InnerProduct
Input
- x:
- Required: yes
- Type: float16
- Description: input tensor
- Restrictions: none
- w:
- Required: yes
- Type: float16
- Description: weight tensor
- Restrictions: none
- b (optional):
- Required: no
- Type: float16, float32
- Description: bias
- Restrictions: none
Attribute
- num_output:
- Required: yes
- Type: int
- Description: number of neurons after InnerProduct
- Restriction: Must be within 32768.
- transpose:
- Required: no
- Type: bool
- Description: transposition switch (default = False)
- Restriction: either True or False
- bias_term:
- Required: no
- Type: bool
- Description: bias switch (default = True)
- Restriction: either True or False
- axis:
- Required: no
- Type: int
- Description: the axis along which to perform InnerProduct, either 1 or 2
- Restriction: Must be 1.
Output
y:
- Required: yes
- Type: float16, float32
- Description: output tensor
- Restrictions: none
Parent topic: Supported Caffe Operators