Skip to content

Conversation

@vmoens
Copy link
Collaborator

@vmoens vmoens commented Sep 24, 2025

Description

Describe your changes in detail.

Motivation and Context

Why is this change required? What problem does it solve?
If it fixes an open issue, please link to the issue here.
You can use the syntax close #15213 if this solves the issue #15213

  • I have raised an issue to propose this change (required for new features and bug fixes)

Types of changes

What types of changes does your code introduce? Remove all that do not apply:

  • Bug fix (non-breaking change which fixes an issue)
  • New feature (non-breaking change which adds core functionality)
  • Breaking change (fix or feature that would cause existing functionality to change)
  • Documentation (update in the documentation)
  • Example (update in the folder of examples)

Checklist

Go over all the following points, and put an x in all the boxes that apply.
If you are unsure about any of these, don't hesitate to ask. We are here to help!

  • I have read the CONTRIBUTION guide (required)
  • My change requires a change to the documentation.
  • I have updated the tests accordingly (required for a bug fix or a new feature).
  • I have updated the documentation accordingly.

@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 Sep 24, 2025
@github-actions
Copy link

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

Total Benchmarks: 233. Improved: $\large\color{#35bf28}14$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 38.8210μs 14.5368μs 68.7908 KOps/s 68.3098 KOps/s $\color{#35bf28}+0.70\%$
test_plain_set_stack_nested 37.6810μs 14.6275μs 68.3645 KOps/s 68.4691 KOps/s $\color{#d91a1a}-0.15\%$
test_plain_set_nested_inplace 42.0310μs 16.0863μs 62.1645 KOps/s 61.4825 KOps/s $\color{#35bf28}+1.11\%$
test_plain_set_stack_nested_inplace 53.0910μs 16.1242μs 62.0186 KOps/s 61.7609 KOps/s $\color{#35bf28}+0.42\%$
test_items 33.1700μs 5.8242μs 171.6959 KOps/s 165.9533 KOps/s $\color{#35bf28}+3.46\%$
test_items_nested 0.5710ms 0.5266ms 1.8990 KOps/s 1.9015 KOps/s $\color{#d91a1a}-0.13\%$
test_items_nested_locked 0.5886ms 0.5294ms 1.8888 KOps/s 1.9005 KOps/s $\color{#d91a1a}-0.61\%$
test_items_nested_leaf 0.1438ms 90.9117μs 10.9997 KOps/s 10.9084 KOps/s $\color{#35bf28}+0.84\%$
test_items_stack_nested 0.5807ms 0.5250ms 1.9046 KOps/s 1.9152 KOps/s $\color{#d91a1a}-0.55\%$
test_items_stack_nested_leaf 0.1271ms 92.9663μs 10.7566 KOps/s 10.9127 KOps/s $\color{#d91a1a}-1.43\%$
test_items_stack_nested_locked 0.6280ms 0.5275ms 1.8959 KOps/s 1.9236 KOps/s $\color{#d91a1a}-1.44\%$
test_keys 25.1100μs 4.1625μs 240.2396 KOps/s 242.5968 KOps/s $\color{#d91a1a}-0.97\%$
test_keys_nested 0.1642ms 0.1165ms 8.5858 KOps/s 8.5745 KOps/s $\color{#35bf28}+0.13\%$
test_keys_nested_locked 2.2734ms 0.1259ms 7.9449 KOps/s 7.9298 KOps/s $\color{#35bf28}+0.19\%$
test_keys_nested_leaf 0.1465ms 0.1071ms 9.3332 KOps/s 9.3380 KOps/s $\color{#d91a1a}-0.05\%$
test_keys_stack_nested 0.1503ms 0.1167ms 8.5712 KOps/s 8.5473 KOps/s $\color{#35bf28}+0.28\%$
test_keys_stack_nested_leaf 0.1490ms 0.1064ms 9.4024 KOps/s 9.3445 KOps/s $\color{#35bf28}+0.62\%$
test_keys_stack_nested_locked 0.1543ms 0.1237ms 8.0848 KOps/s 7.9052 KOps/s $\color{#35bf28}+2.27\%$
test_values 12.3502μs 1.0255μs 975.1325 KOps/s 1.0016 MOps/s $\color{#d91a1a}-2.64\%$
test_values_nested 0.1074ms 46.1197μs 21.6827 KOps/s 21.4521 KOps/s $\color{#35bf28}+1.08\%$
test_values_nested_locked 78.8510μs 49.5588μs 20.1781 KOps/s 20.1938 KOps/s $\color{#d91a1a}-0.08\%$
test_values_nested_leaf 87.9520μs 52.5452μs 19.0312 KOps/s 18.9700 KOps/s $\color{#35bf28}+0.32\%$
test_values_stack_nested 70.4810μs 46.1970μs 21.6464 KOps/s 21.3062 KOps/s $\color{#35bf28}+1.60\%$
test_values_stack_nested_leaf 87.8210μs 52.6122μs 19.0070 KOps/s 18.8610 KOps/s $\color{#35bf28}+0.77\%$
test_values_stack_nested_locked 0.1023ms 49.1719μs 20.3368 KOps/s 20.1933 KOps/s $\color{#35bf28}+0.71\%$
test_membership 4.7933μs 0.8006μs 1.2490 MOps/s 1.2330 MOps/s $\color{#35bf28}+1.30\%$
test_membership_nested 33.3810μs 2.9346μs 340.7626 KOps/s 335.5009 KOps/s $\color{#35bf28}+1.57\%$
test_membership_nested_leaf 23.4610μs 2.9665μs 337.0988 KOps/s 332.3719 KOps/s $\color{#35bf28}+1.42\%$
test_membership_stacked_nested 37.1710μs 2.9609μs 337.7394 KOps/s 334.9832 KOps/s $\color{#35bf28}+0.82\%$
test_membership_stacked_nested_leaf 28.3210μs 2.9662μs 337.1314 KOps/s 333.6837 KOps/s $\color{#35bf28}+1.03\%$
test_membership_nested_last 31.7600μs 4.3895μs 227.8166 KOps/s 228.7356 KOps/s $\color{#d91a1a}-0.40\%$
test_membership_nested_leaf_last 49.6710μs 4.3308μs 230.9058 KOps/s 228.3970 KOps/s $\color{#35bf28}+1.10\%$
test_membership_stacked_nested_last 26.0210μs 4.3455μs 230.1229 KOps/s 227.7076 KOps/s $\color{#35bf28}+1.06\%$
test_membership_stacked_nested_leaf_last 40.3210μs 4.3627μs 229.2141 KOps/s 228.8937 KOps/s $\color{#35bf28}+0.14\%$
test_nested_getleaf 52.2600μs 20.8295μs 48.0089 KOps/s 48.2015 KOps/s $\color{#d91a1a}-0.40\%$
test_nested_get 42.9610μs 19.5620μs 51.1196 KOps/s 50.5763 KOps/s $\color{#35bf28}+1.07\%$
test_stacked_getleaf 47.6210μs 20.6768μs 48.3634 KOps/s 48.2784 KOps/s $\color{#35bf28}+0.18\%$
test_stacked_get 42.9510μs 19.6314μs 50.9389 KOps/s 51.4424 KOps/s $\color{#d91a1a}-0.98\%$
test_nested_getitemleaf 47.1010μs 21.2282μs 47.1071 KOps/s 47.4521 KOps/s $\color{#d91a1a}-0.73\%$
test_nested_getitem 49.9300μs 20.3342μs 49.1783 KOps/s 50.3227 KOps/s $\color{#d91a1a}-2.27\%$
test_stacked_getitemleaf 52.2410μs 21.0784μs 47.4420 KOps/s 47.3649 KOps/s $\color{#35bf28}+0.16\%$
test_stacked_getitem 49.1600μs 20.0591μs 49.8528 KOps/s 49.3394 KOps/s $\color{#35bf28}+1.04\%$
test_lock_nested 0.5451ms 0.4644ms 2.1535 KOps/s 2.1567 KOps/s $\color{#d91a1a}-0.15\%$
test_lock_stack_nested 0.5347ms 0.4600ms 2.1739 KOps/s 2.1467 KOps/s $\color{#35bf28}+1.26\%$
test_unlock_nested 0.4733ms 0.3733ms 2.6790 KOps/s 2.6442 KOps/s $\color{#35bf28}+1.32\%$
test_unlock_stack_nested 0.4206ms 0.3694ms 2.7074 KOps/s 2.6458 KOps/s $\color{#35bf28}+2.33\%$
test_flatten_speed 0.1464ms 0.1172ms 8.5307 KOps/s 8.4387 KOps/s $\color{#35bf28}+1.09\%$
test_unflatten_speed 0.6762ms 0.5678ms 1.7611 KOps/s 1.7445 KOps/s $\color{#35bf28}+0.95\%$
test_common_ops 0.8796ms 0.7281ms 1.3735 KOps/s 1.3491 KOps/s $\color{#35bf28}+1.81\%$
test_creation 69.8010μs 2.5496μs 392.2260 KOps/s 383.4035 KOps/s $\color{#35bf28}+2.30\%$
test_creation_empty 42.9810μs 8.5019μs 117.6203 KOps/s 115.7564 KOps/s $\color{#35bf28}+1.61\%$
test_creation_nested_1 41.2310μs 11.4044μs 87.6852 KOps/s 85.2455 KOps/s $\color{#35bf28}+2.86\%$
test_creation_nested_2 49.2510μs 15.3143μs 65.2985 KOps/s 64.7760 KOps/s $\color{#35bf28}+0.81\%$
test_clone 36.7900μs 13.0710μs 76.5053 KOps/s 76.0523 KOps/s $\color{#35bf28}+0.60\%$
test_getitem[int] 1.2351ms 14.1274μs 70.7846 KOps/s 69.9225 KOps/s $\color{#35bf28}+1.23\%$
test_getitem[slice_int] 0.1478ms 28.5508μs 35.0252 KOps/s 33.2623 KOps/s $\textbf{\color{#35bf28}+5.30\%}$
test_getitem[range] 0.1625ms 48.3780μs 20.6706 KOps/s 20.3434 KOps/s $\color{#35bf28}+1.61\%$
test_getitem[tuple] 0.1434ms 23.9865μs 41.6901 KOps/s 40.9997 KOps/s $\color{#35bf28}+1.68\%$
test_getitem[list] 0.1673ms 43.2507μs 23.1210 KOps/s 21.9762 KOps/s $\textbf{\color{#35bf28}+5.21\%}$
test_setitem_dim[int] 32.0000μs 24.6831μs 40.5135 KOps/s 38.8950 KOps/s $\color{#35bf28}+4.16\%$
test_setitem_dim[slice_int] 72.5320μs 48.7382μs 20.5178 KOps/s 19.3709 KOps/s $\textbf{\color{#35bf28}+5.92\%}$
test_setitem_dim[range] 91.0810μs 66.2119μs 15.1030 KOps/s 14.0593 KOps/s $\textbf{\color{#35bf28}+7.42\%}$
test_setitem_dim[tuple] 70.4110μs 40.7646μs 24.5311 KOps/s 22.8144 KOps/s $\textbf{\color{#35bf28}+7.52\%}$
test_setitem 50.8010μs 17.5049μs 57.1269 KOps/s 55.0679 KOps/s $\color{#35bf28}+3.74\%$
test_set 58.6520μs 16.9404μs 59.0304 KOps/s 57.4464 KOps/s $\color{#35bf28}+2.76\%$
test_set_shared 0.5509ms 0.1971ms 5.0733 KOps/s 4.9137 KOps/s $\color{#35bf28}+3.25\%$
test_update 0.2084ms 21.5982μs 46.3002 KOps/s 46.0394 KOps/s $\color{#35bf28}+0.57\%$
test_update_nested 66.6010μs 33.2807μs 30.0475 KOps/s 29.2055 KOps/s $\color{#35bf28}+2.88\%$
test_update__nested 0.5254ms 33.2282μs 30.0949 KOps/s 29.7318 KOps/s $\color{#35bf28}+1.22\%$
test_set_nested 45.7210μs 18.6963μs 53.4865 KOps/s 52.4418 KOps/s $\color{#35bf28}+1.99\%$
test_set_nested_new 64.7520μs 23.5138μs 42.5282 KOps/s 42.4760 KOps/s $\color{#35bf28}+0.12\%$
test_select 88.0820μs 40.4866μs 24.6995 KOps/s 24.4343 KOps/s $\color{#35bf28}+1.09\%$
test_select_nested 0.1113ms 71.8056μs 13.9265 KOps/s 13.9176 KOps/s $\color{#35bf28}+0.06\%$
test_exclude_nested 0.1481ms 94.5456μs 10.5769 KOps/s 10.5932 KOps/s $\color{#d91a1a}-0.15\%$
test_empty[True] 0.4879ms 0.4230ms 2.3643 KOps/s 2.3676 KOps/s $\color{#d91a1a}-0.14\%$
test_empty[False] 7.1428μs 1.2982μs 770.3074 KOps/s 775.4861 KOps/s $\color{#d91a1a}-0.67\%$
test_to 94.2910μs 63.9222μs 15.6440 KOps/s 15.0012 KOps/s $\color{#35bf28}+4.28\%$
test_to_nonblocking 0.1016ms 58.4414μs 17.1112 KOps/s 16.9827 KOps/s $\color{#35bf28}+0.76\%$
test_unbind_speed 0.4077ms 0.3196ms 3.1291 KOps/s 3.1211 KOps/s $\color{#35bf28}+0.26\%$
test_unbind_speed_stack0 0.3679ms 0.3164ms 3.1605 KOps/s 3.1385 KOps/s $\color{#35bf28}+0.70\%$
test_unbind_speed_stack1 97.6764ms 0.9271ms 1.0787 KOps/s 1.1602 KOps/s $\textbf{\color{#d91a1a}-7.02\%}$
test_split 1.2111ms 1.1269ms 887.4107 Ops/s 753.1893 Ops/s $\textbf{\color{#35bf28}+17.82\%}$
test_chunk 97.7315ms 1.2103ms 826.2380 Ops/s 913.7125 Ops/s $\textbf{\color{#d91a1a}-9.57\%}$
test_consolidate[False-None] 3.9264ms 3.8163ms 262.0324 Ops/s 236.1586 Ops/s $\textbf{\color{#35bf28}+10.96\%}$
test_consolidate[default-None] 2.3139ms 2.1358ms 468.2105 Ops/s 456.6591 Ops/s $\color{#35bf28}+2.53\%$
test_consolidate[reduce-overhead-None] 2.2357ms 2.1263ms 470.3037 Ops/s 453.3014 Ops/s $\color{#35bf28}+3.75\%$
test_consolidate_njt[False-None] 8.7553ms 8.5199ms 117.3727 Ops/s 119.3612 Ops/s $\color{#d91a1a}-1.67\%$
test_to[False-False-None] 2.1268ms 2.0453ms 488.9338 Ops/s 489.3355 Ops/s $\color{#d91a1a}-0.08\%$
test_to[True-False-None] 10.6503ms 1.9290ms 518.4079 Ops/s 531.0416 Ops/s $\color{#d91a1a}-2.38\%$
test_to[within-False-None] 0.1875s 6.9435ms 144.0191 Ops/s 174.3016 Ops/s $\textbf{\color{#d91a1a}-17.37\%}$
test_to[True-default-None] 7.2153ms 6.7829ms 147.4286 Ops/s 145.5964 Ops/s $\color{#35bf28}+1.26\%$
test_to_njt[False-False-None] 8.5184ms 8.2782ms 120.7998 Ops/s 121.2399 Ops/s $\color{#d91a1a}-0.36\%$
test_to_njt[True-False-None] 7.2295ms 7.0016ms 142.8247 Ops/s 142.4472 Ops/s $\color{#35bf28}+0.26\%$
test_to_njt[within-False-None] 15.9533ms 15.6862ms 63.7503 Ops/s 64.6966 Ops/s $\color{#d91a1a}-1.46\%$
test_creation[device0] 0.4020ms 0.1078ms 9.2726 KOps/s 9.2764 KOps/s $\color{#d91a1a}-0.04\%$
test_creation_from_tensor 0.3994ms 0.1089ms 9.1840 KOps/s 9.1499 KOps/s $\color{#35bf28}+0.37\%$
test_add_one[memmap_tensor0] 0.3759ms 6.8484μs 146.0186 KOps/s 146.2241 KOps/s $\color{#d91a1a}-0.14\%$
test_contiguous[memmap_tensor0] 29.4700μs 0.7458μs 1.3409 MOps/s 1.8680 MOps/s $\textbf{\color{#d91a1a}-28.22\%}$
test_stack[memmap_tensor0] 36.2810μs 4.7960μs 208.5084 KOps/s 208.7446 KOps/s $\color{#d91a1a}-0.11\%$
test_memmaptd_index 1.0738ms 0.2939ms 3.4020 KOps/s 3.3912 KOps/s $\color{#35bf28}+0.32\%$
test_memmaptd_index_astensor 0.5349ms 0.3840ms 2.6045 KOps/s 2.6238 KOps/s $\color{#d91a1a}-0.74\%$
test_memmaptd_index_op 0.7984ms 0.6248ms 1.6006 KOps/s 1.6138 KOps/s $\color{#d91a1a}-0.82\%$
test_serialize_model 0.1329s 0.1314s 7.6126 Ops/s 7.6032 Ops/s $\color{#35bf28}+0.12\%$
test_serialize_model_pickle 1.3476s 1.2136s 0.8240 Ops/s 0.8348 Ops/s $\color{#d91a1a}-1.30\%$
test_serialize_weights 0.3000s 0.1547s 6.4646 Ops/s 7.6302 Ops/s $\textbf{\color{#d91a1a}-15.28\%}$
test_serialize_weights_returnearly 0.3758s 67.2500ms 14.8699 Ops/s 11.5462 Ops/s $\textbf{\color{#35bf28}+28.79\%}$
test_serialize_weights_pickle 1.3465s 1.1845s 0.8443 Ops/s 0.8214 Ops/s $\color{#35bf28}+2.78\%$
test_reshape_pytree 0.3588ms 33.4484μs 29.8968 KOps/s 29.6649 KOps/s $\color{#35bf28}+0.78\%$
test_reshape_td 68.7120μs 41.4637μs 24.1175 KOps/s 24.6468 KOps/s $\color{#d91a1a}-2.15\%$
test_view_pytree 0.2216ms 32.7873μs 30.4996 KOps/s 30.7236 KOps/s $\color{#d91a1a}-0.73\%$
test_view_td 0.2110ms 49.8341μs 20.0666 KOps/s 21.0846 KOps/s $\color{#d91a1a}-4.83\%$
test_unbind_pytree 0.2372ms 37.2038μs 26.8789 KOps/s 26.4704 KOps/s $\color{#35bf28}+1.54\%$
test_unbind_td 0.1195ms 49.8881μs 20.0449 KOps/s 20.3720 KOps/s $\color{#d91a1a}-1.61\%$
test_split_pytree 0.2470ms 44.9429μs 22.2504 KOps/s 23.0234 KOps/s $\color{#d91a1a}-3.36\%$
test_split_td 0.1862ms 66.2482μs 15.0947 KOps/s 15.9221 KOps/s $\textbf{\color{#d91a1a}-5.20\%}$
test_add_pytree 0.1922ms 42.6903μs 23.4245 KOps/s 23.1217 KOps/s $\color{#35bf28}+1.31\%$
test_add_td 75.0410μs 52.7247μs 18.9665 KOps/s 18.3191 KOps/s $\color{#35bf28}+3.53\%$
test_compile_add_one_nested[tensordict-compile] 0.2444ms 0.1369ms 7.3037 KOps/s 6.8054 KOps/s $\textbf{\color{#35bf28}+7.32\%}$
test_compile_add_one_nested[tensordict-eager] 0.2907ms 0.1832ms 5.4585 KOps/s 5.4049 KOps/s $\color{#35bf28}+0.99\%$
test_compile_add_one_nested[pytree-compile] 0.1703ms 0.1068ms 9.3656 KOps/s 9.1266 KOps/s $\color{#35bf28}+2.62\%$
test_compile_add_one_nested[pytree-eager] 0.3616ms 0.1803ms 5.5464 KOps/s 5.5386 KOps/s $\color{#35bf28}+0.14\%$
test_compile_copy_nested[tensordict-compile] 67.7810μs 28.5507μs 35.0254 KOps/s 32.0117 KOps/s $\textbf{\color{#35bf28}+9.41\%}$
test_compile_copy_nested[tensordict-eager] 79.0710μs 48.9982μs 20.4089 KOps/s 20.3373 KOps/s $\color{#35bf28}+0.35\%$
test_compile_copy_nested[pytree-compile] 89.9720μs 13.4586μs 74.3017 KOps/s 74.6557 KOps/s $\color{#d91a1a}-0.47\%$
test_compile_copy_nested[pytree-eager] 0.4120ms 72.5493μs 13.7837 KOps/s 13.8310 KOps/s $\color{#d91a1a}-0.34\%$
test_compile_add_one_flat[tensordict-compile] 0.2518ms 0.1660ms 6.0243 KOps/s 5.9445 KOps/s $\color{#35bf28}+1.34\%$
test_compile_add_one_flat[tensordict-eager] 0.3520ms 0.2543ms 3.9330 KOps/s 3.8116 KOps/s $\color{#35bf28}+3.19\%$
test_compile_add_one_flat[tensorclass-compile] 0.1905ms 0.1162ms 8.6026 KOps/s 8.6776 KOps/s $\color{#d91a1a}-0.87\%$
test_compile_add_one_flat[tensorclass-eager] 0.1353ms 68.0806μs 14.6885 KOps/s 14.4919 KOps/s $\color{#35bf28}+1.36\%$
test_compile_add_one_flat[pytree-compile] 0.2036ms 0.1574ms 6.3550 KOps/s 6.2449 KOps/s $\color{#35bf28}+1.76\%$
test_compile_add_one_flat[pytree-eager] 0.7477ms 0.5275ms 1.8958 KOps/s 1.8640 KOps/s $\color{#35bf28}+1.71\%$
test_compile_add_self_flat[tensordict-eager] 0.3924ms 0.3092ms 3.2346 KOps/s 3.1975 KOps/s $\color{#35bf28}+1.16\%$
test_compile_add_self_flat[tensordict-compile] 0.2463ms 0.1631ms 6.1328 KOps/s 5.9228 KOps/s $\color{#35bf28}+3.55\%$
test_compile_add_self_flat[tensorclass-eager] 0.1383ms 82.9583μs 12.0542 KOps/s 11.9924 KOps/s $\color{#35bf28}+0.52\%$
test_compile_add_self_flat[tensorclass-compile] 0.1609ms 0.1138ms 8.7892 KOps/s 8.4991 KOps/s $\color{#35bf28}+3.41\%$
test_compile_add_self_flat[pytree-eager] 0.6309ms 0.4460ms 2.2421 KOps/s 2.2141 KOps/s $\color{#35bf28}+1.26\%$
test_compile_add_self_flat[pytree-compile] 0.2058ms 0.1570ms 6.3682 KOps/s 6.1959 KOps/s $\color{#35bf28}+2.78\%$
test_compile_copy_flat[tensordict-compile] 93.2110μs 22.5904μs 44.2665 KOps/s 42.2236 KOps/s $\color{#35bf28}+4.84\%$
test_compile_copy_flat[tensordict-eager] 70.9710μs 40.4370μs 24.7298 KOps/s 25.2396 KOps/s $\color{#d91a1a}-2.02\%$
test_compile_copy_flat[pytree-compile] 0.1750ms 19.0418μs 52.5159 KOps/s 51.1299 KOps/s $\color{#35bf28}+2.71\%$
test_compile_copy_flat[pytree-eager] 0.3659ms 65.9911μs 15.1535 KOps/s 15.0077 KOps/s $\color{#35bf28}+0.97\%$
test_compile_assign_and_add[tensordict-compile] 1.9355ms 0.5186ms 1.9284 KOps/s 1.9092 KOps/s $\color{#35bf28}+1.00\%$
test_compile_assign_and_add[tensordict-eager] 3.2968ms 3.1889ms 313.5916 Ops/s 307.8191 Ops/s $\color{#35bf28}+1.88\%$
test_compile_assign_and_add[pytree-compile] 1.9071ms 0.5071ms 1.9721 KOps/s 1.9401 KOps/s $\color{#35bf28}+1.65\%$
test_compile_assign_and_add[pytree-eager] 2.9439ms 2.8274ms 353.6855 Ops/s 351.5634 Ops/s $\color{#35bf28}+0.60\%$
test_compile_indexing[tensor-tensordict-compile] 0.2133ms 0.1301ms 7.6848 KOps/s 7.1970 KOps/s $\textbf{\color{#35bf28}+6.78\%}$
test_compile_indexing[tensor-tensordict-eager] 0.2605ms 93.2800μs 10.7204 KOps/s 10.7552 KOps/s $\color{#d91a1a}-0.32\%$
test_compile_indexing[tensor-tensorclass-compile] 0.2555ms 0.1223ms 8.1765 KOps/s 7.8162 KOps/s $\color{#35bf28}+4.61\%$
test_compile_indexing[tensor-tensorclass-eager] 0.3198ms 83.2378μs 12.0138 KOps/s 12.1181 KOps/s $\color{#d91a1a}-0.86\%$
test_compile_indexing[tensor-pytree-compile] 0.1840ms 0.1289ms 7.7562 KOps/s 7.7046 KOps/s $\color{#35bf28}+0.67\%$
test_compile_indexing[tensor-pytree-eager] 0.2835ms 83.7399μs 11.9417 KOps/s 12.1169 KOps/s $\color{#d91a1a}-1.45\%$
test_compile_indexing[slice-tensordict-compile] 0.1900ms 0.1170ms 8.5460 KOps/s 8.6302 KOps/s $\color{#d91a1a}-0.98\%$
test_compile_indexing[slice-tensordict-eager] 0.1909ms 26.1497μs 38.2414 KOps/s 37.3306 KOps/s $\color{#35bf28}+2.44\%$
test_compile_indexing[slice-tensorclass-compile] 0.1612ms 0.1111ms 9.0017 KOps/s 8.9577 KOps/s $\color{#35bf28}+0.49\%$
test_compile_indexing[slice-tensorclass-eager] 0.2136ms 23.5762μs 42.4156 KOps/s 44.7383 KOps/s $\textbf{\color{#d91a1a}-5.19\%}$
test_compile_indexing[slice-pytree-compile] 0.1655ms 0.1148ms 8.7137 KOps/s 8.9747 KOps/s $\color{#d91a1a}-2.91\%$
test_compile_indexing[slice-pytree-eager] 0.2285ms 23.4789μs 42.5915 KOps/s 43.9455 KOps/s $\color{#d91a1a}-3.08\%$
test_compile_indexing[int-tensordict-compile] 0.1799ms 0.1221ms 8.1880 KOps/s 8.5724 KOps/s $\color{#d91a1a}-4.48\%$
test_compile_indexing[int-tensordict-eager] 0.2533ms 26.3790μs 37.9089 KOps/s 38.1829 KOps/s $\color{#d91a1a}-0.72\%$
test_compile_indexing[int-tensorclass-compile] 0.1849ms 0.1149ms 8.7006 KOps/s 8.9444 KOps/s $\color{#d91a1a}-2.73\%$
test_compile_indexing[int-tensorclass-eager] 0.2243ms 23.2628μs 42.9870 KOps/s 43.7747 KOps/s $\color{#d91a1a}-1.80\%$
test_compile_indexing[int-pytree-compile] 0.1855ms 0.1163ms 8.5957 KOps/s 8.9115 KOps/s $\color{#d91a1a}-3.54\%$
test_compile_indexing[int-pytree-eager] 0.2209ms 22.5331μs 44.3792 KOps/s 44.0236 KOps/s $\color{#35bf28}+0.81\%$
test_mod_add[eager] 0.1162ms 46.7765μs 21.3782 KOps/s 20.8609 KOps/s $\color{#35bf28}+2.48\%$
test_mod_add[compile] 0.2519ms 93.4864μs 10.6967 KOps/s 10.5768 KOps/s $\color{#35bf28}+1.13\%$
test_mod_add[compile-overhead] 0.3249ms 0.1817ms 5.5040 KOps/s 5.3204 KOps/s $\color{#35bf28}+3.45\%$
test_mod_wrap[eager] 0.3635ms 0.2877ms 3.4757 KOps/s 3.4501 KOps/s $\color{#35bf28}+0.74\%$
test_mod_wrap[compile] 0.3959ms 0.3330ms 3.0031 KOps/s 2.9917 KOps/s $\color{#35bf28}+0.38\%$
test_mod_wrap[compile-overhead] 7.6171ms 4.1392ms 241.5897 Ops/s 239.0032 Ops/s $\color{#35bf28}+1.08\%$
test_mod_wrap_and_backward[eager] 1.5918ms 1.4924ms 670.0471 Ops/s 670.8177 Ops/s $\color{#d91a1a}-0.11\%$
test_mod_wrap_and_backward[compile] 1.6161ms 1.4701ms 680.2252 Ops/s 681.9966 Ops/s $\color{#d91a1a}-0.26\%$
