MatrixBandPart

tensorflow C++ API

tensorflow::ops::MatrixBandPart

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

Summary

to zero.

Thebandpart is computed as follows: Assumeinputhaskdimensions[I, J, K, ..., M, N], then the output is a tensor with the same shape where

band[i, j, k, ..., m, n] = in_band(m, n) * input[i, j, k, ..., m, n].

The indicator function

in_band(m, n) = (num_lower < 0 || (m-n) <= num_lower)) && (num_upper < 0 || (n-m) <= num_upper).

For example:

``` if 'input' is [[ 0, 1, 2, 3] [-1, 0, 1, 2] [-2, -1, 0, 1] [-3, -2, -1, 0]],

tf.matrix_band_part(input, 1, -1) ==> [[ 0, 1, 2, 3] [-1, 0, 1, 2] [ 0, -1, 0, 1] [ 0, 0, -1, 0]],

tf.matrix_band_part(input, 2, 1) ==> [[ 0, 1, 0, 0] [-1, 0, 1, 0] [-2, -1, 0, 1] [ 0, -2, -1, 0]] ```

Useful special cases:

``` tf.matrix_band_part(input, 0, -1) ==> Upper triangular part. tf.matrix_band_part(input, -1, 0) ==> Lower triangular part. tf.matrix_band_part(input, 0, 0) ==> Diagonal. ```

Arguments:

• scope: A Scope object
• input: Rank k tensor.
• num_lower: 0-D tensor. Number of subdiagonals to keep. If negative, keep entire lower triangle.
• num_upper: 0-D tensor. Number of superdiagonals to keep. If negative, keep entire upper triangle.

Returns:

• Output: Rank k tensor of the same shape as input. The extracted banded tensor.

MatrixBandPart block

Source link :https://github.com/EXPNUNI/enuSpaceTensorflow/blob/master/enuSpaceTensorflow/tf_array_ops.cpp

Argument:

• Scope scope : A Scope object (A scope is generated automatically each page. A scope is not connected.)
• Input x : A 1-D Tensor type of int32,int64(Must be one integer type).

Output:

• y : Output object of MatrixBandPart class object.

Result:

• std::vector(Tensor) result_y: A Tensor.

Using Method

※ 위의 그림과 같이 output에 그어놓은 빨간색 선을 기준으로 위쪽 줄(num_upper)을 또는 아래쪽 줄(num_lower)을 몇번째 까지 표시 할지 정하는 기능을 한다.

※ 위 그림을 표현하면 8 1 0 0 이 된다. 즉 아래로 2줄 위로 1줄을 남기고 나머지는 0으로 만든다는 뜻
-1 9 1 0
-2 -1 10 0
0 -2 -1 11