Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

[BugFix] Fix compile during _check_keys #1239

Merged
merged 2 commits into from
Feb 26, 2025
Merged

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Feb 26, 2025

[ghstack-poisoned]
@facebook-github-bot facebook-github-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 Feb 26, 2025
vmoens added a commit that referenced this pull request Feb 26, 2025
ghstack-source-id: b7549a04d132b88dd19fbb25798a136bae73a607
Pull Request resolved: #1239
[ghstack-poisoned]
vmoens added a commit that referenced this pull request Feb 26, 2025
ghstack-source-id: cfed094e425a60c62617ecfa454d3104ff1f461c
Pull Request resolved: #1239
Comment on lines +1798 to +1801
if is_comp:
return [key for key in keys] # noqa: C416
else:
return list(keys)
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@anijain2305 if you look up, we need to do this for lists and sets because dynamo cannot interpret that these two expressions are identical.
Would it be possible to make set(iterable) and list(iterable) work natively within dynamo?

Copy link

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

Total Benchmarks: 217. Improved: $\large\color{#35bf28}18$. Worsened: $\large\color{#d91a1a}6$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 50.1340μs 20.4916μs 48.8004 KOps/s 48.7382 KOps/s $\color{#35bf28}+0.13\%$
test_plain_set_stack_nested 46.6880μs 20.4182μs 48.9760 KOps/s 48.5332 KOps/s $\color{#35bf28}+0.91\%$
test_plain_set_nested_inplace 81.8700μs 22.1869μs 45.0717 KOps/s 45.0668 KOps/s $\color{#35bf28}+0.01\%$
test_plain_set_stack_nested_inplace 50.6350μs 22.3147μs 44.8136 KOps/s 45.1036 KOps/s $\color{#d91a1a}-0.64\%$
test_items 15.5390μs 4.1257μs 242.3839 KOps/s 239.5733 KOps/s $\color{#35bf28}+1.17\%$
test_items_nested 0.7030ms 0.3992ms 2.5050 KOps/s 2.4401 KOps/s $\color{#35bf28}+2.66\%$
test_items_nested_locked 1.5121ms 0.4008ms 2.4948 KOps/s 2.4158 KOps/s $\color{#35bf28}+3.27\%$
test_items_nested_leaf 0.1449ms 76.2389μs 13.1167 KOps/s 12.8209 KOps/s $\color{#35bf28}+2.31\%$
test_items_stack_nested 0.5653ms 0.4000ms 2.5000 KOps/s 2.4178 KOps/s $\color{#35bf28}+3.40\%$
test_items_stack_nested_leaf 0.1652ms 77.3736μs 12.9243 KOps/s 12.9484 KOps/s $\color{#d91a1a}-0.19\%$
test_items_stack_nested_locked 0.7328ms 0.4033ms 2.4798 KOps/s 2.4043 KOps/s $\color{#35bf28}+3.14\%$
test_keys 23.0130μs 3.4301μs 291.5330 KOps/s 282.1894 KOps/s $\color{#35bf28}+3.31\%$
test_keys_nested 0.2586ms 0.1666ms 6.0037 KOps/s 6.0386 KOps/s $\color{#d91a1a}-0.58\%$
test_keys_nested_locked 1.5666ms 0.1739ms 5.7489 KOps/s 5.7899 KOps/s $\color{#d91a1a}-0.71\%$
test_keys_nested_leaf 0.2315ms 0.1462ms 6.8390 KOps/s 6.8225 KOps/s $\color{#35bf28}+0.24\%$
test_keys_stack_nested 0.3017ms 0.1667ms 5.9972 KOps/s 5.9774 KOps/s $\color{#35bf28}+0.33\%$
test_keys_stack_nested_leaf 0.2345ms 0.1464ms 6.8290 KOps/s 6.8920 KOps/s $\color{#d91a1a}-0.91\%$
test_keys_stack_nested_locked 0.2682ms 0.1739ms 5.7495 KOps/s 5.7555 KOps/s $\color{#d91a1a}-0.10\%$
