Skip to content
New issue

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

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

Already on GitHub? Sign in to your account

[BugFix] Fix pad_sequence for non tensors #784

Merged
merged 4 commits into from
May 22, 2024
Merged

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented May 20, 2024

No description provided.

@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 May 20, 2024
@vmoens vmoens added the bug Something isn't working label May 20, 2024
Copy link

github-actions bot commented May 20, 2024

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

Total Benchmarks: 127. Improved: $\large\color{#35bf28}18$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 41.4270μs 16.5839μs 60.2993 KOps/s 58.6601 KOps/s $\color{#35bf28}+2.79\%$
test_plain_set_stack_nested 53.6090μs 16.7123μs 59.8361 KOps/s 58.0138 KOps/s $\color{#35bf28}+3.14\%$
test_plain_set_nested_inplace 47.2280μs 18.8578μs 53.0284 KOps/s 51.7396 KOps/s $\color{#35bf28}+2.49\%$
test_plain_set_stack_nested_inplace 62.3060μs 18.7433μs 53.3523 KOps/s 51.8974 KOps/s $\color{#35bf28}+2.80\%$
test_items 16.1200μs 2.5791μs 387.7372 KOps/s 385.4395 KOps/s $\color{#35bf28}+0.60\%$
test_items_nested 0.9319ms 0.2640ms 3.7879 KOps/s 3.7595 KOps/s $\color{#35bf28}+0.75\%$
test_items_nested_locked 0.4226ms 0.2626ms 3.8084 KOps/s 3.7345 KOps/s $\color{#35bf28}+1.98\%$
test_items_nested_leaf 0.1344ms 76.9246μs 12.9997 KOps/s 13.1181 KOps/s $\color{#d91a1a}-0.90\%$
test_items_stack_nested 0.4978ms 0.2658ms 3.7625 KOps/s 3.7280 KOps/s $\color{#35bf28}+0.93\%$
test_items_stack_nested_leaf 0.1306ms 76.0439μs 13.1503 KOps/s 13.0301 KOps/s $\color{#35bf28}+0.92\%$
test_items_stack_nested_locked 0.5122ms 0.2673ms 3.7407 KOps/s 3.7216 KOps/s $\color{#35bf28}+0.51\%$
test_keys 23.2930μs 4.0306μs 248.1031 KOps/s 262.9656 KOps/s $\textbf{\color{#d91a1a}-5.65\%}$
test_keys_nested 0.2264ms 0.1398ms 7.1530 KOps/s 7.2834 KOps/s $\color{#d91a1a}-1.79\%$
test_keys_nested_locked 0.7618ms 0.1444ms 6.9262 KOps/s 7.0195 KOps/s $\color{#d91a1a}-1.33\%$
test_keys_nested_leaf 0.2045ms 0.1180ms 8.4773 KOps/s 8.5278 KOps/s $\color{#d91a1a}-0.59\%$
test_keys_stack_nested 0.2397ms 0.1361ms 7.3484 KOps/s 7.2379 KOps/s $\color{#35bf28}+1.53\%$
test_keys_stack_nested_leaf 0.1627ms 0.1145ms 8.7373 KOps/s 8.5426 KOps/s $\color{#35bf28}+2.28\%$
test_keys_stack_nested_locked 0.3503ms 0.1399ms 7.1457 KOps/s 7.0427 KOps/s $\color{#35bf28}+1.46\%$
test_values 6.9378μs 1.1917μs 839.1267 KOps/s 868.9058 KOps/s $\color{#d91a1a}-3.43\%$
test_values_nested 95.9290μs 50.4495μs 19.8218 KOps/s 19.7884 KOps/s $\color{#35bf28}+0.17\%$
test_values_nested_locked 0.1049ms 49.8933μs 20.0428 KOps/s 19.7521 KOps/s $\color{#35bf28}+1.47\%$
test_values_nested_leaf 81.0710μs 45.1099μs 22.1681 KOps/s 21.8005 KOps/s $\color{#35bf28}+1.69\%$
test_values_stack_nested 0.1364ms 51.7092μs 19.3389 KOps/s 19.3973 KOps/s $\color{#d91a1a}-0.30\%$
test_values_stack_nested_leaf 93.3140μs 44.7125μs 22.3651 KOps/s 21.8066 KOps/s $\color{#35bf28}+2.56\%$
test_values_stack_nested_locked 91.3800μs 51.5486μs 19.3992 KOps/s 19.5494 KOps/s $\color{#d91a1a}-0.77\%$
test_membership 14.1260μs 1.3363μs 748.3570 KOps/s 743.1071 KOps/s $\color{#35bf28}+0.71\%$
test_membership_nested 22.0010μs 3.3867μs 295.2709 KOps/s 296.5865 KOps/s $\color{#d91a1a}-0.44\%$
test_membership_nested_leaf 20.9990μs 3.4221μs 292.2214 KOps/s 291.5343 KOps/s $\color{#35bf28}+0.24\%$
test_membership_stacked_nested 21.0090μs 3.3987μs 294.2281 KOps/s 262.8896 KOps/s $\textbf{\color{#35bf28}+11.92\%}$
test_membership_stacked_nested_leaf 21.3890μs 3.4500μs 289.8513 KOps/s 294.3817 KOps/s $\color{#d91a1a}-1.54\%$
test_membership_nested_last 21.9410μs 4.2112μs 237.4620 KOps/s 242.0584 KOps/s $\color{#d91a1a}-1.90\%$
test_membership_nested_leaf_last 33.0010μs 4.2027μs 237.9446 KOps/s 241.0820 KOps/s $\color{#d91a1a}-1.30\%$
test_membership_stacked_nested_last 50.6540μs 13.3075μs 75.1458 KOps/s 207.6671 KOps/s $\textbf{\color{#d91a1a}-63.81\%}$
test_membership_stacked_nested_leaf_last 39.0130μs 13.2489μs 75.4781 KOps/s 208.6986 KOps/s $\textbf{\color{#d91a1a}-63.83\%}$
test_nested_getleaf 43.6210μs 10.5737μs 94.5744 KOps/s 94.1090 KOps/s $\color{#35bf28}+0.49\%$
test_nested_get 30.9380μs 10.0435μs 99.5672 KOps/s 99.8352 KOps/s $\color{#d91a1a}-0.27\%$
test_stacked_getleaf 34.5140μs 10.5656μs 94.6467 KOps/s 93.8753 KOps/s $\color{#35bf28}+0.82\%$
test_stacked_get 32.2300μs 9.7752μs 102.3001 KOps/s 99.8682 KOps/s $\color{#35bf28}+2.44\%$
test_nested_getitemleaf 61.3540μs 11.1459μs 89.7195 KOps/s 89.1397 KOps/s $\color{#35bf28}+0.65\%$
test_nested_getitem 61.0330μs 10.2224μs 97.8241 KOps/s 95.7708 KOps/s $\color{#35bf28}+2.14\%$
test_stacked_getitemleaf 46.5870μs 10.9610μs 91.2327 KOps/s 88.6351 KOps/s $\color{#35bf28}+2.93\%$
test_stacked_getitem 28.7940μs 10.1625μs 98.4011 KOps/s 96.3472 KOps/s $\color{#35bf28}+2.13\%$
test_lock_nested 48.0080ms 0.3925ms 2.5475 KOps/s 2.9038 KOps/s $\textbf{\color{#d91a1a}-12.27\%}$
test_lock_stack_nested 0.4794ms 0.2937ms 3.4050 KOps/s 3.2487 KOps/s $\color{#35bf28}+4.81\%$
test_unlock_nested 0.7935ms 0.3442ms 2.9056 KOps/s 2.5285 KOps/s $\textbf{\color{#35bf28}+14.91\%}$
test_unlock_stack_nested 0.4595ms 0.3008ms 3.3248 KOps/s 3.1529 KOps/s $\textbf{\color{#35bf28}+5.45\%}$
test_flatten_speed 0.1897ms 96.0407μs 10.4123 KOps/s 10.4868 KOps/s $\color{#d91a1a}-0.71\%$
test_unflatten_speed 0.7126ms 0.4023ms 2.4860 KOps/s 2.4500 KOps/s $\color{#35bf28}+1.47\%$
test_common_ops 5.1630ms 0.6884ms 1.4527 KOps/s 1.4178 KOps/s $\color{#35bf28}+2.46\%$
test_creation 49.9630μs 1.8526μs 539.7828 KOps/s 538.9689 KOps/s $\color{#35bf28}+0.15\%$
test_creation_empty 30.1560μs 9.7947μs 102.0965 KOps/s 89.6973 KOps/s $\textbf{\color{#35bf28}+13.82\%}$
test_creation_nested_1 35.4950μs 12.3196μs 81.1713 KOps/s 71.9790 KOps/s $\textbf{\color{#35bf28}+12.77\%}$
test_creation_nested_2 38.8720μs 15.7468μs 63.5050 KOps/s 57.4940 KOps/s $\textbf{\color{#35bf28}+10.46\%}$
test_clone 0.1169ms 13.5278μs 73.9221 KOps/s 75.2015 KOps/s $\color{#d91a1a}-1.70\%$
test_getitem[int] 38.5310μs 11.6206μs 86.0544 KOps/s 87.9922 KOps/s $\color{#d91a1a}-2.20\%$
test_getitem[slice_int] 56.5750μs 23.6937μs 42.2053 KOps/s 43.0103 KOps/s $\color{#d91a1a}-1.87\%$
test_getitem[range] 81.4920μs 58.6472μs 17.0511 KOps/s 17.2838 KOps/s $\color{#d91a1a}-1.35\%$
test_getitem[tuple] 45.4950μs 19.0705μs 52.4369 KOps/s 53.2232 KOps/s $\color{#d91a1a}-1.48\%$
test_getitem[list] 0.1485ms 40.4691μs 24.7102 KOps/s 24.8424 KOps/s $\color{#d91a1a}-0.53\%$
test_setitem_dim[int] 56.2340μs 33.9201μs 29.4810 KOps/s 28.2137 KOps/s $\color{#35bf28}+4.49\%$
test_setitem_dim[slice_int] 0.1093ms 60.5899μs 16.5044 KOps/s 15.4807 KOps/s $\textbf{\color{#35bf28}+6.61\%}$
test_setitem_dim[range] 0.1452ms 82.7129μs 12.0900 KOps/s 11.6203 KOps/s $\color{#35bf28}+4.04\%$
test_setitem_dim[tuple] 76.7330μs 49.1906μs 20.3291 KOps/s 19.9443 KOps/s $\color{#35bf28}+1.93\%$
test_setitem 55.8840μs 19.8271μs 50.4359 KOps/s 50.0629 KOps/s $\color{#35bf28}+0.75\%$
test_set 59.5610μs 19.2556μs 51.9329 KOps/s 50.9882 KOps/s $\color{#35bf28}+1.85\%$
test_set_shared 3.7187ms 0.1393ms 7.1773 KOps/s 7.1889 KOps/s $\color{#d91a1a}-0.16\%$
test_update 81.2310μs 20.7565μs 48.1777 KOps/s 45.7135 KOps/s $\textbf{\color{#35bf28}+5.39\%}$
test_update_nested 70.4610μs 29.2130μs 34.2313 KOps/s 33.6690 KOps/s $\color{#35bf28}+1.67\%$
test_update__nested 71.9740μs 25.1396μs 39.7779 KOps/s 40.0001 KOps/s $\color{#d91a1a}-0.56\%$
test_set_nested 55.2220μs 20.9520μs 47.7281 KOps/s 46.8391 KOps/s $\color{#35bf28}+1.90\%$
test_set_nested_new 75.9110μs 24.9706μs 40.0472 KOps/s 39.6205 KOps/s $\color{#35bf28}+1.08\%$
test_select 88.8660μs 39.6683μs 25.2091 KOps/s 24.8136 KOps/s $\color{#35bf28}+1.59\%$
test_select_nested 0.1364ms 58.3863μs 17.1273 KOps/s 16.6219 KOps/s $\color{#35bf28}+3.04\%$
test_exclude_nested 0.3028ms 0.1216ms 8.2256 KOps/s 8.2290 KOps/s $\color{#d91a1a}-0.04\%$
test_empty[True] 0.7086ms 0.3957ms 2.5274 KOps/s 2.5585 KOps/s $\color{#d91a1a}-1.21\%$
test_empty[False] 5.1314μs 1.0633μs 940.5026 KOps/s 917.7605 KOps/s $\color{#35bf28}+2.48\%$
test_unbind_speed 0.5086ms 0.2588ms 3.8640 KOps/s 3.8708 KOps/s $\color{#d91a1a}-0.18\%$
test_unbind_speed_stack0 0.3351ms 0.2420ms 4.1322 KOps/s 3.9008 KOps/s $\textbf{\color{#35bf28}+5.93\%}$
test_unbind_speed_stack1 62.5120ms 0.7230ms 1.3831 KOps/s 1.2848 KOps/s $\textbf{\color{#35bf28}+7.65\%}$
test_split 66.5642ms 1.6178ms 618.1132 Ops/s 628.0369 Ops/s $\color{#d91a1a}-1.58\%$
test_chunk 2.3476ms 1.5182ms 658.6960 Ops/s 628.6454 Ops/s $\color{#35bf28}+4.78\%$
test_creation[device0] 0.1803ms 80.8923μs 12.3621 KOps/s 11.9449 KOps/s $\color{#35bf28}+3.49\%$
test_creation_from_tensor 0.2036ms 83.9960μs 11.9053 KOps/s 11.7949 KOps/s $\color{#35bf28}+0.94\%$
test_add_one[memmap_tensor0] 73.7070μs 5.4895μs 182.1644 KOps/s 181.4445 KOps/s $\color{#35bf28}+0.40\%$
test_contiguous[memmap_tensor0] 17.0810μs 0.6429μs 1.5555 MOps/s 1.4276 MOps/s $\textbf{\color{#35bf28}+8.96\%}$
test_stack[memmap_tensor0] 27.9020μs 3.5856μs 278.8937 KOps/s 283.2376 KOps/s $\color{#d91a1a}-1.53\%$
test_memmaptd_index 63.3772ms 0.2802ms 3.5690 KOps/s 3.9408 KOps/s $\textbf{\color{#d91a1a}-9.44\%}$
test_memmaptd_index_astensor 0.7293ms 0.3172ms 3.1523 KOps/s 3.0271 KOps/s $\color{#35bf28}+4.14\%$
test_memmaptd_index_op 1.1216ms 0.5917ms 1.6900 KOps/s 1.6056 KOps/s $\textbf{\color{#35bf28}+5.26\%}$
test_serialize_model 0.1633s 0.1065s 9.3876 Ops/s 8.6126 Ops/s $\textbf{\color{#35bf28}+9.00\%}$
test_serialize_model_pickle 0.4480s 0.3789s 2.6391 Ops/s 2.6360 Ops/s $\color{#35bf28}+0.12\%$
test_serialize_weights 0.1686s 0.1058s 9.4540 Ops/s 8.8015 Ops/s $\textbf{\color{#35bf28}+7.41\%}$
test_serialize_weights_returnearly 0.1790s 0.1290s 7.7503 Ops/s 7.9384 Ops/s $\color{#d91a1a}-2.37\%$
test_serialize_weights_pickle 1.1855s 0.7338s 1.3628 Ops/s 1.3375 Ops/s $\color{#35bf28}+1.90\%$
test_serialize_weights_filesystem 98.8958ms 90.9459ms 10.9956 Ops/s 10.8555 Ops/s $\color{#35bf28}+1.29\%$
test_serialize_model_filesystem 99.7521ms 94.5555ms 10.5758 Ops/s 9.7759 Ops/s $\textbf{\color{#35bf28}+8.18\%}$
test_reshape_pytree 55.4540μs 25.2699μs 39.5728 KOps/s 38.9944 KOps/s $\color{#35bf28}+1.48\%$
test_reshape_td 70.8010μs 32.7091μs 30.5725 KOps/s 29.5404 KOps/s $\color{#35bf28}+3.49\%$
test_view_pytree 78.3760μs 25.0356μs 39.9431 KOps/s 38.6643 KOps/s $\color{#35bf28}+3.31\%$
test_view_td 76.1720μs 36.2883μs 27.5571 KOps/s 26.3687 KOps/s $\color{#35bf28}+4.51\%$
test_unbind_pytree 0.1017ms 29.1834μs 34.2661 KOps/s 34.3618 KOps/s $\color{#d91a1a}-0.28\%$
test_unbind_td 0.3740ms 37.5172μs 26.6544 KOps/s 26.0736 KOps/s $\color{#35bf28}+2.23\%$
test_split_pytree 66.2930μs 29.2206μs 34.2224 KOps/s 33.9403 KOps/s $\color{#35bf28}+0.83\%$
test_split_td 0.5555ms 40.9477μs 24.4214 KOps/s 24.6081 KOps/s $\color{#d91a1a}-0.76\%$
test_add_pytree 0.1130ms 34.6527μs 28.8577 KOps/s 28.6990 KOps/s $\color{#35bf28}+0.55\%$
test_add_td 0.1232ms 52.6460μs 18.9948 KOps/s 17.5621 KOps/s $\textbf{\color{#35bf28}+8.16\%}$
test_distributed 0.2304ms 99.5343μs 10.0468 KOps/s 9.7380 KOps/s $\color{#35bf28}+3.17\%$
test_tdmodule 30.5760μs 17.0356μs 58.7005 KOps/s 57.2805 KOps/s $\color{#35bf28}+2.48\%$
test_tdmodule_dispatch 58.4590μs 34.1629μs 29.2715 KOps/s 29.0159 KOps/s $\color{#35bf28}+0.88\%$
test_tdseq 41.9980μs 20.0092μs 49.9770 KOps/s 47.9446 KOps/s $\color{#35bf28}+4.24\%$
test_tdseq_dispatch 70.7720μs 38.8218μs 25.7587 KOps/s 24.4214 KOps/s $\textbf{\color{#35bf28}+5.48\%}$
test_instantiation_functorch 2.1650ms 1.2983ms 770.2340 Ops/s 768.6132 Ops/s $\color{#35bf28}+0.21\%$
test_instantiation_td 1.6252ms 0.9996ms 1.0004 KOps/s 995.5265 Ops/s $\color{#35bf28}+0.49\%$
test_exec_functorch 0.2319ms 0.1608ms 6.2207 KOps/s 5.6822 KOps/s $\textbf{\color{#35bf28}+9.48\%}$
test_exec_functional_call 0.3535ms 0.1519ms 6.5831 KOps/s 6.6057 KOps/s $\color{#d91a1a}-0.34\%$
test_exec_td 0.2182ms 0.1455ms 6.8721 KOps/s 6.7536 KOps/s $\color{#35bf28}+1.75\%$
test_exec_td_decorator 0.8124ms 0.2191ms 4.5634 KOps/s 4.4889 KOps/s $\color{#35bf28}+1.66\%$
test_vmap_mlp_speed[True-True] 0.6857ms 0.4947ms 2.0214 KOps/s 2.0638 KOps/s $\color{#d91a1a}-2.05\%$
test_vmap_mlp_speed[True-False] 0.6044ms 0.4851ms 2.0615 KOps/s 2.0743 KOps/s $\color{#d91a1a}-0.61\%$
test_vmap_mlp_speed[False-True] 0.5124ms 0.3973ms 2.5172 KOps/s 2.5400 KOps/s $\color{#d91a1a}-0.90\%$
test_vmap_mlp_speed[False-False] 0.5710ms 0.3960ms 2.5255 KOps/s 2.5415 KOps/s $\color{#d91a1a}-0.63\%$
test_vmap_mlp_speed_decorator[True-True] 1.2717ms 0.5489ms 1.8217 KOps/s 1.8011 KOps/s $\color{#35bf28}+1.14\%$
test_vmap_mlp_speed_decorator[True-False] 0.9119ms 0.5508ms 1.8156 KOps/s 1.7986 KOps/s $\color{#35bf28}+0.95\%$
test_vmap_mlp_speed_decorator[False-True] 0.6511ms 0.4557ms 2.1944 KOps/s 2.2101 KOps/s $\color{#d91a1a}-0.71\%$
test_vmap_mlp_speed_decorator[False-False] 0.8385ms 0.4564ms 2.1909 KOps/s 2.1990 KOps/s $\color{#d91a1a}-0.37\%$
test_to_module_speed[True] 1.8473ms 1.6691ms 599.1171 Ops/s 576.1980 Ops/s $\color{#35bf28}+3.98\%$
test_to_module_speed[False] 1.7787ms 1.6364ms 611.0951 Ops/s 609.1181 Ops/s $\color{#35bf28}+0.32\%$

