[mednet] Fixes after rebase

pyproject.toml

@@ -427,6 +427,9 @@ lwnet = "mednet.libs.segmentation.config.models.lwnet"
 chasedb1 = "mednet.libs.segmentation.config.data.chasedb1.first_annotator"
 chasedb1-2nd = "mednet.libs.segmentation.config.data.chasedb1.second_annotator"
 
+# cxr8 - cxr
+cxr8 = "mednet.libs.segmentation.config.data.cxr8.default"
+
 # drive dataset - retinography
 drive = "mednet.libs.segmentation.config.data.drive.default"
 

src/mednet/libs/classification/config/data/hivtb/datamodule.py

@@ -15,10 +15,8 @@ from mednet.libs.classification.data.typing import (
 )
 from mednet.libs.common.data.datamodule import CachingDataModule
 from mednet.libs.common.data.image_utils import remove_black_borders
-
 from mednet.libs.common.data.split import make_split
 from mednet.libs.common.data.typing import Sample
-
 from torchvision.transforms.functional import to_tensor
 
 from ....utils.rc import load_rc
@@ -127,7 +125,7 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/classification/config/data/indian/datamodule.py

@@ -11,9 +11,8 @@ import pathlib
 from mednet.libs.common.data.datamodule import CachingDataModule
 from mednet.libs.common.data.split import make_split
 
-from mednet.libs.classification.data.typing import (
-    ClassificationRawDataLoader as _ClassificationRawDataLoader,
-)
+from ....config.data.shenzhen.datamodule import ClassificationRawDataLoader
+
 CONFIGURATION_KEY_DATADIR = "datadir." + (__name__.rsplit(".", 2)[-2])
 """Key to search for in the configuration file for the root directory of this
 database."""
@@ -64,8 +63,9 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(
+                config_variable=CONFIGURATION_KEY_DATADIR
+            ),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )
-

src/mednet/libs/classification/config/data/montgomery/datamodule.py

@@ -15,10 +15,8 @@ from mednet.libs.classification.data.typing import (
 )
 from mednet.libs.common.data.datamodule import CachingDataModule
 from mednet.libs.common.data.image_utils import remove_black_borders
-
 from mednet.libs.common.data.split import make_split
 from mednet.libs.common.data.typing import Sample
-
 from torchvision.transforms.functional import to_tensor
 
 from ....utils.rc import load_rc
@@ -28,7 +26,7 @@ CONFIGURATION_KEY_DATADIR = "datadir." + (__name__.rsplit(".", 2)[-2])
 database."""
 
 
-class RawDataLoader(_BaseRawDataLoader):
+class ClassificationRawDataLoader(_ClassificationRawDataLoader):
     """A specialized raw-data-loader for the Montgomery dataset.
 
     Parameters
@@ -140,7 +138,7 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/classification/config/data/nih_cxr14/datamodule.py

@@ -14,10 +14,8 @@ from mednet.libs.classification.data.typing import (
     ClassificationRawDataLoader as _ClassificationRawDataLoader,
 )
 from mednet.libs.common.data.datamodule import CachingDataModule
-
 from mednet.libs.common.data.split import make_split
 from mednet.libs.common.data.typing import Sample
-
 from torchvision.transforms.functional import to_tensor
 
 from ....utils.rc import load_rc
@@ -86,7 +84,9 @@ class ClassificationRawDataLoader(_ClassificationRawDataLoader):
             # for folder lookup efficiency, data is split into subfolders
             # each original file is on the subfolder `f[:5]/f`, where f
             # is the original file basename
-            file_path = file_path.parent / file_path.name[:5] / file_path.name
+            file_path = pathlib.Path(
+                file_path.parent / file_path.name[:5] / file_path.name
+            )
 
         # N.B.: some NIH CXR-14 images are encoded as color PNGs with an alpha
         # channel.  Most, are grayscale PNGs
@@ -176,7 +176,7 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/classification/config/data/padchest/datamodule.py

@@ -16,10 +16,8 @@ from mednet.libs.classification.data.typing import (
 )
 from mednet.libs.common.data.datamodule import CachingDataModule
 from mednet.libs.common.data.image_utils import remove_black_borders
-
 from mednet.libs.common.data.split import make_split
 from mednet.libs.common.data.typing import Sample
-
 from torchvision.transforms.functional import to_tensor
 
 from ....utils.rc import load_rc
@@ -329,7 +327,7 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/classification/config/data/shenzhen/datamodule.py

@@ -15,10 +15,8 @@ from mednet.libs.classification.data.typing import (
 )
 from mednet.libs.common.data.datamodule import CachingDataModule
 from mednet.libs.common.data.image_utils import remove_black_borders
-
 from mednet.libs.common.data.split import make_split
 from mednet.libs.common.data.typing import Sample
-
 from torchvision.transforms.functional import to_tensor
 
 from ....utils.rc import load_rc
@@ -139,7 +137,7 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/classification/config/data/tbpoc/datamodule.py

@@ -11,10 +11,8 @@ from mednet.libs.classification.data.typing import (
 )
 from mednet.libs.common.data.datamodule import CachingDataModule
 from mednet.libs.common.data.image_utils import remove_black_borders
-
 from mednet.libs.common.data.split import make_split
 from mednet.libs.common.data.typing import Sample
-
 from torchvision.transforms.functional import to_tensor
 
 from ....utils.rc import load_rc
@@ -125,7 +123,7 @@ class DataModule(CachingDataModule):
         assert __package__ is not None
         super().__init__(
             database_split=make_split(__package__, split_filename),
-            raw_data_loader=_ClassificationRawDataLoader(),
-            database_name=__package__.rsplit(".", 1)[1],            
+            raw_data_loader=ClassificationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
             split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/classification/tests/test_hivtb.py

@@ -39,7 +39,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.hivtb", f"{split}.json"),
+        make_split("mednet.libs.classification.config.data.hivtb", f"{split}.json"),
         lengths=lengths,
         prefixes=("HIV-TB_Algorithm_study_X-rays",),
         possible_labels=(0, 1),

src/mednet/libs/classification/tests/test_indian.py

@@ -43,7 +43,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.indian", f"{split}.json"),
+        make_split("mednet.libs.classification.config.data.indian", f"{split}.json"),
         lengths=lengths,
         prefixes=("DatasetA/Training", "DatasetA/Testing"),
         possible_labels=(0, 1),

src/mednet/libs/classification/tests/test_montgomery.py

@@ -41,7 +41,9 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.montgomery", f"{split}.json"),
+        make_split(
+            "mednet.libs.classification.config.data.montgomery", f"{split}.json"
+        ),
         lengths=lengths,
         prefixes=("CXR_png/MCUCXR_0",),
         possible_labels=(0, 1),
@@ -114,12 +116,8 @@ def test_loading(database_checkers, name: str, dataset: str):
 @pytest.mark.skip_if_rc_var_not_set("datadir.montgomery")
 def test_raw_transforms_image_quality(database_checkers, datadir: pathlib.Path):
     datamodule = importlib.import_module(
-<<<<<<< HEAD
-        ".default", "mednet.config.data.montgomery"
-=======
         ".default",
         "mednet.libs.classification.config.data.montgomery",
->>>>>>> b1ea1c0 ([mednet] Start reorganizing into monorepo)
     ).datamodule
 
     datamodule.model_transforms = []

src/mednet/libs/classification/tests/test_nih_cxr14.py

@@ -31,7 +31,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.nih_cxr14", f"{split}"),
+        make_split("mednet.libs.classification.config.data.nih_cxr14", f"{split}"),
         lengths=lengths,
         prefixes=("images/000",),
         possible_labels=(0, 1),

src/mednet/libs/classification/tests/test_padchest.py

@@ -36,7 +36,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.padchest", split),
+        make_split("mednet.libs.classification.config.data.padchest", split),
         lengths=lengths,
         prefixes=("",),
         possible_labels=(0, 1),

src/mednet/libs/classification/tests/test_shenzhen.py

@@ -40,7 +40,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.shenzhen", f"{split}.json"),
+        make_split("mednet.libs.classification.config.data.shenzhen", f"{split}.json"),
         lengths=lengths,
         prefixes=("CXR_png/CHNCXR_0",),
         possible_labels=(0, 1),

src/mednet/libs/classification/tests/test_tbpoc.py

@@ -39,7 +39,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.tbpoc", f"{split}.json"),
+        make_split("mednet.libs.classification.config.data.tbpoc", f"{split}.json"),
         lengths=lengths,
         prefixes=(
             "TBPOC_CXR/TBPOC-",

src/mednet/libs/classification/tests/test_tbx11k.py

@@ -142,7 +142,7 @@ def test_protocol_consistency(
     from mednet.libs.common.data.split import make_split
 
     database_checkers.check_split(
-        make_split("mednet.config.data.tbx11k", f"{split}.json"),
+        make_split("mednet.libs.classification.config.data.tbx11k", f"{split}.json"),
         lengths=lengths,
         prefixes=prefixes,
         possible_labels=(0, 1),

src/mednet/libs/common/models/model.py

@@ -11,8 +11,8 @@ import torch.nn
 import torch.optim.optimizer
 import torch.utils.data
 import torchvision.transforms
+from mednet.libs.common.data.typing import TransformSequence
 
-from medbase.data.typing import TransformSequence
 from .loss_weights import get_positive_weights
 from .typing import Checkpoint
 
@@ -224,7 +224,5 @@ class Model(pl.LightningModule):
                 logger.warning(
                     "Datamodule does not contain a validation dataloader. The training dataloader will be used instead."
                 )
-                validation_weights = get_positive_weights(
-                    datamodule.train_dataloader()
-                )
+                validation_weights = get_positive_weights(datamodule.train_dataloader())
             self._validation_loss_arguments["pos_weight"] = validation_weights

src/mednet/libs/common/tests/test_database_split.py

@@ -3,31 +3,9 @@
 # SPDX-License-Identifier: GPL-3.0-or-later
 """Test code for datasets."""
 
