API Reference#

Ikomia API is divided into 4 main modules, each one may contain sub-modules. A large part of this API is based on C++ bindings, those classes are easily identifiable since we apply a naming convention where all C++ classes begin with the C capital letter.

Utils module

ikomia.utils.pyutils

Module offering various helper tools.

ikomia.utils.pyqtutils

Module providing helper functions to design your own task widget.

ikomia.utils.qtconversion

Module dedicated to wrap widget instance from Qt-based Python frameworks to C++ Qt.

ikomia.utils.displayIO

Module dedicated to the visualization of workflow components.

Core module

ikomia.core.pycore

Module offering core features to handle tasks, I/O, parameters and widgets.

ikomia.core.task

Module dedicated to high-level features around task management.

DataProcess module

ikomia.dataprocess.pydataprocess

Module offering implementation of specialization classes to handle inputs/outputs and tasks involved in Ikomia workflows for concrete use cases.

ikomia.dataprocess.registry

Module dedicated to algorithms management from the Ikomia platform.

ikomia.dataprocess.workflow

Module dedicated to workflow management.

ikomia.dataprocess.datadictIO

Module providing Ikomia workflow I/O implementation for data stored as Python dict.

DNN module

ikomia.dnn.dnntrain

Module dedicated to Deep Learning training.

ikomia.dnn.datasetio

Module dedicated to provide default implementation of Ikomia Deep Learning dataset structure.

ikomia.dnn.dataset

Module providing dataset loaders from various source formats.

ikomia.dnn.torch.models

Model providing helper funtions to create TorchVision Deep Learning models.

ikomia.dnn.torch.datasetmapper

Module providing implementation of a PyTorch dataset mapper for Ikomia dataset structure.