Fourier Transform Functions

TensorFlow provides several operations that you can use to add discrete Fourier transform functions to your graph.

tf.fft(input, name=None)

Compute the 1-dimensional discrete Fourier Transform over the inner-most

dimension of input.

Args:
• input: A Tensor of type complex64. A complex64 tensor.
• name: A name for the operation (optional).
Returns:

A Tensor of type complex64. A complex64 tensor of the same shape as input. The inner-most dimension of input is replaced with its 1D Fourier Transform.

tf.ifft(input, name=None)

Compute the inverse 1-dimensional discrete Fourier Transform over the inner-most

dimension of input.

Args:
• input: A Tensor of type complex64. A complex64 tensor.
• name: A name for the operation (optional).
Returns:

A Tensor of type complex64. A complex64 tensor of the same shape as input. The inner-most dimension of input is replaced with its inverse 1D Fourier Transform.

tf.fft2d(input, name=None)

Compute the 2-dimensional discrete Fourier Transform over the inner-most

2 dimensions of input.

Args:
• input: A Tensor of type complex64. A complex64 tensor.
• name: A name for the operation (optional).
Returns:

A Tensor of type complex64. A complex64 tensor of the same shape as input. The inner-most 2 dimensions of input are replaced with their 2D Fourier Transform.

@compatibility(numpy) Equivalent to np.fft2 @end_compatibility

tf.ifft2d(input, name=None)

Compute the inverse 2-dimensional discrete Fourier Transform over the inner-most

2 dimensions of input.

Args:
• input: A Tensor of type complex64. A complex64 tensor.
• name: A name for the operation (optional).
Returns:

A Tensor of type complex64. A complex64 tensor of the same shape as input. The inner-most 2 dimensions of input are replaced with their inverse 2D Fourier Transform.

@compatibility(numpy) Equivalent to np.ifft2 @end_compatibility

tf.fft3d(input, name=None)

Compute the 3-dimensional discrete Fourier Transform over the inner-most 3

dimensions of input.

Args:
• input: A Tensor of type complex64. A complex64 tensor.
• name: A name for the operation (optional).
Returns:

A Tensor of type complex64. A complex64 tensor of the same shape as input. The inner-most 3 dimensions of input are replaced with their 3D Fourier Transform.

@compatibility(numpy) Equivalent to np.fft3 @end_compatibility

tf.ifft3d(input, name=None)

Compute the inverse 3-dimensional discrete Fourier Transform over the inner-most

3 dimensions of input.

Args:
• input: A Tensor of type complex64. A complex64 tensor.
• name: A name for the operation (optional).
Returns:

A Tensor of type complex64. A complex64 tensor of the same shape as input. The inner-most 3 dimensions of input are replaced with their inverse 3D Fourier Transform.

@compatibility(numpy) Equivalent to np.fft3 @end_compatibility