new Regression()
Creates a set of regression objects using the constructor from emscripten
Properties:
| Name | Type | Description |
|---|---|---|
Module.RegressionCpp |
function | constructor from emscripten |
- Source:
Methods
getNumHiddenLayers() → {Number}
Returns the number of hidden layers in a MLP.
- Source:
Returns:
k values
- Type
- Number
process(input) → {Array}
Deprecated! Use run() instead
Parameters:
| Name | Type | Description |
|---|---|---|
input |
- Source:
Returns:
- Type
- Array
reset() → {Boolean}
Returns the model set to its initial configuration.
- Source:
Returns:
true indicates successful initialization
- Type
- Boolean
run(input) → {Array}
Runs feed-forward regression on input
Parameters:
| Name | Type | Description |
|---|---|---|
input |
Array | An array of features to be processed. Non-arrays are converted. |
- Source:
Returns:
output - One number for each model in the set
- Type
- Array
setNumEpochs(numEpochs)
Sets the number of epochs for MLP training.
Parameters:
| Name | Type | Description |
|---|---|---|
numEpochs |
Number |
- Source:
setNumHiddenLayers(numHiddenLayers)
Sets the number of hidden layers for an MLP.
Parameters:
| Name | Type | Description |
|---|---|---|
numHiddenLayers |
Number |
- Source:
train(trainingSet) → {Boolean}
Trains the models using the input. Starts training from a randomized state.
Parameters:
| Name | Type | Description |
|---|---|---|
trainingSet |
Object | An array of training examples |
- Source:
Returns:
true indicates successful training
- Type
- Boolean