Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Dec 18, 2025

No description provided.

@meta-cla meta-cla bot added the CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. label Dec 18, 2025
@github-actions
Copy link

github-actions bot commented Dec 18, 2025

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

Total Benchmarks: 233. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 37.7610μs 14.3531μs 69.6714 KOps/s 70.3965 KOps/s $\color{#d91a1a}-1.03\%$
test_plain_set_stack_nested 39.3210μs 14.7590μs 67.7554 KOps/s 69.2923 KOps/s $\color{#d91a1a}-2.22\%$
test_plain_set_nested_inplace 77.2010μs 16.0083μs 62.4674 KOps/s 63.2571 KOps/s $\color{#d91a1a}-1.25\%$
test_plain_set_stack_nested_inplace 46.2610μs 15.8539μs 63.0761 KOps/s 62.8456 KOps/s $\color{#35bf28}+0.37\%$
test_items 29.1210μs 5.5489μs 180.2167 KOps/s 178.0220 KOps/s $\color{#35bf28}+1.23\%$
test_items_nested 0.5576ms 0.5094ms 1.9629 KOps/s 1.9124 KOps/s $\color{#35bf28}+2.64\%$
test_items_nested_locked 0.6601ms 0.5131ms 1.9490 KOps/s 1.9206 KOps/s $\color{#35bf28}+1.48\%$
test_items_nested_leaf 0.1284ms 90.8966μs 11.0015 KOps/s 11.0206 KOps/s $\color{#d91a1a}-0.17\%$
test_items_stack_nested 0.7245ms 0.5122ms 1.9523 KOps/s 1.9242 KOps/s $\color{#35bf28}+1.46\%$
test_items_stack_nested_leaf 0.1329ms 92.0365μs 10.8653 KOps/s 10.9437 KOps/s $\color{#d91a1a}-0.72\%$
test_items_stack_nested_locked 0.5820ms 0.5056ms 1.9778 KOps/s 1.9253 KOps/s $\color{#35bf28}+2.73\%$
test_keys 43.2310μs 4.0695μs 245.7300 KOps/s 243.1115 KOps/s $\color{#35bf28}+1.08\%$
test_keys_nested 0.1746ms 0.1141ms 8.7636 KOps/s 8.5824 KOps/s $\color{#35bf28}+2.11\%$
test_keys_nested_locked 2.1835ms 0.1250ms 8.0015 KOps/s 7.9158 KOps/s $\color{#35bf28}+1.08\%$
test_keys_nested_leaf 0.1558ms 0.1058ms 9.4499 KOps/s 9.2991 KOps/s $\color{#35bf28}+1.62\%$
test_keys_stack_nested 0.1850ms 0.1154ms 8.6687 KOps/s 8.5972 KOps/s $\color{#35bf28}+0.83\%$
test_keys_stack_nested_leaf 0.1538ms 0.1063ms 9.4075 KOps/s 9.3066 KOps/s $\color{#35bf28}+1.08\%$
test_keys_stack_nested_locked 0.1677ms 0.1249ms 8.0093 KOps/s 7.9471 KOps/s $\color{#35bf28}+0.78\%$
test_values 5.6582μs 0.9922μs 1.0078 MOps/s 998.3309 KOps/s $\color{#35bf28}+0.95\%$
test_values_nested 75.1610μs 45.9424μs 21.7664 KOps/s 21.5633 KOps/s $\color{#35bf28}+0.94\%$
test_values_nested_locked 92.6910μs 49.3771μs 20.2523 KOps/s 20.1539 KOps/s $\color{#35bf28}+0.49\%$
test_values_nested_leaf 98.5320μs 52.1449μs 19.1773 KOps/s 19.0797 KOps/s $\color{#35bf28}+0.51\%$
test_values_stack_nested 88.1020μs 46.2016μs 21.6443 KOps/s 21.4628 KOps/s $\color{#35bf28}+0.85\%$
test_values_stack_nested_leaf 95.4610μs 52.2412μs 19.1420 KOps/s 19.0692 KOps/s $\color{#35bf28}+0.38\%$
test_values_stack_nested_locked 77.4920μs 48.9272μs 20.4385 KOps/s 20.2378 KOps/s $\color{#35bf28}+0.99\%$
test_membership 11.2835μs 0.8117μs 1.2320 MOps/s 1.2572 MOps/s $\color{#d91a1a}-2.00\%$
test_membership_nested 65.3010μs 2.9764μs 335.9726 KOps/s 338.1028 KOps/s $\color{#d91a1a}-0.63\%$
test_membership_nested_leaf 27.5000μs 2.9664μs 337.1040 KOps/s 334.6422 KOps/s $\color{#35bf28}+0.74\%$
test_membership_stacked_nested 43.2810μs 2.9038μs 344.3736 KOps/s 338.3314 KOps/s $\color{#35bf28}+1.79\%$
test_membership_stacked_nested_leaf 35.3510μs 2.9521μs 338.7448 KOps/s 332.8824 KOps/s $\color{#35bf28}+1.76\%$
test_membership_nested_last 31.8310μs 4.3083μs 232.1075 KOps/s 230.6187 KOps/s $\color{#35bf28}+0.65\%$
test_membership_nested_leaf_last 28.9610μs 4.3442μs 230.1911 KOps/s 228.8243 KOps/s $\color{#35bf28}+0.60\%$
test_membership_stacked_nested_last 47.1100μs 4.3055μs 232.2619 KOps/s 228.0550 KOps/s $\color{#35bf28}+1.84\%$
test_membership_stacked_nested_leaf_last 34.3800μs 4.3284μs 231.0304 KOps/s 229.5229 KOps/s $\color{#35bf28}+0.66\%$
test_nested_getleaf 48.2410μs 20.3603μs 49.1153 KOps/s 48.5081 KOps/s $\color{#35bf28}+1.25\%$
test_nested_get 47.8310μs 19.1356μs 52.2587 KOps/s 51.1976 KOps/s $\color{#35bf28}+2.07\%$
test_stacked_getleaf 39.6710μs 20.2430μs 49.3998 KOps/s 48.8942 KOps/s $\color{#35bf28}+1.03\%$
test_stacked_get 60.6310μs 19.1442μs 52.2351 KOps/s 51.1710 KOps/s $\color{#35bf28}+2.08\%$
test_nested_getitemleaf 71.5710μs 20.5473μs 48.6683 KOps/s 47.7254 KOps/s $\color{#35bf28}+1.98\%$
test_nested_getitem 46.1110μs 19.3613μs 51.6495 KOps/s 50.3541 KOps/s $\color{#35bf28}+2.57\%$
test_stacked_getitemleaf 47.0500μs 20.7312μs 48.2365 KOps/s 47.6057 KOps/s $\color{#35bf28}+1.33\%$
test_stacked_getitem 48.8410μs 19.8696μs 50.3281 KOps/s 50.5152 KOps/s $\color{#d91a1a}-0.37\%$
test_lock_nested 0.5348ms 0.4466ms 2.2390 KOps/s 2.2140 KOps/s $\color{#35bf28}+1.13\%$
test_lock_stack_nested 0.5109ms 0.4527ms 2.2090 KOps/s 2.1965 KOps/s $\color{#35bf28}+0.57\%$
test_unlock_nested 0.4382ms 0.3639ms 2.7484 KOps/s 2.7648 KOps/s $\color{#d91a1a}-0.59\%$
test_unlock_stack_nested 0.3917ms 0.3616ms 2.7652 KOps/s 2.7372 KOps/s $\color{#35bf28}+1.02\%$
test_flatten_speed 0.1471ms 0.1174ms 8.5173 KOps/s 8.5606 KOps/s $\color{#d91a1a}-0.51\%$
test_unflatten_speed 0.6176ms 0.5621ms 1.7789 KOps/s 1.7711 KOps/s $\color{#35bf28}+0.44\%$
test_common_ops 0.8633ms 0.7227ms 1.3836 KOps/s 1.3742 KOps/s $\color{#35bf28}+0.69\%$
test_creation 72.2320μs 2.5704μs 389.0439 KOps/s 388.4780 KOps/s $\color{#35bf28}+0.15\%$
test_creation_empty 57.1910μs 8.5311μs 117.2186 KOps/s 115.5546 KOps/s $\color{#35bf28}+1.44\%$
test_creation_nested_1 32.9810μs 11.4917μs 87.0193 KOps/s 85.9518 KOps/s $\color{#35bf28}+1.24\%$
test_creation_nested_2 44.1600μs 15.3107μs 65.3139 KOps/s 65.0835 KOps/s $\color{#35bf28}+0.35\%$
test_clone 38.5100μs 12.6233μs 79.2186 KOps/s 75.7779 KOps/s $\color{#35bf28}+4.54\%$
test_getitem[int] 1.2087ms 13.5421μs 73.8440 KOps/s 73.1998 KOps/s $\color{#35bf28}+0.88\%$
test_getitem[slice_int] 0.1564ms 23.2258μs 43.0556 KOps/s 42.6190 KOps/s $\color{#35bf28}+1.02\%$
test_getitem[range] 0.1643ms 57.0528μs 17.5276 KOps/s 17.3089 KOps/s $\color{#35bf28}+1.26\%$
test_getitem[tuple] 0.1489ms 23.0777μs 43.3319 KOps/s 42.8121 KOps/s $\color{#35bf28}+1.21\%$
test_getitem[list] 0.1805ms 50.7363μs 19.7098 KOps/s 19.0895 KOps/s $\color{#35bf28}+3.25\%$
