

In [16]:

    
%matplotlib notebook
import numpy as np
import matplotlib.pyplot as plt
import os
import pandas as pd
from pyxrf.model.command_tools import fit_pixel_data_and_save
from pyxrf.api import *

Batch mode to fit data from multiple runs

Users need to have .json file ready in order to do batch mode fitting.



In [24]:

    
# Define working directory and json file
working_dir = '/data/users/2016Q3/Gill_2016Q3/'



In [22]:

    
# Define a list of h5 files which should stay in the working directory.
datalist = np.arange(16715, 16808)
filelist = ['scan2D_'+str(n)+'.h5' for n in datalist]



In [8]:

    
# Parameter file to fit all the data.
param_file = 'parameter_data.json'



In [ ]:

    
# Pixel fitting for all the files. If ic_name is given, 
# data will be also normalized based on ion chamber value. 
for fname in filelist:
    fit_pixel_data_and_save(working_dir, fname, 
                            param_file_name=param_file, 
                            save_txt=True, ic_name='sclr1_ch4')

Combine data into 3D array for reconstruction



In [32]:

    
element_list = ['Cu_K','Fe_K']
d3 = combine_data_to_recon(element_list, datalist, working_dir)



In [33]:

    
d3.keys()









    Out[33]:





['Fe_K', 'Cu_K']



In [35]:

    
# create movie for Fe
create_movie(d3['Fe_K'], 'data3d_Fe.mp4')



In [36]:

    
# create movie for Cu
create_movie(d3['Cu_K'], 'data3d_Cu.mp4')



In [ ]:



In [ ]:



In [ ]: