Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Nov 19, 2020

• Implement C++ ModuleDict (#47707)

  Show description for c542614
  

  ejguan

  authored and

  facebook-github-bot

  committed
  c542614

  Copy full SHA for c542614

• Pass any verbosity from test/run_test.py to pytest (#48204)

  Show description for c4a6df9
  

  samestep

  authored and

  facebook-github-bot

  committed
  c4a6df9

  Copy full SHA for c4a6df9

• batched grad for binary_cross_entropy, symeig (#48057)

  Show description for 370310b
  

  zou3519

  authored and

  facebook-github-bot

  committed
  370310b

  Copy full SHA for 370310b

• Add c10d new frontend to build (#48146)

  Show description for db767b7
  

  gmagogsfm

  authored and

  facebook-github-bot

  committed
  db767b7

  Copy full SHA for db767b7

• [Gradient Compression] PowerSGD comm hook (#48060)

  Show description for daff3a8
  

  Yi Wang

  authored and

  facebook-github-bot

  committed
  daff3a8

  Copy full SHA for daff3a8

• Make softmax and log_softmax handle negative dims, add tests (#48156)

  Show description for 0d8ddb5
  

  Krovatkin

  authored and

  facebook-github-bot

  committed
  0d8ddb5

  Copy full SHA for 0d8ddb5

• T78750158 Support varying size input in numeric suite at 10/30/2020, 3:55:01 PM (#47391)

  Show description for 46d846f
  

  DeepDriving

  authored and

  facebook-github-bot

  committed
  46d846f

  Copy full SHA for 46d846f

• Implement igammac (3rd PR) (#48171)

  Show description for 8819bad
  

  mfkasim1

  authored and

  facebook-github-bot

  committed
  8819bad

  Copy full SHA for 8819bad

• [PT][StaticRuntime] Move prim op impl to ops.cpp (#48210)

  Show description for c5dae33
  

  Hao Lu

  authored and

  facebook-github-bot

  committed
  c5dae33

  Copy full SHA for c5dae33

• [te] Fix pow (#48213)

  Show description for 6da26fe
  

  bertmaher

  authored and

  facebook-github-bot

  committed
  6da26fe

  Copy full SHA for 6da26fe

• [quant][refactor] Move some util functions from torch/quantization/fx/utils.py to torch/quantization/utils.py (#48107)

  Show description for ed57f80
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  ed57f80

  Copy full SHA for ed57f80

• [FX] Refactor unique name handling (#48205)

  Show description for 4316bf9
  

  James Reed

  authored and

  facebook-github-bot

  committed
  4316bf9

  Copy full SHA for 4316bf9

• [PyTorch] Return raw ptr from ThreadLocalDebugInfo::get() (#47796)

  Show description for bef460a
  

  swolchok

  authored and

  facebook-github-bot

  committed
  bef460a

  Copy full SHA for bef460a

• [quant][fix][ez] Fix quant_type classification for fp16, fp16 (#48073)

  Show description for 5883e0b
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  5883e0b

  Copy full SHA for 5883e0b

• [Person Seg] Compress the person seg model (#48008)

  Show description for 773d1f3
  

  xta0

  authored and

  facebook-github-bot

  committed
  773d1f3

  Copy full SHA for 773d1f3

• Get TestTorch.test_empty_meta working again (#48113)

  Show description for a97d059
  

  ezyang

  authored and

  facebook-github-bot

  committed
  a97d059

  Copy full SHA for a97d059

• [PyTorch] Narrow Device to 2 bytes by narrowing DeviceType and DeviceIndex (#47023)

  Show description for 4c9eb57
  

  swolchok

  authored and

  facebook-github-bot

  committed
  4c9eb57

  Copy full SHA for 4c9eb57

• [quant] FakeQuantize inherit from FakeQuantizeBase (#48072)

  Show description for 72918e4
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  72918e4

  Copy full SHA for 72918e4

• Suppress "ioctl points to uninitialised" check (#48187)

