state tests

.gitignore

@@ -1,6 +1,6 @@
 __pycache__
 *.egg-info
 env
-
+.coverage
 .vscode
 

README.md

@@ -8,7 +8,17 @@ The book represents algorithms as:
 1. matlab code
 1. described in English in the textbook body
 
-## Pour Moi
+## Chapters
+
+1. Introduction
+2. Background
+3. Discrete-Time Random Processes
+4. Signal Modeling
+5. The Levinson Recursion
+6. Lattice Filters
+7. Optimum Filters
+8. Spectrum Estimation
+9. Adaptive Filters
 
 ### TODO
 
@@ -25,3 +35,5 @@ The book represents algorithms as:
 - stub files?
 - VS code add-ons
 - use matlab-equivalent psuedo-inverse ("X \ R") operation to match the book
+- switch linting to Ruff?
+- performance benchmarking would be cool

pyssp/__init__.py

@@ -1 +1,10 @@
 """The import root for the book sections."""
+
+# import system as Chapter3
+# import modeling as Chapter4
+# import levinson as Chapter5
+# import lattice as Chapter6
+# import optimal as Chapter7
+# import spectrum as Chapter8
+# import adaptive as Chapter9
+# import state as Appendix

pyssp/modeling.py

@@ -155,7 +155,7 @@ def covm(x, p):
     x0 = x.copy().ravel().reshape(-1, 1)
     N = len(x0)
     if p >= len(x0):
-        raise ValueError("p (all-pole model) too large")
+        raise ValueError(f"{p=} all-pole model too large")
 
     X = convm(x0, p + 1)
     Xq = X[p - 1:N - 1, :p].copy()

tests/test_state.py

@@ -0,0 +1,31 @@
+"""Test the state module."""
+
+
+import numpy as np
+
+from pyssp import state
+
+
+def test_convm():
+    x = np.array([1, 2, 3])
+    p = 4
+
+    expected = np.array([[1, 0, 0, 0],
+                         [2, 1, 0, 0],
+                         [3, 2, 1, 0],
+                         [0, 3, 2, 1],
+                         [0, 0, 3, 2],
+                         [0, 0, 0, 3]], dtype=float)
+    assert np.array_equal(expected, state.convm(x, p))
+
+
+def test_covar():
+    x = np.array([1, 2, 3])
+    p = 4
+
+    expected = np.array([[1, 0, -0.5, 0, 0],
+                         [0, 1, 0, -0.5, 0],
+                         [-0.5, 0, 1, 0, -0.5],
+                         [0, -0.5, 0, 1, 0],
+                         [0, 0, -0.5, 0, 1]], dtype=float)
+    assert np.array_equal(expected, state.covar(x, p))