[ENH] Rework of base series annotator API

extension_templates/annotation.py

@@ -42,16 +42,6 @@ class MySeriesAnnotator(BaseSeriesAnnotator):
 
     Parameters
     ----------
-    fmt : str {"dense", "sparse"}, optional (default="dense")
-        Annotation output format:
-        * If "sparse", a sub-series of labels for only the outliers in X is returned,
-        * If "dense", a series of labels for all values in X is returned.
-    labels : str {"indicator", "score"}, optional (default="indicator")
-        Annotation output labels:
-        * If "indicator", returned values are boolean, indicating whether a value is an
-        outlier,
-        * If "score", returned values are floats, giving the outlier score.
-
     parama : int
         descriptive explanation of parama
     paramb : string, optional (default='default')
@@ -69,6 +59,12 @@ class MySeriesAnnotator(BaseSeriesAnnotator):
     and so on
     """
 
+    # Change the `task` and `learning_type` as needed
+    _tags = {
+        "task": "segmentation",
+        "learning_type": "unsupervised",
+    }
+
     # todo: add any hyper-parameters and components to constructor
     def __init__(
         self,
@@ -77,8 +73,6 @@ def __init__(
         est2=None,
         paramb="default",
         paramc=None,
-        fmt="dense",
-        labels="indicator",
     ):
         # estimators should precede parameters
         #  if estimators have default values, set None and initialize below
@@ -89,8 +83,7 @@ def __init__(
         self.paramb = paramb
         self.paramc = paramc
 
-        # leave this as is
-        super().__init__(fmt=fmt, labels=labels)
+        super().__init__()
 
         # todo: optional, parameter checking logic (if applicable) should happen here
         # if writes derived values to self, should *not* overwrite self.parama etc

sktime/annotation/adapters/_pyod.py

@@ -21,22 +21,17 @@ class PyODAnnotator(BaseSeriesAnnotator):
     estimator : PyOD estimator
         See ``https://pyod.readthedocs.io/en/latest/`` documentation for a detailed
         description of all options.
-    fmt : str {"dense", "sparse"}, optional (default="dense")
-        Annotation output format:
-        * If "sparse", a sub-series of labels for only the outliers in X is returned,
-        * If "dense", a series of labels for all values in X is returned.
-    labels : str {"indicator", "score"}, optional (default="indicator")
-        Annotation output labels:
-        * If "indicator", returned values are boolean, indicating whether a value is an
-        outlier,
-        * If "score", returned values are floats, giving the outlier score.
     """
 
-    _tags = {"python_dependencies": "pyod"}
+    _tags = {
+        "python_dependencies": "pyod",
+        "task": "anomaly_detection",
+        "learning_type": "unsupervised",
+    }
 
-    def __init__(self, estimator, fmt="dense", labels="indicator"):
+    def __init__(self, estimator):
         self.estimator = estimator  # pyod estimator
-        super().__init__(fmt=fmt, labels=labels)
+        super().__init__()
 
     def _fit(self, X, Y=None):
         """Fit to training data.
@@ -77,28 +72,17 @@ def _predict(self, X):
         Returns
         -------
         Y : pd.Series - annotations for sequence X
-            exact format depends on annotation type
         """
-        fmt = self.fmt
-        labels = self.labels
-
         X_np = X.to_numpy()
 
         if len(X_np.shape) == 1:
             X_np = X_np.reshape(-1, 1)
 
         Y_np = self.estimator_.predict(X_np)
 
-        if labels == "score":
-            Y_val_np = self.estimator_.decision_function(X_np)
-        elif labels == "indicator":
-            Y_val_np = Y_np
-
-        if fmt == "dense":
-            Y = pd.Series(Y_val_np, index=X.index)
-        elif fmt == "sparse":
-            Y_loc = np.where(Y_np)
-            Y = pd.Series(Y_val_np[Y_loc], index=X.index[Y_loc])
+        # Convert the output to a sparse format
+        Y_loc = np.where(Y_np)
+        Y = pd.Series(Y_np[Y_loc], index=X.index[Y_loc])
 
         return Y