Commits

Branch selector

 master

User selector

pkulzc

Datepicker

All time

Commit History

Commits on May 15, 2018

• Merge pull request tensorflow#4205 from vladpaunescu/master

  Show description for ea6d6aa
  pkulzc

  authored
  ea6d6aa

  Copy full SHA for ea6d6aa

Commits on May 11, 2018

• Merge pull request tensorflow#4147 - update exporting_models doc

  Show description for 28b9142
  pkulzc

  authored
  28b9142

  Copy full SHA for 28b9142

• Merged commit includes the following changes:

  Show description for 324d6dc
  pkulzc

  committed
  324d6dc

  Copy full SHA for 324d6dc

Commits on May 3, 2018

• Merged commit includes the following changes:

  Show description for 6305421
  pkulzc

  committed
  6305421

  Copy full SHA for 6305421

Commits on May 1, 2018

• Internal changes to slim and object detection (tensorflow#4100)

  Show description for 505f554
  

  pkulzc

  authored and

  sguada

  committed
  505f554

  Copy full SHA for 505f554

Commits on Apr 20, 2018

• Merge pull request tensorflow#2146 to fix data augmentation functions.

  Show description for c86f791
  pkulzc

  authored
  c86f791

  Copy full SHA for c86f791

Commits on Apr 17, 2018

• Merge pull request tensorflow#3973 from pkulzc/master

  Show description for ed4e22b
  pkulzc

  authored
  ed4e22b

  Copy full SHA for ed4e22b

• Sync to master.

  pkulzc

  committed
  13b89b9

  Copy full SHA for 13b89b9

Commits on Apr 16, 2018

• Fix copyright year.

  pkulzc

  committed
  a4b6765

  Copy full SHA for a4b6765

Commits on Apr 13, 2018

• Update to trainer to allow for reading multiclass scores

  Show description for cc08dc8
  pkulzc

  committed
  cc08dc8

  Copy full SHA for cc08dc8

• Contains implementation of Visual Relations Detection evaluation metric (per

  Show description for 1bddd18
  pkulzc

  committed
  1bddd18

  Copy full SHA for 1bddd18

• Modify the ssd meta arch to allow the option of not adding an implicit background class.

  Show description for eccae44
  pkulzc

  committed
  eccae44

  Copy full SHA for eccae44

• Refactor model_tpu_main.py files and move continuous eval loop into model_lib.py

  Show description for a60dd98
  pkulzc

  committed
  a60dd98

  Copy full SHA for a60dd98

• Update create_pascal_tf_record.py and create_pet_tf_record.py

  Show description for f98f000
  pkulzc

  committed
  f98f000

  Copy full SHA for f98f000

• Enforcing that all eval metric ops have valid python strings.

  Show description for 2151c44
  pkulzc

  committed
  2151c44

  Copy full SHA for 2151c44

• Set regularize_depthwise to true in mobilenet_v1_argscope.

  Show description for 0940134
  pkulzc

  committed
  0940134

  Copy full SHA for 0940134

• Refactoring of Mask-RCNN to put all mask prediction code in third stage.

  Show description for b47ca97
  pkulzc

  committed
  b47ca97

  Copy full SHA for b47ca97

• Returning eval_on_train_input_fn from create_estimator_and_inputs(), rather than using train_input_fn in EVAL mode (which will still have data augmentation).

  Show description for 227f41e
  pkulzc

  committed
  227f41e

  Copy full SHA for 227f41e

• Access TPUEstimator and CrossShardOptimizer from tf namesspace.

  Show description for 7e81000
  pkulzc

  committed
  7e81000

  Copy full SHA for 7e81000

• Fix test that was flaky due to randomness

  Show description for b0c5c3b
  pkulzc

  committed
  b0c5c3b

  Copy full SHA for b0c5c3b

• Minor fixes to match git repo.

  Show description for 2624a23
  pkulzc

  committed
  2624a23

  Copy full SHA for 2624a23

• use shape utils for assertion in feature extractor.

  Show description for e25c014
  pkulzc

  committed
  e25c014

  Copy full SHA for e25c014

• Class agnostic masks for mask_rcnn

  Show description for a4d9c3a
  pkulzc

  committed
  a4d9c3a

  Copy full SHA for a4d9c3a

• Add learning rate summary in EVAL mode in model.py

  Show description for bfd15ec
  pkulzc

  committed
  bfd15ec

  Copy full SHA for bfd15ec

• Migrating away from Experiment class, as it is now deprecated. Also, refactoring into a separate model library and binaries.

  Show description for 90cc9ba
  pkulzc

  committed
  90cc9ba

  Copy full SHA for 90cc9ba

• Add classification_loss and localiztion_loss metrics for TPU jobs.

  Show description for 8deba73
  pkulzc

  committed
  8deba73

  Copy full SHA for 8deba73

• Add an option to skip the last striding in mobilenet. The modified network has nominal output stride 16 instead of 32.

  Show description for aedfa2e
  pkulzc

  committed
  aedfa2e

  Copy full SHA for aedfa2e

• Add option to override base feature extractor hyperparams in SSD models. This would allow us to use the same set of hyperparams for the complete feature extractor (base + new layers) if desired.

  Show description for decbad8
  pkulzc

  committed
  decbad8

  Copy full SHA for decbad8

• Adding an attribute to SSD model to indicate which fields in prediction dictionary have a batch dimension. This will be useful for future video models.

  Show description for 45ecc0f
  pkulzc

  committed
  45ecc0f

  Copy full SHA for 45ecc0f

• Internal change.

  Show description for a0c3c44
  pkulzc

  committed
  a0c3c44

  Copy full SHA for a0c3c44

• Sync to latest master.

  pkulzc

  committed
  b668f59

  Copy full SHA for b668f59

Commits on Apr 9, 2018

• Merge pull request tensorflow#3744 from liangxiao05/master

  Show description for e3bfb9b
  pkulzc

  authored
  e3bfb9b

  Copy full SHA for e3bfb9b

• Merge pull request tensorflow#3914 from matiji66/master

  Show description for 2506c9a
  pkulzc

  authored
  2506c9a

  Copy full SHA for 2506c9a

Commits on Apr 4, 2018

• Merge remote-tracking branch 'upstream/master'

  pkulzc

  committed
  d5fc3ef

  Copy full SHA for d5fc3ef

• Merged commit includes the following changes:

  Show description for 6b72b5c
  pkulzc

  committed
  6b72b5c

  Copy full SHA for 6b72b5c

Pagination

Previous

Next