import numpy as np
from pyspedas.tplot_tools import get_data, store_data
[docs]
def tnormalize(variable, newname=None, return_data=False):
"""
Normalize all the vectors stored in a tplot variable
Parameters
----------
variable: str or np.ndarray
tplot variable (or numpy array) containing the vectors to be normalized
newname: str, optional
name of the output variable; If not set, output name will be variable + '_normalize'
Default: None
return_data: bool, optional
If true, return the normalized vectors instead of creating a tplot variable
Default: False
Returns
-------
str or ndarray of float
name of the tplot variable created or normalized vectors if return_data is set
Examples
---------
>>> import pyspedas
>>> pyspedas.store_data('a', data={'x':[1, 2, 3], 'y':[[1,2,3],[4,5,6],[7,8,9]]})
>>> pyspedas.tnormalize('a')
"""
metadata_in = {}
if isinstance(variable, str):
data_in = get_data(variable)
metadata_in = get_data(variable, metadata=True)
data = data_in[1]
times = data_in[0]
else:
data = np.atleast_2d(variable)
times = np.zeros(data.shape[0])
n = np.sqrt(np.nansum(data**2, axis=1))
# to do element-wise division, the magnitude needs to be repeated for each component
norm_reshaped = np.reshape(n, [len(times), 1])
norm_mag = np.repeat(norm_reshaped, len(data[0, :]), axis=1)
data_norm = data/norm_mag
if return_data:
return data_norm
else:
if newname is None:
newname = variable + '_normalized'
store_data(newname, data={'x': times, 'y': data_norm}, attr_dict=metadata_in)
return newname