def test__getitem__(self):
"Dataset: operator 'a[3:5], a[3]'"
#1D dataset
ds = self.Dataset(name="distance",
unit="meter",
attributes={},
shape=[12],
storage=NdArray('double', list(range(12))))
#slicing
self.assertTrue(ds[3:5].storage().compare(NdArray('double', [3, 4])))
#indexing
self.assertAlmostEqual(ds[3] / meter, 3)
#2D dataset
stor = NdArray('double', list(range(12)))
stor.setShape((3, 4))
ds = self.Dataset(name="distance",
unit="meter",
attributes={},
shape=[3, 4],
storage=stor)
#slicing
subarr = ds[1:2, 1:2].storage()
expected = NdArray('double', [5])
expected.setShape((1, 1))
self.assertTrue(subarr.compare(expected))
#indexing
self.assertAlmostEqual(ds[1, 1] / meter, 5)
return
def test_createDataset(self):
"histogram.__init__: createDataset"
name = "intensity"
unit = "1"
shape = [100,200]
types = [ 'double', 'float', 'int', 'unsigned' ]
for datatype in types:
try:
ds = createDataset( name, unit, shape=shape, data_type = datatype )
except:
raise RuntimeError, datatype
self.assertEqual( name, ds.name() )
self.assertEqual( 1, ds.unit() )
self.assertVectorEqual( shape, ds.shape() )
self.assertEqual( ds.typecodeAsC(), datatype )
continue
from histogram.ndarray.NumpyNdArray import NdArray
storage = NdArray( "float", 20000, 0 )
storage.setShape( shape )
ds2 = createDataset( name, unit, storage = storage )
self.assertEqual( name, ds2.name() )
self.assertEqual( 1, ds2.unit() )
self.assertVectorEqual( shape, ds2.shape() )
self.assertEqual( ds2.typecodeAsC(), "float" )
return
def test_createDataset(self):
"histogram.__init__: createDataset"
name = "intensity"
unit = "1"
shape = [100, 200]
types = ['double', 'float', 'int', 'unsigned']
for datatype in types:
try:
ds = createDataset(name, unit, shape=shape, data_type=datatype)
except:
raise RuntimeError(datatype)
self.assertEqual(name, ds.name())
self.assertEqual(1, ds.unit())
self.assertVectorEqual(shape, ds.shape())
self.assertEqual(ds.typecodeAsC(), datatype)
continue
from histogram.ndarray.NumpyNdArray import NdArray
storage = NdArray("float", 20000, 0)
storage.setShape(shape)
ds2 = createDataset(name, unit, storage=storage)
self.assertEqual(name, ds2.name())
self.assertEqual(1, ds2.unit())
self.assertVectorEqual(shape, ds2.shape())
self.assertEqual(ds2.typecodeAsC(), "float")
return
def test__setitem__(self):
"Dataset: operator 'a[3]=4'"
#1D dataset
ds = self.Dataset(name="distance",
unit="meter",
attributes={},
shape=[12],
storage=NdArray('double', list(range(12))))
#set slice
from numpy import array
ds[3:5] = array([1, 2]) * meter
self.assertTrue(ds[3:5].storage().compare(NdArray('double', [1, 2])))
#set item
ds[10] = 11 * meter
self.assertAlmostEqual(ds[10] / meter, 11)
#2D dataset
stor = NdArray('double', list(range(12)))
stor.setShape((3, 4))
ds = self.Dataset(name="distance",
unit="meter",
attributes={},
shape=[3, 4],
storage=stor)
#slicing
ds[1:2, 1:2] = array([[999]]) * meter
subarr = ds[1:2, 1:2].storage()
expected = NdArray('double', [999])
expected.setShape((1, 1))
self.assertTrue(subarr.compare(expected))
#indexing
self.assertAlmostEqual(ds[1, 1] / meter, 999)
ds[1, 1] = 333 * meter
self.assertAlmostEqual(ds[1, 1] / meter, 333)
#set slice with datasets
ds1v = NdArray('double', list(range(12)))
ds1v.setShape((3, 4))
ds1 = self.Dataset(name="distance",
unit="meter",
shape=[3, 4],
storage=ds1v)
ds2v = NdArray('double', list(range(4)))
ds2v.setShape((2, 2))
ds2 = self.Dataset(name="distance",
unit="meter",
shape=[2, 2],
storage=ds2v)
ds1[1:3, 1:3] = ds2
#set slice with one number
ds1v = NdArray('double', list(range(12)))
ds1v.setShape((3, 4))
ds1 = self.Dataset(name="distance",
unit="meter",
shape=[3, 4],
storage=ds1v)
ds1[1:3, 1:3] = 1 * meter
return
def testSum(self):
"Dataset: method 'sum'"
storage = NdArray('double', list(range(6)))
storage.setShape((2, 3))
ds = self.Dataset(name="distance",
unit="meter",
attributes={},
shape=[2, 3],
storage=storage)
self.assertAlmostEqual(ds.sum() / meter, 15.)
ds1 = ds.sum(0)
expected = NdArray('double', [3, 5, 7])
self.assertTrue(ds1.storage().compare(expected))
return
def test_1(**kwds):
from histogram.Axis import Axis
from histogram.NdArrayDataset import Dataset
name = 'test'
unit = '1'
attributes = {
'plottable': True,
'nifty': False,
'pi': 3.14159,
3.14159: 'pi'
}
lengths = [2, 3, 4]
dtype = 6 # double
stor1 = NdArray(dtype, lengths[0] + 1, 1.0)
ax1 = Axis(name + 'ax1', unit, attributes, lengths[0], stor1)
stor2 = NdArray(dtype, lengths[1] + 1, 1.0)
ax2 = Axis(name + 'ax2', unit, attributes, lengths[1], stor2)
stor3 = NdArray(dtype, lengths[2] + 1, 1.0)
ax3 = Axis(name + 'ax3', unit, attributes, lengths[2], stor3)
size = lengths[0] * lengths[1] * lengths[2]
dataStore = NdArray(dtype, size, 1.0)
dataStore.setShape((2, 3, 4))
data = Dataset(name + 'data', unit, attributes, lengths, dataStore)
errorStore = NdArray(dtype, size, 1.0)
errorStore.setShape((2, 3, 4))
error = Dataset(name + 'error', unit, attributes, lengths, errorStore)
from histogram.Histogram import Histogram
hist = Histogram('testHist', data, error, [ax1, ax2, ax3], attributes)
passed = True
hax1 = hist.axisFromId(1)
if hax1 is not ax1:
passed = False
log("hax1 was %s instead of %s" % (hax1, ax1))
return passed
def test_0(**kwds):
from histogram.Axis import Axis
from histogram.NdArrayDataset import Dataset
name = 'test'
unit = '1'
attributes = {
'plottable': True,
'nifty': False,
'pi': 3.14159,
3.14159: 'pi'
}
lengths = [2, 3, 4]
dtype = 6 # double
stor1 = NdArray(dtype, lengths[0] + 1, 1.0)
ax1 = Axis(name + 'ax1', unit, attributes, lengths[0], stor1)
stor2 = NdArray(dtype, lengths[1] + 1, 1.0)
ax2 = Axis(name + 'ax2', unit, attributes, lengths[1], stor2)
stor3 = NdArray(dtype, lengths[2] + 1, 1.0)
ax3 = Axis(name + 'ax3', unit, attributes, lengths[2], stor3)
size = lengths[0] * lengths[1] * lengths[2]
dataStore = NdArray(dtype, size, 1.0)
dataStore.setShape((2, 3, 4))
data = Dataset(name + 'data', unit, attributes, lengths, dataStore)
errorStore = NdArray(dtype, size, 1.0)
errorStore.setShape((2, 3, 4))
error = Dataset(name + 'error', unit, attributes, lengths, errorStore)
from histogram.Histogram import Histogram
histogram = Histogram('testHist', data, error, [ax1, ax2, ax3], attributes)
return True