def _standard_normal(shape, randstate=np.random, dtype=np.float_):
"""Generates a standard normal numpy array of given shape and dtype, i.e.
this function is equivalent to `randstate.randn(*shape)` for real dtype and
`randstate.randn(*shape) + 1.j * randstate.randn(shape)` for complex dtype.
:param tuple shape: Shape of array to be returned
:param randstate: An instance of :class:`numpy.random.RandomState` (default is
``np.random``))
:param dtype: ``np.float_`` (default) or `np.complex_`
Returns
-------
A: An array of given shape and dtype with standard normal entries
"""
if dtype == np.float_:
return randstate.randn(*shape)
elif dtype == np.complex_:
return randstate.randn(*shape) + 1.j * randstate.randn(*shape)
else:
raise ValueError('{} is not a valid dtype.'.format(dtype))
python类float_()的实例源码
def test_operations_typesafety(nr_sites, local_dim, rank, rgen):
# create a real MPA
mpo1 = factory.random_mpa(nr_sites, (local_dim, local_dim), rank,
randstate=rgen, dtype=np.float_)
mpo2 = factory.random_mpa(nr_sites, (local_dim, local_dim), rank,
randstate=rgen, dtype=np.complex_)
assert mpo1.dtype == np.float_
assert mpo2.dtype == np.complex_
assert (mpo1 + mpo1).dtype == np.float_
assert (mpo1 + mpo2).dtype == np.complex_
assert (mpo2 + mpo1).dtype == np.complex_
assert mp.sumup((mpo1, mpo1)).dtype == np.float_
assert mp.sumup((mpo1, mpo2)).dtype == np.complex_
assert mp.sumup((mpo2, mpo1)).dtype == np.complex_
assert (mpo1 - mpo1).dtype == np.float_
assert (mpo1 - mpo2).dtype == np.complex_
assert (mpo2 - mpo1).dtype == np.complex_
mpo1 += mpo2
assert mpo1.dtype == np.complex_
def __init__(self, data, bucket_size=128):
if bucket_size < 1:
raise ValueError("A minimum bucket size of 1 is expected.")
self._data = data
self._n, self._k = self._data.shape
self._nodes = None
self._buckets = []
self._bucket_size = bucket_size
self._node_dtype = numpy.dtype([
('size', numpy.intp),
('bucket', numpy.intp),
('lower_bounds', (numpy.float_, self._k)),
('upper_bounds', (numpy.float_, self._k)),
])
self._neighbour_dtype = numpy.dtype([
('squared_distance', numpy.float_),
('index', numpy.intp),
])
self._build()
def search(self, point, count, radius, sort):
"""Retrieve the neighbours to a point."""
if count is None:
count = self._n
elif count < 1:
return numpy.empty(0, dtype=self._neighbour_dtype)
if radius is None:
radius = numpy.inf
elif radius < 0.0:
return numpy.empty(0, dtype=self._neighbour_dtype)
point = numpy.asarray(point, dtype=numpy.float_)
if count >= self._n:
return self._search_all_within_radius(point, radius, sort)
else:
return self._search_k_nearests(point, count, radius, sort)
def test_empty_tuple_index(self):
# Empty tuple index creates a view
a = np.array([1, 2, 3])
assert_equal(a[()], a)
assert_(a[()].base is a)
a = np.array(0)
assert_(isinstance(a[()], np.int_))
# Regression, it needs to fall through integer and fancy indexing
# cases, so need the with statement to ignore the non-integer error.
with warnings.catch_warnings():
warnings.filterwarnings('ignore', '', DeprecationWarning)
a = np.array([1.])
assert_(isinstance(a[0.], np.float_))
a = np.array([np.array(1)], dtype=object)
assert_(isinstance(a[0.], np.ndarray))
def approx(a, b, fill_value=True, rtol=1e-5, atol=1e-8):
"""
Returns true if all components of a and b are equal to given tolerances.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal. The relative error rtol should
be positive and << 1.0 The absolute error atol comes into play for
those elements of b that are very small or zero; it says how small a
must be also.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.less_equal(umath.absolute(x - y), atol + rtol * umath.absolute(y))
return d.ravel()
def _fix_type(value):
"""convert possible types to str, float, and bool"""
# Because numpy floats can not be pickled to json
if isinstance(value, string_types):
return str(value)
if isinstance(value, float_):
return float(value)
if isinstance(value, bool_):
return bool(value)
if isinstance(value, set):
return list(value)
if isinstance(value, Basic):
return str(value)
if hasattr(value, 'id'):
return str(value.id)
# if value is None:
# return ''
return value
def default(self, obj):
# convert dates and numpy objects in a json serializable format
if isinstance(obj, datetime):
return obj.strftime('%Y-%m-%dT%H:%M:%SZ')
elif isinstance(obj, date):
return obj.strftime('%Y-%m-%d')
elif type(obj) in (np.int_, np.intc, np.intp, np.int8, np.int16,
np.int32, np.int64, np.uint8, np.uint16,
np.uint32, np.uint64):
return int(obj)
elif type(obj) in (np.bool_,):
return bool(obj)
elif type(obj) in (np.float_, np.float16, np.float32, np.float64,
np.complex_, np.complex64, np.complex128):
return float(obj)
# Let the base class default method raise the TypeError
return json.JSONEncoder.default(self, obj)
def event_bounds_expressions(self, event_bounds_exp):
if hasattr(self, 'output_equations'):
assert len(event_bounds_exp)+1 == self.output_equations.shape[0]
if hasattr(self, 'output_equations_functions'):
assert len(event_bounds_exp)+1 == \
self.output_equations_functions.size
if hasattr(self, 'state_equations'):
assert len(event_bounds_exp)+1 == self.state_equations.shape[0]
if hasattr(self, 'state_equations_functions'):
assert len(event_bounds_exp)+1 == \
self.state_equations_functions.size
self._event_bounds_expressions = event_bounds_exp
self.event_bounds = np.array(
[sp.N(bound, subs=self.constants_values)
for bound in event_bounds_exp],
dtype=np.float_
)
def test_empty_tuple_index(self):
# Empty tuple index creates a view
a = np.array([1, 2, 3])
assert_equal(a[()], a)
assert_(a[()].base is a)
a = np.array(0)
assert_(isinstance(a[()], np.int_))
# Regression, it needs to fall through integer and fancy indexing
# cases, so need the with statement to ignore the non-integer error.
with warnings.catch_warnings():
warnings.filterwarnings('ignore', '', DeprecationWarning)
a = np.array([1.])
assert_(isinstance(a[0.], np.float_))
a = np.array([np.array(1)], dtype=object)
assert_(isinstance(a[0.], np.ndarray))
def approx(a, b, fill_value=True, rtol=1e-5, atol=1e-8):
"""
Returns true if all components of a and b are equal to given tolerances.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal. The relative error rtol should
be positive and << 1.0 The absolute error atol comes into play for
those elements of b that are very small or zero; it says how small a
must be also.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.less_equal(umath.absolute(x - y), atol + rtol * umath.absolute(y))
return d.ravel()
def find_a_dominant_color(image):
# K-mean clustering to find the k most dominant color, from:
# http://stackoverflow.com/questions/3241929/python-find-dominant-most-common-color-in-an-image
n_clusters = 5
# Get image into a workable form
im = image.copy()
im = im.resize((150, 150)) # optional, to reduce time
ar = scipy.misc.fromimage(im)
im_shape = ar.shape
ar = ar.reshape(scipy.product(im_shape[:2]), im_shape[2])
ar = np.float_(ar)
# Compute clusters
codes, dist = scipy.cluster.vq.kmeans(ar, n_clusters)
vecs, dist = scipy.cluster.vq.vq(ar, codes) # assign codes
counts, bins = scipy.histogram(vecs, len(codes)) # count occurrences
# Get the indexes of the most frequent, 2nd most frequent, 3rd, ...
sorted_idxs = np.argsort(counts)
# Get the color
peak = codes[sorted_idxs[1]] # get second most frequent color
return [int(i) for i in peak.tolist()] # list comprehension to quickly cast everything to int
test_base.py 文件源码
项目：PyDataLondon29-EmbarrassinglyParallelDAWithAWSLambda
作者: SignalMedia
项目源码
文件源码
阅读 22
收藏 0
点赞 0
评论 0
def test_empty_fancy(self):
empty_farr = np.array([], dtype=np.float_)
empty_iarr = np.array([], dtype=np.int_)
empty_barr = np.array([], dtype=np.bool_)
# pd.DatetimeIndex is excluded, because it overrides getitem and should
# be tested separately.
for idx in [self.strIndex, self.intIndex, self.floatIndex]:
empty_idx = idx.__class__([])
self.assertTrue(idx[[]].identical(empty_idx))
self.assertTrue(idx[empty_iarr].identical(empty_idx))
self.assertTrue(idx[empty_barr].identical(empty_idx))
# np.ndarray only accepts ndarray of int & bool dtypes, so should
# Index.
self.assertRaises(IndexError, idx.__getitem__, empty_farr)
test_constructors.py 文件源码
项目：PyDataLondon29-EmbarrassinglyParallelDAWithAWSLambda
作者: SignalMedia
项目源码
文件源码
阅读 25
收藏 0
点赞 0
评论 0
def test_fromValue(self):
nans = Series(np.NaN, index=self.ts.index)
self.assertEqual(nans.dtype, np.float_)
self.assertEqual(len(nans), len(self.ts))
strings = Series('foo', index=self.ts.index)
self.assertEqual(strings.dtype, np.object_)
self.assertEqual(len(strings), len(self.ts))
d = datetime.now()
dates = Series(d, index=self.ts.index)
self.assertEqual(dates.dtype, 'M8[ns]')
self.assertEqual(len(dates), len(self.ts))
# GH12336
# Test construction of categorical series from value
categorical = Series(0, index=self.ts.index, dtype="category")
expected = Series(0, index=self.ts.index).astype("category")
self.assertEqual(categorical.dtype, 'category')
self.assertEqual(len(categorical), len(self.ts))
tm.assert_series_equal(categorical, expected)
test_indexing.py 文件源码
项目：PyDataLondon29-EmbarrassinglyParallelDAWithAWSLambda
作者: SignalMedia
项目源码
文件源码
阅读 29
收藏 0
点赞 0
评论 0
def test_empty_tuple_index(self):
# Empty tuple index creates a view
a = np.array([1, 2, 3])
assert_equal(a[()], a)
assert_(a[()].base is a)
a = np.array(0)
assert_(isinstance(a[()], np.int_))
# Regression, it needs to fall through integer and fancy indexing
# cases, so need the with statement to ignore the non-integer error.
with warnings.catch_warnings():
warnings.filterwarnings('ignore', '', DeprecationWarning)
a = np.array([1.])
assert_(isinstance(a[0.], np.float_))
a = np.array([np.array(1)], dtype=object)
assert_(isinstance(a[0.], np.ndarray))
testutils.py 文件源码
项目：PyDataLondon29-EmbarrassinglyParallelDAWithAWSLambda
作者: SignalMedia
项目源码
文件源码
阅读 33
收藏 0
点赞 0
评论 0
def almost(a, b, decimal=6, fill_value=True):
"""
Returns True if a and b are equal up to decimal places.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.around(np.abs(x - y), decimal) <= 10.0 ** (-decimal)
return d.ravel()
def test_empty_tuple_index(self):
# Empty tuple index creates a view
a = np.array([1, 2, 3])
assert_equal(a[()], a)
assert_(a[()].base is a)
a = np.array(0)
assert_(isinstance(a[()], np.int_))
# Regression, it needs to fall through integer and fancy indexing
# cases, so need the with statement to ignore the non-integer error.
with warnings.catch_warnings():
warnings.filterwarnings('ignore', '', DeprecationWarning)
a = np.array([1.])
assert_(isinstance(a[0.], np.float_))
a = np.array([np.array(1)], dtype=object)
assert_(isinstance(a[0.], np.ndarray))
def approx(a, b, fill_value=True, rtol=1e-5, atol=1e-8):
"""
Returns true if all components of a and b are equal to given tolerances.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal. The relative error rtol should
be positive and << 1.0 The absolute error atol comes into play for
those elements of b that are very small or zero; it says how small a
must be also.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.less_equal(umath.absolute(x - y), atol + rtol * umath.absolute(y))
return d.ravel()
def test_empty_tuple_index(self):
# Empty tuple index creates a view
a = np.array([1, 2, 3])
assert_equal(a[()], a)
assert_(a[()].base is a)
a = np.array(0)
assert_(isinstance(a[()], np.int_))
# Regression, it needs to fall through integer and fancy indexing
# cases, so need the with statement to ignore the non-integer error.
with warnings.catch_warnings():
warnings.filterwarnings('ignore', '', DeprecationWarning)
a = np.array([1.])
assert_(isinstance(a[0.], np.float_))
a = np.array([np.array(1)], dtype=object)
assert_(isinstance(a[0.], np.ndarray))
def approx(a, b, fill_value=True, rtol=1e-5, atol=1e-8):
"""
Returns true if all components of a and b are equal to given tolerances.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal. The relative error rtol should
be positive and << 1.0 The absolute error atol comes into play for
those elements of b that are very small or zero; it says how small a
must be also.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.less_equal(umath.absolute(x - y), atol + rtol * umath.absolute(y))
return d.ravel()
def kl_divergence(p, q):
"""
Returns KL-divergence of distribution q from distribution p.
The Kullback-Leibler (KL) divergence is defined as
.. math::
\\textrm{KL-divergence}(p, q) :=
\\sum_{x} p(x) \\log{} \\frac{p(x)}{q(x)}
Warning: this function uses numpy's scalar floating point types to
perform the evaluation. Therefore, the result may be non-finite.
For example, if the state x has non-zero probability for distribution p,
but zero probability for distribution q, then the result will be
non-finite.
"""
accum = 0.0
for x in p:
p_x = numpy.float_(p[x])
if p_x != 0.0:
q_x = numpy.float_(q.get(x, 0.0))
accum += p_x * numpy.log(p_x / q_x)
return accum
def default(self, obj):
# convert dates and numpy objects in a json serializable format
if isinstance(obj, datetime):
return obj.strftime('%Y-%m-%dT%H:%M:%SZ')
elif isinstance(obj, date):
return obj.strftime('%Y-%m-%d')
elif type(obj) in [np.int_, np.intc, np.intp, np.int8, np.int16,
np.int32, np.int64, np.uint8, np.uint16,
np.uint32, np.uint64]:
return int(obj)
elif type(obj) in [np.bool_]:
return bool(obj)
elif type(obj) in [np.float_, np.float16, np.float32, np.float64,
np.complex_, np.complex64, np.complex128]:
return float(obj)
# Let the base class default method raise the TypeError
return json.JSONEncoder.default(self, obj)
def approx(a, b, fill_value=True, rtol=1e-5, atol=1e-8):
"""
Returns true if all components of a and b are equal to given tolerances.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal. The relative error rtol should
be positive and << 1.0 The absolute error atol comes into play for
those elements of b that are very small or zero; it says how small a
must be also.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.less_equal(umath.absolute(x - y), atol + rtol * umath.absolute(y))
return d.ravel()
def almost(a, b, decimal=6, fill_value=True):
"""
Returns True if a and b are equal up to decimal places.
If fill_value is True, masked values considered equal. Otherwise,
masked values are considered unequal.
"""
m = mask_or(getmask(a), getmask(b))
d1 = filled(a)
d2 = filled(b)
if d1.dtype.char == "O" or d2.dtype.char == "O":
return np.equal(d1, d2).ravel()
x = filled(masked_array(d1, copy=False, mask=m), fill_value).astype(float_)
y = filled(masked_array(d2, copy=False, mask=m), 1).astype(float_)
d = np.around(np.abs(x - y), decimal) <= 10.0 ** (-decimal)
return d.ravel()
def achisquare(f_obs,f_exp=None):
"""
Calculates a one-way chi square for array of observed frequencies and returns
the result. If no expected frequencies are given, the total N is assumed to
be equally distributed across all groups (NOT RIGHT??)
Usage: achisquare(f_obs, f_exp=None) f_obs = array of observed cell freq.
Returns: chisquare-statistic, associated p-value
"""
k = len(f_obs)
if f_exp == None:
f_exp = N.array([sum(f_obs)/float(k)] * len(f_obs),N.float_)
f_exp = f_exp.astype(N.float_)
chisq = N.add.reduce((f_obs-f_exp)**2 / f_exp)
return chisq, achisqprob(chisq, k-1)
def asquare_of_sums(inarray, dimension=None, keepdims=0):
"""
Adds the values in the passed array, squares that sum, and returns the
result. Dimension can equal None (ravel array first), an integer (the
dimension over which to operate), or a sequence (operate over multiple
dimensions). If keepdims=1, the returned array will have the same
NUMBER of dimensions as the original.
Usage: asquare_of_sums(inarray, dimension=None, keepdims=0)
Returns: the square of the sum over dim(s) in dimension
"""
if dimension == None:
inarray = N.ravel(inarray)
dimension = 0
s = asum(inarray,dimension,keepdims)
if type(s) == N.ndarray:
return s.astype(N.float_)*s
else:
return float(s)*s
def arankdata(inarray):
"""
Ranks the data in inarray, dealing with ties appropritely. Assumes
a 1D inarray. Adapted from Gary Perlman's |Stat ranksort.
Usage: arankdata(inarray)
Returns: array of length equal to inarray, containing rank scores
"""
n = len(inarray)
svec, ivec = ashellsort(inarray)
sumranks = 0
dupcount = 0
newarray = N.zeros(n,N.float_)
for i in range(n):
sumranks = sumranks + i
dupcount = dupcount + 1
if i==n-1 or svec[i] <> svec[i+1]:
averank = sumranks / float(dupcount) + 1
for j in range(i-dupcount+1,i+1):
newarray[ivec[j]] = averank
sumranks = 0
dupcount = 0
return newarray
def load_data(path, seq_length):
with open(path) as file:
content = file.read().strip()
key = sorted(list(set(content)))
dataX = []
dataY = []
for i in range(0, len(content) - seq_length, 1):
seq_in = content[i:i+seq_length]
seq_out = content[i+seq_length]
dataX.append(encode_vals(seq_in, key))
dataY.append(encode(seq_out, key))
X = np.reshape(dataX, (len(dataX), seq_length, len(key)))
X = np.float_(X)
Y = np.asarray(dataY)
return (X, Y, key)
def npy2py_type(npy_type):
int_types = [
np.int_, np.intc, np.intp, np.int8, np.int16, np.int32, np.int64,
np.uint8, np.uint16, np.uint32, np.uint64
]
float_types = [np.float_, np.float16, np.float32, np.float64]
bytes_types = [np.str_, np.string_]
if npy_type in int_types:
return int
if npy_type in float_types:
return float
if npy_type in bytes_types:
return bytes
if hasattr(npy_type, 'char'):
if npy_type.char in ['S', 'a']:
return bytes
raise TypeError
return npy_type
def apply_rescaling(self, frm_in, frm_out, n_avg, virtual_param):
"""Apply rescaling and averaging operations."""
frm_out.i = frm_in.i
# --- perform averaging on histograms
val = np.float_(self.factor) / np.float_(n_avg)
# --- rescale distance histograms
if frm_out.has_key(base.loc_histograms):
X = frm_out.get_data(base.loc_histograms)
dict_util.scale_values(X, val)
# --- multiref: rescale shell XX histogram
if (virtual_param is not None and self.geometry == 'MultiReferenceStructure'):
if frm_out.has_key(base.loc_shell_Hxx):
X = frm_out.get_data(base.loc_shell_Hxx)
dict_util.scale_values(X, val)
# --- rescale length histograms
if frm_out.has_key(base.loc_len_histograms):
X = frm_out.get_data(base.loc_len_histograms)
dict_util.scale_values(X, val)
# ---
frm_out.put_data('log', frm_in.get_data('log'))
frm_out.put_meta(self.get_meta(n_avg=n_avg))
