Merge branch 'master' of https://github.com/gboehl/pydsge

.pre-commit-config.yaml

@@ -29,34 +29,34 @@ repos:
     #hooks:
     #-   id: blacken-docs
         #additional_dependencies: [black]
--   repo: https://github.com/PyCQA/flake8
-    rev: 3.9.2
-    hooks:
-    -   id: flake8
-        args: # arguments to configure flake8
-        # making flake8 compatible with Black
-          - "--max-line-length=88"
-        # ignore errors for Black compatability
-        # E203: ignore whitespace around " : "
-        # T000: ignore TODOs
-        # W503: ignore linebreak before binary operator
-        # PT5006: ignore tuples as default for pytest parametrize
-          - "--ignore=E203,T000,W503,PT006,"
-        additional_dependencies: [
-            flake8-alfred,
-            flake8-bugbear,
-            flake8-builtins,
-            flake8-comprehensions,
-            flake8-docstrings,
-            flake8-eradicate,
-            flake8-print,
-            flake8-pytest-style,
-            flake8-todo,
-            flake8-unused-arguments,
-            pep8-naming,
-            pydocstyle,
-            Pygments,
-        ]
+#-   repo: https://github.com/PyCQA/flake8
+    #rev: 3.9.2
+    #hooks:
+    #-   id: flake8
+        #args: # arguments to configure flake8
+        ## making flake8 compatible with Black
+          #- "--max-line-length=88"
+        ## ignore errors for Black compatability
+        ## E203: ignore whitespace around " : "
+        ## T000: ignore TODOs
+        ## W503: ignore linebreak before binary operator
+        ## PT5006: ignore tuples as default for pytest parametrize
+          #- "--ignore=E203,T000,W503,PT006,"
+        #additional_dependencies: [
+            #flake8-alfred,
+            #flake8-bugbear,
+            #flake8-builtins,
+            #flake8-comprehensions,
+            #flake8-docstrings,
+            #flake8-eradicate,
+            #flake8-print,
+            #flake8-pytest-style,
+            #flake8-todo,
+            #flake8-unused-arguments,
+            #pep8-naming,
+            #pydocstyle,
+            #Pygments,
+        #]
 # -   repo: https://github.com/PyCQA/doc8
 #     rev: 0.9.0
 #     hooks:

README.rst

@@ -12,7 +12,7 @@ pydsge
 
 Contains the functions and classes for solving, filtering and estimating DSGE models at the ZLB (or with other occasionally binding constraints).
 
-A collection of models that can (and were) used with this package can be found in `another repo <https://github.com/gboehl/projectlib/tree/master/yamls>`_.
+A collection of models that can be (and were) used with this package can be found in `another repo <https://github.com/gboehl/projectlib/tree/master/yamls>`_.
 
 Installation
 -------------

docs/modules.rst

@@ -36,13 +36,6 @@ Module Documentation
    :undoc-members:
    :show-inheritance:
 
-``pydsge.partfilt``
--------------------
-.. automodule:: pydsge.partfilt
-   :members:
-   :undoc-members:
-   :show-inheritance:
-
 ``pydsge.symbols``
 ------------------
 .. automodule:: pydsge.symbols

pydsge/clsmethods.py

@@ -375,6 +375,9 @@ def load_data(self, df, start=None, end=None):
 
     d = d.loc[start:end]
 
+    if np.any(d.isna()):
+        raise Exception("Data must not contain `NaN`s.")
+
     self.data = d
     self.fdict["data"] = cpickle.dumps(d, protocol=4)
     self.fdict["obs"] = self.observables

pydsge/stats.py

@@ -543,7 +543,7 @@ def mdd_lp(chain, lprobs, calc_hess=False):
     return mdd
 
 
-def mdd_mhm(chain, lprobs, alpha=0.05, verbose=False, debug=False):
+def mdd_mhm(chain, lprobs, pool=None, alpha=0.05, verbose=False, debug=False):
     """Approximate the marginal data density useing modified harmonic mean."""
 
     from grgrlib.stats import logpdf
@@ -572,14 +572,8 @@ def runner(chunk):
                 res[i] = -np.inf
         return res
 
-    if not debug:
-        if hasattr(self, "pool"):
-            from .estimation import create_pool
-
-            create_pool(self)
-
-        nbatches = self.pool.ncpus
-
+    if not debug and pool is not None:
+        nbatches = pool.ncpus
         batches = pool.imap(runner, np.array_split(chain, nbatches))
         mls = np.vstack(list(batches))
     else: