Hello community,

here is the log from the commit of package python-statsmodels for 
openSUSE:Factory checked in at 2018-08-18 00:03:37
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Comparing /work/SRC/openSUSE:Factory/python-statsmodels (Old)
 and      /work/SRC/openSUSE:Factory/.python-statsmodels.new (New)
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Package is "python-statsmodels"

Sat Aug 18 00:03:37 2018 rev:3 rq:629451 version:0.9.0

Changes:
--------
--- /work/SRC/openSUSE:Factory/python-statsmodels/python-statsmodels.changes    
2017-05-08 19:03:52.648825752 +0200
+++ 
/work/SRC/openSUSE:Factory/.python-statsmodels.new/python-statsmodels.changes   
    2018-08-18 00:03:38.178991736 +0200
@@ -1,0 +2,16 @@
+Tue Aug 14 16:57:55 UTC 2018 - [email protected]
+
+- Update to 0.9.0
+  Highlights:
+  * statespace refactoring, Markov Switching Kim smoother
+  * Bayesian mixed GLM
+  * Gaussian Imputation
+  * new multivariate methods: factor analysis, MANOVA, repeated measures 
within ANOVA
+  * GLM var_weights in addition to freq_weights
+  * Holt-Winters and Exponential Smoothing
+- Add pandas_to_datetime.patch
+  to_datetime has been moved in pandas.
+  Should be in next release
+  From: https://github.com/statsmodels/statsmodels/pull/4640
+
+-------------------------------------------------------------------

Old:
----
  statsmodels-0.8.0.tar.gz

New:
----
  pandas_to_datetime.patch
  statsmodels-0.9.0.tar.gz

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Other differences:
------------------
++++++ python-statsmodels.spec ++++++
--- /var/tmp/diff_new_pack.xFMSYe/_old  2018-08-18 00:03:39.966997078 +0200
+++ /var/tmp/diff_new_pack.xFMSYe/_new  2018-08-18 00:03:39.970997091 +0200
@@ -1,7 +1,7 @@
 #
 # spec file for package python-statsmodels
 #
-# Copyright (c) 2017 SUSE LINUX Products GmbH, Nuernberg, Germany.
+# Copyright (c) 2018 SUSE LINUX Products GmbH, Nuernberg, Germany.
 #
 # All modifications and additions to the file contributed by third parties
 # remain the property of their copyright owners, unless otherwise agreed
@@ -15,21 +15,18 @@
 # Please submit bugfixes or comments via http://bugs.opensuse.org/
 #
 
-%bcond_without tests
 
 %{?!python_module:%define python_module() python-%{**} python3-%{**}}
 Name:           python-statsmodels
-Version:        0.8.0
+Version:        0.9.0
 Release:        0
 Summary:        A Python module that allows users to explore data
 License:        BSD-3-Clause
 Group:          Development/Libraries/Python
 Url:            http://statsmodels.sourceforge.net/
 Source:         
https://files.pythonhosted.org/packages/source/s/statsmodels/statsmodels-%{version}.tar.gz
-BuildRequires:  fdupes
-BuildRequires:  gcc-fortran
-BuildRequires:  fdupes
-BuildRequires:  python-rpm-macros
+# PATCH-FIX-UPSTREAM pandas_to_datetime.patch -- to_datetime has been moved in 
pandas -- https://github.com/statsmodels/statsmodels/pull/4640
+Patch0:         pandas_to_datetime.patch
 BuildRequires:  %{python_module devel}
 BuildRequires:  %{python_module setuptools >= 0.6}
 BuildRequires:  %{python_module matplotlib >= 1.0.0}
@@ -37,15 +34,19 @@
 BuildRequires:  %{python_module pandas >= 0.7.1}
 BuildRequires:  %{python_module patsy >= 0.3.0}
 BuildRequires:  %{python_module scipy >= 0.9.0}
-%if %{with tests}
+BuildRequires:  fdupes
+BuildRequires:  gcc-fortran
+BuildRequires:  python-rpm-macros
+# SECTION test requirements
 BuildRequires:  %{python_module nose}
-%endif
+BuildRequires:  %{python_module pytest}
+# /SECTION
 Requires:       python-numpy >= 1.7.0
 Requires:       python-pandas >= 0.7.1
 Requires:       python-patsy >= 0.3.0
 Requires:       python-scipy >= 0.9.0
 Recommends:     python-matplotlib >= 1.0.0
-BuildRoot:      %{_tmppath}/%{name}-%{version}-build
+
 %python_subpackages
 