test_values 4.5906μs 1.0335μs 967.6036 KOps/s 920.6102 KOps/s $\textbf{\color{#35bf28}+5.10\%}$
test_values_nested 0.1187ms 62.8107μs 15.9209 KOps/s 15.6174 KOps/s $\color{#35bf28}+1.94\%$
test_values_nested_locked 0.1124ms 62.4915μs 16.0022 KOps/s 15.7741 KOps/s $\color{#35bf28}+1.45\%$
test_values_nested_leaf 0.1352ms 72.2993μs 13.8314 KOps/s 13.8639 KOps/s $\color{#d91a1a}-0.23\%$
test_values_stack_nested 0.1100ms 63.1675μs 15.8309 KOps/s 15.7924 KOps/s $\color{#35bf28}+0.24\%$
test_values_stack_nested_leaf 0.1335ms 72.8912μs 13.7191 KOps/s 13.9719 KOps/s $\color{#d91a1a}-1.81\%$
test_values_stack_nested_locked 0.1153ms 62.9658μs 15.8816 KOps/s 15.7728 KOps/s $\color{#35bf28}+0.69\%$
test_membership 9.7780μs 0.8483μs 1.1788 MOps/s 1.1751 MOps/s $\color{#35bf28}+0.32\%$
test_membership_nested 32.4810μs 2.8381μs 352.3504 KOps/s 340.7724 KOps/s $\color{#35bf28}+3.40\%$
test_membership_nested_leaf 29.0740μs 2.9004μs 344.7813 KOps/s 339.5580 KOps/s $\color{#35bf28}+1.54\%$
test_membership_stacked_nested 14.7080μs 2.8424μs 351.8183 KOps/s 344.6237 KOps/s $\color{#35bf28}+2.09\%$
test_membership_stacked_nested_leaf 21.8210μs 2.8442μs 351.5942 KOps/s 344.5630 KOps/s $\color{#35bf28}+2.04\%$
test_membership_nested_last 35.3560μs 4.3052μs 232.2784 KOps/s 229.2606 KOps/s $\color{#35bf28}+1.32\%$
test_membership_nested_leaf_last 57.5080μs 4.3505μs 229.8571 KOps/s 227.9464 KOps/s $\color{#35bf28}+0.84\%$
test_membership_stacked_nested_last 25.3580μs 4.3223μs 231.3591 KOps/s 227.4289 KOps/s $\color{#35bf28}+1.73\%$
test_membership_stacked_nested_leaf_last 24.6660μs 4.3462μs 230.0870 KOps/s 230.7239 KOps/s $\color{#d91a1a}-0.28\%$
test_nested_getleaf 30.5470μs 10.6179μs 94.1804 KOps/s 95.3475 KOps/s $\color{#d91a1a}-1.22\%$
test_nested_get 37.8300μs 10.0593μs 99.4105 KOps/s 99.7200 KOps/s $\color{#d91a1a}-0.31\%$
test_stacked_getleaf 33.9130μs 10.5905μs 94.4246 KOps/s 96.0966 KOps/s $\color{#d91a1a}-1.74\%$
test_stacked_get 28.5140μs 9.9986μs 100.0135 KOps/s 100.8685 KOps/s $\color{#d91a1a}-0.85\%$
test_nested_getitemleaf 34.9150μs 11.2717μs 88.7179 KOps/s 89.4435 KOps/s $\color{#d91a1a}-0.81\%$
test_nested_getitem 31.3990μs 10.8323μs 92.3165 KOps/s 94.5059 KOps/s $\color{#d91a1a}-2.32\%$
test_stacked_getitemleaf 31.8800μs 11.1850μs 89.4051 KOps/s 90.2586 KOps/s $\color{#d91a1a}-0.95\%$
test_stacked_getitem 30.3370μs 10.7474μs 93.0461 KOps/s 94.7080 KOps/s $\color{#d91a1a}-1.75\%$
test_lock_nested 0.8505ms 0.4160ms 2.4036 KOps/s 2.4304 KOps/s $\color{#d91a1a}-1.10\%$
test_lock_stack_nested 0.7677ms 0.4243ms 2.3569 KOps/s 2.3440 KOps/s $\color{#35bf28}+0.55\%$
test_unlock_nested 0.6183ms 0.3373ms 2.9649 KOps/s 2.9331 KOps/s $\color{#35bf28}+1.08\%$
test_unlock_stack_nested 0.6327ms 0.3401ms 2.9407 KOps/s 2.8973 KOps/s $\color{#35bf28}+1.50\%$
test_flatten_speed 0.1815ms 0.1022ms 9.7864 KOps/s 10.0725 KOps/s $\color{#d91a1a}-2.84\%$
test_unflatten_speed 0.9643ms 0.5363ms 1.8647 KOps/s 1.9148 KOps/s $\color{#d91a1a}-2.62\%$
test_common_ops 1.5898ms 0.8104ms 1.2340 KOps/s 1.2268 KOps/s $\color{#35bf28}+0.59\%$
test_creation 32.7810μs 2.4991μs 400.1393 KOps/s 397.8769 KOps/s $\color{#35bf28}+0.57\%$
test_creation_empty 28.1830μs 11.5038μs 86.9281 KOps/s 82.2505 KOps/s $\textbf{\color{#35bf28}+5.69\%}$
test_creation_nested_1 37.6510μs 14.3307μs 69.7801 KOps/s 66.5591 KOps/s $\color{#35bf28}+4.84\%$
test_creation_nested_2 53.1900μs 18.8947μs 52.9250 KOps/s 51.2752 KOps/s $\color{#35bf28}+3.22\%$
test_clone 0.6667ms 13.2292μs 75.5904 KOps/s 73.5655 KOps/s $\color{#35bf28}+2.75\%$
test_getitem[int] 0.8003ms 12.4857μs 80.0915 KOps/s 78.6686 KOps/s $\color{#35bf28}+1.81\%$
test_getitem[slice_int] 0.1500ms 24.0620μs 41.5593 KOps/s 41.0553 KOps/s $\color{#35bf28}+1.23\%$
test_getitem[range] 0.2113ms 49.0438μs 20.3899 KOps/s 19.9155 KOps/s $\color{#35bf28}+2.38\%$
test_getitem[tuple] 0.1386ms 20.3188μs 49.2156 KOps/s 49.8205 KOps/s $\color{#d91a1a}-1.21\%$
test_getitem[list] 0.1576ms 44.2752μs 22.5860 KOps/s 21.7005 KOps/s $\color{#35bf28}+4.08\%$
test_setitem_dim[int] 61.3850μs 25.5826μs 39.0891 KOps/s 38.4412 KOps/s $\color{#35bf28}+1.69\%$
test_setitem_dim[slice_int] 99.6260μs 50.7789μs 19.6932 KOps/s 19.5314 KOps/s $\color{#35bf28}+0.83\%$
test_setitem_dim[range] 0.1259ms 76.4237μs 13.0849 KOps/s 13.0140 KOps/s $\color{#35bf28}+0.54\%$
test_setitem_dim[tuple] 72.4260μs 40.3949μs 24.7556 KOps/s 24.7910 KOps/s $\color{#d91a1a}-0.14\%$
test_setitem 0.1511ms 20.5271μs 48.7160 KOps/s 48.4314 KOps/s $\color{#35bf28}+0.59\%$
test_set 60.3330μs 19.8086μs 50.4832 KOps/s 50.2202 KOps/s $\color{#35bf28}+0.52\%$
test_set_shared 3.2650ms 0.1813ms 5.5168 KOps/s 5.3560 KOps/s $\color{#35bf28}+3.00\%$
test_update 0.1157ms 25.4134μs 39.3493 KOps/s 39.2102 KOps/s $\color{#35bf28}+0.35\%$
test_update_nested 85.6700μs 40.5885μs 24.6375 KOps/s 24.2825 KOps/s $\color{#35bf28}+1.46\%$
test_update__nested 0.5533ms 32.9945μs 30.3081 KOps/s 30.1211 KOps/s $\color{#35bf28}+0.62\%$
test_set_nested 71.7340μs 22.1705μs 45.1050 KOps/s 45.5127 KOps/s $\color{#d91a1a}-0.90\%$
test_set_nested_new 66.8050μs 26.9617μs 37.0896 KOps/s 36.2318 KOps/s $\color{#35bf28}+2.37\%$
test_select 0.1231ms 42.8157μs 23.3559 KOps/s 23.0607 KOps/s $\color{#35bf28}+1.28\%$
test_select_nested 0.2900ms 63.6371μs 15.7141 KOps/s 15.7782 KOps/s $\color{#d91a1a}-0.41\%$
test_exclude_nested 0.1611ms 80.0805μs 12.4874 KOps/s 12.2617 KOps/s $\color{#35bf28}+1.84\%$
test_empty[True] 0.5977ms 0.4036ms 2.4776 KOps/s 2.4180 KOps/s $\color{#35bf28}+2.47\%$
test_empty[False] 7.3788μs 1.3463μs 742.7681 KOps/s 706.0189 KOps/s $\textbf{\color{#35bf28}+5.21\%}$
test_unbind_speed 0.3664ms 0.2706ms 3.6954 KOps/s 3.6529 KOps/s $\color{#35bf28}+1.16\%$
test_unbind_speed_stack0 0.7749ms 0.2669ms 3.7462 KOps/s 3.7575 KOps/s $\color{#d91a1a}-0.30\%$
test_unbind_speed_stack1 0.1013s 0.7323ms 1.3656 KOps/s 1.2119 KOps/s $\textbf{\color{#35bf28}+12.68\%}$
test_split 0.1047s 1.7294ms 578.2362 Ops/s 560.0861 Ops/s $\color{#35bf28}+3.24\%$
test_chunk 0.1094s 1.7356ms 576.1715 Ops/s 616.6393 Ops/s $\textbf{\color{#d91a1a}-6.56\%}$
test_consolidate_njt[False-None] 10.9249ms 8.2313ms 121.4881 Ops/s 109.6028 Ops/s $\textbf{\color{#35bf28}+10.84\%}$
