Merge pull request #44 from pyt-team/dev

Numpy style docstrings

configs/__init__.py

@@ -1 +1 @@
-# this file is needed here to include configs when building project as a package
+"""This file is needed here to include configs when building project as a package."""

configs/test.yaml

@@ -0,0 +1,57 @@
+# @package _global_
+
+# specify here default configuration
+# order of defaults determines the order in which configs override each other
+defaults:
+  - _self_
+  - dataset: graph/ZINC
+  - model: simplicial/sccn
+  - transforms: no_transform
+  - optimizer: adam
+  - scheduler: step_lr
+  - loss: default
+  - evaluator: default
+  - callbacks: default
+  - logger: wandb # set logger here or use command line (e.g. `python train.py logger=tensorboard`)
+  - trainer: cpu
+  - paths: default
+  - extras: default
+  - hydra: default
+
+  # experiment configs allow for version control of specific hyperparameters
+  # e.g. best hyperparameters for given model and datamodule
+  - experiment: null
+
+  # config for hyperparameter optimization
+  - hparams_search: null
+
+  # optional local config for machine/user specific settings
+  # it's optional since it doesn't need to exist and is excluded from version control
+  - optional local: default
+
+  # debugging config (enable through command line, e.g. `python train.py debug=default)
+  - debug: null
+
+# evaluator: ${dataset.parameters.task}
+# callbacks: ${dataset.parameters.task}
+
+# task name, determines output directory path
+task_name: "train"
+
+# tags to help you identify your experiments
+# you can overwrite this in experiment configs
+# overwrite from command line with `python train.py tags="[first_tag, second_tag]"`
+tags: ["dev"]
+
+# set False to skip model training
+train: True
+
+# evaluate on test set, using best model weights achieved during training
+# lightning chooses best weights based on the metric specified in checkpoint callback
+test: True
+
+# simply provide checkpoint path to resume training
+ckpt_path: null
+
+# seed for random number generators in pytorch, numpy and python.random
+seed: 42

notebooks/__init__.py

@@ -0,0 +1 @@
+"""Empty file to make the notebooks folder a package."""

pyproject.toml

@@ -48,9 +48,9 @@ dependencies=[
     "jupyterlab",
     "rich",
     "rootutils",
-    #"toponetx @ git+https://github.com/pyt-team/TopoNetX.git",
-    #"topomodelx @ git+https://github.com/pyt-team/TopoModelX.git",
-    #"topoembedx @ git+https://github.com/pyt-team/TopoEmbedX.git",
+    "toponetx @ git+https://github.com/pyt-team/TopoNetX.git",
+    "topomodelx @ git+https://github.com/pyt-team/TopoModelX.git",
+    "topoembedx @ git+https://github.com/pyt-team/TopoEmbedX.git",
     "lightning",
 ]
 
@@ -73,6 +73,7 @@ test = [
     "coverage",
     "jupyter",
     "mypy",
+    "pytest-mock"
 ]
 
 dev = ["TopoBenchmarkX[test, lint]"]

test/data/dataload/test_Dataloaders.py

@@ -3,24 +3,26 @@
 import hydra
 import rootutils
 import torch
-from hydra import compose, initialize
+from hydra import compose, initialize_config_dir
 
 from topobenchmarkx.data.preprocess.preprocessor import PreProcessor
 from topobenchmarkx.dataloader import TBXDataloader
 from topobenchmarkx.dataloader.utils import to_data_list
 
+from topobenchmarkx.run import initialize_hydra
+
 rootutils.setup_root("./", indicator=".project-root", pythonpath=True)
 
 
 class TestCollateFunction:
     """Test collate_fn."""
 
     def setup_method(self):
-        initialize(
-            version_base="1.3", config_path="../../../configs", job_name="job"
-        )
-        cfg = compose(config_name="run.yaml", overrides=["dataset=graph/ZINC"])
-
+        # initialize_config_dir(
+        #     version_base="1.3", config_dir=str(rootutils.find_root() / "configs"), job_name="job"
+        # )
+        # cfg = compose(config_name="test.yaml", overrides=["dataset=graph/ZINC", "model=simplicial/sccn", "transforms=dataset_defaults/ZINC"])
+        cfg = initialize_hydra()
         graph_loader = hydra.utils.instantiate(cfg.dataset, _recursive_=False)
 
         datasets, dataset_dir = graph_loader.loader.load()

test/data/preprocess/test_preprocessor.py

@@ -13,11 +13,12 @@
 class TestPreProcessor:
 
     @pytest.fixture(autouse=True)
-    def setup_method(self, mocker):
+    def setup_method(self, mocker_fixture):
+        mocker = mocker_fixture
 
         # Setup test parameters
         self.dataset = MagicMock(spec=torch_geometric.data.Dataset)
-        self.data_dir = "/fake/path"
+        self.data_dir = "fake/path"
         self.transforms_config = DictConfig(
             {"transform": {"transform_name": "CellCycleLifting"}}
         )
@@ -61,7 +62,8 @@ def test_init(self):
         assert self.preprocessor.transforms_applied == False
         assert self.preprocessor.data_list == ["0", "0", "0"]
 
-    def test_init_with_transform(self, mocker):
+    def test_init_with_transform(self, mocker_fixture):
+        mocker = mocker_fixture
         val_processed_paths = ["/some/path"]
         params = [
             {"assert_args": ("created_property", "processed_data_dir")},

topobenchmarkx/__init__.py

@@ -1,3 +1,5 @@
+"""TopobenchmarkX: A library for benchmarking of topological models."""
+
 # Import submodules
 from . import (
     data,
@@ -10,6 +12,7 @@
     transforms,
     utils,
 )
+from .run import initialize_hydra
 
 __all__ = [
     "data",
@@ -21,6 +24,7 @@
     "dataloader",
     "datasets",
     "model",
+    "initialize_hydra",
 ]
 
 

topobenchmarkx/__main__.py

@@ -1,3 +1,5 @@
+"""TopoBenchmarkX.__main__ module."""
+
 from .run import main
 
 if __name__ == "__main__":

topobenchmarkx/data/load/loaders.py

@@ -164,8 +164,7 @@ def __init__(self, parameters: DictConfig):
     def load(
         self,
     ) -> torch_geometric.data.Dataset:
-        """
-        Load cell complex dataset.
+        """Load cell complex dataset.
 
         Returns
         -------
@@ -196,8 +195,7 @@ def __init__(self, parameters: DictConfig):
     def load(
         self,
     ) -> torch_geometric.data.Dataset:
-        """
-        Load simplicial dataset.
+        """Load simplicial dataset.
 
         Returns
         -------

topobenchmarkx/data/preprocess/preprocessor.py

@@ -18,8 +18,7 @@
 
 
 class PreProcessor(torch_geometric.data.InMemoryDataset):
-    """
-    Preprocessor for datasets.
+    """Preprocessor for datasets.
 
     Parameters
     ----------
@@ -63,8 +62,7 @@ def __init__(self, dataset, data_dir, transforms_config=None, **kwargs):
 
     @property
     def processed_dir(self) -> str:
-        """
-        Return the path to the processed directory.
+        """Return the path to the processed directory.
 
         Returns
         -------
@@ -78,8 +76,7 @@ def processed_dir(self) -> str:
 
     @property
     def processed_file_names(self) -> str:
-        """
-        Return the name of the processed file.
+        """Return the name of the processed file.
 
         Returns
         -------
@@ -91,8 +88,7 @@ def processed_file_names(self) -> str:
     def instantiate_pre_transform(
         self, data_dir, transforms_config
     ) -> torch_geometric.transforms.Compose:
-        """
-        Instantiate the pre-transforms.
+        """Instantiate the pre-transforms.
 
         Parameters
         ----------
@@ -122,8 +118,7 @@ def instantiate_pre_transform(
     def set_processed_data_dir(
         self, pre_transforms_dict, data_dir, transforms_config
     ) -> None:
-        """
-        Set the processed data directory.
+        """Set the processed data directory.
 
         Parameters
         ----------
@@ -147,9 +142,7 @@ def set_processed_data_dir(
         )
 
     def save_transform_parameters(self) -> None:
-        """
-        Save the transform parameters.
-        """
+        """Save the transform parameters."""
         # Check if root/params_dict.json exists, if not, save it
         path_transform_parameters = os.path.join(
             self.processed_data_dir, "path_transform_parameters_dict.json"
@@ -172,9 +165,7 @@ def save_transform_parameters(self) -> None:
             )
 
     def process(self) -> None:
-        """
-        Method that processes the data.
-        """
+        """Method that processes the data."""
         self.data_list = (
             [self.pre_transform(d) for d in self.data_list]
             if self.pre_transform is not None
@@ -192,8 +183,7 @@ def load_dataset_splits(
     ) -> tuple[
         DataloadDataset, DataloadDataset | None, DataloadDataset | None
     ]:
-        """
-        Load the dataset splits.
+        """Load the dataset splits.
 
         Parameters
         ----------

topobenchmarkx/dataloader/__init__.py

@@ -1,3 +1,5 @@
+"""This module implements the dataloader for the topobenchmarkx package."""
+
 from .dataload_dataset import DataloadDataset
 from .dataloader import TBXDataloader
 

topobenchmarkx/dataloader/dataload_dataset.py

@@ -1,11 +1,15 @@
+"""Dataset class compatible with TBXDataloader."""
+
 import torch_geometric
 
 
 class DataloadDataset(torch_geometric.data.Dataset):
-    r"""Custom dataset to return all the values added to the dataset object.
+    """Custom dataset to return all the values added to the dataset object.
 
-    Args:
-        data_lst (list[torch_geometric.data.Data]): List of torch_geometric.data.Data objects.
+    Parameters
+    ----------
+    data_lst : list[torch_geometric.data.Data]
+        List of torch_geometric.data.Data objects.
     """
 
     def __init__(self, data_lst):
@@ -16,22 +20,28 @@ def __repr__(self):
         return f"{self.__class__.__name__}({len(self.data_lst)})"
 
     def get(self, idx):
-        r"""Get data object from data list.
+        """Get data object from data list.
 
-        Args:
-            idx (int): Index of the data object to get.
+        Parameters
+        ----------
+        idx : int
+            Index of the data object to get.
 
-        Returns:
-            tuple: tuple containing a list of all the values for the data and the corresponding keys.
+        Returns
+        -------
+        tuple
+            Tuple containing a list of all the values for the data and the corresponding keys.
         """
         data = self.data_lst[idx]
         keys = list(data.keys())
         return ([data[key] for key in keys], keys)
 
     def len(self):
-        r"""Return the length of the dataset.
+        """Return the length of the dataset.
 
-        Returns:
-            int: Length of the dataset.
+        Returns
+        -------
+        int
+            Length of the dataset.
         """
         return len(self.data_lst)