test_mod_wrap_and_backward[compile-overhead] 1.4922ms 0.9900ms 1.0101 KOps/s 1.0051 KOps/s $\color{#35bf28}+0.50\%$
test_seq_add[eager] 0.5697ms 0.1493ms 6.6979 KOps/s 6.5960 KOps/s $\color{#35bf28}+1.54\%$
test_seq_add[compile] 0.5862ms 0.1038ms 9.6357 KOps/s 9.2550 KOps/s $\color{#35bf28}+4.11\%$
test_seq_add[compile-overhead] 0.2277ms 0.1398ms 7.1512 KOps/s 6.9137 KOps/s $\color{#35bf28}+3.44\%$
test_seq_wrap[eager] 0.9584ms 0.5050ms 1.9801 KOps/s 1.9166 KOps/s $\color{#35bf28}+3.31\%$
test_seq_wrap[compile] 0.8278ms 0.3510ms 2.8491 KOps/s 2.7590 KOps/s $\color{#35bf28}+3.27\%$
test_seq_wrap[compile-overhead] 0.6805ms 0.2503ms 3.9949 KOps/s 4.0405 KOps/s $\color{#d91a1a}-1.13\%$
test_func_call_runtime[False-eager] 1.2637ms 0.8383ms 1.1930 KOps/s 1.1906 KOps/s $\color{#35bf28}+0.19\%$
test_func_call_runtime[False-compile] 1.3791ms 0.8887ms 1.1253 KOps/s 1.1264 KOps/s $\color{#d91a1a}-0.10\%$
test_func_call_runtime[False-compile-overhead] 0.8365ms 0.4036ms 2.4780 KOps/s 2.4528 KOps/s $\color{#35bf28}+1.03\%$
test_func_call_runtime[True-eager] 1.4927ms 1.0594ms 943.8928 Ops/s 928.0289 Ops/s $\color{#35bf28}+1.71\%$
test_func_call_runtime[True-compile] 1.4373ms 0.9140ms 1.0941 KOps/s 1.0886 KOps/s $\color{#35bf28}+0.50\%$
test_func_call_runtime[True-compile-overhead] 0.5026ms 0.4246ms 2.3552 KOps/s 2.3454 KOps/s $\color{#35bf28}+0.42\%$
test_func_call_cm_runtime[False-eager] 0.8894ms 0.8382ms 1.1930 KOps/s 1.1851 KOps/s $\color{#35bf28}+0.67\%$
test_func_call_cm_runtime[False-compile] 1.3929ms 0.8955ms 1.1167 KOps/s 1.1136 KOps/s $\color{#35bf28}+0.28\%$
test_func_call_cm_runtime[False-compile-overhead] 0.4600ms 0.4044ms 2.4731 KOps/s 2.4421 KOps/s $\color{#35bf28}+1.27\%$
test_func_call_cm_runtime[True-eager] 1.2852ms 1.1916ms 839.1925 Ops/s 825.2463 Ops/s $\color{#35bf28}+1.69\%$
test_func_call_cm_runtime[True-compile] 1.4496ms 0.9465ms 1.0566 KOps/s 1.0539 KOps/s $\color{#35bf28}+0.25\%$
test_func_call_cm_runtime[True-compile-overhead] 0.5215ms 0.4598ms 2.1746 KOps/s 2.1947 KOps/s $\color{#d91a1a}-0.92\%$
test_vmap_func_call_cm_runtime[eager] 2.7517ms 2.2714ms 440.2656 Ops/s 437.3023 Ops/s $\color{#35bf28}+0.68\%$
test_vmap_func_call_cm_runtime[compile] 1.4665ms 0.9634ms 1.0380 KOps/s 1.0387 KOps/s $\color{#d91a1a}-0.07\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.6056ms 0.4534ms 2.2056 KOps/s 2.1764 KOps/s $\color{#35bf28}+1.34\%$
test_distributed 2.9651ms 0.1605ms 6.2315 KOps/s 5.6873 KOps/s $\textbf{\color{#35bf28}+9.57\%}$
test_tdmodule 53.0810μs 27.0558μs 36.9607 KOps/s 37.1922 KOps/s $\color{#d91a1a}-0.62\%$
test_tdmodule_dispatch 69.8320μs 46.3037μs 21.5965 KOps/s 21.2724 KOps/s $\color{#35bf28}+1.52\%$
test_tdseq 37.9310μs 25.4238μs 39.3332 KOps/s 38.5076 KOps/s $\color{#35bf28}+2.14\%$
test_tdseq_dispatch 70.9110μs 47.8680μs 20.8908 KOps/s 20.4978 KOps/s $\color{#35bf28}+1.92\%$
test_instantiation_functorch 2.0584ms 1.9704ms 507.5174 Ops/s 500.1177 Ops/s $\color{#35bf28}+1.48\%$
test_exec_functorch 0.2554ms 0.1779ms 5.6200 KOps/s 5.4593 KOps/s $\color{#35bf28}+2.94\%$
test_exec_functional_call 0.2152ms 0.1585ms 6.3072 KOps/s 6.1036 KOps/s $\color{#35bf28}+3.34\%$
test_exec_td_decorator 0.4334ms 0.2288ms 4.3704 KOps/s 4.3011 KOps/s $\color{#35bf28}+1.61\%$
test_vmap_mlp_speed_decorator[True-True] 0.9196ms 0.7599ms 1.3160 KOps/s 1.3202 KOps/s $\color{#d91a1a}-0.32\%$
test_vmap_mlp_speed_decorator[True-False] 0.9136ms 0.7598ms 1.3161 KOps/s 1.3236 KOps/s $\color{#d91a1a}-0.57\%$
test_vmap_mlp_speed_decorator[False-True] 0.7958ms 0.6537ms 1.5296 KOps/s 1.5361 KOps/s $\color{#d91a1a}-0.42\%$
test_vmap_mlp_speed_decorator[False-False] 0.7957ms 0.6585ms 1.5185 KOps/s 1.5322 KOps/s $\color{#d91a1a}-0.90\%$
test_vmap_transformer_speed_decorator[True-True] 20.8310ms 20.3862ms 49.0528 Ops/s 49.6585 Ops/s $\color{#d91a1a}-1.22\%$
test_vmap_transformer_speed_decorator[True-False] 21.3906ms 20.4422ms 48.9185 Ops/s 49.5959 Ops/s $\color{#d91a1a}-1.37\%$
test_vmap_transformer_speed_decorator[False-True] 20.4324ms 20.0127ms 49.9684 Ops/s 50.0861 Ops/s $\color{#d91a1a}-0.24\%$
test_vmap_transformer_speed_decorator[False-False] 20.7701ms 20.3189ms 49.2153 Ops/s 50.0464 Ops/s $\color{#d91a1a}-1.66\%$
test_to_module_speed[True] 1.5087ms 1.4023ms 713.1219 Ops/s 708.0935 Ops/s $\color{#35bf28}+0.71\%$
test_to_module_speed[False] 1.4829ms 1.3948ms 716.9675 Ops/s 721.5763 Ops/s $\color{#d91a1a}-0.64\%$
test_tc_init 80.1820μs 50.1634μs 19.9349 KOps/s 19.8630 KOps/s $\color{#35bf28}+0.36\%$
test_tc_init_tensor_only 44.0200μs 14.2391μs 70.2292 KOps/s 69.1920 KOps/s $\color{#35bf28}+1.50\%$
test_tc_init_nested 0.1434ms 99.8577μs 10.0142 KOps/s 9.9495 KOps/s $\color{#35bf28}+0.65\%$
test_tc_first_layer_tensor 15.5200μs 1.6868μs 592.8333 KOps/s 595.1445 KOps/s $\color{#d91a1a}-0.39\%$
test_tc_first_layer_tensor_only 3.9401μs 0.6495μs 1.5397 MOps/s 1.5397 MOps/s $-0.00\%$
test_tc_first_layer_tensor_set 28.2500μs 3.9952μs 250.3032 KOps/s 249.8567 KOps/s $\color{#35bf28}+0.18\%$
test_tc_first_layer_tensor_only_set 20.3005μs 2.8097μs 355.9104 KOps/s 335.6244 KOps/s $\textbf{\color{#35bf28}+6.04\%}$
test_tc_first_layer_nontensor 30.1100μs 5.6271μs 177.7105 KOps/s 177.7911 KOps/s $\color{#d91a1a}-0.05\%$
test_tc_second_layer_tensor 24.6510μs 4.0190μs 248.8178 KOps/s 244.4112 KOps/s $\color{#35bf28}+1.80\%$
test_tc_second_layer_nontensor 41.7410μs 7.9487μs 125.8070 KOps/s 126.3308 KOps/s $\color{#d91a1a}-0.41\%$
test_unbind 9.6715ms 9.2982ms 107.5476 Ops/s 107.9356 Ops/s $\color{#d91a1a}-0.36\%$
test_full_like 5.1906ms 4.3852ms 228.0398 Ops/s 59.1510 Ops/s $\textbf{\color{#35bf28}+285.52\%}$
test_zeros_like 17.1643ms 16.7207ms 59.8060 Ops/s 59.5085 Ops/s $\color{#35bf28}+0.50\%$
test_ones_like 17.0731ms 16.7462ms 59.7149 Ops/s 59.2500 Ops/s $\color{#35bf28}+0.78\%$
test_clone 18.0288ms 17.7109ms 56.4625 Ops/s 56.4131 Ops/s $\color{#35bf28}+0.09\%$
test_squeeze 62.3710μs 13.8269μs 72.3228 KOps/s 72.4047 KOps/s $\color{#d91a1a}-0.11\%$
test_unsqueeze 0.1425ms 0.1041ms 9.6107 KOps/s 9.5066 KOps/s $\color{#35bf28}+1.09\%$
test_split 0.4562ms 0.1801ms 5.5538 KOps/s 5.5951 KOps/s $\color{#d91a1a}-0.74\%$
test_permute 0.3007ms 0.2068ms 4.8357 KOps/s 4.9969 KOps/s $\color{#d91a1a}-3.23\%$
test_stack 53.5512ms 51.4144ms 19.4498 Ops/s 19.4099 Ops/s $\color{#35bf28}+0.21\%$
test_cat 51.6182ms 51.2003ms 19.5311 Ops/s 19.4354 Ops/s $\color{#35bf28}+0.49\%$