-<<<<<<< HEAD
-from medbase.data.split import JSONDatabaseSplit
-=======
-from mednet.libs.common.data.split import CSVDatabaseSplit, JSONDatabaseSplit
+from mednet.libs.common.data.split import JSONDatabaseSplit
 
 
-def test_csv_loading(datadir):
-    # tests if we can build a simple CSV loader for the Iris Flower dataset
-    database_split = CSVDatabaseSplit(datadir)
-
-    assert len(database_split["iris-train"]) == 75
-    for k in database_split["iris-train"]:
-        for f in range(4):
-            assert isinstance(k[f], str)  # csv only loads stringd
-        assert isinstance(k[4], str)
-
-    assert len(database_split["iris-test"]) == 75
-    for k in database_split["iris-test"]:
-        for f in range(4):
-            assert isinstance(k[f], str)  # csv only loads stringd
-        assert isinstance(k[4], str)
-        assert k[4] in ("Iris-setosa", "Iris-versicolor", "Iris-virginica")
-
->>>>>>> b1ea1c0 ([mednet] Start reorganizing into monorepo)
-
 def test_json_loading(datadir):
     # tests if we can build a simple JSON loader for the Iris Flower dataset
 

src/mednet/libs/segmentation/config/data/chasedb1/datamodule.py

@@ -3,15 +3,14 @@
 # SPDX-License-Identifier: GPL-3.0-or-later
 """CHASE-DB1 dataset for Vessel Segmentation."""
 
-import importlib.resources
 import os
-from pathlib import Path
+import pathlib
 
 import PIL.Image
 import pkg_resources
 from mednet.libs.common.data.datamodule import CachingDataModule
-from mednet.libs.common.data.split import JSONDatabaseSplit
-from mednet.libs.common.data.typing import DatabaseSplit, Sample
+from mednet.libs.common.data.split import make_split
+from mednet.libs.common.data.typing import Sample
 from mednet.libs.segmentation.data.typing import (
     SegmentationRawDataLoader as _SegmentationRawDataLoader,
 )
@@ -28,16 +27,18 @@ database."""
 class SegmentationRawDataLoader(_SegmentationRawDataLoader):
     """A specialized raw-data-loader for the Chase-db1 dataset."""
 
-    datadir: str
+    datadir: pathlib.Path
     """This variable contains the base directory where the database raw data is
     stored."""
 
     def __init__(self):
-        self.datadir = load_rc().get(
-            CONFIGURATION_KEY_DATADIR, os.path.realpath(os.curdir)
+        self.datadir = pathlib.Path(
+            load_rc().get(CONFIGURATION_KEY_DATADIR, os.path.realpath(os.curdir))
         )
 
-        self._pkg_path = pkg_resources.resource_filename(__name__, "masks")
+        self._pkg_path = pathlib.Path(
+            pkg_resources.resource_filename(__name__, "masks")
+        )
 
     def sample(self, sample: tuple[str, str, str]) -> Sample:
         """Load a single image sample from the disk.
@@ -53,20 +54,16 @@ class SegmentationRawDataLoader(_SegmentationRawDataLoader):
             The sample representation.
         """
 
-        image = PIL.Image.open(Path(self.datadir) / str(sample[0])).convert(
-            mode="RGB"
-        )
+        image = PIL.Image.open(self.datadir / sample[0]).convert(mode="RGB")
         tensor = tv_tensors.Image(to_tensor(image))
         target = tv_tensors.Image(
             to_tensor(
-                PIL.Image.open(Path(self.datadir) / str(sample[1])).convert(
-                    mode="1", dither=None
-                )
+                PIL.Image.open(self.datadir / sample[1]).convert(mode="1", dither=None)
             )
         )
         mask = tv_tensors.Mask(
             to_tensor(
-                PIL.Image.open(Path(self._pkg_path) / str(sample[2])).convert(
+                PIL.Image.open(self._pkg_path / sample[2]).convert(
                     mode="1", dither=None
                 )
             )
@@ -75,24 +72,6 @@ class SegmentationRawDataLoader(_SegmentationRawDataLoader):
         return tensor, dict(target=target, mask=mask, name=sample[0])  # type: ignore[arg-type]
 
 
-def make_split(basename: str) -> DatabaseSplit:
-    """Return a database split for the Chase-db1 database.
-
-    Parameters
-    ----------
-    basename
-        Name of the .json file containing the split to load.
-
-    Returns
-    -------
-        An instance of DatabaseSplit.
-    """
-
-    return JSONDatabaseSplit(
-        importlib.resources.files(__name__.rsplit(".", 1)[0]).joinpath(basename)
-    )
-
-
 class DataModule(CachingDataModule):
     """CHASE-DB1 dataset for Vessel Segmentation.
 
@@ -129,7 +108,10 @@ class DataModule(CachingDataModule):
     """
 
     def __init__(self, split_filename: str):
+        assert __package__ is not None
         super().__init__(
-            database_split=make_split(split_filename),
+            database_split=make_split(__package__, split_filename),
             raw_data_loader=SegmentationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
+            split_name=pathlib.Path(split_filename).stem,
         )

src/mednet/libs/segmentation/config/data/drive/datamodule.py

@@ -3,14 +3,13 @@
 # SPDX-License-Identifier: GPL-3.0-or-later
 """COVD-DRIVE for Vessel Segmentation."""
 
-import importlib.resources
 import os
-from pathlib import Path
+import pathlib
 
 import PIL.Image
 from mednet.libs.common.data.datamodule import CachingDataModule
-from mednet.libs.common.data.split import JSONDatabaseSplit
-from mednet.libs.common.data.typing import DatabaseSplit, Sample
+from mednet.libs.common.data.split import make_split
+from mednet.libs.common.data.typing import Sample
 from mednet.libs.common.models.transforms import crop_image_to_mask
 from mednet.libs.segmentation.data.typing import (
     SegmentationRawDataLoader as _SegmentationRawDataLoader,
@@ -28,13 +27,13 @@ database."""
 class SegmentationRawDataLoader(_SegmentationRawDataLoader):
     """A specialized raw-data-loader for the Drive dataset."""
 
-    datadir: str
+    datadir: pathlib.Path
     """This variable contains the base directory where the database raw data is
     stored."""
 
     def __init__(self):
-        self.datadir = load_rc().get(
-            CONFIGURATION_KEY_DATADIR, os.path.realpath(os.curdir)
+        self.datadir = pathlib.Path(
+            load_rc().get(CONFIGURATION_KEY_DATADIR, os.path.realpath(os.curdir))
         )
 
     def sample(self, sample: tuple[str, str, str]) -> Sample:
@@ -51,20 +50,12 @@ class SegmentationRawDataLoader(_SegmentationRawDataLoader):
             The sample representation.
         """
 
-        image = to_tensor(
-            PIL.Image.open(Path(self.datadir) / str(sample[0])).convert(
-                mode="RGB"
-            )
-        )
+        image = to_tensor(PIL.Image.open(self.datadir / sample[0]).convert(mode="RGB"))
         target = to_tensor(
-            PIL.Image.open(Path(self.datadir) / str(sample[1])).convert(
-                mode="1", dither=None
-            )
+            PIL.Image.open(self.datadir / sample[1]).convert(mode="1", dither=None)
         )
         mask = to_tensor(
-            PIL.Image.open(Path(self.datadir) / str(sample[2])).convert(
-                mode="1", dither=None
-            )
+            PIL.Image.open(self.datadir / sample[2]).convert(mode="1", dither=None)
         )
 
         tensor = tv_tensors.Image(crop_image_to_mask(image, mask))
@@ -74,24 +65,6 @@ class SegmentationRawDataLoader(_SegmentationRawDataLoader):
         return tensor, dict(target=target, mask=mask, name=sample[0])  # type: ignore[arg-type]
 
 
-def make_split(basename: str) -> DatabaseSplit:
-    """Return a database split for the Drive database.
-
-    Parameters
-    ----------
-    basename
-        Name of the .json file containing the split to load.
-
-    Returns
-    -------
-        An instance of DatabaseSplit.
-    """
-
-    return JSONDatabaseSplit(
-        importlib.resources.files(__name__.rsplit(".", 1)[0]).joinpath(basename)
-    )
-
-
 class DataModule(CachingDataModule):
     """DRIVE dataset for Vessel Segmentation.
 
@@ -117,7 +90,10 @@ class DataModule(CachingDataModule):
     """
 
     def __init__(self, split_filename: str):
+        assert __package__ is not None
         super().__init__(
-            database_split=make_split(split_filename),
+            database_split=make_split(__package__, split_filename),
             raw_data_loader=SegmentationRawDataLoader(),
+            database_name=__package__.rsplit(".", 1)[1],
+            split_name=pathlib.Path(split_filename).stem,
         )