test_setitem_dim[int] 50.1610μs 23.1816μs 43.1377 KOps/s 41.4141 KOps/s $\color{#35bf28}+4.16\%$
test_setitem_dim[slice_int] 64.4710μs 42.2152μs 23.6882 KOps/s 22.8436 KOps/s $\color{#35bf28}+3.70\%$
test_setitem_dim[range] 0.1057ms 82.7032μs 12.0914 KOps/s 11.9357 KOps/s $\color{#35bf28}+1.31\%$
test_setitem_dim[tuple] 69.5910μs 39.5561μs 25.2805 KOps/s 24.3824 KOps/s $\color{#35bf28}+3.68\%$
test_setitem 58.6210μs 17.1806μs 58.2053 KOps/s 55.3227 KOps/s $\textbf{\color{#35bf28}+5.21\%}$
test_set 47.9910μs 16.3834μs 61.0374 KOps/s 58.4950 KOps/s $\color{#35bf28}+4.35\%$
test_set_shared 0.6379ms 0.2026ms 4.9359 KOps/s 4.8059 KOps/s $\color{#35bf28}+2.70\%$
test_update 0.2848ms 21.5771μs 46.3453 KOps/s 42.9185 KOps/s $\textbf{\color{#35bf28}+7.98\%}$
test_update_nested 68.5510μs 32.6767μs 30.6029 KOps/s 28.1451 KOps/s $\textbf{\color{#35bf28}+8.73\%}$
test_update__nested 0.4757ms 32.5435μs 30.7281 KOps/s 29.2111 KOps/s $\textbf{\color{#35bf28}+5.19\%}$
test_set_nested 54.9210μs 18.5048μs 54.0400 KOps/s 53.0542 KOps/s $\color{#35bf28}+1.86\%$
test_set_nested_new 59.2410μs 23.3405μs 42.8440 KOps/s 42.1765 KOps/s $\color{#35bf28}+1.58\%$
test_select 91.4910μs 40.3335μs 24.7933 KOps/s 24.8533 KOps/s $\color{#d91a1a}-0.24\%$
test_select_nested 0.1471ms 69.9706μs 14.2917 KOps/s 14.1301 KOps/s $\color{#35bf28}+1.14\%$
test_exclude_nested 0.1453ms 93.2079μs 10.7287 KOps/s 10.6439 KOps/s $\color{#35bf28}+0.80\%$
test_empty[True] 0.4750ms 0.4160ms 2.4038 KOps/s 2.3696 KOps/s $\color{#35bf28}+1.44\%$
test_empty[False] 7.8127μs 1.2682μs 788.5414 KOps/s 800.4181 KOps/s $\color{#d91a1a}-1.48\%$
test_to 0.1055ms 71.6024μs 13.9660 KOps/s 13.6386 KOps/s $\color{#35bf28}+2.40\%$
test_to_nonblocking 0.1084ms 64.2870μs 15.5552 KOps/s 15.8320 KOps/s $\color{#d91a1a}-1.75\%$
test_unbind_speed 0.3423ms 0.3096ms 3.2304 KOps/s 3.2478 KOps/s $\color{#d91a1a}-0.54\%$
test_unbind_speed_stack0 0.3844ms 0.3056ms 3.2728 KOps/s 3.2930 KOps/s $\color{#d91a1a}-0.62\%$
test_unbind_speed_stack1 99.5315ms 0.9055ms 1.1044 KOps/s 1.2097 KOps/s $\textbf{\color{#d91a1a}-8.70\%}$
test_split 1.1721ms 1.1081ms 902.4719 Ops/s 686.1340 Ops/s $\textbf{\color{#35bf28}+31.53\%}$
test_chunk 99.1335ms 1.1686ms 855.6937 Ops/s 958.6752 Ops/s $\textbf{\color{#d91a1a}-10.74\%}$
test_consolidate[False-None] 3.7807ms 3.6930ms 270.7843 Ops/s 268.6169 Ops/s $\color{#35bf28}+0.81\%$
test_consolidate[default-None] 2.4290ms 1.9892ms 502.7216 Ops/s 485.1807 Ops/s $\color{#35bf28}+3.62\%$
test_consolidate[reduce-overhead-None] 1.9909ms 1.9108ms 523.3490 Ops/s 500.0905 Ops/s $\color{#35bf28}+4.65\%$
test_consolidate_njt[False-None] 9.0487ms 8.5696ms 116.6920 Ops/s 118.3143 Ops/s $\color{#d91a1a}-1.37\%$
test_to[False-False-None] 2.4964ms 2.0358ms 491.2100 Ops/s 485.6265 Ops/s $\color{#35bf28}+1.15\%$
test_to[True-False-None] 2.1531ms 1.7761ms 563.0223 Ops/s 555.3077 Ops/s $\color{#35bf28}+1.39\%$
test_to[within-False-None] 5.7926ms 5.5898ms 178.8983 Ops/s 179.2454 Ops/s $\color{#d91a1a}-0.19\%$
test_to[True-default-None] 12.1248ms 11.4794ms 87.1126 Ops/s 86.2423 Ops/s $\color{#35bf28}+1.01\%$
test_to_njt[False-False-None] 8.8532ms 8.3814ms 119.3115 Ops/s 118.9391 Ops/s $\color{#35bf28}+0.31\%$
test_to_njt[True-False-None] 7.6285ms 7.1269ms 140.3127 Ops/s 140.6494 Ops/s $\color{#d91a1a}-0.24\%$
test_to_njt[within-False-None] 16.4008ms 15.9220ms 62.8063 Ops/s 63.1791 Ops/s $\color{#d91a1a}-0.59\%$
test_creation[device0] 0.5445ms 0.1097ms 9.1128 KOps/s 9.3537 KOps/s $\color{#d91a1a}-2.57\%$
test_creation_from_tensor 0.4759ms 0.1140ms 8.7697 KOps/s 9.1823 KOps/s $\color{#d91a1a}-4.49\%$
test_add_one[memmap_tensor0] 0.4498ms 6.5067μs 153.6866 KOps/s 152.2709 KOps/s $\color{#35bf28}+0.93\%$
test_contiguous[memmap_tensor0] 0.4176ms 0.6343μs 1.5765 MOps/s 2.1475 MOps/s $\textbf{\color{#d91a1a}-26.59\%}$
test_stack[memmap_tensor0] 22.8600μs 4.5309μs 220.7049 KOps/s 222.8542 KOps/s $\color{#d91a1a}-0.96\%$
test_memmaptd_index 1.0293ms 0.2688ms 3.7203 KOps/s 3.7242 KOps/s $\color{#d91a1a}-0.10\%$
test_memmaptd_index_astensor 0.7940ms 0.3599ms 2.7784 KOps/s 2.8008 KOps/s $\color{#d91a1a}-0.80\%$
test_memmaptd_index_op 1.0114ms 0.6017ms 1.6619 KOps/s 1.6749 KOps/s $\color{#d91a1a}-0.78\%$
test_serialize_model 0.1353s 0.1333s 7.5002 Ops/s 7.4935 Ops/s $\color{#35bf28}+0.09\%$
test_serialize_model_pickle 1.3460s 1.2119s 0.8252 Ops/s 0.8423 Ops/s $\color{#d91a1a}-2.04\%$
test_serialize_weights 0.1336s 0.1326s 7.5427 Ops/s 7.5419 Ops/s $\color{#35bf28}+0.01\%$
test_serialize_weights_returnearly 0.3999s 67.6429ms 14.7835 Ops/s 12.2043 Ops/s $\textbf{\color{#35bf28}+21.13\%}$
test_serialize_weights_pickle 1.3902s 1.2178s 0.8212 Ops/s 0.8318 Ops/s $\color{#d91a1a}-1.28\%$
test_reshape_pytree 0.3802ms 31.5517μs 31.6940 KOps/s 30.5869 KOps/s $\color{#35bf28}+3.62\%$
test_reshape_td 61.9810μs 36.3425μs 27.5160 KOps/s 27.1456 KOps/s $\color{#35bf28}+1.36\%$
test_view_pytree 0.2239ms 31.4531μs 31.7934 KOps/s 31.7518 KOps/s $\color{#35bf28}+0.13\%$
test_view_td 83.6710μs 43.4736μs 23.0025 KOps/s 22.2612 KOps/s $\color{#35bf28}+3.33\%$
test_unbind_pytree 0.2409ms 36.2058μs 27.6199 KOps/s 26.8931 KOps/s $\color{#35bf28}+2.70\%$
test_unbind_td 0.1732ms 46.6099μs 21.4547 KOps/s 21.2198 KOps/s $\color{#35bf28}+1.11\%$
test_split_pytree 0.2519ms 42.0202μs 23.7981 KOps/s 23.2479 KOps/s $\color{#35bf28}+2.37\%$
test_split_td 0.1039ms 60.9354μs 16.4108 KOps/s 16.3093 KOps/s $\color{#35bf28}+0.62\%$
test_add_pytree 0.2354ms 42.5687μs 23.4915 KOps/s 23.2378 KOps/s $\color{#35bf28}+1.09\%$
test_add_td 94.2820μs 51.9977μs 19.2316 KOps/s 18.9947 KOps/s $\color{#35bf28}+1.25\%$
test_compile_add_one_nested[tensordict-compile] 0.3272ms 0.1799ms 5.5574 KOps/s 5.6696 KOps/s $\color{#d91a1a}-1.98\%$
test_compile_add_one_nested[tensordict-eager] 0.4127ms 0.1864ms 5.3659 KOps/s 5.3511 KOps/s $\color{#35bf28}+0.28\%$
test_compile_add_one_nested[pytree-compile] 0.4686ms 0.1479ms 6.7616 KOps/s 6.2727 KOps/s $\textbf{\color{#35bf28}+7.79\%}$
test_compile_add_one_nested[pytree-eager] 0.4353ms 0.1820ms 5.4953 KOps/s 5.4947 KOps/s $\color{#35bf28}+0.01\%$
test_compile_copy_nested[tensordict-compile] 0.1431ms 26.8378μs 37.2608 KOps/s 36.4688 KOps/s $\color{#35bf28}+2.17\%$
test_compile_copy_nested[tensordict-eager] 0.1130ms 50.2118μs 19.9156 KOps/s 20.0900 KOps/s $\color{#d91a1a}-0.87\%$
test_compile_copy_nested[pytree-compile] 0.1843ms 13.8158μs 72.3811 KOps/s 66.2307 KOps/s $\textbf{\color{#35bf28}+9.29\%}$
test_compile_copy_nested[pytree-eager] 0.3859ms 71.9208μs 13.9042 KOps/s 13.8924 KOps/s $\color{#35bf28}+0.08\%$
test_compile_add_one_flat[tensordict-compile] 0.2948ms 0.2018ms 4.9553 KOps/s 4.5908 KOps/s $\textbf{\color{#35bf28}+7.94\%}$
test_compile_add_one_flat[tensordict-eager] 0.3180ms 0.2562ms 3.9025 KOps/s 3.8361 KOps/s $\color{#35bf28}+1.73\%$
test_compile_add_one_flat[tensorclass-compile] 0.2245ms 0.1494ms 6.6951 KOps/s 6.3751 KOps/s $\textbf{\color{#35bf28}+5.02\%}$
test_compile_add_one_flat[tensorclass-eager] 0.1173ms 70.1650μs 14.2521 KOps/s 13.9851 KOps/s $\color{#35bf28}+1.91\%$
test_compile_add_one_flat[pytree-compile] 0.2575ms 0.1975ms 5.0639 KOps/s 4.8095 KOps/s $\textbf{\color{#35bf28}+5.29\%}$
test_compile_add_one_flat[pytree-eager] 0.7831ms 0.5334ms 1.8746 KOps/s 1.8682 KOps/s $\color{#35bf28}+0.34\%$
test_compile_add_self_flat[tensordict-eager] 0.4878ms 0.3081ms 3.2461 KOps/s 3.2103 KOps/s $\color{#35bf28}+1.12\%$
test_compile_add_self_flat[tensordict-compile] 0.2480ms 0.2029ms 4.9283 KOps/s 4.6929 KOps/s $\textbf{\color{#35bf28}+5.02\%}$
test_compile_add_self_flat[tensorclass-eager] 0.1908ms 84.6795μs 11.8092 KOps/s 11.6650 KOps/s $\color{#35bf28}+1.24\%$
test_compile_add_self_flat[tensorclass-compile] 0.2083ms 0.1507ms 6.6367 KOps/s 6.3651 KOps/s $\color{#35bf28}+4.27\%$
test_compile_add_self_flat[pytree-eager] 0.6608ms 0.4470ms 2.2369 KOps/s 2.2098 KOps/s $\color{#35bf28}+1.23\%$
test_compile_add_self_flat[pytree-compile] 0.2712ms 0.2002ms 4.9938 KOps/s 4.9128 KOps/s $\color{#35bf28}+1.65\%$
test_compile_copy_flat[tensordict-compile] 0.4162ms 23.6246μs 42.3288 KOps/s 40.6426 KOps/s $\color{#35bf28}+4.15\%$
test_compile_copy_flat[tensordict-eager] 72.1910μs 40.4444μs 24.7253 KOps/s 25.4645 KOps/s $\color{#d91a1a}-2.90\%$
test_compile_copy_flat[pytree-compile] 56.5910μs 19.4527μs 51.4067 KOps/s 44.7337 KOps/s $\textbf{\color{#35bf28}+14.92\%}$
test_compile_copy_flat[pytree-eager] 0.3642ms 66.6788μs 14.9973 KOps/s 14.8967 KOps/s $\color{#35bf28}+0.68\%$
test_compile_assign_and_add[tensordict-compile] 1.9685ms 0.2062ms 4.8490 KOps/s 4.8143 KOps/s $\color{#35bf28}+0.72\%$
test_compile_assign_and_add[tensordict-eager] 3.4380ms 3.2531ms 307.3969 Ops/s 311.0039 Ops/s $\color{#d91a1a}-1.16\%$
test_compile_assign_and_add[pytree-compile] 1.9568ms 0.2031ms 4.9226 KOps/s 4.9016 KOps/s $\color{#35bf28}+0.43\%$
test_compile_assign_and_add[pytree-eager] 3.0186ms 2.9029ms 344.4830 Ops/s 345.0529 Ops/s $\color{#d91a1a}-0.17\%$
test_compile_indexing[tensor-tensordict-compile] 0.3649ms 0.1392ms 7.1816 KOps/s 7.0705 KOps/s $\color{#35bf28}+1.57\%$
test_compile_indexing[tensor-tensordict-eager] 0.3319ms 65.2088μs 15.3354 KOps/s 15.4514 KOps/s $\color{#d91a1a}-0.75\%$
test_compile_indexing[tensor-tensorclass-compile] 0.2347ms 0.1329ms 7.5267 KOps/s 7.3921 KOps/s $\color{#35bf28}+1.82\%$
test_compile_indexing[tensor-tensorclass-eager] 0.2418ms 45.9062μs 21.7836 KOps/s 21.9629 KOps/s $\color{#d91a1a}-0.82\%$
test_compile_indexing[tensor-pytree-compile] 0.1836ms 0.1333ms 7.5002 KOps/s 7.3461 KOps/s $\color{#35bf28}+2.10\%$
test_compile_indexing[tensor-pytree-eager] 0.2427ms 45.4692μs 21.9929 KOps/s 21.9952 KOps/s $\color{#d91a1a}-0.01\%$
test_compile_indexing[slice-tensordict-compile] 0.2461ms 83.9087μs 11.9177 KOps/s 11.7556 KOps/s $\color{#35bf28}+1.38\%$
test_compile_indexing[slice-tensordict-eager] 0.2105ms 25.5904μs 39.0771 KOps/s 38.4507 KOps/s $\color{#35bf28}+1.63\%$
test_compile_indexing[slice-tensorclass-compile] 0.1628ms 78.7076μs 12.7053 KOps/s 12.5168 KOps/s $\color{#35bf28}+1.51\%$
test_compile_indexing[slice-tensorclass-eager] 0.2275ms 22.5081μs 44.4285 KOps/s 43.0327 KOps/s $\color{#35bf28}+3.24\%$
test_compile_indexing[slice-pytree-compile] 0.1226ms 79.3507μs 12.6023 KOps/s 12.4257 KOps/s $\color{#35bf28}+1.42\%$
test_compile_indexing[slice-pytree-eager] 0.2304ms 22.4678μs 44.5081 KOps/s 43.3241 KOps/s $\color{#35bf28}+2.73\%$
test_compile_indexing[int-tensordict-compile] 0.1315ms 84.7170μs 11.8040 KOps/s 11.6510 KOps/s $\color{#35bf28}+1.31\%$
test_compile_indexing[int-tensordict-eager] 0.2103ms 25.6052μs 39.0546 KOps/s 38.6531 KOps/s $\color{#35bf28}+1.04\%$
test_compile_indexing[int-tensorclass-compile] 0.1255ms 79.0570μs 12.6491 KOps/s 12.3714 KOps/s $\color{#35bf28}+2.24\%$
test_compile_indexing[int-tensorclass-eager] 0.2164ms 22.4533μs 44.5370 KOps/s 43.3658 KOps/s $\color{#35bf28}+2.70\%$
test_compile_indexing[int-pytree-compile] 0.1336ms 79.9878μs 12.5019 KOps/s 12.3838 KOps/s $\color{#35bf28}+0.95\%$
test_compile_indexing[int-pytree-eager] 0.2781ms 22.5204μs 44.4041 KOps/s 43.2300 KOps/s $\color{#35bf28}+2.72\%$
test_mod_add[eager] 98.4010μs 51.3991μs 19.4556 KOps/s 19.7696 KOps/s $\color{#d91a1a}-1.59\%$
test_mod_add[compile] 0.1946ms 0.1482ms 6.7498 KOps/s 6.6699 KOps/s $\color{#35bf28}+1.20\%$
test_mod_add[compile-overhead] 0.2788ms 0.1924ms 5.1982 KOps/s 5.1431 KOps/s $\color{#35bf28}+1.07\%$
test_mod_wrap[eager] 0.3819ms 0.3063ms 3.2647 KOps/s 3.3155 KOps/s $\color{#d91a1a}-1.53\%$
test_mod_wrap[compile] 0.5036ms 0.3928ms 2.5457 KOps/s 2.5540 KOps/s $\color{#d91a1a}-0.33\%$
test_mod_wrap[compile-overhead] 7.4898ms 3.9952ms 250.2994 Ops/s 255.6882 Ops/s $\color{#d91a1a}-2.11\%$
test_mod_wrap_and_backward[eager] 1.9407ms 1.5539ms 643.5337 Ops/s 587.4016 Ops/s $\textbf{\color{#35bf28}+9.56\%}$
test_mod_wrap_and_backward[compile] 1.7031ms 1.5841ms 631.2698 Ops/s 630.7823 Ops/s $\color{#35bf28}+0.08\%$
test_mod_wrap_and_backward[compile-overhead] 1.3063ms 0.9617ms 1.0399 KOps/s 1.0291 KOps/s $\color{#35bf28}+1.04\%$
test_seq_add[eager] 0.5804ms 0.1593ms 6.2776 KOps/s 6.4050 KOps/s $\color{#d91a1a}-1.99\%$
test_seq_add[compile] 0.6554ms 0.1581ms 6.3244 KOps/s 6.3306 KOps/s $\color{#d91a1a}-0.10\%$
