Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 27, 2018

• Have the SavedModel loader use Session's Make/Run/ReleaseCallable() API instead of Run(), to avoid leaving behind non-GC'ed state after model initialization.

  Show description for 19defc6
  

  Christopher Olston

  authored and

  tensorflower-gardener

  committed
  19defc6

  Copy full SHA for 19defc6

• SpaceToBatchND should pad with zero_point when inference_type is uint8

  Show description for 82293a9
  

  Suharsh Sivakumar

  authored and

  tensorflower-gardener

  committed
  82293a9

  Copy full SHA for 82293a9

• Layers/Networks now have a default implementation for static shape inference in eager execution (when possible).

  Show description for 9beaf70
  

  fchollet

  authored and

  tensorflower-gardener

  committed
  9beaf70

  Copy full SHA for 9beaf70

• Add build option to exclude TensorRT from cuda build.

  Show description for 65c2baa
  

  aaroey

  authored and

  tensorflower-gardener

  committed
  65c2baa

  Copy full SHA for 65c2baa

• Simulate eager variable resoration in tf.keras.Model.load_weights when graph building

  Show description for 21d94be
  

  allenlavoie

  authored and

  tensorflower-gardener

  committed
  21d94be

  Copy full SHA for 21d94be

• Add low and high as properties to quantized distribution.

  Show description for c7fc5b0
  

  Dustin Tran

  authored and

  tensorflower-gardener

  committed
  c7fc5b0

  Copy full SHA for c7fc5b0

• Update backward pass to save memory in eager mode.

  Show description for f83263d
  

  lxuechen

  authored and

  tensorflower-gardener

  committed
  f83263d

  Copy full SHA for f83263d

• Use `SaveableObject.restore` in init_from_checkpoint instead of `assign`.

  Show description for eea66ee
  

  guptapriya

  authored and

  tensorflower-gardener

  committed
  eea66ee

  Copy full SHA for eea66ee

Commits on Jul 26, 2018

• Implement constant buffer allocation for XLA:GPU

  Show description for 4009f82
  

  Sanjoy Das

  authored and

  tensorflower-gardener

  committed
  4009f82

  Copy full SHA for 4009f82

• Fix: When sample_weight_mode is a list/dict set default sample_weight values so that we do not require sample_weight to be set during training/eval

  Show description for e336ee6
  

  pavithrasv

  authored and

  tensorflower-gardener

  committed
  e336ee6

  Copy full SHA for e336ee6

• Remove tensorflow/opensource_only/Core as it is not used by tensorflow.

  Show description for 403845d
  tensorflower-gardener

  committed
  403845d

  Copy full SHA for 403845d

• [TF:XLA] Align the two implementations of CanShareOperandBufferWithUser.

  Show description for 111745b
  tensorflower-gardener

  committed
  111745b

  Copy full SHA for 111745b

• Avoid using cuda_fp16.h for fp16 definition.

  Show description for fad8d28
  

  gunan

  authored and

  tensorflower-gardener

  committed
  fad8d28

  Copy full SHA for fad8d28

• [TF:XLA] Remove unused parameter.

  Show description for 00dd73b
  tensorflower-gardener

  committed
  00dd73b

  Copy full SHA for 00dd73b

• [tf.data] Switch `batch_and_drop_remainder` to use the fused op.

  Show description for 0c55fdd
  

  mrry

  authored and

  tensorflower-gardener

  committed
  0c55fdd

  Copy full SHA for 0c55fdd

• Don't make remote copy call when both send/recv devices are the same.

  Show description for 4e0c568
  

  Akshay Modi

  authored and

  tensorflower-gardener

  committed
  4e0c568

  Copy full SHA for 4e0c568

• Increase the input dimension size from 4 to 6 to address the RetinaNet model

  Show description for 62df725
  tensorflower-gardener

  committed
  62df725

  Copy full SHA for 62df725

• Keras Tensorboard Callback - enable metrics logging in Eager mode

  Show description for d4cb01f
  tensorflower-gardener

  committed
  d4cb01f

  Copy full SHA for d4cb01f

• Test tf.constant directly (still doesn't work with scalars)

  Show description for 46332e1
  tensorflower-gardener

  committed
  46332e1

  Copy full SHA for 46332e1

• This change adds an attribute to the _HostCast op to have the same interface as the Cast op.

  Show description for 6aa9b86
  tensorflower-gardener

  committed
  6aa9b86

  Copy full SHA for 6aa9b86

• [XLA] Fix BF16 propagation type adjustment

  Show description for 1a306c9
  

  ukoxyz

  authored and

  tensorflower-gardener

  committed
  1a306c9

  Copy full SHA for 1a306c9

• [XLA:GPU] Don't fuse random things into fusions that become a cublas call

  Show description for 6f4e330
  

  d0k

  authored and

  tensorflower-gardener

  committed
  6f4e330

  Copy full SHA for 6f4e330

• Full set of unittests for the Eager delegate.

  Show description for 58841d7
  tensorflower-gardener

  committed
  58841d7

  Copy full SHA for 58841d7

• ResourceVariables shouldn't need twice the memory when initializing.

  Show description for ca69ddc
  

  alextp

  authored and

  tensorflower-gardener

  committed
  ca69ddc

  Copy full SHA for ca69ddc

• Delete unused code in CXX11/src/NeuralNetworks

  Show description for 6356357
  tensorflower-gardener

  committed
  6356357

  Copy full SHA for 6356357

• Restore tower local variables correctly in init_from_checkpoint.

  Show description for 53449eb
  

  guptapriya

  authored and

  tensorflower-gardener

  committed
  53449eb

  Copy full SHA for 53449eb

• PUBLIC: Allow user passing training/evaluation/prediction_hooks from tf.estimator.EstimatorSpec.

  Show description for b5ee7b1
  

  Youlong Cheng

  authored and

  tensorflower-gardener

  committed
  b5ee7b1

  Copy full SHA for b5ee7b1

• [TF:XLA] Bump open source llvm revision to r338012

  Show description for 4fd159c
  

  Sanjoy Das

  authored and

  tensorflower-gardener

  committed
  4fd159c

  Copy full SHA for 4fd159c

• More changes for TFLite portability

  Show description for 20154fe
  

  jdduke

  authored and

  tensorflower-gardener

  committed
  20154fe

  Copy full SHA for 20154fe

• Add example Unity plugin for the experimental TFLite C API.

  Show description for 99deb5e
  

  jdduke

  authored and

  tensorflower-gardener

  committed
  99deb5e

  Copy full SHA for 99deb5e

• Remove the gen_locally tag which is no more needed.

  Show description for b4cc9c3
  

  aaroey

  authored and

  tensorflower-gardener

  committed
  b4cc9c3

  Copy full SHA for b4cc9c3

• shard some estimator tests

  Show description for 7e4d0d1
  

  tanzhenyu

  authored and

  tensorflower-gardener

  committed
  7e4d0d1

  Copy full SHA for 7e4d0d1

• Boosted trees: Revealing pruning mode as one of the parameters for a gbdt estimator

  Show description for 91ff408
  tensorflower-gardener

  committed
  91ff408

  Copy full SHA for 91ff408

• Ergonomic fixes to tensorflow::dump_graph::*

  Show description for c83525a
  

  Sanjoy Das

  authored and

  tensorflower-gardener

  committed
  c83525a

  Copy full SHA for c83525a

• Make model totally defunable.

  Show description for e91c597
  

  lxuechen

  authored and

  tensorflower-gardener

  committed
  e91c597

  Copy full SHA for e91c597

Pagination

Previous

Next