@github-actions
Copy link

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

Total Benchmarks: 233. Improved: $\large\color{#35bf28}19$. Worsened: $\large\color{#d91a1a}9$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 40.6720μs 15.1548μs 65.9858 KOps/s 66.5767 KOps/s $\color{#d91a1a}-0.89\%$
test_plain_set_stack_nested 40.1220μs 15.2782μs 65.4526 KOps/s 65.8654 KOps/s $\color{#d91a1a}-0.63\%$
test_plain_set_nested_inplace 43.5420μs 16.4935μs 60.6300 KOps/s 61.0748 KOps/s $\color{#d91a1a}-0.73\%$
test_plain_set_stack_nested_inplace 46.4120μs 16.5813μs 60.3090 KOps/s 60.8169 KOps/s $\color{#d91a1a}-0.84\%$
test_items 29.7510μs 6.1703μs 162.0678 KOps/s 163.7082 KOps/s $\color{#d91a1a}-1.00\%$
test_items_nested 0.6419ms 0.5416ms 1.8463 KOps/s 1.8793 KOps/s $\color{#d91a1a}-1.76\%$
test_items_nested_locked 0.5938ms 0.5437ms 1.8393 KOps/s 1.8611 KOps/s $\color{#d91a1a}-1.17\%$
test_items_nested_leaf 0.1340ms 96.2654μs 10.3879 KOps/s 10.4763 KOps/s $\color{#d91a1a}-0.84\%$
test_items_stack_nested 0.5817ms 0.5365ms 1.8639 KOps/s 1.8662 KOps/s $\color{#d91a1a}-0.13\%$
test_items_stack_nested_leaf 0.1327ms 96.7715μs 10.3336 KOps/s 10.4488 KOps/s $\color{#d91a1a}-1.10\%$
test_items_stack_nested_locked 0.5931ms 0.5393ms 1.8543 KOps/s 1.8616 KOps/s $\color{#d91a1a}-0.39\%$
test_keys 30.5510μs 4.2553μs 235.0012 KOps/s 235.1747 KOps/s $\color{#d91a1a}-0.07\%$
test_keys_nested 0.1570ms 0.1201ms 8.3262 KOps/s 8.3875 KOps/s $\color{#d91a1a}-0.73\%$
test_keys_nested_locked 1.9851ms 0.1298ms 7.7052 KOps/s 7.8291 KOps/s $\color{#d91a1a}-1.58\%$
test_keys_nested_leaf 0.1482ms 0.1104ms 9.0578 KOps/s 9.1150 KOps/s $\color{#d91a1a}-0.63\%$
test_keys_stack_nested 0.1536ms 0.1197ms 8.3560 KOps/s 8.3891 KOps/s $\color{#d91a1a}-0.40\%$
test_keys_stack_nested_leaf 0.1459ms 0.1096ms 9.1250 KOps/s 9.1298 KOps/s $\color{#d91a1a}-0.05\%$
test_keys_stack_nested_locked 0.1698ms 0.1283ms 7.7926 KOps/s 7.7311 KOps/s $\color{#35bf28}+0.80\%$
test_values 6.6822μs 1.0412μs 960.4586 KOps/s 981.1411 KOps/s $\color{#d91a1a}-2.11\%$
test_values_nested 83.4940μs 47.8657μs 20.8918 KOps/s 20.7586 KOps/s $\color{#35bf28}+0.64\%$
test_values_nested_locked 78.2640μs 50.5562μs 19.7800 KOps/s 19.6000 KOps/s $\color{#35bf28}+0.92\%$
test_values_nested_leaf 86.6140μs 54.0071μs 18.5161 KOps/s 18.3396 KOps/s $\color{#35bf28}+0.96\%$
test_values_stack_nested 76.1330μs 47.8215μs 20.9111 KOps/s 20.7002 KOps/s $\color{#35bf28}+1.02\%$
test_values_stack_nested_leaf 0.1132ms 53.9489μs 18.5361 KOps/s 18.4496 KOps/s $\color{#35bf28}+0.47\%$
test_values_stack_nested_locked 0.1069ms 49.5270μs 20.1910 KOps/s 19.6569 KOps/s $\color{#35bf28}+2.72\%$
test_membership 4.9252μs 0.8340μs 1.1990 MOps/s 1.1657 MOps/s $\color{#35bf28}+2.85\%$
test_membership_nested 28.8320μs 3.1530μs 317.1542 KOps/s 314.1117 KOps/s $\color{#35bf28}+0.97\%$
test_membership_nested_leaf 30.6210μs 3.1505μs 317.4065 KOps/s 314.9331 KOps/s $\color{#35bf28}+0.79\%$
test_membership_stacked_nested 38.8720μs 3.1512μs 317.3443 KOps/s 314.4920 KOps/s $\color{#35bf28}+0.91\%$
test_membership_stacked_nested_leaf 19.9210μs 3.1504μs 317.4203 KOps/s 314.4273 KOps/s $\color{#35bf28}+0.95\%$
test_membership_nested_last 32.7020μs 4.6173μs 216.5788 KOps/s 217.7331 KOps/s $\color{#d91a1a}-0.53\%$
test_membership_nested_leaf_last 33.1410μs 4.6107μs 216.8866 KOps/s 220.1222 KOps/s $\color{#d91a1a}-1.47\%$
test_membership_stacked_nested_last 42.3120μs 4.5856μs 218.0724 KOps/s 219.0881 KOps/s $\color{#d91a1a}-0.46\%$
test_membership_stacked_nested_leaf_last 31.5410μs 4.5931μs 217.7184 KOps/s 219.6132 KOps/s $\color{#d91a1a}-0.86\%$
test_nested_getleaf 51.4620μs 21.5858μs 46.3268 KOps/s 46.1577 KOps/s $\color{#35bf28}+0.37\%$
test_nested_get 86.4640μs 20.4827μs 48.8217 KOps/s 49.0213 KOps/s $\color{#d91a1a}-0.41\%$
test_stacked_getleaf 51.1930μs 21.5319μs 46.4428 KOps/s 46.6303 KOps/s $\color{#d91a1a}-0.40\%$
test_stacked_get 40.6320μs 20.5224μs 48.7274 KOps/s 48.4057 KOps/s $\color{#35bf28}+0.66\%$
test_nested_getitemleaf 51.7120μs 21.9044μs 45.6530 KOps/s 45.8888 KOps/s $\color{#d91a1a}-0.51\%$
test_nested_getitem 46.2420μs 20.8238μs 48.0220 KOps/s 48.3203 KOps/s $\color{#d91a1a}-0.62\%$
test_stacked_getitemleaf 54.2530μs 22.0621μs 45.3267 KOps/s 45.5479 KOps/s $\color{#d91a1a}-0.49\%$
test_stacked_getitem 40.1620μs 20.8808μs 47.8908 KOps/s 47.9419 KOps/s $\color{#d91a1a}-0.11\%$
test_lock_nested 0.5652ms 0.4733ms 2.1128 KOps/s 2.1288 KOps/s $\color{#d91a1a}-0.75\%$
test_lock_stack_nested 0.5122ms 0.4745ms 2.1075 KOps/s 2.1043 KOps/s $\color{#35bf28}+0.15\%$
test_unlock_nested 0.4734ms 0.3840ms 2.6041 KOps/s 2.6353 KOps/s $\color{#d91a1a}-1.19\%$
test_unlock_stack_nested 0.4194ms 0.3803ms 2.6297 KOps/s 2.6362 KOps/s $\color{#d91a1a}-0.25\%$
test_flatten_speed 0.1682ms 0.1211ms 8.2598 KOps/s 8.1663 KOps/s $\color{#35bf28}+1.14\%$
test_unflatten_speed 0.6762ms 0.5865ms 1.7051 KOps/s 1.7065 KOps/s $\color{#d91a1a}-0.08\%$
test_common_ops 0.8622ms 0.7324ms 1.3653 KOps/s 1.3466 KOps/s $\color{#35bf28}+1.39\%$
test_creation 68.4330μs 2.7648μs 361.6909 KOps/s 373.2897 KOps/s $\color{#d91a1a}-3.11\%$
test_creation_empty 29.7210μs 9.0357μs 110.6727 KOps/s 109.9042 KOps/s $\color{#35bf28}+0.70\%$
test_creation_nested_1 40.2520μs 12.1104μs 82.5734 KOps/s 82.4181 KOps/s $\color{#35bf28}+0.19\%$
test_creation_nested_2 44.3920μs 15.9433μs 62.7222 KOps/s 61.8017 KOps/s $\color{#35bf28}+1.49\%$
test_clone 48.6120μs 13.0261μs 76.7690 KOps/s 77.0295 KOps/s $\color{#d91a1a}-0.34\%$
test_getitem[int] 1.2244ms 14.5121μs 68.9080 KOps/s 69.8811 KOps/s $\color{#d91a1a}-1.39\%$
test_getitem[slice_int] 0.1485ms 29.1146μs 34.3470 KOps/s 34.5945 KOps/s $\color{#d91a1a}-0.72\%$
test_getitem[range] 0.1749ms 52.6036μs 19.0101 KOps/s 19.7545 KOps/s $\color{#d91a1a}-3.77\%$
test_getitem[tuple] 0.1443ms 24.6521μs 40.5645 KOps/s 40.8797 KOps/s $\color{#d91a1a}-0.77\%$
test_getitem[list] 0.1642ms 43.9405μs 22.7580 KOps/s 22.5111 KOps/s $\color{#35bf28}+1.10\%$
test_setitem_dim[int] 47.5320μs 25.4172μs 39.3434 KOps/s 40.1161 KOps/s $\color{#d91a1a}-1.93\%$
