Documentation

Concat

Input

x:
  • Required: yes
  • Type: float16, fp32, int8, int16, int32, int64, uint8, uint16, uint32, uint64
  • Description: concatenates multiple input blobs to one single output blob. x is a variable-length list of tensors.
  • Restriction: Keep the number of bottoms within 16.

Attribute

  • axis:
    • Required: no
    • Type: int
    • Description: the axis along which to concatenate (default = 1). May be negative to index from the end.
    • Restriction: Must be within the rank range.
  • concat_dim:
    • Required: no
    • Type: int
    • Description: alias for axis (default = 1)
    • Restriction: Does not support negative axis indexing.

Output

y:
  • Required: yes
  • Type: float16, fp32, int8, int16, int32, int64, uint8, uint16, uint32, uint64
  • Description: concatenates multiple input blobs to one single output blob. x is a variable-length list of tensors.
  • Restrictions: none