Fixed #316 Add Maximum Window Size

Author: seanlaw

stumpy/aamp.py

@@ -280,7 +280,7 @@ def aamp(T_A, m, T_B=None, ignore_trivial=True):
     if T_B.ndim != 1:  # pragma: no cover
         raise ValueError(f"T_B is {T_B.ndim}-dimensional and must be 1-dimensional. ")
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
     if ignore_trivial is False and core.are_arrays_equal(T_A, T_B):  # pragma: no cover
         logger.warning("Arrays T_A, T_B are equal, which implies a self-join.")

stumpy/aamped.py

@@ -71,7 +71,7 @@ def aamped(dask_client, T_A, m, T_B=None, ignore_trivial=True):
     if T_B.ndim != 1:  # pragma: no cover
         raise ValueError(f"T_B is {T_B.ndim}-dimensional and must be 1-dimensional. ")
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
     if ignore_trivial is False and core.are_arrays_equal(T_A, T_B):  # pragma: no cover
         logger.warning("Arrays T_A, T_B are equal, which implies a self-join.")

stumpy/core.py

@@ -247,15 +247,20 @@ def are_distances_too_small(a, threshold=10e-6):  # pragma: no cover
     return False
 
 
-def check_window_size(m):
+def check_window_size(m, max_size=None):
     """
-    Check the window size and ensure that it is greater than or equal to 3
+    Check the window size and ensure that it is greater than or equal to 3 and, if
+    `max_size` is provided, ensure that the window size is less than or equal to the
+    `max_size`
 
     Parameters
     ----------
     m : int
         Window size
 
+    max_size : int, default None
+        The maximum window size allowed
+
     Returns
     -------
     None
@@ -274,6 +279,9 @@ def check_window_size(m):
             """,
         )
 
+    if max_size is not None and m > max_size:
+        raise ValueError(f"The window size must be less than or equal to {max_size}")
+
 
 def sliding_dot_product(Q, T):
     """
@@ -979,6 +987,12 @@ def mass(Q, T, M_T=None, Σ_T=None):
     if T.ndim != 1:  # pragma: no cover
         raise ValueError(f"T is {T.ndim}-dimensional and must be 1-dimensional. ")
 
+    if m > n:  # pragma: no cover
+        raise ValueError(
+            f"The length of `Q` ({len(Q)}) must be less than or equal to "
+            f"the length of `T` ({len(T)}). "
+        )
+
     distance_profile = np.empty(n - m + 1)
     if np.any(~np.isfinite(Q)):
         distance_profile[:] = np.inf
@@ -1104,6 +1118,12 @@ def mass_absolute(Q, T):
     if T.ndim != 1:  # pragma: no cover
         raise ValueError(f"T is {T.ndim}-dimensional and must be 1-dimensional. ")
 
+    if m > n:  # pragma: no cover
+        raise ValueError(
+            f"The length of `Q` ({len(Q)}) must be less than or equal to "
+            f"the length of `T` ({len(T)}). "
+        )
+
     distance_profile = np.empty(n - m + 1)
     if np.any(~np.isfinite(Q)):
         distance_profile[:] = np.inf

stumpy/gpu_aamp.py

@@ -431,7 +431,7 @@ def gpu_aamp(T_A, m, T_B=None, ignore_trivial=True, device_id=0):
             "For multidimensional STUMP use `stumpy.mstump` or `stumpy.mstumped`"
         )
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
     if ignore_trivial is False and core.are_arrays_equal(T_A, T_B):  # pragma: no cover
         logger.warning("Arrays T_A, T_B are equal, which implies a self-join.")

stumpy/gpu_stump.py

@@ -456,7 +456,7 @@ def gpu_stump(T_A, m, T_B=None, ignore_trivial=True, device_id=0):
             "For multidimensional STUMP use `stumpy.mstump` or `stumpy.mstumped`"
         )
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
     if ignore_trivial is False and core.are_arrays_equal(T_A, T_B):  # pragma: no cover
         logger.warning("Arrays T_A, T_B are equal, which implies a self-join.")

stumpy/mstump.py

@@ -836,7 +836,7 @@ def mstump(T, m, include=None, discords=False):
         err = f"T is {T_A.ndim}-dimensional and must be at least 1-dimensional"
         raise ValueError(f"{err}")
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[1], T_B.shape[1]))
 
     if include is not None:
         include = _preprocess_include(include)

stumpy/mstumped.py

@@ -80,7 +80,7 @@ def mstumped(dask_client, T, m, include=None, discords=False):
         err = f"T is {T_A.ndim}-dimensional and must be at least 1-dimensional"
         raise ValueError(f"{err}")
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[1], T_B.shape[1]))
 
     if include is not None:
         include = np.asarray(include)

stumpy/scrump.py

@@ -197,7 +197,7 @@ def prescrump(T_A, m, T_B=None, s=None):
     core.check_dtype(T_B)
     T_B[np.isinf(T_B)] = np.nan
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
     μ_Q, σ_Q = core.compute_mean_std(T_A, m)
     M_T, Σ_T = core.compute_mean_std(T_B, m)
@@ -377,7 +377,7 @@ def __init__(
                 "For multidimensional STUMP use `stumpy.mstump` or `stumpy.mstumped`"
             )
 
-        core.check_window_size(self._m)
+        core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
         if self._ignore_trivial is False and core.are_arrays_equal(
             self._T_A, self._T_B

stumpy/stamp.py

@@ -136,7 +136,7 @@ def stamp(T_A, T_B, m, ignore_trivial=False):
     if T_B.ndim != 1:  # pragma: no cover
         raise ValueError(f"T_B is {T_B.ndim}-dimensional and must be 1-dimensional. ")
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
     subseq_T_A = core.rolling_window(T_A, m)
     excl_zone = int(np.ceil(m / 2))
 

stumpy/stomp.py

@@ -86,7 +86,7 @@ def _stomp(T_A, m, T_B=None, ignore_trivial=True):
     if T_B.ndim != 1:  # pragma: no cover
         raise ValueError(f"T_B is {T_B.ndim}-dimensional and must be 1-dimensional. ")
 
-    core.check_window_size(m)
+    core.check_window_size(m, max_size=min(T_A.shape[0], T_B.shape[0]))
 
     if ignore_trivial is False and core.are_arrays_equal(T_A, T_B):  # pragma: no cover
         logger.warning("Arrays T_A, T_B are equal, which implies a self-join.")