CUQIpy 1.3.0.post0.dev300__tar.gz → 1.3.0.post0.dev371__tar.gz

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/CUQIpy.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: CUQIpy
-Version: 1.3.0.post0.dev300
+Version: 1.3.0.post0.dev371
 Summary: Computational Uncertainty Quantification for Inverse problems in Python
 Maintainer-email: "Nicolai A. B. Riis" <nabr@dtu.dk>, "Jakob S. Jørgensen" <jakj@dtu.dk>, "Amal M. Alghamdi" <amaal@dtu.dk>, Chao Zhang <chaz@dtu.dk>
 License:                                  Apache License

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: CUQIpy
-Version: 1.3.0.post0.dev300
+Version: 1.3.0.post0.dev371
 Summary: Computational Uncertainty Quantification for Inverse problems in Python
 Maintainer-email: "Nicolai A. B. Riis" <nabr@dtu.dk>, "Jakob S. Jørgensen" <jakj@dtu.dk>, "Amal M. Alghamdi" <amaal@dtu.dk>, Chao Zhang <chaz@dtu.dk>
 License:                                  Apache License

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/cuqi/_version.py

@@ -8,11 +8,11 @@ import json
 
 version_json = '''
 {
- "date": "2025-09-06T13:37:24+0300",
+ "date": "2025-09-12T10:46:36+0200",
  "dirty": false,
  "error": null,
- "full-revisionid": "2071a1f316a1486c3d6c3b95fa7b0b3318c7f461",
- "version": "1.3.0.post0.dev300"
+ "full-revisionid": "1bd165981a9e7200ec642e2845ffde05a7e8a168",
+ "version": "1.3.0.post0.dev371"
 }
 '''  # END VERSION_JSON
 

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/cuqi/distribution/_distribution.py

@@ -105,7 +105,7 @@ class Distribution(Density, ABC):
                 f"Inconsistent distribution geometry attribute {self._geometry} and inferred "
                 f"dimension from distribution variables {inferred_dim}."
             )
-        
+
         # If Geometry dimension is None, update it with the inferred dimension
         if inferred_dim and self._geometry.par_dim is None: 
             self.geometry = inferred_dim
@@ -117,7 +117,7 @@ class Distribution(Density, ABC):
         # We do not use self.name to potentially infer it from python stack.
         if self._name: 
             self._geometry._variable_name = self._name
-            
+
         return self._geometry
 
     @geometry.setter
@@ -160,7 +160,7 @@ class Distribution(Density, ABC):
                     f"{self.logd.__qualname__}: To evaluate the log density all conditioning variables and main"
                     f" parameter must be specified. Conditioning variables are: {cond_vars}"
                 )
-            
+
             # Check if all conditioning variables are specified
             all_cond_vars_specified = all([key in kwargs for key in cond_vars])
             if not all_cond_vars_specified:
@@ -168,7 +168,7 @@ class Distribution(Density, ABC):
                     f"{self.logd.__qualname__}: To evaluate the log density all conditioning variables must be"
                     f" specified. Conditioning variables are: {cond_vars}"
                 )
-            
+
             # Extract exactly the conditioning variables from kwargs
             cond_kwargs = {key: kwargs[key] for key in cond_vars}
 
@@ -186,7 +186,7 @@ class Distribution(Density, ABC):
         # Not conditional distribution, simply evaluate log density directly
         else:
             return super().logd(*args, **kwargs)
-        
+
     def _logd(self, *args):
         return self.logpdf(*args) # Currently all distributions implement logpdf so we simply call this method.
 
