Squeeze

Description

Removes dimensions of size 1 from the input.

Input

One input:

x: tensor of type float16, float, double, uint8, uint16, uint32, uint64, int8, int16, int32, int64 or bool.

Output

One output:

y: tensor. Has an identical data type to that of the input.

Attribute

axes: int32 or int64 list. Must be one-dimensional. Must be in the range [-r, r - 1]. r indicates the rank of the input tensor, and a negative number indicates that the dimension is computed from the end.

ONNX Opset Support

Opset v8/v9/v10/v11/v12/v13/v14/v15/v16/v17/v18