test_creation[device0] 0.1939ms 91.6293μs 10.9135 KOps/s 11.0539 KOps/s $\color{#d91a1a}-1.27\%$
test_creation_from_tensor 0.2823ms 94.7987μs 10.5487 KOps/s 10.5514 KOps/s $\color{#d91a1a}-0.03\%$
test_add_one[memmap_tensor0] 0.1147ms 5.1943μs 192.5184 KOps/s 198.0917 KOps/s $\color{#d91a1a}-2.81\%$
test_contiguous[memmap_tensor0] 12.3340μs 0.5117μs 1.9544 MOps/s 1.8738 MOps/s $\color{#35bf28}+4.30\%$
test_stack[memmap_tensor0] 30.1570μs 3.5410μs 282.4059 KOps/s 289.9516 KOps/s $\color{#d91a1a}-2.60\%$
test_memmaptd_index 1.4236ms 0.2299ms 4.3490 KOps/s 4.2277 KOps/s $\color{#35bf28}+2.87\%$
test_memmaptd_index_astensor 0.4993ms 0.3155ms 3.1691 KOps/s 3.0665 KOps/s $\color{#35bf28}+3.35\%$
test_memmaptd_index_op 1.1202ms 0.5860ms 1.7065 KOps/s 1.7019 KOps/s $\color{#35bf28}+0.27\%$
test_serialize_model 0.2369s 0.1366s 7.3196 Ops/s 8.5451 Ops/s $\textbf{\color{#d91a1a}-14.34\%}$
test_serialize_model_pickle 0.5053s 0.4038s 2.4768 Ops/s 2.5909 Ops/s $\color{#d91a1a}-4.41\%$
test_serialize_weights 0.1196s 0.1138s 8.7851 Ops/s 8.5992 Ops/s $\color{#35bf28}+2.16\%$
test_serialize_weights_returnearly 0.1850s 0.1593s 6.2788 Ops/s 6.4799 Ops/s $\color{#d91a1a}-3.10\%$
test_serialize_weights_pickle 0.4710s 0.4111s 2.4325 Ops/s 2.3477 Ops/s $\color{#35bf28}+3.61\%$
test_serialize_weights_filesystem 0.1550s 0.1449s 6.8998 Ops/s 6.8947 Ops/s $\color{#35bf28}+0.07\%$
test_serialize_model_filesystem 0.1585s 0.1481s 6.7521 Ops/s 6.4684 Ops/s $\color{#35bf28}+4.39\%$
test_reshape_pytree 58.9010μs 26.3383μs 37.9675 KOps/s 37.9687 KOps/s $-0.00\%$
test_reshape_td 76.9640μs 31.8933μs 31.3546 KOps/s 30.4564 KOps/s $\color{#35bf28}+2.95\%$
test_view_pytree 83.5070μs 25.7934μs 38.7696 KOps/s 37.9462 KOps/s $\color{#35bf28}+2.17\%$
test_view_td 0.1089ms 38.2232μs 26.1621 KOps/s 25.1298 KOps/s $\color{#35bf28}+4.11\%$
test_unbind_pytree 58.0680μs 28.7539μs 34.7779 KOps/s 33.5111 KOps/s $\color{#35bf28}+3.78\%$
test_unbind_td 0.3593ms 39.2578μs 25.4726 KOps/s 24.6672 KOps/s $\color{#35bf28}+3.27\%$
test_split_pytree 60.6530μs 28.5924μs 34.9743 KOps/s 33.8125 KOps/s $\color{#35bf28}+3.44\%$
test_split_td 0.5328ms 45.2449μs 22.1019 KOps/s 22.0008 KOps/s $\color{#35bf28}+0.46\%$
test_add_pytree 0.1071ms 34.7866μs 28.7467 KOps/s 27.8947 KOps/s $\color{#35bf28}+3.05\%$
test_add_td 0.1416ms 55.3916μs 18.0533 KOps/s 17.2897 KOps/s $\color{#35bf28}+4.42\%$
test_compile_add_one_nested[tensordict-compile] 0.1758ms 65.6970μs 15.2214 KOps/s 14.7561 KOps/s $\color{#35bf28}+3.15\%$
test_compile_add_one_nested[tensordict-eager] 0.3445ms 0.1684ms 5.9376 KOps/s 5.8556 KOps/s $\color{#35bf28}+1.40\%$
test_compile_add_one_nested[pytree-compile] 0.1415ms 44.6421μs 22.4004 KOps/s 21.9132 KOps/s $\color{#35bf28}+2.22\%$
test_compile_add_one_nested[pytree-eager] 0.2995ms 0.1167ms 8.5702 KOps/s 8.4697 KOps/s $\color{#35bf28}+1.19\%$
test_compile_copy_nested[tensordict-compile] 75.4710μs 27.5197μs 36.3376 KOps/s 34.6908 KOps/s $\color{#35bf28}+4.75\%$
test_compile_copy_nested[tensordict-eager] 0.1126ms 59.1084μs 16.9181 KOps/s 16.9913 KOps/s $\color{#d91a1a}-0.43\%$
