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 is_memmap attribute for memmap_like and memmap #625

Merged
merged 1 commit into from
Jan 17, 2024
Merged

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Jan 17, 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 Jan 17, 2024
@vmoens vmoens added the bug Something isn't working label Jan 17, 2024
@vmoens vmoens merged commit a9234fc into main Jan 17, 2024
28 of 32 checks passed
@vmoens vmoens deleted the fix-memmap branch January 17, 2024 16:32
Copy link

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

Total Benchmarks: 124. Improved: $\large\color{#35bf28}15$. Worsened: $\large\color{#d91a1a}5$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 34.4340μs 16.6617μs 60.0178 KOps/s 55.9725 KOps/s $\textbf{\color{#35bf28}+7.23\%}$
test_plain_set_stack_nested 0.1868ms 0.1424ms 7.0202 KOps/s 6.9070 KOps/s $\color{#35bf28}+1.64\%$
test_plain_set_nested_inplace 46.6470μs 19.3143μs 51.7750 KOps/s 48.5309 KOps/s $\textbf{\color{#35bf28}+6.68\%}$
test_plain_set_stack_nested_inplace 0.3083ms 0.1746ms 5.7289 KOps/s 5.6378 KOps/s $\color{#35bf28}+1.62\%$
test_items 13.5350μs 2.4385μs 410.0865 KOps/s 381.9471 KOps/s $\textbf{\color{#35bf28}+7.37\%}$
test_items_nested 1.3589ms 0.2697ms 3.7084 KOps/s 3.6617 KOps/s $\color{#35bf28}+1.28\%$
test_items_nested_locked 0.4423ms 0.2701ms 3.7024 KOps/s 3.6904 KOps/s $\color{#35bf28}+0.33\%$
test_items_nested_leaf 0.6718ms 0.1665ms 6.0045 KOps/s 5.9220 KOps/s $\color{#35bf28}+1.39\%$
test_items_stack_nested 1.9967ms 1.3056ms 765.9336 Ops/s 754.9595 Ops/s $\color{#35bf28}+1.45\%$
test_items_stack_nested_leaf 2.2811ms 1.2223ms 818.1294 Ops/s 841.3454 Ops/s $\color{#d91a1a}-2.76\%$
test_items_stack_nested_locked 1.4401ms 0.8783ms 1.1386 KOps/s 1.1375 KOps/s $\color{#35bf28}+0.09\%$
test_keys 19.4660μs 3.8389μs 260.4892 KOps/s 261.1619 KOps/s $\color{#d91a1a}-0.26\%$
test_keys_nested 47.6947ms 0.1570ms 6.3704 KOps/s 6.6824 KOps/s $\color{#d91a1a}-4.67\%$
test_keys_nested_locked 1.9651ms 0.1531ms 6.5296 KOps/s 6.5649 KOps/s $\color{#d91a1a}-0.54\%$
test_keys_nested_leaf 0.2637ms 0.1297ms 7.7087 KOps/s 7.6885 KOps/s $\color{#35bf28}+0.26\%$
test_keys_stack_nested 1.4932ms 1.2671ms 789.1735 Ops/s 784.4176 Ops/s $\color{#35bf28}+0.61\%$
test_keys_stack_nested_leaf 1.8152ms 1.2567ms 795.7620 Ops/s 790.5553 Ops/s $\color{#35bf28}+0.66\%$
test_keys_stack_nested_locked 1.2476ms 0.8210ms 1.2180 KOps/s 1.2407 KOps/s $\color{#d91a1a}-1.82\%$
test_values 4.9968μs 1.1622μs 860.4505 KOps/s 857.8881 KOps/s $\color{#35bf28}+0.30\%$
test_values_nested 0.1077ms 51.8876μs 19.2724 KOps/s 19.4522 KOps/s $\color{#d91a1a}-0.92\%$
test_values_nested_locked 92.6830μs 51.4664μs 19.4302 KOps/s 19.3546 KOps/s $\color{#35bf28}+0.39\%$
test_values_nested_leaf 87.3730μs 45.8966μs 21.7881 KOps/s 21.5177 KOps/s $\color{#35bf28}+1.26\%$
test_values_stack_nested 1.5814ms 1.0324ms 968.5861 Ops/s 950.7331 Ops/s $\color{#35bf28}+1.88\%$
test_values_stack_nested_leaf 1.2551ms 1.0184ms 981.9478 Ops/s 954.8843 Ops/s $\color{#35bf28}+2.83\%$
test_values_stack_nested_locked 1.1345ms 0.6145ms 1.6273 KOps/s 1.6417 KOps/s $\color{#d91a1a}-0.88\%$
test_membership 15.0480μs 1.3721μs 728.8240 KOps/s 730.1985 KOps/s $\color{#d91a1a}-0.19\%$