test_setitem_dim[slice_int] 79.3340μs 51.3510μs 19.4738 KOps/s 20.2518 KOps/s $\color{#d91a1a}-3.84\%$
test_setitem_dim[range] 97.8850μs 72.4103μs 13.8102 KOps/s 14.3538 KOps/s $\color{#d91a1a}-3.79\%$
test_setitem_dim[tuple] 74.6040μs 41.7936μs 23.9271 KOps/s 23.6161 KOps/s $\color{#35bf28}+1.32\%$
test_setitem 49.5530μs 18.8402μs 53.0780 KOps/s 55.0070 KOps/s $\color{#d91a1a}-3.51\%$
test_set 63.8230μs 16.7783μs 59.6010 KOps/s 58.2736 KOps/s $\color{#35bf28}+2.28\%$
test_set_shared 0.4965ms 0.2018ms 4.9559 KOps/s 4.9367 KOps/s $\color{#35bf28}+0.39\%$
test_update 0.2080ms 22.8325μs 43.7973 KOps/s 44.6277 KOps/s $\color{#d91a1a}-1.86\%$
test_update_nested 67.8040μs 34.9683μs 28.5973 KOps/s 29.3282 KOps/s $\color{#d91a1a}-2.49\%$
test_update__nested 0.4577ms 33.6734μs 29.6971 KOps/s 29.6368 KOps/s $\color{#35bf28}+0.20\%$
test_set_nested 52.6220μs 19.6170μs 50.9762 KOps/s 51.9398 KOps/s $\color{#d91a1a}-1.86\%$
test_set_nested_new 73.0640μs 24.6738μs 40.5289 KOps/s 41.2491 KOps/s $\color{#d91a1a}-1.75\%$
test_select 75.1030μs 42.0730μs 23.7682 KOps/s 24.2384 KOps/s $\color{#d91a1a}-1.94\%$
test_select_nested 0.1046ms 74.6277μs 13.3998 KOps/s 13.2937 KOps/s $\color{#35bf28}+0.80\%$
test_exclude_nested 0.1296ms 98.7355μs 10.1281 KOps/s 10.0635 KOps/s $\color{#35bf28}+0.64\%$
test_empty[True] 0.4769ms 0.4346ms 2.3010 KOps/s 2.2810 KOps/s $\color{#35bf28}+0.88\%$
test_empty[False] 9.6502μs 1.3222μs 756.2972 KOps/s 736.7159 KOps/s $\color{#35bf28}+2.66\%$
test_to 96.2540μs 65.8419μs 15.1879 KOps/s 15.0521 KOps/s $\color{#35bf28}+0.90\%$
test_to_nonblocking 97.7540μs 62.2038μs 16.0762 KOps/s 17.1339 KOps/s $\textbf{\color{#d91a1a}-6.17\%}$
test_unbind_speed 0.3935ms 0.3261ms 3.0667 KOps/s 3.0811 KOps/s $\color{#d91a1a}-0.47\%$
test_unbind_speed_stack0 0.3913ms 0.3248ms 3.0792 KOps/s 3.1291 KOps/s $\color{#d91a1a}-1.59\%$
test_unbind_speed_stack1 96.9303ms 0.9455ms 1.0576 KOps/s 1.1488 KOps/s $\textbf{\color{#d91a1a}-7.93\%}$
test_split 1.2119ms 1.1570ms 864.2689 Ops/s 738.7414 Ops/s $\textbf{\color{#35bf28}+16.99\%}$
test_chunk 97.1512ms 1.2405ms 806.1225 Ops/s 892.0849 Ops/s $\textbf{\color{#d91a1a}-9.64\%}$
test_consolidate[False-None] 4.0396ms 3.9029ms 256.2218 Ops/s 232.5355 Ops/s $\textbf{\color{#35bf28}+10.19\%}$
test_consolidate[default-None] 2.3422ms 2.1966ms 455.2392 Ops/s 441.1371 Ops/s $\color{#35bf28}+3.20\%$
test_consolidate[reduce-overhead-None] 2.2784ms 2.1962ms 455.3226 Ops/s 438.6176 Ops/s $\color{#35bf28}+3.81\%$
test_consolidate_njt[False-None] 8.9415ms 8.7525ms 114.2530 Ops/s 113.9090 Ops/s $\color{#35bf28}+0.30\%$
test_to[False-False-None] 2.1240ms 2.0423ms 489.6504 Ops/s 499.0357 Ops/s $\color{#d91a1a}-1.88\%$
test_to[True-False-None] 2.2475ms 1.9545ms 511.6523 Ops/s 524.1441 Ops/s $\color{#d91a1a}-2.38\%$
test_to[within-False-None] 0.1859s 7.0396ms 142.0531 Ops/s 169.5666 Ops/s $\textbf{\color{#d91a1a}-16.23\%}$
test_to[True-default-None] 7.4027ms 6.9016ms 144.8940 Ops/s 141.2132 Ops/s $\color{#35bf28}+2.61\%$
test_to_njt[False-False-None] 8.7309ms 8.4469ms 118.3867 Ops/s 117.6325 Ops/s $\color{#35bf28}+0.64\%$
test_to_njt[True-False-None] 7.3871ms 7.1795ms 139.2851 Ops/s 135.0744 Ops/s $\color{#35bf28}+3.12\%$
test_to_njt[within-False-None] 16.2184ms 15.9445ms 62.7175 Ops/s 43.9169 Ops/s $\textbf{\color{#35bf28}+42.81\%}$
test_creation[device0] 0.3968ms 0.1136ms 8.8038 KOps/s 9.0799 KOps/s $\color{#d91a1a}-3.04\%$
test_creation_from_tensor 0.3873ms 0.1112ms 8.9891 KOps/s 9.0401 KOps/s $\color{#d91a1a}-0.56\%$
test_add_one[memmap_tensor0] 0.3481ms 6.7830μs 147.4284 KOps/s 156.9686 KOps/s $\textbf{\color{#d91a1a}-6.08\%}$
test_contiguous[memmap_tensor0] 15.1510μs 0.8024μs 1.2463 MOps/s 1.7331 MOps/s $\textbf{\color{#d91a1a}-28.09\%}$
test_stack[memmap_tensor0] 23.0210μs 4.8643μs 205.5803 KOps/s 212.8889 KOps/s $\color{#d91a1a}-3.43\%$
test_memmaptd_index 1.0740ms 0.2926ms 3.4174 KOps/s 3.4275 KOps/s $\color{#d91a1a}-0.29\%$
test_memmaptd_index_astensor 0.5410ms 0.3867ms 2.5858 KOps/s 2.5795 KOps/s $\color{#35bf28}+0.24\%$
test_memmaptd_index_op 0.8769ms 0.6283ms 1.5917 KOps/s 1.6179 KOps/s $\color{#d91a1a}-1.62\%$
test_serialize_model 0.1377s 0.1319s 7.5802 Ops/s 7.6333 Ops/s $\color{#d91a1a}-0.70\%$
test_serialize_model_pickle 2.0759s 1.3657s 0.7322 Ops/s 0.8257 Ops/s $\textbf{\color{#d91a1a}-11.32\%}$
test_serialize_weights 0.2982s 0.1548s 6.4617 Ops/s 7.6454 Ops/s $\textbf{\color{#d91a1a}-15.48\%}$
test_serialize_weights_returnearly 0.3853s 68.1315ms 14.6775 Ops/s 14.2570 Ops/s $\color{#35bf28}+2.95\%$
test_serialize_weights_pickle 1.3751s 1.2147s 0.8232 Ops/s 0.8216 Ops/s $\color{#35bf28}+0.19\%$
test_reshape_pytree 0.3548ms 34.4242μs 29.0493 KOps/s 28.9072 KOps/s $\color{#35bf28}+0.49\%$
test_reshape_td 65.5730μs 41.6811μs 23.9917 KOps/s 24.4743 KOps/s $\color{#d91a1a}-1.97\%$
test_view_pytree 0.2297ms 33.4651μs 29.8819 KOps/s 29.8655 KOps/s $\color{#35bf28}+0.05\%$
test_view_td 73.0530μs 47.8845μs 20.8836 KOps/s 20.8345 KOps/s $\color{#35bf28}+0.24\%$
test_unbind_pytree 0.2451ms 38.0191μs 26.3025 KOps/s 26.2389 KOps/s $\color{#35bf28}+0.24\%$
test_unbind_td 0.1097ms 48.7999μs 20.4918 KOps/s 20.1554 KOps/s $\color{#35bf28}+1.67\%$
test_split_pytree 0.2532ms 44.3762μs 22.5346 KOps/s 21.8906 KOps/s $\color{#35bf28}+2.94\%$
test_split_td 0.1872ms 66.2755μs 15.0885 KOps/s 15.2078 KOps/s $\color{#d91a1a}-0.78\%$
test_add_pytree 0.2309ms 43.6959μs 22.8855 KOps/s 22.8963 KOps/s $\color{#d91a1a}-0.05\%$
test_add_td 88.0650μs 56.1860μs 17.7980 KOps/s 18.3230 KOps/s $\color{#d91a1a}-2.87\%$
test_compile_add_one_nested[tensordict-compile] 0.1942ms 0.1394ms 7.1750 KOps/s 6.6763 KOps/s $\textbf{\color{#35bf28}+7.47\%}$
test_compile_add_one_nested[tensordict-eager] 0.2921ms 0.1875ms 5.3339 KOps/s 5.2953 KOps/s $\color{#35bf28}+0.73\%$
test_compile_add_one_nested[pytree-compile] 0.1438ms 0.1089ms 9.1863 KOps/s 8.8365 KOps/s $\color{#35bf28}+3.96\%$
test_compile_add_one_nested[pytree-eager] 0.3649ms 0.1815ms 5.5104 KOps/s 5.6241 KOps/s $\color{#d91a1a}-2.02\%$
test_compile_copy_nested[tensordict-compile] 59.4930μs 29.0684μs 34.4016 KOps/s 30.6817 KOps/s $\textbf{\color{#35bf28}+12.12\%}$
