FullyConnectedLayer.md

CFullyConnectedLayer Class

• CFullyConnectedLayer Class
  • Settings
    • Output vector length
    • Using the free terms
  • Trainable parameters
    • Weight matrix
    • Free terms
  • Inputs
  • Outputs

This class implements a fully connected layer. The layer multiplies each of the input vectors by the weight matrix and adds the free term vector to the result.

Settings

Output vector length

void SetNumberOfElements(int newNumberOfElements);

Sets the length of each vector in the output.

Using the free terms

void SetZeroFreeTerm(bool isZeroFreeTerm);

Specifies if the free terms should be used. If you set this value to true, the free terms vector will be set to all zeros and won't be trained. By default, this value is set to false.

Trainable parameters

Weight matrix

CPtr<CDnnBlob> GetWeightsData() const;

The weight matrix is a blob of the dimensions:

• BatchLength * BatchWidth * ListSize is equal to GetNumberOfElements()
• Height, Width, Depth, and Channels are equal to the same dimensions of the first input

Free terms

CPtr<CDnnBlob> GetFreeTermData() const;

The free terms are represented by a blob of the total size equal to GetNumberOfElements().

Inputs

Each input accepts a blob with a set of vectors, of the dimensions:

• BatchLength * BatchWidth * ListSize is the number of vectors in the set.
• Height * Width * Depth * Channels is each vector length. It should be the same for all inputs.

Outputs

For each input, the corresponding output contains a blob with the result, of the dimensions:

• BatchLength is equal to the input BatchLength
• BatchWidth is equal to the input BatchWidth
• ListSize is equal to the input ListSize
• Height, Width, and Depth are equal to 1
• Channels is equal to GetNumberOfElements()