test_membership_nested 20.3880μs 3.4168μs 292.6679 KOps/s 278.6126 KOps/s $\textbf{\color{#35bf28}+5.04\%}$
test_membership_nested_leaf 21.1290μs 3.4952μs 286.1041 KOps/s 282.7801 KOps/s $\color{#35bf28}+1.18\%$
test_membership_stacked_nested 44.5530μs 11.7410μs 85.1714 KOps/s 84.8475 KOps/s $\color{#35bf28}+0.38\%$
test_membership_stacked_nested_leaf 47.0880μs 11.9289μs 83.8301 KOps/s 83.6621 KOps/s $\color{#35bf28}+0.20\%$
test_membership_nested_last 26.5890μs 6.6196μs 151.0672 KOps/s 149.5127 KOps/s $\color{#35bf28}+1.04\%$
test_membership_nested_leaf_last 34.2240μs 6.5400μs 152.9059 KOps/s 148.0255 KOps/s $\color{#35bf28}+3.30\%$
test_membership_stacked_nested_last 0.2450ms 0.1724ms 5.8003 KOps/s 5.7374 KOps/s $\color{#35bf28}+1.10\%$
test_membership_stacked_nested_leaf_last 37.1600μs 13.9517μs 71.6757 KOps/s 71.3327 KOps/s $\color{#35bf28}+0.48\%$
test_nested_getleaf 41.4680μs 10.6395μs 93.9894 KOps/s 90.8068 KOps/s $\color{#35bf28}+3.50\%$
test_nested_get 45.4750μs 10.0190μs 99.8108 KOps/s 99.5607 KOps/s $\color{#35bf28}+0.25\%$
test_stacked_getleaf 0.6823ms 0.3905ms 2.5607 KOps/s 2.5374 KOps/s $\color{#35bf28}+0.92\%$
test_stacked_get 0.5833ms 0.3589ms 2.7860 KOps/s 2.7605 KOps/s $\color{#35bf28}+0.92\%$
test_nested_getitemleaf 46.0460μs 10.5431μs 94.8487 KOps/s 94.0743 KOps/s $\color{#35bf28}+0.82\%$
test_nested_getitem 34.3950μs 10.0578μs 99.4254 KOps/s 99.7822 KOps/s $\color{#d91a1a}-0.36\%$
test_stacked_getitemleaf 0.5991ms 0.3894ms 2.5683 KOps/s 2.5187 KOps/s $\color{#35bf28}+1.97\%$
test_stacked_getitem 0.6911ms 0.3625ms 2.7589 KOps/s 2.7518 KOps/s $\color{#35bf28}+0.26\%$
test_lock_nested 1.3427ms 0.4015ms 2.4905 KOps/s 2.5337 KOps/s $\color{#d91a1a}-1.71\%$
test_lock_stack_nested 75.7031ms 6.2882ms 159.0278 Ops/s 162.6571 Ops/s $\color{#d91a1a}-2.23\%$
test_unlock_nested 66.7956ms 0.4653ms 2.1491 KOps/s 2.5547 KOps/s $\textbf{\color{#d91a1a}-15.88\%}$
test_unlock_stack_nested 75.6839ms 5.8986ms 169.5316 Ops/s 170.5289 Ops/s $\color{#d91a1a}-0.58\%$
test_flatten_speed 0.6791ms 0.3711ms 2.6948 KOps/s 2.7299 KOps/s $\color{#d91a1a}-1.28\%$
test_unflatten_speed 0.6372ms 0.4577ms 2.1848 KOps/s 2.1807 KOps/s $\color{#35bf28}+0.19\%$
test_common_ops 1.1760ms 0.6827ms 1.4648 KOps/s 1.4280 KOps/s $\color{#35bf28}+2.58\%$
test_creation 27.3610μs 1.9139μs 522.4865 KOps/s 534.1128 KOps/s $\color{#d91a1a}-2.18\%$
test_creation_empty 32.2800μs 9.9533μs 100.4690 KOps/s 90.3770 KOps/s $\textbf{\color{#35bf28}+11.17\%}$
test_creation_nested_1 39.4630μs 12.5143μs 79.9085 KOps/s 72.9816 KOps/s $\textbf{\color{#35bf28}+9.49\%}$
test_creation_nested_2 92.0320μs 15.7127μs 63.6428 KOps/s 58.3998 KOps/s $\textbf{\color{#35bf28}+8.98\%}$
test_clone 0.2351ms 13.0783μs 76.4626 KOps/s 78.3414 KOps/s $\color{#d91a1a}-2.40\%$
test_getitem[int] 33.8030μs 11.0299μs 90.6629 KOps/s 91.9688 KOps/s $\color{#d91a1a}-1.42\%$
test_getitem[slice_int] 66.6040μs 22.3882μs 44.6664 KOps/s 44.6886 KOps/s $\color{#d91a1a}-0.05\%$
test_getitem[range] 96.0990μs 41.5532μs 24.0656 KOps/s 24.4390 KOps/s $\color{#d91a1a}-1.53\%$
test_getitem[tuple] 83.8460μs 18.0524μs 55.3942 KOps/s 54.4216 KOps/s $\color{#35bf28}+1.79\%$
test_getitem[list] 85.1590μs 36.5928μs 27.3278 KOps/s 27.7849 KOps/s $\color{#d91a1a}-1.65\%$