test_compile_copy_nested[tensordict-eager] 0.1722ms 50.5058μs 19.7997 KOps/s 19.9969 KOps/s $\color{#d91a1a}-0.99\%$
test_compile_copy_nested[pytree-compile] 63.1530μs 13.7865μs 72.5345 KOps/s 73.2301 KOps/s $\color{#d91a1a}-0.95\%$
test_compile_copy_nested[pytree-eager] 0.4089ms 75.9037μs 13.1746 KOps/s 13.1518 KOps/s $\color{#35bf28}+0.17\%$
test_compile_add_one_flat[tensordict-compile] 0.2046ms 0.1634ms 6.1198 KOps/s 5.7992 KOps/s $\textbf{\color{#35bf28}+5.53\%}$
test_compile_add_one_flat[tensordict-eager] 0.3240ms 0.2596ms 3.8514 KOps/s 3.7980 KOps/s $\color{#35bf28}+1.41\%$
test_compile_add_one_flat[tensorclass-compile] 0.1639ms 0.1177ms 8.4932 KOps/s 8.4609 KOps/s $\color{#35bf28}+0.38\%$
test_compile_add_one_flat[tensorclass-eager] 0.1412ms 69.9795μs 14.2899 KOps/s 14.2099 KOps/s $\color{#35bf28}+0.56\%$
test_compile_add_one_flat[pytree-compile] 0.2023ms 0.1578ms 6.3371 KOps/s 6.0699 KOps/s $\color{#35bf28}+4.40\%$
test_compile_add_one_flat[pytree-eager] 0.7334ms 0.5184ms 1.9292 KOps/s 1.9858 KOps/s $\color{#d91a1a}-2.85\%$
test_compile_add_self_flat[tensordict-eager] 0.4202ms 0.3100ms 3.2253 KOps/s 3.1576 KOps/s $\color{#35bf28}+2.15\%$
test_compile_add_self_flat[tensordict-compile] 0.2145ms 0.1648ms 6.0671 KOps/s 5.8156 KOps/s $\color{#35bf28}+4.32\%$
test_compile_add_self_flat[tensorclass-eager] 0.1420ms 87.0961μs 11.4816 KOps/s 11.5181 KOps/s $\color{#d91a1a}-0.32\%$
test_compile_add_self_flat[tensorclass-compile] 0.1582ms 0.1159ms 8.6257 KOps/s 8.2839 KOps/s $\color{#35bf28}+4.13\%$
test_compile_add_self_flat[pytree-eager] 0.6331ms 0.4360ms 2.2936 KOps/s 2.3410 KOps/s $\color{#d91a1a}-2.03\%$
test_compile_add_self_flat[pytree-compile] 0.1958ms 0.1595ms 6.2708 KOps/s 6.0901 KOps/s $\color{#35bf28}+2.97\%$
test_compile_copy_flat[tensordict-compile] 53.4130μs 23.5243μs 42.5093 KOps/s 41.1176 KOps/s $\color{#35bf28}+3.38\%$
test_compile_copy_flat[tensordict-eager] 78.0030μs 41.2106μs 24.2656 KOps/s 24.8330 KOps/s $\color{#d91a1a}-2.28\%$
test_compile_copy_flat[pytree-compile] 0.1004ms 19.1364μs 52.2565 KOps/s 49.2104 KOps/s $\textbf{\color{#35bf28}+6.19\%}$
test_compile_copy_flat[pytree-eager] 0.3660ms 68.4607μs 14.6069 KOps/s 14.3814 KOps/s $\color{#35bf28}+1.57\%$
test_compile_assign_and_add[tensordict-compile] 2.0027ms 0.4671ms 2.1407 KOps/s 1.8515 KOps/s $\textbf{\color{#35bf28}+15.62\%}$
test_compile_assign_and_add[tensordict-eager] 3.4178ms 3.2156ms 310.9808 Ops/s 310.6087 Ops/s $\color{#35bf28}+0.12\%$
test_compile_assign_and_add[pytree-compile] 2.0035ms 0.5273ms 1.8964 KOps/s 1.9012 KOps/s $\color{#d91a1a}-0.25\%$
test_compile_assign_and_add[pytree-eager] 2.9886ms 2.7830ms 359.3185 Ops/s 364.5537 Ops/s $\color{#d91a1a}-1.44\%$
test_compile_indexing[tensor-tensordict-compile] 0.2443ms 0.1331ms 7.5113 KOps/s 7.3344 KOps/s $\color{#35bf28}+2.41\%$
test_compile_indexing[tensor-tensordict-eager] 0.2627ms 97.0291μs 10.3062 KOps/s 10.5951 KOps/s $\color{#d91a1a}-2.73\%$
test_compile_indexing[tensor-tensorclass-compile] 0.1983ms 0.1248ms 8.0109 KOps/s 7.8806 KOps/s $\color{#35bf28}+1.65\%$
test_compile_indexing[tensor-tensorclass-eager] 0.2716ms 79.6899μs 12.5486 KOps/s 12.4152 KOps/s $\color{#35bf28}+1.07\%$
test_compile_indexing[tensor-pytree-compile] 0.1865ms 0.1258ms 7.9492 KOps/s 7.8251 KOps/s $\color{#35bf28}+1.59\%$
test_compile_indexing[tensor-pytree-eager] 0.3463ms 81.7204μs 12.2368 KOps/s 12.0681 KOps/s $\color{#35bf28}+1.40\%$
test_compile_indexing[slice-tensordict-compile] 0.1642ms 0.1223ms 8.1786 KOps/s 8.3037 KOps/s $\color{#d91a1a}-1.51\%$
test_compile_indexing[slice-tensordict-eager] 0.2149ms 26.5415μs 37.6768 KOps/s 36.8880 KOps/s $\color{#35bf28}+2.14\%$
test_compile_indexing[slice-tensorclass-compile] 0.1505ms 0.1136ms 8.8059 KOps/s 8.6358 KOps/s $\color{#35bf28}+1.97\%$
test_compile_indexing[slice-tensorclass-eager] 0.2090ms 23.2323μs 43.0435 KOps/s 42.8627 KOps/s $\color{#35bf28}+0.42\%$
test_compile_indexing[slice-pytree-compile] 0.1572ms 0.1145ms 8.7340 KOps/s 8.5848 KOps/s $\color{#35bf28}+1.74\%$
test_compile_indexing[slice-pytree-eager] 0.2372ms 23.3102μs 42.8997 KOps/s 42.9349 KOps/s $\color{#d91a1a}-0.08\%$
test_compile_indexing[int-tensordict-compile] 0.1690ms 0.1207ms 8.2836 KOps/s 8.2068 KOps/s $\color{#35bf28}+0.94\%$
test_compile_indexing[int-tensordict-eager] 0.1995ms 26.1456μs 38.2474 KOps/s 37.7105 KOps/s $\color{#35bf28}+1.42\%$
test_compile_indexing[int-tensorclass-compile] 0.1552ms 0.1147ms 8.7157 KOps/s 8.5797 KOps/s $\color{#35bf28}+1.59\%$
test_compile_indexing[int-tensorclass-eager] 0.2210ms 23.2456μs 43.0190 KOps/s 43.5432 KOps/s $\color{#d91a1a}-1.20\%$
test_compile_indexing[int-pytree-compile] 0.1631ms 0.1166ms 8.5789 KOps/s 8.6454 KOps/s $\color{#d91a1a}-0.77\%$
test_compile_indexing[int-pytree-eager] 0.2578ms 23.3352μs 42.8538 KOps/s 43.2014 KOps/s $\color{#d91a1a}-0.80\%$
test_mod_add[eager] 86.8750μs 48.6316μs 20.5628 KOps/s 20.3955 KOps/s $\color{#35bf28}+0.82\%$
test_mod_add[compile] 0.2571ms 94.8281μs 10.5454 KOps/s 10.2106 KOps/s $\color{#35bf28}+3.28\%$
test_mod_add[compile-overhead] 0.3276ms 0.1833ms 5.4568 KOps/s 5.1067 KOps/s $\textbf{\color{#35bf28}+6.86\%}$
test_mod_wrap[eager] 0.4110ms 0.2940ms 3.4010 KOps/s 3.3945 KOps/s $\color{#35bf28}+0.19\%$
test_mod_wrap[compile] 0.4102ms 0.3502ms 2.8557 KOps/s 2.8800 KOps/s $\color{#d91a1a}-0.84\%$
test_mod_wrap[compile-overhead] 7.4847ms 4.0957ms 244.1614 Ops/s 240.5002 Ops/s $\color{#35bf28}+1.52\%$
test_mod_wrap_and_backward[eager] 1.6749ms 1.5122ms 661.2923 Ops/s 643.8737 Ops/s $\color{#35bf28}+2.71\%$
test_mod_wrap_and_backward[compile] 1.8813ms 1.5863ms 630.4148 Ops/s 654.4108 Ops/s $\color{#d91a1a}-3.67\%$
test_mod_wrap_and_backward[compile-overhead] 1.6268ms 1.1112ms 899.9643 Ops/s 996.8619 Ops/s $\textbf{\color{#d91a1a}-9.72\%}$
test_seq_add[eager] 0.2066ms 0.1513ms 6.6093 KOps/s 6.2084 KOps/s $\textbf{\color{#35bf28}+6.46\%}$
test_seq_add[compile] 0.1571ms 0.1064ms 9.3973 KOps/s 8.7092 KOps/s $\textbf{\color{#35bf28}+7.90\%}$
test_seq_add[compile-overhead] 0.1966ms 0.1440ms 6.9459 KOps/s 6.5697 KOps/s $\textbf{\color{#35bf28}+5.73\%}$
