CLN: remove the internal Benchmark class

benchmarks/benchmarks/bench_app.py

@@ -1,13 +1,11 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark
-
 import numpy as np
 
 from six.moves import xrange
 
 
-class LaplaceInplace(Benchmark):
+class LaplaceInplace(object):
     params = ['inplace', 'normal']
     param_names = ['update']
 
@@ -53,7 +51,7 @@ def time_it(self, update):
         self.run()
 
 
-class MaxesOfDots(Benchmark):
+class MaxesOfDots(object):
     def setup(self):
         np.random.seed(1)
         nsubj = 5

benchmarks/benchmarks/bench_core.py

@@ -1,11 +1,9 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark
-
 import numpy as np
 
 
-class Core(Benchmark):
+class Core(object):
     def setup(self):
         self.l100 = range(100)
         self.l50 = range(50)
@@ -76,7 +74,7 @@ def time_tril_l10x10(self):
         np.tril(self.l10x10)
 
 
-class Temporaries(Benchmark):
+class Temporaries(object):
     def setup(self):
         self.amid = np.ones(50000)
         self.bmid = np.ones(50000)
@@ -96,7 +94,7 @@ def time_large2(self):
         (self.alarge + self.blarge) - 2
 
 
-class CorrConv(Benchmark):
+class CorrConv(object):
     params = [[50, 1000, 1e5],
               [10, 100, 1000, 1e4],
               ['valid', 'same', 'full']]
@@ -113,7 +111,7 @@ def time_convolve(self, size1, size2, mode):
         np.convolve(self.x1, self.x2, mode=mode)
 
 
-class CountNonzero(Benchmark):
+class CountNonzero(object):
     param_names = ['numaxes', 'size', 'dtype']
     params = [
         [1, 2, 3],
@@ -137,7 +135,7 @@ def time_count_nonzero_multi_axis(self, numaxes, size, dtype):
                 self.x.ndim - 1, self.x.ndim - 2))
 
 
-class PackBits(Benchmark):
+class PackBits(object):
     param_names = ['dtype']
     params = [[bool, np.uintp]]
     def setup(self, dtype):
@@ -154,7 +152,7 @@ def time_packbits_axis1(self, dtype):
         np.packbits(self.d2, axis=1)
 
 
-class UnpackBits(Benchmark):
+class UnpackBits(object):
     def setup(self):
         self.d = np.ones(10000, dtype=np.uint8)
         self.d2 = np.ones((200, 1000), dtype=np.uint8)
@@ -169,6 +167,6 @@ def time_unpackbits_axis1(self):
         np.unpackbits(self.d2, axis=1)
 
 
-class Indices(Benchmark):
+class Indices(object):
     def time_indices(self):
         np.indices((1000, 500))

benchmarks/benchmarks/bench_function_base.py

@@ -1,11 +1,9 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark
-
 import numpy as np
 
 
-class Histogram1D(Benchmark):
+class Histogram1D(object):
     def setup(self):
         self.d = np.linspace(0, 100, 100000)
 
@@ -19,7 +17,7 @@ def time_fine_binning(self):
         np.histogram(self.d, 10000, (0, 100))
 
 
-class Histogram2D(Benchmark):
+class Histogram2D(object):
     def setup(self):
         self.d = np.linspace(0, 100, 200000).reshape((-1,2))
 
@@ -33,7 +31,7 @@ def time_fine_binning(self):
         np.histogramdd(self.d, (10000, 10000), ((0, 100), (0, 100)))
 
 
-class Bincount(Benchmark):
+class Bincount(object):
     def setup(self):
         self.d = np.arange(80000, dtype=np.intp)
         self.e = self.d.astype(np.float64)
@@ -45,7 +43,7 @@ def time_weights(self):
         np.bincount(self.d, weights=self.e)
 
 
-class Median(Benchmark):
+class Median(object):
     def setup(self):
         self.e = np.arange(10000, dtype=np.float32)
         self.o = np.arange(10001, dtype=np.float32)
@@ -69,7 +67,7 @@ def time_odd_small(self):
         np.median(self.o[:500], overwrite_input=True)
 
