Added glm.py examples in docstrings

src/nemos/basis.py

@@ -204,6 +204,7 @@ def fit(self, X: FeatureMatrix, y=None):
         -------
         self :
             The transformer object.
+
         """
         self._basis._set_kernel(*self._unpack_inputs(X))
         return self

src/nemos/glm.py

@@ -104,6 +104,25 @@ class GLM(BaseRegressor):
     TypeError
         If provided `regularizer` or `observation_model` are not valid.
 
+    Examples
+    --------
+    >>> import nemos as nmo
+
+    >>> # define simple GLM model
+    >>> model = nmo.glm.GLM()
+    >>> print("Regularizer type: ", type(model.regularizer))
+    Regularizer type:  <class 'nemos.regularizer.UnRegularized'>
+    >>> print("Observation model: ", type(model.observation_model))
+    Observation model:  <class 'nemos.observation_models.PoissonObservations'>
+
+
+    >>> # define GLM model of PoissonObservations model with soft-plus NL
+    >>> observation_models = nmo.observation_models.PoissonObservations(jax.nn.softplus)
+    >>> model = nmo.glm.GLM(observation_model=observation_models, solver_name="LBFGS")
+    >>> print("Regularizer type: ", type(model.regularizer))
+    Regularizer type:  <class 'nemos.regularizer.UnRegularized'>
+    >>> print("Observation model: ", type(model.observation_model))
+    Observation model:  <class 'nemos.observation_models.PoissonObservations'>
     """
 
     def __init__(
@@ -307,6 +326,21 @@ def predict(self, X: DESIGN_INPUT_TYPE) -> jnp.ndarray:
             - If `X` is not three-dimensional.
             - If there's an inconsistent number of features between spike basis coefficients and `X`.
 
+        Examples
+        --------
+        >>> # example input
+        >>> import numpy as np
+        >>> X, y = np.random.normal(size=(10, 2)), np.random.uniform(size=10)
+        >>> Xnew = np.random.normal(size=(20, ) + X.shape[1:])
+
+        >>> # define and fit a GLM
+        >>> import nemos as nmo
+        >>> model = nmo.glm.GLM()
+        >>> model = model.fit(X, y)
+
+        >>> # predict new spike data
+        >>> predicted_spikes = model.predict(Xnew)
+
         See Also
         --------
         - [score](./#nemos.glm.GLM.score)
@@ -407,6 +441,20 @@ def score(
             If X structure doesn't match the params, and if X and y have different
             number of samples.
 
+        Examples
+        --------
+        >>> # example input
+        >>> import numpy as np
+        >>> X, y = np.random.normal(size=(10, 2)), np.random.uniform(size=10)
+
+        >>> import nemos as nmo
+        >>> model = nmo.glm.GLM()
+        >>> model = model.fit(X, y)
+
+        >>> # get model score
+        >>> log_likelihood_score = model.score(X, y)
+        >>> pseudo_r2_score = model.score(X, y, score_type='pseudo-r2-McFadden')
+
         Notes
         -----
         The log-likelihood is not on a standard scale, its value is influenced by many factors,
@@ -605,6 +653,20 @@ def fit(
             - If `init_params` are not array-like
             - If `init_params[i]` cannot be converted to jnp.ndarray for all i
 
+        Examples
+        -------
+        >>> # example input
+        >>> import numpy as np
+        >>> X, y = np.random.normal(size=(10, 2)), np.random.uniform(size=10)
+
+        >>> # fit a ridge regression Poisson GLM
+        >>> import nemos as nmo
+        >>> model = nmo.glm.GLM(regularizer="Ridge", regularizer_strength=0.1)
+        >>> model = model.fit(X, y)
+
+        >>> # get model weights
+        >>> model_weights = model.coef_
+
         """
         # validate the inputs & initialize solver
         init_params = self.initialize_params(X, y, init_params=init_params)
@@ -686,7 +748,7 @@ def simulate(
         Returns
         -------
         simulated_activity :
-            Simulated activity (spike counts for PoissonGLMs) for the neuron over time.
+            Simulated activity (spike counts for Poisson GLMs) for the neuron over time.
             Shape: (n_time_bins, ).
         firing_rates :
             Simulated rates for the neuron over time. Shape, (n_time_bins, ).
@@ -698,6 +760,21 @@ def simulate(
         ValueError
             - If the instance has not been previously fitted.
 
+        Examples
+        --------
+        >>> # example input
+        >>> import numpy as np
+        >>> X, y = np.random.normal(size=(10, 2)), np.random.uniform(size=10)
+        >>> Xnew = np.random.normal(size=(20, ) + X.shape[1:])
+
+        >>> # define and fit model
+        >>> import nemos as nmo
+        >>> model = nmo.glm.GLM()
+        >>> model = model.fit(X, y)
+
+        >>> # generate spikes and rates
+        >>> random_key = jax.random.key(123)
+        >>> spikes, rates = model.simulate(random_key, Xnew)
 
         See Also
         --------
@@ -1376,6 +1453,26 @@ def fit(
         - If the mask is a `FeaturePytree`, then `"feature_name"` is a predictor of neuron `j` if
         `feature_mask["feature_name"][j] == 1`.
 
+        Examples
+        --------
+        >>> # Generate sample data
+        >>> import jax.numpy as jnp
+        >>> import numpy as np
+        >>> from nemos.glm import PopulationGLM
+
+        >>> # Define predictors (X), weights, and neural activity (y)
+        >>> num_samples, num_features, num_neurons = 100, 3, 2
+        >>> X = np.random.normal(size=(num_samples, num_features))
+        >>> weights = np.array([[ 0.5,  0. ], [-0.5, -0.5], [ 0. ,  1. ]])
+        >>> y = np.random.poisson(np.exp(X.dot(weights)))
+
+        >>> # Define a feature mask, shape (num_features, num_neurons)
+        >>> feature_mask = jnp.array([[1, 0], [1, 1], [0, 1]])
+
+        >>> # Create and fit the model
+        >>> model = PopulationGLM(feature_mask=feature_mask).fit(X, y)
+        >>> print(model.coef_.shape)
+        (3, 2)
         """
         return super().fit(X, y, init_params)