Merge branch 'master' into gh12541

Author: mralgos

appveyor.yml

@@ -65,7 +65,8 @@ install:
 
   # install our build environment
   - cmd: conda config --set show_channel_urls true --set always_yes true --set changeps1 false
-  - cmd: conda update -q conda
+  #- cmd: conda update -q conda
+  - cmd: conda install conda=4.2.15
   - cmd: conda config --set ssl_verify false
 
   # add the pandas channel *before* defaults to have defaults take priority

asv_bench/benchmarks/frame_methods.py

@@ -433,6 +433,20 @@ def time_frame_from_records_generator_nrows(self):
 
 
 
+#-----------------------------------------------------------------------------
+# nunique
+
+class frame_nunique(object):
+
+    def setup(self):
+        self.data = np.random.randn(10000, 1000)
+        self.df = DataFrame(self.data)
+
+    def time_frame_nunique(self):
+        self.df.nunique()
+
+
+
 #-----------------------------------------------------------------------------
 # duplicated
 

asv_bench/benchmarks/groupby.py

@@ -251,6 +251,22 @@ def time_groupby_int_count(self):
         self.df.groupby(['key1', 'key2']).count()
 
 
+#----------------------------------------------------------------------
+# nunique() speed
+
+class groupby_nunique(object):
+
+    def setup(self):
+        self.n = 10000
+        self.df = DataFrame({'key1': randint(0, 500, size=self.n),
+                             'key2': randint(0, 100, size=self.n),
+                             'ints': randint(0, 1000, size=self.n),
+                             'ints2': randint(0, 1000, size=self.n), })
+
+    def time_groupby_nunique(self):
+        self.df.groupby(['key1', 'key2']).nunique()
+
+
 #----------------------------------------------------------------------
 # group with different functions per column
 

ci/ironcache/get.py

@@ -8,10 +8,9 @@ RET=0
 
 if [ "$LINT" ]; then
 
-    # pandas/rpy is deprecated and will be removed.
     # pandas/src is C code, so no need to search there.
     echo "Linting  *.py"
-    flake8 pandas --filename=*.py --exclude pandas/rpy,pandas/src
+    flake8 pandas --filename=*.py --exclude pandas/src
     if [ $? -ne "0" ]; then
         RET=1
     fi

ci/ironcache/put.py

@@ -3,8 +3,8 @@
 ls "$HOME/.cache/"
 
 PYX_CACHE_DIR="$HOME/.cache/pyxfiles"
-pyx_file_list=`find ${TRAVIS_BUILD_DIR} -name "*.pyx" -o -name "*.pxd"`
-pyx_cache_file_list=`find ${PYX_CACHE_DIR} -name "*.pyx" -o -name "*.pxd"`
+pyx_file_list=`find ${TRAVIS_BUILD_DIR} -name "*.pyx" -o -name "*.pxd" -o -name "*.pxi.in"`
+pyx_cache_file_list=`find ${PYX_CACHE_DIR} -name "*.pyx" -o -name "*.pxd" -o -name "*.pxi.in"`
 
 CACHE_File="$HOME/.cache/cython_files.tar"
 

ci/lint.sh

@@ -1,4 +1,4 @@
 python-dateutil
 pytz
-numpy=1.9.3
+numpy=1.10*
 cython

ci/prep_cython_cache.sh

@@ -1,6 +1,6 @@
 python-dateutil
 pytz
-numpy=1.9.3
+numpy=1.10*
 openpyxl
 xlsxwriter
 xlrd

ci/requirements-3.4_SLOW.build

@@ -4,4 +4,4 @@ source activate pandas
 
 echo "install 34_slow"
 
-conda install -n pandas -c conda-forge/label/rc -c conda-forge matplotlib
+conda install -n pandas -c conda-forge matplotlib

ci/requirements-3.4_SLOW.run

@@ -1 +0,0 @@
-xlwt

ci/requirements-3.4_SLOW.sh

@@ -5,10 +5,10 @@ scipy
 openpyxl
 xlsxwriter
 xlrd
-# xlwt (installed via pip)
+xlwt
 numexpr
 pytables
-# matplotlib (not avail on defaults ATM)
+matplotlib
 lxml
 html5lib
 jinja2