@@ -216,7 +216,7 @@ class Distribution(Density, ABC):
         # Get samples from the distribution sample method
         s = self._sample(N,*args,**kwargs)
 
-        #Store samples in cuqi samples object if more than 1 sample
+        # Store samples in cuqi samples object if more than 1 sample
         if N==1:
             if len(s) == 1 and isinstance(s,np.ndarray): #Extract single value from numpy array
                 s = s.ravel()[0]
@@ -264,7 +264,7 @@ class Distribution(Density, ABC):
         # Go through every mutable variable and assign value from kwargs if present
         for var_key in mutable_vars:
 
-            #If keyword directly specifies new value of variable we simply reassign
+            # If keyword directly specifies new value of variable we simply reassign
             if var_key in kwargs:
                 setattr(new_dist, var_key, kwargs.get(var_key))
                 processed_kwargs.add(var_key)
@@ -291,9 +291,18 @@ class Distribution(Density, ABC):
 
                 elif len(var_args)>0:                      #Some keywords found
                     # Define new partial function with partially defined args
-                    func = partial(var_val, **var_args)
+                    if (
+                        hasattr(var_val, "_supports_partial_eval")
+                        and var_val._supports_partial_eval
+                    ):
+                        func = var_val(**var_args)
+                    else:
+                        # If the callable does not support partial evaluation,
+                        # we use the partial function to set the variable
+                        func = partial(var_val, **var_args)
+
                     setattr(new_dist, var_key, func)
-                
+
                 # Store processed keywords
                 processed_kwargs.update(var_args.keys())
 
@@ -329,7 +338,7 @@ class Distribution(Density, ABC):
 
     def get_conditioning_variables(self):
         """Return the conditioning variables of this distribution (if any)."""
-        
+
         # Get all mutable variables
         mutable_vars = self.get_mutable_variables()
 
@@ -338,7 +347,7 @@ class Distribution(Density, ABC):
 
         # Add any variables defined through callable functions
         cond_vars += get_indirect_variables(self)
-        
+
         return cond_vars
 
     def get_mutable_variables(self):
@@ -347,10 +356,10 @@ class Distribution(Density, ABC):
         # If mutable variables are already cached, return them
         if hasattr(self, '_mutable_vars'):
             return self._mutable_vars
-        
+
         # Define list of ignored attributes and properties
         ignore_vars = ['name', 'is_symmetric', 'geometry', 'dim']
-        
+
         # Get public attributes
         attributes = get_writeable_attributes(self)
 
@@ -396,7 +405,7 @@ class Distribution(Density, ABC):
                         raise ValueError(f"{self._condition.__qualname__}: {ordered_keys[index]} passed as both argument and keyword argument.\nArguments follow the listed conditioning variable order: {self.get_conditioning_variables()}")
                     kwargs[ordered_keys[index]] = arg
         return kwargs
-    
+
     def _check_geometry_consistency(self):
         """ Checks that the geometry of the distribution is consistent by calling the geometry property. Should be called at the end of __init__ of subclasses. """
         self.geometry
@@ -411,4 +420,4 @@ class Distribution(Density, ABC):
     def rv(self):
         """ Return a random variable object representing the distribution. """
         from cuqi.experimental.algebra import RandomVariable
-        return RandomVariable(self)
+        return RandomVariable(self)

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/cuqi/experimental/mcmc/_rto.py

@@ -36,21 +36,48 @@ class LinearRTO(Sampler):
     tol : float
         Tolerance of the inner CGLS solver. *Optional*.
 
+    inner_initial_point : string or np.ndarray or cuqi.array.CUQIArray
+        Initial point for the inner optimization problem. Can be "previous_sample" (default), "MAP", or a specific numpy or cuqi array. *Optional*.
+        
     callback : callable, optional
         A function that will be called after each sampling step. It can be useful for monitoring the sampler during sampling.
         The function should take three arguments: the sampler object, the index of the current sampling step, the total number of requested samples. The last two arguments are integers. An example of the callback function signature is: `callback(sampler, sample_index, num_of_samples)`.
         
