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

[Versioning] M1 compatibility #523

Merged
merged 26 commits into from
Oct 26, 2023
Merged

[Versioning] M1 compatibility #523

merged 26 commits into from
Oct 26, 2023

Conversation

vmoens
Copy link
Contributor

@vmoens vmoens commented Sep 7, 2023

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 Sep 7, 2023
@github-actions
Copy link

github-actions bot commented Sep 7, 2023

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

Total Benchmarks: 105. Improved: $\large\color{#35bf28}5$. Worsened: $\large\color{#d91a1a}4$.

Expand to view detailed results
Name Max Mean Ops Ops on Repo HEAD Change
test_plain_set_nested 31.9000μs 20.0412μs 49.8972 KOps/s 50.5111 KOps/s $\color{#d91a1a}-1.22\%$
test_plain_set_stack_nested 0.2216ms 0.1827ms 5.4720 KOps/s 5.4716 KOps/s $+0.01\%$
test_plain_set_nested_inplace 36.5000μs 23.4817μs 42.5864 KOps/s 42.7936 KOps/s $\color{#d91a1a}-0.48\%$
test_plain_set_stack_nested_inplace 0.2426ms 0.2200ms 4.5461 KOps/s 4.6147 KOps/s $\color{#d91a1a}-1.49\%$
test_items 22.1000μs 3.1178μs 320.7395 KOps/s 321.6457 KOps/s $\color{#d91a1a}-0.28\%$
test_items_nested 0.4273ms 0.4009ms 2.4943 KOps/s 2.6972 KOps/s $\textbf{\color{#d91a1a}-7.52\%}$
test_items_nested_locked 0.4582ms 0.4009ms 2.4944 KOps/s 2.6876 KOps/s $\textbf{\color{#d91a1a}-7.19\%}$
test_items_nested_leaf 1.2133ms 0.2388ms 4.1873 KOps/s 4.6300 KOps/s $\textbf{\color{#d91a1a}-9.56\%}$
test_items_stack_nested 1.9870ms 1.9221ms 520.2665 Ops/s 533.0955 Ops/s $\color{#d91a1a}-2.41\%$
test_items_stack_nested_leaf 1.7629ms 1.7232ms 580.3178 Ops/s 589.0901 Ops/s $\color{#d91a1a}-1.49\%$
test_items_stack_nested_locked 2.3906ms 0.9773ms 1.0233 KOps/s 1.0268 KOps/s $\color{#d91a1a}-0.34\%$
test_keys 26.5000μs 4.5841μs 218.1451 KOps/s 222.6771 KOps/s $\color{#d91a1a}-2.04\%$
test_keys_nested 0.7183ms 0.1744ms 5.7334 KOps/s 5.5030 KOps/s $\color{#35bf28}+4.19\%$
test_keys_nested_locked 0.2202ms 0.1722ms 5.8068 KOps/s 5.8362 KOps/s $\color{#d91a1a}-0.50\%$
test_keys_nested_leaf 0.2845ms 0.1704ms 5.8701 KOps/s 5.8914 KOps/s $\color{#d91a1a}-0.36\%$
test_keys_stack_nested 1.8108ms 1.7392ms 574.9912 Ops/s 577.6947 Ops/s $\color{#d91a1a}-0.47\%$
test_keys_stack_nested_leaf 1.8100ms 1.7378ms 575.4351 Ops/s 580.5824 Ops/s $\color{#d91a1a}-0.89\%$
test_keys_stack_nested_locked 1.0898ms 0.7886ms 1.2680 KOps/s 1.2799 KOps/s $\color{#d91a1a}-0.93\%$
test_values 17.6000μs 1.3270μs 753.5682 KOps/s 802.8355 KOps/s $\textbf{\color{#d91a1a}-6.14\%}$
test_values_nested 0.1074ms 64.4708μs 15.5109 KOps/s 15.5668 KOps/s $\color{#d91a1a}-0.36\%$
test_values_nested_locked 0.1146ms 63.9400μs 15.6397 KOps/s 15.4369 KOps/s $\color{#35bf28}+1.31\%$
test_values_nested_leaf 78.6010μs 56.2251μs 17.7857 KOps/s 17.5059 KOps/s $\color{#35bf28}+1.60\%$
test_values_stack_nested 1.5831ms 1.5177ms 658.8771 Ops/s 656.9659 Ops/s $\color{#35bf28}+0.29\%$
test_values_stack_nested_leaf 1.5581ms 1.5016ms 665.9686 Ops/s 667.6121 Ops/s $\color{#d91a1a}-0.25\%$
test_values_stack_nested_locked 0.6758ms 0.6289ms 1.5900 KOps/s 1.5761 KOps/s $\color{#35bf28}+0.88\%$
test_membership 37.2000μs 1.7741μs 563.6717 KOps/s 530.7397 KOps/s $\textbf{\color{#35bf28}+6.20\%}$
test_membership_nested 23.8000μs 3.7261μs 268.3798 KOps/s 269.4976 KOps/s $\color{#d91a1a}-0.41\%$
test_membership_nested_leaf 17.1000μs 3.6878μs 271.1657 KOps/s 268.8748 KOps/s $\color{#35bf28}+0.85\%$
test_membership_stacked_nested 40.6000μs 14.7662μs 67.7220 KOps/s 68.5283 KOps/s $\color{#d91a1a}-1.18\%$
test_membership_stacked_nested_leaf 39.1000μs 14.7923μs 67.6025 KOps/s 68.0489 KOps/s $\color{#d91a1a}-0.66\%$
test_membership_nested_last 33.1010μs 7.4598μs 134.0514 KOps/s 135.0378 KOps/s $\color{#d91a1a}-0.73\%$
test_membership_nested_leaf_last 36.3010μs 7.5852μs 131.8355 KOps/s 133.7045 KOps/s $\color{#d91a1a}-1.40\%$
test_membership_stacked_nested_last 0.2623ms 0.2231ms 4.4819 KOps/s 4.5063 KOps/s $\color{#d91a1a}-0.54\%$
test_membership_stacked_nested_leaf_last 39.2000μs 16.9923μs 58.8501 KOps/s 59.2821 KOps/s $\color{#d91a1a}-0.73\%$
test_nested_getleaf 54.0000μs 15.4792μs 64.6027 KOps/s 64.1219 KOps/s $\color{#35bf28}+0.75\%$
test_nested_get 38.3000μs 14.6744μs 68.1460 KOps/s 67.5006 KOps/s $\color{#35bf28}+0.96\%$
test_stacked_getleaf 0.9313ms 0.8182ms 1.2222 KOps/s 1.2240 KOps/s $\color{#d91a1a}-0.14\%$
test_stacked_get 0.8226ms 0.7845ms 1.2747 KOps/s 1.2828 KOps/s $\color{#d91a1a}-0.63\%$
test_nested_getitemleaf 36.5010μs 15.4028μs 64.9231 KOps/s 64.1367 KOps/s $\color{#35bf28}+1.23\%$
test_nested_getitem 43.1000μs 14.6431μs 68.2915 KOps/s 67.2074 KOps/s $\color{#35bf28}+1.61\%$
test_stacked_getitemleaf 0.9185ms 0.8208ms 1.2183 KOps/s 1.2229 KOps/s $\color{#d91a1a}-0.38\%$
test_stacked_getitem 0.8219ms 0.7891ms 1.2673 KOps/s 1.2791 KOps/s $\color{#d91a1a}-0.92\%$
test_lock_nested 51.6775ms 1.4310ms 698.8324 Ops/s 719.5292 Ops/s $\color{#d91a1a}-2.88\%$
test_lock_stack_nested 72.5015ms 16.6775ms 59.9610 Ops/s 55.7714 Ops/s $\textbf{\color{#35bf28}+7.51\%}$
test_unlock_nested 50.5781ms 1.4261ms 701.2301 Ops/s 685.4328 Ops/s $\color{#35bf28}+2.30\%$
test_unlock_stack_nested 70.8144ms 17.1720ms 58.2343 Ops/s 54.6399 Ops/s $\textbf{\color{#35bf28}+6.58\%}$
test_flatten_speed 1.0339ms 0.9818ms 1.0186 KOps/s 1.0255 KOps/s $\color{#d91a1a}-0.67\%$
test_unflatten_speed 1.7544ms 1.7191ms 581.7065 Ops/s 592.7705 Ops/s $\color{#d91a1a}-1.87\%$
test_common_ops 1.1545ms 1.0103ms 989.7858 Ops/s 989.4965 Ops/s $\color{#35bf28}+0.03\%$
test_creation 24.7000μs 5.9918μs 166.8959 KOps/s 164.9914 KOps/s $\color{#35bf28}+1.15\%$
test_creation_empty 27.8000μs 13.1173μs 76.2352 KOps/s 75.4443 KOps/s $\color{#35bf28}+1.05\%$
test_creation_nested_1 41.7000μs 22.5891μs 44.2691 KOps/s 44.0203 KOps/s $\color{#35bf28}+0.57\%$
test_creation_nested_2 45.2000μs 25.7345μs 38.8583 KOps/s 39.3510 KOps/s $\color{#d91a1a}-1.25\%$
test_clone 94.4000μs 24.0399μs 41.5974 KOps/s 42.3205 KOps/s $\color{#d91a1a}-1.71\%$
test_getitem[int] 47.0000μs 26.6113μs 37.5780 KOps/s 36.8477 KOps/s $\color{#35bf28}+1.98\%$
test_getitem[slice_int] 73.5000μs 50.7437μs 19.7069 KOps/s 19.6310 KOps/s $\color{#35bf28}+0.39\%$
test_getitem[range] 0.2107ms 74.6154μs 13.4021 KOps/s 13.2842 KOps/s $\color{#35bf28}+0.89\%$
test_getitem[tuple] 60.4000μs 41.4194μs 24.1433 KOps/s 23.5990 KOps/s $\color{#35bf28}+2.31\%$
test_getitem[list] 0.2296ms 69.8883μs 14.3085 KOps/s 13.9549 KOps/s $\color{#35bf28}+2.53\%$
test_setitem_dim[int] 48.8000μs 33.4270μs 29.9159 KOps/s 30.2430 KOps/s $\color{#d91a1a}-1.08\%$
test_setitem_dim[slice_int] 86.1000μs 59.3587μs 16.8467 KOps/s 16.8164 KOps/s $\color{#35bf28}+0.18\%$
test_setitem_dim[range] 0.1164ms 78.3120μs 12.7694 KOps/s 12.5918 KOps/s $\color{#35bf28}+1.41\%$
test_setitem_dim[tuple] 53.5000μs 49.8323μs 20.0673 KOps/s 20.0583 KOps/s $\color{#35bf28}+0.04\%$
test_setitem 83.6000μs 29.0096μs 34.4713 KOps/s 34.6115 KOps/s $\color{#d91a1a}-0.41\%$
test_set 81.7000μs 28.2198μs 35.4361 KOps/s 35.4349 KOps/s $+0.00\%$
test_set_shared 2.5548ms 0.1707ms 5.8597 KOps/s 5.9537 KOps/s $\color{#d91a1a}-1.58\%$
test_update 0.1365ms 31.2667μs 31.9829 KOps/s 31.8606 KOps/s $\color{#35bf28}+0.38\%$
test_update_nested 0.1090ms 48.8831μs 20.4570 KOps/s 20.7936 KOps/s $\color{#d91a1a}-1.62\%$
test_set_nested 88.7010μs 30.2643μs 33.0422 KOps/s 32.6305 KOps/s $\color{#35bf28}+1.26\%$
test_set_nested_new 0.1136ms 49.0185μs 20.4004 KOps/s 20.8234 KOps/s $\color{#d91a1a}-2.03\%$
test_select 0.1612ms 95.5640μs 10.4642 KOps/s 10.7843 KOps/s $\color{#d91a1a}-2.97\%$
test_unbind_speed 0.6452ms 0.6223ms 1.6069 KOps/s 1.6057 KOps/s $\color{#35bf28}+0.08\%$
test_unbind_speed_stack0 59.3599ms 7.7565ms 128.9241 Ops/s 128.1869 Ops/s $\color{#35bf28}+0.58\%$
test_unbind_speed_stack1 5.5200μs 0.9362μs 1.0681 MOps/s 884.6065 KOps/s $\textbf{\color{#35bf28}+20.75\%}$
test_creation[device0] 0.4623ms 0.3322ms 3.0104 KOps/s 2.9990 KOps/s $\color{#35bf28}+0.38\%$
test_creation_from_tensor 2.5206ms 0.3769ms 2.6534 KOps/s 2.6548 KOps/s $\color{#d91a1a}-0.05\%$
test_add_one[memmap_tensor0] 1.4125ms 30.9106μs 32.3513 KOps/s 32.4794 KOps/s $\color{#d91a1a}-0.39\%$
test_contiguous[memmap_tensor0] 29.8000μs 8.6391μs 115.7532 KOps/s 116.2193 KOps/s $\color{#d91a1a}-0.40\%$
test_stack[memmap_tensor0] 59.3000μs 25.7950μs 38.7671 KOps/s 38.7750 KOps/s $\color{#d91a1a}-0.02\%$
test_memmaptd_index 0.3419ms 0.3016ms 3.3159 KOps/s 3.3421 KOps/s $\color{#d91a1a}-0.78\%$
test_memmaptd_index_astensor 1.3155ms 1.1235ms 890.0772 Ops/s 901.5832 Ops/s $\color{#d91a1a}-1.28\%$
test_memmaptd_index_op 2.4196ms 2.3608ms 423.5870 Ops/s 424.3092 Ops/s $\color{#d91a1a}-0.17\%$
test_reshape_pytree 78.6000μs 32.4380μs 30.8280 KOps/s 31.0319 KOps/s $\color{#d91a1a}-0.66\%$
test_reshape_td 85.7010μs 39.8719μs 25.0803 KOps/s 25.1542 KOps/s $\color{#d91a1a}-0.29\%$
test_view_pytree 0.1099ms 31.7723μs 31.4740 KOps/s 31.4660 KOps/s $\color{#35bf28}+0.03\%$
test_view_td 28.1000μs 8.6565μs 115.5206 KOps/s 115.4040 KOps/s $\color{#35bf28}+0.10\%$
test_unbind_pytree 69.4010μs 37.7432μs 26.4949 KOps/s 26.8274 KOps/s $\color{#d91a1a}-1.24\%$
test_unbind_td 0.1105ms 91.5321μs 10.9251 KOps/s 10.7533 KOps/s $\color{#35bf28}+1.60\%$
test_split_pytree 68.3000μs 36.8646μs 27.1263 KOps/s 27.3854 KOps/s $\color{#d91a1a}-0.95\%$
test_split_td 0.7593ms 0.1038ms 9.6360 KOps/s 9.5299 KOps/s $\color{#35bf28}+1.11\%$
test_add_pytree 0.1066ms 45.3036μs 22.0733 KOps/s 22.4552 KOps/s $\color{#d91a1a}-1.70\%$
test_add_td 0.1338ms 71.8255μs 13.9226 KOps/s 14.1210 KOps/s $\color{#d91a1a}-1.40\%$
test_distributed 26.1000μs 8.1780μs 122.2788 KOps/s 124.4720 KOps/s $\color{#d91a1a}-1.76\%$
test_tdmodule 1.3235ms 26.7418μs 37.3946 KOps/s 38.8015 KOps/s $\color{#d91a1a}-3.63\%$
test_tdmodule_dispatch 0.2216ms 48.7342μs 20.5195 KOps/s 20.3897 KOps/s $\color{#35bf28}+0.64\%$
test_tdseq 45.1000μs 26.6103μs 37.5795 KOps/s 37.6648 KOps/s $\color{#d91a1a}-0.23\%$
test_tdseq_dispatch 0.1676ms 52.5278μs 19.0375 KOps/s 19.0033 KOps/s $\color{#35bf28}+0.18\%$
test_instantiation_functorch 1.9694ms 1.5476ms 646.1569 Ops/s 645.1065 Ops/s $\color{#35bf28}+0.16\%$
test_instantiation_td 2.0406ms 1.2683ms 788.4454 Ops/s 791.5549 Ops/s $\color{#d91a1a}-0.39\%$
test_exec_functorch 0.2143ms 0.1843ms 5.4250 KOps/s 5.4371 KOps/s $\color{#d91a1a}-0.22\%$
test_exec_td 0.2056ms 0.1770ms 5.6502 KOps/s 5.7706 KOps/s $\color{#d91a1a}-2.09\%$
test_vmap_mlp_speed[True-True] 6.4489ms 1.0577ms 945.4818 Ops/s 900.3915 Ops/s $\textbf{\color{#35bf28}+5.01\%}$
test_vmap_mlp_speed[True-False] 0.6759ms 0.5337ms 1.8739 KOps/s 1.8678 KOps/s $\color{#35bf28}+0.33\%$
test_vmap_mlp_speed[False-True] 5.9250ms 0.9118ms 1.0967 KOps/s 1.1057 KOps/s $\color{#d91a1a}-0.81\%$
test_vmap_mlp_speed[False-False] 5.7083ms 0.4292ms 2.3297 KOps/s 2.3320 KOps/s $\color{#d91a1a}-0.10\%$

Comment on lines 18 to 20
strategy:
matrix:
python_version: ["3.8", "3.9", "3.10", "3.11"]
Copy link
Contributor

Choose a reason for hiding this comment

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

Let's remove this part. The idea is to use only one Python version to avoid wasting resources on every PR.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Got it thanks! So this is a test that wheels can be built, but how do we actually build them for all python versions, then?

@vmoens vmoens added the ciflow/binaries/all Build all wheels label Oct 5, 2023
@vmoens vmoens merged commit e2de5ef into main Oct 26, 2023
42 of 46 checks passed
@vmoens vmoens deleted the m1_wheels branch October 26, 2023 20:40
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
CI ciflow/binaries/all Build all wheels CLA Signed This label is managed by the Facebook bot. Authors need to sign the CLA before a PR can be reviewed. versioning
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants