdmodule_dispatch	53.3410μs	30.0704μs	33.2553 KOps/s	31.3113 KOps/s	$\textbf{\color{#35bf28}+6.21\%}$
test_tdseq	37.8400μs	15.9109μs	62.8499 KOps/s	56.5135 KOps/s	$\textbf{\color{#35bf28}+11.21\%}$
test_tdseq_dispatch	54.7310μs	31.8573μs	31.3900 KOps/s	28.4562 KOps/s	$\textbf{\color{#35bf28}+10.31\%}$
test_instantiation_functorch	2.0141ms	1.8627ms	536.8466 Ops/s	530.8640 Ops/s	$\color{#35bf28}+1.13\%$
test_exec_functorch	0.2668ms	0.2093ms	4.7772 KOps/s	4.7214 KOps/s	$\color{#35bf28}+1.18\%$
test_exec_functional_call	0.2520ms	0.2092ms	4.7808 KOps/s	4.6584 KOps/s	$\color{#35bf28}+2.63\%$
test_exec_td_decorator	0.4421ms	0.2676ms	3.7368 KOps/s	3.7168 KOps/s	$\color{#35bf28}+0.54\%$
test_vmap_mlp_speed_decorator[True-True]	0.8354ms	0.7107ms	1.4070 KOps/s	1.4522 KOps/s	$\color{#d91a1a}-3.11\%$
test_vmap_mlp_speed_decorator[True-False]	0.8927ms	0.7115ms	1.4055 KOps/s	1.4608 KOps/s	$\color{#d91a1a}-3.79\%$
test_vmap_mlp_speed_decorator[False-True]	0.7600ms	0.6207ms	1.6111 KOps/s	1.6654 KOps/s	$\color{#d91a1a}-3.26\%$
test_vmap_mlp_speed_decorator[False-False]	0.7755ms	0.6269ms	1.5951 KOps/s	1.6647 KOps/s	$\color{#d91a1a}-4.18\%$
test_vmap_transformer_speed_decorator[True-True]	20.0367ms	19.4618ms	51.3827 Ops/s	51.1022 Ops/s	$\color{#35bf28}+0.55\%$
test_vmap_transformer_speed_decorator[True-False]	20.2373ms	19.5138ms	51.2458 Ops/s	51.1248 Ops/s	$\color{#35bf28}+0.24\%$
test_vmap_transformer_speed_decorator[False-True]	20.5781ms	19.3984ms	51.5508 Ops/s	51.6004 Ops/s	$\color{#d91a1a}-0.10\%$
test_vmap_transformer_speed_decorator[False-False]	19.4538ms	19.3764ms	51.6093 Ops/s	51.3569 Ops/s	$\color{#35bf28}+0.49\%$
test_to_module_speed[True]	1.2691ms	0.9998ms	1.0002 KOps/s	994.2266 Ops/s	$\color{#35bf28}+0.60\%$
test_to_module_speed[False]	1.4137ms	0.9737ms	1.0271 KOps/s	1.0162 KOps/s	$\color{#35bf28}+1.07\%$
test_tc_init	74.1410μs	34.7403μs	28.7851 KOps/s	27.3324 KOps/s	$\textbf{\color{#35bf28}+5.31\%}$
test_tc_init_nested	0.1329ms	71.2356μs	14.0379 KOps/s	13.4444 KOps/s	$\color{#35bf28}+4.41\%$
test_tc_first_layer_tensor	10.3416μs	0.6703μs	1.4918 MOps/s	1.4770 MOps/s	$\color{#35bf28}+1.01\%$
test_tc_first_layer_nontensor	25.7010μs	2.2314μs	448.1436 KOps/s	446.5971 KOps/s	$\color{#35bf28}+0.35\%$
test_tc_second_layer_tensor	11.6778μs	1.3431μs	744.5461 KOps/s	728.0689 KOps/s	$\color{#35bf28}+2.26\%$
test_tc_second_layer_nontensor	22.2600μs	2.9199μs	342.4771 KOps/s	339.8466 KOps/s	$\color{#35bf28}+0.77\%$
test_unbind	0.1835s	11.9713ms	83.5331 Ops/s	93.2731 Ops/s	$\textbf{\color{#d91a1a}-10.44\%}$
test_full_like	0.6545ms	0.5733ms	1.7444 KOps/s	1.7440 KOps/s	$\color{#35bf28}+0.02\%$
test_zeros_like	0.2872ms	0.1979ms	5.0532 KOps/s	5.0555 KOps/s	$\color{#d91a1a}-0.05\%$
test_ones_like	0.2353ms	0.1977ms	5.0576 KOps/s	5.0540 KOps/s	$\color{#35bf28}+0.07\%$
test_clone	0.4450ms	0.4146ms	2.4118 KOps/s	2.4100 KOps/s	$\color{#35bf28}+0.07\%$
test_squeeze	33.5610μs	9.7537μs	102.5252 KOps/s	101.6714 KOps/s	$\color{#35bf28}+0.84\%$
test_unsqueeze	0.2234ms	74.3563μs	13.4488 KOps/s	13.6498 KOps/s	$\color{#d91a1a}-1.47\%$
test_split	0.4252ms	0.1552ms	6.4443 KOps/s	6.3827 KOps/s	$\color{#35bf28}+0.96\%$
test_permute	0.2214ms	0.1805ms	5.5414 KOps/s	5.4155 KOps/s	$\color{#35bf28}+2.33\%$
test_stack	1.2510ms	0.8598ms	1.1631 KOps/s	1.2530 KOps/s	$\textbf{\color{#d91a1a}-7.18\%}$
test_cat	1.2531ms	1.2311ms	812.2777 Ops/s	812.1694 Ops/s	$\color{#35bf28}+0.01\%$