CI: Run ASV on Travis for failed benchmarks

create asv.sh

add to travis

Changed file permission

Add else statement and asv machine config

move to separate travis build

CI: Run ASV on Travis for failed benchmarks

create asv.sh

add to travis

move to separate travis build

Change build order and change to asv dev

Put asv before doc

Fix some failing benchmarks

Grep for failed benchmarks

Add travis wait

Don't grep output yet

Remove travis_wait for now

fix some benchmarks

Now grep output

add travis wait 40 to ci/asv.sh

Add tee

Remove wait

Author: mroeschke

.travis.yml

@@ -73,6 +73,10 @@ matrix:
       env:
         - JOB="3.6_NUMPY_DEV" TEST_ARGS="--skip-slow --skip-network" PANDAS_TESTING_MODE="deprecate"
     # In allow_failures
+    - dist: trusty
+      env:
+        - JOB="3.6_ASV" ASV=true
+    # In allow_failures
     - dist: trusty
       env:
         - JOB="3.6_DOC" DOC=true
@@ -93,6 +97,9 @@ matrix:
       - dist: trusty
         env:
           - JOB="3.6_NUMPY_DEV" TEST_ARGS="--skip-slow --skip-network" PANDAS_TESTING_MODE="deprecate"
+      - dist: trusty
+        env:
+          - JOB="3.6_ASV" ASV=true
       - dist: trusty
         env:
           - JOB="3.6_DOC" DOC=true
@@ -128,6 +135,7 @@ script:
   - ci/script_single.sh
   - ci/script_multi.sh
   - ci/lint.sh
+  - ci/asv.sh
   - echo "checking imports"
   - source activate pandas && python ci/check_imports.py
   - echo "script done"

asv_bench/benchmarks/frame_methods.py

@@ -141,8 +141,8 @@ class Repr(object):
     def setup(self):
         nrows = 10000
         data = np.random.randn(nrows, 10)
-        idx = MultiIndex.from_arrays(np.tile(np.random.randn(3, nrows / 100),
-                                             100))
+        arrays = np.tile(np.random.randn(3, int(nrows / 100)), 100)
+        idx = MultiIndex.from_arrays(arrays)
         self.df3 = DataFrame(data, index=idx)
         self.df4 = DataFrame(data, index=np.random.randn(nrows))
         self.df_tall = DataFrame(np.random.randn(nrows, 10))

asv_bench/benchmarks/groupby.py

@@ -467,7 +467,7 @@ class SumMultiLevel(object):
 
     def setup(self):
         N = 50
-        self.df = DataFrame({'A': range(N) * 2,
+        self.df = DataFrame({'A': list(range(N)) * 2,
                              'B': range(N * 2),
                              'C': 1}).set_index(['A', 'B'])
 

asv_bench/benchmarks/indexing.py

@@ -295,7 +295,8 @@ def setup(self):
     def time_insert(self):
         np.random.seed(1234)
         for i in range(100):
-            self.df.insert(0, i, np.random.randn(self.N))
+            self.df.insert(0, i, np.random.randn(self.N),
+                           allow_duplicates=True)
 
     def time_assign_with_setitem(self):
         np.random.seed(1234)

asv_bench/benchmarks/reindex.py

@@ -167,10 +167,6 @@ def setup(self):
         col_array2 = col_array.copy()
         col_array2[:, :10000] = np.nan
         self.col_array_list = list(col_array)
-        self.col_array_list2 = list(col_array2)
 
     def time_lib_fast_zip(self):
         lib.fast_zip(self.col_array_list)
-
-    def time_lib_fast_zip_fillna(self):
-        lib.fast_zip_fillna(self.col_array_list2)

asv_bench/benchmarks/reshape.py

@@ -104,9 +104,9 @@ def setup(self):
         self.letters = list('ABCD')
         yrvars = [l + str(num)
                   for l, num in product(self.letters, range(1, nyrs + 1))]
-
+        columns = [str(i) for i in range(nidvars)] + yrvars
         self.df = DataFrame(np.random.randn(N, nidvars + len(yrvars)),
-                            columns=list(range(nidvars)) + yrvars)
+                            columns=columns)
         self.df['id'] = self.df.index
 
     def time_wide_to_long_big(self):

asv_bench/benchmarks/timeseries.py

@@ -365,7 +365,7 @@ class ToDatetimeCache(object):
 
     def setup(self, cache):
         N = 10000
-        self.unique_numeric_seconds = range(N)
+        self.unique_numeric_seconds = list(range(N))
         self.dup_numeric_seconds = [1000] * N
         self.dup_string_dates = ['2000-02-11'] * N
         self.dup_string_with_tz = ['2000-02-11 15:00:00-0800'] * N

ci/asv.sh

@@ -0,0 +1,27 @@
+#!/bin/bash
+
+echo "inside $0"
+
+source activate pandas
+
+if [ "$ASV" ]; then
+    echo "Check for failed asv benchmarks"
+
+    cd asv_bench
+
+    asv machine --yes
+
+    RET=0
+
+    time asv dev | tee >(grep "failed")
+
+    if [ $? = "0" ]; then
+        RET=1
+    fi
+    echo "Check for failed asv benchmarks DONE"
+else
+    RET=0
+    echo "NOT checking for failed asv benchmarks"
+fi
+
+exit $RET

ci/requirements-3.6_ASV.build

@@ -0,0 +1,5 @@
+python=3.6*
+python-dateutil
+pytz
+numpy=1.13*
+cython

ci/requirements-3.6_ASV.run

@@ -0,0 +1,25 @@
+ipython
+ipykernel
+ipywidgets
+sphinx=1.5*
+nbconvert
+nbformat
+notebook
+matplotlib
+seaborn
+scipy
+lxml
+beautifulsoup4
+html5lib
+pytables
+python-snappy
+openpyxl
+xlrd
+xlwt
+xlsxwriter
+sqlalchemy
+numexpr
+bottleneck
+statsmodels
+xarray
+pyqt

ci/requirements-3.6_ASV.sh

@@ -0,0 +1,7 @@
+#!/bin/bash
+
+source activate pandas
+
+echo "[install ASV_BUILD deps]"
+
+pip install git+https://github.com/spacetelescope/asv

ci/script_multi.sh

@@ -37,6 +37,9 @@ if [ "$PIP_BUILD_TEST" ] || [ "$CONDA_BUILD_TEST" ]; then
 elif [ "$DOC" ]; then
     echo "We are not running pytest as this is a doc-build"
 
+elif [ "$ASV" ]; then
+    echo "We are not running pytest as this is an asv-build"
+
 elif [ "$COVERAGE" ]; then
     echo pytest -s -n 2 -m "not single" --cov=pandas --cov-report xml:/tmp/cov-multiple.xml --junitxml=/tmp/multiple.xml --strict $TEST_ARGS pandas
     pytest -s -n 2 -m "not single" --cov=pandas --cov-report xml:/tmp/cov-multiple.xml --junitxml=/tmp/multiple.xml --strict $TEST_ARGS pandas

ci/script_single.sh

@@ -22,6 +22,9 @@ if [ "$PIP_BUILD_TEST" ] || [ "$CONDA_BUILD_TEST" ]; then
 elif [ "$DOC" ]; then
     echo "We are not running pytest as this is a doc-build"
 
+elif [ "$ASV" ]; then
+    echo "We are not running pytest as this is an asv-build"
+
 elif [ "$COVERAGE" ]; then
     echo pytest -s -m "single" --strict --cov=pandas --cov-report xml:/tmp/cov-single.xml --junitxml=/tmp/single.xml $TEST_ARGS pandas
     pytest -s -m "single" --strict --cov=pandas --cov-report xml:/tmp/cov-single.xml --junitxml=/tmp/single.xml $TEST_ARGS pandas