Source code for foxes.models.vertical_profiles.data_profile
import numpy as np
import pandas as pd
from scipy.interpolate import interp1d
from foxes.core import VerticalProfile
[docs]
class DataProfile(VerticalProfile):
"""
A profile based on numerical data.
Attributes
----------
var: float
The value
data_z: numpy.ndarray
The z values, shape: (n_z,)
data_v: numpy.ndarray
The variable values, shape: (n_z,)
interp_pars: dict
Additional parameters for interpolation
:group: models.vertical_profiles
"""
[docs]
def __init__(
self,
data_source,
variable,
col_z=None,
col_var=None,
pd_read_pars={},
**interp_pars
):
"""
Constructor
Parameters
----------
data_source: str or numpy.ndarray or pandas.DataFrame
The profile data
variable: float
The value
col_z: str or int, optional
The column of z data
col_var: str or int, optional
The column of variable data
pd_read_pars: dict
Additional parameters for pandas.read_csv()
interp_pars: dict, optional
Additional parameters for interpolation
"""
super().__init__()
self.var = variable
self.interp_pars = interp_pars
if isinstance(data_source, np.ndarray):
col_z = col_z if col_z is not None else 0
col_var = col_var if col_var is not None else -1
self.data_z = data_source[col_z]
self.data_v = data_source[col_var]
else:
if isinstance(data_source, pd.DataFrame):
data = data_source
else:
data = pd.read_csv(data_source, **pd_read_pars)
col_var = col_var if col_var is not None else variable
self.data_v = data[col_var].to_numpy()
if col_z is None:
self.data_z = data.index.to_numpy()
else:
self.data_z = data[col_z].to_numpy()
if not np.all(np.diff(self.data_z) > 0):
inds = np.argsort(self.data_z)
self.data_z = self.data_z[inds]
self.data_v = self.data_v[inds]
[docs]
def calculate(self, data, heights):
"""
Run the profile calculation.
Parameters
----------
data: dict
The input data
heights: numpy.ndarray
The evaluation heights
Returns
-------
results: numpy.ndarray
The profile results, same
shape as heights
"""
return interp1d(self.data_z, self.data_v, **self.interp_pars)(heights)