Companion

object Companion

Functions

fromOutput
Link copied to clipboard
fun fromOutput(finalLayer: Layer): Functional

Creates the Functional model.

loadDefaultModelConfiguration
Link copied to clipboard
fun loadDefaultModelConfiguration(modelDirectory: File, inputShape: IntArray? = null): Functional

Loads a Functional model from json file with name 'modelConfig.json' with model configuration located in modelDirectory.

loadModelConfiguration
Link copied to clipboard
fun loadModelConfiguration(configuration: File, inputShape: IntArray? = null): Functional

Loads a Functional model from json file with model configuration.

loadModelLayersFromConfiguration
Link copied to clipboard
fun loadModelLayersFromConfiguration(configuration: File, inputShape: IntArray? = null): List<Layer>

Loads a Functional model layers from json file with model configuration.

loadModelLayersFromDefaultConfiguration
Link copied to clipboard
fun loadModelLayersFromDefaultConfiguration(modelDirectory: File, inputShape: IntArray? = null): List<Layer>

Loads a Functional model layers from json file with name 'modelConfig.json' with model configuration located in modelDirectory.

of
Link copied to clipboard
fun of(vararg layers: Layer, noInput: Boolean = false): Functional
fun of(layers: List<Layer>, noInput: Boolean = false): Functional

Creates the Functional model.

fun of(pretrainedModel: GraphTrainableModel, topModel: GraphTrainableModel): Functional

Creates the Functional model from two models: pretrainedModel and topModel. All layers of pretrainedModel will be frozen automatically. The input of the topModel will be connected to the output of the pretrainedModel.