Commits

Branch selector

 master

User selector

lezcano

Datepicker

All time

Commit History

Commits on Sep 1, 2023

• [dynamo] Avoid recompilation when the PyTorch function accepts scalars (pytorch#108162)

  Show description for 2a6ef9b
  

  lezcano

  authored and

  pytorchmergebot

  committed
  2a6ef9b

  Copy full SHA for 2a6ef9b

Commits on Aug 31, 2023

• Add refs/decomps for dot/vdot (pytorch#108194)

  Show description for 239ee76
  

  lezcano

  authored and

  pytorchmergebot

  committed
  239ee76

  Copy full SHA for 239ee76

• Add reference for linalg.vecdot (pytorch#108188)

  Show description for 239fed7
  

  lezcano

  authored and

  pytorchmergebot

  committed
  239fed7

  Copy full SHA for 239fed7

Commits on Aug 29, 2023

• Add AMP support to linalg.vecdot. (pytorch#108165)

  Show description for 86bc50a
  

  lezcano

  authored and

  pytorchmergebot

  committed
  86bc50a

  Copy full SHA for 86bc50a

Commits on Aug 23, 2023

• Revert the removal of a SampleInput for gather (pytorch#107776)

  Show description for 977aba7
  

  lezcano

  authored and

  pytorchmergebot

  committed
  977aba7

  Copy full SHA for 977aba7

• [dynamo] Wrap ndarray dunder methods (pytorch#107689)

  Show description for 207b06d
  

  lezcano

  authored and

  pytorchmergebot

  committed
  207b06d

  Copy full SHA for 207b06d

• [dynamo] Fix ndarray.__pow__ (pytorch#107746)

  Show description for b5c90ba
  

  lezcano

  authored and

  pytorchmergebot

  committed
  b5c90ba

  Copy full SHA for b5c90ba

• Wrap indirect indexing on CUDA (pytorch#105055)

  Show description for 2b6249e
  

  lezcano

  authored and

  pytorchmergebot

  committed
  2b6249e

  Copy full SHA for 2b6249e

• Dispatch take_along_axis to gather (pytorch#107711)

  Show description for fada052
  

  lezcano

  authored and

  pytorchmergebot

  committed
  fada052

  Copy full SHA for fada052

• [dynamo] np.sort(complex) is not implemented (pytorch#107710)

  Show description for 62113a2
  

  lezcano

  authored and

  pytorchmergebot

  committed
  62113a2

  Copy full SHA for 62113a2

• Support negative indices in ndarray.__getitem__ (pytorch#107688)

  Show description for 2fc8283
  

  lezcano

  authored and

  pytorchmergebot

  committed
  2fc8283

  Copy full SHA for 2fc8283

• Add a flag that allows breaking on NumPy ops (pytorch#107687)

  Show description for db39a81
  

  lezcano

  authored and

  pytorchmergebot

  committed
  db39a81

  Copy full SHA for db39a81

Commits on Aug 22, 2023

• Quick return when there's nothing to bound in bound_sympy (pytorch#107549)

  Show description for f131016
  

  lezcano

  authored and

  pytorchmergebot

  committed
  f131016

  Copy full SHA for f131016

• Wrap indirect indexing on CUDA (pytorch#105055)

  Show description for 85c673e
  

  lezcano

  authored and

  pytorchmergebot

  committed
  85c673e

  Copy full SHA for 85c673e

Commits on Aug 21, 2023

• Guard numpy imports in the dynamo folder (pytorch#107299)

  Show description for 612c8a8
  

  lezcano

  authored and

  pytorchmergebot

  committed
  612c8a8

  Copy full SHA for 612c8a8

• Don't use thrust::log(complex) in CUDA as it takes a FOREVER to compile (pytorch#107559)

  Show description for 6dea992
  

  lezcano

  authored and

  pytorchmergebot

  committed
  6dea992

  Copy full SHA for 6dea992

Commits on Aug 20, 2023

• Enable a number inductor of tests on CPU (pytorch#107465)

  Show description for c1cc74c
  

  lezcano

  authored and

  pytorchmergebot

  committed
  c1cc74c

  Copy full SHA for c1cc74c

Commits on Aug 19, 2023

• Trace through Tensor slots (pytorch#107159)

  Show description for 4eac43d
  

  lezcano

  authored and

  pytorchmergebot

  committed
  4eac43d

  Copy full SHA for 4eac43d

Commits on Aug 14, 2023

• [Inductor] Make softshrink composite implicit (pytorch#107052)

  Show description for 2c5f96d
  

  lezcano

  authored and

  pytorchmergebot

  committed
  2c5f96d

  Copy full SHA for 2c5f96d

• Simplify sign lowering in triton (pytorch#107051)

  Show description for 6d89957
  

  lezcano

  authored and

  pytorchmergebot

  committed
  6d89957

  Copy full SHA for 6d89957

• Make hardshrink's decomp composite implicit (pytorch#107039)

  Show description for 3b1254e
  

  lezcano

  authored and

  pytorchmergebot

  committed
  3b1254e

  Copy full SHA for 3b1254e

• Simplify some decompositions. (pytorch#107038)

  Show description for 45c7880
  

  lezcano

  authored and

  pytorchmergebot

  committed
  45c7880

  Copy full SHA for 45c7880

Commits on Aug 11, 2023

• NumPy support in torch.compile (pytorch#106211)

  Show description for a9dca53
  

  lezcano

  authored and

  pytorchmergebot

  committed
  a9dca53

  Copy full SHA for a9dca53

Commits on Jul 26, 2023

• Update matmul decomp to match eager (pytorch#105850)

  Show description for 36ae359
  

  lezcano

  authored and

  pytorchmergebot

  committed
  36ae359

  Copy full SHA for 36ae359

• Fix the docs for cosine_similarity (pytorch#104772)

  Show description for 9bde7f4
  

  lezcano

  authored and

  pytorchmergebot

  committed
  9bde7f4

  Copy full SHA for 9bde7f4

• Fuse ops in eager cosine_similarity while keeping the stability and the gradients (pytorch#104771)

  Show description for fff4a9d
  

  lezcano

  authored and

  pytorchmergebot

  committed
  fff4a9d

  Copy full SHA for fff4a9d

Commits on Jul 17, 2023

• Better comparisons for np.ndarrays in dynamo (pytorch#105333)

  Show description for a26afb9
  

  lezcano

  authored and

  pytorchmergebot

  committed
  a26afb9

  Copy full SHA for a26afb9

• Assert that we can compute the bounds for guards using rational numbers (pytorch#105139)

  Show description for 3152fea
  

  lezcano

  authored and

  pytorchmergebot

  committed
  3152fea

  Copy full SHA for 3152fea

• Prefer bound_sympy over sympy_interp (pytorch#105138)

  Show description for 34c91a7
  

  lezcano

  authored and

  pytorchmergebot

  committed
  34c91a7

  Copy full SHA for 34c91a7

• Bound just size variables in bound_sympy (pytorch#105155)

  Show description for eae99b0
  

  lezcano

  authored and

  pytorchmergebot

  committed
  eae99b0

  Copy full SHA for eae99b0

Commits on Jul 14, 2023

• ValueRange analysis for indirect indexing (pytorch#102611)

  Show description for c099b7e
  

  lezcano

  authored and

  pytorchmergebot

  committed
  c099b7e

  Copy full SHA for c099b7e

• Fix ranges for range vars (pytorch#104987)

  Show description for 87a3ed5
  

  lezcano

  authored and

  pytorchmergebot

  committed
  87a3ed5

  Copy full SHA for 87a3ed5

• Remove unnecessary casting in triton (pytorch#104975)

  Show description for 88dcecd
  

  lezcano

  authored and

  pytorchmergebot

  committed
  88dcecd

  Copy full SHA for 88dcecd

Commits on Jul 13, 2023

• Perform value range analysis with rationals when possible (pytorch#105137)

  Show description for d1fedad
  

  lezcano

  authored and

  pytorchmergebot

  committed
  d1fedad

  Copy full SHA for d1fedad

Commits on Jul 12, 2023

• Fix bug when an index appears in two expressions (pytorch#104886)

  Show description for 398606e
  

  lezcano

  authored and

  pytorchmergebot

  committed
  398606e

  Copy full SHA for 398606e

Pagination

Previous

Next