Copy link

github-actions bot commented May 20, 2024

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

Total Benchmarks: 135. Improved: $\large\color{#35bf28}25$. Worsened: $\large\color{#d91a1a}7$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 0.1180ms 11.7857μs 84.8484 KOps/s 75.4316 KOps/s $\textbf{\color{#35bf28}+12.48\%}$
test_plain_set_stack_nested 32.0410μs 11.8148μs 84.6397 KOps/s 74.7832 KOps/s $\textbf{\color{#35bf28}+13.18\%}$
test_plain_set_nested_inplace 0.3320ms 12.9376μs 77.2938 KOps/s 69.2309 KOps/s $\textbf{\color{#35bf28}+11.65\%}$
test_plain_set_stack_nested_inplace 36.0900μs 13.1758μs 75.8965 KOps/s 68.5403 KOps/s $\textbf{\color{#35bf28}+10.73\%}$
test_items 0.1955ms 4.6158μs 216.6478 KOps/s 211.2682 KOps/s $\color{#35bf28}+2.55\%$
test_items_nested 0.5251ms 0.3373ms 2.9648 KOps/s 2.9154 KOps/s $\color{#35bf28}+1.69\%$
test_items_nested_locked 0.5534ms 0.3413ms 2.9301 KOps/s 2.9128 KOps/s $\color{#35bf28}+0.59\%$
test_items_nested_leaf 0.2666ms 82.7038μs 12.0913 KOps/s 12.1912 KOps/s $\color{#d91a1a}-0.82\%$
test_items_stack_nested 0.5891ms 0.3392ms 2.9485 KOps/s 2.9136 KOps/s $\color{#35bf28}+1.20\%$
test_items_stack_nested_leaf 0.1137ms 85.6250μs 11.6788 KOps/s 11.8674 KOps/s $\color{#d91a1a}-1.59\%$
test_items_stack_nested_locked 0.5315ms 0.3430ms 2.9155 KOps/s 2.9136 KOps/s $\color{#35bf28}+0.06\%$
test_keys 0.2073ms 4.3266μs 231.1279 KOps/s 231.4414 KOps/s $\color{#d91a1a}-0.14\%$
test_keys_nested 0.2713ms 67.2050μs 14.8798 KOps/s 14.8590 KOps/s $\color{#35bf28}+0.14\%$
test_keys_nested_locked 0.8125ms 71.4109μs 14.0035 KOps/s 13.7120 KOps/s $\color{#35bf28}+2.13\%$
test_keys_nested_leaf 0.2463ms 57.0274μs 17.5354 KOps/s 17.2982 KOps/s $\color{#35bf28}+1.37\%$
test_keys_stack_nested 0.2562ms 66.5804μs 15.0194 KOps/s 14.9448 KOps/s $\color{#35bf28}+0.50\%$
test_keys_stack_nested_leaf 0.2582ms 57.5674μs 17.3710 KOps/s 17.2467 KOps/s $\color{#35bf28}+0.72\%$
test_keys_stack_nested_locked 0.2689ms 71.1388μs 14.0570 KOps/s 13.8561 KOps/s $\color{#35bf28}+1.45\%$
test_values 62.3447μs 1.8236μs 548.3530 KOps/s 550.5162 KOps/s $\color{#d91a1a}-0.39\%$
test_values_nested 0.2826ms 35.2665μs 28.3555 KOps/s 28.3697 KOps/s $\color{#d91a1a}-0.05\%$
test_values_nested_locked 0.2248ms 36.9901μs 27.0342 KOps/s 26.9004 KOps/s $\color{#35bf28}+0.50\%$
test_values_nested_leaf 49.6410μs 31.4137μs 31.8332 KOps/s 31.8625 KOps/s $\color{#d91a1a}-0.09\%$
test_values_stack_nested 0.2289ms 35.9203μs 27.8394 KOps/s 27.8918 KOps/s $\color{#d91a1a}-0.19\%$
test_values_stack_nested_leaf 66.1220μs 32.2252μs 31.0317 KOps/s 30.9666 KOps/s $\color{#35bf28}+0.21\%$
test_values_stack_nested_locked 0.2555ms 38.0606μs 26.2739 KOps/s 26.6536 KOps/s $\color{#d91a1a}-1.42\%$
test_membership 4.0800μs 0.7357μs 1.3593 MOps/s 1.4311 MOps/s $\textbf{\color{#d91a1a}-5.02\%}$
test_membership_nested 0.2027ms 2.5299μs 395.2657 KOps/s 390.5827 KOps/s $\color{#35bf28}+1.20\%$
test_membership_nested_leaf 34.9710μs 2.5426μs 393.2981 KOps/s 392.9659 KOps/s $\color{#35bf28}+0.08\%$
test_membership_stacked_nested 0.1965ms 2.5780μs 387.8966 KOps/s 389.5755 KOps/s $\color{#d91a1a}-0.43\%$
test_membership_stacked_nested_leaf 38.1610μs 2.5329μs 394.8047 KOps/s 388.1502 KOps/s $\color{#35bf28}+1.71\%$
test_membership_nested_last 0.2095ms 3.0777μs 324.9129 KOps/s 322.8446 KOps/s $\color{#35bf28}+0.64\%$
test_membership_nested_leaf_last 67.1810μs 3.0568μs 327.1361 KOps/s 326.9711 KOps/s $\color{#35bf28}+0.05\%$
test_membership_stacked_nested_last 27.1200μs 4.7907μs 208.7357 KOps/s 325.0082 KOps/s $\textbf{\color{#d91a1a}-35.78\%}$
test_membership_stacked_nested_leaf_last 22.0300μs 4.7933μs 208.6242 KOps/s 327.2492 KOps/s $\textbf{\color{#d91a1a}-36.25\%}$
test_nested_getleaf 0.1998ms 8.3120μs 120.3075 KOps/s 119.9847 KOps/s $\color{#35bf28}+0.27\%$
test_nested_get 45.3210μs 7.8645μs 127.1532 KOps/s 127.5676 KOps/s $\color{#d91a1a}-0.32\%$
test_stacked_getleaf 0.2199ms 8.3760μs 119.3888 KOps/s 119.7528 KOps/s $\color{#d91a1a}-0.30\%$
test_stacked_get 0.2162ms 7.8663μs 127.1241 KOps/s 127.2349 KOps/s $\color{#d91a1a}-0.09\%$
test_nested_getitemleaf 0.1912ms 8.4819μs 117.8985 KOps/s 117.3192 KOps/s $\color{#35bf28}+0.49\%$
test_nested_getitem 0.2056ms 7.9944μs 125.0876 KOps/s 124.8552 KOps/s $\color{#35bf28}+0.19\%$
test_stacked_getitemleaf 0.2193ms 8.5589μs 116.8371 KOps/s 117.1246 KOps/s $\color{#d91a1a}-0.25\%$
test_stacked_getitem 31.2000μs 8.0183μs 124.7147 KOps/s 124.8374 KOps/s $\color{#d91a1a}-0.10\%$
test_lock_nested 61.4312ms 0.4147ms 2.4115 KOps/s 2.3984 KOps/s $\color{#35bf28}+0.55\%$
test_lock_stack_nested 0.5160ms 0.3086ms 3.2409 KOps/s 3.2958 KOps/s $\color{#d91a1a}-1.67\%$
test_unlock_nested 0.7452ms 0.3595ms 2.7813 KOps/s 2.8739 KOps/s $\color{#d91a1a}-3.22\%$
test_unlock_stack_nested 0.4663ms 0.3161ms 3.1631 KOps/s 3.2116 KOps/s $\color{#d91a1a}-1.51\%$
test_flatten_speed 0.3695ms 0.1020ms 9.8043 KOps/s 9.7790 KOps/s $\color{#35bf28}+0.26\%$
test_unflatten_speed 0.4902ms 0.2907ms 3.4402 KOps/s 3.5108 KOps/s $\color{#d91a1a}-2.01\%$
test_common_ops 1.0026ms 0.5369ms 1.8626 KOps/s 1.7285 KOps/s $\textbf{\color{#35bf28}+7.76\%}$
test_creation 34.2710μs 1.6275μs 614.4362 KOps/s 610.1826 KOps/s $\color{#35bf28}+0.70\%$
test_creation_empty 23.9200μs 6.5864μs 151.8287 KOps/s 105.8219 KOps/s $\textbf{\color{#35bf28}+43.48\%}$
test_creation_nested_1 0.1849ms 8.4370μs 118.5259 KOps/s 89.2660 KOps/s $\textbf{\color{#35bf28}+32.78\%}$
test_creation_nested_2 0.2168ms 10.5485μs 94.8005 KOps/s 74.2135 KOps/s $\textbf{\color{#35bf28}+27.74\%}$
test_clone 0.1718ms 11.8234μs 84.5778 KOps/s 87.0414 KOps/s $\color{#d91a1a}-2.83\%$
test_getitem[int] 32.7900μs 11.2983μs 88.5085 KOps/s 93.1895 KOps/s $\textbf{\color{#d91a1a}-5.02\%}$
test_getitem[slice_int] 0.2226ms 21.1263μs 47.3344 KOps/s 49.1209 KOps/s $\color{#d91a1a}-3.64\%$
test_getitem[range] 62.6310μs 45.3410μs 22.0551 KOps/s 22.3979 KOps/s $\color{#d91a1a}-1.53\%$
test_getitem[tuple] 56.6810μs 19.2657μs 51.9056 KOps/s 54.2456 KOps/s $\color{#d91a1a}-4.31\%$
test_getitem[list] 0.2425ms 33.4160μs 29.9258 KOps/s 31.0679 KOps/s $\color{#d91a1a}-3.68\%$
test_setitem_dim[int] 43.3710μs 26.6809μs 37.4799 KOps/s 33.7923 KOps/s $\textbf{\color{#35bf28}+10.91\%}$
test_setitem_dim[slice_int] 76.2610μs 45.4707μs 21.9922 KOps/s 20.2155 KOps/s $\textbf{\color{#35bf28}+8.79\%}$
test_setitem_dim[range] 0.1576ms 62.2101μs 16.0746 KOps/s 15.1032 KOps/s $\textbf{\color{#35bf28}+6.43\%}$
test_setitem_dim[tuple] 61.4610μs 40.1784μs 24.8890 KOps/s 22.3806 KOps/s $\textbf{\color{#35bf28}+11.21\%}$
test_setitem 48.9110μs 15.7120μs 63.6458 KOps/s 60.5330 KOps/s $\textbf{\color{#35bf28}+5.14\%}$
test_set 0.2225ms 15.1777μs 65.8861 KOps/s 61.7270 KOps/s $\textbf{\color{#35bf28}+6.74\%}$
test_set_shared 76.5147ms 0.1133ms 8.8236 KOps/s 8.9960 KOps/s $\color{#d91a1a}-1.92\%$
test_update 0.2142ms 16.3986μs 60.9807 KOps/s 53.9156 KOps/s $\textbf{\color{#35bf28}+13.10\%}$
test_update_nested 0.2175ms 21.8975μs 45.6674 KOps/s 42.4959 KOps/s $\textbf{\color{#35bf28}+7.46\%}$
test_update__nested 0.1292ms 22.6505μs 44.1492 KOps/s 44.7623 KOps/s $\color{#d91a1a}-1.37\%$
