Automl

API reference for the AUCMEDI AutoML pipeline.

The mentality behind AutoML is to ensure easy application, integration and maintenance of complex medical image classification pipelines. AUCMEDI provides a fast and intuitive interface through AUCMEDI AutoML for building, application and sharing of state-of-the-art medical image classification models.

The AutoML pipelines are categorized into the following modes: training, prediction and evaluation.

• The console entry aucmedi refers to aucmedi.automl.main:main.
• The Argparse interface for CLI is defined in aucmedi.automl.cli
• Each AutoML mode is implemented as a code block defining the AUCMEDI pipeline.

Info

AutoML Mode	Argparse	Code Block (Pipeline)
training	CLI - Training	Block - Train
prediction	CLI - Prediction	Block - Predict
evaluation	CLI - Evaluation	Block - Evaluate

More information can be found in the docs: Documentation - AutoML