config.rst

.. Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
..
.. SPDX-License-Identifier: GPL-3.0-or-later

.. _ptbench.config:

Preset Configurations
---------------------

This module contains preset configurations for baseline CNN architectures and
datamodules.


.. _ptbench.config.models:

Models
======

.. autosummary::
   :toctree: api/models/config
   :template: config.rst

   ptbench.models.config.alexnet
   ptbench.models.config.alexnet_pretrained
   ptbench.models.config.densenet
   ptbench.models.config.densenet_pretrained
   ptbench.models.config.densenet_rs
   ptbench.models.config.logistic_regression
   ptbench.models.config.mlp
   ptbench.models.config.pasa


.. _ptbench.config.datamodules:

Preset Datamodules
==================

Datamodules include iterative accessors to raw data
(:ref:`ptbench.setup.databases`) including data pre-processing to a minimum
input standard, if applicable.  Use these datamodules for training and evaluating
your models.

.. autosummary::
   :toctree: api/data/config
   :template: config.rst

   ptbench.data.indian.default
   ptbench.data.montgomery.default
   ptbench.data.nih_cxr14.default
   ptbench.data.padchest.idiap
   ptbench.data.shenzhen.default
   ptbench.data.tbx11k.v1_healthy_vs_atb
   ptbench.data.tbx11k.v2_others_vs_atb


.. _ptbench.config.datamodules.remix:

Remix Datamodules
=================

We provide some aggregated datamodules to facilitate cross-database development.

.. autosummary::
   :toctree: api/data/config
   :template: config.rst

   ptbench.data.montgomery_shenzhen.default
   ptbench.data.montgomery_shenzhen_indian.default
   ptbench.data.montgomery_shenzhen_indian_padchest.default
   ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_healthy_vs_atb
   ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_others_vs_atb
   ptbench.data.nih_cxr14_padchest.idiap


.. _ptbench.config.datamodules.folds:

Cross-validation Datamodules
============================

We support cross-validation with precise preset folds.  In this section, you
will find the configuration for the first fold (fold-0) for all supported
datamodules.  Nine other folds are available for every configuration (from 1 to
9), making up 10 folds per supported datamodule.


.. autosummary::
   :toctree: api/data/config
   :template: config.rst

   ptbench.data.hivtb.fold_0
   ptbench.data.indian.fold_0
   ptbench.data.montgomery.fold_0
   ptbench.data.montgomery_shenzhen.fold_0
   ptbench.data.montgomery_shenzhen_indian.fold_0
   ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_0
   ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_0
   ptbench.data.shenzhen.fold_0
   ptbench.data.tbpoc.fold_0
   ptbench.data.tbx11k.v1_fold_0
   ptbench.data.tbx11k.v2_fold_0

.. include:: links.rst