NetParam
The NetParam class bundles parameter weights and biases together.
Value parameters
- b
-
the bias/intercept vector
- w
-
the weight matrix
Attributes
- Graph
-
- Supertypes
-
trait Serializabletrait Producttrait Equalsclass Objecttrait Matchableclass AnyShow all
Members list
Value members
Concrete methods
Multiply matrix x by the weight matrix w and add the bias vector b, unless the bias is null.
Multiply matrix x by the weight matrix w and add the bias vector b, unless the bias is null.
Value parameters
- x
-
the matrix to multiply by
Attributes
Multiply matrix x by the weight matrix w and add the bias vector b. This is the right associative version (NetParam on the left).
Multiply matrix x by the weight matrix w and add the bias vector b. This is the right associative version (NetParam on the left).
Value parameters
- x
-
the matrix to multiply by
Attributes
Add and assign the changes to the parameters (weights and biases).
Add and assign the changes to the parameters (weights and biases).
Value parameters
- c
-
the change to the parameters
Attributes
Add and assign the changes to the parameters (weights and biases).
Add and assign the changes to the parameters (weights and biases).
Value parameters
- cb
-
the change to the baises
- cw
-
the change to the weights
Attributes
Subtract and assign the changes to the parameters (weights and biases).
Subtract and assign the changes to the parameters (weights and biases).
Value parameters
- c
-
the change to the parameters
Attributes
Subtract and assign the changes to the parameters (weights and biases).
Subtract and assign the changes to the parameters (weights and biases).
Value parameters
- cb
-
the change to the baises
- cw
-
the change to the weights
Attributes
Return whether this and NetParam c are approximately equal.
Return whether this and NetParam c are approximately equal.
Value parameters
- c
-
the other network parameters
Attributes
Make a copy of the network parameters.
Make a copy of the network parameters.
Attributes
Multiply vector x by the weight matrix w and add the bias vector b.
Multiply vector x by the weight matrix w and add the bias vector b.
Value parameters
- x
-
the vector to multiply by
Attributes
Set/assign NetParam c to this (needed for val cases).
Set/assign NetParam c to this (needed for val cases).
Attributes
Set/assign (cw, cb) to this (needed for val cases).
Set/assign (cw, cb) to this (needed for val cases).
Attributes
Convert this NetParam object to a matrix where the first row is the bias vector b and the rest of the rows contain the weight matrix w.
Convert this NetParam object to a matrix where the first row is the bias vector b and the rest of the rows contain the weight matrix w.
Attributes
Convert this NetParam object to a string showing this (b) parameter's weight matrix b.w and bias vector b.b.
Convert this NetParam object to a string showing this (b) parameter's weight matrix b.w and bias vector b.b.
Attributes
- Definition Classes
-
Any
Make a trimmed copy of the network parameters.
Make a trimmed copy of the network parameters.
Value parameters
- dim
-
the first dimension of weight matrix
- dim2
-
the second dimension of weight matrix and dimension of bias vector
Attributes
Update/assign NetParam c to this.
Update/assign NetParam c to this.
Attributes
Inherited methods
Attributes
- Inherited from:
- Product
Attributes
- Inherited from:
- Product