Commits

Branch selector

 main

User selector

jerryzh168

Datepicker

All time

Commit History

Commits on Apr 26, 2024

• [parametrization] fix requires_grad propagation (pytorch#124888)

  Show description for 74afccd
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  74afccd

  Copy full SHA for 74afccd

Commits on Mar 27, 2024

• [quant][pt2e] Enable observer sharing between different quantization specs (pytorch#122734)

  Show description for 5af839f
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  5af839f

  Copy full SHA for 5af839f

Commits on Mar 19, 2024

• [quant][pt2e] Add support for conv transpose + bn + {relu} weights fusion in PTQ (pytorch#122046)

  Show description for 901ba2b
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  901ba2b

  Copy full SHA for 901ba2b

Commits on Mar 9, 2024

• [quant] Add error check for input_edge annotation (pytorch#121536)

  Show description for a6a67da
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  a6a67da

  Copy full SHA for a6a67da

Commits on Feb 9, 2024

• [quant][pt2e][bc-breaking] Set fold_quantize to True in convert_pt2e (pytorch#119425)

  Show description for 7082e24
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  7082e24

  Copy full SHA for 7082e24

Commits on Feb 7, 2024

• [quant][pt2e][bc-breaking] Remove fold_quantize flag (pytorch#118701)

  Show description for 482d952
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  482d952

  Copy full SHA for 482d952

Commits on Jan 30, 2024

• [quant][bc-breaking] Turn on fold_quantize by default (pytorch#118605)

  Show description for 82a7460
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  82a7460

  Copy full SHA for 82a7460

Commits on Jan 24, 2024

• [quant][pt2e] Add fold_quantize=True for all convert_pt2e calls (pytorch#117797)

  Show description for af1ebc4
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  af1ebc4

  Copy full SHA for af1ebc4

Commits on Jan 17, 2024

• [quant] Support custom qmin/qmax for activation and weight for xnnpack quantizer (pytorch#117305)

  Show description for 8f1bc87
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  8f1bc87

  Copy full SHA for 8f1bc87

Commits on Jan 13, 2024

• Add uint1 to uint7 dtypes (pytorch#117208)

  Show description for 3e397ce
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  3e397ce

  Copy full SHA for 3e397ce

Commits on Jan 8, 2024

• [quant][be] enable xnnpack_quantizer tests to run in internal CI (pytorch#116911)

  Show description for 28e2e12
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  28e2e12

  Copy full SHA for 28e2e12

Commits on Jan 3, 2024

• [quant][pt2e] Preserve numeric_debug_handle in quantization flows (pytorch#116477)

  Show description for 41f265b
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  41f265b

  Copy full SHA for 41f265b

Commits on Dec 28, 2023

• [quant][be] Skip conv-bn folding when there are no batchnorm ops (pytorch#116440)

  Show description for 8173d98
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  8173d98

  Copy full SHA for 8173d98

Commits on Dec 22, 2023

• [quant][pt2e] Relax constraints on dtype and qscheme to allow for customizations (pytorch#116287)

  Show description for db25462
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  db25462

  Copy full SHA for db25462

Commits on Dec 8, 2023

• [quant][pt2e] Add convert callback to Observer module (pytorch#115001)

  Show description for cc8f6f5
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  cc8f6f5

  Copy full SHA for cc8f6f5

Commits on Dec 7, 2023

• [quant][pt2e] XNNPACKQuantizer skip inserting observers for non-float Tensors (pytorch#114999)

  Show description for ecba053
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  ecba053

  Copy full SHA for ecba053

• [quant][be] Add a test for per channel quant for groupwise conv (pytorch#115224)

  Show description for a93b9ee
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  a93b9ee

  Copy full SHA for a93b9ee

Commits on Dec 4, 2023

• [quant][pt2e][xnnpack] Add support for QAT dynamic quantization for linear in XNNPACKQuantizer (pytorch#113288)

  Show description for 1474dad
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  1474dad

  Copy full SHA for 1474dad

Commits on Dec 2, 2023

• [quant][pt2e][xnnpack] XNNPACKQuantizer skip quantization for input and output to workaround histogram observer problem (pytorch#113405)

  Show description for 8f16401
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  8f16401

  Copy full SHA for 8f16401

Commits on Dec 1, 2023

• [quant][pt2e] Add generate_numeric_debug_handle pass (pytorch#114315)

  Show description for 64fd706
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  64fd706

  Copy full SHA for 64fd706

Commits on Nov 29, 2023

• [quant][pt2e] Fix the order for implicit sharing code (pytorch#114704)

  Show description for d2f4215
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  d2f4215

  Copy full SHA for d2f4215

Commits on Nov 22, 2023

• [quant][pt2e] Refactor some internal code for observer insertion (pytorch#113500)

  Show description for d6578b3
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  d6578b3

  Copy full SHA for d6578b3

• [reland][quant][pt2e] Refactor insert observer to do sharing checking in the same place (pytorch#113458) (pytorch#113920)

  Show description for a785fbe
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  a785fbe

  Copy full SHA for a785fbe

Commits on Nov 15, 2023

• [quant][pt2e] Refactor insert observer to do sharing checking in the same place (pytorch#113458)

  Show description for 585e315
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  585e315

  Copy full SHA for 585e315

Commits on Nov 9, 2023

• [quant][pt2e] Add transform_for_annotation method in Quantizer (pytorch#113115)

  Show description for 501d118
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  501d118

  Copy full SHA for 501d118

Commits on Nov 8, 2023

• [qunat][pt2e] Support allow_implicit_sharing flag (pytorch#112929)

  Show description for 12c257c
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  12c257c

  Copy full SHA for 12c257c

Commits on Nov 2, 2023

• [quant][docs] Add x86 inductor quant docs (pytorch#112648)

  Show description for 6929ebf
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  6929ebf

  Copy full SHA for 6929ebf

Commits on Oct 31, 2023

• [reland][quant][pt2e][be] Cleanup observer insertion logic (pytorch#111828) (pytorch#112453)

  Show description for 3db0095
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  3db0095

  Copy full SHA for 3db0095

Commits on Oct 25, 2023

• [quant][pt2e][be] Cleanup observer insertion logic (pytorch#111828)

  Show description for bf998a2
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  bf998a2

  Copy full SHA for bf998a2

• [quant][be] Clean up prepare code (pytorch#111827)

  Show description for 6e2dfb3
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  6e2dfb3

  Copy full SHA for 6e2dfb3

Commits on Oct 20, 2023

• [quant][pt2e] Actually support transitive sharing for SharedQuantizationSpec (pytorch#111172)

  Show description for 43c211f
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  43c211f

  Copy full SHA for 43c211f

Commits on Oct 12, 2023

• [quant][pt2e][be] Move xnnpack quantizer tests to separate file (pytorch#111004)

  Show description for e0ddc3f
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  e0ddc3f

  Copy full SHA for e0ddc3f

Commits on Oct 11, 2023

• [quant][pt2e] Disable remove_qconfig (pytorch#111000)

  Show description for d589106
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  d589106

  Copy full SHA for d589106

Commits on Oct 10, 2023

• [fx][subgraph_matcher] Add a matcher that supports name to node map (pytorch#110743)

  Show description for 7a69e3d
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  7a69e3d

  Copy full SHA for 7a69e3d

Commits on Oct 7, 2023

• [release_notes] Some updates after 2.1 release (pytorch#110771)

  Show description for 5cc1a38
  

  jerryzh168

  authored and

  pytorchmergebot

  committed
  5cc1a38

  Copy full SHA for 5cc1a38

Pagination

Previous

Next