     """
-    def __init__(self, target=None, initial_point=None, maxit=10, tol=1e-6, **kwargs):
+    def __init__(self, target=None, initial_point=None, maxit=10, tol=1e-6, inner_initial_point="previous_sample", **kwargs):
 
         super().__init__(target=target, initial_point=initial_point, **kwargs)
 
         # Other parameters
         self.maxit = maxit
         self.tol = tol
+        self.inner_initial_point = inner_initial_point
 
     def _initialize(self):
         self._precompute()
+        self._compute_map()
+
+    @property
+    def inner_initial_point(self):
+        if isinstance(self._inner_initial_point, str):
+            if self._inner_initial_point == "previous_sample":
+                return self.current_point
+            elif self._inner_initial_point == "map":
+                return self._map
+        else:
+            return self._inner_initial_point
+
+    @inner_initial_point.setter
+    def inner_initial_point(self, value):
+        is_correct_string = (isinstance(value, str) and
+                             (value.lower() == "previous_sample" or
+                              value.lower() == "map"))
+        if is_correct_string:
+            self._inner_initial_point = value.lower()
+        elif isinstance(value, (np.ndarray, cuqi.array.CUQIarray)):
+            self._inner_initial_point = value
+        else:
+            raise ValueError("Invalid value for inner_initial_point. Choose either 'previous_sample', 'MAP', or provide a numpy array/cuqi array.")
 
     @property
     def prior(self):
@@ -78,6 +105,10 @@ class LinearRTO(Sampler):
         elif isinstance(self.target, cuqi.distribution.MultipleLikelihoodPosterior):
             return self.target.models    
 
+    def _compute_map(self):
+        sim = CGLS(self.M, self.b_tild, self.current_point, self.maxit, self.tol)            
+        self._map, _ = sim.solve()
+
     def _precompute(self):
         L1 = [likelihood.distribution.sqrtprec for likelihood in self.likelihoods]
         L2 = self.prior.sqrtprec
@@ -114,7 +145,7 @@ class LinearRTO(Sampler):
 
     def step(self):
         y = self.b_tild + np.random.randn(len(self.b_tild))
-        sim = CGLS(self.M, y, self.current_point, self.maxit, self.tol)            
+        sim = CGLS(self.M, y, self.inner_initial_point, self.maxit, self.tol)            
         self.current_point, _ = sim.solve()
         acc = 1
         return acc
@@ -203,12 +234,15 @@ class RegularizedLinearRTO(LinearRTO):
     solver : string
         Options are "FISTA" (default for a single constraint or regularization), "ADMM" (default and the only option for multiple constraints or regularizations), "ScipyLinearLSQ" and "ScipyMinimizer". Note "ScipyLinearLSQ" and "ScipyMinimizer" can only be used with `RegularizedGaussian` of a single `box` or `nonnegativity` constraint. *Optional*.
 
+    inner_initial_point : string or np.ndarray or cuqi.array.CUQIArray
+        Initial point for the inner optimization problem. Can be "previous_sample" (default), "MAP", or a specific numpy or cuqi array. *Optional*.
+
     callback : callable, optional
         A function that will be called after each sampling step. It can be useful for monitoring the sampler during sampling.
         The function should take three arguments: the sampler object, the index of the current sampling step, the total number of requested samples. The last two arguments are integers. An example of the callback function signature is: `callback(sampler, sample_index, num_of_samples)`.
         
     """
-    def __init__(self, target=None, initial_point=None, maxit=100, inner_max_it=10, stepsize="automatic", penalty_parameter=10, abstol=1e-10, adaptive=True, solver=None, inner_abstol=None, **kwargs):
+    def __init__(self, target=None, initial_point=None, maxit=100, inner_max_it=10, stepsize="automatic", penalty_parameter=10, abstol=1e-10, adaptive=True, solver=None, inner_abstol=None, inner_initial_point="previous_sample", **kwargs):
         
         super().__init__(target=target, initial_point=initial_point, **kwargs)
 
@@ -221,6 +255,7 @@ class RegularizedLinearRTO(LinearRTO):
         self.inner_max_it = inner_max_it
         self.penalty_parameter = penalty_parameter
         self.solver = solver
+        self.inner_initial_point = inner_initial_point
 
     def _initialize(self):
         super()._initialize()
@@ -228,6 +263,7 @@ class RegularizedLinearRTO(LinearRTO):
             self.solver = "FISTA" if callable(self.proximal) else "ADMM"
         if self.solver == "FISTA":
             self._stepsize = self._choose_stepsize()
+        self._compute_map_regularized()
 