  Show description for efeb988
  

  malfet

  authored and

  facebook-github-bot

  committed
  efeb988

  Copy full SHA for efeb988

• [quant][fix] Fix quant type classification for float_qparam qconfig (#48069)

  Show description for 576fa09
  

  jerryzh168

  authored and

  facebook-github-bot

  committed
  576fa09

  Copy full SHA for 576fa09

• Introducing winograd transformed fp16 nnpack to PT for unet 106 (#47925)

  Show description for f0f8b97
  

  kimishpatel

  authored and

  facebook-github-bot

  committed
  f0f8b97

  Copy full SHA for f0f8b97

• [PyTorch] Make RecordFunction::active private (#47549)

  Show description for 383abf1
  

  swolchok

  authored and

  facebook-github-bot

  committed
  383abf1

  Copy full SHA for 383abf1

• [PyTorch][JIT] Skip unnecessary refcounting in TensorType::merge (#47959)

  Show description for 1bafff2
  

  swolchok

  authored and

  facebook-github-bot

  committed
  1bafff2

  Copy full SHA for 1bafff2

• Removing non-thread-safe log statement from ReinitializeTensor (#48185)

  Show description for 0f89be6
  

  Max Ouellet

  authored and

  facebook-github-bot

  committed
  0f89be6

  Copy full SHA for 0f89be6

• pass strict_fuser_check for recursive fusion (#47221)

  Show description for 4360486
  

  jjsjann123

  authored and

  facebook-github-bot

  committed
  4360486

  Copy full SHA for 4360486

• Revert D24853669: [pytorch][PR] Migrate `eig` from the TH to Aten (CUDA)

  Show description for ea1e78a
  

  Mike Ruberry

  authored and

  facebook-github-bot

  committed
  ea1e78a

  Copy full SHA for ea1e78a

Commits on Nov 18, 2020

• fft: Generalize fill with conjugate symmetry and use complex dtypes (#46908)

  Show description for 2fbd70d
  

  peterbell10

  authored and

  facebook-github-bot

  committed
  2fbd70d

  Copy full SHA for 2fbd70d

• move Tensor comparisons back to C (#48018)

  Show description for 0639387
  

  Taylor Robie

  authored and

  facebook-github-bot

  committed
  0639387

  Copy full SHA for 0639387

• move aten::round to lite interpreter (#45931)

  Show description for ed4dd86
  

  linbinyu

  authored and

  facebook-github-bot

  committed
  ed4dd86

  Copy full SHA for ed4dd86

• [pytorch][codegen] simplify python signature creation logic (#47977)

  Show description for a36e646
  

  ljk53

  authored and

  facebook-github-bot

  committed
  a36e646

  Copy full SHA for a36e646

• [pytorch][codegen] simplify dunder method check in gen_python_functions.py (#47976)

  Show description for 5eaf856
  

  ljk53

  authored and

  facebook-github-bot

  committed
  5eaf856

  Copy full SHA for 5eaf856

• [pytorch][codegen] remove dead code in gen_variable_type.py (#47975)

  Show description for 5243456
  

  ljk53

  authored and

  facebook-github-bot

  committed
  5243456

  Copy full SHA for 5243456

• [tensorexpr] Switch cpp tests to pure gtest (#48160)

  Show description for 07657b6
  

  bertmaher

  authored and

  facebook-github-bot

  committed
  07657b6

  Copy full SHA for 07657b6

• [te][benchmark] Add more optimized versions of gemm (#48159)

  Show description for 464d23e
  

  bertmaher

  authored and

  facebook-github-bot

  committed
  464d23e

  Copy full SHA for 464d23e

• [te] Make BUILD_TENSOREXPR_BENCHMARK a real CMake option (#48158)

  Show description for 8a996dd
  

  bertmaher

  authored and

  facebook-github-bot

  committed
  8a996dd

  Copy full SHA for 8a996dd

Pagination

Previous

Next