From 179427210442441d766d3005ae09381fc5288747 Mon Sep 17 00:00:00 2001
From: Andre Anjos <andre.dos.anjos@gmail.com>
Date: Thu, 3 Aug 2023 17:54:27 +0200
Subject: [PATCH] [doc] Make it pass
---
doc/api.rst | 72 +--
doc/conf.py | 1 +
doc/config.rst | 32 +-
doc/links.rst | 1 +
doc/results/index.rst | 30 +-
doc/usage/evaluation.rst | 4 +-
pyproject.toml | 14 +-
src/ptbench/data/datamodule.py | 72 ++-
src/ptbench/data/hivtb/datamodule.py | 42 +-
src/ptbench/data/hivtb/fold_0.py | 2 +-
src/ptbench/data/hivtb/fold_1.py | 2 +-
src/ptbench/data/hivtb/fold_2.py | 2 +-
src/ptbench/data/hivtb/fold_3.py | 2 +-
src/ptbench/data/hivtb/fold_4.py | 2 +-
src/ptbench/data/hivtb/fold_5.py | 2 +-
src/ptbench/data/hivtb/fold_6.py | 2 +-
src/ptbench/data/hivtb/fold_7.py | 2 +-
src/ptbench/data/hivtb/fold_8.py | 2 +-
src/ptbench/data/hivtb/fold_9.py | 2 +-
src/ptbench/data/indian/datamodule.py | 22 +-
src/ptbench/data/indian/fold_0.py | 5 +
src/ptbench/data/indian/fold_1.py | 5 +
src/ptbench/data/indian/fold_2.py | 5 +
src/ptbench/data/indian/fold_3.py | 5 +
src/ptbench/data/indian/fold_4.py | 5 +
src/ptbench/data/indian/fold_5.py | 5 +
src/ptbench/data/indian/fold_6.py | 5 +
src/ptbench/data/indian/fold_7.py | 5 +
src/ptbench/data/indian/fold_8.py | 5 +
src/ptbench/data/indian/fold_9.py | 5 +
src/ptbench/data/montgomery/datamodule.py | 44 +-
src/ptbench/data/montgomery/fold_0.py | 4 +
src/ptbench/data/montgomery/fold_1.py | 4 +
src/ptbench/data/montgomery/fold_2.py | 4 +
src/ptbench/data/montgomery/fold_3.py | 4 +
src/ptbench/data/montgomery/fold_4.py | 4 +
src/ptbench/data/montgomery/fold_5.py | 4 +
src/ptbench/data/montgomery/fold_6.py | 4 +
src/ptbench/data/montgomery/fold_7.py | 4 +
src/ptbench/data/montgomery/fold_8.py | 4 +
src/ptbench/data/montgomery/fold_9.py | 4 +
.../data/montgomery_shenzhen/default.py | 5 +
.../data/montgomery_shenzhen/fold_0.py | 5 +
.../data/montgomery_shenzhen/fold_1.py | 5 +
.../data/montgomery_shenzhen/fold_2.py | 5 +
.../data/montgomery_shenzhen/fold_3.py | 5 +
.../data/montgomery_shenzhen/fold_4.py | 5 +
.../data/montgomery_shenzhen/fold_5.py | 5 +
.../data/montgomery_shenzhen/fold_6.py | 5 +
.../data/montgomery_shenzhen/fold_7.py | 5 +
.../data/montgomery_shenzhen/fold_8.py | 5 +
.../data/montgomery_shenzhen/fold_9.py | 5 +
.../montgomery_shenzhen_indian/datamodule.py | 3 +-
.../montgomery_shenzhen_indian/default.py | 5 +
.../data/montgomery_shenzhen_indian/fold_0.py | 6 +
.../data/montgomery_shenzhen_indian/fold_1.py | 5 +
.../data/montgomery_shenzhen_indian/fold_2.py | 6 +
.../data/montgomery_shenzhen_indian/fold_3.py | 6 +
.../data/montgomery_shenzhen_indian/fold_4.py | 6 +
.../data/montgomery_shenzhen_indian/fold_5.py | 6 +
.../data/montgomery_shenzhen_indian/fold_6.py | 6 +
.../data/montgomery_shenzhen_indian/fold_7.py | 5 +
.../data/montgomery_shenzhen_indian/fold_8.py | 6 +
.../data/montgomery_shenzhen_indian/fold_9.py | 5 +
.../datamodule.py | 3 +-
src/ptbench/data/nih_cxr14/datamodule.py | 36 +-
src/ptbench/data/nih_cxr14/default.py | 2 +-
.../data/nih_cxr14_padchest/datamodule.py | 3 +-
src/ptbench/data/padchest/datamodule.py | 436 +++++++++---------
src/ptbench/data/padchest/idiap.py | 2 +-
src/ptbench/data/shenzhen/datamodule.py | 47 +-
src/ptbench/data/shenzhen/default.py | 4 +-
src/ptbench/data/shenzhen/fold_0.py | 4 +
src/ptbench/data/shenzhen/fold_1.py | 4 +
src/ptbench/data/shenzhen/fold_2.py | 4 +
src/ptbench/data/shenzhen/fold_3.py | 4 +
src/ptbench/data/shenzhen/fold_4.py | 4 +
src/ptbench/data/shenzhen/fold_5.py | 4 +
src/ptbench/data/shenzhen/fold_6.py | 4 +
src/ptbench/data/shenzhen/fold_7.py | 4 +
src/ptbench/data/shenzhen/fold_8.py | 4 +
src/ptbench/data/shenzhen/fold_9.py | 4 +
src/ptbench/data/tbpoc/datamodule.py | 49 +-
src/ptbench/data/tbpoc/fold_0.py | 2 +-
src/ptbench/data/tbpoc/fold_1.py | 2 +-
src/ptbench/data/tbpoc/fold_2.py | 2 +-
src/ptbench/data/tbpoc/fold_3.py | 2 +-
src/ptbench/data/tbpoc/fold_4.py | 2 +-
src/ptbench/data/tbpoc/fold_5.py | 2 +-
src/ptbench/data/tbpoc/fold_6.py | 2 +-
src/ptbench/data/tbpoc/fold_7.py | 2 +-
src/ptbench/data/tbpoc/fold_8.py | 2 +-
src/ptbench/data/tbpoc/fold_9.py | 2 +-
src/ptbench/data/tbx11k/datamodule.py | 129 ++----
src/ptbench/data/tbx11k/v1_healthy_vs_atb.py | 24 +
src/ptbench/data/tbx11k/v2_others_vs_atb.py | 24 +
src/ptbench/data/typing.py | 17 +-
src/ptbench/models/alexnet.py | 10 +-
src/ptbench/models/densenet.py | 10 +-
src/ptbench/models/normalizer.py | 2 +-
src/ptbench/models/pasa.py | 8 +-
src/ptbench/scripts/train_analysis.py | 14 +-
src/ptbench/utils/checkpointer.py | 12 +-
103 files changed, 804 insertions(+), 652 deletions(-)
diff --git a/doc/api.rst b/doc/api.rst
index 0114d9ff..41fc340f 100644
--- a/doc/api.rst
+++ b/doc/api.rst
@@ -22,41 +22,49 @@ Auxiliary classes and methods to define raw dataset iterators.
.. autosummary::
:toctree: api/data
- ptbench.data.sample
- ptbench.data.dataset
- ptbench.data.utils
- ptbench.data.loader
- ptbench.data.transforms
- ptbench.configs.datasets
+ ptbench.data.augmentations
+ ptbench.data.datamodule
+ ptbench.data.image_utils
+ ptbench.data.split
+ ptbench.data.typing
+.. _ptbench.api.rawdata:
+
+Basic Datamodules
+-----------------
+
+Pytorch_ :py:class:`torch.utils.data.DataLoader` access through lightning_
+:py:class:`lightning.pytorch.core.LightningDataModule`.
+
+.. autosummary::
+ :toctree: api/data/raw
+
+ ptbench.data.hivtb.datamodule
+ ptbench.data.indian.datamodule
+ ptbench.data.montgomery.datamodule
+ ptbench.data.nih_cxr14.datamodule
+ ptbench.data.padchest.datamodule
+ ptbench.data.shenzhen.datamodule
+ ptbench.data.tbpoc.datamodule
+ ptbench.data.tbx11k.datamodule
-.. _ptbench.api.data.raw:
-Raw Dataset Access
-------------------
+.. _ptbench.api.remixdata:
-Direct data-access through iterators.
+Remixed Datamodules
+-------------------
+
+Pytorch_ :py:class:`torch.utils.data.DataLoader` access through lightning_
+:py:class:`lightning.pytorch.core.LightningDataModule`.
.. autosummary::
:toctree: api/data/raw
- ptbench.data.hivtb_RS
- ptbench.data.tbpoc
- ptbench.data.montgomery_RS
- ptbench.data.padchest
- ptbench.data.hivtb
- ptbench.data.indian_RS
- ptbench.data.shenzhen_RS
- ptbench.data.tbpoc_RS
- ptbench.data.shenzhen
- ptbench.data.montgomery
- ptbench.data.indian
- ptbench.data.nih_cxr14_re
- ptbench.data.padchest_RS
- ptbench.data.tbx11k_simplified
- ptbench.data.tbx11k_simplified_RS
- ptbench.data.tbx11k_simplified_v2
- ptbench.data.tbx11k_simplified_v2_RS
+ ptbench.data.montgomery_shenzhen.datamodule
+ ptbench.data.montgomery_shenzhen_indian.datamodule
+ ptbench.data.montgomery_shenzhen_indian_tbx11k.datamodule
+ ptbench.data.montgomery_shenzhen_indian_padchest.datamodule
+ ptbench.data.nih_cxr14_padchest.datamodule
.. _ptbench.api.models:
@@ -69,12 +77,11 @@ CNN and other models implemented.
.. autosummary::
:toctree: api/models
+ ptbench.models.pasa
ptbench.models.alexnet
ptbench.models.densenet
- ptbench.models.densenet_rs
- ptbench.models.logistic_regression
ptbench.models.normalizer
- ptbench.models.pasa
+ ptbench.models.logistic_regression
ptbench.models.signs_to_tb
@@ -88,11 +95,14 @@ Functions to actuate on the data.
.. autosummary::
:toctree: api/engine
+ ptbench.engine.device
+ ptbench.engine.callbacks
ptbench.engine.trainer
ptbench.engine.predictor
ptbench.engine.evaluator
+
.. _ptbench.api.utils:
Various utilities
@@ -106,6 +116,7 @@ Reusable auxiliary functions.
ptbench.utils.checkpointer
ptbench.utils.download
ptbench.utils.grad_cams
+ ptbench.utils.image
ptbench.utils.measure
ptbench.utils.model_serialization
ptbench.utils.model_zoo
@@ -114,6 +125,7 @@ Reusable auxiliary functions.
ptbench.utils.resources
ptbench.utils.summary
ptbench.utils.table
+ ptbench.utils.tensorboard
.. include:: links.rst
diff --git a/doc/conf.py b/doc/conf.py
index 49b7ceac..b69ef1b9 100644
--- a/doc/conf.py
+++ b/doc/conf.py
@@ -120,6 +120,7 @@ auto_intersphinx_packages = [
"pandas",
"pillow",
"psutil",
+ "scipy",
"torch",
"torchvision",
"lightning",
diff --git a/doc/config.rst b/doc/config.rst
index b0774e35..34025f7f 100644
--- a/doc/config.rst
+++ b/doc/config.rst
@@ -11,24 +11,26 @@ This module contains preset configurations for baseline CNN architectures and
datamodules.
+.. _ptbench.config.models:
+
Models
======
.. autosummary::
- :toctree: api/configs/models
+ :toctree: api/models/config
:template: config.rst
- ptbench.configs.models.alexnet
- ptbench.configs.models.alexnet_pretrained
- ptbench.configs.models.densenet
- ptbench.configs.models.densenet_pretrained
- ptbench.configs.models.densenet_rs
- ptbench.configs.models.logistic_regression
- ptbench.configs.models.pasa
- ptbench.configs.models.signs_to_tb
+ 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.pasa
+ ptbench.models.config.signs_to_tb
-.. _ptbench.configs.datamodules:
+.. _ptbench.config.datamodules:
Preset Datamodules
==================
@@ -39,7 +41,7 @@ input standard, if applicable. Use these datamodules for training and evaluatin
your models.
.. autosummary::
- :toctree: api/configs/datamodules
+ :toctree: api/data/config
:template: config.rst
ptbench.data.indian.default
@@ -51,7 +53,7 @@ your models.
ptbench.data.tbx11k.v2_others_vs_atb
-.. _ptbench.configs.datamodules.remix:
+.. _ptbench.config.datamodules.remix:
Remix Datamodules
=================
@@ -59,7 +61,7 @@ Remix Datamodules
We provide some aggregated datamodules to facilitate cross-database development.
.. autosummary::
- :toctree: api/configs/datamodules
+ :toctree: api/data/config
:template: config.rst
ptbench.data.montgomery_shenzhen.default
@@ -70,7 +72,7 @@ We provide some aggregated datamodules to facilitate cross-database development.
ptbench.data.nih_cxr14_padchest.idiap
-.. _ptbench.configs.datamodules.folds:
+.. _ptbench.config.datamodules.folds:
Cross-validation Datamodules
============================
@@ -82,7 +84,7 @@ datamodules. Nine other folds are available for every configuration (from 1 to
.. autosummary::
- :toctree: api/configs/datamodules
+ :toctree: api/data/config
:template: config.rst
ptbench.data.hivtb.fold_0
diff --git a/doc/links.rst b/doc/links.rst
index 92ecafd5..d773d1ae 100644
--- a/doc/links.rst
+++ b/doc/links.rst
@@ -11,6 +11,7 @@
.. _pip: https://pip.pypa.io/en/stable/
.. _mamba: https://mamba.readthedocs.io/en/latest/index.html
.. _pytorch: https://pytorch.org
+.. _lightning: https://lightning.ai
.. Raw data websites
.. _montgomery: https://lhncbc.nlm.nih.gov/publication/pub9931
diff --git a/doc/results/index.rst b/doc/results/index.rst
index 29ad8a10..0217fe6c 100644
--- a/doc/results/index.rst
+++ b/doc/results/index.rst
@@ -113,37 +113,37 @@ Thresholds used:
:scale: 50%
:alt: Testing sets ROC curves for Pasa model trained on normalized-kfold MC
- :py:mod:`Pasa <ptbench.configs.models.pasa>`: Pasa trained on normalized-kfold MC
+ :py:mod:`Pasa <ptbench.models.config.pasa>`: Pasa trained on normalized-kfold MC
- .. figure:: img/compare_pasa_mc_ch_kfold_500.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for Pasa model trained on normalized-kfold MC-CH
- :py:mod:`Pasa <ptbench.configs.models.pasa>`: Pasa trained on normalized-kfold MC-CH
+ :py:mod:`Pasa <ptbench.models.config.pasa>`: Pasa trained on normalized-kfold MC-CH
- .. figure:: img/compare_pasa_mc_ch_in_kfold_500.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for Pasa model trained on normalized-kfold MC-CH-IN
- :py:mod:`Pasa <ptbench.configs.models.pasa>`: Pasa trained on normalized-kfold MC-CH-IN
+ :py:mod:`Pasa <ptbench.models.config.pasa>`: Pasa trained on normalized-kfold MC-CH-IN
* - .. figure:: img/compare_densenet_mc_kfold_2000.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC
- :py:mod:`DenseNet <ptbench.configs.models.densenet>`: DenseNet trained on normalized-kfold MC
+ :py:mod:`DenseNet <ptbench.models.config.densenet>`: DenseNet trained on normalized-kfold MC
- .. figure:: img/compare_densenet_mc_ch_kfold_2000.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC-CH
- :py:mod:`DenseNet <ptbench.configs.models.densenet>`: DenseNet trained on normalized-kfold MC-CH
+ :py:mod:`DenseNet <ptbench.models.config.densenet>`: DenseNet trained on normalized-kfold MC-CH
- .. figure:: img/compare_densenet_mc_ch_in_kfold_2000.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC-CH-IN
- :py:mod:`DenseNet <ptbench.configs.models.densenet>`: DenseNet trained on normalized-kfold MC-CH-IN
+ :py:mod:`DenseNet <ptbench.models.config.densenet>`: DenseNet trained on normalized-kfold MC-CH-IN
DenseNet-121 (pretrained on ImageNet)
"""""""""""""""""""""""""""""""""""""
@@ -180,19 +180,19 @@ Thresholds used:
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC
- :py:mod:`DenseNet <ptbench.configs.models.densenet>` DenseNet trained on normalized-kfold MC
+ :py:mod:`DenseNet <ptbench.models.config.densenet>` DenseNet trained on normalized-kfold MC
- .. figure:: img/compare_densenetpreIN_mc_ch_kfold_600.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC-CH
- :py:mod:`DenseNet <ptbench.configs.models.densenet>` DenseNet trained on normalized-kfold MC-CH
+ :py:mod:`DenseNet <ptbench.models.config.densenet>` DenseNet trained on normalized-kfold MC-CH
- .. figure:: img/compare_densenetpreIN_mc_ch_ch_kfold_600.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC-CH-IN
- :py:mod:`DenseNet <ptbench.configs.models.densenet>` DenseNet trained on normalized-kfold MC-CH-IN
+ :py:mod:`DenseNet <ptbench.models.config.densenet>` DenseNet trained on normalized-kfold MC-CH-IN
Logistic Regression Classifier
""""""""""""""""""""""""""""""
@@ -229,19 +229,19 @@ Thresholds used:
:scale: 50%
:alt: Testing sets ROC curves for LogReg model trained on normalized-kfold MC
- :py:mod:`LogReg <ptbench.configs.models.logistic_regression>`: LogReg trained on normalized-kfold MC
+ :py:mod:`LogReg <ptbench.models.config.logistic_regression>`: LogReg trained on normalized-kfold MC
- .. figure:: img/compare_logreg_mc_ch_kfold_100.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for LogReg model trained on normalized-kfold MC-CH
- :py:mod:`LogReg <ptbench.configs.models.logistic_regression>`: LogReg trained on normalized-kfold MC-CH
+ :py:mod:`LogReg <ptbench.models.config.logistic_regression>`: LogReg trained on normalized-kfold MC-CH
- .. figure:: img/compare_logreg_mc_ch_in_kfold_100.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for LogReg model trained on normalized-kfold MC-CH-IN
- :py:mod:`LogReg <ptbench.configs.models.logistic_regression>`: LogReg trained on normalized-kfold MC-CH-IN
+ :py:mod:`LogReg <ptbench.models.config.logistic_regression>`: LogReg trained on normalized-kfold MC-CH-IN
DenseNet-121 (pretrained on ImageNet and NIH CXR14)
"""""""""""""""""""""""""""""""""""""""""""""""""""
@@ -278,19 +278,19 @@ Thresholds used:
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC (pretrained on NIH)
- :py:mod:`DenseNet <ptbench.configs.models.densenet>`: DenseNet trained on normalized-kfold MC (pretrained on NIH)
+ :py:mod:`DenseNet <ptbench.models.config.densenet>`: DenseNet trained on normalized-kfold MC (pretrained on NIH)
- .. figure:: img/compare_densenetpre_mc_ch_kfold_300.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC-CH (pretrained on NIH)
- :py:mod:`DenseNet <ptbench.configs.models.densenet>`: DenseNet trained on normalized-kfold MC-CH (pretrained on NIH)
+ :py:mod:`DenseNet <ptbench.models.config.densenet>`: DenseNet trained on normalized-kfold MC-CH (pretrained on NIH)
- .. figure:: img/compare_densenetpre_mc_ch_in_kfold_300.jpg
:align: center
:scale: 50%
:alt: Testing sets ROC curves for DenseNet model trained on normalized-kfold MC-CH-IN (pretrained on NIH)
- :py:mod:`DenseNet <ptbench.configs.models.densenet>`: DenseNet trained on normalized-kfold MC-CH-IN (pretrained on NIH)
+ :py:mod:`DenseNet <ptbench.models.config.densenet>`: DenseNet trained on normalized-kfold MC-CH-IN (pretrained on NIH)
Global sensitivity analysis (relevance)
diff --git a/doc/usage/evaluation.rst b/doc/usage/evaluation.rst
index d8b15040..8d98ca92 100644
--- a/doc/usage/evaluation.rst
+++ b/doc/usage/evaluation.rst
@@ -35,8 +35,8 @@ the pre-trained model.
.. tip::
An option to generate grad-CAMs is available for the :py:mod:`DensenetRS
- <ptbench.configs.models_datasets.densenet_rs>` model. To activate it, use
- the ``--grad-cams`` argument.
+ <ptbench.models.config.densenet_rs>` model. To activate it, use the
+ ``--grad-cams`` argument.
.. tip::
diff --git a/pyproject.toml b/pyproject.toml
index 56544f73..67bd370c 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -70,13 +70,13 @@ ptbench = "ptbench.scripts.cli:cli"
[project.entry-points."ptbench.config"]
# models
-pasa = "ptbench.configs.models.pasa"
-signs-to-tb = "ptbench.configs.models.signs_to_tb"
-logistic-regression = "ptbench.configs.models.logistic_regression"
-alexnet = "ptbench.configs.models.alexnet"
-alexnet-pretrained = "ptbench.configs.models.alexnet_pretrained"
-densenet = "ptbench.configs.models.densenet"
-densenet-pretrained = "ptbench.configs.models.densenet_pretrained"
+pasa = "ptbench.models.config.pasa"
+signs-to-tb = "ptbench.models.config.signs_to_tb"
+logistic-regression = "ptbench.models.config.logistic_regression"
+alexnet = "ptbench.models.config.alexnet"
+alexnet-pretrained = "ptbench.models.config.alexnet_pretrained"
+densenet = "ptbench.models.config.densenet"
+densenet-pretrained = "ptbench.models.config.densenet_pretrained"
# montgomery dataset (and cross-validation folds)
montgomery = "ptbench.data.montgomery.default"
diff --git a/src/ptbench/data/datamodule.py b/src/ptbench/data/datamodule.py
index 0785e233..0ab3c36c 100644
--- a/src/ptbench/data/datamodule.py
+++ b/src/ptbench/data/datamodule.py
@@ -36,15 +36,12 @@ def _sample_size_bytes(s: Sample) -> int:
Parameters
----------
-
s
The sample to be analyzed
Returns
-------
-
- size
The size in bytes occupied by this sample
"""
@@ -74,7 +71,6 @@ class _DelayedLoadingDataset(Dataset):
Parameters
----------
-
raw_dataset
An iterable containing the raw dataset samples representing one of the
database split datasets.
@@ -132,7 +128,6 @@ def _apply_loader_and_transforms(
Parameters
----------
-
info
The sample information, as loaded from its raw dataset dictionary
@@ -147,8 +142,6 @@ def _apply_loader_and_transforms(
Returns
-------
-
- sample
The loaded and transformed sample.
"""
sample = load(info)
@@ -165,7 +158,6 @@ class _CachedDataset(Dataset):
Parameters
----------
-
raw_dataset
An iterable containing the raw dataset samples representing one of the
database split datasets.
@@ -243,7 +235,6 @@ class _ConcatDataset(Dataset):
Parameters
----------
-
datasets
An iterable over pre-instantiated datasets.
"""
@@ -323,8 +314,8 @@ def _make_balanced_random_sampler(
Dataset 2 proportionally less likely.
This function assumes targets are stored on a dictionary entry named
- ``target`` inside the metadata information for the :py:type:``Sample``, and
- that its value is integer.
+ ``target`` inside the metadata information for the
+ :py:data:`.typing.Sample`, and that its value is integer.
We then instantiate a pytorch sampler using the inverse probabilities (the
more samples of a class, the less likely it becomes to be sampled.
@@ -332,7 +323,6 @@ def _make_balanced_random_sampler(
Parameters
----------
-
dataset
An instance of torch Dataset.
:py:class:`torch.utils.data.ConcatDataset` are supported.
@@ -344,15 +334,12 @@ def _make_balanced_random_sampler(
Returns
-------
-
- sampler
A sampler, to be used in a dataloader equipped with the same dataset
used to calculate the relative sample weights.
Raises
------
-
RuntimeError
If requested to balance a dataset (single, not-concatenated) without an
existing target.
@@ -434,23 +421,22 @@ class ConcatDataModule(lightning.LightningDataModule):
Parameters
----------
-
splits
A dictionary that contains string keys representing dataset names, and
values that are iterables over a 2-tuple containing an iterable over
arbitrary, user-configurable sample representations (potentially on
- disk or permanent storage), and :py:class:`RawDataLoader` (or "sample")
- loader objects, which concretely implement a mechanism to load such
- samples in memory, from permanent storage.
+ disk or permanent storage), and :py:class:`.typing.RawDataLoader` (or
+ "sample") loader objects, which concretely implement a mechanism to
+ load such samples in memory, from permanent storage.
Sample representations on permanent storage may be of any iterable
format (e.g. list, dictionary, etc.), for as long as the assigned
- :py:class:`RawDataLoader` can properly handle it.
+ :py:class:`.typing.RawDataLoader` can properly handle it.
.. tip::
To check the split and the loader function works correctly, you may
- use :py:func:`..dataset.check_database_split_loading`.
+ use :py:func:`.split.check_database_split_loading`.
This class expects at least one entry called ``train`` to exist in the
input dictionary. Optional entries are ``validation``, and ``test``.
@@ -507,24 +493,24 @@ class ConcatDataModule(lightning.LightningDataModule):
multiprocessing data loading. Set to 0 to enable as many data loading
instances as processing cores as available in the system. Set to >= 1
to enable that many multiprocessing instances for data loading.
-
-
- Attributes
- ----------
-
- model_transforms
- A list of transforms (torch modules) that will be applied after
- raw-data-loading, and just before data is fed into the model or
- eventual data-augmentation transformations for all data loaders
- produced by this data module. This part of the pipeline receives data
- as output by the raw-data-loader, or model-related transforms (e.g.
- resize adaptions), if any is specified. If data is cached, it is
- cached **after** model-transforms are applied, as that is a potential
- memory saver (e.g., if it contains a resizing operation to smaller
- images).
"""
- DatasetDictionary = dict[str, Dataset]
+ DatasetDictionary: typing.TypeAlias = dict[str, Dataset]
+ """A dictionary of datasets mapping names to actual datasets."""
+
+ model_transforms: list[Transform] | None
+ """Transforms required to fit data into the model.
+
+ A list of transforms (torch modules) that will be applied after raw-
+ data-loading. and just before data is fed into the model or eventual
+ data-augmentation transformations for all data loaders produced by
+ this data module. This part of the pipeline receives data as output
+ by the raw-data-loader, or model-related transforms (e.g. resize
+ adaptions), if any is specified. If data is cached, it is cached
+ **after** model-transforms are applied, as that is a potential
+ memory saver (e.g., if it contains a resizing operation to smaller
+ images).
+ """
def __init__(
self,
@@ -653,7 +639,6 @@ class ConcatDataModule(lightning.LightningDataModule):
Parameters
----------
-
batch_size
Number of samples in every **training** batch (this parameter affects
memory requirements for the network). If the number of samples in the
@@ -696,7 +681,6 @@ class ConcatDataModule(lightning.LightningDataModule):
Parameters
----------
-
name
Name of the dataset to setup.
"""
@@ -769,7 +753,6 @@ class ConcatDataModule(lightning.LightningDataModule):
Parameters
----------
-
stage
Name of the stage to which the setup is applicable. Can be one of
``fit``, ``validate``, ``test`` or ``predict``. Each stage
@@ -808,7 +791,6 @@ class ConcatDataModule(lightning.LightningDataModule):
Parameters
----------
-
stage
Name of the stage to which the teardown is applicable. Can be one of
``fit``, ``validate``, ``test`` or ``predict``. Each stage
@@ -903,19 +885,19 @@ class CachingDataModule(ConcatDataModule):
Parameters
----------
-
database_split
A dictionary that contains string keys representing dataset names, and
values that are iterables over sample representations (potentially on
- disk). These objects are passed to an unique :py:class:`RawDataLoader`
- for loading the :py:class:`Sample` data (and metadata) in memory. It
+ disk). These objects are passed to an unique
+ :py:class:`.typing.RawDataLoader` for loading the
+ :py:data:`.typing.Sample` data (and metadata) in memory. It
therefore assumes the whole split is homogeneous and can be loaded in
the same way.
.. tip::
To check the split and the loader function works correctly, you may
- use :py:func:`..dataset.check_database_split_loading`.
+ use :py:func:`.split.check_database_split_loading`.
This class expects at least one entry called ``train`` to exist in the
input dictionary. Optional entries are ``validation``, and ``test``.
diff --git a/src/ptbench/data/hivtb/datamodule.py b/src/ptbench/data/hivtb/datamodule.py
index b5b84ec4..2f1dcb85 100644
--- a/src/ptbench/data/hivtb/datamodule.py
+++ b/src/ptbench/data/hivtb/datamodule.py
@@ -19,17 +19,11 @@ from ..typing import Sample
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the HIV-TB dataset.
-
- Attributes
- ----------
-
- datadir
- This variable contains the base directory where the database raw data
- is stored.
- """
+ """A specialized raw-data-loader for the HIV-TB dataset."""
datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
def __init__(self):
self.datadir = load_rc().get(
@@ -41,8 +35,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -50,8 +43,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
image = PIL.Image.open(os.path.join(self.datadir, sample[0])).convert(
@@ -73,8 +64,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -82,8 +72,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- label
The integer label associated with the sample
"""
return sample[1]
@@ -108,23 +96,23 @@ class DataModule(CachingDataModule):
* Raw data input (on disk):
- * BMP (BMP3) and JPEG grayscale images encoded as 8-bit RGB, with
- varying resolution
+ * BMP (BMP3) and JPEG grayscale images encoded as 8-bit RGB, with
+ varying resolution
* Output image:
- * Transforms:
+ * Transforms:
- * Load raw BMP or JPEG with :py:mod:`PIL`
- * Remove black borders
- * Convert to torch tensor
- * Torch center cropping to get square image
+ * Load raw BMP or JPEG with :py:mod:`PIL`
+ * Remove black borders
+ * Convert to torch tensor
+ * Torch center cropping to get square image
* Final specifications
- * Grayscale, encoded as a single plane tensor, 32-bit floats,
- square at 2048 x 2048 pixels
- * Labels: 0 (healthy), 1 (active tuberculosis)
+ * Grayscale, encoded as a single plane tensor, 32-bit floats,
+ square at 2048 x 2048 pixels
+ * Labels: 0 (healthy), 1 (active tuberculosis)
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/hivtb/fold_0.py b/src/ptbench/data/hivtb/fold_0.py
index 57d77952..6919907d 100644
--- a/src/ptbench/data/hivtb/fold_0.py
+++ b/src/ptbench/data/hivtb/fold_0.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
"""HIV-TB dataset for TB detection (cross validation fold 0).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_1.py b/src/ptbench/data/hivtb/fold_1.py
index c91a968f..1f4b1dd0 100644
--- a/src/ptbench/data/hivtb/fold_1.py
+++ b/src/ptbench/data/hivtb/fold_1.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
"""HIV-TB dataset for TB detection (cross validation fold 1).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_2.py b/src/ptbench/data/hivtb/fold_2.py
index 323e80a0..9c9fbe65 100644
--- a/src/ptbench/data/hivtb/fold_2.py
+++ b/src/ptbench/data/hivtb/fold_2.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
"""HIV-TB dataset for TB detection (cross validation fold 2).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_3.py b/src/ptbench/data/hivtb/fold_3.py
index 1eed4c05..ef07c591 100644
--- a/src/ptbench/data/hivtb/fold_3.py
+++ b/src/ptbench/data/hivtb/fold_3.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
"""HIV-TB dataset for TB detection (cross validation fold 3).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_4.py b/src/ptbench/data/hivtb/fold_4.py
index 9cfa6186..6683006d 100644
--- a/src/ptbench/data/hivtb/fold_4.py
+++ b/src/ptbench/data/hivtb/fold_4.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
"""HIV-TB dataset for TB detection (cross validation fold 4).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_5.py b/src/ptbench/data/hivtb/fold_5.py
index 591fef37..cf67833a 100644
--- a/src/ptbench/data/hivtb/fold_5.py
+++ b/src/ptbench/data/hivtb/fold_5.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
"""HIV-TB dataset for TB detection (cross validation fold 5).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_6.py b/src/ptbench/data/hivtb/fold_6.py
index fb5e1614..94614fc4 100644
--- a/src/ptbench/data/hivtb/fold_6.py
+++ b/src/ptbench/data/hivtb/fold_6.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
"""HIV-TB dataset for TB detection (cross validation fold 6).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_7.py b/src/ptbench/data/hivtb/fold_7.py
index d64db483..259446c8 100644
--- a/src/ptbench/data/hivtb/fold_7.py
+++ b/src/ptbench/data/hivtb/fold_7.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
"""HIV-TB dataset for TB detection (cross validation fold 7).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_8.py b/src/ptbench/data/hivtb/fold_8.py
index 8a0f87d1..d243db08 100644
--- a/src/ptbench/data/hivtb/fold_8.py
+++ b/src/ptbench/data/hivtb/fold_8.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
"""HIV-TB dataset for TB detection (cross validation fold 8).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/hivtb/fold_9.py b/src/ptbench/data/hivtb/fold_9.py
index d92de50e..340bc661 100644
--- a/src/ptbench/data/hivtb/fold_9.py
+++ b/src/ptbench/data/hivtb/fold_9.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
"""HIV-TB dataset for TB detection (cross validation fold 9).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.hivtb.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/indian/datamodule.py b/src/ptbench/data/indian/datamodule.py
index f6017cad..ee53da80 100644
--- a/src/ptbench/data/indian/datamodule.py
+++ b/src/ptbench/data/indian/datamodule.py
@@ -33,23 +33,23 @@ class DataModule(CachingDataModule):
* Raw data input (on disk):
- * PNG RGB 8-bit depth images with "inverted" grayscale scale
- * Variable width and height
+ * PNG RGB 8-bit depth images with "inverted" grayscale scale
+ * Variable width and height
* Output image:
- * Transforms:
+ * Transforms:
- * Load raw PNG with :py:mod:`PIL`
- * Remove black borders
- * Convert to torch tensor
- * Torch center cropping to get square image
+ * Load raw PNG with :py:mod:`PIL`
+ * Remove black borders
+ * Convert to torch tensor
+ * Torch center cropping to get square image
- * Final specifications:
+ * Final specifications:
- * Grayscale, encoded as a single plane tensor, 32-bit floats,
- square, with varying resolutions, depending on the input raw image
- * Labels: 0 (healthy), 1 (active tuberculosis)
+ * Grayscale, encoded as a single plane tensor, 32-bit floats,
+ square, with varying resolutions, depending on the input raw image
+ * Labels: 0 (healthy), 1 (active tuberculosis)
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/indian/fold_0.py b/src/ptbench/data/indian/fold_0.py
index 91711d38..635b542f 100644
--- a/src/ptbench/data/indian/fold_0.py
+++ b/src/ptbench/data/indian/fold_0.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 0).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_1.py b/src/ptbench/data/indian/fold_1.py
index 17372695..5a3a0213 100644
--- a/src/ptbench/data/indian/fold_1.py
+++ b/src/ptbench/data/indian/fold_1.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 1).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_2.py b/src/ptbench/data/indian/fold_2.py
index 91346efa..cbf1aee4 100644
--- a/src/ptbench/data/indian/fold_2.py
+++ b/src/ptbench/data/indian/fold_2.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 2).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_3.py b/src/ptbench/data/indian/fold_3.py
index 4156c3ee..369c9528 100644
--- a/src/ptbench/data/indian/fold_3.py
+++ b/src/ptbench/data/indian/fold_3.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 3).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_4.py b/src/ptbench/data/indian/fold_4.py
index 6d5b73d1..e9137b65 100644
--- a/src/ptbench/data/indian/fold_4.py
+++ b/src/ptbench/data/indian/fold_4.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 4).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_5.py b/src/ptbench/data/indian/fold_5.py
index 27a7442d..d6f34d69 100644
--- a/src/ptbench/data/indian/fold_5.py
+++ b/src/ptbench/data/indian/fold_5.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 5).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_6.py b/src/ptbench/data/indian/fold_6.py
index 1270cd58..a293530d 100644
--- a/src/ptbench/data/indian/fold_6.py
+++ b/src/ptbench/data/indian/fold_6.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 6).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_7.py b/src/ptbench/data/indian/fold_7.py
index e7b5a1c3..22ea3439 100644
--- a/src/ptbench/data/indian/fold_7.py
+++ b/src/ptbench/data/indian/fold_7.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 7).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_8.py b/src/ptbench/data/indian/fold_8.py
index 1abbe7b2..77cf20ee 100644
--- a/src/ptbench/data/indian/fold_8.py
+++ b/src/ptbench/data/indian/fold_8.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 8).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/indian/fold_9.py b/src/ptbench/data/indian/fold_9.py
index 71de470d..a0f881bc 100644
--- a/src/ptbench/data/indian/fold_9.py
+++ b/src/ptbench/data/indian/fold_9.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
+"""Indian collection dataset for computer-aided diagnosis (cross validation
+fold 9).
+
+See :py:class:`.indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/datamodule.py b/src/ptbench/data/montgomery/datamodule.py
index 12ead6e8..f3ac8ffc 100644
--- a/src/ptbench/data/montgomery/datamodule.py
+++ b/src/ptbench/data/montgomery/datamodule.py
@@ -19,17 +19,11 @@ from ..typing import Sample
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the Montgomery dataset.
-
- Attributes
- ----------
-
- datadir
- This variable contains the base directory where the database raw data
- is stored.
- """
+ """A specialized raw-data-loader for the Montgomery dataset."""
datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
def __init__(self):
self.datadir = load_rc().get(
@@ -41,8 +35,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -50,8 +43,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
# N.B.: Montgomery images are encoded as grayscale PNGs, so no need to
@@ -73,8 +64,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -82,8 +72,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- label
The integer label associated with the sample
"""
return sample[1]
@@ -111,23 +99,23 @@ class DataModule(CachingDataModule):
* Raw data input (on disk):
- * PNG images 8 bit grayscale
- * resolution: fixed to one of the cases above
+ * PNG images 8 bit grayscale
+ * resolution: fixed to one of the cases above
* Output image:
- * Transforms:
+ * Transforms:
- * Load raw PNG with :py:mod:`PIL`
- * Remove black borders
- * Convert to torch tensor
- * Torch center cropping to get square image
+ * Load raw PNG with :py:mod:`PIL`
+ * Remove black borders
+ * Convert to torch tensor
+ * Torch center cropping to get square image
- * Final specifications
+ * Final specifications
- * Grayscale, encoded as a single plane tensor, 32-bit floats,
- square at 4020 x 4020 pixels
- * Labels: 0 (healthy), 1 (active tuberculosis)
+ * Grayscale, encoded as a single plane tensor, 32-bit floats,
+ square at 4020 x 4020 pixels
+ * Labels: 0 (healthy), 1 (active tuberculosis)
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/montgomery/fold_0.py b/src/ptbench/data/montgomery/fold_0.py
index 91711d38..a271fa0f 100644
--- a/src/ptbench/data/montgomery/fold_0.py
+++ b/src/ptbench/data/montgomery/fold_0.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
+"""Montgomery datamodule for TB detection (cross validation fold 0).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_1.py b/src/ptbench/data/montgomery/fold_1.py
index 17372695..626cf2d0 100644
--- a/src/ptbench/data/montgomery/fold_1.py
+++ b/src/ptbench/data/montgomery/fold_1.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
+"""Montgomery datamodule for TB detection (cross validation fold 1).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_2.py b/src/ptbench/data/montgomery/fold_2.py
index 91346efa..b6146d06 100644
--- a/src/ptbench/data/montgomery/fold_2.py
+++ b/src/ptbench/data/montgomery/fold_2.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
+"""Montgomery datamodule for TB detection (cross validation fold 2).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_3.py b/src/ptbench/data/montgomery/fold_3.py
index 4156c3ee..f1b1c53e 100644
--- a/src/ptbench/data/montgomery/fold_3.py
+++ b/src/ptbench/data/montgomery/fold_3.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
+"""Montgomery datamodule for TB detection (cross validation fold 3).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_4.py b/src/ptbench/data/montgomery/fold_4.py
index 6d5b73d1..eaa5fd27 100644
--- a/src/ptbench/data/montgomery/fold_4.py
+++ b/src/ptbench/data/montgomery/fold_4.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
+"""Montgomery datamodule for TB detection (cross validation fold 4).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_5.py b/src/ptbench/data/montgomery/fold_5.py
index 27a7442d..142ae715 100644
--- a/src/ptbench/data/montgomery/fold_5.py
+++ b/src/ptbench/data/montgomery/fold_5.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
+"""Montgomery datamodule for TB detection (cross validation fold 5).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_6.py b/src/ptbench/data/montgomery/fold_6.py
index 1270cd58..715d9b84 100644
--- a/src/ptbench/data/montgomery/fold_6.py
+++ b/src/ptbench/data/montgomery/fold_6.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
+"""Montgomery datamodule for TB detection (cross validation fold 6).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_7.py b/src/ptbench/data/montgomery/fold_7.py
index e7b5a1c3..fc53e24e 100644
--- a/src/ptbench/data/montgomery/fold_7.py
+++ b/src/ptbench/data/montgomery/fold_7.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
+"""Montgomery datamodule for TB detection (cross validation fold 7).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_8.py b/src/ptbench/data/montgomery/fold_8.py
index 1abbe7b2..2b917a6b 100644
--- a/src/ptbench/data/montgomery/fold_8.py
+++ b/src/ptbench/data/montgomery/fold_8.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
+"""Montgomery datamodule for TB detection (cross validation fold 8).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery/fold_9.py b/src/ptbench/data/montgomery/fold_9.py
index 71de470d..f404ace1 100644
--- a/src/ptbench/data/montgomery/fold_9.py
+++ b/src/ptbench/data/montgomery/fold_9.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
+"""Montgomery datamodule for TB detection (cross validation fold 9).
+
+See :py:class:`.montgomery.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/default.py b/src/ptbench/data/montgomery_shenzhen/default.py
index 2b8a8fb2..c332f4be 100644
--- a/src/ptbench/data/montgomery_shenzhen/default.py
+++ b/src/ptbench/data/montgomery_shenzhen/default.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("default.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (default
+split).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_0.py b/src/ptbench/data/montgomery_shenzhen/fold_0.py
index 91711d38..f3e8ef02 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_0.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_0.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 0).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_1.py b/src/ptbench/data/montgomery_shenzhen/fold_1.py
index 17372695..f6d73de3 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_1.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_1.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 1).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_2.py b/src/ptbench/data/montgomery_shenzhen/fold_2.py
index 91346efa..9b956052 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_2.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_2.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 2).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_3.py b/src/ptbench/data/montgomery_shenzhen/fold_3.py
index 4156c3ee..826e0abb 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_3.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_3.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 3).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_4.py b/src/ptbench/data/montgomery_shenzhen/fold_4.py
index 6d5b73d1..f7261a43 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_4.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_4.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 4).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_5.py b/src/ptbench/data/montgomery_shenzhen/fold_5.py
index 27a7442d..7dfb6f90 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_5.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_5.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 5).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_6.py b/src/ptbench/data/montgomery_shenzhen/fold_6.py
index 1270cd58..57f19130 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_6.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_6.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 6).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_7.py b/src/ptbench/data/montgomery_shenzhen/fold_7.py
index e7b5a1c3..3fc7c8bf 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_7.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_7.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 7).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_8.py b/src/ptbench/data/montgomery_shenzhen/fold_8.py
index 1abbe7b2..7b631543 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_8.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_8.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 8).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen/fold_9.py b/src/ptbench/data/montgomery_shenzhen/fold_9.py
index 71de470d..565b71e7 100644
--- a/src/ptbench/data/montgomery_shenzhen/fold_9.py
+++ b/src/ptbench/data/montgomery_shenzhen/fold_9.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
+"""Aggregated datamodule composed of Montgomery and Shenzhen datasets (cross
+validation fold 9).
+
+See :py:class:`.montgomery_shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/datamodule.py b/src/ptbench/data/montgomery_shenzhen_indian/datamodule.py
index ea2d2a4f..3c555a62 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/datamodule.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/datamodule.py
@@ -12,7 +12,8 @@ from ..shenzhen.datamodule import make_split as make_shenzhen_split
class DataModule(ConcatDataModule):
- """Aggregated datamodule composed of Montgomery and Shenzhen datasets."""
+ """Aggregated datamodule composed of Montgomery, Shenzhen and Indian
+ datasets."""
def __init__(self, split_filename: str):
montgomery_loader = MontgomeryLoader()
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/default.py b/src/ptbench/data/montgomery_shenzhen_indian/default.py
index 2b8a8fb2..e4a05f63 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/default.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/default.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("default.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets.
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_0.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_0.py
index 91711d38..6f08e25a 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_0.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_0.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 0).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_1.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_1.py
index 17372695..3d7529cf 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_1.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_1.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 1).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_2.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_2.py
index 91346efa..2e914281 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_2.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_2.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 2).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_3.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_3.py
index 4156c3ee..dfd4c9cd 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_3.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_3.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 3).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_4.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_4.py
index 6d5b73d1..84e66cef 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_4.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_4.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 4).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_5.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_5.py
index 27a7442d..650292f7 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_5.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_5.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 5).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_6.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_6.py
index 1270cd58..8f673689 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_6.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_6.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 6).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_7.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_7.py
index e7b5a1c3..33eed540 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_7.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_7.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 7).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_8.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_8.py
index 1abbe7b2..4de80f69 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_8.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_8.py
@@ -5,3 +5,9 @@
from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 8).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical
+details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian/fold_9.py b/src/ptbench/data/montgomery_shenzhen_indian/fold_9.py
index 71de470d..f33e691b 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian/fold_9.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian/fold_9.py
@@ -5,3 +5,8 @@
from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
+"""Aggregated datamodule composed of Montgomery, Shenzhen and Indian datasets
+(cross validation fold 9).
+
+See :py:class:`.montgomery_shenzhen_indian.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_padchest/datamodule.py b/src/ptbench/data/montgomery_shenzhen_indian_padchest/datamodule.py
index 2fdcfc67..9f6a3b09 100644
--- a/src/ptbench/data/montgomery_shenzhen_indian_padchest/datamodule.py
+++ b/src/ptbench/data/montgomery_shenzhen_indian_padchest/datamodule.py
@@ -38,7 +38,8 @@ class DataModule(ConcatDataModule):
(montgomery_split["validation"], montgomery_loader),
(shenzhen_split["validation"], shenzhen_loader),
(indian_split["validation"], indian_loader),
- (padchest_split["validation"], padchest_loader),
+ # there is no validation set on padchest...
+ # (padchest_split["validation"], padchest_loader),
],
"test": [
(montgomery_split["test"], montgomery_loader),
diff --git a/src/ptbench/data/nih_cxr14/datamodule.py b/src/ptbench/data/nih_cxr14/datamodule.py
index 1594be18..58c828b4 100644
--- a/src/ptbench/data/nih_cxr14/datamodule.py
+++ b/src/ptbench/data/nih_cxr14/datamodule.py
@@ -18,26 +18,22 @@ from ..typing import Sample
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the NIH CXR-14 dataset.
+ """A specialized raw-data-loader for the NIH CXR-14 dataset."""
- Attributes
- ----------
+ datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
- datadir
- This variable contains the base directory where the database raw data
- is stored.
+ idiap_file_organisation: bool
+ """If should use the Idiap's filesystem organisation when looking up data.
- idiap_file_organisation
- This variable will be ``True``, if the user has set the configuration
- parameter ``nih_cxr14.idiap_file_organisation`` in the global
- configuration file. It will cause internal loader to search for files
- in a slightly different folder structure, that was adapted to Idiap's
- requirements (number of files per folder to be less than 10k).
+ This variable will be ``True``, if the user has set the configuration
+ parameter ``nih_cxr14.idiap_file_organisation`` in the global configuration
+ file. It will cause internal loader to search for files in a slightly
+ different folder structure, that was adapted to Idiap's requirements
+ (number of files per folder to be less than 10k).
"""
- datadir: str
- idiap_file_organisation: bool
-
def __init__(self):
rc = load_rc()
self.datadir = rc.get("datadir.nih_cxr14", os.path.realpath(os.curdir))
@@ -50,8 +46,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -59,8 +54,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
file_path = sample[0] # default
@@ -91,8 +84,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -100,8 +92,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- labels
The integer labels associated with the sample
"""
return sample[1]
diff --git a/src/ptbench/data/nih_cxr14/default.py b/src/ptbench/data/nih_cxr14/default.py
index b9f7b6c5..af4f2172 100644
--- a/src/ptbench/data/nih_cxr14/default.py
+++ b/src/ptbench/data/nih_cxr14/default.py
@@ -11,5 +11,5 @@ datamodule = DataModule("default.json.bz2")
* Validation samples: 6350
* Test samples: 4054
-See :py:class:`DataModule` for technical details.
+See :py:class:`.nih_cxr14.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/nih_cxr14_padchest/datamodule.py b/src/ptbench/data/nih_cxr14_padchest/datamodule.py
index 335679bf..f1ce1f91 100644
--- a/src/ptbench/data/nih_cxr14_padchest/datamodule.py
+++ b/src/ptbench/data/nih_cxr14_padchest/datamodule.py
@@ -31,7 +31,8 @@ class DataModule(ConcatDataModule):
],
"test": [
(cxr14_split["test"], cxr14_loader),
- (padchest_split["test"], padchest_loader),
+ # there is no test set on padchest
+ # (padchest_split["test"], padchest_loader),
],
}
)
diff --git a/src/ptbench/data/padchest/datamodule.py b/src/ptbench/data/padchest/datamodule.py
index a09c5e67..69a37ea8 100644
--- a/src/ptbench/data/padchest/datamodule.py
+++ b/src/ptbench/data/padchest/datamodule.py
@@ -20,17 +20,11 @@ from ..typing import Sample
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the PadChest dataset.
-
- Attributes
- ----------
-
- datadir
- This variable contains the base directory where the database raw data
- is stored.
- """
+ """A specialized raw-data-loader for the PadChest dataset."""
datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
def __init__(self):
rc = load_rc()
@@ -41,8 +35,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -50,8 +43,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
# N.B.: PadChest images are encoded as 16-bit grayscale images
@@ -73,8 +64,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -82,8 +72,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- labels
The integer labels associated with the sample
"""
return sample[1]
@@ -117,213 +105,213 @@ class DataModule(CachingDataModule):
* Labels: [PADCHEST-2019]_
* Output image:
- * Transforms:
-
- * Load raw 16-bit PNG with :py:mod:`PIL`
- * Remove excess black borders
- * Convert image to 32-bit floats between 0. and 1.
- * Convert to tensor
- * Center crop, preserving the minimum dimension (height or width)
-
- * Final specifications
-
- * Grayscale, encoded as a 1-plane 32-bit float image, square with
- varying resolutions depending on the raw input image
- * Labels, in order (some of which may not be present in all splits):
-
- * COPD signs
- * Chilaiditi sign
- * NSG tube
- * abnormal foreign body
- * abscess
- * adenopathy
- * air bronchogram
- * air fluid level
- * air trapping
- * alveolar pattern
- * aortic aneurysm
- * aortic atheromatosis
- * aortic button enlargement
- * aortic elongation
- * aortic endoprosthesis
- * apical pleural thickening
- * artificial aortic heart valve
- * artificial heart valve
- * artificial mitral heart valve
- * asbestosis signs
- * ascendent aortic elongation
- * atelectasis
- * atelectasis basal
- * atypical pneumonia
- * axial hyperostosis
- * azygoesophageal recess shift
- * azygos lobe
- * blastic bone lesion
- * bone cement
- * bone metastasis
- * breast mass
- * bronchiectasis
- * bronchovascular markings
- * bullas
- * calcified adenopathy
- * calcified densities
- * calcified fibroadenoma
- * calcified granuloma
- * calcified mediastinal adenopathy
- * calcified pleural plaques
- * calcified pleural thickening
- * callus rib fracture
- * cardiomegaly
- * catheter
- * cavitation
- * central vascular redistribution
- * central venous catheter
- * central venous catheter via jugular vein
- * central venous catheter via subclavian vein
- * central venous catheter via umbilical vein
- * cervical rib
- * chest drain tube
- * chronic changes
- * clavicle fracture
- * consolidation
- * costochondral junction hypertrophy
- * costophrenic angle blunting
- * cyst
- * dai
- * descendent aortic elongation
- * dextrocardia
- * diaphragmatic eventration
- * double J stent
- * dual chamber device
- * electrical device
- * emphysema
- * empyema
- * end on vessel
- * endoprosthesis
- * endotracheal tube
- * esophagic dilatation
- * exclude
- * external foreign body
- * fibrotic band
- * fissure thickening
- * flattened diaphragm
- * fracture
- * gastrostomy tube
- * goiter
- * granuloma
- * ground glass pattern
- * gynecomastia
- * heart insufficiency
- * heart valve calcified
- * hemidiaphragm elevation
- * hiatal hernia
- * hilar congestion
- * hilar enlargement
- * humeral fracture
- * humeral prosthesis
- * hydropneumothorax
- * hyperinflated lung
- * hypoexpansion
- * hypoexpansion basal
- * increased density
- * infiltrates
- * interstitial pattern
- * kerley lines
- * kyphosis
- * laminar atelectasis
- * lepidic adenocarcinoma
- * lipomatosis
- * lobar atelectasis
- * loculated fissural effusion
- * loculated pleural effusion
- * lung metastasis
- * lung vascular paucity
- * lymphangitis carcinomatosa
- * lytic bone lesion
- * major fissure thickening
- * mammary prosthesis
- * mass
- * mastectomy
- * mediastinal enlargement
- * mediastinal mass
- * mediastinal shift
- * mediastinic lipomatosis
- * metal
- * miliary opacities
- * minor fissure thickening
- * multiple nodules
- * nephrostomy tube
- * nipple shadow
- * nodule
- * non axial articular degenerative changes
- * normal
- * obesity
- * osteopenia
- * osteoporosis
- * osteosynthesis material
- * pacemaker
- * pectum carinatum
- * pectum excavatum
- * pericardial effusion
- * pleural effusion
- * pleural mass
- * pleural plaques
- * pleural thickening
- * pneumomediastinum
- * pneumonia
- * pneumoperitoneo
- * pneumothorax
- * post radiotherapy changes
- * prosthesis
- * pseudonodule
- * pulmonary artery enlargement
- * pulmonary artery hypertension
- * pulmonary edema
- * pulmonary fibrosis
- * pulmonary hypertension
- * pulmonary mass
- * pulmonary venous hypertension
- * reservoir central venous catheter
- * respiratory distress
- * reticular interstitial pattern
- * reticulonodular interstitial pattern
- * rib fracture
- * right sided aortic arch
- * round atelectasis
- * sclerotic bone lesion
- * scoliosis
- * segmental atelectasis
- * single chamber device
- * soft tissue mass
- * sternoclavicular junction hypertrophy
- * sternotomy
- * subacromial space narrowing
- * subcutaneous emphysema
- * suboptimal study
- * superior mediastinal enlargement
- * supra aortic elongation
- * surgery
- * surgery breast
- * surgery heart
- * surgery humeral
- * surgery lung
- * surgery neck
- * suture material
- * thoracic cage deformation
- * total atelectasis
- * tracheal shift
- * tracheostomy tube
- * tuberculosis
- * tuberculosis sequelae
- * unchanged
- * vascular hilar enlargement
- * vascular redistribution
- * ventriculoperitoneal drain tube
- * vertebral anterior compression
- * vertebral compression
- * vertebral degenerative changes
- * vertebral fracture
- * volume loss
+ * Transforms:
+
+ * Load raw 16-bit PNG with :py:mod:`PIL`
+ * Remove excess black borders
+ * Convert image to 32-bit floats between 0. and 1.
+ * Convert to tensor
+ * Center crop, preserving the minimum dimension (height or width)
+
+ * Final specifications
+
+ * Grayscale, encoded as a 1-plane 32-bit float image, square with
+ varying resolutions depending on the raw input image
+ * Labels, in order (some of which may not be present in all splits):
+
+ * COPD signs
+ * Chilaiditi sign
+ * NSG tube
+ * abnormal foreign body
+ * abscess
+ * adenopathy
+ * air bronchogram
+ * air fluid level
+ * air trapping
+ * alveolar pattern
+ * aortic aneurysm
+ * aortic atheromatosis
+ * aortic button enlargement
+ * aortic elongation
+ * aortic endoprosthesis
+ * apical pleural thickening
+ * artificial aortic heart valve
+ * artificial heart valve
+ * artificial mitral heart valve
+ * asbestosis signs
+ * ascendent aortic elongation
+ * atelectasis
+ * atelectasis basal
+ * atypical pneumonia
+ * axial hyperostosis
+ * azygoesophageal recess shift
+ * azygos lobe
+ * blastic bone lesion
+ * bone cement
+ * bone metastasis
+ * breast mass
+ * bronchiectasis
+ * bronchovascular markings
+ * bullas
+ * calcified adenopathy
+ * calcified densities
+ * calcified fibroadenoma
+ * calcified granuloma
+ * calcified mediastinal adenopathy
+ * calcified pleural plaques
+ * calcified pleural thickening
+ * callus rib fracture
+ * cardiomegaly
+ * catheter
+ * cavitation
+ * central vascular redistribution
+ * central venous catheter
+ * central venous catheter via jugular vein
+ * central venous catheter via subclavian vein
+ * central venous catheter via umbilical vein
+ * cervical rib
+ * chest drain tube
+ * chronic changes
+ * clavicle fracture
+ * consolidation
+ * costochondral junction hypertrophy
+ * costophrenic angle blunting
+ * cyst
+ * dai
+ * descendent aortic elongation
+ * dextrocardia
+ * diaphragmatic eventration
+ * double J stent
+ * dual chamber device
+ * electrical device
+ * emphysema
+ * empyema
+ * end on vessel
+ * endoprosthesis
+ * endotracheal tube
+ * esophagic dilatation
+ * exclude
+ * external foreign body
+ * fibrotic band
+ * fissure thickening
+ * flattened diaphragm
+ * fracture
+ * gastrostomy tube
+ * goiter
+ * granuloma
+ * ground glass pattern
+ * gynecomastia
+ * heart insufficiency
+ * heart valve calcified
+ * hemidiaphragm elevation
+ * hiatal hernia
+ * hilar congestion
+ * hilar enlargement
+ * humeral fracture
+ * humeral prosthesis
+ * hydropneumothorax
+ * hyperinflated lung
+ * hypoexpansion
+ * hypoexpansion basal
+ * increased density
+ * infiltrates
+ * interstitial pattern
+ * kerley lines
+ * kyphosis
+ * laminar atelectasis
+ * lepidic adenocarcinoma
+ * lipomatosis
+ * lobar atelectasis
+ * loculated fissural effusion
+ * loculated pleural effusion
+ * lung metastasis
+ * lung vascular paucity
+ * lymphangitis carcinomatosa
+ * lytic bone lesion
+ * major fissure thickening
+ * mammary prosthesis
+ * mass
+ * mastectomy
+ * mediastinal enlargement
+ * mediastinal mass
+ * mediastinal shift
+ * mediastinic lipomatosis
+ * metal
+ * miliary opacities
+ * minor fissure thickening
+ * multiple nodules
+ * nephrostomy tube
+ * nipple shadow
+ * nodule
+ * non axial articular degenerative changes
+ * normal
+ * obesity
+ * osteopenia
+ * osteoporosis
+ * osteosynthesis material
+ * pacemaker
+ * pectum carinatum
+ * pectum excavatum
+ * pericardial effusion
+ * pleural effusion
+ * pleural mass
+ * pleural plaques
+ * pleural thickening
+ * pneumomediastinum
+ * pneumonia
+ * pneumoperitoneo
+ * pneumothorax
+ * post radiotherapy changes
+ * prosthesis
+ * pseudonodule
+ * pulmonary artery enlargement
+ * pulmonary artery hypertension
+ * pulmonary edema
+ * pulmonary fibrosis
+ * pulmonary hypertension
+ * pulmonary mass
+ * pulmonary venous hypertension
+ * reservoir central venous catheter
+ * respiratory distress
+ * reticular interstitial pattern
+ * reticulonodular interstitial pattern
+ * rib fracture
+ * right sided aortic arch
+ * round atelectasis
+ * sclerotic bone lesion
+ * scoliosis
+ * segmental atelectasis
+ * single chamber device
+ * soft tissue mass
+ * sternoclavicular junction hypertrophy
+ * sternotomy
+ * subacromial space narrowing
+ * subcutaneous emphysema
+ * suboptimal study
+ * superior mediastinal enlargement
+ * supra aortic elongation
+ * surgery
+ * surgery breast
+ * surgery heart
+ * surgery humeral
+ * surgery lung
+ * surgery neck
+ * suture material
+ * thoracic cage deformation
+ * total atelectasis
+ * tracheal shift
+ * tracheostomy tube
+ * tuberculosis
+ * tuberculosis sequelae
+ * unchanged
+ * vascular hilar enlargement
+ * vascular redistribution
+ * ventriculoperitoneal drain tube
+ * vertebral anterior compression
+ * vertebral compression
+ * vertebral degenerative changes
+ * vertebral fracture
+ * volume loss
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/padchest/idiap.py b/src/ptbench/data/padchest/idiap.py
index ea8b3dbd..52c52123 100644
--- a/src/ptbench/data/padchest/idiap.py
+++ b/src/ptbench/data/padchest/idiap.py
@@ -8,7 +8,7 @@ datamodule = DataModule("idiap.json.bz2")
"""Padchest dataset for computer-aided diagnosis (``idiap`` split).
This split contains all images in the database. Read documentation of
-:py:class:`DataModule` for technical details.
+:py:class:`.padchest.datamodule.DataModule` for technical details.
* Split reference: ours
* Training samples: 96'269
diff --git a/src/ptbench/data/shenzhen/datamodule.py b/src/ptbench/data/shenzhen/datamodule.py
index 0596007e..221bc869 100644
--- a/src/ptbench/data/shenzhen/datamodule.py
+++ b/src/ptbench/data/shenzhen/datamodule.py
@@ -19,20 +19,11 @@ from ..typing import Sample
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the Shenzen dataset.
-
- Attributes
- ----------
-
- datadir
- This variable contains the base directory where the database raw data
- is stored.
-
- transform
- Transforms that are always applied to the loaded raw images.
- """
+ """A specialized raw-data-loader for the Shenzhen dataset."""
datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
def __init__(self, config_variable: str = "datadir.shenzhen"):
self.datadir = load_rc().get(
@@ -44,8 +35,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -53,8 +43,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
# N.B.: Image.convert("L") is required to normalize grayscale back to
@@ -78,8 +66,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -87,8 +74,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- label
The integer label associated with the sample
"""
return sample[1]
@@ -117,23 +102,23 @@ class DataModule(CachingDataModule):
* Raw data input (on disk):
- * PNG 8-bit RGB images (grayscale, but encoded as RGB images with
- "inverted" grayscale scale requiring special treatment).
- * Variable width and height of 3000 x 3000 pixels or less
+ * PNG 8-bit RGB images (grayscale, but encoded as RGB images with
+ "inverted" grayscale scale requiring special treatment).
+ * Variable width and height of 3000 x 3000 pixels or less
* Output image:
- * Transforms:
+ * Transforms:
- * Load raw PNG with :py:mod:`PIL`
- * Remove black borders
- * Torch center cropping to get square image
+ * Load raw PNG with :py:mod:`PIL`
+ * Remove black borders
+ * Torch center cropping to get square image
- * Final specifications:
+ * Final specifications:
- * Grayscale, encoded as a single plane tensor, 32-bit floats,
- square with varying resolutions, depending on the input image
- * Labels: 0 (healthy), 1 (active tuberculosis)
+ * Grayscale, encoded as a single plane tensor, 32-bit floats,
+ square with varying resolutions, depending on the input image
+ * Labels: 0 (healthy), 1 (active tuberculosis)
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/shenzhen/default.py b/src/ptbench/data/shenzhen/default.py
index 93517c03..0a73847a 100644
--- a/src/ptbench/data/shenzhen/default.py
+++ b/src/ptbench/data/shenzhen/default.py
@@ -5,9 +5,11 @@
from .datamodule import DataModule
datamodule = DataModule("default.json")
-"""Default Shenzen TB database split.
+"""Default Shenzhen TB database split.
* Training samples: 64% of TB and healthy CXR (including labels)
* Validation samples: 16% of TB and healthy CXR (including labels)
* Test samples: 20% of TB and healthy CXR (including labels)
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/shenzhen/fold_0.py b/src/ptbench/data/shenzhen/fold_0.py
index 91711d38..dc895f78 100644
--- a/src/ptbench/data/shenzhen/fold_0.py
+++ b/src/ptbench/data/shenzhen/fold_0.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 0).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_1.py b/src/ptbench/data/shenzhen/fold_1.py
index 17372695..ce062462 100644
--- a/src/ptbench/data/shenzhen/fold_1.py
+++ b/src/ptbench/data/shenzhen/fold_1.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 1).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_2.py b/src/ptbench/data/shenzhen/fold_2.py
index 91346efa..3f061c37 100644
--- a/src/ptbench/data/shenzhen/fold_2.py
+++ b/src/ptbench/data/shenzhen/fold_2.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 2).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_3.py b/src/ptbench/data/shenzhen/fold_3.py
index 4156c3ee..e3bb067e 100644
--- a/src/ptbench/data/shenzhen/fold_3.py
+++ b/src/ptbench/data/shenzhen/fold_3.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 3).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_4.py b/src/ptbench/data/shenzhen/fold_4.py
index 6d5b73d1..f0cb843b 100644
--- a/src/ptbench/data/shenzhen/fold_4.py
+++ b/src/ptbench/data/shenzhen/fold_4.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 4).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_5.py b/src/ptbench/data/shenzhen/fold_5.py
index 27a7442d..6a27ac51 100644
--- a/src/ptbench/data/shenzhen/fold_5.py
+++ b/src/ptbench/data/shenzhen/fold_5.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 5).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_6.py b/src/ptbench/data/shenzhen/fold_6.py
index 1270cd58..302d2b8d 100644
--- a/src/ptbench/data/shenzhen/fold_6.py
+++ b/src/ptbench/data/shenzhen/fold_6.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 6).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_7.py b/src/ptbench/data/shenzhen/fold_7.py
index e7b5a1c3..a07f4d29 100644
--- a/src/ptbench/data/shenzhen/fold_7.py
+++ b/src/ptbench/data/shenzhen/fold_7.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 7).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_8.py b/src/ptbench/data/shenzhen/fold_8.py
index 1abbe7b2..0c5e6d22 100644
--- a/src/ptbench/data/shenzhen/fold_8.py
+++ b/src/ptbench/data/shenzhen/fold_8.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 8).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/shenzhen/fold_9.py b/src/ptbench/data/shenzhen/fold_9.py
index 71de470d..bb73ea16 100644
--- a/src/ptbench/data/shenzhen/fold_9.py
+++ b/src/ptbench/data/shenzhen/fold_9.py
@@ -5,3 +5,7 @@
from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
+"""Shenzhen datamodule for computer-aided diagnosis (cross validation fold 9).
+
+See :py:class:`.shenzhen.datamodule.DataModule` for technical details.
+"""
diff --git a/src/ptbench/data/tbpoc/datamodule.py b/src/ptbench/data/tbpoc/datamodule.py
index 31e2aac4..f564e232 100644
--- a/src/ptbench/data/tbpoc/datamodule.py
+++ b/src/ptbench/data/tbpoc/datamodule.py
@@ -19,20 +19,11 @@ from ..typing import Sample
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the Shenzen dataset.
-
- Attributes
- ----------
-
- datadir
- This variable contains the base directory where the database raw data
- is stored.
-
- transform
- Transforms that are always applied to the loaded raw images.
- """
+ """A specialized raw-data-loader for the Shenzen dataset."""
datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
def __init__(self, config_variable: str = "datadir.tbpoc"):
self.datadir = load_rc().get(
@@ -44,8 +35,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -53,8 +43,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
# images from TBPOC are encoded as grayscale JPEGs, no need to
@@ -76,8 +64,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, and an integer, representing the
sample label.
@@ -85,8 +72,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- label
The integer label associated with the sample
"""
return sample[1]
@@ -111,24 +96,24 @@ class DataModule(CachingDataModule):
* Raw data input (on disk):
- * JPEG 8-bit Grayscale images
- * resolution: fixed to one of the cases above
+ * JPEG 8-bit Grayscale images
+ * resolution: fixed to one of the cases above
* Output image:
- * Transforms:
+ * Transforms:
- * Load raw grayscale jpeg with :py:mod:`PIL`
- * Remove black borders
- * Convert to torch tensor
- * Torch center cropping to get square image
+ * Load raw grayscale jpeg with :py:mod:`PIL`
+ * Remove black borders
+ * Convert to torch tensor
+ * Torch center cropping to get square image
- * Final specifications:
+ * Final specifications:
- * Grayscale, encoded as a single plane tensor, 32-bit floats,
- square with varying resolutions (2048 x 2048 being the maximum),
- but also depending on black borders' sizes on the input image.
- * Labels: 0 (healthy), 1 (active tuberculosis)
+ * Grayscale, encoded as a single plane tensor, 32-bit floats,
+ square with varying resolutions (2048 x 2048 being the maximum),
+ but also depending on black borders' sizes on the input image.
+ * Labels: 0 (healthy), 1 (active tuberculosis)
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/tbpoc/fold_0.py b/src/ptbench/data/tbpoc/fold_0.py
index 775f64cf..2beb07fd 100644
--- a/src/ptbench/data/tbpoc/fold_0.py
+++ b/src/ptbench/data/tbpoc/fold_0.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-0.json")
"""TB-POC dataset for TB detection (cross validation fold 0).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_1.py b/src/ptbench/data/tbpoc/fold_1.py
index 6f0f137f..338d99b2 100644
--- a/src/ptbench/data/tbpoc/fold_1.py
+++ b/src/ptbench/data/tbpoc/fold_1.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-1.json")
"""TB-POC dataset for TB detection (cross validation fold 1).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_2.py b/src/ptbench/data/tbpoc/fold_2.py
index 662fd32c..9df72b54 100644
--- a/src/ptbench/data/tbpoc/fold_2.py
+++ b/src/ptbench/data/tbpoc/fold_2.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-2.json")
"""TB-POC dataset for TB detection (cross validation fold 2).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_3.py b/src/ptbench/data/tbpoc/fold_3.py
index c52b8c2e..514bf12c 100644
--- a/src/ptbench/data/tbpoc/fold_3.py
+++ b/src/ptbench/data/tbpoc/fold_3.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-3.json")
"""TB-POC dataset for TB detection (cross validation fold 3).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_4.py b/src/ptbench/data/tbpoc/fold_4.py
index 6de0dc13..d4f87280 100644
--- a/src/ptbench/data/tbpoc/fold_4.py
+++ b/src/ptbench/data/tbpoc/fold_4.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-4.json")
"""TB-POC dataset for TB detection (cross validation fold 4).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_5.py b/src/ptbench/data/tbpoc/fold_5.py
index bdca5a36..2df9a7ff 100644
--- a/src/ptbench/data/tbpoc/fold_5.py
+++ b/src/ptbench/data/tbpoc/fold_5.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-5.json")
"""TB-POC dataset for TB detection (cross validation fold 5).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_6.py b/src/ptbench/data/tbpoc/fold_6.py
index c17ba0ba..5d4fd08a 100644
--- a/src/ptbench/data/tbpoc/fold_6.py
+++ b/src/ptbench/data/tbpoc/fold_6.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-6.json")
"""TB-POC dataset for TB detection (cross validation fold 6).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_7.py b/src/ptbench/data/tbpoc/fold_7.py
index 4310f2f4..3b0b137f 100644
--- a/src/ptbench/data/tbpoc/fold_7.py
+++ b/src/ptbench/data/tbpoc/fold_7.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-7.json")
"""TB-POC dataset for TB detection (cross validation fold 7).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_8.py b/src/ptbench/data/tbpoc/fold_8.py
index d7fa5d10..f0304467 100644
--- a/src/ptbench/data/tbpoc/fold_8.py
+++ b/src/ptbench/data/tbpoc/fold_8.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-8.json")
"""TB-POC dataset for TB detection (cross validation fold 8).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbpoc/fold_9.py b/src/ptbench/data/tbpoc/fold_9.py
index f37e1f36..327c7156 100644
--- a/src/ptbench/data/tbpoc/fold_9.py
+++ b/src/ptbench/data/tbpoc/fold_9.py
@@ -7,5 +7,5 @@ from .datamodule import DataModule
datamodule = DataModule("fold-9.json")
"""TB-POC dataset for TB detection (cross validation fold 9).
-See :py:class:`DataModule` for technical details.
+See :py:class:`.tbpoc.datamodule.DataModule` for technical details.
"""
diff --git a/src/ptbench/data/tbx11k/datamodule.py b/src/ptbench/data/tbx11k/datamodule.py
index 45785ddb..45cab752 100644
--- a/src/ptbench/data/tbx11k/datamodule.py
+++ b/src/ptbench/data/tbx11k/datamodule.py
@@ -47,17 +47,11 @@ finding locations, as described above.
class RawDataLoader(_BaseRawDataLoader):
- """A specialized raw-data-loader for the TBX11k dataset.
-
- Attributes
- ----------
-
- datadir
- This variable contains the base directory where the database raw data
- is stored.
- """
+ """A specialized raw-data-loader for the TBX11k dataset."""
datadir: str
+ """This variable contains the base directory where the database raw data is
+ stored."""
def __init__(self):
self.datadir = load_rc().get(
@@ -69,8 +63,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, an integer, representing the
sample label, and possible radiological findings represented by
@@ -79,8 +72,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- sample
The sample representation
"""
image = PIL.Image.open(os.path.join(self.datadir, sample[0]))
@@ -98,8 +89,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, an integer, representing the
sample label, and possible radiological findings represented by
@@ -108,8 +98,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- label
The integer label associated with the sample
"""
return sample[1]
@@ -121,8 +109,7 @@ class RawDataLoader(_BaseRawDataLoader):
Parameters
----------
-
- sample:
+ sample
A tuple containing the path suffix, within the dataset root folder,
where to find the image to be loaded, an integer, representing the
sample label, and possible radiological findings represented by
@@ -131,8 +118,6 @@ class RawDataLoader(_BaseRawDataLoader):
Returns
-------
-
- annotations
Bounding box annotations, if any available with the sample.
"""
return sample[2] if len(sample) > 2 else [] # type: ignore
@@ -194,91 +179,55 @@ class DataModule(CachingDataModule):
- imgs/tb/tb1104.png
- imgs/tb/tb1143.png
- - Original train dataset samples:
- - Healthy: 3000
- - Sick (but no TB): 3000
- - Active TB only: 473
- - Latent TB only: 103
- - Both active and latent TB: 23
- - Unknown: 1
- - Total: 6600
-
- - Original validation dataset samples:
- - Healthy: 800
- - Sick (but no TB): 800
- - Latent TB only: 36
- - Active TB only: 157
- - Both active and latent TB: 7
- - Total: 1800
-
- - Original test dataset samples:
- - Unknown: 3302
- - Total: 3302
-
- * Because the test set does not have annotations, we generate train,
- validation and test datasets as such:
+ * Original train dataset samples:
- - The original validation dataset becomes our test set.
- - The original train dataset is split into new train and validation
- datasets (validation ratio = 0.203 w.r.t. original train dataset size).
- The selection of samples is stratified (see comments through our split
- code, which is shipped alongside this file.)
-
- * Split v1 contains healthy subjects against active TB cases (total samples
- = 4430):
-
- - ``train`` dataset samples:
- - Healthy: 2390
- - Active TB only: 377
- - Total: 2767
-
- - ``validation`` dataset samples:
- - Healthy: 610
- - Active TB only: 96
- - Total: 706
+ - Healthy: 3000
+ - Sick (but no TB): 3000
+ - Active TB only: 473
+ - Latent TB only: 103
+ - Both active and latent TB: 23
+ - Unknown: 1
+ - Total: 6600
- - ``test`` dataset samples:
- - Healthy: 800
- - Active TB only: 157
- - Total: 957
+ * Original validation dataset samples:
- * Split v2 contains healthy, sick (no TB), and latent TB subjects against
- active TB cases (total samples = 8369):
+ - Healthy: 800
+ - Sick (but no TB): 800
+ - Latent TB only: 36
+ - Active TB only: 157
+ - Both active and latent TB: 7
+ - Total: 1800
- - ``train`` dataset samples:
- - Healthy, Sick or Latent TB: 4864
- - Active TB only: 377
- - Total: 5241
+ * Original test dataset samples:
- - ``validation`` dataset samples:
- - Healthy, Sick or Latent TB: 1239
- - Active TB only: 96
- - Total: 1335
+ - Unknown: 3302
+ - Total: 3302
- - ``test`` dataset samples:
- - Healthy, Sick or Latent TB: 1636
- - Active TB only: 157
- - Total: 1793
+ * Because the test set does not have annotations, we generated train,
+ validation and test datasets as such:
+ - The original validation dataset becomes our test set.
+ - The original train dataset is split into new train and validation
+ datasets (validation ratio = 0.203 w.r.t. original train dataset size).
+ The selection of samples is stratified (see comments through our split
+ code, which is shipped alongside this file.)
Data specifications:
- * Raw data input (on disk):
-
- * PNG images 8 bits RGB, 512 x 512 pixels
+ * Raw data input (on disk): PNG images 8 bits RGB, 512 x 512 pixels
* Output image:
- * Transforms:
+ * Transforms:
- * Load raw PNG with :py:mod:`PIL`
+ - Load raw PNG with :py:mod:`PIL`
- * Final specifications:
+ * Final specifications:
- * RGB, encoded as a 3-plane tensor using 32-bit floats, square
- (512x512 pixels)
- * Labels: 0 (healthy, latent tb or sick but no tb depending on the
- protocol), 1 (active tuberculosis)
+ - RGB, encoded as a 3-plane tensor using 32-bit floats, square
+ (512x512 pixels)
+ - Labels: 0 (healthy, latent tb or sick but no tb depending on the
+ protocol), 1 (active tuberculosis)
"""
def __init__(self, split_filename: str):
diff --git a/src/ptbench/data/tbx11k/v1_healthy_vs_atb.py b/src/ptbench/data/tbx11k/v1_healthy_vs_atb.py
index 9a5cf8e3..fe0de82a 100644
--- a/src/ptbench/data/tbx11k/v1_healthy_vs_atb.py
+++ b/src/ptbench/data/tbx11k/v1_healthy_vs_atb.py
@@ -5,3 +5,27 @@
from .datamodule import DataModule
datamodule = DataModule("v1-healthy-vs-atb.json")
+"""TBX11k dataset for TB detection. Split ``v1`` (healthy against active TB
+cases).
+
+Split v1 contains healthy subjects against active TB cases (total samples =
+4430):
+
+* ``train`` dataset samples:
+
+ - Healthy: 2390
+ - Active TB only: 377
+ - Total: 2767
+
+* ``validation`` dataset samples:
+
+ - Healthy: 610
+ - Active TB only: 96
+ - Total: 706
+
+* ``test`` dataset samples:
+
+ - Healthy: 800
+ - Active TB only: 157
+ - Total: 957
+"""
diff --git a/src/ptbench/data/tbx11k/v2_others_vs_atb.py b/src/ptbench/data/tbx11k/v2_others_vs_atb.py
index ccb98222..5bee7bd0 100644
--- a/src/ptbench/data/tbx11k/v2_others_vs_atb.py
+++ b/src/ptbench/data/tbx11k/v2_others_vs_atb.py
@@ -5,3 +5,27 @@
from .datamodule import DataModule
datamodule = DataModule("v2-others-vs-atb.json")
+"""TBX11k dataset for TB detection. Split ``v1`` (everything else against
+active TB cases).
+
+Split v2 contains healthy, sick (no TB), and latent TB subjects against
+active TB cases (total samples = 8369):
+
+* ``train`` dataset samples:
+
+ - Healthy, Sick or Latent TB: 4864
+ - Active TB only: 377
+ - Total: 5241
+
+* ``validation`` dataset samples:
+
+ - Healthy, Sick or Latent TB: 1239
+ - Active TB only: 96
+ - Total: 1335
+
+* ``test`` dataset samples:
+
+ - Healthy, Sick or Latent TB: 1636
+ - Active TB only: 157
+ - Total: 1793
+"""
diff --git a/src/ptbench/data/typing.py b/src/ptbench/data/typing.py
index 6f41b39e..c1df54c6 100644
--- a/src/ptbench/data/typing.py
+++ b/src/ptbench/data/typing.py
@@ -52,9 +52,9 @@ DatabaseSplit: typing.TypeAlias = collections.abc.Mapping[
]
"""The definition of a database split.
-A database split maps dataset (subset) names to sequences of objects
-that, through :py:class:`RawDataLoader`s, eventually become
-:py:class:`Sample`s in the processing pipeline.
+A database split maps dataset (subset) names to sequences of objects that,
+through a :py:class:`RawDataLoader`, eventually becomes a :py:data:`.Sample` in
+the processing pipeline.
"""
ConcatDatabaseSplit: typing.TypeAlias = collections.abc.Mapping[
@@ -63,12 +63,11 @@ ConcatDatabaseSplit: typing.TypeAlias = collections.abc.Mapping[
]
"""The definition of a complex database split composed of several other splits.
-A database split maps dataset (subset) names to sequences of objects
-that, through :py:class:`RawDataLoader`s, eventually become
-:py:class:`Sample`s in the processing pipeline. Objects of this subtype
-allow the construction of complex splits composed of cannibalized parts
-of other splits. Each split may be assigned a different
-:py:class:`RawDataLoader`.
+A database split maps dataset (subset) names to sequences of objects that,
+through a :py:class:`.RawDataLoader`, eventually becomes a :py:data:`.Sample` in
+the processing pipeline. Objects of this subtype allow the construction of
+complex splits composed of cannibalized parts of other splits. Each split may
+be assigned a different :py:class:`.RawDataLoader`.
"""
diff --git a/src/ptbench/models/alexnet.py b/src/ptbench/models/alexnet.py
index 5daed2a3..9096f608 100644
--- a/src/ptbench/models/alexnet.py
+++ b/src/ptbench/models/alexnet.py
@@ -27,7 +27,6 @@ class Alexnet(pl.LightningModule):
Parameters
----------
-
train_loss
The loss to be used during the training.
@@ -124,8 +123,7 @@ class Alexnet(pl.LightningModule):
Parameters
----------
-
- checkpoint:
+ checkpoint
Loaded checkpoint
"""
checkpoint["normalizer"] = self.normalizer
@@ -136,8 +134,7 @@ class Alexnet(pl.LightningModule):
Parameters
----------
-
- checkpoint:
+ checkpoint
Loaded checkpoint
"""
logger.info("Restoring normalizer from checkpoint.")
@@ -151,8 +148,7 @@ class Alexnet(pl.LightningModule):
Parameters
----------
-
- dataloader: :py:class:`torch.utils.data.DataLoader`
+ dataloader
A torch Dataloader from which to compute the mean and std.
Will not be used if the model is pretrained.
"""
diff --git a/src/ptbench/models/densenet.py b/src/ptbench/models/densenet.py
index 0663b60b..97ebaf78 100644
--- a/src/ptbench/models/densenet.py
+++ b/src/ptbench/models/densenet.py
@@ -25,7 +25,6 @@ class Densenet(pl.LightningModule):
Parameters
----------
-
train_loss
The loss to be used during the training.
@@ -124,8 +123,7 @@ class Densenet(pl.LightningModule):
Parameters
----------
-
- checkpoint:
+ checkpoint
Loaded checkpoint
"""
checkpoint["normalizer"] = self.normalizer
@@ -136,8 +134,7 @@ class Densenet(pl.LightningModule):
Parameters
----------
-
- checkpoint:
+ checkpoint
Loaded checkpoint
"""
logger.info("Restoring normalizer from checkpoint.")
@@ -151,8 +148,7 @@ class Densenet(pl.LightningModule):
Parameters
----------
-
- dataloader: :py:class:`torch.utils.data.DataLoader`
+ dataloader
A torch Dataloader from which to compute the mean and std.
Will not be used if the model is pretrained.
"""
diff --git a/src/ptbench/models/normalizer.py b/src/ptbench/models/normalizer.py
index 147efc7e..576f21cc 100644
--- a/src/ptbench/models/normalizer.py
+++ b/src/ptbench/models/normalizer.py
@@ -1,7 +1,7 @@
# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
#
# SPDX-License-Identifier: GPL-3.0-or-later
-"""A network model that prefixes a subtract/divide step to any other module."""
+"""Functions to compute normalisation factors based on dataloaders."""
import torch
import torch.nn
diff --git a/src/ptbench/models/pasa.py b/src/ptbench/models/pasa.py
index 3f10757f..5d6e20b4 100644
--- a/src/ptbench/models/pasa.py
+++ b/src/ptbench/models/pasa.py
@@ -31,7 +31,6 @@ class Pasa(pl.LightningModule):
Parameters
----------
-
train_loss
The loss to be used during the training.
@@ -204,8 +203,7 @@ class Pasa(pl.LightningModule):
Parameters
----------
-
- checkpoint:
+ checkpoint
Loaded checkpoint
"""
checkpoint["normalizer"] = self.normalizer
@@ -216,8 +214,7 @@ class Pasa(pl.LightningModule):
Parameters
----------
-
- checkpoint:
+ checkpoint
Loaded checkpoint
"""
logger.info("Restoring normalizer from checkpoint.")
@@ -228,7 +225,6 @@ class Pasa(pl.LightningModule):
Parameters
----------
-
dataloader
A torch Dataloader from which to compute the mean and std
"""
diff --git a/src/ptbench/scripts/train_analysis.py b/src/ptbench/scripts/train_analysis.py
index 4061164f..6fd623f8 100644
--- a/src/ptbench/scripts/train_analysis.py
+++ b/src/ptbench/scripts/train_analysis.py
@@ -24,12 +24,14 @@ def create_figures(df: pandas.DataFrame) -> list[plt.figure]:
It is assumed that some metric names are of the form <metric>/<subset>.
All subsets for a metric will be displayed on the same figure.
+
Parameters
----------
df:
Pandas dataframe containing the data to plot.
+
Returns
-------
@@ -115,17 +117,7 @@ def train_analysis(
**_,
) -> None:
"""Creates a plot for each metric in the training logs and saves them in a
- pdf file.
-
- Parameters
- ----------
-
- logdir:
- Directory containing tensorboard event files.
-
- output_pdf:
- The pdf file in which to save the plots.
- """
+ pdf file."""
from matplotlib.backends.backend_pdf import PdfPages
diff --git a/src/ptbench/utils/checkpointer.py b/src/ptbench/utils/checkpointer.py
index 67190e2f..88cdfbb7 100644
--- a/src/ptbench/utils/checkpointer.py
+++ b/src/ptbench/utils/checkpointer.py
@@ -17,23 +17,25 @@ def get_checkpoint(
Can return the best or last checkpoint, or a checkpoint at a specific path.
Ensures the checkpoint exists, raising an error if it is not the case.
- If resume_from is None, checks the output directory if a checkpoint already exists and returns it.
- If no checkpoint is found, returns None.
+ If resume_from is ``None``, checks the output directory if a checkpoint
+ already exists and returns it. If no checkpoint is found, returns ``None``.
+
Parameters
----------
- output_folder:
+ output_folder
Directory in which checkpoints are stored.
- resume_from:
+ resume_from
Which model to get. Can be one of "best", "last", or a path to a checkpoint.
If None, gets the last checkpoint if it exists, otherwise returns None
+
Returns
-------
- checkpoint_file:
+ checkpoint_file
Path to the requested checkpoint or None.
"""
last_checkpoint_path = os.path.join(output_folder, "model_final_epoch.ckpt")
--
GitLab