From 816e8b809c670de618bcf41f1f998b3d0caefe49 Mon Sep 17 00:00:00 2001
From: Andre Anjos <andre.dos.anjos@gmail.com>
Date: Mon, 31 Jul 2023 12:00:24 +0200
Subject: [PATCH] [data.montgomery_shenzhen_indian_tbx11k] Implement it
---
pyproject.toml | 46 ++++----
src/ptbench/data/mc_ch_in_11k/__init__.py | 3 -
src/ptbench/data/mc_ch_in_11k/default.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_0.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_1.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_2.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_3.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_4.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_5.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_6.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_7.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_8.py | 101 ------------------
src/ptbench/data/mc_ch_in_11k/fold_9.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/__init__.py | 3 -
src/ptbench/data/mc_ch_in_11kv2/default.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_0.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_1.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_2.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_3.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_4.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_5.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_6.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_7.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_8.py | 101 ------------------
src/ptbench/data/mc_ch_in_11kv2/fold_9.py | 101 ------------------
.../__init__.py | 0
.../datamodule.py | 50 +++++++++
.../v1_fold_0.py | 7 ++
.../v1_fold_1.py | 7 ++
.../v1_fold_2.py | 7 ++
.../v1_fold_3.py | 7 ++
.../v1_fold_4.py | 7 ++
.../v1_fold_5.py | 7 ++
.../v1_fold_6.py | 7 ++
.../v1_fold_7.py | 7 ++
.../v1_fold_8.py | 7 ++
.../v1_fold_9.py | 7 ++
.../v1_healthy_vs_atb.py | 7 ++
.../v2_fold_0.py | 7 ++
.../v2_fold_1.py | 7 ++
.../v2_fold_2.py | 7 ++
.../v2_fold_3.py | 7 ++
.../v2_fold_4.py | 7 ++
.../v2_fold_5.py | 7 ++
.../v2_fold_6.py | 7 ++
.../v2_fold_7.py | 7 ++
.../v2_fold_8.py | 7 ++
.../v2_fold_9.py | 7 ++
.../v2_others_vs_atb.py | 7 ++
49 files changed, 226 insertions(+), 2252 deletions(-)
delete mode 100644 src/ptbench/data/mc_ch_in_11k/__init__.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/default.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_0.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_1.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_2.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_3.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_4.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_5.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_6.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_7.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_8.py
delete mode 100644 src/ptbench/data/mc_ch_in_11k/fold_9.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/__init__.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/default.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_0.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_1.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_2.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_3.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_4.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_5.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_6.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_7.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_8.py
delete mode 100644 src/ptbench/data/mc_ch_in_11kv2/fold_9.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/__init__.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/datamodule.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_0.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_1.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_2.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_3.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_4.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_5.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_6.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_7.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_8.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_9.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_healthy_vs_atb.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_0.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_1.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_2.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_3.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_4.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_5.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_6.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_7.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_8.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_9.py
create mode 100644 src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_others_vs_atb.py
diff --git a/pyproject.toml b/pyproject.toml
index 6549e5c2..d76e3979 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -170,30 +170,28 @@ montgomery-shenzhen-indian-f8 = "ptbench.data.montgomery_shenzhen_indian.fold_8"
montgomery-shenzhen-indian-f9 = "ptbench.data.montgomery_shenzhen_indian.fold_9"
# montgomery-shenzhen-indian-tbx11k aggregated dataset
-mc_ch_in_11k = "ptbench.data.mc_ch_in_11k.default"
-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"
-mc_ch_in_11k_f3 = "ptbench.data.mc_ch_in_11k.fold_3"
-mc_ch_in_11k_f4 = "ptbench.data.mc_ch_in_11k.fold_4"
-mc_ch_in_11k_f5 = "ptbench.data.mc_ch_in_11k.fold_5"
-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"
-
-# montgomery-shenzhen-indian-tbx11kv2 aggregated dataset
-mc_ch_in_11kv2 = "ptbench.data.mc_ch_in_11kv2.default"
-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"
-mc_ch_in_11kv2_f3 = "ptbench.data.mc_ch_in_11kv2.fold_3"
-mc_ch_in_11kv2_f4 = "ptbench.data.mc_ch_in_11kv2.fold_4"
-mc_ch_in_11kv2_f5 = "ptbench.data.mc_ch_in_11kv2.fold_5"
-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"
+montgomery-shenzhen-indian-tbx11k-v1 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_healthy_vs_atb"
+montgomery-shenzhen-indian-tbx11k-v1-f0 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_0"
+montgomery-shenzhen-indian-tbx11k-v1-f1 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_1"
+montgomery-shenzhen-indian-tbx11k-v1-f2 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_2"
+montgomery-shenzhen-indian-tbx11k-v1-f3 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_3"
+montgomery-shenzhen-indian-tbx11k-v1-f4 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_4"
+montgomery-shenzhen-indian-tbx11k-v1-f5 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_5"
+montgomery-shenzhen-indian-tbx11k-v1-f6 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_6"
+montgomery-shenzhen-indian-tbx11k-v1-f7 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_7"
+montgomery-shenzhen-indian-tbx11k-v1-f8 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_8"
+montgomery-shenzhen-indian-tbx11k-v1-f9 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v1_fold_9"
+montgomery-shenzhen-indian-tbx11k-v2 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_others_vs_atb"
+montgomery-shenzhen-indian-tbx11k-v2-f0 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_0"
+montgomery-shenzhen-indian-tbx11k-v2-f1 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_1"
+montgomery-shenzhen-indian-tbx11k-v2-f2 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_2"
+montgomery-shenzhen-indian-tbx11k-v2-f3 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_3"
+montgomery-shenzhen-indian-tbx11k-v2-f4 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_4"
+montgomery-shenzhen-indian-tbx11k-v2-f5 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_5"
+montgomery-shenzhen-indian-tbx11k-v2-f6 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_6"
+montgomery-shenzhen-indian-tbx11k-v2-f7 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_7"
+montgomery-shenzhen-indian-tbx11k-v2-f8 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_8"
+montgomery-shenzhen-indian-tbx11k-v2-f9 = "ptbench.data.montgomery_shenzhen_indian_tbx11k.v2_fold_9"
# tbpoc dataset (and cross-validation folds)
tbpoc_f0 = "ptbench.data.tbpoc.fold_0"
diff --git a/src/ptbench/data/mc_ch_in_11k/__init__.py b/src/ptbench/data/mc_ch_in_11k/__init__.py
deleted file mode 100644
index 662d5c13..00000000
--- a/src/ptbench/data/mc_ch_in_11k/__init__.py
+++ /dev/null
@@ -1,3 +0,0 @@
-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
diff --git a/src/ptbench/data/mc_ch_in_11k/default.py b/src/ptbench/data/mc_ch_in_11k/default.py
deleted file mode 100644
index 454521a7..00000000
--- a/src/ptbench/data/mc_ch_in_11k/default.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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."""
-
-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.default import datamodule as indian_datamodule
-from ..montgomery.default import datamodule as mc_datamodule
-from ..shenzhen.default import datamodule as ch_datamodule
-from ..tbx11k_simplified.default 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_0.py b/src/ptbench/data/mc_ch_in_11k/fold_0.py
deleted file mode 100644
index 1adce163..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_0.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 0)"""
-
-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.fold_0 import datamodule as indian_datamodule
-from ..montgomery.fold_0 import datamodule as mc_datamodule
-from ..shenzhen.fold_0 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_0 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_1.py b/src/ptbench/data/mc_ch_in_11k/fold_1.py
deleted file mode 100644
index 584a65c4..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_1.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 1)"""
-
-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.fold_1 import datamodule as indian_datamodule
-from ..montgomery.fold_1 import datamodule as mc_datamodule
-from ..shenzhen.fold_1 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_1 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_2.py b/src/ptbench/data/mc_ch_in_11k/fold_2.py
deleted file mode 100644
index 05c0234e..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_2.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 2)"""
-
-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.fold_2 import datamodule as indian_datamodule
-from ..montgomery.fold_2 import datamodule as mc_datamodule
-from ..shenzhen.fold_2 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_2 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_3.py b/src/ptbench/data/mc_ch_in_11k/fold_3.py
deleted file mode 100644
index 80a544be..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_3.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 3)"""
-
-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.fold_3 import datamodule as indian_datamodule
-from ..montgomery.fold_3 import datamodule as mc_datamodule
-from ..shenzhen.fold_3 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_3 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_4.py b/src/ptbench/data/mc_ch_in_11k/fold_4.py
deleted file mode 100644
index 5860e1dd..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_4.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 4)"""
-
-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.fold_4 import datamodule as indian_datamodule
-from ..montgomery.fold_4 import datamodule as mc_datamodule
-from ..shenzhen.fold_4 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_4 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_5.py b/src/ptbench/data/mc_ch_in_11k/fold_5.py
deleted file mode 100644
index 93fac65b..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_5.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 5)"""
-
-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.fold_5 import datamodule as indian_datamodule
-from ..montgomery.fold_5 import datamodule as mc_datamodule
-from ..shenzhen.fold_5 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_5 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_6.py b/src/ptbench/data/mc_ch_in_11k/fold_6.py
deleted file mode 100644
index 44a79a42..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_6.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 6)"""
-
-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.fold_6 import datamodule as indian_datamodule
-from ..montgomery.fold_6 import datamodule as mc_datamodule
-from ..shenzhen.fold_6 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_6 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_7.py b/src/ptbench/data/mc_ch_in_11k/fold_7.py
deleted file mode 100644
index d955a02e..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_7.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 7)"""
-
-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.fold_7 import datamodule as indian_datamodule
-from ..montgomery.fold_7 import datamodule as mc_datamodule
-from ..shenzhen.fold_7 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_7 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_8.py b/src/ptbench/data/mc_ch_in_11k/fold_8.py
deleted file mode 100644
index 77753c11..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_8.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 8)"""
-
-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.fold_8 import datamodule as indian_datamodule
-from ..montgomery.fold_8 import datamodule as mc_datamodule
-from ..shenzhen.fold_8 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_8 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
diff --git a/src/ptbench/data/mc_ch_in_11k/fold_9.py b/src/ptbench/data/mc_ch_in_11k/fold_9.py
deleted file mode 100644
index 45e88b27..00000000
--- a/src/ptbench/data/mc_ch_in_11k/fold_9.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 9)"""
-
-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.fold_9 import datamodule as indian_datamodule
-from ..montgomery.fold_9 import datamodule as mc_datamodule
-from ..shenzhen.fold_9 import datamodule as ch_datamodule
-from ..tbx11k_simplified.fold_9 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/__init__.py b/src/ptbench/data/mc_ch_in_11kv2/__init__.py
deleted file mode 100644
index 662d5c13..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/__init__.py
+++ /dev/null
@@ -1,3 +0,0 @@
-# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
diff --git a/src/ptbench/data/mc_ch_in_11kv2/default.py b/src/ptbench/data/mc_ch_in_11kv2/default.py
deleted file mode 100644
index 1eea460b..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/default.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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."""
-
-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.default import datamodule as indian_datamodule
-from ..montgomery.default import datamodule as mc_datamodule
-from ..shenzhen.default import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.default 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_0.py b/src/ptbench/data/mc_ch_in_11kv2/fold_0.py
deleted file mode 100644
index 738164a7..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_0.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 0)"""
-
-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.fold_0 import datamodule as indian_datamodule
-from ..montgomery.fold_0 import datamodule as mc_datamodule
-from ..shenzhen.fold_0 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_0 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_1.py b/src/ptbench/data/mc_ch_in_11kv2/fold_1.py
deleted file mode 100644
index bfc2dbfc..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_1.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 1)"""
-
-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.fold_1 import datamodule as indian_datamodule
-from ..montgomery.fold_1 import datamodule as mc_datamodule
-from ..shenzhen.fold_1 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_1 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_2.py b/src/ptbench/data/mc_ch_in_11kv2/fold_2.py
deleted file mode 100644
index 06c7c1ab..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_2.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 2)"""
-
-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.fold_2 import datamodule as indian_datamodule
-from ..montgomery.fold_2 import datamodule as mc_datamodule
-from ..shenzhen.fold_2 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_2 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_3.py b/src/ptbench/data/mc_ch_in_11kv2/fold_3.py
deleted file mode 100644
index 11309b1c..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_3.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 3)"""
-
-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.fold_3 import datamodule as indian_datamodule
-from ..montgomery.fold_3 import datamodule as mc_datamodule
-from ..shenzhen.fold_3 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_3 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_4.py b/src/ptbench/data/mc_ch_in_11kv2/fold_4.py
deleted file mode 100644
index 0f53ed63..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_4.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 4)"""
-
-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.fold_4 import datamodule as indian_datamodule
-from ..montgomery.fold_4 import datamodule as mc_datamodule
-from ..shenzhen.fold_4 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_4 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_5.py b/src/ptbench/data/mc_ch_in_11kv2/fold_5.py
deleted file mode 100644
index df7dc0e8..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_5.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 5)"""
-
-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.fold_5 import datamodule as indian_datamodule
-from ..montgomery.fold_5 import datamodule as mc_datamodule
-from ..shenzhen.fold_5 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_5 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_6.py b/src/ptbench/data/mc_ch_in_11kv2/fold_6.py
deleted file mode 100644
index ef246cd8..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_6.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 6)"""
-
-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.fold_6 import datamodule as indian_datamodule
-from ..montgomery.fold_6 import datamodule as mc_datamodule
-from ..shenzhen.fold_6 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_6 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_7.py b/src/ptbench/data/mc_ch_in_11kv2/fold_7.py
deleted file mode 100644
index 41d3ba3e..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_7.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 7)"""
-
-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.fold_7 import datamodule as indian_datamodule
-from ..montgomery.fold_7 import datamodule as mc_datamodule
-from ..shenzhen.fold_7 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_7 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_8.py b/src/ptbench/data/mc_ch_in_11kv2/fold_8.py
deleted file mode 100644
index a6842d47..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_8.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 8)"""
-
-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.fold_8 import datamodule as indian_datamodule
-from ..montgomery.fold_8 import datamodule as mc_datamodule
-from ..shenzhen.fold_8 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_8 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
diff --git a/src/ptbench/data/mc_ch_in_11kv2/fold_9.py b/src/ptbench/data/mc_ch_in_11kv2/fold_9.py
deleted file mode 100644
index ba69b788..00000000
--- a/src/ptbench/data/mc_ch_in_11kv2/fold_9.py
+++ /dev/null
@@ -1,101 +0,0 @@
-# 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 (cross validation fold 9)"""
-
-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.fold_9 import datamodule as indian_datamodule
-from ..montgomery.fold_9 import datamodule as mc_datamodule
-from ..shenzhen.fold_9 import datamodule as ch_datamodule
-from ..tbx11k_simplified_v2.fold_9 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
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/__init__.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/__init__.py
new file mode 100644
index 00000000..e69de29b
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/datamodule.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/datamodule.py
new file mode 100644
index 00000000..02599187
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/datamodule.py
@@ -0,0 +1,50 @@
+# Copyright © 2022 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from ..datamodule import ConcatDataModule
+from ..indian.datamodule import RawDataLoader as IndianLoader
+from ..indian.datamodule import make_split as make_indian_split
+from ..montgomery.datamodule import RawDataLoader as MontgomeryLoader
+from ..montgomery.datamodule import make_split as make_montgomery_split
+from ..shenzhen.datamodule import RawDataLoader as ShenzhenLoader
+from ..shenzhen.datamodule import make_split as make_shenzhen_split
+from ..tbx11k.datamodule import RawDataLoader as TBX11kLoader
+from ..tbx11k.datamodule import make_split as make_tbx11k_split
+
+
+class DataModule(ConcatDataModule):
+ """Aggregated datamodule composed of Montgomery and Shenzhen datasets."""
+
+ def __init__(self, split_filename: str, tbx11_split_filename: str):
+ montgomery_loader = MontgomeryLoader()
+ montgomery_split = make_montgomery_split(split_filename)
+ shenzhen_loader = ShenzhenLoader()
+ shenzhen_split = make_shenzhen_split(split_filename)
+ indian_loader = IndianLoader()
+ indian_split = make_indian_split(split_filename)
+ tbx11k_loader = TBX11kLoader()
+ tbx11k_split = make_tbx11k_split(tbx11_split_filename)
+
+ super().__init__(
+ splits={
+ "train": [
+ (montgomery_split["train"], montgomery_loader),
+ (shenzhen_split["train"], shenzhen_loader),
+ (indian_split["train"], indian_loader),
+ (tbx11k_split["train"], tbx11k_loader),
+ ],
+ "validation": [
+ (montgomery_split["validation"], montgomery_loader),
+ (shenzhen_split["validation"], shenzhen_loader),
+ (indian_split["validation"], indian_loader),
+ (tbx11k_split["validation"], tbx11k_loader),
+ ],
+ "test": [
+ (montgomery_split["test"], montgomery_loader),
+ (shenzhen_split["test"], shenzhen_loader),
+ (indian_split["test"], indian_loader),
+ (tbx11k_split["test"], tbx11k_loader),
+ ],
+ }
+ )
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_0.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_0.py
new file mode 100644
index 00000000..42620697
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_0.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-0.json", "v1-fold-0.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_1.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_1.py
new file mode 100644
index 00000000..c7f11aad
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_1.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-1.json", "v1-fold-1.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_2.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_2.py
new file mode 100644
index 00000000..8c94b1e9
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_2.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-2.json", "v1-fold-2.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_3.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_3.py
new file mode 100644
index 00000000..c90704d1
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_3.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-3.json", "v1-fold-3.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_4.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_4.py
new file mode 100644
index 00000000..5d243746
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_4.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-4.json", "v1-fold-4.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_5.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_5.py
new file mode 100644
index 00000000..65aa7840
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_5.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-5.json", "v1-fold-5.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_6.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_6.py
new file mode 100644
index 00000000..bc175ac2
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_6.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-6.json", "v1-fold-6.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_7.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_7.py
new file mode 100644
index 00000000..3b7d7f71
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_7.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-7.json", "v1-fold-7.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_8.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_8.py
new file mode 100644
index 00000000..20ac8e5c
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_8.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-8.json", "v1-fold-8.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_9.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_9.py
new file mode 100644
index 00000000..413f141d
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_fold_9.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-9.json", "v1-fold-9.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_healthy_vs_atb.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_healthy_vs_atb.py
new file mode 100644
index 00000000..2f5e2226
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v1_healthy_vs_atb.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("default.json", "v1-healthy-vs-atb.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_0.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_0.py
new file mode 100644
index 00000000..634aff84
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_0.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-0.json", "v2-others-vs-atb.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_1.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_1.py
new file mode 100644
index 00000000..10de943e
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_1.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-1.json", "v2-fold-1.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_2.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_2.py
new file mode 100644
index 00000000..12062ae6
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_2.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-2.json", "v2-fold-2.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_3.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_3.py
new file mode 100644
index 00000000..453c2b8c
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_3.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-3.json", "v2-fold-3.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_4.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_4.py
new file mode 100644
index 00000000..d31da770
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_4.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-4.json", "v2-fold-4.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_5.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_5.py
new file mode 100644
index 00000000..1c50c2e7
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_5.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-5.json", "v2-fold-5.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_6.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_6.py
new file mode 100644
index 00000000..0f15a0fa
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_6.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-6.json", "v2-fold-6.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_7.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_7.py
new file mode 100644
index 00000000..10829456
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_7.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-7.json", "v2-fold-7.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_8.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_8.py
new file mode 100644
index 00000000..a90e1111
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_8.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-8.json", "v2-fold-8.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_9.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_9.py
new file mode 100644
index 00000000..a6cb7d51
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_fold_9.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("fold-9.json", "v2-fold-9.json")
diff --git a/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_others_vs_atb.py b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_others_vs_atb.py
new file mode 100644
index 00000000..bef1efce
--- /dev/null
+++ b/src/ptbench/data/montgomery_shenzhen_indian_tbx11k/v2_others_vs_atb.py
@@ -0,0 +1,7 @@
+# SPDX-FileCopyrightText: Copyright © 2023 Idiap Research Institute <contact@idiap.ch>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+from .datamodule import DataModule
+
+datamodule = DataModule("default.json", "v2-others-vs-atb.json")
--
GitLab