test_compile_copy_nested[pytree-compile] 0.1574ms 78.3848μs 12.7576 KOps/s 12.7298 KOps/s $\color{#35bf28}+0.22\%$
test_compile_copy_nested[pytree-eager] 0.1222ms 65.5597μs 15.2533 KOps/s 14.9828 KOps/s $\color{#35bf28}+1.81\%$
test_compile_add_one_flat[tensordict-compile] 0.2709ms 0.1061ms 9.4224 KOps/s 9.0740 KOps/s $\color{#35bf28}+3.84\%$
test_compile_add_one_flat[tensordict-eager] 0.3639ms 0.2116ms 4.7259 KOps/s 4.6088 KOps/s $\color{#35bf28}+2.54\%$
test_compile_add_one_flat[tensorclass-compile] 0.1054ms 46.2139μs 21.6385 KOps/s 21.2022 KOps/s $\color{#35bf28}+2.06\%$
test_compile_add_one_flat[tensorclass-eager] 0.1323ms 65.0619μs 15.3700 KOps/s 14.7082 KOps/s $\color{#35bf28}+4.50\%$
test_compile_add_one_flat[pytree-compile] 0.1705ms 99.9882μs 10.0012 KOps/s 9.8425 KOps/s $\color{#35bf28}+1.61\%$
test_compile_add_one_flat[pytree-eager] 0.3685ms 0.1983ms 5.0438 KOps/s 4.8602 KOps/s $\color{#35bf28}+3.78\%$
test_compile_add_self_flat[tensordict-eager] 0.4844ms 0.2288ms 4.3703 KOps/s 4.2839 KOps/s $\color{#35bf28}+2.02\%$
test_compile_add_self_flat[tensordict-compile] 0.2100ms 0.1072ms 9.3327 KOps/s 9.3517 KOps/s $\color{#d91a1a}-0.20\%$
test_compile_add_self_flat[tensorclass-eager] 0.1590ms 63.6163μs 15.7193 KOps/s 15.5752 KOps/s $\color{#35bf28}+0.93\%$
test_compile_add_self_flat[tensorclass-compile] 0.1112ms 47.2343μs 21.1711 KOps/s 20.4679 KOps/s $\color{#35bf28}+3.44\%$
test_compile_add_self_flat[pytree-eager] 0.2302ms 0.1543ms 6.4814 KOps/s 6.2096 KOps/s $\color{#35bf28}+4.38\%$
test_compile_add_self_flat[pytree-compile] 0.2473ms 0.1007ms 9.9302 KOps/s 9.6490 KOps/s $\color{#35bf28}+2.91\%$
test_compile_copy_flat[tensordict-compile] 53.4400μs 20.3583μs 49.1200 KOps/s 46.3985 KOps/s $\textbf{\color{#35bf28}+5.87\%}$
test_compile_copy_flat[tensordict-eager] 0.1257ms 68.4565μs 14.6078 KOps/s 14.5459 KOps/s $\color{#35bf28}+0.43\%$
test_compile_copy_flat[pytree-compile] 0.1800ms 79.2451μs 12.6191 KOps/s 12.2944 KOps/s $\color{#35bf28}+2.64\%$
test_compile_copy_flat[pytree-eager] 0.1395ms 66.5984μs 15.0154 KOps/s 14.4644 KOps/s $\color{#35bf28}+3.81\%$
test_compile_assign_and_add[tensordict-compile] 0.3049ms 0.2098ms 4.7670 KOps/s 4.6154 KOps/s $\color{#35bf28}+3.29\%$
test_compile_assign_and_add[tensordict-eager] 1.6002ms 1.3637ms 733.2913 Ops/s 722.6040 Ops/s $\color{#35bf28}+1.48\%$
test_compile_assign_and_add[pytree-compile] 0.3764ms 0.2033ms 4.9183 KOps/s 4.7666 KOps/s $\color{#35bf28}+3.18\%$
test_compile_assign_and_add[pytree-eager] 0.9146ms 0.8083ms 1.2371 KOps/s 1.1947 KOps/s $\color{#35bf28}+3.55\%$
test_compile_assign_and_add_stack[compile] 0.5184ms 0.4425ms 2.2597 KOps/s 2.1653 KOps/s $\color{#35bf28}+4.36\%$
test_compile_assign_and_add_stack[eager] 2.8648ms 2.6050ms 383.8794 Ops/s 360.5961 Ops/s $\textbf{\color{#35bf28}+6.46\%}$
test_compile_indexing[tensor-tensordict-compile] 0.1039ms 36.5830μs 27.3351 KOps/s 25.7673 KOps/s $\textbf{\color{#35bf28}+6.08\%}$
test_compile_indexing[tensor-tensordict-eager] 0.5876ms 31.9465μs 31.3024 KOps/s 30.4099 KOps/s $\color{#35bf28}+2.93\%$