 
-class Percentile(Benchmark):
+class Percentile(object):
     def setup(self):
         self.e = np.arange(10000, dtype=np.float32)
         self.o = np.arange(10001, dtype=np.float32)
@@ -81,7 +79,7 @@ def time_percentile(self):
         np.percentile(self.e, [25, 35, 55, 65, 75])
 
 
-class Select(Benchmark):
+class Select(object):
     def setup(self):
         self.d = np.arange(20000)
         self.e = self.d.copy()
@@ -95,7 +93,7 @@ def time_select_larger(self):
         np.select(self.cond_large, ([self.d, self.e] * 10))
 
 
-class Sort(Benchmark):
+class Sort(object):
     def setup(self):
         self.e = np.arange(10000, dtype=np.float32)
         self.o = np.arange(10001, dtype=np.float32)
@@ -138,7 +136,7 @@ def time_argsort_random(self):
         self.o.argsort()
 
 
-class Where(Benchmark):
+class Where(object):
     def setup(self):
         self.d = np.arange(20000)
         self.e = self.d.copy()

benchmarks/benchmarks/bench_indexing.py

@@ -1,6 +1,6 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark, get_squares_, get_indexes_, get_indexes_rand_
+from .common import get_squares_, get_indexes_, get_indexes_rand_
 
 from os.path import join as pjoin
 import shutil
@@ -11,7 +11,7 @@
 from tempfile import mkdtemp
 
 
-class Indexing(Benchmark):
+class Indexing(object):
     params = [["indexes_", "indexes_rand_"],
               ['I', ':,I', 'np.ix_(I, I)'],
               ['', '=1']]
@@ -38,7 +38,7 @@ def time_op(self, indexes, sel, op):
         self.func()
 
 
-class IndexingSeparate(Benchmark):
+class IndexingSeparate(object):
     def setup(self):
         self.tmp_dir = mkdtemp()
         self.fp = memmap(pjoin(self.tmp_dir, 'tmp.dat'),
@@ -58,7 +58,7 @@ def time_mmap_fancy_indexing(self):
             self.fp[self.indexes]
 
 
-class IndexingStructured0D(Benchmark):
+class IndexingStructured0D(object):
     def setup(self):
         self.dt = np.dtype([('a', 'f4', 256)])
 

benchmarks/benchmarks/bench_io.py

@@ -1,12 +1,12 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark, get_squares
+from .common import get_squares
 
 import numpy as np
 from io import StringIO
 
 
-class Copy(Benchmark):
+class Copy(object):
     params = ["int8", "int16", "float32", "float64",
               "complex64", "complex128"]
     param_names = ['type']
@@ -31,7 +31,7 @@ def time_strided_assign(self, typename):
         self.dflat[::2] = 2
 
 
-class CopyTo(Benchmark):
+class CopyTo(object):
     def setup(self):
         self.d = np.ones(50000)
         self.e = self.d.copy()
@@ -57,14 +57,14 @@ def time_copyto_8_dense(self):
         np.copyto(self.d, self.e, where=self.im8)
 
 
-class Savez(Benchmark):
+class Savez(object):
     def setup(self):
         self.squares = get_squares()
 
     def time_vb_savez_squares(self):
         np.savez('tmp.npz', self.squares)
 
-class LoadtxtCSVComments(Benchmark):
+class LoadtxtCSVComments(object):
     # benchmarks for np.loadtxt comment handling
     # when reading in CSV files
 
@@ -93,7 +93,7 @@ def time_comment_loadtxt_csv(self, num_lines):
                    delimiter=u',')
         self.data_comments.seek(0)
 
-class LoadtxtCSVdtypes(Benchmark):
+class LoadtxtCSVdtypes(object):
     # benchmarks for np.loadtxt operating with
     # different dtypes parsed / cast from CSV files
 
@@ -118,7 +118,7 @@ def time_loadtxt_dtypes_csv(self, dtype, num_lines):
                    dtype=dtype)
         self.csv_data.seek(0)
 
-class LoadtxtCSVStructured(Benchmark):
+class LoadtxtCSVStructured(object):
     # benchmarks for np.loadtxt operating with
     # a structured data type & CSV file
 
@@ -141,7 +141,7 @@ def time_loadtxt_csv_struct_dtype(self):
         self.csv_data.seek(0)
 
 
-class LoadtxtCSVSkipRows(Benchmark):
+class LoadtxtCSVSkipRows(object):
     # benchmarks for loadtxt row skipping when
     # reading in csv file data; a similar benchmark
     # is present in the pandas asv suite
@@ -162,7 +162,7 @@ def time_skiprows_csv(self, skiprows):
                    delimiter=',',
                    skiprows=skiprows)
 
-class LoadtxtReadUint64Integers(Benchmark):
+class LoadtxtReadUint64Integers(object):
     # pandas has a similar CSV reading benchmark
     # modified to suit np.loadtxt
 
@@ -188,7 +188,7 @@ def time_read_uint64_neg_values(self, size):
         np.loadtxt(self.data2)
         self.data2.seek(0)
 
-class LoadtxtUseColsCSV(Benchmark):
+class LoadtxtUseColsCSV(object):
     # benchmark selective column reading from CSV files
     # using np.loadtxt
 
@@ -208,7 +208,7 @@ def time_loadtxt_usecols_csv(self, usecols):
                    usecols=usecols)
         self.csv_data.seek(0)
 
-class LoadtxtCSVDateTime(Benchmark):
+class LoadtxtCSVDateTime(object):
     # benchmarks for np.loadtxt operating with
     # datetime data in a CSV file
 

benchmarks/benchmarks/bench_lib.py

@@ -1,15 +1,13 @@
-"""Benchmarks for `numpy.lib`."""
+"""objects for `numpy.lib`."""
 
 
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark
-
 import numpy as np
 
 
-class Pad(Benchmark):
-    """Benchmarks for `numpy.pad`."""
+class Pad(object):
+    """objects for `numpy.pad`."""
 
     param_names = ["shape", "pad_width", "mode"]
     params = [

benchmarks/benchmarks/bench_linalg.py

@@ -1,11 +1,11 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark, get_squares_, get_indexes_rand, TYPES1
+from .common import get_squares_, get_indexes_rand, TYPES1
 
 import numpy as np
 
 
-class Eindot(Benchmark):
+class Eindot(object):
     def setup(self):
         self.a = np.arange(60000.0).reshape(150, 400)
         self.ac = self.a.copy()
@@ -73,7 +73,7 @@ def time_tensordot_a_b_axes_1_0_0_1(self):
         np.tensordot(self.a3, self.b3, axes=([1, 0], [0, 1]))
 
 
-class Linalg(Benchmark):
+class Linalg(object):
     params = [['svd', 'pinv', 'det', 'norm'],
               TYPES1]
     param_names = ['op', 'type']
@@ -100,7 +100,7 @@ def time_op(self, op, typename):
         self.func(self.a)
 
 
-class Lstsq(Benchmark):
+class Lstsq(object):
     def setup(self):
         self.a = get_squares_()['float64']
         self.b = get_indexes_rand()[:100].astype(np.float64)

benchmarks/benchmarks/bench_ma.py

@@ -1,11 +1,9 @@
 from __future__ import absolute_import, division, print_function
 
-from .common import Benchmark
-
 import numpy as np
 
 
-class MA(Benchmark):
+class MA(object):
     def setup(self):
         self.l100 = range(100)
         self.t100 = ([True] * 100)
@@ -20,7 +18,7 @@ def time_masked_array_l100_t100(self):
         np.ma.masked_array(self.l100, self.t100)
 
 
-class Indexing(Benchmark):
+class Indexing(object):
     param_names = ['masked', 'ndim', 'size']
     params = [[True, False],
               [1, 2],
@@ -47,7 +45,7 @@ def time_1d(self, masked, ndim, size):
         self.m[self.idx_1d]
 
 
-class UFunc(Benchmark):
+class UFunc(object):
     param_names = ['a_masked', 'b_masked', 'size']
     params = [[True, False],
               [True, False],
@@ -79,7 +77,7 @@ def time_2d(self, a_masked, b_masked, size):
         np.ma.add(self.a_2d, self.b_2d)
 
 
-class Concatenate(Benchmark):
+class Concatenate(object):
     param_names = ['mode', 'n']
     params = [
         ['ndarray', 'unmasked',