test_seq_wrap[eager] 0.5919ms 0.5184ms 1.9290 KOps/s 1.8328 KOps/s $\textbf{\color{#35bf28}+5.25\%}$
test_seq_wrap[compile] 0.4007ms 0.3563ms 2.8066 KOps/s 2.8007 KOps/s $\color{#35bf28}+0.21\%$
test_seq_wrap[compile-overhead] 0.3120ms 0.2573ms 3.8858 KOps/s 3.7417 KOps/s $\color{#35bf28}+3.85\%$
test_func_call_runtime[False-eager] 0.9612ms 0.8984ms 1.1131 KOps/s 1.1568 KOps/s $\color{#d91a1a}-3.78\%$
test_func_call_runtime[False-compile] 0.9792ms 0.9046ms 1.1055 KOps/s 1.0896 KOps/s $\color{#35bf28}+1.46\%$
test_func_call_runtime[False-compile-overhead] 0.4644ms 0.4202ms 2.3799 KOps/s 2.3154 KOps/s $\color{#35bf28}+2.79\%$
test_func_call_runtime[True-eager] 1.2233ms 1.1037ms 906.0814 Ops/s 901.3350 Ops/s $\color{#35bf28}+0.53\%$
test_func_call_runtime[True-compile] 1.0530ms 0.9290ms 1.0764 KOps/s 1.0660 KOps/s $\color{#35bf28}+0.98\%$
test_func_call_runtime[True-compile-overhead] 0.4889ms 0.4435ms 2.2548 KOps/s 2.2425 KOps/s $\color{#35bf28}+0.55\%$
test_func_call_cm_runtime[False-eager] 1.2515ms 0.8941ms 1.1184 KOps/s 1.1651 KOps/s $\color{#d91a1a}-4.01\%$
test_func_call_cm_runtime[False-compile] 1.1040ms 0.9173ms 1.0901 KOps/s 1.0869 KOps/s $\color{#35bf28}+0.30\%$
test_func_call_cm_runtime[False-compile-overhead] 0.4806ms 0.4235ms 2.3615 KOps/s 2.3620 KOps/s $\color{#d91a1a}-0.02\%$
test_func_call_cm_runtime[True-eager] 1.3282ms 1.2278ms 814.4509 Ops/s 790.5977 Ops/s $\color{#35bf28}+3.02\%$
test_func_call_cm_runtime[True-compile] 1.0459ms 0.9692ms 1.0318 KOps/s 973.3280 Ops/s $\textbf{\color{#35bf28}+6.00\%}$
test_func_call_cm_runtime[True-compile-overhead] 0.5274ms 0.4767ms 2.0976 KOps/s 2.1008 KOps/s $\color{#d91a1a}-0.15\%$
test_vmap_func_call_cm_runtime[eager] 2.8496ms 2.3184ms 431.3406 Ops/s 430.3346 Ops/s $\color{#35bf28}+0.23\%$
test_vmap_func_call_cm_runtime[compile] 1.1434ms 0.9816ms 1.0188 KOps/s 959.9870 Ops/s $\textbf{\color{#35bf28}+6.12\%}$
test_vmap_func_call_cm_runtime[compile-overhead] 0.5282ms 0.4715ms 2.1208 KOps/s 1.9982 KOps/s $\textbf{\color{#35bf28}+6.14\%}$
test_distributed 0.5434ms 0.1523ms 6.5658 KOps/s 6.4949 KOps/s $\color{#35bf28}+1.09\%$
test_tdmodule 59.7130μs 27.2683μs 36.6727 KOps/s 35.9230 KOps/s $\color{#35bf28}+2.09\%$
test_tdmodule_dispatch 68.8730μs 47.8089μs 20.9166 KOps/s 20.7889 KOps/s $\color{#35bf28}+0.61\%$
test_tdseq 43.8020μs 26.1462μs 38.2465 KOps/s 37.8561 KOps/s $\color{#35bf28}+1.03\%$
test_tdseq_dispatch 70.2930μs 48.9212μs 20.4411 KOps/s 20.0988 KOps/s $\color{#35bf28}+1.70\%$
test_instantiation_functorch 2.1355ms 2.0676ms 483.6541 Ops/s 486.5904 Ops/s $\color{#d91a1a}-0.60\%$
test_exec_functorch 0.2242ms 0.1796ms 5.5667 KOps/s 5.4912 KOps/s $\color{#35bf28}+1.38\%$
test_exec_functional_call 0.2099ms 0.1613ms 6.2014 KOps/s 5.9654 KOps/s $\color{#35bf28}+3.96\%$
test_exec_td_decorator 0.4465ms 0.2340ms 4.2732 KOps/s 4.0952 KOps/s $\color{#35bf28}+4.34\%$
test_vmap_mlp_speed_decorator[True-True] 0.9401ms 0.7775ms 1.2862 KOps/s 1.2608 KOps/s $\color{#35bf28}+2.02\%$
test_vmap_mlp_speed_decorator[True-False] 0.9401ms 0.7711ms 1.2968 KOps/s 1.2750 KOps/s $\color{#35bf28}+1.71\%$
test_vmap_mlp_speed_decorator[False-True] 0.8089ms 0.6665ms 1.5004 KOps/s 1.4413 KOps/s $\color{#35bf28}+4.10\%$
test_vmap_mlp_speed_decorator[False-False] 0.8182ms 0.6706ms 1.4911 KOps/s 1.4551 KOps/s $\color{#35bf28}+2.48\%$
test_vmap_transformer_speed_decorator[True-True] 21.1799ms 20.5566ms 48.6462 Ops/s 48.6032 Ops/s $\color{#35bf28}+0.09\%$
test_vmap_transformer_speed_decorator[True-False] 21.2488ms 20.4870ms 48.8115 Ops/s 48.1947 Ops/s $\color{#35bf28}+1.28\%$
test_vmap_transformer_speed_decorator[False-True] 20.4058ms 20.2862ms 49.2947 Ops/s 49.0350 Ops/s $\color{#35bf28}+0.53\%$
test_vmap_transformer_speed_decorator[False-False] 20.6190ms 20.3388ms 49.1672 Ops/s 48.4259 Ops/s $\color{#35bf28}+1.53\%$
test_to_module_speed[True] 2.2862ms 1.4740ms 678.4451 Ops/s 669.1348 Ops/s $\color{#35bf28}+1.39\%$
test_to_module_speed[False] 1.9398ms 1.4351ms 696.7923 Ops/s 689.4610 Ops/s $\color{#35bf28}+1.06\%$
test_tc_init 88.4340μs 50.9028μs 19.6453 KOps/s 19.5909 KOps/s $\color{#35bf28}+0.28\%$
test_tc_init_tensor_only 60.5230μs 14.7182μs 67.9433 KOps/s 66.4923 KOps/s $\color{#35bf28}+2.18\%$
test_tc_init_nested 0.1426ms 0.1004ms 9.9569 KOps/s 9.8059 KOps/s $\color{#35bf28}+1.54\%$
test_tc_first_layer_tensor 25.9510μs 1.7756μs 563.2035 KOps/s 559.2115 KOps/s $\color{#35bf28}+0.71\%$
test_tc_first_layer_tensor_only 5.9446μs 0.6861μs 1.4575 MOps/s 1.4702 MOps/s $\color{#d91a1a}-0.86\%$
test_tc_first_layer_tensor_set 40.7820μs 4.2103μs 237.5129 KOps/s 240.5460 KOps/s $\color{#d91a1a}-1.26\%$
test_tc_first_layer_tensor_only_set 0.1066ms 2.9906μs 334.3800 KOps/s 318.9639 KOps/s $\color{#35bf28}+4.83\%$
test_tc_first_layer_nontensor 28.7510μs 5.8342μs 171.4021 KOps/s 170.8840 KOps/s $\color{#35bf28}+0.30\%$
test_tc_second_layer_tensor 27.0310μs 4.2492μs 235.3396 KOps/s 233.1815 KOps/s $\color{#35bf28}+0.93\%$
test_tc_second_layer_nontensor 31.9520μs 8.2840μs 120.7142 KOps/s 120.4234 KOps/s $\color{#35bf28}+0.24\%$
test_unbind 0.2535s 13.4034ms 74.6082 Ops/s 70.8476 Ops/s $\textbf{\color{#35bf28}+5.31\%}$
test_full_like 17.7612ms 4.5775ms 218.4609 Ops/s 227.2720 Ops/s $\color{#d91a1a}-3.88\%$
test_zeros_like 5.8576ms 4.2514ms 235.2183 Ops/s 228.0539 Ops/s $\color{#35bf28}+3.14\%$
test_ones_like 4.7918ms 4.3384ms 230.5014 Ops/s 227.0879 Ops/s $\color{#35bf28}+1.50\%$
test_clone 6.8968ms 6.5761ms 152.0663 Ops/s 151.4787 Ops/s $\color{#35bf28}+0.39\%$
test_squeeze 0.1702ms 14.4698μs 69.1095 KOps/s 64.5731 KOps/s $\textbf{\color{#35bf28}+7.03\%}$
test_unsqueeze 0.2520ms 0.1047ms 9.5552 KOps/s 9.0263 KOps/s $\textbf{\color{#35bf28}+5.86\%}$
test_split 0.2334ms 0.1821ms 5.4921 KOps/s 5.3857 KOps/s $\color{#35bf28}+1.98\%$
test_permute 0.2545ms 0.2019ms 4.9534 KOps/s 4.8935 KOps/s $\color{#35bf28}+1.22\%$
test_stack 52.1028ms 51.7426ms 19.3264 Ops/s 19.2937 Ops/s $\color{#35bf28}+0.17\%$
test_cat 52.0258ms 51.5243ms 19.4083 Ops/s 19.2960 Ops/s $\color{#35bf28}+0.58\%$

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