TFModelHub

class TFModelHub(cacheDirectory: File) : ModelHub

This class provides methods for loading Keras models, its weights and ImageNet labels (for prediction purposes) to the local cacheDirectory.

Since

0.2

Parameters

cacheDirectory

The directory for all loaded models. It should be created before model loading and should have all required permissions for file writing/reading on your OS.

Constructors

TFModelHub

fun TFModelHub(cacheDirectory: File)

Functions

get

operator fun <T : InferenceModel, U : InferenceModel> get(modelType: ModelType<T, U>): U

loadClassLabels

fun loadClassLabels(): MutableMap<Int, String>

Forms mapping of class label to class name for the ImageNet dataset.

loadModel

open override fun <T : InferenceModel, U : InferenceModel> loadModel(modelType: ModelType<T, U>, loadingMode: LoadingMode): T

Loads model configuration without weights.

loadPretrainedModel

fun <T : InferenceModel, U : InferenceModel> loadPretrainedModel(modelType: ModelType<T, U>, loadingMode: LoadingMode = LoadingMode.SKIP_LOADING_IF_EXISTS): U

Loads pretrained model of modelType from the ModelHub in loadingMode.

loadWeights

fun loadWeights(modelType: ModelType<*, *>, loadingMode: LoadingMode = LoadingMode.SKIP_LOADING_IF_EXISTS): HdfFile

Loads model weights.

Properties

cacheDirectory

val cacheDirectory: File

The directory for all loaded models and datasets. It should be created before model loading and should have all required permissions for file writing/reading on your OS.