Google I/O returns May 18-20! Reserve space and build your schedule Register now

Array Ops

Summary

Classes

tensorflow:: ops:: BatchToSpace

BatchToSpace for 4-D tensors of type T.

tensorflow:: ops:: BatchToSpaceND

BatchToSpace for N-D tensors of type T.

tensorflow:: ops:: Bitcast

Bitcasts a tensor from one type to another without copying data.

tensorflow:: ops:: BroadcastDynamicShape

Return the shape of s0 op s1 with broadcast.

tensorflow:: ops:: BroadcastTo

Broadcast an array for a compatible shape.

tensorflow:: ops:: CheckNumerics

Checks a tensor for NaN and Inf values.

tensorflow:: ops:: Concat

Concatenates tensors along one dimension.

tensorflow:: ops:: ConjugateTranspose

Shuffle dimensions of x according to a permutation and conjugate the result.

tensorflow:: ops:: DebugGradientIdentity

Identity op for gradient debugging.

tensorflow:: ops:: DebugGradientRefIdentity

Identity op for gradient debugging.

tensorflow:: ops:: DeepCopy

Makes a copy of x .

tensorflow:: ops:: DepthToSpace

DepthToSpace for tensors of type T.

tensorflow:: ops:: Dequantize

Dequantize the 'input' tensor into a float or bfloat16 Tensor .

tensorflow:: ops:: Diag

Returns a diagonal tensor with a given diagonal values.

tensorflow:: ops:: DiagPart

Returns the diagonal part of the tensor.

tensorflow:: ops:: EditDistance

Computes the (possibly normalized) Levenshtein Edit Distance.

tensorflow:: ops:: Empty

Creates a tensor with the given shape.

tensorflow:: ops:: EnsureShape

Ensures that the tensor's shape matches the expected shape.

tensorflow:: ops:: ExpandDims

Inserts a dimension of 1 into a tensor's shape.

tensorflow:: ops:: ExtractImagePatches

Extract patches from images and put them in the "depth" output dimension.

tensorflow:: ops:: ExtractVolumePatches

Extract patches from input and put them in the "depth" output dimension.

tensorflow:: ops:: FakeQuantWithMinMaxArgs

Fake-quantize the 'inputs' tensor, type float to 'outputs' tensor of same type.

tensorflow:: ops:: FakeQuantWithMinMaxArgsGradient

Compute gradients for a FakeQuantWithMinMaxArgs operation.

tensorflow:: ops:: FakeQuantWithMinMaxVars

Fake-quantize the 'inputs' tensor of type float via global float scalars.

tensorflow:: ops:: FakeQuantWithMinMaxVarsGradient

Compute gradients for a FakeQuantWithMinMaxVars operation.

tensorflow:: ops:: FakeQuantWithMinMaxVarsPerChannel

Fake-quantize the 'inputs' tensor of type float via per-channel floats.

tensorflow:: ops:: FakeQuantWithMinMaxVarsPerChannelGradient

Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation.

tensorflow:: ops:: Fill

Creates a tensor filled with a scalar value.

tensorflow:: ops:: Fingerprint

Generates fingerprint values.

tensorflow:: ops:: Gather

Gather slices from params according to indices .

tensorflow:: ops:: GatherNd

Gather slices from params into a Tensor with shape specified by indices .

tensorflow:: ops:: GatherV2

Gather slices from params axis axis according to indices .

tensorflow:: ops:: GuaranteeConst

Gives a guarantee to the TF runtime that the input tensor is a constant.

tensorflow:: ops:: Identity

Return a tensor with the same shape and contents as the input tensor or value.

tensorflow:: ops:: IdentityN

Returns a list of tensors with the same shapes and contents as the input.

tensorflow:: ops:: ImmutableConst

Returns immutable tensor from memory region.

tensorflow:: ops:: InplaceAdd

Adds v into specified rows of x.

tensorflow:: ops:: InplaceSub

Subtracts v into specified rows of x .

tensorflow:: ops:: InplaceUpdate

Updates specified rows 'i' with values 'v'.

tensorflow:: ops:: InvertPermutation

Computes the inverse permutation of a tensor.

tensorflow:: ops:: MatrixBandPart

Copy a tensor setting everything outside a central band in each innermost matrix to zero.

tensorflow:: ops:: MatrixDiag

Returns a batched diagonal tensor with a given batched diagonal values.

tensorflow:: ops:: MatrixDiagPart

Returns the batched diagonal part of a batched tensor.

tensorflow:: ops:: MatrixDiagPartV2

Returns the batched diagonal part of a batched tensor.

tensorflow:: ops:: MatrixDiagPartV3

Returns the batched diagonal part of a batched tensor.

tensorflow:: ops:: MatrixDiagV2

Returns a batched diagonal tensor with given batched diagonal values.

tensorflow:: ops:: MatrixDiagV3

Returns a batched diagonal tensor with given batched diagonal values.

tensorflow:: ops:: MatrixSetDiag

Returns a batched matrix tensor with new batched diagonal values.

tensorflow:: ops:: MatrixSetDiagV2

Returns a batched matrix tensor with new batched diagonal values.

tensorflow:: ops:: MatrixSetDiagV3

Returns a batched matrix tensor with new batched diagonal values.

tensorflow:: ops:: MirrorPad

Pads a tensor with mirrored values.

tensorflow:: ops:: OneHot

Returns a one-hot tensor.

tensorflow:: ops:: OnesLike

Returns a tensor of ones with the same shape and type as x.

tensorflow:: ops:: Pad

Pads a tensor with zeros.

tensorflow:: ops:: PadV2

Pads a tensor.

tensorflow:: ops:: ParallelConcat

Concatenates a list of N tensors along the first dimension.

tensorflow:: ops:: Placeholder

A placeholder op for a value that will be fed into the computation.

tensorflow:: ops:: PlaceholderWithDefault

A placeholder op that passes through input when its output is not fed.

tensorflow:: ops:: PreventGradient

An identity op that triggers an error if a gradient is requested.

tensorflow:: ops:: QuantizeAndDequantizeV2

Quantizes then dequantizes a tensor.

tensorflow:: ops:: QuantizeAndDequantizeV3

Quantizes then dequantizes a tensor.

tensorflow:: ops:: QuantizeAndDequantizeV4

Returns the gradient of QuantizeAndDequantizeV4 .

tensorflow:: ops:: QuantizeAndDequantizeV4Grad

Returns the gradient of QuantizeAndDequantizeV4 .

tensorflow:: ops:: QuantizeV2

Quantize the 'input' tensor of type float to 'output' tensor of type 'T'.

tensorflow:: ops:: QuantizedConcat

Concatenates quantized tensors along one dimension.

tensorflow:: ops:: QuantizedInstanceNorm

Quantized Instance normalization.

tensorflow:: ops:: SetDiff1D

Computes the difference between two lists of numbers or strings.

tensorflow:: ops:: Stack

Packs a list of N rank- R tensors into one rank- (R+1) tensor.

tensorflow:: ops:: Where

Reshapes a quantized tensor as per the Reshape op.

tensorflow:: ops:: ZerosLike

Returns a tensor of zeros with the same shape and type as x.