test_seq_add[compile-overhead] 0.3128ms 0.2103ms 4.7557 KOps/s 5.0185 KOps/s $\textbf{\color{#d91a1a}-5.24\%}$
test_seq_wrap[eager] 0.6896ms 0.5701ms 1.7542 KOps/s 1.8891 KOps/s $\textbf{\color{#d91a1a}-7.14\%}$
test_seq_wrap[compile] 0.5461ms 0.4274ms 2.3397 KOps/s 2.4612 KOps/s $\color{#d91a1a}-4.94\%$
test_seq_wrap[compile-overhead] 0.7158ms 0.3041ms 3.2884 KOps/s 3.2671 KOps/s $\color{#35bf28}+0.65\%$
test_func_call_runtime[False-eager] 1.3380ms 0.8811ms 1.1349 KOps/s 1.1515 KOps/s $\color{#d91a1a}-1.44\%$
test_func_call_runtime[False-compile] 1.4673ms 0.9280ms 1.0776 KOps/s 1.0800 KOps/s $\color{#d91a1a}-0.22\%$
test_func_call_runtime[False-compile-overhead] 0.6045ms 0.4815ms 2.0770 KOps/s 2.0928 KOps/s $\color{#d91a1a}-0.75\%$
test_func_call_runtime[True-eager] 1.3058ms 1.1141ms 897.6094 Ops/s 902.7760 Ops/s $\color{#d91a1a}-0.57\%$
test_func_call_runtime[True-compile] 1.0969ms 0.9426ms 1.0609 KOps/s 1.0544 KOps/s $\color{#35bf28}+0.62\%$
test_func_call_runtime[True-compile-overhead] 0.5671ms 0.4987ms 2.0053 KOps/s 1.9878 KOps/s $\color{#35bf28}+0.88\%$
test_func_call_cm_runtime[False-eager] 1.0115ms 0.8912ms 1.1221 KOps/s 1.1494 KOps/s $\color{#d91a1a}-2.38\%$
test_func_call_cm_runtime[False-compile] 1.0187ms 0.9285ms 1.0771 KOps/s 1.0336 KOps/s $\color{#35bf28}+4.20\%$
test_func_call_cm_runtime[False-compile-overhead] 0.6534ms 0.4845ms 2.0639 KOps/s 2.0772 KOps/s $\color{#d91a1a}-0.64\%$
test_func_call_cm_runtime[True-eager] 1.4493ms 1.2483ms 801.0771 Ops/s 804.0398 Ops/s $\color{#d91a1a}-0.37\%$
test_func_call_cm_runtime[True-compile] 1.1026ms 0.9885ms 1.0116 KOps/s 1.0215 KOps/s $\color{#d91a1a}-0.96\%$
test_func_call_cm_runtime[True-compile-overhead] 0.5940ms 0.5362ms 1.8648 KOps/s 1.8564 KOps/s $\color{#35bf28}+0.46\%$
test_vmap_func_call_cm_runtime[eager] 2.8187ms 2.3421ms 426.9610 Ops/s 418.9674 Ops/s $\color{#35bf28}+1.91\%$
test_vmap_func_call_cm_runtime[compile] 1.0641ms 0.9951ms 1.0050 KOps/s 1.0111 KOps/s $\color{#d91a1a}-0.61\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.5835ms 0.5306ms 1.8848 KOps/s 1.8753 KOps/s $\color{#35bf28}+0.50\%$
test_distributed 0.8395ms 0.1535ms 6.5130 KOps/s 6.4651 KOps/s $\color{#35bf28}+0.74\%$
test_tdmodule 45.8100μs 26.8004μs 37.3129 KOps/s 36.0510 KOps/s $\color{#35bf28}+3.50\%$
test_tdmodule_dispatch 75.5510μs 46.0778μs 21.7024 KOps/s 20.9933 KOps/s $\color{#35bf28}+3.38\%$
test_tdseq 48.4210μs 26.4247μs 37.8434 KOps/s 38.2083 KOps/s $\color{#d91a1a}-0.96\%$
test_tdseq_dispatch 70.5810μs 49.6157μs 20.1549 KOps/s 20.1809 KOps/s $\color{#d91a1a}-0.13\%$
test_instantiation_functorch 2.0616ms 1.9884ms 502.9141 Ops/s 507.0151 Ops/s $\color{#d91a1a}-0.81\%$
test_exec_functorch 0.2253ms 0.1765ms 5.6663 KOps/s 5.6006 KOps/s $\color{#35bf28}+1.17\%$
test_exec_functional_call 0.2089ms 0.1586ms 6.3068 KOps/s 6.3610 KOps/s $\color{#d91a1a}-0.85\%$
test_exec_td_decorator 0.4434ms 0.2324ms 4.3030 KOps/s 4.3353 KOps/s $\color{#d91a1a}-0.74\%$
test_vmap_mlp_speed_decorator[True-True] 0.9887ms 0.7985ms 1.2523 KOps/s 1.2645 KOps/s $\color{#d91a1a}-0.97\%$
test_vmap_mlp_speed_decorator[True-False] 1.0011ms 0.7970ms 1.2547 KOps/s 1.2602 KOps/s $\color{#d91a1a}-0.43\%$
test_vmap_mlp_speed_decorator[False-True] 0.8836ms 0.6869ms 1.4559 KOps/s 1.4676 KOps/s $\color{#d91a1a}-0.80\%$
test_vmap_mlp_speed_decorator[False-False] 0.8319ms 0.6848ms 1.4603 KOps/s 1.4690 KOps/s $\color{#d91a1a}-0.59\%$
test_vmap_transformer_speed_decorator[True-True] 20.6462ms 20.5166ms 48.7411 Ops/s 48.7294 Ops/s $\color{#35bf28}+0.02\%$
test_vmap_transformer_speed_decorator[True-False] 21.1887ms 20.5395ms 48.6867 Ops/s 48.6452 Ops/s $\color{#35bf28}+0.09\%$
test_vmap_transformer_speed_decorator[False-True] 21.0078ms 20.3327ms 49.1818 Ops/s 49.1970 Ops/s $\color{#d91a1a}-0.03\%$
test_vmap_transformer_speed_decorator[False-False] 20.4759ms 20.3441ms 49.1542 Ops/s 49.1753 Ops/s $\color{#d91a1a}-0.04\%$
test_to_module_speed[True] 1.4878ms 1.4093ms 709.5694 Ops/s 709.8788 Ops/s $\color{#d91a1a}-0.04\%$
test_to_module_speed[False] 1.4854ms 1.3850ms 722.0025 Ops/s 723.3665 Ops/s $\color{#d91a1a}-0.19\%$
test_tc_init 79.2010μs 49.7817μs 20.0877 KOps/s 20.0206 KOps/s $\color{#35bf28}+0.33\%$
test_tc_init_tensor_only 36.4810μs 14.4214μs 69.3412 KOps/s 71.1863 KOps/s $\color{#d91a1a}-2.59\%$
test_tc_init_nested 0.1422ms 0.1005ms 9.9542 KOps/s 10.1533 KOps/s $\color{#d91a1a}-1.96\%$
test_tc_first_layer_tensor 25.1300μs 1.6906μs 591.4912 KOps/s 600.0769 KOps/s $\color{#d91a1a}-1.43\%$
test_tc_first_layer_tensor_only 3.1571μs 0.6478μs 1.5436 MOps/s 1.5622 MOps/s $\color{#d91a1a}-1.19\%$
test_tc_first_layer_tensor_set 36.6810μs 3.9412μs 253.7286 KOps/s 255.7710 KOps/s $\color{#d91a1a}-0.80\%$
test_tc_first_layer_tensor_only_set 35.3810μs 2.9359μs 340.6099 KOps/s 348.6504 KOps/s $\color{#d91a1a}-2.31\%$
test_tc_first_layer_nontensor 40.1810μs 5.6524μs 176.9167 KOps/s 177.2748 KOps/s $\color{#d91a1a}-0.20\%$
test_tc_second_layer_tensor 25.3510μs 4.0470μs 247.0971 KOps/s 249.7451 KOps/s $\color{#d91a1a}-1.06\%$
test_tc_second_layer_nontensor 47.7310μs 7.9916μs 125.1320 KOps/s 125.0927 KOps/s $\color{#35bf28}+0.03\%$
test_unbind 0.2572s 13.0833ms 76.4336 Ops/s 67.0412 Ops/s $\textbf{\color{#35bf28}+14.01\%}$
test_full_like 5.5609ms 4.3048ms 232.2981 Ops/s 59.3366 Ops/s $\textbf{\color{#35bf28}+291.49\%}$
test_zeros_like 4.4923ms 4.3690ms 228.8827 Ops/s 59.5699 Ops/s $\textbf{\color{#35bf28}+284.23\%}$
test_ones_like 4.5886ms 4.3889ms 227.8486 Ops/s 60.0703 Ops/s $\textbf{\color{#35bf28}+279.30\%}$
test_clone 6.7734ms 6.5294ms 153.1523 Ops/s 56.3420 Ops/s $\textbf{\color{#35bf28}+171.83\%}$
test_squeeze 0.1711ms 13.4989μs 74.0803 KOps/s 74.6943 KOps/s $\color{#d91a1a}-0.82\%$
test_unsqueeze 0.2594ms 0.1062ms 9.4184 KOps/s 9.5009 KOps/s $\color{#d91a1a}-0.87\%$
test_split 0.2310ms 0.1766ms 5.6614 KOps/s 5.5866 KOps/s $\color{#35bf28}+1.34\%$
test_permute 0.2546ms 0.1986ms 5.0357 KOps/s 5.0320 KOps/s $\color{#35bf28}+0.07\%$
test_stack 51.8125ms 51.4710ms 19.4284 Ops/s 19.4487 Ops/s $\color{#d91a1a}-0.10\%$
test_cat 51.8291ms 51.4438ms 19.4387 Ops/s 19.4184 Ops/s $\color{#35bf28}+0.10\%$

@github-actions
Copy link

github-actions bot commented Dec 18, 2025

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

Total Benchmarks: 233. Improved: $\large\color{#35bf28}20$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 31.7010μs 14.9746μs 66.7799 KOps/s 67.0173 KOps/s $\color{#d91a1a}-0.35\%$
test_plain_set_stack_nested 39.1210μs 15.3018μs 65.3517 KOps/s 66.5049 KOps/s $\color{#d91a1a}-1.73\%$
test_plain_set_nested_inplace 45.4910μs 16.5784μs 60.3196 KOps/s 59.5189 KOps/s $\color{#35bf28}+1.35\%$
test_plain_set_stack_nested_inplace 45.1910μs 16.8148μs 59.4715 KOps/s 59.7780 KOps/s $\color{#d91a1a}-0.51\%$
test_items 31.3900μs 5.8585μs 170.6909 KOps/s 164.9835 KOps/s $\color{#35bf28}+3.46\%$
test_items_nested 0.5807ms 0.5340ms 1.8726 KOps/s 1.8599 KOps/s $\color{#35bf28}+0.68\%$
test_items_nested_locked 0.6154ms 0.5424ms 1.8437 KOps/s 1.8456 KOps/s $\color{#d91a1a}-0.10\%$
test_items_nested_leaf 0.1341ms 95.9895μs 10.4178 KOps/s 10.4044 KOps/s $\color{#35bf28}+0.13\%$
test_items_stack_nested 0.5890ms 0.5303ms 1.8856 KOps/s 1.8900 KOps/s $\color{#d91a1a}-0.23\%$
test_items_stack_nested_leaf 0.1287ms 97.4601μs 10.2606 KOps/s 10.6119 KOps/s $\color{#d91a1a}-3.31\%$
test_items_stack_nested_locked 0.5889ms 0.5405ms 1.8501 KOps/s 1.8473 KOps/s $\color{#35bf28}+0.15\%$
test_keys 29.9600μs 4.2185μs 237.0493 KOps/s 235.3205 KOps/s $\color{#35bf28}+0.73\%$
test_keys_nested 0.1627ms 0.1197ms 8.3575 KOps/s 8.3361 KOps/s $\color{#35bf28}+0.26\%$
test_keys_nested_locked 2.1496ms 0.1295ms 7.7210 KOps/s 7.7282 KOps/s $\color{#d91a1a}-0.09\%$
test_keys_nested_leaf 0.1658ms 0.1110ms 9.0100 KOps/s 9.0372 KOps/s $\color{#d91a1a}-0.30\%$
test_keys_stack_nested 0.1667ms 0.1211ms 8.2550 KOps/s 8.3555 KOps/s $\color{#d91a1a}-1.20\%$
test_keys_stack_nested_leaf 0.1436ms 0.1109ms 9.0147 KOps/s 9.0750 KOps/s $\color{#d91a1a}-0.66\%$
test_keys_stack_nested_locked 0.1757ms 0.1296ms 7.7170 KOps/s 7.8918 KOps/s $\color{#d91a1a}-2.21\%$
test_values 6.9582μs 1.0179μs 982.3950 KOps/s 981.8793 KOps/s $\color{#35bf28}+0.05\%$
test_values_nested 83.9130μs 48.0550μs 20.8095 KOps/s 20.8230 KOps/s $\color{#d91a1a}-0.06\%$
test_values_nested_locked 79.0820μs 50.9579μs 19.6241 KOps/s 19.5061 KOps/s $\color{#35bf28}+0.60\%$
test_values_nested_leaf 84.6720μs 54.2068μs 18.4479 KOps/s 18.4730 KOps/s $\color{#d91a1a}-0.14\%$
test_values_stack_nested 82.8120μs 47.9742μs 20.8446 KOps/s 21.0657 KOps/s $\color{#d91a1a}-1.05\%$
test_values_stack_nested_leaf 91.2220μs 54.3773μs 18.3900 KOps/s 18.3403 KOps/s $\color{#35bf28}+0.27\%$
test_values_stack_nested_locked 84.8820μs 50.9867μs 19.6130 KOps/s 19.6415 KOps/s $\color{#d91a1a}-0.15\%$
test_membership 5.0835μs 0.8510μs 1.1751 MOps/s 1.1677 MOps/s $\color{#35bf28}+0.63\%$
test_membership_nested 37.1710μs 3.1722μs 315.2380 KOps/s 316.5651 KOps/s $\color{#d91a1a}-0.42\%$
test_membership_nested_leaf 30.0410μs 3.1493μs 317.5328 KOps/s 317.0283 KOps/s $\color{#35bf28}+0.16\%$
test_membership_stacked_nested 29.5210μs 3.1948μs 313.0052 KOps/s 315.9360 KOps/s $\color{#d91a1a}-0.93\%$
test_membership_stacked_nested_leaf 35.4810μs 3.1663μs 315.8305 KOps/s 314.6942 KOps/s $\color{#35bf28}+0.36\%$
test_membership_nested_last 31.0210μs 4.5956μs 217.5983 KOps/s 216.5488 KOps/s $\color{#35bf28}+0.48\%$
test_membership_nested_leaf_last 52.0910μs 4.6465μs 215.2155 KOps/s 214.8825 KOps/s $\color{#35bf28}+0.16\%$
test_membership_stacked_nested_last 29.7810μs 4.5876μs 217.9781 KOps/s 217.4376 KOps/s $\color{#35bf28}+0.25\%$
test_membership_stacked_nested_leaf_last 36.7610μs 4.6193μs 216.4845 KOps/s 215.3930 KOps/s $\color{#35bf28}+0.51\%$
test_nested_getleaf 44.4410μs 21.4190μs 46.6875 KOps/s 46.0710 KOps/s $\color{#35bf28}+1.34\%$
test_nested_get 42.0310μs 20.5011μs 48.7780 KOps/s 48.9338 KOps/s $\color{#d91a1a}-0.32\%$
test_stacked_getleaf 67.4110μs 21.7194μs 46.0418 KOps/s 47.1502 KOps/s $\color{#d91a1a}-2.35\%$
test_stacked_get 42.3710μs 20.6434μs 48.4416 KOps/s 48.6043 KOps/s $\color{#d91a1a}-0.33\%$
test_nested_getitemleaf 55.8120μs 22.1195μs 45.2090 KOps/s 44.5709 KOps/s $\color{#35bf28}+1.43\%$
test_nested_getitem 68.1920μs 20.7951μs 48.0882 KOps/s 46.6847 KOps/s $\color{#35bf28}+3.01\%$
test_stacked_getitemleaf 47.2510μs 22.0634μs 45.3239 KOps/s 44.7565 KOps/s $\color{#35bf28}+1.27\%$
test_stacked_getitem 71.1220μs 20.5553μs 48.6493 KOps/s 46.1080 KOps/s $\textbf{\color{#35bf28}+5.51\%}$
test_lock_nested 0.5922ms 0.4700ms 2.1275 KOps/s 2.1225 KOps/s $\color{#35bf28}+0.24\%$
test_lock_stack_nested 0.5256ms 0.4757ms 2.1023 KOps/s 2.0890 KOps/s $\color{#35bf28}+0.64\%$
test_unlock_nested 0.4777ms 0.3842ms 2.6030 KOps/s 2.5854 KOps/s $\color{#35bf28}+0.68\%$
test_unlock_stack_nested 0.4282ms 0.3813ms 2.6225 KOps/s 2.5801 KOps/s $\color{#35bf28}+1.64\%$
test_flatten_speed 0.1636ms 0.1221ms 8.1890 KOps/s 8.2340 KOps/s $\color{#d91a1a}-0.55\%$
test_unflatten_speed 0.6958ms 0.5955ms 1.6794 KOps/s 1.6751 KOps/s $\color{#35bf28}+0.25\%$
test_common_ops 0.8854ms 0.7530ms 1.3280 KOps/s 1.3252 KOps/s $\color{#35bf28}+0.21\%$
test_creation 0.1060ms 2.7368μs 365.3898 KOps/s 366.6763 KOps/s $\color{#d91a1a}-0.35\%$
test_creation_empty 29.8310μs 9.0557μs 110.4280 KOps/s 109.9362 KOps/s $\color{#35bf28}+0.45\%$
test_creation_nested_1 36.9710μs 12.1884μs 82.0451 KOps/s 82.1385 KOps/s $\color{#d91a1a}-0.11\%$
test_creation_nested_2 40.2210μs 16.1642μs 61.8652 KOps/s 61.5259 KOps/s $\color{#35bf28}+0.55\%$
test_clone 59.6220μs 13.4006μs 74.6238 KOps/s 72.4033 KOps/s $\color{#35bf28}+3.07\%$
test_getitem[int] 1.1611ms 14.0791μs 71.0274 KOps/s 70.7793 KOps/s $\color{#35bf28}+0.35\%$
test_getitem[slice_int] 0.1396ms 24.6837μs 40.5125 KOps/s 40.4260 KOps/s $\color{#35bf28}+0.21\%$
test_getitem[range] 0.1740ms 59.3883μs 16.8383 KOps/s 16.7370 KOps/s $\color{#35bf28}+0.61\%$
test_getitem[tuple] 0.1461ms 24.2370μs 41.2592 KOps/s 40.5546 KOps/s $\color{#35bf28}+1.74\%$
test_getitem[list] 0.2289ms 53.4289μs 18.7165 KOps/s 18.5951 KOps/s $\color{#35bf28}+0.65\%$
test_setitem_dim[int] 42.6810μs 24.7174μs 40.4573 KOps/s 39.9520 KOps/s $\color{#35bf28}+1.26\%$
test_setitem_dim[slice_int] 82.5320μs 45.2412μs 22.1038 KOps/s 21.9788 KOps/s $\color{#35bf28}+0.57\%$
test_setitem_dim[range] 0.1249ms 87.0580μs 11.4866 KOps/s 11.5584 KOps/s $\color{#d91a1a}-0.62\%$
test_setitem_dim[tuple] 72.6120μs 41.2410μs 24.2477 KOps/s 23.6951 KOps/s $\color{#35bf28}+2.33\%$
test_setitem 54.6020μs 18.4564μs 54.1816 KOps/s 52.8188 KOps/s $\color{#35bf28}+2.58\%$
test_set 48.6410μs 17.6511μs 56.6537 KOps/s 54.8341 KOps/s $\color{#35bf28}+3.32\%$
test_set_shared 0.4942ms 0.2055ms 4.8672 KOps/s 4.8216 KOps/s $\color{#35bf28}+0.95\%$
test_update 0.1938ms 22.9962μs 43.4855 KOps/s 43.7730 KOps/s $\color{#d91a1a}-0.66\%$
test_update_nested 65.3810μs 34.9804μs 28.5874 KOps/s 27.8939 KOps/s $\color{#35bf28}+2.49\%$
test_update__nested 0.4702ms 34.2746μs 29.1761 KOps/s 27.6292 KOps/s $\textbf{\color{#35bf28}+5.60\%}$
test_set_nested 54.4810μs 19.5447μs 51.1647 KOps/s 49.9660 KOps/s $\color{#35bf28}+2.40\%$
test_set_nested_new 60.8020μs 24.8354μs 40.2651 KOps/s 39.8404 KOps/s $\color{#35bf28}+1.07\%$
test_select 72.4520μs 42.9285μs 23.2945 KOps/s 23.3333 KOps/s $\color{#d91a1a}-0.17\%$
test_select_nested 0.1052ms 75.0182μs 13.3301 KOps/s 13.5740 KOps/s $\color{#d91a1a}-1.80\%$
test_exclude_nested 0.1460ms 99.6140μs 10.0388 KOps/s 10.0740 KOps/s $\color{#d91a1a}-0.35\%$
test_empty[True] 0.5073ms 0.4439ms 2.2528 KOps/s 2.2864 KOps/s $\color{#d91a1a}-1.47\%$
test_empty[False] 8.5775μs 1.3212μs 756.8755 KOps/s 773.3547 KOps/s $\color{#d91a1a}-2.13\%$
test_to 0.1039ms 71.2405μs 14.0370 KOps/s 13.5993 KOps/s $\color{#35bf28}+3.22\%$
test_to_nonblocking 0.1068ms 65.2804μs 15.3185 KOps/s 15.1083 KOps/s $\color{#35bf28}+1.39\%$
test_unbind_speed 0.3617ms 0.3274ms 3.0543 KOps/s 3.0918 KOps/s $\color{#d91a1a}-1.21\%$
test_unbind_speed_stack0 0.3798ms 0.3229ms 3.0968 KOps/s 3.1247 KOps/s $\color{#d91a1a}-0.89\%$
test_unbind_speed_stack1 98.7278ms 0.8621ms 1.1600 KOps/s 1.1524 KOps/s $\color{#35bf28}+0.66\%$
test_split 98.4472ms 1.2852ms 778.0862 Ops/s 641.6221 Ops/s $\textbf{\color{#35bf28}+21.27\%}$
test_chunk 98.4405ms 1.2168ms 821.8410 Ops/s 911.1402 Ops/s $\textbf{\color{#d91a1a}-9.80\%}$
test_consolidate[False-None] 3.9602ms 3.8855ms 257.3660 Ops/s 255.9695 Ops/s $\color{#35bf28}+0.55\%$
test_consolidate[default-None] 2.2582ms 2.0582ms 485.8574 Ops/s 452.8836 Ops/s $\textbf{\color{#35bf28}+7.28\%}$
test_consolidate[reduce-overhead-None] 2.4107ms 1.9863ms 503.4440 Ops/s 472.8283 Ops/s $\textbf{\color{#35bf28}+6.48\%}$
test_consolidate_njt[False-None] 9.2565ms 8.8905ms 112.4792 Ops/s 111.3932 Ops/s $\color{#35bf28}+0.97\%$
test_to[False-False-None] 2.5179ms 2.0958ms 477.1530 Ops/s 472.7540 Ops/s $\color{#35bf28}+0.93\%$
test_to[True-False-None] 0.1731s 2.2336ms 447.7110 Ops/s 521.0770 Ops/s $\textbf{\color{#d91a1a}-14.08\%}$
test_to[within-False-None] 6.2323ms 5.8647ms 170.5110 Ops/s 170.1554 Ops/s $\color{#35bf28}+0.21\%$
test_to[True-default-None] 12.6188ms 12.0688ms 82.8583 Ops/s 81.1198 Ops/s $\color{#35bf28}+2.14\%$
test_to_njt[False-False-None] 9.0427ms 8.7058ms 114.8655 Ops/s 112.8306 Ops/s $\color{#35bf28}+1.80\%$
test_to_njt[True-False-None] 7.7648ms 7.4298ms 134.5936 Ops/s 132.6296 Ops/s $\color{#35bf28}+1.48\%$
test_to_njt[within-False-None] 16.6713ms 16.2492ms 61.5415 Ops/s 60.0803 Ops/s $\color{#35bf28}+2.43\%$
test_creation[device0] 0.3629ms 0.1131ms 8.8454 KOps/s 8.7503 KOps/s $\color{#35bf28}+1.09\%$
test_creation_from_tensor 0.3527ms 0.1139ms 8.7787 KOps/s 8.7775 KOps/s $\color{#35bf28}+0.01\%$
test_add_one[memmap_tensor0] 0.4245ms 6.8368μs 146.2681 KOps/s 136.5787 KOps/s $\textbf{\color{#35bf28}+7.09\%}$
test_contiguous[memmap_tensor0] 29.5910μs 0.7502μs 1.3330 MOps/s 2.0370 MOps/s $\textbf{\color{#d91a1a}-34.56\%}$
test_stack[memmap_tensor0] 26.5610μs 4.5884μs 217.9396 KOps/s 214.7511 KOps/s $\color{#35bf28}+1.48\%$
test_memmaptd_index 1.0243ms 0.2795ms 3.5773 KOps/s 3.4754 KOps/s $\color{#35bf28}+2.93\%$
test_memmaptd_index_astensor 0.8113ms 0.3752ms 2.6652 KOps/s 2.6192 KOps/s $\color{#35bf28}+1.76\%$
test_memmaptd_index_op 1.0556ms 0.6257ms 1.5983 KOps/s 1.5866 KOps/s $\color{#35bf28}+0.74\%$
test_serialize_model 0.1349s 0.1338s 7.4741 Ops/s 7.5184 Ops/s $\color{#d91a1a}-0.59\%$
test_serialize_model_pickle 1.3493s 1.2109s 0.8258 Ops/s 0.8355 Ops/s $\color{#d91a1a}-1.16\%$
test_serialize_weights 0.3034s 0.1578s 6.3383 Ops/s 7.5357 Ops/s $\textbf{\color{#d91a1a}-15.89\%}$
test_serialize_weights_returnearly 0.4204s 69.8042ms 14.3258 Ops/s 11.2891 Ops/s $\textbf{\color{#35bf28}+26.90\%}$
test_serialize_weights_pickle 1.3511s 1.1851s 0.8438 Ops/s 0.8202 Ops/s $\color{#35bf28}+2.88\%$
test_reshape_pytree 0.3730ms 33.5166μs 29.8359 KOps/s 29.7196 KOps/s $\color{#35bf28}+0.39\%$
test_reshape_td 65.0820μs 39.0359μs 25.6174 KOps/s 26.1398 KOps/s $\color{#d91a1a}-2.00\%$
test_view_pytree 0.2214ms 32.9967μs 30.3060 KOps/s 30.4573 KOps/s $\color{#d91a1a}-0.50\%$
test_view_td 0.1146ms 46.3865μs 21.5580 KOps/s 21.5685 KOps/s $\color{#d91a1a}-0.05\%$
test_unbind_pytree 0.2425ms 37.8610μs 26.4124 KOps/s 26.2699 KOps/s $\color{#35bf28}+0.54\%$
test_unbind_td 91.6530μs 48.6940μs 20.5364 KOps/s 20.3835 KOps/s $\color{#35bf28}+0.75\%$
test_split_pytree 0.2559ms 44.3133μs 22.5666 KOps/s 22.2485 KOps/s $\color{#35bf28}+1.43\%$
test_split_td 0.1451ms 65.4322μs 15.2830 KOps/s 14.9817 KOps/s $\color{#35bf28}+2.01\%$
test_add_pytree 0.2280ms 45.9193μs 21.7773 KOps/s 22.3749 KOps/s $\color{#d91a1a}-2.67\%$
test_add_td 0.1217ms 57.0127μs 17.5400 KOps/s 18.1858 KOps/s $\color{#d91a1a}-3.55\%$
test_compile_add_one_nested[tensordict-compile] 0.2899ms 0.1788ms 5.5941 KOps/s 5.2864 KOps/s $\textbf{\color{#35bf28}+5.82\%}$
test_compile_add_one_nested[tensordict-eager] 0.2806ms 0.1944ms 5.1437 KOps/s 5.1330 KOps/s $\color{#35bf28}+0.21\%$
test_compile_add_one_nested[pytree-compile] 0.4882ms 0.1499ms 6.6696 KOps/s 6.2600 KOps/s $\textbf{\color{#35bf28}+6.54\%}$
test_compile_add_one_nested[pytree-eager] 0.4379ms 0.1891ms 5.2870 KOps/s 5.2245 KOps/s $\color{#35bf28}+1.20\%$
test_compile_copy_nested[tensordict-compile] 0.1214ms 27.7038μs 36.0961 KOps/s 32.9895 KOps/s $\textbf{\color{#35bf28}+9.42\%}$
