def select_or_selectable_to_frame(el, bind=None, dshape=None, **kwargs):
bind = getbind(el, bind)
if bind.dialect.name == 'postgresql':
buf = StringIO()
append(CSV(None, buffer=buf), el, bind=bind, **kwargs)
buf.seek(0)
datetime_fields = []
other_dtypes = {}
optional_string_fields = []
try:
fields = dshape.measure.fields
except AttributeError:
fields = [(0, dshape.measure)]
for field, dtype in fields:
if isdatelike(dtype):
datetime_fields.append(field)
elif isinstance(dtype, Option):
ty = dtype.ty
if ty in datashape.integral:
other_dtypes[field] = 'float64'
else:
other_dtypes[field] = ty.to_numpy_dtype()
if ty == string:
optional_string_fields.append(field)
else:
other_dtypes[field] = dtype.to_numpy_dtype()
df = pd.read_csv(
buf,
parse_dates=datetime_fields,
dtype=other_dtypes,
skip_blank_lines=False,
escapechar=kwargs.get('escapechar', '\\'),
)
# read_csv really wants missing values to be NaN, but for
# string (object) columns, we want None to be missing
for field in optional_string_fields:
df.loc[df[field].isnull(), field] = None
return df
columns, rows = batch(el, bind=bind)
dtypes = {}
try:
fields = dshape.measure.fields
except AttributeError:
fields = [(columns[0], dshape.measure)]
for field, dtype in fields:
if isinstance(dtype, Option):
ty = dtype.ty
if ty in datashape.integral:
dtypes[field] = 'float64'
else:
dtypes[field] = ty.to_numpy_dtype()
else:
dtypes[field] = dtype.to_numpy_dtype()
return pd.DataFrame(
np.array(list(map(tuple, rows)),
dtype=[(str(c), dtypes[c]) for c in columns]))
""" Vector norm
See np.linalg.norm
"""
if ord is None or ord == 'fro':
ord = 2
if ord == inf:
return max(abs(expr), axis=axis, keepdims=keepdims)
elif ord == -inf:
return min(abs(expr), axis=axis, keepdims=keepdims)
elif ord == 1:
return sum(abs(expr), axis=axis, keepdims=keepdims)
elif ord % 2 == 0:
return sum(expr**ord, axis=axis, keepdims=keepdims)**(1.0 / ord)
return sum(abs(expr)**ord, axis=axis, keepdims=keepdims)**(1.0 / ord)
dshape_method_list.extend([
(iscollection, set([count, nelements])),
(lambda ds:
(iscollection(ds) and
(isstring(ds) or isnumeric(ds) or isboolean(ds) or isdatelike(ds))),
set([min, max])),
(lambda ds: len(ds.shape) == 1, set([nrows, nunique])),
(lambda ds: iscollection(ds) and isboolean(ds), set([any, all])),
(lambda ds: iscollection(ds) and (isnumeric(ds) or isboolean(ds)),
set([mean, sum, std, var, vnorm])),
])
method_properties.update([nrows])
See np.linalg.norm
"""
if ord is None or ord == 'fro':
ord = 2
if ord == inf:
return max(abs(expr), axis=axis, keepdims=keepdims)
elif ord == -inf:
return min(abs(expr), axis=axis, keepdims=keepdims)
elif ord == 1:
return sum(abs(expr), axis=axis, keepdims=keepdims)
elif ord % 2 == 0:
return sum(expr ** ord, axis=axis, keepdims=keepdims) ** (1.0 / ord)
return sum(abs(expr) ** ord, axis=axis, keepdims=keepdims) ** (1.0 / ord)
dshape_method_list.extend([
(iscollection, set([count, nelements])),
(lambda ds: (iscollection(ds) and
(isstring(ds) or isnumeric(ds) or isboolean(ds) or
isdatelike(ds) or isinstance(ds, TimeDelta))),
set([min, max])),
(lambda ds: len(ds.shape) == 1,
set([nrows, nunique])),
(lambda ds: iscollection(ds) and isboolean(ds),
set([any, all])),
(lambda ds: iscollection(ds) and (isnumeric(ds) or isboolean(ds)),
set([mean, sum, std, var, vnorm])),
])
method_properties.update([nrows])
See np.linalg.norm
"""
if ord is None or ord == 'fro':
ord = 2
if ord == inf:
return max(abs(expr), axis=axis, keepdims=keepdims)
elif ord == -inf:
return min(abs(expr), axis=axis, keepdims=keepdims)
elif ord == 1:
return sum(abs(expr), axis=axis, keepdims=keepdims)
elif ord % 2 == 0:
return sum(expr ** ord, axis=axis, keepdims=keepdims) ** (1.0 / ord)
return sum(abs(expr) ** ord, axis=axis, keepdims=keepdims) ** (1.0 / ord)
dshape_method_list.extend([
(iscollection, set([count, nelements])),
(lambda ds: (iscollection(ds) and
(isstring(ds) or isnumeric(ds) or isboolean(ds) or
isdatelike(ds))),
set([min, max])),
(lambda ds: len(ds.shape) == 1,
set([nrows, nunique])),
(lambda ds: iscollection(ds) and isboolean(ds),
set([any, all])),
(lambda ds: iscollection(ds) and (isnumeric(ds) or isboolean(ds)),
set([mean, sum, std, var, vnorm])),
])
method_properties.update([nrows])
""" Vector norm
See np.linalg.norm
"""
if ord is None or ord == 'fro':
ord = 2
if ord == inf:
return max(abs(expr), axis=axis, keepdims=keepdims)
elif ord == -inf:
return min(abs(expr), axis=axis, keepdims=keepdims)
elif ord == 1:
return sum(abs(expr), axis=axis, keepdims=keepdims)
elif ord % 2 == 0:
return sum(expr**ord, axis=axis, keepdims=keepdims)**(1.0 / ord)
return sum(abs(expr)**ord, axis=axis, keepdims=keepdims)**(1.0 / ord)
dshape_method_list.extend([
(iscollection, set([count, nelements])),
(lambda ds:
(iscollection(ds) and (isstring(ds) or isnumeric(ds) or isboolean(
ds) or isdatelike(ds) or isinstance(ds, TimeDelta))), set([min,
max])),
(lambda ds: len(ds.shape) == 1, set([nrows, nunique])),
(lambda ds: iscollection(ds) and isboolean(ds), set([any, all])),
(lambda ds: iscollection(ds) and (isnumeric(ds) or isboolean(ds)),
set([mean, sum, std, var, vnorm])),
])
method_properties.update([nrows])
def select_or_selectable_to_frame(el, bind=None, dshape=None, **kwargs):
bind = getbind(el, bind)
if bind.dialect.name == 'postgresql':
buf = StringIO()
append(CSV(None, buffer=buf), el, bind=bind, **kwargs)
buf.seek(0)
datetime_fields = []
other_dtypes = {}
optional_string_fields = []
try:
fields = dshape.measure.fields
except AttributeError:
fields = [(0, dshape.measure)]
for n, (field, dtype) in enumerate(fields):
if isdatelike(dtype):
datetime_fields.append(field)
elif isinstance(dtype, Option):
ty = dtype.ty
if ty in datashape.integral:
other_dtypes[field] = 'float64'
else:
other_dtypes[field] = ty.to_numpy_dtype()
if ty == string:
# work with integer column indices for the
# optional_string columns because we don't always
# know the column name and then the lookup will fail
# in the loop below.
optional_string_fields.append(n)
else:
other_dtypes[field] = dtype.to_numpy_dtype()
df = pd.read_csv(
buf,
parse_dates=datetime_fields,
dtype=other_dtypes,
skip_blank_lines=False,
escapechar=kwargs.get('escapechar', '\\'),
)
# read_csv really wants missing values to be NaN, but for
# string (object) columns, we want None to be missing
columns = df.columns
for field_ix in optional_string_fields:
# use ``df.loc[bool, df.columns[field_ix]]`` because you cannot do
# boolean slicing with ``df.iloc``.
field = columns[field_ix]
df.loc[df[field].isnull(), field] = None
return df
columns, rows = batch(el, bind=bind)
dtypes = {}
try:
fields = dshape.measure.fields
except AttributeError:
fields = [(columns[0], dshape.measure)]
for field, dtype in fields:
if isinstance(dtype, Option):
ty = dtype.ty
if ty in datashape.integral:
dtypes[field] = 'float64'
else:
dtypes[field] = ty.to_numpy_dtype()
else:
dtypes[field] = dtype.to_numpy_dtype()
return pd.DataFrame(np.array(list(map(tuple, rows)),
dtype=[(str(c), dtypes[c]) for c in columns]))
def select_or_selectable_to_frame(el, bind=None, dshape=None, **kwargs):
bind = getbind(el, bind)
if bind.dialect.name == 'postgresql':
buf = StringIO()
append(CSV(None, buffer=buf), el, bind=bind, **kwargs)
buf.seek(0)
datetime_fields = []
other_dtypes = {}
optional_string_fields = []
try:
fields = dshape.measure.fields
except AttributeError:
fields = [(0, dshape.measure)]
for n, (field, dtype) in enumerate(fields):
if isdatelike(dtype):
datetime_fields.append(field)
elif isinstance(dtype, Option):
ty = dtype.ty
if ty in datashape.integral:
other_dtypes[field] = 'float64'
else:
other_dtypes[field] = ty.to_numpy_dtype()
if ty == string:
# work with integer column indices for the
# optional_string columns because we don't always
# know the column name and then the lookup will fail
# in the loop below.
optional_string_fields.append(n)
else:
other_dtypes[field] = dtype.to_numpy_dtype()
df = pd.read_csv(
buf,
parse_dates=datetime_fields,
dtype=other_dtypes,
skip_blank_lines=False,
escapechar=kwargs.get('escapechar', '\\'),
)
# read_csv really wants missing values to be NaN, but for
# string (object) columns, we want None to be missing
columns = df.columns
for field_ix in optional_string_fields:
# use ``df.loc[bool, df.columns[field_ix]]`` because you cannot do
# boolean slicing with ``df.iloc``.
field = columns[field_ix]
df.loc[df[field].isnull(), field] = None
return df
columns, rows = batch(el, bind=bind)
dtypes = {}
try:
fields = dshape.measure.fields
except AttributeError:
fields = [(columns[0], dshape.measure)]
for field, dtype in fields:
if isinstance(dtype, Option):
ty = dtype.ty
if ty in datashape.integral:
dtypes[field] = 'float64'
else:
dtypes[field] = ty.to_numpy_dtype()
else:
dtypes[field] = dtype.to_numpy_dtype()
return pd.DataFrame(
np.array(list(map(tuple, rows)),
dtype=[(str(c), dtypes[c]) for c in columns]))
