[DOC] Shapelet classifier notebook extensions & minor docstring corre…

…ctions (#1930)

* Corrected typos while doing a code review

* improved for clarity

* Typo corrections

* Some more typo corrections

* Correcting same typos in different docstring

* minor docstring corrections

* minor docstring typo correction

* small typo

* modified code to highlight the univariate example

* Further discussed shapelet distance visualisation

* Attempted a more high level description of the transform mapping

* Minor correction relating to last commit

* changed ! to . to avoid confusion with factorial

* Tried to make alpha similarity explanation more explicit by referencing overlap

* found small typo in docstring while going through notebook

* we don't know if the generated subsequence is a shapelet because we havent evaluated its discriminative ability

* more minor markdown text corrections

* My previous addition about the shapelet transform table was wrong - corrected.

* Describing the shapelet graphs

* Automatic `pre-commit` fixes

* fixing precommit error

---------

Co-authored-by: IRKnyazev <[email protected]>

aeon/classification/shapelet_based/_rdst.py

@@ -1,7 +1,7 @@
 """Random Dilated Shapelet Transform (RDST) Classifier.
 
 A Random Dilated Shapelet Transform classifier pipeline that simply performs a random
-shapelet dilated transform and build (by default) a ridge classifier on the output.
+shapelet dilated transform and builds (by default) a ridge classifier on the output.
 """
 
 __maintainer__ = ["baraline"]
@@ -28,20 +28,20 @@ class RDSTClassifier(BaseClassifier):
     Implementation of the random dilated shapelet transform classifier pipeline
     along the lines of [1]_, [2]_. Transforms the data using the
     `RandomDilatedShapeletTransform` and then builds a `RidgeClassifierCV` classifier
-    with standard scalling.
+    with standard scaling.
 
     Parameters
     ----------
     max_shapelets : int, default=10000
-        The maximum number of shapelet to keep for the final transformation.
-        A lower number of shapelets can be kept if alpha similarity have discarded the
+        The maximum number of shapelets to keep for the final transformation.
+        A lower number of shapelets can be kept if alpha similarity has discarded the
         whole dataset.
     shapelet_lengths : array, default=None
-        The set of possible length for shapelets. Each shapelet length is uniformly
-        drawn from this set. If None, the shapelets length will be equal to
+        The set of possible lengths for shapelets. Each shapelet length is uniformly
+        drawn from this set. If None, the shapelet length will be equal to
         min(max(2,n_timepoints//2),11).
     proba_normalization : float, default=0.8
-        This probability (between 0 and 1) indicate the chance of each shapelet to be
+        This probability (between 0 and 1) indicates the chance of each shapelet to be
         initialized such as it will use a z-normalized distance, inducing either scale
         sensitivity or invariance. A value of 1 would mean that all shapelets will use
         a z-normalized distance.
@@ -50,22 +50,22 @@ class RDSTClassifier(BaseClassifier):
         Occurrence feature. If None, the 5th and the 10th percentiles (i.e. [5,10])
         will be used.
     alpha_similarity : float, default=0.5
-        The strenght of the alpha similarity pruning. The higher the value, the lower
-        the allowed number of common indexes with previously sampled shapelets
-        when sampling a new candidate with the same dilation parameter.
-        It can cause the number of sampled shapelets to be lower than max_shapelets if
-        the whole search space has been covered. The default is 0.5, and the maximum is
-        1. Value above it have no effect for now.
+        The strength of the alpha similarity pruning. The higher the value, the fewer
+        common indexes with previously sampled shapelets are allowed when sampling a
+        new candidate with the same dilation parameter. It can cause the number of
+        sampled shapelets to be lower than max_shapelets if the whole search space has
+        been covered. The default is 0.5, and the maximum is 1. Values above it have
+        no effect for now.
     use_prime_dilations : bool, default=False
-        If True, restrict the value of the shapelet dilation parameter to be prime
+        If True, restricts the value of the shapelet dilation parameter to be prime
         values. This can greatly speed-up the algorithm for long time series and/or
-        short shapelet length, possibly at the cost of some accuracy.
+        short shapelet lengths, possibly at the cost of some accuracy.
     distance: str="manhattan"
         Name of the distance function to be used. By default this is the
         manhattan distance. Other distances from the aeon distance modules can be used.
     estimator : BaseEstimator or None, default=None
         Base estimator for the ensemble, can be supplied a sklearn `BaseEstimator`. If
-        `None` a default `RidgeClassifierCV` classifier is used with standard scalling.
+        `None` a default `RidgeClassifierCV` classifier is used with standard scaling.
     save_transformed_data : bool, default=False
         If True, the transformed training dataset for all classifiers will be saved.
     class_weight{“balanced”, “balanced_subsample”}, dict or list of dicts, default=None
@@ -234,7 +234,7 @@ def _predict(self, X) -> np.ndarray:
         Parameters
         ----------
         X: np.ndarray shape (n_cases, n_channels, n_timepoints)
-            The data to make prediction for.
+            The data to make predictions for.
 
         Returns
         -------
@@ -246,12 +246,12 @@ def _predict(self, X) -> np.ndarray:
         return self._estimator.predict(X_t)
 
     def _predict_proba(self, X) -> np.ndarray:
-        """Predicts labels probabilities for sequences in X.
+        """Predicts label probabilities for sequences in X.
 
         Parameters
         ----------
         X: np.ndarray shape (n_cases, n_channels, n_timepoints)
-            The data to make predict probabilities for.
+            The data to predict probabilities for.
 
         Returns
         -------

aeon/transformations/collection/shapelet_based/_dilated_shapelet_transform.py

@@ -1,6 +1,6 @@
 """Dilated Shapelet transformers.
 
-A modification of the classic Shapelet Transform which add a dilation parameter to
+A modification of the classic Shapelet Transform which adds a dilation parameter to
 Shapelets.
 """
 
@@ -38,33 +38,33 @@ class RandomDilatedShapeletTransform(BaseCollectionTransformer):
         - Length is randomly selected from shapelet_lengths parameter
         - Dilation is sampled as a function the shapelet length and time series length
         - Normalization is chosen randomly given the probability given as parameter
-        - Value is sampled randomly from an input time series given the length and
+        - Start value is sampled randomly from an input time series given the length and
         dilation parameter.
         - Threshold is randomly chosen between two percentiles of the distribution
         of the distance vector between the shapelet and another time series. This time
-        serie is drawn from the same class if classes are given during fit. Otherwise,
+        series is drawn from the same class if classes are given during fit. Otherwise,
         a random sample will be used. If there is only one sample per class, the same
         sample will be used.
     Then, once the set of shapelets have been initialized, we extract the shapelet
     features from each pair of shapelets and input series. Three features are extracted:
         - min d(S,X): the minimum value of the distance vector between a shapelet S and
         a time series X.
         - argmin d(S,X): the location of the minumum.
-        - SO(d(S,X), threshold): The number of point in the distance vector that are
+        - SO(d(S,X), threshold): The number of points in the distance vector that are
         bellow the threshold parameter of the shapelet.
 
     Parameters
     ----------
     max_shapelets : int, default=10000
-        The maximum number of shapelet to keep for the final transformation.
-        A lower number of shapelets can be kept if alpha similarity have discarded the
+        The maximum number of shapelets to keep for the final transformation.
+        A lower number of shapelets can be kept if alpha similarity has discarded the
         whole dataset.
     shapelet_lengths : array, default=None
-        The set of possible length for shapelets. Each shapelet length is uniformly
-        drawn from this set. If None, the shapelets length will be equal to
+        The set of possible lengths for shapelets. Each shapelet length is uniformly
+        drawn from this set. If None, the shapelet length will be equal to
         min(max(2,n_timepoints//2),11).
     proba_normalization : float, default=0.8
-        This probability (between 0 and 1) indicate the chance of each shapelet to be
+        This probability (between 0 and 1) indicates the chance of each shapelet to be
         initialized such as it will use a z-normalized distance, inducing either scale
         sensitivity or invariance. A value of 1 would mean that all shapelets will use
         a z-normalized distance.
@@ -73,16 +73,16 @@ class RandomDilatedShapeletTransform(BaseCollectionTransformer):
         Occurrence feature. If None, the 5th and the 10th percentiles (i.e. [5,10])
         will be used.
     alpha_similarity : float, default=0.5
-        The strength of the alpha similarity pruning. The higher the value, the lower
-        the allowed number of common indexes with previously sampled shapelets
-        when sampling a new candidate with the same dilation parameter.
-        It can cause the number of sampled shapelets to be lower than max_shapelets if
-        the whole search space has been covered. The default is 0.5, and the maximum is
-        1. Value above it have no effect for now.
+        The strength of the alpha similarity pruning. The higher the value, the fewer
+        common indexes with previously sampled shapelets are allowed when sampling a
+        new candidate with the same dilation parameter. It can cause the number of
+        sampled shapelets to be lower than max_shapelets if the whole search space
+        has been covered. The default is 0.5, and the maximum is 1. Values above it
+        have no effect for now.
     use_prime_dilations : bool, default=False
-        If True, restrict the value of the shapelet dilation parameter to be prime
+        If True, restricts the value of the shapelet dilation parameter to be prime
         values. This can greatly speed up the algorithm for long time series and/or
-        short shapelet length, possibly at the cost of some accuracy.
+        short shapelet lengths, possibly at the cost of some accuracy.
     distance: str="manhattan"
         Name of the distance function to be used. By default this is the
         manhattan distance. Other distances from the aeon distance modules can be used.
@@ -98,7 +98,7 @@ class RandomDilatedShapeletTransform(BaseCollectionTransformer):
             - shapelet values
             - length parameter
             - dilation parameter
-            - treshold parameter
+            - threshold parameter
             - normalization parameter
             - mean parameter
             - standard deviation parameter
@@ -109,8 +109,8 @@ class RandomDilatedShapeletTransform(BaseCollectionTransformer):
 
     Notes
     -----
-    This implementation use all the features for multivariate shapelets, without
-    affecting a random feature subsets to each shapelet as done in the original
+    This implementation uses all the features for multivariate shapelets, without
+    affecting a random feature subset to each shapelet as done in the original
     implementation. See `convst
     https://github.com/baraline/convst/blob/main/convst/transformers/rdst.py`_.
 
@@ -717,7 +717,7 @@ def dilated_shapelet_transform(X, shapelets, distance):
     -------
     X_new : array, shape=(n_cases, 3*n_shapelets)
         The transformed input time series with each shapelet extracting 3
-        feature from the distance vector computed on each time series.
+        features from the distance vector computed on each time series.
 
     """
     (

aeon/visualisation/estimator/_shapelets.py

@@ -798,7 +798,7 @@ def visualize_best_shapelets_one_class(
         """
         Plot the n_shp best candidates for the class_id.
 
-        Visualize best macth on two random samples and how the shapelet discriminate
+        Visualize best match on two random samples and how the shapelet discriminate
         (X,y) with boxplots.
 
         Parameters