added mp.pool in _generate_samples (#150)

# Pull Request Template

## Description

added `n_jobs` (`1` by default) to _generate_samples, leveraging mp.pool
to parallelize generation of bootstraps.
closes #149 

## Type of change

Please delete options that are not relevant.

- [x ] New feature (non-breaking change which adds functionality)
- [ x] This change requires a documentation update

## Checklist:

- [x ] My code follows the style guidelines of this project
- [ x] I have performed a self-review of my own code
- [ x] I have commented my code, particularly in hard-to-understand
areas
- [ x] I have made corresponding changes to the documentation

src/tsbootstrap/base_bootstrap.py

@@ -2,6 +2,7 @@
 
 import inspect
 from collections.abc import Callable
+from multiprocessing import Pool
 from numbers import Integral
 from typing import Optional
 
@@ -153,33 +154,59 @@ def _generate_samples(
         X: np.ndarray,
         return_indices: bool = False,
         y=None,
+        n_jobs: int = 1,
     ):
         """Generate bootstrapped samples directly.
 
         Parameters
         ----------
-        X : array-like of shape (n_samples, n_features)
+        X : array-like of shape (n_timepoints, n_features)
             The input samples.
+        return_indices : bool, default=False
+            If True, a second output is retured, integer locations of
+            index references for the bootstrap sample, in reference to original indices.
+            Indexed values do are not necessarily identical with bootstrapped values.
+        y : array-like of shape (n_timepoints, n_features_exog), default=None
+            Exogenous time series to use in bootstrapping.
+        n_jobs : int, default=1
+            The number of jobs to run in parallel.
 
         Yields
         ------
         Iterator[np.ndarray]
             An iterator over the bootstrapped samples.
-
         """
-        for _ in range(self.config.n_bootstraps):
-            indices, data = self._generate_samples_single_bootstrap(X=X, y=y)
-            data = np.concatenate(data, axis=0)
-
-            # hack to fix known issue with non-concatenated index sets
-            # see bug issue #81
-            if isinstance(indices, list):
-                indices = np.concatenate(indices, axis=0)
-
-            if return_indices:
-                yield data, indices  # type: ignore
-            else:
-                yield data
+        if n_jobs == 1:
+            # Run bootstrap generation sequentially in the main process
+            for _ in range(self.config.n_bootstraps):
+                indices, data = self._generate_samples_single_bootstrap(X, y)
+                data = np.concatenate(data, axis=0)
+                if return_indices:
+                    # hack to fix known issue with non-concatenated index sets
+                    # see bug issue #81
+                    if isinstance(indices, list):
+                        indices = np.concatenate(indices, axis=0)
+                    yield data, indices
+                else:
+                    yield data
+        else:
+            # Use multiprocessing to handle bootstrapping
+            args = [(X, y) for _ in range(self.config.n_bootstraps)]
+            with Pool(n_jobs) as pool:
+                results = pool.starmap(
+                    self._generate_samples_single_bootstrap, args
+                )
+
+            for indices, data in results:
+                data = np.concatenate(data, axis=0)
+                if return_indices:
+                    # hack to fix known issue with non-concatenated index sets
+                    # see bug issue #81
+                    if isinstance(indices, list):
+                        indices = np.concatenate(indices, axis=0)
+                    yield data, indices
+                else:
+                    yield data
 
     def _generate_samples_single_bootstrap(self, X: np.ndarray, y=None):
         """Generate list of bootstraps for a single bootstrap iteration."""

src/tsbootstrap/block_bootstrap.py

@@ -124,7 +124,7 @@ def _generate_blocks(self, X: np.ndarray):
 
         Parameters
         ----------
-        X : array-like of shape (n_samples, n_features)
+        X : array-like of shape (n_timepoints, n_features)
             The input samples.
 
         Returns
@@ -165,7 +165,7 @@ def _generate_samples_single_bootstrap(self, X: np.ndarray, y=None):
 
         Parameters
         ----------
-        X : array-like of shape (n_samples, n_features)
+        X : array-like of shape (n_timepoints, n_features)
             The input samples.
 
         Returns
@@ -295,7 +295,7 @@ def _generate_samples_single_bootstrap(self, X: np.ndarray, y=None):
 
         Parameters
         ----------
-        X : array-like of shape (n_samples, n_features)
+        X : array-like of shape (n_timepoints, n_features)
             The input samples.
 
         Returns

src/tsbootstrap/tests/test_all_bootstraps.py

@@ -131,9 +131,9 @@ def test_bootstrap_input_output_contract(self, object_instance, scenario):
             )
 
         if not all(bs.ndim == 2 for bs in bss):
+            print([bs.shape for bs in bss])
             raise ValueError(
-                f"{cls_name}.bootstrap yielded arrays with unexpected number of "
-                "dimensions. All bootstrap samples should have 2 dimensions."
+                f"{cls_name}.bootstrap yielded arrays with unexpected number of dimensions. All bootstrap samples should have 2 dimensions."
             )
 
         if not all(bs.shape[0] == n_timepoints for bs in bss):
@@ -218,9 +218,9 @@ def test_bootstrap_test_ratio(self, object_instance, scenario, test_ratio):
             )
 
         if not all(bs.ndim == 2 for bs in bss):
+            print([bs.shape for bs in bss])
             raise ValueError(
-                f"{cls_name}.bootstrap yielded arrays with unexpected number of "
-                "dimensions. All bootstrap samples should have 2 dimensions."
+                f"{cls_name}.bootstrap yielded arrays with unexpected number of dimensions. All bootstrap samples should have 2 dimensions."
             )
 
         if not all(bs.shape[0] == expected_length for bs in bss):

src/tsbootstrap/tsfit.py

@@ -234,11 +234,13 @@ def set_params(self, **params):
                 self.model_params[key] = value
         return self
 
+    '''
     def __repr__(self):
         """
         Official string representation of a TSFit object.
         """
         return f"TSFit(order={self.order}, model_type='{self.model_type}')"
+    '''
 
     def fit(self, X: np.ndarray, y=None) -> TSFit:
         """
@@ -247,7 +249,7 @@ def fit(self, X: np.ndarray, y=None) -> TSFit:
         Parameters
         ----------
         X : np.ndarray
-            Input data of shape (n_samples, n_features).
+            Input data of shape (n_timepoints, n_features).
         y : np.ndarray, optional
             Exogenous variables, by default None.
 
@@ -566,7 +568,7 @@ def predict(self, X: np.ndarray, y=None, n_steps: int = 1) -> np.ndarray:
         Parameters
         ----------
         X : np.ndarray
-            Input data of shape (n_samples, n_features).
+            Input data of shape (n_timepoints, n_features).
         y : np.ndarray, optional
             Exogenous variables, by default None.
         n_steps : int, optional