Add param isconstant to stimp (#878)

* add test function

* add param isconstant to stimp

* add test function for stimped

* minor fixes

* fix non-normalized

Author: NimaSarajpoor

stumpy/stimp.py

@@ -50,16 +50,16 @@ class _stimp:
     T : numpy.ndarray
         The time series or sequence for which to compute the pan matrix profile
 
-    m_start : int, default 3
+    min_m : int, default 3
         The starting (or minimum) subsequence window size for which a matrix profile
         may be computed
 
-    m_stop : int, default None
+    max_m : int, default None
         The stopping (or maximum) subsequence window size for which a matrix profile
-        may be computed. When `m_stop = Non`, this is set to the maximum allowable
+        may be computed. When `max_m = Non`, this is set to the maximum allowable
         subsequence window size
 
-    m_step : int, default 1
+    step : int, default 1
         The step between subsequence window sizes
 
     percentage : float, default 0.01
@@ -87,6 +87,14 @@ class _stimp:
     mp_func : function, default stump
         The matrix profile function to use when `percentage = 1.0`
 
+    T_subseq_isconstant_func : function, default None
+        A custom, user-defined function that returns a boolean array that indicates
+        whether a subsequence in `T` is constant (True). The function must only take
+        two arguments, `a`, a 1-D array, and `w`, the window size, while additional
+        arguments may be specified by currying the user-defined function using
+        `functools.partial`. Any subsequence with at least one np.nan/np.inf will
+        automatically have its corresponding value set to False in this boolean array.
+
     Attributes
     ----------
     PAN_ : numpy.ndarray
@@ -122,6 +130,7 @@ def __init__(
         client=None,
         device_id=None,
         mp_func=stump,
+        T_subseq_isconstant_func=None,
     ):
         """
         Initialize the `stimp` object and compute the Pan Matrix Profile
@@ -167,6 +176,15 @@ def __init__(
 
         mp_func : function, default stump
             The matrix profile function to use when `percentage = 1.0`
+
+        T_subseq_isconstant_func : function, default None
+            A custom, user-defined function that returns a boolean array that indicates
+            whether a subsequence in `T` is constant (True). The function must only take
+            two arguments, `a`, a 1-D array, and `w`, the window size, while additional
+            arguments may be specified by currying the user-defined function using
+            `functools.partial`. Any subsequence with at least one np.nan/np.inf will
+            automatically have its corresponding value set to False in this boolean
+            array.
         """
         self._T = T.copy()
         if max_m is None:
@@ -189,6 +207,16 @@ def __init__(
             mp_func, client=client, device_id=device_id
         )
 
+        if T_subseq_isconstant_func is None:
+            T_subseq_isconstant_func = core._rolling_isconstant
+        if not callable(T_subseq_isconstant_func):  # pragma: no cover
+            msg = (
+                "`T_subseq_isconstant_func` was expected to be a callable function "
+                + f"but {type(T_subseq_isconstant_func)} was found."
+            )
+            raise ValueError(msg)
+        self._T_subseq_isconstant_func = T_subseq_isconstant_func
+
         self._PAN = np.full(
             (self._M.shape[0], self._T.shape[0]), fill_value=np.inf, dtype=np.float64
         )
@@ -223,6 +251,7 @@ def update(self):
                     percentage=self._percentage,
                     pre_scrump=self._pre_scrump,
                     k=1,
+                    T_A_subseq_isconstant=self._T_subseq_isconstant_func,
                 )
                 approx.update()
                 self._PAN[
@@ -233,6 +262,7 @@ def update(self):
                     self._T,
                     m,
                     ignore_trivial=True,
+                    T_A_subseq_isconstant=self._T_subseq_isconstant_func,
                 )
                 self._PAN[
                     self._bfs_indices[self._n_processed], : out[:, 0].shape[0]
@@ -347,7 +377,7 @@ def M_(self):
 
 @core.non_normalized(
     aamp_stimp,
-    exclude=["pre_scrump", "normalize", "p", "pre_scraamp"],
+    exclude=["pre_scrump", "normalize", "p", "T_subseq_isconstant_func", "pre_scraamp"],
     replace={"pre_scrump": "pre_scraamp"},
 )
 class stimp(_stimp):
@@ -361,16 +391,16 @@ class stimp(_stimp):
     T : numpy.ndarray
         The time series or sequence for which to compute the pan matrix profile
 
-    m_start : int, default 3
+    min_m : int, default 3
         The starting (or minimum) subsequence window size for which a matrix profile
         may be computed
 
-    m_stop : int, default None
+    max_m : int, default None
         The stopping (or maximum) subsequence window size for which a matrix profile
-        may be computed. When `m_stop = Non`, this is set to the maximum allowable
+        may be computed. When `max_m = Non`, this is set to the maximum allowable
         subsequence window size
 
-    m_step : int, default 1
+    step : int, default 1
         The step between subsequence window sizes
 
     percentage : float, default 0.01
@@ -393,6 +423,14 @@ class stimp(_stimp):
         The p-norm to apply for computing the Minkowski distance. This parameter is
         ignored when `normalize == True`.
 
+    T_subseq_isconstant_func : function, default None
+        A custom, user-defined function that returns a boolean array that indicates
+        whether a subsequence in `T` is constant (True). The function must only take
+        two arguments, `a`, a 1-D array, and `w`, the window size, while additional
+        arguments may be specified by currying the user-defined function using
+        `functools.partial`. Any subsequence with at least one np.nan/np.inf will
+        automatically have its corresponding value set to False in this boolean array.
+
     Attributes
     ----------
     PAN_ : numpy.ndarray
@@ -442,6 +480,7 @@ def __init__(
         pre_scrump=True,
         normalize=True,
         p=2.0,
+        T_subseq_isconstant_func=None,
     ):
         """
         Initialize the `stimp` object and compute the Pan Matrix Profile
@@ -483,6 +522,15 @@ def __init__(
         p : float, default 2.0
             The p-norm to apply for computing the Minkowski distance. This parameter is
             ignored when `normalize == True`.
+
+        T_subseq_isconstant_func : function, default None
+            A custom, user-defined function that returns a boolean array that indicates
+            whether a subsequence in `T` is constant (True). The function must only take
+            two arguments, `a`, a 1-D array, and `w`, the window size, while additional
+            arguments may be specified by currying the user-defined function using
+            `functools.partial`. Any subsequence with at least one np.nan/np.inf will
+            automatically have its corresponding value set to False in this boolean
+            array.
         """
         super().__init__(
             T,
@@ -492,12 +540,13 @@ def __init__(
             percentage=percentage,
             pre_scrump=pre_scrump,
             mp_func=stump,
+            T_subseq_isconstant_func=T_subseq_isconstant_func,
         )
 
 
 @core.non_normalized(
     aamp_stimped,
-    exclude=["pre_scrump", "normalize", "p", "pre_scraamp"],
+    exclude=["pre_scrump", "normalize", "p", "T_subseq_isconstant_func", "pre_scraamp"],
     replace={"pre_scrump": "pre_scraamp"},
 )
 class stimped(_stimp):
@@ -516,16 +565,16 @@ class stimped(_stimp):
     T : numpy.ndarray
         The time series or sequence for which to compute the pan matrix profile
 
-    m_start : int, default 3
+    min_m : int, default 3
         The starting (or minimum) subsequence window size for which a matrix profile
         may be computed
 
-    m_stop : int, default None
+    max_m : int, default None
         The stopping (or maximum) subsequence window size for which a matrix profile
-        may be computed. When `m_stop = Non`, this is set to the maximum allowable
+        may be computed. When `max_m = Non`, this is set to the maximum allowable
         subsequence window size
 
-    m_step : int, default 1
+    step : int, default 1
         The step between subsequence window sizes
 
     normalize : bool, default True
@@ -537,6 +586,15 @@ class stimped(_stimp):
         The p-norm to apply for computing the Minkowski distance. This parameter is
         ignored when `normalize == True`.
 
+    T_subseq_isconstant_func : function, default None
+            A custom, user-defined function that returns a boolean array that indicates
+            whether a subsequence in `T` is constant (True). The function must only take
+            two arguments, `a`, a 1-D array, and `w`, the window size, while additional
+            arguments may be specified by currying the user-defined function using
+            `functools.partial`. Any subsequence with at least one np.nan/np.inf will
+            automatically have its corresponding value set to False in this boolean
+            array.
+
     Attributes
     ----------
     PAN_ : numpy.ndarray
@@ -590,6 +648,7 @@ def __init__(
         step=1,
         normalize=True,
         p=2.0,
+        T_subseq_isconstant_func=None,
     ):
         """
         Initialize the `stimp` object and compute the Pan Matrix Profile
@@ -625,6 +684,15 @@ def __init__(
         p : float, default 2.0
             The p-norm to apply for computing the Minkowski distance. This parameter is
             ignored when `normalize == True`.
+
+        T_subseq_isconstant_func : function, default None
+            A custom, user-defined function that returns a boolean array that indicates
+            whether a subsequence in `T` is constant (True). The function must only take
+            two arguments, `a`, a 1-D array, and `w`, the window size, while additional
+            arguments may be specified by currying the user-defined function using
+            `functools.partial`. Any subsequence with at least one np.nan/np.inf will
+            automatically have its corresponding value set to False in this boolean
+            array.
         """
         super().__init__(
             T,
@@ -635,4 +703,5 @@ def __init__(
             pre_scrump=False,
             client=client,
             mp_func=stumped,
+            T_subseq_isconstant_func=T_subseq_isconstant_func,
         )

tests/test_stimp.py

@@ -1,3 +1,5 @@
+import functools
+
 import naive
 import numpy as np
 import numpy.testing as npt
@@ -223,3 +225,138 @@ def test_stimped(T, dask_cluster):
         naive.replace_inf(cmp_pan)
 
         npt.assert_almost_equal(ref_pan, cmp_pan)
+
+
+def test_stimp_1_percent_with_isconstant():
+    T = np.random.uniform(-1, 1, [64])
+    isconstant_func = functools.partial(
+        naive.isconstant_func_stddev_threshold, stddev_threshold=0.5
+    )
+
+    threshold = 0.2
+    percentage = 0.01
+    min_m = 3
+    n = T.shape[0] - min_m + 1
+
+    seed = np.random.randint(100000)
+
+    np.random.seed(seed)
+    pan = stimp(
+        T,
+        min_m=min_m,
+        max_m=None,
+        step=1,
+        percentage=percentage,
+        pre_scrump=True,
+        # normalize=True,
+        T_subseq_isconstant_func=isconstant_func,
+    )
+
+    for i in range(n):
+        pan.update()
+
+    ref_PAN = np.full((pan.M_.shape[0], T.shape[0]), fill_value=np.inf)
+
+    np.random.seed(seed)
+    for idx, m in enumerate(pan.M_[:n]):
+        zone = int(np.ceil(m / 4))
+        s = zone
+        tmp_P, tmp_I = naive.prescrump(
+            T,
+            m,
+            T,
+            s=s,
+            exclusion_zone=zone,
+            T_A_subseq_isconstant=isconstant_func,
+            T_B_subseq_isconstant=isconstant_func,
+        )
+        ref_P, ref_I, _, _ = naive.scrump(
+            T,
+            m,
+            T,
+            percentage,
+            zone,
+            True,
+            s,
+            T_A_subseq_isconstant=isconstant_func,
+            T_B_subseq_isconstant=isconstant_func,
+        )
+        naive.merge_topk_PI(ref_P, tmp_P, ref_I, tmp_I)
+        ref_PAN[pan._bfs_indices[idx], : ref_P.shape[0]] = ref_P
+
+    # Compare raw pan
+    cmp_PAN = pan._PAN
+
+    naive.replace_inf(ref_PAN)
+    naive.replace_inf(cmp_PAN)
+
+    npt.assert_almost_equal(ref_PAN, cmp_PAN)
+
+    # Compare transformed pan
+    cmp_pan = pan.PAN_
+    ref_pan = naive.transform_pan(
+        pan._PAN, pan._M, threshold, pan._bfs_indices, pan._n_processed
+    )
+
+    naive.replace_inf(ref_pan)
+    naive.replace_inf(cmp_pan)
+
+    npt.assert_almost_equal(ref_pan, cmp_pan)
+
+
+@pytest.mark.filterwarnings("ignore:\\s+Port 8787 is already in use:UserWarning")
+def test_stimped_with_isconstant(dask_cluster):
+    T = np.random.uniform(-1, 1, [64])
+    isconstant_func = functools.partial(
+        naive.isconstant_func_stddev_threshold, stddev_threshold=0.5
+    )
+
+    with Client(dask_cluster) as dask_client:
+        threshold = 0.2
+        min_m = 3
+        n = T.shape[0] - min_m + 1
+
+        pan = stimped(
+            dask_client,
+            T,
+            min_m=min_m,
+            max_m=None,
+            step=1,
+            # normalize=True,
+            T_subseq_isconstant_func=isconstant_func,
+        )
+
+        for i in range(n):
+            pan.update()
+
+        ref_PAN = np.full((pan.M_.shape[0], T.shape[0]), fill_value=np.inf)
+
+        for idx, m in enumerate(pan.M_[:n]):
+            zone = int(np.ceil(m / 4))
+            ref_mp = naive.stump(
+                T,
+                m,
+                T_B=None,
+                exclusion_zone=zone,
+                T_A_subseq_isconstant=isconstant_func,
+            )
+            ref_PAN[pan._bfs_indices[idx], : ref_mp.shape[0]] = ref_mp[:, 0]
+
+        # Compare raw pan
+        cmp_PAN = pan._PAN
+
+        naive.replace_inf(ref_PAN)
+        naive.replace_inf(cmp_PAN)
+
+        npt.assert_almost_equal(ref_PAN, cmp_PAN)
+
+        # Compare transformed pan
+        cmp_pan = pan.PAN_
+        ref_pan = naive.transform_pan(
+            pan._PAN, pan._M, threshold, pan._bfs_indices, pan._n_processed
+        )
+
+        naive.replace_inf(ref_pan)
+        naive.replace_inf(cmp_pan)
+
+        npt.assert_almost_equal(ref_pan, cmp_pan)