GatherND

Description

Slices and outputs the input data.

Input

Two inputs:

data: tensor of type float16, float, double, int32, or int64.

indices: int64 tensor, rank q>=1.

Output

One output:

output: tensor with the rank of q + r - indices_shape[-1] - 1.

Property

batch_dims: int, number of batch processing axes. Currently, the value can only be 0.

Constraints

The double input is not supported when the atc tool parameter --precision_mode=must_keep_origin_dtype is used.

ONNX Opset Support

Opset v11/v12/v13/v14/v15/v16/v17/v18