     @property
     def solver(self):
@@ -272,15 +308,16 @@ class RegularizedLinearRTO(LinearRTO):
     def prior(self):
         return self.target.prior.gaussian
 
-    def step(self):
-        y = self.b_tild + np.random.randn(len(self.b_tild))
+    def _compute_map_regularized(self):
+        self._map = self._customized_step(self.b_tild, self.initial_point)
 
+    def _customized_step(self, y, x0):
         if self.solver == "FISTA":
             sim = FISTA(self.M, y, self.proximal,
-                        self.current_point, maxit = self.maxit, stepsize = self._stepsize, abstol = self.abstol, adaptive = self.adaptive)         
+                        x0, maxit = self.maxit, stepsize = self._stepsize, abstol = self.abstol, adaptive = self.adaptive)         
         elif self.solver == "ADMM":
             sim = ADMM(self.M, y, self.proximal,
-                        self.current_point, self.penalty_parameter, maxit = self.maxit, inner_max_it = self.inner_max_it, adaptive = self.adaptive)
+                        x0, self.penalty_parameter, maxit = self.maxit, inner_max_it = self.inner_max_it, adaptive = self.adaptive)
         elif self.solver == "ScipyLinearLSQ":
             A_op = sp.sparse.linalg.LinearOperator((sum([llh.distribution.dim for llh in self.likelihoods])+self.target.prior.dim, self.target.prior.dim),
                                     matvec=lambda x: self.M(x, 1),
@@ -297,10 +334,17 @@ class RegularizedLinearRTO(LinearRTO):
             bounds = [(self.target.prior._box_bounds[0][i], self.target.prior._box_bounds[1][i]) for i in range(self.target.prior.dim)]
             # Note that the objective function is defined as 0.5*||Mx-y||^2, 
             # and the corresponding gradient (gradfunc) is given by M^T(Mx-y).
-            sim = ScipyMinimizer(lambda x: 0.5*np.sum((self.M(x, 1)-y)**2), self.current_point, gradfunc=lambda x: self.M(self.M(x, 1) - y, 2), bounds=bounds, tol=self.abstol, options={"maxiter": self.maxit})
+            sim = ScipyMinimizer(lambda x: 0.5*np.sum((self.M(x, 1)-y)**2), x0, gradfunc=lambda x: self.M(self.M(x, 1) - y, 2), bounds=bounds, tol=self.abstol, options={"maxiter": self.maxit})
         else:
             raise ValueError("Choice of solver not supported.")
+        
+        sol, _ = sim.solve()
+        return sol
+
+    def step(self):
+        y = self.b_tild + np.random.randn(len(self.b_tild))
+
+        self.current_point = self._customized_step(y, self.inner_initial_point)
 
-        self.current_point, _ = sim.solve()
         acc = 1
         return acc

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/cuqi/experimental/mcmc/_sampler.py

@@ -148,6 +148,16 @@ class Sampler(ABC):
         if self._target is not None:
             self.validate_target()
 
+    @property
+    def current_point(self):
+        """ The current point of the sampler. """
+        return self._current_point
+
+    @current_point.setter
+    def current_point(self, value):
+        """ Set the current point of the sampler. """
+        self._current_point = value
+
     # ------------ Public methods ------------
     def get_samples(self) -> Samples:
         """ Return the samples. The internal data-structure for the samples is a dynamic list so this creates a copy. """

{cuqipy-1.3.0.post0.dev300 → cuqipy-1.3.0.post0.dev371}/cuqi/likelihood/_likelihood.py

@@ -212,4 +212,4 @@ class UserDefinedLikelihood(object):
         return get_non_default_args(self.logpdf_func)
 
     def __repr__(self) -> str:
-        return "CUQI {} function. Parameters {}.".format(self.__class__.__name__,self.get_parameter_names())
+        return "CUQI {} function. Parameters {}.".format(self.__class__.__name__,self.get_parameter_names())