Merge branch 'refactor/rename_base_model_module' into 'dev'

Rename base model module

See merge request cdd/QSPRpred!181

CHANGELOG.md

@@ -18,6 +18,8 @@ From v3.0.1 to v3.0.2
 
 ## Changes
 
+- The module containing the sole model base class (`QSPRModel`) was renamed
+  from `models` to `model`.
 - Restrictions on `numpy` versions were removed to allow for more flexibility in
   package installations. However, the `BorutaFilter` feature selection method does not
   function with `numpy` versions 1.24.0 and above. Therefore, this functionality now

docs/features.rst

@@ -122,7 +122,7 @@ Overview of available features
 
     .. dropdown:: Models
 
-        :class:`~qsprpred.models.models.QSPRModel`: Base class for models.
+        :class:`~qsprpred.models.model.QSPRModel`: Base class for models.
 
         Models are used to predict properties of molecules.
         A general example can be found in the `quick start tutorial <https://github.com/CDDLeiden/QSPRpred/blob/main/tutorials/quick_start.ipynb>`_.

qsprpred/benchmarks/replica.py

@@ -12,7 +12,7 @@
 from ..data.sources.data_source import DataSource
 from ..logs import logger
 from ..models.hyperparam_optimization import HyperparameterOptimization
-from ..models.models import QSPRModel
+from ..models.model import QSPRModel
 from ..models.monitors import NullMonitor
 from ..tasks import TargetProperty
 from ..utils.serialization import JSONSerializable

qsprpred/benchmarks/settings/benchmark.py

@@ -6,7 +6,7 @@
 from ...data.descriptors.sets import DescriptorSet
 from ...data.sources.data_source import DataSource
 from ...models.hyperparam_optimization import HyperparameterOptimization
-from ...models.models import QSPRModel
+from ...models.model import QSPRModel
 from ...tasks import TargetProperty
 from ...utils.serialization import JSONSerializable
 

qsprpred/data/descriptors/sets.py

@@ -529,7 +529,7 @@ def __init__(self, model: Type["QSPRModel"] | str):
         """
         super().__init__()
         if isinstance(model, str):
-            from ...models.models import QSPRModel
+            from ...models.model import QSPRModel
 
             self.model = QSPRModel.fromFile(model)
         else:
@@ -544,7 +544,7 @@ def __getstate__(self):
 
     def __setstate__(self, state):
         super().__setstate__(state)
-        from ...models.models import QSPRModel
+        from ...models.model import QSPRModel
 
         self.model = QSPRModel.fromFile(self.model)
 

qsprpred/extra/gpu/models/chemprop.py

@@ -18,7 +18,7 @@
 from ....data.tables.qspr import QSPRDataset
 from ....logs import logger
 from ....models.early_stopping import EarlyStoppingMode, early_stopping
-from ....models.models import QSPRModel
+from ....models.model import QSPRModel
 from ....models.monitors import BaseMonitor, FitMonitor
 
 

qsprpred/extra/gpu/models/dnn.py

@@ -16,7 +16,7 @@
 from ....data.tables.qspr import QSPRDataset
 from ....extra.gpu.models.neural_network import STFullyConnected
 from ....models.early_stopping import EarlyStoppingMode, early_stopping
-from ....models.models import QSPRModel
+from ....models.model import QSPRModel
 from ....models.monitors import BaseMonitor, FitMonitor
 
 

qsprpred/extra/gpu/models/pyboost.py

@@ -24,7 +24,7 @@
 from ....data.sampling.splits import DataSplit
 from ....data.tables.qspr import QSPRDataset
 from ....models.early_stopping import EarlyStoppingMode, early_stopping
-from ....models.models import QSPRModel
+from ....models.model import QSPRModel
 from ....models.monitors import BaseMonitor, FitMonitor
 from ....tasks import ModelTasks
 

qsprpred/extra/models/pcm.py

@@ -6,14 +6,14 @@
 from typing import Callable
 
 import numpy as np
+from rdkit import Chem
+from rdkit.Chem import Mol
 
 from qsprpred.extra.data.tables.pcm import PCMDataSet
 from ..data.descriptors.sets import ProteinDescriptorSet
 from ...data.tables.mol import MoleculeTable
-from ...models.models import QSPRModel
+from ...models.model import QSPRModel
 from ...models.scikit_learn import SklearnModel
-from rdkit import Chem
-from rdkit.Chem import Mol
 
 
 class PCMModel(QSPRModel, ABC):

qsprpred/extra/models/random.py

@@ -11,7 +11,7 @@
 from qsprpred.models.monitors import FitMonitor
 
 from ...data.tables.qspr import QSPRDataset
-from ...models.models import QSPRModel
+from ...models.model import QSPRModel
 
 
 class RandomDistributionAlgorithm(ABC):

qsprpred/models/__init__.py

@@ -10,7 +10,7 @@
     HyperparameterOptimization,
 )
 from .metrics import SklearnMetrics
-from .models import QSPRModel
+from .model import QSPRModel
 
 from .monitors import (
     BaseMonitor,

qsprpred/models/assessment/methods.py

@@ -13,7 +13,7 @@
 from qsprpred.logs import logger
 from qsprpred.models.early_stopping import EarlyStoppingMode
 from qsprpred.models.metrics import SklearnMetrics
-from qsprpred.models.models import QSPRModel
+from qsprpred.models.model import QSPRModel
 from qsprpred.models.monitors import AssessorMonitor, BaseMonitor
 
 

qsprpred/models/hyperparam_optimization.py

@@ -11,7 +11,7 @@
 from qsprpred.models.assessment.methods import ModelAssessor
 from ..data import QSPRDataset
 from ..logs import logger
-from ..models.models import QSPRModel
+from ..models.model import QSPRModel
 from ..models.monitors import BaseMonitor, HyperparameterOptimizationMonitor
 
 

qsprpred/models/monitors.py

@@ -9,7 +9,7 @@
 from rdkit import Chem
 from rdkit.Chem import Draw
 
-from .models import QSPRModel
+from .model import QSPRModel
 from ..data.tables.qspr import QSPRDataset
 from ..utils.serialization import JSONSerializable
 

qsprpred/models/scikit_learn.py

@@ -14,7 +14,7 @@
 from sklearn.exceptions import NotFittedError
 from sklearn.utils.validation import check_is_fitted
 
-from .models import QSPRModel
+from .model import QSPRModel
 from ..data.tables.qspr import QSPRDataset
 from ..logs import logger
 from ..tasks import ModelTasks

qsprpred/plotting/base_plot.py

@@ -3,7 +3,7 @@
 from abc import ABC, abstractmethod
 from typing import Any
 
-from ..models.models import QSPRModel
+from ..models.model import QSPRModel
 
 
 class ModelPlot(ABC):
@@ -21,6 +21,7 @@ class ModelPlot(ABC):
         indPaths (dict[QSPRModel, str]):
             dictionary of models mapped to their independent test set results paths
     """
+
     def __init__(self, models: list[QSPRModel]):
         """Initialize the base class for all model plots.
 
@@ -55,13 +56,10 @@ def checkModel(self, model: QSPRModel) -> tuple[str, str]:
         ind_path = f"{self.modelOuts[model]}.ind.tsv"
         if model.task not in self.getSupportedTasks():
             raise ValueError("Unsupported model type: %s" % model.task)
-        if not os.path.exists(
-            model.metaFile
-        ):
+        if not os.path.exists(model.metaFile):
             raise ValueError(
                 "Model output file does not exist: %s. "
-                "Have you evaluated and saved the model, yet?" %
-                model.metaFile
+                "Have you evaluated and saved the model, yet?" % model.metaFile
             )
         if not os.path.exists(cv_path):
             raise ValueError(

qsprpred/plotting/classification.py

@@ -24,7 +24,7 @@
 )
 
 from ..models.metrics import calibration_error
-from ..models.models import QSPRModel
+from ..models.model import QSPRModel
 from ..plotting.base_plot import ModelPlot
 from ..tasks import ModelTasks
 

qsprpred/predict_CLI.py

@@ -11,7 +11,7 @@
 import pandas as pd
 
 from .logs.utils import backup_files, enable_file_logger
-from .models.models import QSPRModel
+from .models.model import QSPRModel
 
 
 def QSPRArgParser(txt=None):
@@ -120,23 +120,28 @@ def QSPR_predict(args):
                     for i in range(predictions[idx].shape[1]):
                         results.update(
                             {
-                                f"preds_{predictor.name}_{target.name}_class_{i}":
-                                    predictions[idx][:, i].flatten()
+                                f"preds_{predictor.name}_{target.name}_class_{i}": predictions[
+                                    idx
+                                ][
+                                    :, i
+                                ].flatten()
                             }
                         )
                 else:
                     for i in range(predictions.shape[1]):
                         results.update(
                             {
-                                f"preds_{predictor.name}_{target.name}_class_{i}":
-                                    predictions[:, i].flatten()
+                                f"preds_{predictor.name}_{target.name}_class_{i}": predictions[
+                                    :, i
+                                ].flatten()
                             }
                         )
             else:
                 results.update(
                     {
-                        f"preds_{predictor.name}_{target.name}":
-                            predictions[:, idx].flatten()
+                        f"preds_{predictor.name}_{target.name}": predictions[
+                            :, idx
+                        ].flatten()
                     }
                 )
 

tutorials/advanced/modelling/custom_models.ipynb

@@ -220,7 +220,7 @@
     "This significantly speeds up the algorithm, so let's continue with the second implementation. \n",
     "\n",
     "Now for the implementation of the `QSPRModel` Gzip KNN, we find the base class for models `QSPRModels` in the `interfaces.py` file of the `models` module.\n",
-    "In the `models.py` file of the `models` module you can also find an example of a model that inherits from the base class `QSPRSklearn`, which uses estimators from the `scikit-learn` package.\n",
+    "In the `model.py` file of the `models` module you can also find an example of a model that inherits from the base class `QSPRSklearn`, which uses estimators from the `scikit-learn` package.\n",
     "\n",
     "There are six abstract methods that need to be implemented from `QSPRModel` class in our new class `GzipKNNModel`:\n",
     "- `fit`, which trains the model\n",