test_set_nested 0.1131ms 16.1081μs 62.0806 KOps/s 58.3681 KOps/s $\textbf{\color{#35bf28}+6.36\%}$
test_set_nested_new 0.2223ms 18.7618μs 53.2997 KOps/s 51.1078 KOps/s $\color{#35bf28}+4.29\%$
test_select 91.0310μs 31.3071μs 31.9416 KOps/s 30.1293 KOps/s $\textbf{\color{#35bf28}+6.02\%}$
test_select_nested 0.6634ms 55.1468μs 18.1334 KOps/s 17.9634 KOps/s $\color{#35bf28}+0.95\%$
test_exclude_nested 0.3002ms 0.1138ms 8.7864 KOps/s 8.9642 KOps/s $\color{#d91a1a}-1.98\%$
test_empty[True] 0.4872ms 0.3561ms 2.8080 KOps/s 2.8569 KOps/s $\color{#d91a1a}-1.71\%$
test_empty[False] 19.2304μs 0.8742μs 1.1439 MOps/s 1.1492 MOps/s $\color{#d91a1a}-0.46\%$
test_to 0.1296ms 77.0816μs 12.9733 KOps/s 13.3823 KOps/s $\color{#d91a1a}-3.06\%$
test_to_nonblocking 0.2266ms 58.7991μs 17.0070 KOps/s 16.6472 KOps/s $\color{#35bf28}+2.16\%$
test_unbind_speed 1.6787ms 0.2770ms 3.6103 KOps/s 3.7073 KOps/s $\color{#d91a1a}-2.62\%$
test_unbind_speed_stack0 0.4908ms 0.2779ms 3.5982 KOps/s 3.7044 KOps/s $\color{#d91a1a}-2.87\%$
test_unbind_speed_stack1 80.7729ms 0.8263ms 1.2102 KOps/s 1.2192 KOps/s $\color{#d91a1a}-0.74\%$
test_split 2.3171ms 1.5821ms 632.0787 Ops/s 655.7112 Ops/s $\color{#d91a1a}-3.60\%$
test_chunk 79.8232ms 1.7017ms 587.6585 Ops/s 604.1492 Ops/s $\color{#d91a1a}-2.73\%$
test_creation[device0] 0.2952ms 56.6279μs 17.6591 KOps/s 18.2297 KOps/s $\color{#d91a1a}-3.13\%$
test_creation_from_tensor 0.1950ms 53.7169μs 18.6161 KOps/s 18.9820 KOps/s $\color{#d91a1a}-1.93\%$
test_add_one[memmap_tensor0] 75.7820μs 6.9101μs 144.7148 KOps/s 153.1289 KOps/s $\textbf{\color{#d91a1a}-5.49\%}$
test_contiguous[memmap_tensor0] 24.0500μs 0.6317μs 1.5829 MOps/s 1.5915 MOps/s $\color{#d91a1a}-0.54\%$
test_stack[memmap_tensor0] 0.2069ms 4.8819μs 204.8366 KOps/s 221.7851 KOps/s $\textbf{\color{#d91a1a}-7.64\%}$
test_memmaptd_index 1.3253ms 0.2925ms 3.4188 KOps/s 3.4523 KOps/s $\color{#d91a1a}-0.97\%$
test_memmaptd_index_astensor 0.7014ms 0.3653ms 2.7374 KOps/s 2.7642 KOps/s $\color{#d91a1a}-0.97\%$
test_memmaptd_index_op 1.1454ms 0.6259ms 1.5976 KOps/s 1.5019 KOps/s $\textbf{\color{#35bf28}+6.37\%}$
test_serialize_model 0.1888s 0.1102s 9.0749 Ops/s 8.5527 Ops/s $\textbf{\color{#35bf28}+6.11\%}$
test_serialize_model_pickle 1.3515s 1.2358s 0.8092 Ops/s 0.8080 Ops/s $\color{#35bf28}+0.15\%$
test_serialize_weights 0.1887s 0.1068s 9.3670 Ops/s 9.5589 Ops/s $\color{#d91a1a}-2.01\%$
test_serialize_weights_returnearly 0.1833s 87.8276ms 11.3859 Ops/s 10.8706 Ops/s $\color{#35bf28}+4.74\%$
test_serialize_weights_pickle 1.3593s 1.2528s 0.7982 Ops/s 0.8087 Ops/s $\color{#d91a1a}-1.29\%$
test_reshape_pytree 0.1637ms 23.5708μs 42.4253 KOps/s 43.6275 KOps/s $\color{#d91a1a}-2.76\%$
test_reshape_td 58.9510μs 31.4427μs 31.8039 KOps/s 33.2129 KOps/s $\color{#d91a1a}-4.24\%$
test_view_pytree 0.2426ms 23.6978μs 42.1980 KOps/s 43.5819 KOps/s $\color{#d91a1a}-3.18\%$
test_view_td 0.3461ms 37.0561μs 26.9861 KOps/s 29.6790 KOps/s $\textbf{\color{#d91a1a}-9.07\%}$
test_unbind_pytree 0.2691ms 29.4356μs 33.9724 KOps/s 34.7552 KOps/s $\color{#d91a1a}-2.25\%$
test_unbind_td 0.5002ms 41.6070μs 24.0344 KOps/s 24.4101 KOps/s $\color{#d91a1a}-1.54\%$
test_split_pytree 0.2322ms 31.3153μs 31.9333 KOps/s 32.0916 KOps/s $\color{#d91a1a}-0.49\%$
test_split_td 0.6403ms 40.0237μs 24.9852 KOps/s 25.8889 KOps/s $\color{#d91a1a}-3.49\%$
test_add_pytree 0.2570ms 33.9393μs 29.4644 KOps/s 30.4181 KOps/s $\color{#d91a1a}-3.14\%$
test_add_td 0.1413ms 45.8175μs 21.8257 KOps/s 19.1679 KOps/s $\textbf{\color{#35bf28}+13.87\%}$
