Commits

Breadcrumbs

History for

onmaster

User selector

All users

All time

Commit History

Commits on Apr 2, 2018

• [FasterRCNN] include anchors at the boundary

  ppwwyyxx

  committed
  92a9315

  Copy full SHA for 92a9315

Commits on Mar 31, 2018

• [FasterRCNN] pad roi

  ppwwyyxx

  committed
  07fcdcc

  Copy full SHA for 07fcdcc

Commits on Mar 11, 2018

• update docs

  ppwwyyxx

  committed
  f130c10

  Copy full SHA for f130c10

Commits on Mar 4, 2018

• small update & bump version

  ppwwyyxx

  committed
  4f52bcf

  Copy full SHA for 4f52bcf

Commits on Feb 23, 2018

• update docs

  ppwwyyxx

  committed
  910cfae

  Copy full SHA for 910cfae

Commits on Feb 7, 2018

• Fallback to tf.contrib.layers.variance_scaling_initializer.

  ppwwyyxx

  committed
  6eb4b24

  Copy full SHA for 6eb4b24

Commits on Feb 6, 2018

• [FasterRCNN] add notes about using different dataset (#632)

  ppwwyyxx

  committed
  271ffad

  Copy full SHA for 271ffad

Commits on Dec 29, 2017

• Support internal_update in BN

  ppwwyyxx

  committed
  6b10019

  Copy full SHA for 6b10019

• Allow zero weight_decay in ImagenetModel

  ppwwyyxx

  committed
  3552703

  Copy full SHA for 3552703

Commits on Dec 28, 2017

• [FasterRCNN] use numpy iou implementation to avoid running TF in multiple processes (#546)

  ppwwyyxx

  committed
  8411d8c

  Copy full SHA for 8411d8c

Commits on Dec 12, 2017

• [FasterRCNN] Disable prefetch for now (fix #546)

  ppwwyyxx

  committed
  bf683cd

  Copy full SHA for bf683cd

Commits on Nov 26, 2017

• Add Mask-RCNN implementation

  ppwwyyxx

  committed
  1a12ccd

  Copy full SHA for 1a12ccd

Commits on Nov 19, 2017

• [FasterRCNN] remove unused arg; fix bug in decode bbox

  ppwwyyxx

  committed
  7556cc1

  Copy full SHA for 7556cc1

Commits on Nov 17, 2017

• [FasterRCNN] suppress tf logging; fix nan precision (#494)

  ppwwyyxx

  committed
  7f55d50

  Copy full SHA for 7f55d50

Commits on Nov 15, 2017

• [FasterRCNN] disable cropsize bug and disable fork

  ppwwyyxx

  committed
  15b35d9

  Copy full SHA for 15b35d9

Commits on Nov 14, 2017

• Use stricter linter in examples. Clean-up imports

  ppwwyyxx

  committed
  2fa4989

  Copy full SHA for 2fa4989

Commits on Nov 13, 2017

• [FasterRCNN] code clean-up and docs

  ppwwyyxx

  committed
  2766bb5

  Copy full SHA for 2766bb5

Commits on Nov 11, 2017

• [FasterRCNN] move GAP to fastrcnn_head

  ppwwyyxx

  committed
  b3f6e31

  Copy full SHA for b3f6e31

• optionally include mask in train_dataflow

  ppwwyyxx

  committed
  29c81dd

  Copy full SHA for 29c81dd

• [FasterRCNN] clean-up segmentation code

  ppwwyyxx

  committed
  42c9b8a

  Copy full SHA for 42c9b8a

• [FasterRCNN] put training preprocessing together

  ppwwyyxx

  committed
  671b64c

  Copy full SHA for 671b64c

Commits on Nov 9, 2017

• [FasterRCNN] rename a config; prefer arguments over constants

  ppwwyyxx

  committed
  cc0f07a

  Copy full SHA for cc0f07a

Commits on Oct 22, 2017

• [FasterRCNN] update LR schedule

  ppwwyyxx

  committed
  b0297e6

  Copy full SHA for b0297e6

Commits on Oct 11, 2017

• update readme

  ppwwyyxx

  committed
  6f3ed35

  Copy full SHA for 6f3ed35

• fix fasterrcnn linting

  ppwwyyxx

  committed
  0ef5061

  Copy full SHA for 0ef5061

• initial commit of FasterRCNN

  ppwwyyxx

  committed
  2344630

  Copy full SHA for 2344630

Loading