Commits

Branch selector

 master

User selector

teijeong

Datepicker

All time

Commit History

Commits on Jan 17, 2022

• fix a typo

  Show description for 888bab7
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  888bab7

  Copy full SHA for 888bab7

Commits on Jan 7, 2022

• Added FusedBatchNormV3 op to list.

  Show description for afdfd2a
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  afdfd2a

  Copy full SHA for afdfd2a

Commits on Dec 13, 2021

• Fix whole-model verify mode to produce correct debug model when there are unquantizable ops

  Show description for fd97c76
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  fd97c76

  Copy full SHA for fd97c76

• [cleanup] Rename `quantized_op` to `quantizing_op`

  Show description for 1cada4f
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  1cada4f

  Copy full SHA for 1cada4f

Commits on Dec 9, 2021

• [Cleanup] Change struct to class for QuantizationPattern

  Show description for 025a88e
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  025a88e

  Copy full SHA for 025a88e

Commits on Dec 1, 2021

• Cleanup unnecessary code.

  Show description for 5681a9e
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  5681a9e

  Copy full SHA for 5681a9e

Commits on Nov 10, 2021

• PR #50615: [TFLite] Avoid passing negative quantized multipliers to MultiplyByQuantizedMultiplier

  Show description for e5a86e1
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  e5a86e1

  Copy full SHA for e5a86e1

• Handle tfr.cast canonicalization when the input and output cast elements types are different

  Show description for c8a99cc
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  c8a99cc

  Copy full SHA for c8a99cc

Commits on Oct 29, 2021

• Fix title MLIR converter README

  Show description for 837f4a5
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  837f4a5

  Copy full SHA for 837f4a5

Commits on Oct 21, 2021

• Improved logic to respect `rename_to` in API definitions in tfr_gen

  Show description for 4b27282
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  4b27282

  Copy full SHA for 4b27282

Commits on Oct 20, 2021

• Fix failing test. Increasing steps seems to resolves the issue.

  Show description for dcacb7a
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  dcacb7a

  Copy full SHA for dcacb7a

Commits on Oct 1, 2021

• Remove warnings regarding type inference during the build

  Show description for 119917f
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  119917f

  Copy full SHA for 119917f

• Add tfds update command to quantization debugger colab.

  Show description for af4115f
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  af4115f

  Copy full SHA for af4115f

Commits on Sep 17, 2021

• Change long long literals to use ll instead of LL, to improve readability

  Show description for d62bab8
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  d62bab8

  Copy full SHA for d62bab8

Commits on Sep 15, 2021

• Remove unused functions

  Show description for 1f956fe
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  1f956fe

  Copy full SHA for 1f956fe

Commits on Sep 12, 2021

• Documentation for quantization debugger and selective quantization

  Show description for af20155
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  af20155

  Copy full SHA for af20155

Commits on Sep 9, 2021

• Preserve converter's optimization value.

  Show description for d2e0e79
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  d2e0e79

  Copy full SHA for d2e0e79

• Add QuantizationDebugger to Python API

  Show description for 0165a05
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  0165a05

  Copy full SHA for 0165a05

Commits on Sep 7, 2021

• Remove volatile QDQ for input and outputs in MLIR quantizer

  Show description for bfe7329
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  bfe7329

  Copy full SHA for bfe7329

Commits on Sep 3, 2021

• Only handle static shapes in optimiza_op_order.cc

  Show description for cb162f1
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  cb162f1

  Copy full SHA for cb162f1

Commits on Aug 24, 2021

• Add bias adjustments when the scale for the bias would cause overflow.

  Show description for 01108a0
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  01108a0

  Copy full SHA for 01108a0

Commits on Aug 23, 2021

• Expand calibrated ranges that are too narrow

  Show description for 44bcb14
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  44bcb14

  Copy full SHA for 44bcb14

Commits on Aug 21, 2021

• Add bias adjustments when the scale for the bias would cause overflow.

  Show description for 3994a73
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  3994a73

  Copy full SHA for 3994a73

Commits on Aug 6, 2021

• Fix post processing in flatbuffer import to traverse all users of a constant

  Show description for 6534013
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  6534013

  Copy full SHA for 6534013

Commits on Aug 3, 2021

• Resolve outputs with fixed element types correctly

  Show description for 8c3a8a1
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  8c3a8a1

  Copy full SHA for 8c3a8a1

• [TFR] Improve error message on undefined default value

  Show description for 62f0dee
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  62f0dee

  Copy full SHA for 62f0dee

Commits on Jul 30, 2021

• Add handling of ParseExample ops to LiftTfliteFlexOps pass

  Show description for 41eae57
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  41eae57

  Copy full SHA for 41eae57

• Add out_type attribute to TFR ops, when the output type information is not readily available yet needed.

  Show description for 04bf8d7
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  04bf8d7

  Copy full SHA for 04bf8d7

Commits on Jul 29, 2021

• Move LiftTfLiteFlexOps pass to tensorflow/compiler/mlir/lite

  Show description for adff352
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  adff352

  Copy full SHA for adff352

Commits on Jul 27, 2021

• Add correct import logic for string constants

  Show description for 879b758
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  879b758

  Copy full SHA for 879b758

• Change RewriteQuantizedIO pass to module-level pass

  Show description for 9bf0615
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  9bf0615

  Copy full SHA for 9bf0615

Commits on Jul 24, 2021

• Change QuantScaleFactor op to match constant pattern, and QuantRescale op to handle zero point correctly.

  Show description for 33bbc58
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  33bbc58

  Copy full SHA for 33bbc58

• internal code change

  Show description for c0c1f18
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  c0c1f18

  Copy full SHA for c0c1f18

Commits on Jul 21, 2021

• Remove quantization parameters from input and output values

  Show description for 2dae99d
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  2dae99d

  Copy full SHA for 2dae99d

• Minor refactoring: change `struct` to `class`

  Show description for bd35438
  

  teijeong

  authored and

  tensorflower-gardener

  committed
  bd35438

  Copy full SHA for bd35438

Pagination

Previous

Next