Beta function by 0x00b1 · Pull Request #78031 · pytorch/pytorch

0x00b1 · 2022-05-21T01:24:24Z

Euler beta function:

torch.special.beta(input, other, *, out=None) → Tensor

reentrant_gamma and reentrant_ln_gamma implementations (using Stirling’s approximation) are provided. I started working on this before I realized we were missing a gamma implementation (despite providing incomplete gamma implementations). Uses the coefficients computed by Steve Moshier to replicate SciPy’s implementation. Likewise, it mimics SciPy’s behavior (instead of the behavior in Cephes).

facebook-github-bot · 2022-05-21T01:24:29Z

🔗 Helpful links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/78031
📄 Preview Python docs built from this PR
📄 Preview C++ docs built from this PR
❓Need help or want to give feedback on the CI? Visit our office hours
↩️ [fb-only] Re-run with SSH instructions

❌ 1 New Failures

As of commit d17c6b4 (more details on the Dr. CI page):

Expand to see more

1/1 failures introduced in this PR

🕵️ 1 new failure recognized by patterns

The following CI failures do not appear to be due to upstream breakages

pull / linux-xenial-py3.7-gcc5.4 / test (backwards_compat, 1, 1, linux.2xlarge) (1/1)

Step: "Test" (full log | diagnosis details | 🔁 rerun)

2022-05-24T15:55:44.7985799Z The PR is introduc...m to confirm whether this change is wanted or not.

2022-05-24T15:55:44.7973563Z processing existing schema:  text(__torch__.torch.classes.profiling.SourceRef _0) -> (str _0)
2022-05-24T15:55:44.7974699Z processing existing schema:  count(__torch__.torch.classes.profiling.InstructionStats _0) -> (int _0)
2022-05-24T15:55:44.7975953Z processing existing schema:  duration_ns(__torch__.torch.classes.profiling.InstructionStats _0) -> (int _0)
2022-05-24T15:55:44.7977346Z processing existing schema:  source(__torch__.torch.classes.profiling.SourceStats _0) -> (__torch__.torch.classes.profiling.SourceRef _0)
2022-05-24T15:55:44.7978716Z processing existing schema:  line_map(__torch__.torch.classes.profiling.SourceStats _0) -> (Dict(int, __torch__.torch.classes.profiling.InstructionStats) _0)
2022-05-24T15:55:44.7979688Z processing existing schema:  __init__(__torch__.torch.classes.profiling._ScriptProfile _0) -> (NoneType _0)
2022-05-24T15:55:44.7981319Z processing existing schema:  enable(__torch__.torch.classes.profiling._ScriptProfile _0) -> (NoneType _0)
2022-05-24T15:55:44.7982211Z processing existing schema:  disable(__torch__.torch.classes.profiling._ScriptProfile _0) -> (NoneType _0)
2022-05-24T15:55:44.7983936Z processing existing schema:  _dump_stats(__torch__.torch.classes.profiling._ScriptProfile _0) -> (__torch__.torch.classes.profiling.SourceStats[] _0)
2022-05-24T15:55:44.7984974Z processing existing schema:  __init__(__torch__.torch.classes.dist_rpc.WorkerInfo _0, str _1, int _2) -> (NoneType _0)
2022-05-24T15:55:44.7985799Z The PR is introducing backward incompatible changes to the operator library. Please contact PyTorch team to confirm whether this change is wanted or not. 
2022-05-24T15:55:44.7986084Z 
2022-05-24T15:55:44.7986144Z Broken ops: [
2022-05-24T15:55:44.7986491Z 	prims::uniform(int[] shape, *, Scalar low, Scalar high, int dtype, Device device) -> (Tensor)
2022-05-24T15:55:44.7986928Z 	prims::empty_strided(int[] shape, int[] strides, *, int dtype, Device device, bool requires_grad) -> (Tensor)
2022-05-24T15:55:44.7997324Z 	prims::var(Tensor inp, int[]? dims, *, int correction, int? output_dtype=None) -> (Tensor)
2022-05-24T15:55:44.7997647Z 	prims::item(Tensor a) -> (Scalar)
2022-05-24T15:55:44.7997978Z 	prims::where(Tensor pred, Tensor a, Tensor b) -> (Tensor)
2022-05-24T15:55:44.7998290Z 	prims::cat(Tensor[] tensors, int dim) -> (Tensor)
2022-05-24T15:55:44.7998588Z 	prims::zeta(Tensor self, Tensor other) -> (Tensor)
2022-05-24T15:55:44.7998866Z 	prims::log10(Tensor self) -> (Tensor)

This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

aten/src/ATen/native/Math.h

aten/src/ATen/native/native_functions.yaml

docs/source/special.rst

torch/testing/_internal/common_methods_invocations.py

mruberry

Cool! Test failure is irrelevant.

I think this is fine to land as-is EXCEPT see my question about citing sources/licenses for the function implementations. If we need citations we should get them in with this PR

0x00b1 · 2022-05-24T21:01:21Z

@pytorchbot merge this please

github-actions · 2022-05-24T21:55:29Z

Hey @0x00b1.
You've committed this PR, but it does not have both a 'release notes: ...' and 'topics: ...' label. Please add one of each to the PR. The 'release notes: ...' label should represent the part of PyTorch that this PR changes (fx, autograd, distributed, etc) and the 'topics: ...' label should represent the kind of PR it is (not user facing, new feature, bug fix, perf improvement, etc). The list of valid labels can be found here for the 'release notes: ...' and here for the 'topics: ...'.
For changes that are 'topic: not user facing' there is no need for a release notes label.

jeffdaily · 2022-05-24T22:14:02Z

This broke the ROCm build on trunk.

https://ossci-raw-job-status.s3.amazonaws.com/log/6581698413

jeffdaily · 2022-05-24T22:20:39Z

The beta template function is only defined for the jiterator, not the non-jiterator code path.

suo · 2022-05-24T22:53:50Z

@pytorchbot revert -m "broke trunk, see the above message" -c nosignal

suo · 2022-05-24T22:54:12Z

If you'd like to run rocm tests on this PR, add the ciflow/trunk label.

This reverts commit da16450. Reverted #78031 on behalf of https://github.com/suo due to broke trunk, see the above message

0x00b1 · 2022-05-24T23:43:51Z

thanks @suo and @jeffdaily

facebook-github-bot added the cla signed label May 21, 2022

0x00b1 force-pushed the beta branch 2 times, most recently from 3061248 to 5bb1fa4 Compare May 21, 2022 15:43

0x00b1 changed the title ~~torch.special.beta~~ Beta function May 21, 2022

0x00b1 force-pushed the beta branch from fbd8807 to 83a41c7 Compare May 21, 2022 18:43

0x00b1 mentioned this pull request May 22, 2022

Gamma and Related Functions #78065

Open

25 tasks

0x00b1 force-pushed the beta branch 2 times, most recently from 79aff9b to a3bb1a2 Compare May 23, 2022 15:16

0x00b1 marked this pull request as ready for review May 23, 2022 18:00

0x00b1 requested review from albanD, anjali411, bdhirsh, mruberry, ngimel and soulitzer as code owners May 23, 2022 18:00

albanD removed their request for review May 23, 2022 18:01