test_setitem_dim[int] 59.5310μs 29.3431μs 34.0795 KOps/s 28.5906 KOps/s $\textbf{\color{#35bf28}+19.20\%}$
test_setitem_dim[slice_int] 96.6500μs 56.6043μs 17.6665 KOps/s 17.1823 KOps/s $\color{#35bf28}+2.82\%$
test_setitem_dim[range] 0.1283ms 75.3117μs 13.2782 KOps/s 12.9670 KOps/s $\color{#35bf28}+2.40\%$
test_setitem_dim[tuple] 79.5280μs 44.8599μs 22.2916 KOps/s 20.5659 KOps/s $\textbf{\color{#35bf28}+8.39\%}$
test_setitem 0.1141ms 19.0928μs 52.3757 KOps/s 50.1680 KOps/s $\color{#35bf28}+4.40\%$
test_set 0.1984ms 18.8567μs 53.0316 KOps/s 52.0903 KOps/s $\color{#35bf28}+1.81\%$
test_set_shared 4.1899ms 0.1417ms 7.0563 KOps/s 7.1094 KOps/s $\color{#d91a1a}-0.75\%$
test_update 0.1611ms 21.4909μs 46.5313 KOps/s 44.6699 KOps/s $\color{#35bf28}+4.17\%$
test_update_nested 0.2350ms 29.6215μs 33.7593 KOps/s 33.3758 KOps/s $\color{#35bf28}+1.15\%$
test_set_nested 0.1951ms 21.1577μs 47.2641 KOps/s 47.3471 KOps/s $\color{#d91a1a}-0.18\%$
test_set_nested_new 0.1323ms 24.4338μs 40.9269 KOps/s 40.5520 KOps/s $\color{#35bf28}+0.92\%$
test_select 0.2042ms 38.7619μs 25.7985 KOps/s 25.3281 KOps/s $\color{#35bf28}+1.86\%$
test_select_nested 0.1348ms 62.4142μs 16.0220 KOps/s 16.0958 KOps/s $\color{#d91a1a}-0.46\%$
test_exclude_nested 0.2289ms 0.1122ms 8.9108 KOps/s 8.7427 KOps/s $\color{#35bf28}+1.92\%$
test_empty[True] 0.7141ms 0.3298ms 3.0326 KOps/s 3.0174 KOps/s $\color{#35bf28}+0.50\%$
test_empty[False] 6.6144μs 1.0271μs 973.6410 KOps/s 967.0463 KOps/s $\color{#35bf28}+0.68\%$
test_unbind_speed 0.5408ms 0.3186ms 3.1389 KOps/s 3.1628 KOps/s $\color{#d91a1a}-0.76\%$
test_unbind_speed_stack0 73.8753ms 4.1599ms 240.3876 Ops/s 244.9394 Ops/s $\color{#d91a1a}-1.86\%$
test_unbind_speed_stack1 2.4876μs 0.6217μs 1.6086 MOps/s 1.5576 MOps/s $\color{#35bf28}+3.28\%$
test_split 2.2974ms 1.4814ms 675.0526 Ops/s 670.9192 Ops/s $\color{#35bf28}+0.62\%$
test_chunk 69.2918ms 1.5780ms 633.6983 Ops/s 630.0922 Ops/s $\color{#35bf28}+0.57\%$
test_creation[device0] 0.1647ms 99.2559μs 10.0750 KOps/s 9.7256 KOps/s $\color{#35bf28}+3.59\%$
test_creation_from_tensor 5.1116ms 80.7864μs 12.3783 KOps/s 12.3269 KOps/s $\color{#35bf28}+0.42\%$
test_add_one[memmap_tensor0] 0.2787ms 5.2533μs 190.3582 KOps/s 194.7270 KOps/s $\color{#d91a1a}-2.24\%$
test_contiguous[memmap_tensor0] 8.5260μs 0.6608μs 1.5134 MOps/s 1.5743 MOps/s $\color{#d91a1a}-3.87\%$
test_stack[memmap_tensor0] 66.1240μs 3.5530μs 281.4504 KOps/s 291.9814 KOps/s $\color{#d91a1a}-3.61\%$
test_memmaptd_index 0.9655ms 0.2178ms 4.5904 KOps/s 4.6569 KOps/s $\color{#d91a1a}-1.43\%$
test_memmaptd_index_astensor 0.6640ms 0.2776ms 3.6021 KOps/s 3.6264 KOps/s $\color{#d91a1a}-0.67\%$
test_memmaptd_index_op 1.1501ms 0.5764ms 1.7349 KOps/s 1.7368 KOps/s $\color{#d91a1a}-0.11\%$
test_serialize_model 0.1063s 99.6164ms 10.0385 Ops/s 9.1971 Ops/s $\textbf{\color{#35bf28}+9.15\%}$
test_serialize_model_pickle 0.4649s 0.3764s 2.6567 Ops/s 2.5994 Ops/s $\color{#35bf28}+2.20\%$
test_serialize_weights 0.1652s 0.1054s 9.4888 Ops/s 10.3392 Ops/s $\textbf{\color{#d91a1a}-8.22\%}$
test_serialize_weights_returnearly 0.1779s 0.1330s 7.5183 Ops/s 8.1974 Ops/s $\textbf{\color{#d91a1a}-8.28\%}$
test_serialize_weights_pickle 1.0638s 0.6146s 1.6270 Ops/s 2.3706 Ops/s $\textbf{\color{#d91a1a}-31.37\%}$
test_serialize_weights_filesystem 0.1564s 97.8737ms 10.2172 Ops/s 10.1417 Ops/s $\color{#35bf28}+0.75\%$
test_serialize_model_filesystem 96.6894ms 90.0105ms 11.1098 Ops/s 10.6800 Ops/s $\color{#35bf28}+4.02\%$
test_reshape_pytree 53.9210μs 23.6230μs 42.3316 KOps/s 43.6526 KOps/s $\color{#d91a1a}-3.03\%$
test_reshape_td 65.9330μs 29.4199μs 33.9906 KOps/s 33.7020 KOps/s $\color{#35bf28}+0.86\%$
test_view_pytree 77.0040μs 23.4090μs 42.7186 KOps/s 43.7905 KOps/s $\color{#d91a1a}-2.45\%$
test_view_td 24.1950μs 4.9200μs 203.2512 KOps/s 208.0294 KOps/s $\color{#d91a1a}-2.30\%$
test_unbind_pytree 69.4890μs 26.9543μs 37.0998 KOps/s 37.8292 KOps/s $\color{#d91a1a}-1.93\%$
test_unbind_td 0.1026ms 50.7645μs 19.6988 KOps/s 20.1593 KOps/s $\color{#d91a1a}-2.28\%$
test_split_pytree 74.4590μs 26.9052μs 37.1675 KOps/s 38.1073 KOps/s $\color{#d91a1a}-2.47\%$
test_split_td 0.5482ms 40.9821μs 24.4009 KOps/s 25.1034 KOps/s $\color{#d91a1a}-2.80\%$
test_add_pytree 70.9520μs 32.1921μs 31.0635 KOps/s 31.7393 KOps/s $\color{#d91a1a}-2.13\%$
test_add_td 0.1193ms 49.4654μs 20.2161 KOps/s 19.1879 KOps/s $\textbf{\color{#35bf28}+5.36\%}$
test_distributed 0.1826ms 97.7849μs 10.2265 KOps/s 9.8725 KOps/s $\color{#35bf28}+3.59\%$
test_tdmodule 0.1039ms 22.0059μs 45.4423 KOps/s 43.3070 KOps/s $\color{#35bf28}+4.93\%$
test_tdmodule_dispatch 0.2537ms 40.1418μs 24.9117 KOps/s 23.8164 KOps/s $\color{#35bf28}+4.60\%$
test_tdseq 38.5520μs 24.7727μs 40.3671 KOps/s 37.0966 KOps/s $\textbf{\color{#35bf28}+8.82\%}$
test_tdseq_dispatch 0.1470ms 44.4391μs 22.5027 KOps/s 21.2120 KOps/s $\textbf{\color{#35bf28}+6.08\%}$
test_instantiation_functorch 1.5122ms 1.2890ms 775.8073 Ops/s 776.2469 Ops/s $\color{#d91a1a}-0.06\%$
test_instantiation_td 67.0510ms 1.0726ms 932.3502 Ops/s 1.0155 KOps/s $\textbf{\color{#d91a1a}-8.19\%}$
test_exec_functorch 0.2688ms 0.1567ms 6.3813 KOps/s 6.3164 KOps/s $\color{#35bf28}+1.03\%$
test_exec_functional_call 0.2776ms 0.1432ms 6.9850 KOps/s 6.7593 KOps/s $\color{#35bf28}+3.34\%$
test_exec_td 0.2003ms 0.1400ms 7.1452 KOps/s 6.7982 KOps/s $\textbf{\color{#35bf28}+5.10\%}$
test_exec_td_decorator 0.9327ms 0.1769ms 5.6543 KOps/s 5.5625 KOps/s $\color{#35bf28}+1.65\%$
test_vmap_mlp_speed[True-True] 1.1437ms 0.8874ms 1.1269 KOps/s 1.1108 KOps/s $\color{#35bf28}+1.44\%$
test_vmap_mlp_speed[True-False] 0.7145ms 0.4780ms 2.0921 KOps/s 2.0854 KOps/s $\color{#35bf28}+0.32\%$
test_vmap_mlp_speed[False-True] 1.4505ms 0.7769ms 1.2872 KOps/s 1.2715 KOps/s $\color{#35bf28}+1.23\%$
test_vmap_mlp_speed[False-False] 0.6293ms 0.3934ms 2.5419 KOps/s 2.5565 KOps/s $\color{#d91a1a}-0.57\%$
test_vmap_mlp_speed_decorator[True-True] 2.8667ms 2.3564ms 424.3810 Ops/s 405.6111 Ops/s $\color{#35bf28}+4.63\%$
test_vmap_mlp_speed_decorator[True-False] 0.9025ms 0.5259ms 1.9014 KOps/s 1.8829 KOps/s $\color{#35bf28}+0.98\%$
test_vmap_mlp_speed_decorator[False-True] 2.5128ms 1.9171ms 521.6117 Ops/s 487.7678 Ops/s $\textbf{\color{#35bf28}+6.94\%}$
test_vmap_mlp_speed_decorator[False-False] 0.7102ms 0.3981ms 2.5117 KOps/s 2.4715 KOps/s $\color{#35bf28}+1.63\%$

Copy link

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

Total Benchmarks: 132. Improved: $\large\color{#35bf28}10$. Worsened: $\large\color{#d91a1a}11$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 96.1620μs 14.1807μs 70.5183 KOps/s 74.8430 KOps/s $\textbf{\color{#d91a1a}-5.78\%}$
test_plain_set_stack_nested 0.1797ms 0.1169ms 8.5509 KOps/s 8.4923 KOps/s $\color{#35bf28}+0.69\%$
test_plain_set_nested_inplace 91.6520μs 15.5176μs 64.4430 KOps/s 68.1774 KOps/s $\textbf{\color{#d91a1a}-5.48\%}$
test_plain_set_stack_nested_inplace 0.2245ms 0.1450ms 6.8973 KOps/s 6.8277 KOps/s $\color{#35bf28}+1.02\%$
test_items 23.3700μs 4.6607μs 214.5591 KOps/s 210.0284 KOps/s $\color{#35bf28}+2.16\%$
test_items_nested 0.4213ms 0.3366ms 2.9706 KOps/s 2.9388 KOps/s $\color{#35bf28}+1.08\%$
test_items_nested_locked 0.4371ms 0.3399ms 2.9424 KOps/s 2.9173 KOps/s $\color{#35bf28}+0.86\%$
test_items_nested_leaf 0.2773ms 0.1990ms 5.0242 KOps/s 4.9779 KOps/s $\color{#35bf28}+0.93\%$
test_items_stack_nested 1.4015ms 1.3098ms 763.4690 Ops/s 753.8026 Ops/s $\color{#35bf28}+1.28\%$
test_items_stack_nested_leaf 1.2552ms 1.1409ms 876.5209 Ops/s 869.2505 Ops/s $\color{#35bf28}+0.84\%$
test_items_stack_nested_locked 1.0293ms 0.9126ms 1.0957 KOps/s 1.0660 KOps/s $\color{#35bf28}+2.79\%$
test_keys 73.1610μs 4.5398μs 220.2732 KOps/s 219.0770 KOps/s $\color{#35bf28}+0.55\%$
test_keys_nested 0.9077ms 94.8653μs 10.5413 KOps/s 10.5528 KOps/s $\color{#d91a1a}-0.11\%$
test_keys_nested_locked 0.1807ms 98.6052μs 10.1415 KOps/s 10.2244 KOps/s $\color{#d91a1a}-0.81\%$
test_keys_nested_leaf 0.1816ms 78.7840μs 12.6929 KOps/s 12.7848 KOps/s $\color{#d91a1a}-0.72\%$
test_keys_stack_nested 1.2799ms 1.1349ms 881.1583 Ops/s 874.8880 Ops/s $\color{#35bf28}+0.72\%$
test_keys_stack_nested_leaf 1.2413ms 1.1287ms 886.0038 Ops/s 890.7162 Ops/s $\color{#d91a1a}-0.53\%$
test_keys_stack_nested_locked 0.8804ms 0.7370ms 1.3568 KOps/s 1.3460 KOps/s $\color{#35bf28}+0.80\%$
test_values 24.5037μs 1.8961μs 527.4001 KOps/s 530.6369 KOps/s $\color{#d91a1a}-0.61\%$
test_values_nested 0.1212ms 45.0608μs 22.1922 KOps/s 22.2415 KOps/s $\color{#d91a1a}-0.22\%$
test_values_nested_locked 0.1219ms 47.3531μs 21.1179 KOps/s 21.2241 KOps/s $\color{#d91a1a}-0.50\%$
test_values_nested_leaf 0.1143ms 39.5691μs 25.2722 KOps/s 25.5554 KOps/s $\color{#d91a1a}-1.11\%$
test_values_stack_nested 1.1140ms 0.9547ms 1.0475 KOps/s 1.0344 KOps/s $\color{#35bf28}+1.27\%$
test_values_stack_nested_leaf 1.0298ms 0.9451ms 1.0581 KOps/s 1.0496 KOps/s $\color{#35bf28}+0.81\%$
test_values_stack_nested_locked 0.7256ms 0.5860ms 1.7064 KOps/s 1.6692 KOps/s $\color{#35bf28}+2.23\%$
test_membership 27.3800μs 1.0464μs 955.6907 KOps/s 1.0613 MOps/s $\textbf{\color{#d91a1a}-9.95\%}$
test_membership_nested 19.7700μs 2.8363μs 352.5726 KOps/s 339.0042 KOps/s $\color{#35bf28}+4.00\%$
test_membership_nested_leaf 78.7810μs 2.8430μs 351.7430 KOps/s 337.9132 KOps/s $\color{#35bf28}+4.09\%$
test_membership_stacked_nested 30.2800μs 10.9930μs 90.9668 KOps/s 90.4887 KOps/s $\color{#35bf28}+0.53\%$
test_membership_stacked_nested_leaf 87.0420μs 10.9515μs 91.3120 KOps/s 90.5809 KOps/s $\color{#35bf28}+0.81\%$
test_membership_nested_last 78.4110μs 5.2269μs 191.3163 KOps/s 187.2737 KOps/s $\color{#35bf28}+2.16\%$
test_membership_nested_leaf_last 33.6610μs 5.2242μs 191.4152 KOps/s 186.9806 KOps/s $\color{#35bf28}+2.37\%$
test_membership_stacked_nested_last 0.2454ms 0.1417ms 7.0579 KOps/s 7.0491 KOps/s $\color{#35bf28}+0.12\%$
test_membership_stacked_nested_leaf_last 84.2320μs 12.9147μs 77.4314 KOps/s 77.6560 KOps/s $\color{#d91a1a}-0.29\%$
test_nested_getleaf 76.4910μs 8.3976μs 119.0822 KOps/s 119.5049 KOps/s $\color{#d91a1a}-0.35\%$
test_nested_get 31.9010μs 7.8788μs 126.9229 KOps/s 126.4363 KOps/s $\color{#35bf28}+0.38\%$
test_stacked_getleaf 0.4235ms 0.3199ms 3.1260 KOps/s 3.1384 KOps/s $\color{#d91a1a}-0.39\%$
test_stacked_get 0.3599ms 0.2816ms 3.5506 KOps/s 3.5267 KOps/s $\color{#35bf28}+0.68\%$
test_nested_getitemleaf 22.3210μs 8.3903μs 119.1855 KOps/s 118.9504 KOps/s $\color{#35bf28}+0.20\%$
test_nested_getitem 78.2810μs 7.9701μs 125.4694 KOps/s 125.6250 KOps/s $\color{#d91a1a}-0.12\%$
test_stacked_getitemleaf 0.4042ms 0.3193ms 3.1322 KOps/s 3.1397 KOps/s $\color{#d91a1a}-0.24\%$
test_stacked_getitem 0.3654ms 0.2825ms 3.5399 KOps/s 3.5243 KOps/s $\color{#35bf28}+0.44\%$
test_lock_nested 7.4725ms 0.4038ms 2.4767 KOps/s 2.5639 KOps/s $\color{#d91a1a}-3.40\%$
test_lock_stack_nested 92.9596ms 6.5317ms 153.1006 Ops/s 154.2725 Ops/s $\color{#d91a1a}-0.76\%$
test_unlock_nested 0.7932ms 0.3936ms 2.5404 KOps/s 2.5576 KOps/s $\color{#d91a1a}-0.67\%$
test_unlock_stack_nested 91.3162ms 6.8604ms 145.7649 Ops/s 144.0492 Ops/s $\color{#35bf28}+1.19\%$
test_flatten_speed 79.5166ms 0.2821ms 3.5453 KOps/s 3.8186 KOps/s $\textbf{\color{#d91a1a}-7.16\%}$
test_unflatten_speed 0.4434ms 0.3613ms 2.7674 KOps/s 2.7635 KOps/s $\color{#35bf28}+0.14\%$
test_common_ops 1.0719ms 0.6231ms 1.6050 KOps/s 1.6438 KOps/s $\color{#d91a1a}-2.36\%$
test_creation 33.5310μs 1.5571μs 642.2181 KOps/s 638.0338 KOps/s $\color{#35bf28}+0.66\%$
test_creation_empty 80.0610μs 9.6482μs 103.6467 KOps/s 126.4656 KOps/s $\textbf{\color{#d91a1a}-18.04\%}$
test_creation_nested_1 28.3210μs 11.3890μs 87.8041 KOps/s 103.1415 KOps/s $\textbf{\color{#d91a1a}-14.87\%}$
test_creation_nested_2 84.9710μs 13.7657μs 72.6443 KOps/s 82.1491 KOps/s $\textbf{\color{#d91a1a}-11.57\%}$
test_clone 0.1394ms 13.6506μs 73.2566 KOps/s 67.0750 KOps/s $\textbf{\color{#35bf28}+9.22\%}$
test_getitem[int] 28.5900μs 10.6877μs 93.5653 KOps/s 91.6828 KOps/s $\color{#35bf28}+2.05\%$
test_getitem[slice_int] 43.9310μs 21.0350μs 47.5399 KOps/s 46.1233 KOps/s $\color{#35bf28}+3.07\%$
test_getitem[range] 0.1158ms 37.0014μs 27.0260 KOps/s 25.6042 KOps/s $\textbf{\color{#35bf28}+5.55\%}$
test_getitem[tuple] 38.6610μs 18.8088μs 53.1667 KOps/s 53.3267 KOps/s $\color{#d91a1a}-0.30\%$
test_getitem[list] 0.1274ms 33.6636μs 29.7057 KOps/s 28.4656 KOps/s $\color{#35bf28}+4.36\%$
test_setitem_dim[int] 44.7210μs 28.5707μs 35.0009 KOps/s 34.6709 KOps/s $\color{#35bf28}+0.95\%$
test_setitem_dim[slice_int] 71.3110μs 49.9165μs 20.0335 KOps/s 19.3703 KOps/s $\color{#35bf28}+3.42\%$
test_setitem_dim[range] 83.4110μs 63.8715μs 15.6564 KOps/s 15.5022 KOps/s $\color{#35bf28}+1.00\%$
test_setitem_dim[tuple] 66.1620μs 43.3433μs 23.0716 KOps/s 22.6612 KOps/s $\color{#35bf28}+1.81\%$
test_setitem 0.1359ms 18.9593μs 52.7446 KOps/s 51.1939 KOps/s $\color{#35bf28}+3.03\%$
test_set 0.1366ms 18.4789μs 54.1159 KOps/s 52.6752 KOps/s $\color{#35bf28}+2.74\%$
test_set_shared 2.6257ms 0.1049ms 9.5350 KOps/s 9.3414 KOps/s $\color{#35bf28}+2.07\%$
test_update 0.1344ms 21.7568μs 45.9625 KOps/s 47.3070 KOps/s $\color{#d91a1a}-2.84\%$
test_update_nested 0.1307ms 28.4711μs 35.1234 KOps/s 35.8840 KOps/s $\color{#d91a1a}-2.12\%$
test_set_nested 0.1251ms 19.7298μs 50.6848 KOps/s 49.4558 KOps/s $\color{#35bf28}+2.49\%$
test_set_nested_new 0.1258ms 22.5455μs 44.3548 KOps/s 43.6769 KOps/s $\color{#35bf28}+1.55\%$
test_select 86.5110μs 35.3260μs 28.3078 KOps/s 27.9455 KOps/s $\color{#35bf28}+1.30\%$
test_select_nested 0.1390ms 54.4722μs 18.3580 KOps/s 18.0400 KOps/s $\color{#35bf28}+1.76\%$
test_exclude_nested 0.1837ms 0.1065ms 9.3897 KOps/s 9.1252 KOps/s $\color{#35bf28}+2.90\%$
test_empty[True] 0.3947ms 0.3185ms 3.1392 KOps/s 3.1475 KOps/s $\color{#d91a1a}-0.26\%$
test_empty[False] 2.8281μs 0.8632μs 1.1584 MOps/s 1.1717 MOps/s $\color{#d91a1a}-1.13\%$
test_to 72.6210μs 51.9788μs 19.2386 KOps/s 17.1307 KOps/s $\textbf{\color{#35bf28}+12.31\%}$
test_to_nonblocking 0.3692ms 33.1683μs 30.1493 KOps/s 27.9029 KOps/s $\textbf{\color{#35bf28}+8.05\%}$
test_unbind_speed 0.6459ms 0.3114ms 3.2113 KOps/s 3.2330 KOps/s $\color{#d91a1a}-0.67\%$
test_unbind_speed_stack0 85.4683ms 3.6632ms 272.9834 Ops/s 265.1146 Ops/s $\color{#35bf28}+2.97\%$
test_unbind_speed_stack1 1.3191μs 0.5368μs 1.8629 MOps/s 1.8538 MOps/s $\color{#35bf28}+0.49\%$
test_split 83.3899ms 1.7090ms 585.1528 Ops/s 592.1482 Ops/s $\color{#d91a1a}-1.18\%$
test_chunk 82.5232ms 1.6841ms 593.7807 Ops/s 601.8780 Ops/s $\color{#d91a1a}-1.35\%$
test_creation[device0] 0.3532ms 74.3981μs 13.4412 KOps/s 13.4817 KOps/s $\color{#d91a1a}-0.30\%$
test_creation_from_tensor 0.4016ms 54.9819μs 18.1878 KOps/s 17.5662 KOps/s $\color{#35bf28}+3.54\%$
test_add_one[memmap_tensor0] 82.6820μs 7.2407μs 138.1085 KOps/s 124.6011 KOps/s $\textbf{\color{#35bf28}+10.84\%}$
test_contiguous[memmap_tensor0] 23.7800μs 0.6389μs 1.5651 MOps/s 1.5669 MOps/s $\color{#d91a1a}-0.11\%$
test_stack[memmap_tensor0] 20.6000μs 4.7680μs 209.7332 KOps/s 213.8396 KOps/s $\color{#d91a1a}-1.92\%$
test_memmaptd_index 1.1894ms 0.2602ms 3.8439 KOps/s 3.8665 KOps/s $\color{#d91a1a}-0.59\%$
test_memmaptd_index_astensor 0.7426ms 0.3139ms 3.1860 KOps/s 3.1324 KOps/s $\color{#35bf28}+1.71\%$
test_memmaptd_index_op 0.9774ms 0.6482ms 1.5428 KOps/s 1.5645 KOps/s $\color{#d91a1a}-1.39\%$
test_serialize_model 0.1755s 99.0270ms 10.0983 Ops/s 9.5676 Ops/s $\textbf{\color{#35bf28}+5.55\%}$
test_serialize_model_pickle 1.3499s 1.2369s 0.8085 Ops/s 0.8058 Ops/s $\color{#35bf28}+0.33\%$
test_serialize_weights 0.1696s 96.2012ms 10.3949 Ops/s 9.8667 Ops/s $\textbf{\color{#35bf28}+5.35\%}$
test_serialize_weights_returnearly 0.2682s 71.5044ms 13.9851 Ops/s 14.4662 Ops/s $\color{#d91a1a}-3.33\%$
test_serialize_weights_pickle 1.3516s 1.2363s 0.8088 Ops/s 0.8076 Ops/s $\color{#35bf28}+0.16\%$
test_reshape_pytree 46.4710μs 25.0727μs 39.8840 KOps/s 39.6918 KOps/s $\color{#35bf28}+0.48\%$
test_reshape_td 50.9810μs 29.4588μs 33.9457 KOps/s 34.1736 KOps/s $\color{#d91a1a}-0.67\%$
test_view_pytree 88.8810μs 24.3970μs 40.9886 KOps/s 41.2294 KOps/s $\color{#d91a1a}-0.58\%$
test_view_td 23.0710μs 4.2333μs 236.2232 KOps/s 238.5402 KOps/s $\color{#d91a1a}-0.97\%$
test_unbind_pytree 59.3000μs 30.8167μs 32.4499 KOps/s 32.7997 KOps/s $\color{#d91a1a}-1.07\%$
test_unbind_td 73.0710μs 49.5843μs 20.1677 KOps/s 20.1518 KOps/s $\color{#35bf28}+0.08\%$
test_split_pytree 60.2210μs 29.2731μs 34.1611 KOps/s 35.0433 KOps/s $\color{#d91a1a}-2.52\%$
