Array Ops

BatchToSpace for 4-D tensors of type T.

BatchToSpace for N-D tensors of type T.

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

Return the shape of s0 op s1 with broadcast.

Broadcast an array for a compatible shape.

Checks a tensor for NaN and Inf values.

Concatenates tensors along one dimension.

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

Identity op for gradient debugging.

Identity op for gradient debugging.

Makes a copy of x.

DepthToSpace for tensors of type T.

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

Returns a diagonal tensor with a given diagonal values.

Returns the diagonal part of the tensor.

Computes the (possibly normalized) Levenshtein Edit Distance.

Creates a tensor with the given shape.

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

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

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

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

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

Compute gradients for a FakeQuantWithMinMaxArgs operation.

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

Compute gradients for a FakeQuantWithMinMaxVars operation.

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

Compute gradients for a FakeQuantWithMinMaxVarsPerChannel operation.

Creates a tensor filled with a scalar value.

Generates fingerprint values.

Gather slices from params according to indices.

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

Gather slices from params axis axis according to indices.

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

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

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

Returns immutable tensor from memory region.

Adds v into specified rows of x.

Subtracts v into specified rows of x.

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

Computes the inverse permutation of a tensor.

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

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

Returns the batched diagonal part of a batched tensor.

Returns the batched diagonal part of a batched tensor.

Returns the batched diagonal part of a batched tensor.

Returns a batched diagonal tensor with given batched diagonal values.

Returns a batched diagonal tensor with given batched diagonal values.

Returns a batched matrix tensor with new batched diagonal values.

Returns a batched matrix tensor with new batched diagonal values.

Returns a batched matrix tensor with new batched diagonal values.

Pads a tensor with mirrored values.

Returns a one-hot tensor.

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

Pads a tensor with zeros.

Pads a tensor.

Concatenates a list of N tensors along the first dimension.

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

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

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

Quantizes then dequantizes a tensor.

Quantizes then dequantizes a tensor.

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

Concatenates quantized tensors along one dimension.

Quantized Instance normalization.

Computes the difference between two lists of numbers or strings.

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

Reshapes a quantized tensor as per the Reshape op.

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