test_compile_indexing[tensor-tensorclass-compile] 88.2560μs 30.4005μs 32.8942 KOps/s 32.0794 KOps/s $\color{#35bf28}+2.54\%$
test_compile_indexing[tensor-tensorclass-eager] 78.8380μs 22.5219μs 44.4013 KOps/s 42.5995 KOps/s $\color{#35bf28}+4.23\%$
test_compile_indexing[tensor-pytree-compile] 87.5440μs 30.7976μs 32.4701 KOps/s 31.2246 KOps/s $\color{#35bf28}+3.99\%$
test_compile_indexing[tensor-pytree-eager] 72.6960μs 22.6793μs 44.0932 KOps/s 42.2946 KOps/s $\color{#35bf28}+4.25\%$
test_compile_indexing[slice-tensordict-compile] 0.1206ms 51.9955μs 19.2324 KOps/s 18.6318 KOps/s $\color{#35bf28}+3.22\%$
test_compile_indexing[slice-tensordict-eager] 0.3648ms 19.7396μs 50.6597 KOps/s 49.4469 KOps/s $\color{#35bf28}+2.45\%$
test_compile_indexing[slice-tensorclass-compile] 0.1416ms 44.9019μs 22.2708 KOps/s 21.6021 KOps/s $\color{#35bf28}+3.10\%$
test_compile_indexing[slice-tensorclass-eager] 57.8380μs 18.4520μs 54.1947 KOps/s 52.3740 KOps/s $\color{#35bf28}+3.48\%$
test_compile_indexing[slice-pytree-compile] 0.1497ms 47.0913μs 21.2354 KOps/s 21.4586 KOps/s $\color{#d91a1a}-1.04\%$
test_compile_indexing[slice-pytree-eager] 75.0610μs 18.2443μs 54.8118 KOps/s 52.4465 KOps/s $\color{#35bf28}+4.51\%$
test_compile_indexing[int-tensordict-compile] 0.1497ms 53.2861μs 18.7666 KOps/s 18.2902 KOps/s $\color{#35bf28}+2.60\%$
test_compile_indexing[int-tensordict-eager] 1.0114ms 19.8180μs 50.4593 KOps/s 49.2241 KOps/s $\color{#35bf28}+2.51\%$
test_compile_indexing[int-tensorclass-compile] 0.1087ms 45.3089μs 22.0707 KOps/s 20.9497 KOps/s $\textbf{\color{#35bf28}+5.35\%}$
test_compile_indexing[int-tensorclass-eager] 74.0180μs 18.1689μs 55.0391 KOps/s 52.7586 KOps/s $\color{#35bf28}+4.32\%$
test_compile_indexing[int-pytree-compile] 96.5510μs 45.3753μs 22.0384 KOps/s 21.1658 KOps/s $\color{#35bf28}+4.12\%$
test_compile_indexing[int-pytree-eager] 0.1086ms 18.3031μs 54.6355 KOps/s 52.3058 KOps/s $\color{#35bf28}+4.45\%$
test_mod_add[eager] 89.7080μs 35.6324μs 28.0643 KOps/s 26.7223 KOps/s $\textbf{\color{#35bf28}+5.02\%}$
test_mod_add[compile] 0.1165ms 62.1524μs 16.0895 KOps/s 14.5598 KOps/s $\textbf{\color{#35bf28}+10.51\%}$
test_mod_add[compile-overhead] 0.1255ms 61.5806μs 16.2389 KOps/s 14.5999 KOps/s $\textbf{\color{#35bf28}+11.23\%}$
test_mod_wrap[eager] 0.4313ms 0.2199ms 4.5465 KOps/s 4.3519 KOps/s $\color{#35bf28}+4.47\%$
test_mod_wrap[compile] 1.8640ms 0.2160ms 4.6290 KOps/s 4.3413 KOps/s $\textbf{\color{#35bf28}+6.63\%}$
test_mod_wrap[compile-overhead] 0.4150ms 0.2159ms 4.6321 KOps/s 4.4243 KOps/s $\color{#35bf28}+4.70\%$
test_mod_wrap_and_backward[eager] 16.2665ms 13.1961ms 75.7801 Ops/s 88.4217 Ops/s $\textbf{\color{#d91a1a}-14.30\%}$
test_mod_wrap_and_backward[compile] 13.5411ms 11.7239ms 85.2962 Ops/s 87.4091 Ops/s $\color{#d91a1a}-2.42\%$
test_mod_wrap_and_backward[compile-overhead] 22.5302ms 12.1873ms 82.0528 Ops/s 84.7142 Ops/s $\color{#d91a1a}-3.14\%$
test_seq_add[eager] 0.2713ms 0.1192ms 8.3877 KOps/s 8.4200 KOps/s $\color{#d91a1a}-0.38\%$