test_split_td 0.6904ms 39.6581μs 25.2155 KOps/s 24.4477 KOps/s $\color{#35bf28}+3.14\%$
test_add_pytree 59.0010μs 37.1692μs 26.9040 KOps/s 26.1073 KOps/s $\color{#35bf28}+3.05\%$
test_add_td 76.0910μs 49.6005μs 20.1611 KOps/s 19.2172 KOps/s $\color{#35bf28}+4.91\%$
test_distributed 11.4325ms 96.9660μs 10.3129 KOps/s 14.1154 KOps/s $\textbf{\color{#d91a1a}-26.94\%}$
test_tdmodule 95.3920μs 18.4667μs 54.1516 KOps/s 57.4486 KOps/s $\textbf{\color{#d91a1a}-5.74\%}$
test_tdmodule_dispatch 0.2389ms 35.1243μs 28.4703 KOps/s 30.6312 KOps/s $\textbf{\color{#d91a1a}-7.05\%}$
test_tdseq 46.1710μs 21.5386μs 46.4282 KOps/s 48.7267 KOps/s $\color{#d91a1a}-4.72\%$
test_tdseq_dispatch 54.0400μs 38.0561μs 26.2770 KOps/s 27.9510 KOps/s $\textbf{\color{#d91a1a}-5.99\%}$
test_instantiation_functorch 1.9864ms 1.6656ms 600.3728 Ops/s 584.4285 Ops/s $\color{#35bf28}+2.73\%$
test_instantiation_td 1.6860ms 1.1611ms 861.2391 Ops/s 853.7146 Ops/s $\color{#35bf28}+0.88\%$
test_exec_functorch 0.2037ms 0.1617ms 6.1856 KOps/s 6.0402 KOps/s $\color{#35bf28}+2.41\%$
test_exec_functional_call 0.1927ms 0.1607ms 6.2228 KOps/s 5.8750 KOps/s $\textbf{\color{#35bf28}+5.92\%}$
test_exec_td 0.1816ms 0.1518ms 6.5855 KOps/s 6.1841 KOps/s $\textbf{\color{#35bf28}+6.49\%}$
test_exec_td_decorator 0.9106ms 0.1894ms 5.2807 KOps/s 5.0546 KOps/s $\color{#35bf28}+4.47\%$
test_vmap_mlp_speed[True-True] 1.1525ms 1.0972ms 911.3782 Ops/s 871.4312 Ops/s $\color{#35bf28}+4.58\%$
test_vmap_mlp_speed[True-False] 0.6996ms 0.6543ms 1.5283 KOps/s 1.4772 KOps/s $\color{#35bf28}+3.46\%$
test_vmap_mlp_speed[False-True] 1.0414ms 1.0056ms 994.4103 Ops/s 947.8678 Ops/s $\color{#35bf28}+4.91\%$
test_vmap_mlp_speed[False-False] 0.6193ms 0.5843ms 1.7114 KOps/s 1.6534 KOps/s $\color{#35bf28}+3.51\%$
test_vmap_mlp_speed_decorator[True-True] 3.1931ms 2.5054ms 399.1300 Ops/s 385.0193 Ops/s $\color{#35bf28}+3.66\%$
test_vmap_mlp_speed_decorator[True-False] 1.0937ms 0.6996ms 1.4294 KOps/s 1.4317 KOps/s $\color{#d91a1a}-0.16\%$
test_vmap_mlp_speed_decorator[False-True] 2.4896ms 2.1009ms 475.9910 Ops/s 475.9540 Ops/s $+0.01\%$
test_vmap_mlp_speed_decorator[False-False] 0.9296ms 0.6053ms 1.6522 KOps/s 1.6632 KOps/s $\color{#d91a1a}-0.66\%$
test_vmap_transformer_speed[True-True] 12.8843ms 12.2889ms 81.3745 Ops/s 81.5073 Ops/s $\color{#d91a1a}-0.16\%$
test_vmap_transformer_speed[True-False] 8.4462ms 8.1102ms 123.3017 Ops/s 119.8246 Ops/s $\color{#35bf28}+2.90\%$
test_vmap_transformer_speed[False-True] 12.7345ms 12.2158ms 81.8611 Ops/s 79.8927 Ops/s $\color{#35bf28}+2.46\%$
test_vmap_transformer_speed[False-False] 8.1065ms 7.9935ms 125.1023 Ops/s 121.1250 Ops/s $\color{#35bf28}+3.28\%$
test_vmap_transformer_speed_decorator[True-True] 76.3212ms 75.4119ms 13.2605 Ops/s 11.9110 Ops/s $\textbf{\color{#35bf28}+11.33\%}$
test_vmap_transformer_speed_decorator[True-False] 21.3235ms 19.4251ms 51.4797 Ops/s 50.2115 Ops/s $\color{#35bf28}+2.53\%$
test_vmap_transformer_speed_decorator[False-True] 69.3207ms 68.3789ms 14.6244 Ops/s 14.4339 Ops/s $\color{#35bf28}+1.32\%$
test_vmap_transformer_speed_decorator[False-False] 0.1213s 20.9383ms 47.7594 Ops/s 46.2761 Ops/s $\color{#35bf28}+3.21\%$

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.

2 participants