Logo
master

Getting Started

  • Getting started
    • Backends
    • Installation

Tutorials:

  • Delira Introduction
    • Loading Data
      • The Dataset
      • The Dataloader
      • The Datamanager
      • Sampler
    • Models
      • __init__
      • closure
      • prepare_batch
    • Abstract Networks for specific Backends
      • PyTorch
        • forward
        • prepare_batch
        • closure example
      • Other examples
    • Training
      • Parameters
      • Trainer
      • Experiment
    • Logging
      • MultiStreamHandler
      • Logging with Visdom - The trixi Loggers
    • More Examples
  • Classification with Delira - A very short introduction
    • Logging and Visualization
    • Data Preparation
      • Loading
      • Augmentation
    • Training
    • See Also
  • Generative Adversarial Nets with Delira - A very short introduction
    • HyperParameters
    • Logging and Visualization
    • Data Preparation
      • Loading
      • Augmentation
    • Training
    • See Also
  • Segmentation in 2D using U-Nets with Delira - A very short introduction
    • Logging and Visualization
    • Data Praparation
      • Loading
      • Augmentation
    • Training
    • See Also
  • Segmentation in 3D using U-Nets with Delira - A very short introduction
    • Logging and Visualization
    • Data Praparation
      • Loading
      • Augmentation
    • Training
    • See Also
  • How To: Integrate your own Computation Backend
    • Model Definitions
      • TorchScript Limitations
      • Merging TorchScript into our Abstract Class
      • Actual Implementation
        • Class Signature and __init__-Method
        • __call__-Method
        • closure-Method
        • prepare_batch-Method
        • forward-Method
        • Putting it all together
    • Saving and loading
      • Saving
      • Loading
    • A Trainer to train
      • Things to change:
        • The Default Arguments
        • Resuming Training
        • Saving and Loading
      • A Whole Trainer
    • Wrapping it all in an Experiment
    • Testing it

API Documentation:

  • API Documentation
    • Delira
      • Data Loading
        • Arbitrary Data
        • Nii
        • Sampler
      • IO
      • Logging
      • Models
        • Chainer
        • SciKit-Learn
        • TensorFLow Eager Execution
        • TensorFlow Graph Execution
        • PyTorch
        • TorchScript
      • Training
        • Parameters
        • Network Trainer
        • Predictor
        • Experiment
        • Backends
        • Callbacks
        • Losses
        • Metrics
        • Utilities
      • Utilities
      • Backend Resolution
      • Debug Mode
      • Class Hierarchy Diagrams
  • GitHub
delira
  • Docs »
  • API Documentation »
  • Delira »
  • Training »
  • Backends
  • Edit on GitHub

BackendsΒΆ

The following section contains all backends which are implemented, developed and maintained for usage with delira.

A single backend usually contains at least a trainer, an experiment and some models (which are capsuled in the `models<../../models/models>`_ section.

  • Chainer
    • ChainerNetworkTrainer
    • ChainerExperiment
    • convert_chainer_to_numpy
    • create_chainer_optims_default
  • SciKit-Learn
    • SklearnEstimatorTrainer
    • SklearnExperiment
    • create_sklearn_optims_default
  • TensorFlow Eager Execution
    • TfEagerNetworkTrainer
    • TfEagerExperiment
    • create_tfeager_optims_default
    • convert_tfeager_to_numpy
  • Tensorflow Graph Execution
    • TfGraphNetworkTrainer
    • TfGraphExperiment
    • initialize_uninitialized
  • PyTorch
    • PyTorchNetworkTrainer
    • PyTorchExperiment
    • create_pytorch_optims_default
    • convert_torch_to_numpy
  • TorchScript
    • TorchScriptNetworkTrainer
    • TorchScriptExperiment
Next Previous

© Copyright 2019, Justus Schock, Michael Baumgartner, Oliver Rippel, Christoph Haarburger Revision 3b7794be.

Built with Sphinx using a theme provided by Read the Docs.