test_seq_add[compile] 0.1482ms 74.9893μs 13.3352 KOps/s 12.9700 KOps/s $\color{#35bf28}+2.82\%$
test_seq_add[compile-overhead] 0.1395ms 74.3756μs 13.4453 KOps/s 13.3541 KOps/s $\color{#35bf28}+0.68\%$
test_seq_wrap[eager] 0.6804ms 0.4427ms 2.2591 KOps/s 2.2047 KOps/s $\color{#35bf28}+2.47\%$
test_seq_wrap[compile] 0.8575ms 0.2358ms 4.2411 KOps/s 4.1004 KOps/s $\color{#35bf28}+3.43\%$
test_seq_wrap[compile-overhead] 0.3695ms 0.2348ms 4.2592 KOps/s 4.0987 KOps/s $\color{#35bf28}+3.92\%$
test_func_call_runtime[False-eager] 0.8493ms 0.5360ms 1.8657 KOps/s 1.8201 KOps/s $\color{#35bf28}+2.50\%$
test_func_call_runtime[False-compile] 0.5337ms 0.4367ms 2.2900 KOps/s 2.2561 KOps/s $\color{#35bf28}+1.50\%$
test_func_call_runtime[False-compile-overhead] 0.5693ms 0.4370ms 2.2882 KOps/s 2.2300 KOps/s $\color{#35bf28}+2.61\%$
test_func_call_runtime[True-eager] 0.9265ms 0.7460ms 1.3405 KOps/s 1.3075 KOps/s $\color{#35bf28}+2.52\%$
test_func_call_runtime[True-compile] 0.5580ms 0.4567ms 2.1898 KOps/s 2.1518 KOps/s $\color{#35bf28}+1.76\%$
test_func_call_runtime[True-compile-overhead] 0.5667ms 0.4557ms 2.1944 KOps/s 2.1225 KOps/s $\color{#35bf28}+3.39\%$
test_func_call_cm_runtime[False-eager] 0.8138ms 0.5311ms 1.8828 KOps/s 1.8352 KOps/s $\color{#35bf28}+2.59\%$
test_func_call_cm_runtime[False-compile] 0.5441ms 0.4350ms 2.2989 KOps/s 2.2510 KOps/s $\color{#35bf28}+2.13\%$
test_func_call_cm_runtime[False-compile-overhead] 0.5447ms 0.4364ms 2.2916 KOps/s 2.2079 KOps/s $\color{#35bf28}+3.79\%$
test_func_call_cm_runtime[True-eager] 1.0067ms 0.8770ms 1.1403 KOps/s 1.0918 KOps/s $\color{#35bf28}+4.44\%$
test_func_call_cm_runtime[True-compile] 0.9365ms 0.7851ms 1.2738 KOps/s 1.2216 KOps/s $\color{#35bf28}+4.27\%$
test_func_call_cm_runtime[True-compile-overhead] 0.9472ms 0.7852ms 1.2736 KOps/s 1.2385 KOps/s $\color{#35bf28}+2.84\%$
test_vmap_func_call_cm_runtime[eager] 2.6894ms 1.8871ms 529.9111 Ops/s 522.6864 Ops/s $\color{#35bf28}+1.38\%$
test_vmap_func_call_cm_runtime[compile] 1.0445ms 0.5344ms 1.8712 KOps/s 1.7830 KOps/s $\color{#35bf28}+4.95\%$
test_vmap_func_call_cm_runtime[compile-overhead] 0.8667ms 0.5299ms 1.8872 KOps/s 1.8268 KOps/s $\color{#35bf28}+3.30\%$
test_distributed 0.2597ms 0.1255ms 7.9675 KOps/s 7.7825 KOps/s $\color{#35bf28}+2.38\%$
test_tdmodule 46.5970μs 27.0341μs 36.9903 KOps/s 37.1967 KOps/s $\color{#d91a1a}-0.55\%$
test_tdmodule_dispatch 92.1620μs 52.8496μs 18.9216 KOps/s 20.2775 KOps/s $\textbf{\color{#d91a1a}-6.69\%}$
test_tdseq 51.1650μs 27.9944μs 35.7215 KOps/s 34.7696 KOps/s $\color{#35bf28}+2.74\%$
test_tdseq_dispatch 81.3620μs 52.5041μs 19.0461 KOps/s 18.2825 KOps/s $\color{#35bf28}+4.18\%$
test_instantiation_functorch 1.9799ms 1.5222ms 656.9588 Ops/s 656.9803 Ops/s $-0.00\%$
test_exec_functorch 0.2972ms 0.1736ms 5.7593 KOps/s 5.4718 KOps/s $\textbf{\color{#35bf28}+5.25\%}$
test_exec_functional_call 0.2880ms 0.1707ms 5.8583 KOps/s 5.7716 KOps/s $\color{#35bf28}+1.50\%$
test_exec_td_decorator 0.4717ms 0.2277ms 4.3909 KOps/s 4.3038 KOps/s $\color{#35bf28}+2.02\%$
test_vmap_mlp_speed_decorator[True-True] 0.8882ms 0.6545ms 1.5278 KOps/s 1.5297 KOps/s $\color{#d91a1a}-0.12\%$
