interval callable

Author: MatthewMiddlehurst

tsml/interval_based/_interval_pipelines.py

@@ -35,9 +35,10 @@ class RandomIntervalClassifier(ClassifierMixin, BaseTimeSeriesEstimator):
 
     Parameters
     ----------
-    n_intervals : int, default=100,
+    n_intervals : int or callable, default=100,
         The number of intervals of random length, position and dimension to be
-        extracted.
+        extracted.  Input should be an int or a function that takes a 3D np.ndarray
+        input and returns an int.
     min_interval_length : int, default=3
         The minimum length of extracted intervals. Minimum value of 3.
     max_interval_length : int, default=3
@@ -74,9 +75,9 @@ class RandomIntervalClassifier(ClassifierMixin, BaseTimeSeriesEstimator):
     ----------
     n_instances_ : int
         The number of train cases in the training set.
-    n_channels : int
+    n_channels_ : int
         The number of dimensions per case in the training set.
-    n_timepoints : int
+    n_timepoints_ : int
         The length of each series in the training set.
     n_classes_ : int
         Number of classes. Extracted from the data.
@@ -277,9 +278,10 @@ class RandomIntervalRegressor(RegressorMixin, BaseTimeSeriesEstimator):
 
     Parameters
     ----------
-    n_intervals : int, default=100,
+    n_intervals : int or callable, default=100,
         The number of intervals of random length, position and dimension to be
-        extracted.
+        extracted.  Input should be an int or a function that takes a 3D np.ndarray
+        input and returns an int.
     min_interval_length : int, default=3
         The minimum length of extracted intervals. Minimum value of 3.
     max_interval_length : int, default=3

tsml/interval_based/tests/test_interval_pipelines.py

@@ -0,0 +1,19 @@
+"""Tests for the interval pipeline classes."""
+
+from tsml.interval_based import RandomIntervalClassifier
+from tsml.utils.testing import generate_3d_test_data
+
+
+def test_random_interval_callable():
+    """Test RandomIntervalClassifier with a callable n_intervals."""
+    X, y = generate_3d_test_data()
+
+    def interval_func(X):
+        return int(X.shape[2] / 5)
+
+    est = RandomIntervalClassifier(
+        n_intervals=interval_func,
+    )
+    est.fit(X, y)
+
+    assert est._transformer._n_intervals == 2

tsml/transformations/_interval_extraction.py

@@ -48,9 +48,10 @@ class RandomIntervalTransformer(TransformerMixin, BaseTimeSeriesEstimator):
 
     Parameters
     ----------
-    n_intervals : int, default=100,
+    n_intervals : int or callable, default=100,
         The number of intervals of random length, position and dimension to be
-        extracted.
+        extracted.  Input should be an int or a function that takes a 3D np.ndarray
+        input and returns an int.
     min_interval_length : int, default=3
         The minimum length of extracted intervals. Minimum value of 3.
     max_interval_length : int, default=3
@@ -150,7 +151,7 @@ def fit_transform(self, X, y=None):
                 rng.randint(np.iinfo(np.int32).max),
                 True,
             )
-            for _ in range(self.n_intervals)
+            for _ in range(self._n_intervals)
         )
 
         (
@@ -176,7 +177,7 @@ def fit_transform(self, X, y=None):
                 removed_idx.append(i)
 
         Xt = transformed_intervals[0]
-        for i in range(1, self.n_intervals):
+        for i in range(1, self._n_intervals):
             if i not in removed_idx:
                 Xt = np.hstack((Xt, transformed_intervals[i]))
 
@@ -261,6 +262,11 @@ def _fit_setup(self, X):
 
         self.n_instances_, self.n_dims_, self.series_length_ = X.shape
 
+        if callable(self.n_intervals):
+            self._n_intervals = self.n_intervals(X)
+        else:
+            self._n_intervals = self.n_intervals
+
         self._min_interval_length = self.min_interval_length
         if self.min_interval_length < 3:
             self._min_interval_length = 3