Package org.jetbrains.kotlinx.dl.api.inference.keras.loaders

Different Neural Networks were trained on ImageNet dataset with different image preprocessing. The main types of preprocessing widely used in keras.applications are presented in this enumeration.

Loading strategy for remote loading.

This is an abstract class which provides methods for loading models, its weights and labels (for prediction purposes) to the local cacheDirectory.

Basic interface for models loaded from S3.

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

Supported models for inference and transfer learning, trained on ImageNet dataset.

Caffe-style preprocessing.

Returns top-5 labels for the given data encoded with mapping imageNetClassLabels.

Returns topK labels for the given data encoded with mapping imageNetClassLabels.

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

Common preprocessing functions for the Neural Networks trained on ImageNet and whose weights are available with the keras.application.

Reshapes inputData according imageShape.

Torch-style preprocessing.