 %description
@@ -59,11 +60,12 @@
 
 %prep
 %setup -q -n statsmodels-%{version}
+find . -type f -name "*.py" -exec sed -i 's/\r$//' {} \;
+%patch0 -p1
 find statsmodels -type f -name "*.py" -exec sed -i "/#! \/usr\/bin\/env 
python/d" {} \;
 find statsmodels -type f -name "*.py" -exec sed -i "/#!\/usr\/bin\/env 
python/d" {} \;
 find statsmodels -type f -name "*.py" -exec sed -i "/#! \/usr\/bin\/env 
python3/d" {} \;
 find statsmodels -type f -name "*.py" -exec sed -i "/#!\/usr\/bin\/env 
python3/d" {} \;
-find . -type f -name "*.py" -exec sed -i 's/\r$//' {} \;
 find . -type f -name "*.ipynb" -exec sed -i 's/\r$//' {} \;
 sed -i 's/\r$//' COPYRIGHTS.txt
 sed -i 's/\r$//' LICENSE.txt
@@ -90,9 +92,9 @@
 %endif
 
 %files %{python_files}
-%defattr(-,root,root)
-%doc COPYRIGHTS.txt LICENSE.txt README.rst README_l1.txt
+%doc README.rst README_l1.txt
 %doc examples/
+%license COPYRIGHTS.txt LICENSE.txt
 %{python_sitearch}/statsmodels/
 %{python_sitearch}/statsmodels-%{version}-py*.egg-info
 

++++++ pandas_to_datetime.patch ++++++
>From 0219b20103cd92c8716a815e7822383c319110b3 Mon Sep 17 00:00:00 2001
From: Matthew Brett <[email protected]>
Date: Wed, 16 May 2018 10:57:37 +0100
Subject: [PATCH 1/4] BF: DataTimeIndex.to_datetime removed in pandas

https://github.com/pandas-dev/pandas/commit/537c06d150d0942454494b08773567eb245dd200
removed the to_datetime method from DateTimeIndex.  Use pd.to_datetime
instead.
---
 statsmodels/tsa/tests/test_tsa_indexes.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/statsmodels/tsa/tests/test_tsa_indexes.py 
b/statsmodels/tsa/tests/test_tsa_indexes.py
index e46862554b..f5ad603a75 100644
--- a/statsmodels/tsa/tests/test_tsa_indexes.py
+++ b/statsmodels/tsa/tests/test_tsa_indexes.py
@@ -85,7 +85,7 @@ class RangeIndex(object):
     (pd.Series(x), y) for x, y in list_datestr_indexes]
 
 numpy_datetime_indexes = [
-    (x.to_datetime().to_pydatetime(), x.freq)
+    (pd.to_datetime(x).to_pydatetime(), x.freq)
     for x in base_date_indexes]
 list_datetime_indexes = [
     (x.tolist(), y) for x, y in numpy_datetime_indexes]

>From e0191a6366a7fe9ca6747054403c1206ffeed8db Mon Sep 17 00:00:00 2001
From: Matthew Brett <[email protected]>
Date: Wed, 16 May 2018 13:37:48 +0100
Subject: [PATCH 2/4] MAINT: slightly increase tolerance in Poisson test

See:
https://github.com/statsmodels/statsmodels/issues/4639#issuecomment-389493480
---
 statsmodels/genmod/tests/test_glm.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/statsmodels/genmod/tests/test_glm.py 
b/statsmodels/genmod/tests/test_glm.py
index 7a845c882a..1689793b34 100644
--- a/statsmodels/genmod/tests/test_glm.py
+++ b/statsmodels/genmod/tests/test_glm.py
@@ -204,7 +204,7 @@ def test_compare_discrete(self):
         score1 = res1.model.score(res1.params * 0.98)
         assert_allclose(score1, score_obs1.sum(0), atol=1e-20)
         score0 = res1.model.score(res1.params)
-        assert_allclose(score0, np.zeros(score_obs1.shape[1]), atol=1e-7)
+        assert_allclose(score0, np.zeros(score_obs1.shape[1]), atol=5e-7)
 
         hessian1 = res1.model.hessian(res1.params * 0.98, observed=False)
         hessiand = resd.model.hessian(resd.params * 0.98)

>From bd2985d781484c2a2ce47a3db107690bbe688acd Mon Sep 17 00:00:00 2001
From: Matthew Brett <[email protected]>
Date: Wed, 16 May 2018 14:16:11 +0100
Subject: [PATCH 3/4] BF: format keyword removed in pandas 0.23

It appears to have been ignored previously.

See: https://travis-ci.org/MacPython/statsmodels-wheels/jobs/379687450#L5821
---
 statsmodels/datasets/co2/data.py | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/statsmodels/datasets/co2/data.py b/statsmodels/datasets/co2/data.py
index 6cdb5e304e..c062897bf4 100644
--- a/statsmodels/datasets/co2/data.py
+++ b/statsmodels/datasets/co2/data.py
@@ -62,8 +62,7 @@ def load_pandas():
     data = load()
     # pandas <= 0.12.0 fails in the to_datetime regex on Python 3
     index = pd.DatetimeIndex(start=data.data['date'][0].decode('utf-8'),
-                             periods=len(data.data), format='%Y%m%d',
-                             freq='W-SAT')
+                             periods=len(data.data), freq='W-SAT')
     dataset = pd.DataFrame(data.data['co2'], index=index, columns=['co2'])
     #NOTE: this is how I got the missing values in co2.csv
     #new_index = pd.DatetimeIndex(start='1958-3-29', end=index[-1],

>From f1d5eddd44d61099a9f8f90c40c35716a8346cf7 Mon Sep 17 00:00:00 2001
From: Matthew Brett <[email protected]>
Date: Wed, 16 May 2018 14:23:26 +0100
Subject: [PATCH 4/4] BF: pandas more fussy about values, index lengths

Pass two values for length 2 index.

See:
https://travis-ci.org/MacPython/statsmodels-wheels/jobs/379687450#L6067
---
 statsmodels/tsa/statespace/tests/test_mlemodel.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/statsmodels/tsa/statespace/tests/test_mlemodel.py 
b/statsmodels/tsa/statespace/tests/test_mlemodel.py
index 360cfbd6be..bd0c1a565d 100644
--- a/statsmodels/tsa/statespace/tests/test_mlemodel.py
+++ b/statsmodels/tsa/statespace/tests/test_mlemodel.py
@@ -669,7 +669,7 @@ def test_pandas_endog():
     mod.filter([])
 
     # Example : pandas.Series, string datatype
-    endog = pd.Series(['a'], index=dates)
+    endog = pd.Series(['a', 'b'], index=dates)
     # raises error due to direct type casting check in Statsmodels base classes
     assert_raises(ValueError, check_endog, endog, **kwargs)
 
++++++ statsmodels-0.8.0.tar.gz -> statsmodels-0.9.0.tar.gz ++++++
/work/SRC/openSUSE:Factory/python-statsmodels/statsmodels-0.8.0.tar.gz 
/work/SRC/openSUSE:Factory/.python-statsmodels.new/statsmodels-0.9.0.tar.gz 
differ: char 5, line 1


Reply via email to