Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 7, 2024

• v0.2.3

  calad0i

  committed
  36d1f2d

• Add Permute layer

  calad0i

  committed
  412e4a7

Commits on Jul 22, 2024

• add batchnorm layer (train only)

  calad0i

  committed
  a4ffb35

• support 'parallel_factor' for multi dim dense

  calad0i

  committed
  0049e96

Commits on Jul 12, 2024

• plain keras attn layer

  calad0i

  committed
  e01036e

Commits on Jun 27, 2024

• v0.2.2

  calad0i

  committed
  e2657ca

Commits on Jun 17, 2024

• "type: ignore"

  calad0i

  committed
  717b56f

• allow nested layers

  calad0i

  committed
  eafebc7

• allow arbitrary quantizer in qkeras

  calad0i

  committed
  2b85706

Commits on May 14, 2024

• paper link again

  calad0i

  committed
  419d895

• compatibility with np_config.enable_numpy_behavior()

  calad0i

  committed
  8157ad2

Commits on May 6, 2024

• + arxiv badge

  calad0i

  committed
  bb5fe40

Commits on Apr 26, 2024

• v0.2.1

  calad0i

  committed
  599a5e4

• update docs

  calad0i

  committed
  c5b7d39

• fix error msg in test

  calad0i

  committed
  a5f3a3f

• reduce cover factor to ensure overflow in tests

  calad0i

  committed
  bb4f5ec

• fix: wrong pool layer req bw cause overflow

  calad0i

  committed
  2ba9ded

• regiser all layers serializable

  calad0i

  committed
  02d4b73

Commits on Apr 6, 2024

• update rescription regarding tf version

  calad0i

  committed
  f67870b

• tf 2.15 test passed, change dependency; bump to 0.2.0.post1

  calad0i

  committed
  e6dc16e

• update logo

  calad0i

  committed
  dbd4936

Commits on Feb 25, 2024

• minor int bit deduce change

  calad0i

  committed
  caa0ea7

Commits on Feb 21, 2024

• v0.2.0

  calad0i

  committed
  253e051

• minor doc string change

  calad0i

  committed
  4d611fa

• update docs

  calad0i

  committed
  b50d38c

• update tests, allow bit-accuract avg pool

  calad0i

  committed
  3e8d084

• workaround: allow synthesis model with empty layers

  calad0i

  committed
  af68896

• fix: exact bw, handle layer removed at first passs in layer_xf

  calad0i

  committed
  94861c1

• minor change: 0bit number format

  calad0i

  committed
  eae5fb9

Commits on Jan 26, 2024

• fix exact bits format

  calad0i

  committed
  f99bf02

• pass parallelization_factor, rename rule change

  calad0i

  committed
  77c882e

Commits on Jan 25, 2024

• more exact bops computation

  calad0i

  committed
  100e35a

Commits on Jan 21, 2024

• minor fix for non-biased conv/dense

  calad0i

  committed
  3721922

Commits on Jan 19, 2024

• more reasonable max LUT size

  calad0i

  committed
  bc0353a

Commits on Jan 15, 2024

• minor fix

  calad0i

  committed
  0be26b2

Pagination

PreviousNext