CUQIpy 1.1.1.post0.dev90__tar.gz → 1.2.0__tar.gz

{cuqipy-1.1.1.post0.dev90 → cuqipy-1.2.0}/CUQIpy.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: CUQIpy
-Version: 1.1.1.post0.dev90
+Version: 1.2.0
 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-1.1.1.post0.dev90 → cuqipy-1.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: CUQIpy
-Version: 1.1.1.post0.dev90
+Version: 1.2.0
 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-1.1.1.post0.dev90 → cuqipy-1.2.0}/cuqi/_version.py

@@ -8,11 +8,11 @@ import json
 
 version_json = '''
 {
- "date": "2024-09-13T08:36:06+0200",
+ "date": "2024-09-13T10:02:02+0200",
  "dirty": false,
  "error": null,
- "full-revisionid": "22a43b19a36622c11b96970eff2e8ab6776a7142",
- "version": "1.1.1.post0.dev90"
+ "full-revisionid": "59ae0d0a693ec03a3282cffb9dc148e0729b3468",
+ "version": "1.2.0"
 }
 '''  # END VERSION_JSON
 

{cuqipy-1.1.1.post0.dev90 → cuqipy-1.2.0}/cuqi/experimental/mcmc/_cwmh.py

@@ -158,7 +158,9 @@ class CWMH(ProposalBasedSampler):
 
             # accept/reject
             u_theta = np.log(np.random.rand())
-            if (u_theta <= alpha): # accept
+            if (u_theta <= alpha) and \
+               (not np.isnan(target_eval_star)) and \
+               (not np.isinf(target_eval_star)):
                 x_t[j] = x_all_components[j]
                 target_eval_t = target_eval_star
                 acc[j] = 1

{cuqipy-1.1.1.post0.dev90 → cuqipy-1.2.0}/cuqi/experimental/mcmc/_hmc.py

@@ -256,7 +256,10 @@ class NUTS(Sampler):
 
             # Metropolis step
             alpha2 = min(1, (n_prime/n)) #min(0, np.log(n_p) - np.log(n))
-            if (s_prime == 1) and (np.random.rand() <= alpha2):
+            if (s_prime == 1) and \
+                (np.random.rand() <= alpha2) and \
+                (not np.isnan(logd_prime)) and \
+                (not np.isinf(logd_prime)):
                 self.current_point = point_prime
                 self.current_target_logd = logd_prime
                 self.current_target_grad = np.copy(grad_prime)

{cuqipy-1.1.1.post0.dev90 → cuqipy-1.2.0}/cuqi/experimental/mcmc/_langevin_algorithm.py

@@ -102,10 +102,14 @@ class ULA(Sampler): # Refactor to Proposal-based sampler?
         scalar
             1 (accepted)
         """
-        self.current_point = x_star
-        self.current_target_logd = target_eval_star
-        self.current_target_grad = target_grad_star
-        acc = 1
+        acc = 0
+        if (not np.isnan(target_eval_star)) and \
+           (not np.isinf(target_eval_star)):
+            self.current_point = x_star
+            self.current_target_logd = target_eval_star
+            self.current_target_grad = target_grad_star
+            acc = 1
+            
         return acc
 
     def step(self):
@@ -211,7 +215,9 @@ class MALA(ULA): # Refactor to Proposal-based sampler?
         # accept/reject with Metropolis
         acc = 0
         log_u = np.log(np.random.rand())
-        if (log_u <= log_alpha) and (np.isnan(target_eval_star) == False):
+        if (log_u <= log_alpha) and \
+           (not np.isnan(target_eval_star)) and \
+           (not np.isinf(target_eval_star)):
             self.current_point = x_star
             self.current_target_logd = target_eval_star
             self.current_target_grad = target_grad_star

{cuqipy-1.1.1.post0.dev90 → cuqipy-1.2.0}/cuqi/experimental/mcmc/_mh.py

@@ -57,7 +57,9 @@ class MH(ProposalBasedSampler):
         # accept/reject
         u_theta = np.log(np.random.rand())
         acc = 0
-        if (u_theta <= alpha):
+        if (u_theta <= alpha) and \
+           (not np.isnan(target_eval_star)) and \
+           (not np.isinf(target_eval_star)):
             self.current_point = x_star
             self.current_target_logd = target_eval_star
             acc = 1