Skip to content

Conversation

ioeric
Copy link
Contributor

@ioeric ioeric commented Sep 4, 2017

o Removed some internal/outdated ops.
['ApplyGradientDescent', 'ApplyAdagrad', '_ArrayToList', 'ArgMax', 'VariableV2', 'StridedSliceAssign', 'DummyReadResource', '_TPUSend', 'ApplyRMSProp', 'ApplyAdam', '_UnsafeReadVariable', 'Variable', '_TPURecv', '_ListToArray', 'Placeholder', 'Assign', 'ApplyMomentum']
o Added newly supported ops.
['ResourceApplyRMSProp', 'ResourceApplyMomentum', 'InplaceAdd', 'ResourceApplyAdam', 'TensorArrayCloseV3', 'Cross', 'PadV2', 'StackPopV2', 'DepthwiseConv2dNativeBackpropFilter', 'SeluGrad', 'Selu', 'Sin', 'Empty', 'InplaceUpdate', 'ResourceApplyAdagrad', 'TensorArrayGatherV3', 'RecvBarnaCoreActivations', 'ControlTrigger', 'TensorArraySizeV3', 'ResourceStridedSliceAssign', 'ResourceGather', 'SendBarnaCoreGradients', 'TensorArrayWriteV3', 'Cos', 'ResourceApplyGradientDescent', 'UnsortedSegmentSum', 'FusedBatchNormGrad', 'StackV2', 'While', 'MirrorPad', 'DepthwiseConv2dNativeBackpropInput', 'ResourceApplyFtrl', 'XlaWhile', 'TensorArrayV3', 'TensorArrayReadV3', 'TensorArrayScatterV3', 'FusedBatchNorm', 'TensorArraySplitV3', 'TensorArrayGradV3', 'StackCloseV2', 'StackPushV2', 'TensorArrayConcatV3', 'IsFinite', 'ResourceApplyFtrlV2']
o One op a line to make it easier to eyeball.

Copy link
Member

@chihuahua chihuahua left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A bit tedious, but in the PR description, could you enumerate the ops added + removed?

Approving.

@ioeric
Copy link
Contributor Author

ioeric commented Sep 5, 2017

Done.

@yunxing
Copy link

yunxing commented Sep 5, 2017

LGTM

@chihuahua chihuahua merged commit 41fa861 into tensorflow:master Sep 6, 2017
jart pushed a commit to jart/tensorboard that referenced this pull request Sep 6, 2017
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants