Remove rgb datasets

pyproject.toml

@@ -114,7 +114,6 @@ indian_f8 = "ptbench.data.indian.datamodules:fold_8"
 indian_f9 = "ptbench.data.indian.datamodules:fold_9"
 # TBX11K simplified dataset split 1 (and cross-validation folds)
 tbx11k_simplified = "ptbench.data.tbx11k_simplified.default"
-tbx11k_simplified_rgb = "ptbench.data.tbx11k_simplified.rgb"
 tbx11k_simplified_f0 = "ptbench.data.tbx11k_simplified.fold_0"
 tbx11k_simplified_f1 = "ptbench.data.tbx11k_simplified.fold_1"
 tbx11k_simplified_f2 = "ptbench.data.tbx11k_simplified.fold_2"
@@ -125,19 +124,8 @@ tbx11k_simplified_f6 = "ptbench.data.tbx11k_simplified.fold_6"
 tbx11k_simplified_f7 = "ptbench.data.tbx11k_simplified.fold_7"
 tbx11k_simplified_f8 = "ptbench.data.tbx11k_simplified.fold_8"
 tbx11k_simplified_f9 = "ptbench.data.tbx11k_simplified.fold_9"
-tbx11k_simplified_f0_rgb = "ptbench.data.tbx11k_simplified.fold_0_rgb"
-tbx11k_simplified_f1_rgb = "ptbench.data.tbx11k_simplified.fold_1_rgb"
-tbx11k_simplified_f2_rgb = "ptbench.data.tbx11k_simplified.fold_2_rgb"
-tbx11k_simplified_f3_rgb = "ptbench.data.tbx11k_simplified.fold_3_rgb"
-tbx11k_simplified_f4_rgb = "ptbench.data.tbx11k_simplified.fold_4_rgb"
-tbx11k_simplified_f5_rgb = "ptbench.data.tbx11k_simplified.fold_5_rgb"
-tbx11k_simplified_f6_rgb = "ptbench.data.tbx11k_simplified.fold_6_rgb"
-tbx11k_simplified_f7_rgb = "ptbench.data.tbx11k_simplified.fold_7_rgb"
-tbx11k_simplified_f8_rgb = "ptbench.data.tbx11k_simplified.fold_8_rgb"
-tbx11k_simplified_f9_rgb = "ptbench.data.tbx11k_simplified.fold_9_rgb"
 # TBX11K simplified dataset split 2 (and cross-validation folds)
 tbx11k_simplified_v2 = "ptbench.data.tbx11k_simplified_v2.default"
-tbx11k_simplified_v2_rgb = "ptbench.data.tbx11k_simplified_v2.rgb"
 tbx11k_simplified_v2_f0 = "ptbench.data.tbx11k_simplified_v2.fold_0"
 tbx11k_simplified_v2_f1 = "ptbench.data.tbx11k_simplified_v2.fold_1"
 tbx11k_simplified_v2_f2 = "ptbench.data.tbx11k_simplified_v2.fold_2"
@@ -148,19 +136,8 @@ tbx11k_simplified_v2_f6 = "ptbench.data.tbx11k_simplified_v2.fold_6"
 tbx11k_simplified_v2_f7 = "ptbench.data.tbx11k_simplified_v2.fold_7"
 tbx11k_simplified_v2_f8 = "ptbench.data.tbx11k_simplified_v2.fold_8"
 tbx11k_simplified_v2_f9 = "ptbench.data.tbx11k_simplified_v2.fold_9"
-tbx11k_simplified_v2_f0_rgb = "ptbench.data.tbx11k_simplified_v2.fold_0_rgb"
-tbx11k_simplified_v2_f1_rgb = "ptbench.data.tbx11k_simplified_v2.fold_1_rgb"
-tbx11k_simplified_v2_f2_rgb = "ptbench.data.tbx11k_simplified_v2.fold_2_rgb"
-tbx11k_simplified_v2_f3_rgb = "ptbench.data.tbx11k_simplified_v2.fold_3_rgb"
-tbx11k_simplified_v2_f4_rgb = "ptbench.data.tbx11k_simplified_v2.fold_4_rgb"
-tbx11k_simplified_v2_f5_rgb = "ptbench.data.tbx11k_simplified_v2.fold_5_rgb"
-tbx11k_simplified_v2_f6_rgb = "ptbench.data.tbx11k_simplified_v2.fold_6_rgb"
-tbx11k_simplified_v2_f7_rgb = "ptbench.data.tbx11k_simplified_v2.fold_7_rgb"
-tbx11k_simplified_v2_f8_rgb = "ptbench.data.tbx11k_simplified_v2.fold_8_rgb"
-tbx11k_simplified_v2_f9_rgb = "ptbench.data.tbx11k_simplified_v2.fold_9_rgb"
 # montgomery-shenzhen aggregated dataset
 mc_ch = "ptbench.data.mc_ch.default"
-mc_ch_rgb = "ptbench.data.mc_ch.rgb"
 mc_ch_f0 = "ptbench.data.mc_ch.fold_0"
 mc_ch_f1 = "ptbench.data.mc_ch.fold_1"
 mc_ch_f2 = "ptbench.data.mc_ch.fold_2"
@@ -171,19 +148,8 @@ mc_ch_f6 = "ptbench.data.mc_ch.fold_6"
 mc_ch_f7 = "ptbench.data.mc_ch.fold_7"
 mc_ch_f8 = "ptbench.data.mc_ch.fold_8"
 mc_ch_f9 = "ptbench.data.mc_ch.fold_9"
-mc_ch_f0_rgb = "ptbench.data.mc_ch.fold_0_rgb"
-mc_ch_f1_rgb = "ptbench.data.mc_ch.fold_1_rgb"
-mc_ch_f2_rgb = "ptbench.data.mc_ch.fold_2_rgb"
-mc_ch_f3_rgb = "ptbench.data.mc_ch.fold_3_rgb"
-mc_ch_f4_rgb = "ptbench.data.mc_ch.fold_4_rgb"
-mc_ch_f5_rgb = "ptbench.data.mc_ch.fold_5_rgb"
-mc_ch_f6_rgb = "ptbench.data.mc_ch.fold_6_rgb"
-mc_ch_f7_rgb = "ptbench.data.mc_ch.fold_7_rgb"
-mc_ch_f8_rgb = "ptbench.data.mc_ch.fold_8_rgb"
-mc_ch_f9_rgb = "ptbench.data.mc_ch.fold_9_rgb"
 # montgomery-shenzhen-indian aggregated dataset
 mc_ch_in = "ptbench.data.mc_ch_in.default"
-mc_ch_in_rgb = "ptbench.data.mc_ch_in.rgb"
 mc_ch_in_f0 = "ptbench.data.mc_ch_in.fold_0"
 mc_ch_in_f1 = "ptbench.data.mc_ch_in.fold_1"
 mc_ch_in_f2 = "ptbench.data.mc_ch_in.fold_2"
@@ -194,19 +160,8 @@ mc_ch_in_f6 = "ptbench.data.mc_ch_in.fold_6"
 mc_ch_in_f7 = "ptbench.data.mc_ch_in.fold_7"
 mc_ch_in_f8 = "ptbench.data.mc_ch_in.fold_8"
 mc_ch_in_f9 = "ptbench.data.mc_ch_in.fold_9"
-mc_ch_in_f0_rgb = "ptbench.data.mc_ch_in.fold_0_rgb"
-mc_ch_in_f1_rgb = "ptbench.data.mc_ch_in.fold_1_rgb"
-mc_ch_in_f2_rgb = "ptbench.data.mc_ch_in.fold_2_rgb"
-mc_ch_in_f3_rgb = "ptbench.data.mc_ch_in.fold_3_rgb"
-mc_ch_in_f4_rgb = "ptbench.data.mc_ch_in.fold_4_rgb"
-mc_ch_in_f5_rgb = "ptbench.data.mc_ch_in.fold_5_rgb"
-mc_ch_in_f6_rgb = "ptbench.data.mc_ch_in.fold_6_rgb"
-mc_ch_in_f7_rgb = "ptbench.data.mc_ch_in.fold_7_rgb"
-mc_ch_in_f8_rgb = "ptbench.data.mc_ch_in.fold_8_rgb"
-mc_ch_in_f9_rgb = "ptbench.data.mc_ch_in.fold_9_rgb"
 # montgomery-shenzhen-indian-tbx11k aggregated dataset
 mc_ch_in_11k = "ptbench.data.mc_ch_in_11k.default"
-mc_ch_in_11k_rgb = "ptbench.data.mc_ch_in_11k.rgb"
 mc_ch_in_11k_f0 = "ptbench.data.mc_ch_in_11k.fold_0"
 mc_ch_in_11k_f1 = "ptbench.data.mc_ch_in_11k.fold_1"
 mc_ch_in_11k_f2 = "ptbench.data.mc_ch_in_11k.fold_2"
@@ -217,19 +172,8 @@ mc_ch_in_11k_f6 = "ptbench.data.mc_ch_in_11k.fold_6"
 mc_ch_in_11k_f7 = "ptbench.data.mc_ch_in_11k.fold_7"
 mc_ch_in_11k_f8 = "ptbench.data.mc_ch_in_11k.fold_8"
 mc_ch_in_11k_f9 = "ptbench.data.mc_ch_in_11k.fold_9"
-mc_ch_in_11k_f0_rgb = "ptbench.data.mc_ch_in_11k.fold_0_rgb"
-mc_ch_in_11k_f1_rgb = "ptbench.data.mc_ch_in_11k.fold_1_rgb"
-mc_ch_in_11k_f2_rgb = "ptbench.data.mc_ch_in_11k.fold_2_rgb"
-mc_ch_in_11k_f3_rgb = "ptbench.data.mc_ch_in_11k.fold_3_rgb"
-mc_ch_in_11k_f4_rgb = "ptbench.data.mc_ch_in_11k.fold_4_rgb"
-mc_ch_in_11k_f5_rgb = "ptbench.data.mc_ch_in_11k.fold_5_rgb"
-mc_ch_in_11k_f6_rgb = "ptbench.data.mc_ch_in_11k.fold_6_rgb"
-mc_ch_in_11k_f7_rgb = "ptbench.data.mc_ch_in_11k.fold_7_rgb"
-mc_ch_in_11k_f8_rgb = "ptbench.data.mc_ch_in_11k.fold_8_rgb"
-mc_ch_in_11k_f9_rgb = "ptbench.data.mc_ch_in_11k.fold_9_rgb"
 # montgomery-shenzhen-indian-tbx11kv2 aggregated dataset
 mc_ch_in_11kv2 = "ptbench.data.mc_ch_in_11kv2.default"
-mc_ch_in_11kv2_rgb = "ptbench.data.mc_ch_in_11kv2.rgb"
 mc_ch_in_11kv2_f0 = "ptbench.data.mc_ch_in_11kv2.fold_0"
 mc_ch_in_11kv2_f1 = "ptbench.data.mc_ch_in_11kv2.fold_1"
 mc_ch_in_11kv2_f2 = "ptbench.data.mc_ch_in_11kv2.fold_2"
@@ -240,16 +184,6 @@ mc_ch_in_11kv2_f6 = "ptbench.data.mc_ch_in_11kv2.fold_6"
 mc_ch_in_11kv2_f7 = "ptbench.data.mc_ch_in_11kv2.fold_7"
 mc_ch_in_11kv2_f8 = "ptbench.data.mc_ch_in_11kv2.fold_8"
 mc_ch_in_11kv2_f9 = "ptbench.data.mc_ch_in_11kv2.fold_9"
-mc_ch_in_11kv2_f0_rgb = "ptbench.data.mc_ch_in_11kv2.fold_0_rgb"
-mc_ch_in_11kv2_f1_rgb = "ptbench.data.mc_ch_in_11kv2.fold_1_rgb"
-mc_ch_in_11kv2_f2_rgb = "ptbench.data.mc_ch_in_11kv2.fold_2_rgb"
-mc_ch_in_11kv2_f3_rgb = "ptbench.data.mc_ch_in_11kv2.fold_3_rgb"
-mc_ch_in_11kv2_f4_rgb = "ptbench.data.mc_ch_in_11kv2.fold_4_rgb"
-mc_ch_in_11kv2_f5_rgb = "ptbench.data.mc_ch_in_11kv2.fold_5_rgb"
-mc_ch_in_11kv2_f6_rgb = "ptbench.data.mc_ch_in_11kv2.fold_6_rgb"
-mc_ch_in_11kv2_f7_rgb = "ptbench.data.mc_ch_in_11kv2.fold_7_rgb"
-mc_ch_in_11kv2_f8_rgb = "ptbench.data.mc_ch_in_11kv2.fold_8_rgb"
-mc_ch_in_11kv2_f9_rgb = "ptbench.data.mc_ch_in_11kv2.fold_9_rgb"
 # tbpoc dataset (and cross-validation folds)
 tbpoc_f0 = "ptbench.data.tbpoc.fold_0"
 tbpoc_f1 = "ptbench.data.tbpoc.fold_1"
@@ -261,16 +195,6 @@ tbpoc_f6 = "ptbench.data.tbpoc.fold_6"
 tbpoc_f7 = "ptbench.data.tbpoc.fold_7"
 tbpoc_f8 = "ptbench.data.tbpoc.fold_8"
 tbpoc_f9 = "ptbench.data.tbpoc.fold_9"
-tbpoc_f0_rgb = "ptbench.data.tbpoc.fold_0_rgb"
-tbpoc_f1_rgb = "ptbench.data.tbpoc.fold_1_rgb"
-tbpoc_f2_rgb = "ptbench.data.tbpoc.fold_2_rgb"
-tbpoc_f3_rgb = "ptbench.data.tbpoc.fold_3_rgb"
-tbpoc_f4_rgb = "ptbench.data.tbpoc.fold_4_rgb"
-tbpoc_f5_rgb = "ptbench.data.tbpoc.fold_5_rgb"
-tbpoc_f6_rgb = "ptbench.data.tbpoc.fold_6_rgb"
-tbpoc_f7_rgb = "ptbench.data.tbpoc.fold_7_rgb"
-tbpoc_f8_rgb = "ptbench.data.tbpoc.fold_8_rgb"
-tbpoc_f9_rgb = "ptbench.data.tbpoc.fold_9_rgb"
 # hivtb dataset (and cross-validation folds)
 hivtb_f0 = "ptbench.data.hivtb.fold_0"
 hivtb_f1 = "ptbench.data.hivtb.fold_1"
@@ -282,19 +206,8 @@ hivtb_f6 = "ptbench.data.hivtb.fold_6"
 hivtb_f7 = "ptbench.data.hivtb.fold_7"
 hivtb_f8 = "ptbench.data.hivtb.fold_8"
 hivtb_f9 = "ptbench.data.hivtb.fold_9"
-hivtb_f0_rgb = "ptbench.data.hivtb.fold_0_rgb"
-hivtb_f1_rgb = "ptbench.data.hivtb.fold_1_rgb"
-hivtb_f2_rgb = "ptbench.data.hivtb.fold_2_rgb"
-hivtb_f3_rgb = "ptbench.data.hivtb.fold_3_rgb"
-hivtb_f4_rgb = "ptbench.data.hivtb.fold_4_rgb"
-hivtb_f5_rgb = "ptbench.data.hivtb.fold_5_rgb"
-hivtb_f6_rgb = "ptbench.data.hivtb.fold_6_rgb"
-hivtb_f7_rgb = "ptbench.data.hivtb.fold_7_rgb"
-hivtb_f8_rgb = "ptbench.data.hivtb.fold_8_rgb"
-hivtb_f9_rgb = "ptbench.data.hivtb.fold_9_rgb"
 # montgomery-shenzhen-indian-padchest aggregated dataset
 mc_ch_in_pc = "ptbench.data.mc_ch_in_pc.default"
-mc_ch_in_pc_rgb = "ptbench.data.mc_ch_in_pc.rgb"
 # NIH CXR14 (relabeled)
 nih_cxr14 = "ptbench.data.nih_cxr14_re.default"
 nih_cxr14_cm = "ptbench.data.nih_cxr14_re.cardiomegaly"
@@ -304,7 +217,6 @@ nih_cxr14_pc_idiap = "ptbench.data.nih_cxr14_re_pc.idiap"
 padchest_idiap = "ptbench.data.padchest.idiap"
 padchest_tb_idiap = "ptbench.data.padchest.tb_idiap"
 padchest_no_tb_idiap = "ptbench.data.padchest.no_tb_idiap"
-padchest_tb_idiap_rgb = "ptbench.data.padchest.tb_idiap_rgb"
 padchest_cm_idiap = "ptbench.data.padchest.cardiomegaly_idiap"
 
 [tool.setuptools]

src/ptbench/data/indian/rgb.py

-# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Indian dataset for TB detection (default protocol, converted in RGB)
-
-* Split reference: [INDIAN-2013]_ with 20% of train set for the validation set
-* This configuration resolution: 512 x 512 (default)
-* See :py:mod:`ptbench.data.indian` for dataset details
-"""
-
-from clapper.logging import setup
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule
-from . import _maker
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        self.dataset = _maker("default", RGB=True)
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/mc_ch/rgb.py

-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Aggregated dataset composed of Montgomery and Shenzhen (RGB) datasets."""
-
-from clapper.logging import setup
-from torch.utils.data.dataset import ConcatDataset
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule, get_dataset_from_module
-from ..montgomery.rgb import datamodule as mc_datamodule
-from ..shenzhen.rgb import datamodule as ch_datamodule
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        self.train_batch_size = train_batch_size
-        self.predict_batch_size = predict_batch_size
-        self.drop_incomplete_batch = drop_incomplete_batch
-        self.multiproc_kwargs = multiproc_kwargs
-
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        # Instantiate other datamodules and get their datasets
-        module_args = {
-            "train_batch_size": self.train_batch_size,
-            "predict_batch_size": self.predict_batch_size,
-            "drop_incomplete_batch": self.drop_incomplete_batch,
-            "multiproc_kwargs": self.multiproc_kwargs,
-        }
-
-        mc = get_dataset_from_module(mc_datamodule, stage, **module_args)
-        ch = get_dataset_from_module(ch_datamodule, stage, **module_args)
-
-        # Combine datasets
-        self.dataset = {}
-        self.dataset["__train__"] = ConcatDataset(
-            [mc["__train__"], ch["__train__"]]
-        )
-        self.dataset["train"] = ConcatDataset([mc["train"], ch["train"]])
-        self.dataset["__valid__"] = ConcatDataset(
-            [mc["__valid__"], ch["__valid__"]]
-        )
-        self.dataset["validation"] = ConcatDataset(
-            [mc["validation"], ch["validation"]]
-        )
-        self.dataset["test"] = ConcatDataset([mc["test"], ch["test"]])
-
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/mc_ch_in/rgb.py

-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Aggregated dataset composed of Montgomery, Shenzhen and Indian (RGB)
-datasets."""
-
-from clapper.logging import setup
-from torch.utils.data.dataset import ConcatDataset
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule, get_dataset_from_module
-from ..indian.rgb import datamodule as indian_datamodule
-from ..montgomery.rgb import datamodule as mc_datamodule
-from ..shenzhen.rgb import datamodule as ch_datamodule
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        self.train_batch_size = train_batch_size
-        self.predict_batch_size = predict_batch_size
-        self.drop_incomplete_batch = drop_incomplete_batch
-        self.multiproc_kwargs = multiproc_kwargs
-
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        # Instantiate other datamodules and get their datasets
-
-        module_args = {
-            "train_batch_size": self.train_batch_size,
-            "predict_batch_size": self.predict_batch_size,
-            "drop_incomplete_batch": self.drop_incomplete_batch,
-            "multiproc_kwargs": self.multiproc_kwargs,
-        }
-
-        mc = get_dataset_from_module(mc_datamodule, stage, **module_args)
-        ch = get_dataset_from_module(ch_datamodule, stage, **module_args)
-        indian = get_dataset_from_module(
-            indian_datamodule, stage, **module_args
-        )
-
-        # Combine datasets
-        self.dataset = {}
-        self.dataset["__train__"] = ConcatDataset(
-            [mc["__train__"], ch["__train__"], indian["__train__"]]
-        )
-        self.dataset["train"] = ConcatDataset(
-            [mc["train"], ch["train"], indian["train"]]
-        )
-        self.dataset["__valid__"] = ConcatDataset(
-            [mc["__valid__"], ch["__valid__"], indian["__valid__"]]
-        )
-        self.dataset["validation"] = ConcatDataset(
-            [mc["validation"], ch["validation"], indian["validation"]]
-        )
-        self.dataset["test"] = ConcatDataset(
-            [mc["test"], ch["test"], indian["test"]]
-        )
-
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/mc_ch_in_11k/rgb.py

-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Aggregated dataset composed of Montgomery, Shenzhen, Indian and the default
-TBX11K-simplified datasets (RGB)"""
-
-from clapper.logging import setup
-from torch.utils.data.dataset import ConcatDataset
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule, get_dataset_from_module
-from ..indian.rgb import datamodule as indian_datamodule
-from ..montgomery.rgb import datamodule as mc_datamodule
-from ..shenzhen.rgb import datamodule as ch_datamodule
-from ..tbx11k_simplified.rgb import datamodule as tbx11k_datamodule
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        self.train_batch_size = train_batch_size
-        self.predict_batch_size = predict_batch_size
-        self.drop_incomplete_batch = drop_incomplete_batch
-        self.multiproc_kwargs = multiproc_kwargs
-
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        # Instantiate other datamodules and get their datasets
-
-        module_args = {
-            "train_batch_size": self.train_batch_size,
-            "predict_batch_size": self.predict_batch_size,
-            "drop_incomplete_batch": self.drop_incomplete_batch,
-            "multiproc_kwargs": self.multiproc_kwargs,
-        }
-
-        mc = get_dataset_from_module(mc_datamodule, stage, **module_args)
-        ch = get_dataset_from_module(ch_datamodule, stage, **module_args)
-        indian = get_dataset_from_module(
-            indian_datamodule, stage, **module_args
-        )
-        tbx11k = get_dataset_from_module(
-            tbx11k_datamodule, stage, **module_args
-        )
-
-        # Combine datasets
-        self.dataset = {}
-        self.dataset["__train__"] = ConcatDataset(
-            [
-                mc["__train__"],
-                ch["__train__"],
-                indian["__train__"],
-                tbx11k["__train__"],
-            ]
-        )
-        self.dataset["train"] = ConcatDataset(
-            [mc["train"], ch["train"], indian["train"], tbx11k["train"]]
-        )
-        self.dataset["__valid__"] = ConcatDataset(
-            [
-                mc["__valid__"],
-                ch["__valid__"],
-                indian["__valid__"],
-                tbx11k["__valid__"],
-            ]
-        )
-        self.dataset["validation"] = ConcatDataset(
-            [
-                mc["validation"],
-                ch["validation"],
-                indian["validation"],
-                tbx11k["validation"],
-            ]
-        )
-        self.dataset["test"] = ConcatDataset(
-            [mc["test"], ch["test"], indian["test"], tbx11k["test"]]
-        )
-
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/mc_ch_in_11kv2/rgb.py

-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Aggregated dataset composed of Montgomery, Shenzhen, Indian and the default
-TBX11K-simplified datasets (RGB)"""
-
-from clapper.logging import setup
-from torch.utils.data.dataset import ConcatDataset
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule, get_dataset_from_module
-from ..indian.rgb import datamodule as indian_datamodule
-from ..montgomery.rgb import datamodule as mc_datamodule
-from ..shenzhen.rgb import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.rgb import datamodule as tbx11kv2_datamodule
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        self.train_batch_size = train_batch_size
-        self.predict_batch_size = predict_batch_size
-        self.drop_incomplete_batch = drop_incomplete_batch
-        self.multiproc_kwargs = multiproc_kwargs
-
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        # Instantiate other datamodules and get their datasets
-
-        module_args = {
-            "train_batch_size": self.train_batch_size,
-            "predict_batch_size": self.predict_batch_size,
-            "drop_incomplete_batch": self.drop_incomplete_batch,
-            "multiproc_kwargs": self.multiproc_kwargs,
-        }
-
-        mc = get_dataset_from_module(mc_datamodule, stage, **module_args)
-        ch = get_dataset_from_module(ch_datamodule, stage, **module_args)
-        indian = get_dataset_from_module(
-            indian_datamodule, stage, **module_args
-        )
-        tbx11kv2 = get_dataset_from_module(
-            tbx11kv2_datamodule, stage, **module_args
-        )
-
-        # Combine datasets
-        self.dataset = {}
-        self.dataset["__train__"] = ConcatDataset(
-            [
-                mc["__train__"],
-                ch["__train__"],
-                indian["__train__"],
-                tbx11kv2["__train__"],
-            ]
-        )
-        self.dataset["train"] = ConcatDataset(
-            [mc["train"], ch["train"], indian["train"], tbx11kv2["train"]]
-        )
-        self.dataset["__valid__"] = ConcatDataset(
-            [
-                mc["__valid__"],
-                ch["__valid__"],
-                indian["__valid__"],
-                tbx11kv2["__valid__"],
-            ]
-        )
-        self.dataset["validation"] = ConcatDataset(
-            [
-                mc["validation"],
-                ch["validation"],
-                indian["validation"],
-                tbx11kv2["validation"],
-            ]
-        )
-        self.dataset["test"] = ConcatDataset(
-            [mc["test"], ch["test"], indian["test"], tbx11kv2["test"]]
-        )
-
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/mc_ch_in_pc/rgb.py

-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Aggregated dataset composed of Montgomery, Shenzhen, Indian and Padchest
-(RGB) datasets."""
-
-from clapper.logging import setup
-from torch.utils.data.dataset import ConcatDataset
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule, get_dataset_from_module
-from ..indian.rgb import datamodule as indian_datamodule
-from ..montgomery.rgb import datamodule as mc_datamodule
-from ..padchest.tb_idiap_rgb import datamodule as pc_datamodule
-from ..shenzhen.rgb import datamodule as ch_datamodule
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        self.train_batch_size = train_batch_size
-        self.predict_batch_size = predict_batch_size
-        self.drop_incomplete_batch = drop_incomplete_batch
-        self.multiproc_kwargs = multiproc_kwargs
-
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        # Instantiate other datamodules and get their datasets
-
-        module_args = {
-            "train_batch_size": self.train_batch_size,
-            "predict_batch_size": self.predict_batch_size,
-            "drop_incomplete_batch": self.drop_incomplete_batch,
-            "multiproc_kwargs": self.multiproc_kwargs,
-        }
-
-        mc = get_dataset_from_module(mc_datamodule, stage, **module_args)
-        ch = get_dataset_from_module(ch_datamodule, stage, **module_args)
-        indian = get_dataset_from_module(
-            indian_datamodule, stage, **module_args
-        )
-        pc = get_dataset_from_module(pc_datamodule, stage, **module_args)
-
-        # Combine datasets
-        self.dataset = {}
-        self.dataset["__train__"] = ConcatDataset(
-            [
-                mc["__train__"],
-                ch["__train__"],
-                indian["__train__"],
-                pc["__train__"],
-            ]
-        )
-        self.dataset["train"] = ConcatDataset(
-            [mc["train"], ch["train"], indian["train"], pc["train"]]
-        )
-        self.dataset["__valid__"] = ConcatDataset(
-            [
-                mc["__valid__"],
-                ch["__valid__"],
-                indian["__valid__"],
-                pc["__valid__"],
-            ]
-        )
-        self.dataset["test"] = ConcatDataset(
-            [mc["test"], ch["test"], indian["test"], pc["test"]]
-        )
-
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/padchest/tb_idiap_rgb.py

-# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""Padchest tuberculosis (idiap protocol, rgb) dataset for computer-aided
-diagnosis.
-
-The 125 healthy images are the first 125 padchest images with the following
-parameters: Label = "Normal", MethodLabel = "Physician", Projection = "PA"
-
-* Split reference: first 80% of TB and healthy CXR for "train", rest for "test"
-* See :py:mod:`ptbench.data.padchest` for dataset details
-* This configuration resolution: 224 x 224 (default)
-"""
-
-from clapper.logging import setup
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule
-from . import _maker
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        self.dataset = _maker("tb_idiap", resize_size=256, cc_size=224)
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/tbx11k_simplified/rgb.py

-# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""TBX11k simplified dataset for TB detection (default protocol)
-
-* Split reference: first 62.5% of TB and healthy CXR for "train" 15.9% for
-* "validation", 21.6% for "test"
-* This split only consists of healthy and active TB samples
-* "Latent TB" or "sick & non-TB" samples are not included in this configuration
-* This configuration resolution: 512 x 512 (default)
-* See :py:mod:`ptbench.data.tbx11k_simplified` for dataset details
-"""
-
-from clapper.logging import setup
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule
-from . import _maker
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        self.dataset = _maker("default", RGB=True)
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule

src/ptbench/data/tbx11k_simplified_v2/rgb.py

-# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-"""TBX11k simplified dataset for TB detection (default protocol, converted in
-RGB)
-
-* Split reference: first 62.6% of CXR for "train", 16% for "validation",
-* 21.4% for "test"
-* This split consists of non-TB and active TB samples
-* "healthy", "latent TB", and "sick & non-TB" samples are all merged under the label "non-TB"
-* This configuration resolution: 512 x 512 (default)
-* See :py:mod:`ptbench.data.tbx11k_simplified_v2` for dataset details
-"""
-
-from clapper.logging import setup
-
-from .. import return_subsets
-from ..base_datamodule import BaseDataModule
-from . import _maker
-
-logger = setup(__name__.split(".")[0], format="%(levelname)s: %(message)s")
-
-
-class DefaultModule(BaseDataModule):
-    def __init__(
-        self,
-        train_batch_size=1,
-        predict_batch_size=1,
-        drop_incomplete_batch=False,
-        multiproc_kwargs=None,
-    ):
-        super().__init__(
-            train_batch_size=train_batch_size,
-            predict_batch_size=predict_batch_size,
-            drop_incomplete_batch=drop_incomplete_batch,
-            multiproc_kwargs=multiproc_kwargs,
-        )
-
-    def setup(self, stage: str):
-        self.dataset = _maker("default", RGB=True)
-        (
-            self.train_dataset,
-            self.validation_dataset,
-            self.extra_validation_datasets,
-            self.predict_dataset,
-        ) = return_subsets(self.dataset)
-
-
-datamodule = DefaultModule