test_vmap_mlp_speed_decorator[True-False] 1.0365ms 0.6529ms 1.5315 KOps/s 1.5293 KOps/s $\color{#35bf28}+0.15\%$
test_vmap_mlp_speed_decorator[False-True] 0.6654ms 0.5258ms 1.9019 KOps/s 1.9034 KOps/s $\color{#d91a1a}-0.08\%$
test_vmap_mlp_speed_decorator[False-False] 0.8412ms 0.5281ms 1.8935 KOps/s 1.8938 KOps/s $\color{#d91a1a}-0.02\%$
test_to_module_speed[True] 1.8470ms 1.3131ms 761.5423 Ops/s 737.3298 Ops/s $\color{#35bf28}+3.28\%$
test_to_module_speed[False] 2.2713ms 1.2794ms 781.6151 Ops/s 763.2454 Ops/s $\color{#35bf28}+2.41\%$
test_tc_init 85.1400μs 46.4687μs 21.5199 KOps/s 21.3473 KOps/s $\color{#35bf28}+0.81\%$
test_tc_init_nested 0.1562ms 93.1025μs 10.7408 KOps/s 10.6247 KOps/s $\color{#35bf28}+1.09\%$
test_tc_first_layer_tensor 20.9990μs 1.5611μs 640.5765 KOps/s 626.8905 KOps/s $\color{#35bf28}+2.18\%$
test_tc_first_layer_nontensor 18.6350μs 4.8135μs 207.7472 KOps/s 210.5283 KOps/s $\color{#d91a1a}-1.32\%$
test_tc_second_layer_tensor 23.8750μs 2.9128μs 343.3079 KOps/s 343.1805 KOps/s $\color{#35bf28}+0.04\%$
test_tc_second_layer_nontensor 26.1290μs 6.1651μs 162.2035 KOps/s 163.7507 KOps/s $\color{#d91a1a}-0.94\%$
test_unbind 0.2222s 11.6824ms 85.5985 Ops/s 66.8411 Ops/s $\textbf{\color{#35bf28}+28.06\%}$
test_full_like 9.3666ms 8.6465ms 115.6537 Ops/s 129.1693 Ops/s $\textbf{\color{#d91a1a}-10.46\%}$
test_zeros_like 4.4818ms 2.7154ms 368.2755 Ops/s 214.6220 Ops/s $\textbf{\color{#35bf28}+71.59\%}$
test_ones_like 0.1977s 4.1609ms 240.3331 Ops/s 188.0417 Ops/s $\textbf{\color{#35bf28}+27.81\%}$
test_clone 6.7673ms 5.3468ms 187.0269 Ops/s 142.9239 Ops/s $\textbf{\color{#35bf28}+30.86\%}$
test_squeeze 63.2180μs 12.9131μs 77.4409 KOps/s 78.9136 KOps/s $\color{#d91a1a}-1.87\%$
test_unsqueeze 0.1667ms 93.5494μs 10.6895 KOps/s 10.8810 KOps/s $\color{#d91a1a}-1.76\%$
test_split 0.4465ms 0.1960ms 5.1018 KOps/s 5.2380 KOps/s $\color{#d91a1a}-2.60\%$
test_permute 0.3129ms 0.2018ms 4.9554 KOps/s 5.0477 KOps/s $\color{#d91a1a}-1.83\%$
test_stack 41.9423ms 26.5690ms 37.6378 Ops/s 39.9268 Ops/s $\textbf{\color{#d91a1a}-5.73\%}$
test_cat 30.1328ms 25.4868ms 39.2360 Ops/s 39.6862 Ops/s $\color{#d91a1a}-1.13\%$

@vmoens vmoens added bug Something isn't working suitable for minor labels Feb 26, 2025
@vmoens vmoens merged commit f60dbc8 into gh/vmoens/50/base Feb 26, 2025
52 of 53 checks passed
vmoens added a commit that referenced this pull request Feb 26, 2025
ghstack-source-id: cfed094e425a60c62617ecfa454d3104ff1f461c
Pull Request resolved: #1239
@vmoens vmoens deleted the gh/vmoens/50/head branch February 26, 2025 11:03
vmoens added a commit that referenced this pull request Feb 26, 2025
ghstack-source-id: cfed094e425a60c62617ecfa454d3104ff1f461c
Pull Request resolved: #1239

(cherry picked from commit 2ad9f95)
# for free to join this conversation on GitHub. Already have an account? # to comment
Labels
bug Something isn't working CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. suitable for minor
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants