[object Object][object Object][object Object]undefined[object Object]

• Description: For an input tensor [object Object] with dimension r≥1 and an input tensor [object Object] with dimension q≥1, collects data slices to the output tensor out with dimension (q–1) + (r – indices_shape[-1]). indices is a q-dimensional integer tensor, which can be regarded as a q–1 dimensional special tensor consisting of index pairs (each index pair is a one-dimensional tensor with a length of indices_shape[-1], and each index pair points to a slice in self).

• Computing logic:

  • If indices_shape[-1] > r, the scenario is invalid.
  • If indices_shape[-1] = r, the dimension of the output tensor out is q–1, that is, the shape of out is [indices_shape[0:q–1]], and the elements in out are the elements at the positions of the index pairs of self. (See example 1).
  • If indices_shape[-1] < r, the dimension of the output tensor out is (q–1) + (r – indices_shape[-1]). Assume that c = indices_shape[-1], that is, the shape of out is [indices_shape[0:q–1],self_shape[c:r]]. [object Object] consists of the slices at the positions of the index pairs of [object Object]. (See examples 2, 3, and 4.)

  The constraints of r, q, and indices_shape[-1] are as follows:

  • r ≥ 1 and q ≥ 1 must be met.
  • The value of indices_shape[-1] must be within [1, r].
  • Each element of [object Object] must be within the range of [–s, s – 1] (s is the value of each axis of self_shape), that is, –self_shape[i] ≤ indices[...,i] ≤ self_shape[i]–1.

• Example:

  [object Object]

[object Object]

Each operator has calls. First, aclnnGatherNdGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnGatherNd is called to perform computation.

[object Object]

• Parameters

  [object Object]
  • [object Object]Atlas A2 training products/Atlas A2 inference products[object Object] and [object Object]Atlas A3 training products/Atlas A3 inference products[object Object]: The data type cannot be DOUBLE, INT16, UINT16, UINT32, or UINT64.

• Returns

  [object Object]: status code. For details, see .

  The first-phase API implements input parameter verification. The following errors may be thrown.

  [object Object]

[object Object]

• Parameters

  [object Object]

• Returns:

  [object Object]: status code. For details, see .

[object Object]

• Deterministic computation:
  • aclnnGatherNd defaults to a deterministic implementation.

[object Object]

The following example is for reference only. For details, see .