Commits

Branch selector

 master

User selector

ssnl

All time

Commit History

Commits on Oct 30, 2017

• make sparse (new) functions conform that storage is not NULL (pytorch#3381)

  

  ssnl

  authored and

  soumith

  committedOct 30, 2017
  8cd0df0

Commits on Oct 28, 2017

• add tests for the previously failing coalesce case

  

  ssnl

  authored and

  ezyang

  committedOct 28, 2017
  4f33b13

• fix sparse tensor coalesce

  

  ssnl

  authored and

  ezyang

  committedOct 28, 2017
  0b89a68

• fix pynew gpu_guards

  

  ssnl

  authored and

  ezyang

  committedOct 28, 2017
  bb7b630

• test sparse dp, broadcast_coalesced, reduce_add_coalesced

  

  ssnl

  authored and

  ezyang

  committedOct 28, 2017
  91a8d33

• sparse broadcast_coalesce and reduce_add_coalesced

  

  ssnl

  authored and

  ezyang

  committedOct 28, 2017
  01be4d6

• fix sparse tensor .cpu()

  

  ssnl

  authored and

  ezyang

  committedOct 28, 2017
  3a0aee7

• Add sparseTensor.new wrapper bindings (pytorch#3329)

  

  ssnl

  authored and

  apaszke

  committedOct 28, 2017
  9107110

Commits on Oct 27, 2017

• raw text (pytorch#3327)

  

  ssnl

  authored and

  soumith

  committedOct 27, 2017
  de1f4e6

Commits on Oct 26, 2017

• Dynamically find min log scale pytorch#3289

  

  ssnl

  authored and

  soumith

  committedOct 26, 2017
  86d0c24

• Add zero, zeros_like, _dimI and _dimV for sparse tensors (pytorch#3271)

  

  ssnl

  authored and

  apaszke

  committedOct 26, 2017
  bdeee47

Commits on Oct 24, 2017

• update

  

  ssnl

  authored and

  apaszke

  committedOct 24, 2017
  0b8b9cf

• smarter backend option

  

  ssnl

  authored and

  zdevito

  committedOct 24, 2017
  d131219

• sparse cuda and get device (pytorch#122)

  

  ssnl

  authored and

  soumith

  committedOct 24, 2017
  1a0e4e1

Commits on Oct 20, 2017

• isContiguous problems (pytorch#3148)

  

  ssnl

  authored and

  ezyang

  committedOct 20, 2017
  634c831

Commits on Oct 16, 2017

• Change device_id to device in python land (pytorch#3133)

  

  ssnl

  authored and

  soumith

  committedOct 16, 2017
  fce3ed1

• dense buffer (pytorch#3139)

  

  ssnl

  authored and

  soumith

  committedOct 16, 2017
  ba05dc5

Commits on Oct 14, 2017

• doc (pytorch#3110)

  

  ssnl

  authored and

  soumith

  committedOct 14, 2017
  6dc67ae

• Limit print scale by sys.float_info (pytorch#3113)

  

  ssnl

  authored and

  soumith

  committedOct 14, 2017
  38f87cc

Commits on Oct 13, 2017

• More shape checking for ConvNd (pytorch#3052)

  

  ssnl

  authored and

  soumith

  committedOct 13, 2017
  8f26d6a

Commits on Oct 11, 2017

• Add random_ for cuda, fix random_ for cpu (pytorch#3042)

  

  ssnl

  authored and

  soumith

  committedOct 11, 2017
  2000ba0

• Fix a typo in optim.rst (pytorch#3069)

  

  ssnl

  authored and

  soumith

  committedOct 11, 2017
  9260f0e

• Add document on how Module.cuda() and optims should work together (pytorch#3056)

  

  ssnl

  authored and

  soumith

  committedOct 11, 2017
  828048f

Commits on Oct 10, 2017

• add kwargs_only defaults for sorted and largest

  

  ssnl

  authored and

  soumith

  committedOct 10, 2017
  a479487

Commits on Oct 7, 2017

• Fix isContiguousDim (pytorch#3011)

  

  ssnl

  authored and

  soumith

  committedOct 7, 2017
  dc1b4ff

Commits on Oct 6, 2017

• assert reflection padding in range (pytorch#3008)

  

  ssnl

  authored and

  soumith

  committedOct 6, 2017
  0eec332

• lint

  

  ssnl

  authored and

  soumith

  committedOct 6, 2017
  b2f5ccf

• Fix two legacy modules clearing input tensor in clearState

  

  ssnl

  authored and

  soumith

  committedOct 6, 2017
  0c29575

• Update all existing nn tests to new args format; Move all randomness inside tests

  

  ssnl

  authored and

  soumith

  committedOct 6, 2017
  ecdb86e

• Remove top-level seed setting

  

  ssnl

  authored and

  soumith

  committedOct 6, 2017
  b6e1dd2

• Change TestCase args to accept value, size or fn for constructor_args, input and target

  

  ssnl

  authored and

  soumith

  committedOct 6, 2017
  c76e290

Commits on Oct 4, 2017

• Fix BN size check in eval mode (pytorch#2977)

  

  ssnl

  authored and

  soumith

  committedOct 4, 2017
  ba766ef

• Fix scatter size check (pytorch#2960)

  

  ssnl

  authored and

  soumith

  committedOct 4, 2017
  f608208

Commits on Oct 3, 2017

• Raise error when each channel only has 1 value in batch norm (pytorch#2961)

  

  ssnl

  authored and

  soumith

  committedOct 3, 2017
  faa6fdf

• generates non-equal random tensor for max pool

  

  ssnl

  authored and

  soumith

  committedOct 3, 2017
  1ff34a0

Pagination

PreviousNext