Docstring fixes

.github/workflows/python-package-conda.yml

@@ -14,24 +14,25 @@ jobs:
     - uses: actions/checkout@v2
     - uses: mamba-org/setup-micromamba@v1
       with:
-        environment-file: environment.yml
+        environment-file: environment_torch.yml
         activate-environment: test
     - shell: bash -l {0}
       run: |
-        pip install --upgrade keras
         conda info
         conda list
         conda config --show-sources
         conda config --show
         printenv | sort
-    - name: Lint with flake8
+    - name: Lint with ruff
       shell: bash -l {0}
       run: |
-        micromamba install flake8
+        micromamba install ruff
         # stop the build if there are Python syntax errors or undefined names
-        flake8 . --count --select=E9,F63,F7,F82 --show-source --statistics
+        ruff . --count --select=E9,F63,F7,F82 --show-source --statistics
         # exit-zero treats all errors as warnings. The GitHub editor is 127 chars wide
-        flake8 . --count --exit-zero --max-complexity=100 --max-line-length=127 --statistics
+        ruff . --count --exit-zero --max-complexity=100 --max-line-length=127 --statistics
+        # Checking documentation errors
+        ruff . --count --select=D --exit-zero --max-complexity=100 --max-line-length=127 --statistics
     - name: Test with pytest
       shell: bash -l {0}
       run: |

mlguess/keras/data.py

@@ -86,8 +86,7 @@ def preprocess_data(
     groups=[],
     seed=1000,
 ):
-    """
-    Function to select features and scale data for ML
+    """Function to select features and scale data for ML
     Args:
         data (dictionary of dataframes for training and validation data):
         input_features (list): Input features

mlguess/keras/deprecated/losses.py

@@ -15,8 +15,8 @@
 
 
 class DirichletEvidentialLoss(keras.losses.Loss):
-    """
-    Loss function for an evidential categorical model.
+    """Loss function for an evidential categorical model.
+
     Args:
         callback (list): List of callbacks.
         name (str): reference name
@@ -76,8 +76,7 @@ def __call__(self, y, output, sample_weight=None):
         return ops.mean(A + B + C)
 
 class EvidentialRegressionLoss(keras.losses.Loss):
-    """
-    Loss function for an evidential regression model. The total loss is the Negative Log Likelihood of the
+    """Loss function for an evidential regression model. The total loss is the Negative Log Likelihood of the
     Normal Inverse Gamma summed with the error and scaled by the evidential coefficient. The coefficient has a strong
     influence on the uncertainty predictions (less so for the predictions themselves) of the model and must be tuned
     for individual datasets.
@@ -120,8 +119,8 @@ def get_config(self):
         return config
 
 def gaussian_nll(y, y_pred, reduce=True):
-    """
-    Loss function for a parametric Gaussian Loss.
+    """Loss function for a parametric Gaussian Loss.
+
     Args:
         y: Training data targets
         y_pred: Model predicitons
@@ -139,8 +138,7 @@ def gaussian_nll(y, y_pred, reduce=True):
 
 class EvidentialRegressionCoupledLoss(keras.losses.Loss):
     def __init__(self, r=1.0, coeff=1.0):
-        """
-        implementation of the loss from meinert and lavin that fixes issues with the original
+        """Implementation of the loss from meinert and lavin that fixes issues with the original
         evidential loss for regression. The loss couples the virtual evidence values with coefficient r.
         In this new loss, the regularizer is unnecessary.
         """

mlguess/keras/deprecated/models.py

@@ -19,8 +19,8 @@
 import logging
 
 class BaseRegressor(object):
-    """
-    A base class for regression models.
+    """A base class for regression models.
+
     Attributes:
         hidden_layers: Number of hidden layers
         hidden_neurons: Number of neurons in each hidden layer
@@ -97,8 +97,7 @@ def __init__(
         self.history = None
 
     def build_neural_network(self, inputs, outputs, last_layer="Dense"):
-        """
-        Create Keras neural network model and compile it.
+        """Create Keras neural network model and compile it.
 
         Args:
             inputs (int): Number of input predictor variables.
@@ -169,8 +168,7 @@ def build_neural_network(self, inputs, outputs, last_layer="Dense"):
         )
 
     def build_from_sequential(self, model, optimizer="adam", loss="mse", metrics=None):
-        """
-        Build the neural network model using a Keras Sequential model.
+        """Build the neural network model using a Keras Sequential model.
 
         Args:
             model (tf.keras.Sequential): Keras Sequential model to use.
@@ -204,8 +202,8 @@ def fit(
         shuffle=True,
         **kwargs,
     ):
-        """
-        Fit the regression model.
+        """Fit the regression model.
+
         Args:
             x: Input data
             y: Target data
@@ -217,7 +215,6 @@ def fit(
             use_multiprocessing: If True, use ProcessPoolExecutor to load data, which is faster but can cause issues with certain GPU setups. If False, use a ThreadPoolExecutor.
             **kwargs: Additional arguments to be passed to the `fit` method
         """
-
         if self.model is None:
             raise ValueError("Model has not been built. Call build_neural_network first.")
         if self.verbose:
@@ -240,8 +237,7 @@ def fit(
         )
 
     def save_model(self):
-        """
-        Save the trained model to a file.
+        """Save the trained model to a file.
         """
         if not os.path.exists(self.save_path):
             os.makedirs(self.save_path)
@@ -259,8 +255,7 @@ def save_model(self):
 
     @classmethod
     def load_model(cls, conf):
-        """
-        Load a trained model using args from a configuration
+        """Load a trained model using args from a configuration
         """
         # Check if weights file exists
         weights = os.path.join(conf["model"]["save_path"], "best.h5")
@@ -306,7 +301,7 @@ def load_model(cls, conf):
         return model_class
 
     def mae(self, y_true, y_pred):
-        """ Compute the MAE """
+        """Compute the MAE"""
         num_splits = y_pred.shape[-1]
         if num_splits == 4:
             mu, _, _, _ = ops.split(y_pred, num_splits, axis=-1)
@@ -317,7 +312,7 @@ def mae(self, y_true, y_pred):
         return keras.metrics.mean_absolute_error(y_true, mu)
 
     def mse(self, y_true, y_pred):
-        """ Compute the MSE """
+        """Compute the MSE"""
         num_splits = y_pred.shape[-1]
         if num_splits == 4:
             mu, _, _, _ = ops.split(y_pred, num_splits, axis=-1)
@@ -329,8 +324,7 @@ def mse(self, y_true, y_pred):
         return keras.metrics.mean_squared_error(y_true, mu)
 
     def predict(self, x, scaler=None, batch_size=None):
-        """
-        Predict target values for input data.
+        """Predict target values for input data.
 
         Args:
             x (numpy.ndarray): Input data.
@@ -350,8 +344,7 @@ def predict(self, x, scaler=None, batch_size=None):
         return y_out
 
     def predict_ensemble(self, x, batch_size=None, scaler=None, num_outputs=1):
-        """
-        Predicts outcomes using an ensemble of trained Keras models.
+        """Predicts outcomes using an ensemble of trained Keras models.
 
         Args:
             x (numpy.ndarray): Input data for predictions.
@@ -429,8 +422,7 @@ def predict_ensemble(self, x, batch_size=None, scaler=None, num_outputs=1):
         return ensemble_mu, ensemble_ale, ensemble_epi
 
     def predict_monte_carlo(self, x_test, forward_passes, scaler=None, batch_size=None, num_outputs=1):
-        """
-        Perform Monte Carlo dropout predictions for the model.
+        """Perform Monte Carlo dropout predictions for the model.
 
         Args:
             x_test (numpy.ndarray): Input data for prediction.
@@ -442,7 +434,6 @@ def predict_monte_carlo(self, x_test, forward_passes, scaler=None, batch_size=No
         Returns:
             tuple: Tuple of arrays containing predicted target values and specified uncertainties.
         """
-
         n_samples = x_test.shape[0]
         pred_size = self.model.output_shape[-1]
         _batch_size = self.batch_size if batch_size is None else batch_size
@@ -531,8 +522,7 @@ def __init__(
 
 
 class GaussianRegressorDNN(BaseRegressor):
-    """
-    A Dense Neural Network Model that can support arbitrary numbers of hidden layers
+    """A Dense Neural Network Model that can support arbitrary numbers of hidden layers
     and provides evidential uncertainty estimation.
     Inherits from BaseRegressor.
 
@@ -581,8 +571,7 @@ def __init__(
         metrics=None,
         eps=1e-7
     ):
-        """
-        Initialize the EvidentialRegressorDNN.
+        """Initialize the EvidentialRegressorDNN.
 
         Args:
             coupling_coef: Coupling coeffient for loss fix
@@ -617,8 +606,7 @@ def __init__(
         self.loss = gaussian_nll
 
     def build_neural_network(self, inputs, outputs, last_layer="DenseNormal"):
-        """
-        Create Keras neural network model and compile it.
+        """Create Keras neural network model and compile it.
 
         Args:
             inputs (int): Number of input predictor variables.
@@ -714,8 +702,7 @@ def predict_monte_carlo(self, x_test, forward_passes, scaler=None, batch_size=No
 
 
 class EvidentialRegressorDNN(BaseRegressor):
-    """
-    A Dense Neural Network Model that can support arbitrary numbers of hidden layers
+    """A Dense Neural Network Model that can support arbitrary numbers of hidden layers
     and provides evidential uncertainty estimation.
     Inherits from BaseRegressor.
 
@@ -765,8 +752,7 @@ def __init__(
         metrics=None,
         eps=1e-7
     ):
-        """
-        Initialize the EvidentialRegressorDNN.
+        """Initialize the EvidentialRegressorDNN.
 
         Args:
             coupling_coef: Coupling coeffient for loss fix
@@ -817,8 +803,7 @@ def __init__(
         logging.info(f"Using loss: {loss}")
 
     def build_neural_network(self, inputs, outputs):
-        """
-        Create Keras neural network model and compile it.
+        """Create Keras neural network model and compile it.
 
         Args:
             inputs (int): Number of input predictor variables.
@@ -932,8 +917,8 @@ def predict_monte_carlo(
 
 
 class CategoricalDNN(object):
-    """
-    A Dense Neural Network Model that can support arbitrary numbers of hidden layers.
+    """A Dense Neural Network Model that can support arbitrary numbers of hidden layers.
+
     Attributes:
         hidden_layers: Number of hidden layers
         hidden_neurons: Number of neurons in each hidden layer
@@ -1022,8 +1007,8 @@ def __init__(
         self.steps_per_epoch = steps_per_epoch
 
     def build_neural_network(self, inputs, outputs):
-        """
-        Create Keras neural network model and compile it.
+        """Create Keras neural network model and compile it.
+
         Args:
             inputs (int): Number of input predictor variables
             outputs (int): Number of output predictor variables
@@ -1081,8 +1066,7 @@ def build_neural_network(self, inputs, outputs):
         self.model.compile(optimizer=self.optimizer_obj, loss=self.loss)
 
     def build_from_sequential(self, model, optimizer="adam", loss="mse", metrics=None):
-        """
-        Build the neural network model using a Keras Sequential model.
+        """Build the neural network model using a Keras Sequential model.
 
         Args:
             model (tf.keras.Sequential): Keras Sequential model to use.

mlguess/keras/layers.py

@@ -7,8 +7,7 @@
 
 @keras.saving.register_keras_serializable()
 class DenseNormalGamma(layers.Layer):
-    """
-    Implements dense output layer for a deep evidential regression model.
+    """Implements dense output layer for a deep evidential regression model.
     Reference: https://www.mit.edu/~amini/pubs/pdf/deep-evidential-regression.pdf
     Source: https://github.com/aamini/evidential-deep-learning
 
@@ -36,8 +35,7 @@ def __init__(self, units: int,
         self.eps = eps
 
     def evidence(self, x):
-        """
-        Converts values from continuous space to greater than 0 using a softplus activation function.
+        """Converts values from continuous space to greater than 0 using a softplus activation function.
 
         Args:
             x: input value
@@ -65,8 +63,7 @@ def get_config(self):
 
 
 class DenseNormal(layers.Layer):
-    """
-    Dense output layer for a Gaussian distribution regression neural network.
+    """Dense output layer for a Gaussian distribution regression neural network.
 
     Args:
         units (int): Output size of regression tasks

mlguess/keras/losses.py

@@ -57,8 +57,7 @@ def loss(y, y_pred):
 
 @keras.saving.register_keras_serializable()
 def evidential_reg_loss(evi_coef):
-    """
-    Loss function for an evidential regression model. The total loss is the Negative Log Likelihood of the
+    """Loss function for an evidential regression model. The total loss is the Negative Log Likelihood of the
     Normal Inverse Gamma summed with the error and scaled by the evidential coefficient. The coefficient has a strong
     influence on the uncertainty predictions (less so for the predictions themselves) of the model and must be tuned
     for individual datasets.
@@ -97,8 +96,8 @@ def loss(y, y_pred):
 
 @keras.saving.register_keras_serializable()
 def gaussian_nll(y, y_pred, reduce=True):
-    """
-    Loss function for a parametric Gaussian Loss.
+    """Loss function for a parametric Gaussian Loss.
+
     Args:
         y: Training data targets
         y_pred: Model predicitons
@@ -116,8 +115,7 @@ def gaussian_nll(y, y_pred, reduce=True):
 
 class EvidentialRegressionCoupledLoss(keras.losses.Loss):
     def __init__(self, r=1.0, coeff=1.0):
-        """
-        implementation of the loss from meinert and lavin that fixes issues with the original
+        """Implementation of the loss from meinert and lavin that fixes issues with the original
         evidential loss for regression. The loss couples the virtual evidence values with coefficient r.
         In this new loss, the regularizer is unnecessary.
         """