Merge pull request #148 from lcgraham/master

Adds sorting for tails and creation of simple indicator functions

bet/Comm.py

@@ -24,70 +24,84 @@ def __init__(self):
 
     def Get_size(self):
         """
+        
         :rtype: int
         :returns: 1
+        
         """
         return 1
 
     def Get_rank(self):
         """
+        
         :rtype: int
-        :returns 0
+        :returns: 0
+        
         """
         return 0
 
-    def allgather(self, val, val2=None):
+    def allgather(self, val):
         """
         :param object val: object to allgather
+        
         :rtype: object
         :returns: val
+        
         """
         return val
 
-    def gather(self, val1, val2=None, root=0):
+    def gather(self, val1, root=0):
         """
         :param object val1: object to gather
-        :param object val2: object to gather
         :param int root: 0
+        
         :rtype: object
         :returns: val1
+        
         """
         return [val1]
 
     def allreduce(self, val1, op=None):
         """
         :param object val1: object to allreduce
+        
         :rtype: object
         :returns: val1
+        
         """
         return val1
 
     def bcast(self, val, root=0):
         """
         :param object val: object to broadcast
         :param int root: 0
+        
         :rtype: object
         :returns: val
+        
         """
         return val
 
-    def scatter(self, val1, val2=None, root=0):
+    def scatter(self, val1, root=0):
         """
         :param object val1: object to scatter
-        :param object val2: object to scatter
         :param int root: 0
+        
         :rtype: object
         :returns: val1
+        
         """
-        if isinstance(val1, collections.Iterable) and len(val1)==1:
+        if isinstance(val1, collections.Iterable) and len(val1) == 1:
             val1 = val1[0]
         return val1
 
     def Allgather(self, val, val2=None):
         """
         :param object val: object to Allgather
+        
         :rtype: object
         :returns: val
+        
         """
         return val
 
@@ -96,17 +110,21 @@ def Allreduce(self, val1, val2=None, op=None):
         :param object val1: object to Allreduce
         :param object val2: object to Allreduce
         :param op: None
+        
         :rtype: object
         :returns: val1
+        
         """
         return val1
 
     def Bcast(self, val, root=0):
         """
         :param object val: object to gather
         :param int root: 0
+        
         :rtype: object
         :returns: val
+        
         """
         return val
 
@@ -115,8 +133,10 @@ def Scatter(self, val1, val2, root=0):
         :param object val1: object to Scatter
         :param object val2: object to Scatter
         :param int root: 0
+        
         :rtype: object
         :returns: val1
+        
         """
         return val1
 

bet/__init__.py

@@ -1,3 +1,29 @@
 # Copyright (C) 2014-2015 The BET Development Team
 
-__all__ = ['sampling', 'calculateP', 'postProcess', 'sensitivity', 'util', 'Comm']
+"""
+Butler, Estep, Tavener Method
+
+This package provides tools for solving stochastic inverse problems using a
+measure-theoretic. It is named for the developers of the key algorithm in
+:mod:`bet.calculateP.calculateP`.
+
+Comm :mod:`~bet.Comm` provides a work around for users who do not which to install
+    :program:``mpi4py``.
+
+util :mod:`~bet.util` provides some general use methods for creating grids,
+    checking/fixing dimensions, and globalizing arrays.
+
+calculateP :mod:`~bet.calculateP` provides tools to approximate probabilities.
+
+sampling :mod:`~bet.sampling` provides various sampling algorithms.
+
+sensitivity :mod:`~bet.sensitivity` provides tools for approximating
+    derivatives and optimally choosing quantities of interest.
+
+postProcess :mod:`~bet.postProcess` provides plotting tools and tools to sort
+    samples by probabilities.
+
+"""
+
+__all__ = ['sampling', 'calculateP', 'postProcess', 'sensitivity', 'util', 
+    'Comm']

bet/calculateP/__init__.py

@@ -12,5 +12,7 @@
     generating points for cells that define a regular grid for use by
     :meth:`numpy.histogramdd` and for determining their volumes, etc. This
     module is only for use by :mod:`~bet.calculateP.simpleFunP`.  
+* :mod:`~bet.calculateP.indicatorFunctions` provides methods for creating
+    indicator functions for use by various other classes.
 """
-__all__ = ['calculateP', 'simpleFunP', 'voronoiHistogram']
+__all__ = ['calculateP', 'simpleFunP', 'voronoiHistogram', 'indicatorFunctions']

bet/calculateP/calculateP.py

@@ -29,7 +29,8 @@ def emulate_iid_lebesgue(lam_domain, num_l_emulate):
     :returns: a set of samples for emulation
 
     """
-    num_l_emulate = (num_l_emulate/comm.size) + (comm.rank < num_l_emulate%comm.size)
+    num_l_emulate = (num_l_emulate/comm.size) + \
+            (comm.rank < num_l_emulate%comm.size)
     lam_width = lam_domain[:, 1] - lam_domain[:, 0]
     lambda_emulate = lam_width*np.random.random((num_l_emulate,
         lam_domain.shape[0]))+lam_domain[:, 0] 
@@ -63,11 +64,11 @@ def prob_emulated(samples, data, rho_D_M, d_distr_samples,
         samples = np.expand_dims(samples, axis=1) 
     if len(data.shape) == 1:
         data = np.expand_dims(data, axis=1) 
-    if type(lambda_emulate) == type(None):
+    if lambda_emulate is None:
         lambda_emulate = samples
     if len(d_distr_samples.shape) == 1:
         d_distr_samples = np.expand_dims(d_distr_samples, axis=1)
-    if type(d_Tree) == type(None):
+    if d_Tree is None:
         d_Tree = spatial.KDTree(d_distr_samples)
 
     # Determine which inputs go to which M bins using the QoI
@@ -121,7 +122,7 @@ def prob(samples, data, rho_D_M, d_distr_samples, d_Tree=None):
         data = np.expand_dims(data, axis=1) 
     if len(d_distr_samples.shape) == 1:
         d_distr_samples = np.expand_dims(d_distr_samples, axis=1)
-    if type(d_Tree) == type(None):
+    if d_Tree is None:
         d_Tree = spatial.KDTree(d_distr_samples)
 
     # Set up local arrays for parallelism
@@ -183,11 +184,11 @@ def prob_mc(samples, data, rho_D_M, d_distr_samples,
         samples = np.expand_dims(samples, axis=1) 
     if len(data.shape) == 1:
         data = np.expand_dims(data, axis=1) 
-    if type(lambda_emulate) == type(None):
+    if lambda_emulate is None:
         lambda_emulate = samples
     if len(d_distr_samples.shape) == 1:
         d_distr_samples = np.expand_dims(d_distr_samples, axis=1)
-    if type(d_Tree) == type(None):
+    if d_Tree is None:
         d_Tree = spatial.KDTree(d_distr_samples)
 
     # Determine which inputs go to which M bins using the QoI