test_distributed 1.7955ms 70.3940μs 14.2058 KOps/s 14.3070 KOps/s $\color{#d91a1a}-0.71\%$
test_tdmodule 35.3210μs 13.5688μs 73.6982 KOps/s 60.0874 KOps/s $\textbf{\color{#35bf28}+22.65\%}$
test_tdmodule_dispatch 98.7720μs 26.5388μs 37.6807 KOps/s 31.1749 KOps/s $\textbf{\color{#35bf28}+20.87\%}$
test_tdseq 99.8130μs 16.0005μs 62.4981 KOps/s 55.2995 KOps/s $\textbf{\color{#35bf28}+13.02\%}$
test_tdseq_dispatch 85.2520μs 29.6936μs 33.6773 KOps/s 29.1156 KOps/s $\textbf{\color{#35bf28}+15.67\%}$
test_instantiation_functorch 1.7355ms 1.5421ms 648.4873 Ops/s 654.5744 Ops/s $\color{#d91a1a}-0.93\%$
test_instantiation_td 80.3771ms 1.1571ms 864.2444 Ops/s 875.0616 Ops/s $\color{#d91a1a}-1.24\%$
test_exec_functorch 0.2811ms 0.1507ms 6.6348 KOps/s 6.8114 KOps/s $\color{#d91a1a}-2.59\%$
test_exec_functional_call 0.2769ms 0.1381ms 7.2411 KOps/s 7.5374 KOps/s $\color{#d91a1a}-3.93\%$
test_exec_td 0.2553ms 0.1362ms 7.3444 KOps/s 7.5824 KOps/s $\color{#d91a1a}-3.14\%$
test_exec_td_decorator 0.5013ms 0.2097ms 4.7682 KOps/s 4.8397 KOps/s $\color{#d91a1a}-1.48\%$
test_vmap_mlp_speed[True-True] 1.3380ms 0.5908ms 1.6926 KOps/s 1.7066 KOps/s $\color{#d91a1a}-0.82\%$
test_vmap_mlp_speed[True-False] 0.7499ms 0.5825ms 1.7167 KOps/s 1.7190 KOps/s $\color{#d91a1a}-0.13\%$
test_vmap_mlp_speed[False-True] 0.7170ms 0.5347ms 1.8704 KOps/s 1.9497 KOps/s $\color{#d91a1a}-4.07\%$
test_vmap_mlp_speed[False-False] 0.6835ms 0.5173ms 1.9331 KOps/s 1.9423 KOps/s $\color{#d91a1a}-0.48\%$
test_vmap_mlp_speed_decorator[True-True] 0.9819ms 0.6411ms 1.5598 KOps/s 1.5445 KOps/s $\color{#35bf28}+0.99\%$
test_vmap_mlp_speed_decorator[True-False] 0.8044ms 0.6406ms 1.5609 KOps/s 1.5450 KOps/s $\color{#35bf28}+1.03\%$
test_vmap_mlp_speed_decorator[False-True] 0.7575ms 0.5726ms 1.7465 KOps/s 1.7078 KOps/s $\color{#35bf28}+2.26\%$
test_vmap_mlp_speed_decorator[False-False] 0.8363ms 0.5877ms 1.7016 KOps/s 1.6982 KOps/s $\color{#35bf28}+0.19\%$
test_vmap_transformer_speed[True-True] 8.7055ms 8.0861ms 123.6684 Ops/s 127.2367 Ops/s $\color{#d91a1a}-2.80\%$
test_vmap_transformer_speed[True-False] 9.4322ms 7.9121ms 126.3884 Ops/s 129.4994 Ops/s $\color{#d91a1a}-2.40\%$
test_vmap_transformer_speed[False-True] 7.9702ms 7.7005ms 129.8620 Ops/s 130.5745 Ops/s $\color{#d91a1a}-0.55\%$
test_vmap_transformer_speed[False-False] 8.0039ms 7.7310ms 129.3498 Ops/s 130.2982 Ops/s $\color{#d91a1a}-0.73\%$
test_vmap_transformer_speed_decorator[True-True] 19.5075ms 18.8100ms 53.1631 Ops/s 52.9448 Ops/s $\color{#35bf28}+0.41\%$
test_vmap_transformer_speed_decorator[True-False] 19.2927ms 18.9983ms 52.6364 Ops/s 52.7389 Ops/s $\color{#d91a1a}-0.19\%$
test_vmap_transformer_speed_decorator[False-True] 19.5244ms 18.8288ms 53.1102 Ops/s 53.2177 Ops/s $\color{#d91a1a}-0.20\%$
test_vmap_transformer_speed_decorator[False-False] 19.4017ms 18.8468ms 53.0595 Ops/s 53.3147 Ops/s $\color{#d91a1a}-0.48\%$
test_to_module_speed[True] 1.6484ms 1.5157ms 659.7799 Ops/s 640.1558 Ops/s $\color{#35bf28}+3.07\%$
test_to_module_speed[False] 1.6266ms 1.5010ms 666.2043 Ops/s 651.1295 Ops/s $\color{#35bf28}+2.32\%$

@vmoens vmoens merged commit 3fa2224 into main May 22, 2024
33 of 37 checks passed
@vmoens vmoens deleted the fix-nontensor-pad_seq branch May 22, 2024 21:43
Sign up for free to join this conversation on GitHub. Already have an account? Sign in 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.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

[BUG] tensordict.pad_sequence silently ignores non-tensor attributes in tensorclasses or TensorDicts
2 participants