Neurons.DenseColMatrix, Neurons.DenseRowMatrix, Neurons.DenseVector, Neurons.Input, Neurons.Linear, Neurons.Matrix, Neurons.Maxout, Neurons.MaxoutDropout, Neurons.Output, Neurons.Rectifier, Neurons.RectifierDropout, Neurons.Softmax, Neurons.SparseRowMatrix, Neurons.SparseVector, Neurons.Tanh, Neurons.TanhDropout, Neurons.Vector
Modifier and Type | Method and Description |
---|---|
protected void |
bprop()
Back propagation
|
protected void |
fprop(long seed,
boolean training)
Forward propagation
|
void |
setInput(long seed,
double[] data)
One of two methods to set layer input values.
|
void |
setInput(long seed,
double[] nums,
int numcat,
int[] cats)
The second method used to set input layer values.
|
protected void bprop()
Neurons
protected void fprop(long seed, boolean training)
Neurons
public void setInput(long seed, double[] data)
seed
- For seeding the RNG inside (for input dropout)data
- Data (training columns and responses) to extract the training columns
from to be mapped into the input neuron layerpublic void setInput(long seed, double[] nums, int numcat, int[] cats)
seed
- For seeding the RNG inside (for input dropout)nums
- Array containing numerical values, can be NaNnumcat
- Number of horizontalized categorical non-zero values (i.e., those not being the first factor of a class)cats
- Array of indices, the first numcat values are the input layer unit (==column) indices for the non-zero categorical values
(This allows this array to be re-usable by the caller, without re-allocating each time)