test_compile_copy_nested[tensordict-eager] 85.1420μs 53.1363μs 18.8195 KOps/s 18.8500 KOps/s $\color{#d91a1a}-0.16\%$
test_compile_copy_nested[pytree-compile] 0.1415ms 14.6622μs 68.2026 KOps/s 67.1996 KOps/s $\color{#35bf28}+1.49\%$
test_compile_copy_nested[pytree-eager] 0.4025ms 76.0520μs 13.1489 KOps/s 13.0719 KOps/s $\color{#35bf28}+0.59\%$
test_compile_add_one_flat[tensordict-compile] 0.2481ms 0.2067ms 4.8382 KOps/s 4.6242 KOps/s $\color{#35bf28}+4.63\%$
test_compile_add_one_flat[tensordict-eager] 0.3060ms 0.2627ms 3.8067 KOps/s 3.7975 KOps/s $\color{#35bf28}+0.24\%$
test_compile_add_one_flat[tensorclass-compile] 0.2076ms 0.1533ms 6.5224 KOps/s 6.3243 KOps/s $\color{#35bf28}+3.13\%$
test_compile_add_one_flat[tensorclass-eager] 0.1152ms 71.3627μs 14.0129 KOps/s 13.9184 KOps/s $\color{#35bf28}+0.68\%$
test_compile_add_one_flat[pytree-compile] 0.2502ms 0.2028ms 4.9304 KOps/s 4.7314 KOps/s $\color{#35bf28}+4.20\%$
test_compile_add_one_flat[pytree-eager] 0.8056ms 0.5492ms 1.8207 KOps/s 1.7900 KOps/s $\color{#35bf28}+1.72\%$
test_compile_add_self_flat[tensordict-eager] 0.3839ms 0.3171ms 3.1540 KOps/s 3.1487 KOps/s $\color{#35bf28}+0.17\%$
test_compile_add_self_flat[tensordict-compile] 0.2516ms 0.2098ms 4.7674 KOps/s 4.6019 KOps/s $\color{#35bf28}+3.59\%$
test_compile_add_self_flat[tensorclass-eager] 0.1341ms 87.9828μs 11.3659 KOps/s 11.3619 KOps/s $\color{#35bf28}+0.03\%$
test_compile_add_self_flat[tensorclass-compile] 0.2342ms 0.1565ms 6.3910 KOps/s 6.2353 KOps/s $\color{#35bf28}+2.50\%$
test_compile_add_self_flat[pytree-eager] 0.6856ms 0.4567ms 2.1894 KOps/s 2.1392 KOps/s $\color{#35bf28}+2.35\%$
test_compile_add_self_flat[pytree-compile] 0.2506ms 0.2036ms 4.9117 KOps/s 4.7601 KOps/s $\color{#35bf28}+3.18\%$
test_compile_copy_flat[tensordict-compile] 0.1301ms 24.2169μs 41.2934 KOps/s 40.7719 KOps/s $\color{#35bf28}+1.28\%$
test_compile_copy_flat[tensordict-eager] 73.5520μs 41.3914μs 24.1596 KOps/s 23.7967 KOps/s $\color{#35bf28}+1.52\%$
test_compile_copy_flat[pytree-compile] 0.1068ms 21.3382μs 46.8642 KOps/s 46.4816 KOps/s $\color{#35bf28}+0.82\%$
test_compile_copy_flat[pytree-eager] 0.3572ms 69.7631μs 14.3342 KOps/s 14.2696 KOps/s $\color{#35bf28}+0.45\%$
test_compile_assign_and_add[tensordict-compile] 2.0668ms 0.2107ms 4.7470 KOps/s 4.7022 KOps/s $\color{#35bf28}+0.95\%$
test_compile_assign_and_add[tensordict-eager] 3.5047ms 3.3210ms 301.1134 Ops/s 303.3430 Ops/s $\color{#d91a1a}-0.74\%$
test_compile_assign_and_add[pytree-compile] 2.0438ms 0.2083ms 4.8018 KOps/s 4.7411 KOps/s $\color{#35bf28}+1.28\%$
test_compile_assign_and_add[pytree-eager] 3.0896ms 2.9311ms 341.1652 Ops/s 339.3342 Ops/s $\color{#35bf28}+0.54\%$
test_compile_indexing[tensor-tensordict-compile] 0.2173ms 0.1429ms 6.9958 KOps/s 6.9269 KOps/s $\color{#35bf28}+0.99\%$
test_compile_indexing[tensor-tensordict-eager] 0.3005ms 67.4163μs 14.8332 KOps/s 14.8499 KOps/s $\color{#d91a1a}-0.11\%$
test_compile_indexing[tensor-tensorclass-compile] 0.2371ms 0.1380ms 7.2477 KOps/s 6.9857 KOps/s $\color{#35bf28}+3.75\%$
test_compile_indexing[tensor-tensorclass-eager] 0.2630ms 47.2799μs 21.1506 KOps/s 20.9767 KOps/s $\color{#35bf28}+0.83\%$
test_compile_indexing[tensor-pytree-compile] 0.1977ms 0.1379ms 7.2512 KOps/s 6.9298 KOps/s $\color{#35bf28}+4.64\%$
test_compile_indexing[tensor-pytree-eager] 0.2431ms 46.7359μs 21.3968 KOps/s 20.7801 KOps/s $\color{#35bf28}+2.97\%$
test_compile_indexing[slice-tensordict-compile] 0.2208ms 85.7144μs 11.6666 KOps/s 11.0382 KOps/s $\textbf{\color{#35bf28}+5.69\%}$
test_compile_indexing[slice-tensordict-eager] 0.2108ms 27.0155μs 37.0158 KOps/s 36.3769 KOps/s $\color{#35bf28}+1.76\%$
test_compile_indexing[slice-tensorclass-compile] 0.1822ms 80.7195μs 12.3886 KOps/s 11.9533 KOps/s $\color{#35bf28}+3.64\%$
test_compile_indexing[slice-tensorclass-eager] 0.2336ms 24.0251μs 41.6231 KOps/s 41.6552 KOps/s $\color{#d91a1a}-0.08\%$
test_compile_indexing[slice-pytree-compile] 0.1308ms 80.8781μs 12.3643 KOps/s 12.0453 KOps/s $\color{#35bf28}+2.65\%$
test_compile_indexing[slice-pytree-eager] 0.2471ms 23.7294μs 42.1418 KOps/s 42.0246 KOps/s $\color{#35bf28}+0.28\%$
test_compile_indexing[int-tensordict-compile] 0.1389ms 87.0653μs 11.4856 KOps/s 11.1520 KOps/s $\color{#35bf28}+2.99\%$
test_compile_indexing[int-tensordict-eager] 0.2077ms 26.6858μs 37.4731 KOps/s 37.1812 KOps/s $\color{#35bf28}+0.78\%$
test_compile_indexing[int-tensorclass-compile] 0.1312ms 80.9494μs 12.3534 KOps/s 11.7015 KOps/s $\textbf{\color{#35bf28}+5.57\%}$
test_compile_indexing[int-tensorclass-eager] 0.2362ms 23.6728μs 42.2425 KOps/s 41.6690 KOps/s $\color{#35bf28}+1.38\%$
test_compile_indexing[int-pytree-compile] 0.1252ms 80.9112μs 12.3592 KOps/s 11.6189 KOps/s $\textbf{\color{#35bf28}+6.37\%}$
test_compile_indexing[int-pytree-eager] 0.2291ms 23.7610μs 42.0859 KOps/s 41.7476 KOps/s $\color{#35bf28}+0.81\%$
test_mod_add[eager] 91.3230μs 52.1568μs 19.1729 KOps/s 18.8433 KOps/s $\color{#35bf28}+1.75\%$
test_mod_add[compile] 0.2375ms 0.1510ms 6.6237 KOps/s 6.4423 KOps/s $\color{#35bf28}+2.81\%$
test_mod_add[compile-overhead] 0.5750ms 0.1985ms 5.0371 KOps/s 5.0048 KOps/s $\color{#35bf28}+0.65\%$
test_mod_wrap[eager] 0.3953ms 0.3100ms 3.2260 KOps/s 3.0871 KOps/s $\color{#35bf28}+4.50\%$
test_mod_wrap[compile] 0.5511ms 0.4029ms 2.4822 KOps/s 2.4101 KOps/s $\color{#35bf28}+2.99\%$
test_mod_wrap[compile-overhead] 7.4749ms 3.9905ms 250.5947 Ops/s 290.5497 Ops/s $\textbf{\color{#d91a1a}-13.75\%}$
test_mod_wrap_and_backward[eager] 1.6974ms 1.5772ms 634.0210 Ops/s 596.2820 Ops/s $\textbf{\color{#35bf28}+6.33\%}$
test_mod_wrap_and_backward[compile] 1.7253ms 1.6294ms 613.7177 Ops/s 564.5481 Ops/s $\textbf{\color{#35bf28}+8.71\%}$
test_mod_wrap_and_backward[compile-overhead] 1.3575ms 0.9901ms 1.0100 KOps/s 893.3878 Ops/s $\textbf{\color{#35bf28}+13.05\%}$
test_seq_add[eager] 0.2181ms 0.1600ms 6.2493 KOps/s 6.2078 KOps/s $\color{#35bf28}+0.67\%$
test_seq_add[compile] 0.2620ms 0.1606ms 6.2259 KOps/s 6.0513 KOps/s $\color{#35bf28}+2.88\%$
test_seq_add[compile-overhead] 0.2649ms 0.2055ms 4.8667 KOps/s 4.8620 KOps/s $\color{#35bf28}+0.10\%$
test_seq_wrap[eager] 0.6302ms 0.5476ms 1.8263 KOps/s 1.7652 KOps/s $\color{#35bf28}+3.46\%$
test_seq_wrap[compile] 0.5429ms 0.4408ms 2.2687 KOps/s 2.3062 KOps/s $\color{#d91a1a}-1.63\%$
test_seq_wrap[compile-overhead] 0.3962ms 0.3218ms 3.1073 KOps/s 3.1568 KOps/s $\color{#d91a1a}-1.57\%$
test_func_call_runtime[False-eager] 0.9673ms 0.8889ms 1.1250 KOps/s 1.0604 KOps/s $\textbf{\color{#35bf28}+6.09\%}$
test_func_call_runtime[False-compile] 1.0386ms 0.9487ms 1.0541 KOps/s 1.0201 KOps/s $\color{#35bf28}+3.33\%$
test_func_call_runtime[False-compile-overhead] 0.5581ms 0.4980ms 2.0079 KOps/s 1.9853 KOps/s $\color{#35bf28}+1.14\%$
test_func_call_runtime[True-eager] 1.2694ms 1.1255ms 888.4896 Ops/s 862.3208 Ops/s $\color{#35bf28}+3.03\%$
test_func_call_runtime[True-compile] 1.1103ms 0.9703ms 1.0306 KOps/s 1.0004 KOps/s $\color{#35bf28}+3.03\%$
test_func_call_runtime[True-compile-overhead] 0.5694ms 0.5182ms 1.9296 KOps/s 1.8950 KOps/s $\color{#35bf28}+1.82\%$
test_func_call_cm_runtime[False-eager] 0.9521ms 0.8746ms 1.1434 KOps/s 1.0610 KOps/s $\textbf{\color{#35bf28}+7.76\%}$
test_func_call_cm_runtime[False-compile] 1.1010ms 0.9493ms 1.0534 KOps/s 1.0381 KOps/s $\color{#35bf28}+1.47\%$
test_func_call_cm_runtime[False-compile-overhead] 0.5773ms 0.5015ms 1.9942 KOps/s 1.9674 KOps/s $\color{#35bf28}+1.36\%$
test_func_call_cm_runtime[True-eager] 1.3756ms 1.2723ms 785.9772 Ops/s 753.9375 Ops/s $\color{#35bf28}+4.25\%$
test_func_call_cm_runtime[True-compile] 1.0981ms 1.0032ms 996.7925 Ops/s 990.1574 Ops/s $\color{#35bf28}+0.67\%$
test_func_call_cm_runtime[True-compile-overhead] 0.6129ms 0.5512ms 1.8141 KOps/s 1.7952 KOps/s $\color{#35bf28}+1.05\%$
test_vmap_func_call_cm_runtime[eager] 2.9053ms 2.3889ms 418.5991 Ops/s 411.5573 Ops/s $\color{#35bf28}+1.71\%$
test_vmap_func_call_cm_runtime[compile] 1.0751ms 1.0155ms 984.7486 Ops/s 976.0139 Ops/s $\color{#35bf28}+0.89\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.5995ms 0.5449ms 1.8353 KOps/s 1.8066 KOps/s $\color{#35bf28}+1.59\%$
test_distributed 0.4913ms 0.1549ms 6.4563 KOps/s 5.5989 KOps/s $\textbf{\color{#35bf28}+15.31\%}$
test_tdmodule 63.7520μs 27.7092μs 36.0890 KOps/s 35.6654 KOps/s $\color{#35bf28}+1.19\%$
test_tdmodule_dispatch 78.6520μs 48.4752μs 20.6291 KOps/s 20.5181 KOps/s $\color{#35bf28}+0.54\%$
test_tdseq 56.2710μs 26.9124μs 37.1577 KOps/s 36.8815 KOps/s $\color{#35bf28}+0.75\%$
test_tdseq_dispatch 70.5520μs 50.6320μs 19.7504 KOps/s 19.6316 KOps/s $\color{#35bf28}+0.60\%$
test_instantiation_functorch 2.5151ms 2.0910ms 478.2501 Ops/s 477.0532 Ops/s $\color{#35bf28}+0.25\%$
test_exec_functorch 0.2212ms 0.1854ms 5.3931 KOps/s 5.3533 KOps/s $\color{#35bf28}+0.74\%$
test_exec_functional_call 0.6026ms 0.1621ms 6.1686 KOps/s 6.0275 KOps/s $\color{#35bf28}+2.34\%$
test_exec_td_decorator 0.6784ms 0.2429ms 4.1166 KOps/s 4.0501 KOps/s $\color{#35bf28}+1.64\%$
test_vmap_mlp_speed_decorator[True-True] 1.2356ms 0.8164ms 1.2249 KOps/s 1.2146 KOps/s $\color{#35bf28}+0.84\%$
test_vmap_mlp_speed_decorator[True-False] 1.0096ms 0.8159ms 1.2257 KOps/s 1.2125 KOps/s $\color{#35bf28}+1.08\%$
test_vmap_mlp_speed_decorator[False-True] 1.1106ms 0.7027ms 1.4231 KOps/s 1.4100 KOps/s $\color{#35bf28}+0.93\%$
test_vmap_mlp_speed_decorator[False-False] 0.8584ms 0.7014ms 1.4257 KOps/s 1.4103 KOps/s $\color{#35bf28}+1.09\%$
test_vmap_transformer_speed_decorator[True-True] 21.3159ms 20.9232ms 47.7939 Ops/s 47.5614 Ops/s $\color{#35bf28}+0.49\%$
test_vmap_transformer_speed_decorator[True-False] 21.4034ms 20.9592ms 47.7117 Ops/s 47.4691 Ops/s $\color{#35bf28}+0.51\%$
test_vmap_transformer_speed_decorator[False-True] 21.1813ms 20.7486ms 48.1960 Ops/s 48.0115 Ops/s $\color{#35bf28}+0.38\%$
test_vmap_transformer_speed_decorator[False-False] 21.1364ms 20.7498ms 48.1933 Ops/s 47.9960 Ops/s $\color{#35bf28}+0.41\%$
test_to_module_speed[True] 1.9032ms 1.4843ms 673.7321 Ops/s 657.7840 Ops/s $\color{#35bf28}+2.42\%$
test_to_module_speed[False] 1.5897ms 1.4524ms 688.5324 Ops/s 661.6551 Ops/s $\color{#35bf28}+4.06\%$
test_tc_init 81.2320μs 51.8260μs 19.2953 KOps/s 19.5178 KOps/s $\color{#d91a1a}-1.14\%$
test_tc_init_tensor_only 42.5610μs 15.1593μs 65.9663 KOps/s 66.2778 KOps/s $\color{#d91a1a}-0.47\%$
test_tc_init_nested 0.1385ms 0.1020ms 9.7997 KOps/s 9.8438 KOps/s $\color{#d91a1a}-0.45\%$
test_tc_first_layer_tensor 27.2110μs 1.7758μs 563.1375 KOps/s 559.0892 KOps/s $\color{#35bf28}+0.72\%$
test_tc_first_layer_tensor_only 3.6101μs 0.6638μs 1.5064 MOps/s 1.4814 MOps/s $\color{#35bf28}+1.69\%$
test_tc_first_layer_tensor_set 30.7310μs 4.1680μs 239.9249 KOps/s 241.8889 KOps/s $\color{#d91a1a}-0.81\%$
test_tc_first_layer_tensor_only_set 29.7000μs 3.1264μs 319.8522 KOps/s 331.0226 KOps/s $\color{#d91a1a}-3.37\%$
test_tc_first_layer_nontensor 23.6110μs 6.0127μs 166.3160 KOps/s 167.8177 KOps/s $\color{#d91a1a}-0.89\%$
test_tc_second_layer_tensor 40.0010μs 4.3817μs 228.2203 KOps/s 232.0858 KOps/s $\color{#d91a1a}-1.67\%$
test_tc_second_layer_nontensor 33.6610μs 8.3987μs 119.0663 KOps/s 119.6827 KOps/s $\color{#d91a1a}-0.52\%$
test_unbind 0.2537s 15.2427ms 65.6051 Ops/s 64.2536 Ops/s $\color{#35bf28}+2.10\%$
test_full_like 4.5152ms 4.3883ms 227.8769 Ops/s 134.6259 Ops/s $\textbf{\color{#35bf28}+69.27\%}$
test_zeros_like 4.5139ms 4.3754ms 228.5523 Ops/s 228.1913 Ops/s $\color{#35bf28}+0.16\%$
test_ones_like 7.5427ms 4.3988ms 227.3349 Ops/s 233.8510 Ops/s $\color{#d91a1a}-2.79\%$
test_clone 6.8396ms 6.5528ms 152.6064 Ops/s 152.9611 Ops/s $\color{#d91a1a}-0.23\%$
test_squeeze 63.9820μs 14.2226μs 70.3105 KOps/s 69.9303 KOps/s $\color{#35bf28}+0.54\%$
test_unsqueeze 0.1718ms 0.1086ms 9.2099 KOps/s 9.1760 KOps/s $\color{#35bf28}+0.37\%$
test_split 0.2419ms 0.1849ms 5.4074 KOps/s 5.4278 KOps/s $\color{#d91a1a}-0.38\%$
test_permute 0.2777ms 0.2049ms 4.8799 KOps/s 4.8774 KOps/s $\color{#35bf28}+0.05\%$
test_stack 51.8703ms 51.5614ms 19.3944 Ops/s 19.4347 Ops/s $\color{#d91a1a}-0.21\%$
test_cat 51.7914ms 51.5651ms 19.3929 Ops/s 19.4461 Ops/s $\color{#d91a1a}-0.27\%$

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants