Ticket #3726: sage-3726-part5.patch

a/sage/stats/hmm/all.py

@@ -1 +1 @@
-from hmm  import DiscreteHiddenMarkovModel
-Normal
+Gaussian

a/sage/stats/hmm/chmm.pxd

@@ -74 +74 @@
-        # IDs of predecessor states
-        int *in_id
-        # transition probs to successor states. It is a
-COS
+cos
-        double **out_a
-        # transition probs from predecessor states. It is a
-COS
+cos
-        double **in_a
-        # number of  successor states
-        int out_states
@@ -97 +97 @@
-        int xPosition
-        # y coordinate position for graph representation plotting
-        int yPosition
+
+
+    double **ighmm_cmatrix_alloc (int nrows, int ncols)
-
-    #############################################################
-    # Continous Hidden Markov Model
@@ -196 +199 @@
-    cdef ghmm_cmodel* m
-    cdef bint initialized
-
-Normal
+Gaussian
-    pass
-
-

a/sage/stats/hmm/chmm.pyx

@@ -12 +12 @@
-        # Set number of states
-        self.m.N = self.A.nrows()
-
-Normal
+Gaussian
-    """
+    Create a Gaussian Hidden Markov Model.  The probability
+    distribution associated with each state is a Gaussian
+    distribution.
+
+    GaussianHiddenMarkovModel(A, B, pi, name)
+
+    INPUT:
+        A  -- matrix; the transition matrix (n x n)
+        B  -- list of n pairs (mu, sigma) that define the
+              Gaussian distributions associated to each state
+        pi -- list of floats that sums to 1.0; these are
+              the initial probabilities of each hidden state
+        name -- (default: None); 
+
-    EXAMPLES:
-T
+Define t
-        sage: A = [[0,1,0],[0.5,0,0.5],[0.3,0.3,0.4]]
-
-    Parameters of the normal emission distributions in pairs of (mu, sigma):
@@ -24 +38 @@
-    The initial probabilities per state:
-        sage: pi = [1,0,0]
-
-HMM
-Normal
+Gaussian hidden Markov model
+Gaussian
-    """
-    def __init__(self, A, B, pi, name=None):
-        ContinuousHiddenMarkovModel.__init__(self, A, B, pi)
-
-
+
+
-        self.m.M = 1
-
-        # Set the model type to continuous
@@ -57 +72 @@
-
-        for i in range(self.m.N):
-            # Parameters of normal distribution
-
+  
-            # Get a reference to the i-th state for convenience of the notation below.
-            state = &(states[i])
-
-
+
+
+
+
+
-            e = <ghmm_c_emission*> safe_malloc(sizeof(ghmm_c_emission))
-
+     
-            e.dimension = 1
-
+ 
-            e.variance.val = sigma
-            # fixing of emissions is deactivated by default            
-
-
-
-
-
+
+
+
+
+
+
+
-
-        # Set states
-        self.m.s = states
-
+        self.m.class_change = NULL
+
-        self.initialized = True
-
-    def __dealloc__(self):
-        return
-        if self.initialized:
-            ghmm_cmodel_free(&self.m)
-
@@ -92 +113 @@
-            string
-
-        EXAMPLES:
-Normal
+Gaussian
-            sage: a.__repr__()
-            "Discrete Hidden Markov Model (2 states, 2 outputs)\nInitial probabilities: [0.5, 0.5]\nTransition matrix:\n[0.1 0.9]\n[0.1 0.9]\nEmission matrix:\n[0.9 0.1]\n[0.1 0.9]\nEmission symbols: [3/4, 'abc']"
-        """
-Normal
+Gaussian
-            ' ' + self.m.name if self.m.name else '',
-            self.m.N, self.m.M)
-#
-#
-#s += '\nEmission matrix:\n%s'%self.emission_matrix
+
+
+s += '\nEmission parameters:\n%s'%self.emission_parameters
-        return s
-
+    def initial_probabilities(self):
+        """
+        Return the list of initial state probabilities.
+
+        OUTPUT:
+            list of floats
+        
+        EXAMPLES:
+            sage: m = hmm.GaussianHiddenMarkovModel([[0.0,1.0,0],[0.5,0.0,0.5],[0.3,0.3,0.4]], [(0.0,1.0), (-1.0,0.5), (1.0,0.2)], [0.4,0.3,0.3])
+            sage: m.initial_probabilities()
+            [0.4, 0.3, 0.3]
+        """
+        cdef Py_ssize_t i
+        return [self.m.s[i].pi for i in range(self.m.N)]
+    
+    def transition_matrix(self, list_only=True):
+        """
+        Return the hidden state transition matrix. 
+        
+        EXAMPLES:
+            sage: m = hmm.GaussianHiddenMarkovModel([[0.0,1.0,0],[0.5,0.0,0.5],[0.3,0.3,0.4]], [(0.0,1.0), (-1.0,0.5), (1.0,0.2)], [1,0,0])
+            sage: m.transition_matrix()
+            [0.9 0.1]
+            [0.9 0.1]
+        """
+        cdef Py_ssize_t i, j
+        for i from 0 <= i < self.m.N:
+            for j from 0 <= j < self.m.s[i].out_states:
+                self.A.set_unsafe_double(i,j,self.m.s[i].out_a[0][j])
+        return self.A
+
+    def emission_parameters(self):
+        """
+        Return the emission probability matrix.
+        
+        EXAMPLES:
+            sage: m = hmm.GaussianHiddenMarkovModel([[0.0,1.0,0],[0.5,0.0,0.5],[0.3,0.3,0.4]], [(0.0,1.0), (-1.0,0.5), (1.0,0.2)], [0.1,0.4,0.5])
+            sage: m.emission_parameters()
+            [(0.0, 1.0), (-1.0, 0.5), (1.0, 0.20000...)]
+        """
+        cdef Py_ssize_t i, j
+        v = []
+        for i from 0 <= i < self.m.N:
+            v.append((self.m.s[i].e.mean.val, self.m.s[i].e.variance.val))
+        return v

a/sage/stats/hmm/hmm.pyx

@@ -200 +200 @@
-        """
-        Return the list of initial state probabilities.
-        
+        OUTPUT:
+            list of floats
+
-        EXAMPLES:
-            sage: a = hmm.DiscreteHiddenMarkovModel([[0.9,0.1],[0.9,0.1]], [[0.5,0.5,0,0],[0,0,.5,.5]], [0.5,0.5], [1,-1,3,-3])
-            sage: a.initial_probabilities()