commenting on what to do in cases of using the ipynb either via north or locally

docker/examples/optcs/VisualizeOPTCSData.01.ipynb

 %% Cell type:markdown id: tags:
 
 # Exemplar visualization of results from optical spectroscopy (here ellipsometry)
 
 %% Cell type:code id: tags:
 
 ``` python
 import numpy as np
 import matplotlib.pyplot as plt
 import os, sys
 import json
 import shutil
 ```
 
 %% Cell type:markdown id: tags:
 
 ### Get a dataset
 
 %% Cell type:markdown id: tags:
 
 Spectrum of Si wafer with 2nm SiO2.
 
 %% Cell type:code id: tags:
 
 ``` python
+#comment the next line out in case you have the 190801_Si_SiO2_2nm_test_mat.txt file already visible in your nomadOASIS
+#leave the next line untouched if you are testing this notebook locally and have 190801_Si_SiO2_2nm_test.zip there
 shutil.unpack_archive('190801_Si_SiO2_2nm_test.zip')
 ! ls
 ```
 
 %% Output
 
     190801_Si_SiO2_2nm_test.dat		190801_Si_SiO2_2nm_test.zip
     190801_Si_SiO2_2nm_test_mat_fehler.txt	README.txt
     190801_Si_SiO2_2nm_test_mat.txt		VisualizeOPTCSData.01.ipynb
     190801_Si_SiO2_2nm_test.SE
 
 %% Cell type:markdown id: tags:
 
 ### Parsing from file formats
 
 %% Cell type:markdown id: tags:
 
 Chris Sturm explained that the dataset was collected with a JaWoollam instrument via CompleteEASE software https://www.jawoollam.com/ellipsometry-software/completeease
 To the best of his knowledge there is not yet an open-source parser for translating the proprietary *.SE file format that CompleteEASE creates into an open format.
 So Chris used another proprietary software, WVase https://www.jawoollam.com/ellipsometry-software/wvase, to transcode the *.SE into *.dat, which is a set of ASCII tables with metadata.
 From this *.dat he uses a custom parser to extract results into a format with which it is easier for him to work with to discuss the experiment. This format is an ASCII table where the individual spectra are stack as blocks of table glued together. This format he refers to as *_mat.txt and *_mat_fehler.txt specifically where the second format gives the uncertainties. This is what we work with in this example.
 
 %% Cell type:code id: tags:
 
 ``` python
 MYDATASET = '190801_Si_SiO2_2nm_test'
 ```
 
 %% Cell type:code id: tags:
 
 ``` python
 import pandas as pd
 data = pd.read_csv(MYDATASET+'_mat.txt', delimiter='\s+')
 ```
 
 %% Cell type:markdown id: tags:
 
 ### Data wrangling and cleaning
 
 %% Cell type:markdown id: tags:
 
 Two relevant observations wrt to formatting:
 * The column names in the header section get parsed incorrectly because the strings themselves contain space delimiters.
 * Some columns do not contain relevant measured data but dummy data instead these are according to Chris:
 * Drehwinkel/ x Position
 * y-Position
 * lf. Proben Nummber
 * SE Type
 * and all mm<i,j> columns
 
 %% Cell type:markdown id: tags:
 
 The original header line was:
 Energie	Einfallswinkel	Drehwinkel/ x Position y-Position	lf. Proben Nummer	SE-Type	Psi	Delta	mm11	mm12	mm13	mm14	mm21	mm22	mm23	mm24	mm31	mm32	mm33	mm34	mm41	mm42	mm43	mm44
 All we are interested for now is the
 
 %% Cell type:code id: tags:
 
 ``` python
 #data['mm34'] = data['mm34'].replace(0.0, np.nan)
 data = data.drop('Drehwinkel/', 1)
 data = data.drop('x', 1)
 data = data.drop('Position', 1)
 data = data.drop('y-Position', 1)
 for i in [1,2,3,4]:
     for j in [1,2,3,4]:
         columnname = 'mm'+str(i)+str(j)
         data = data.drop(columnname, 1)
 data = data.drop('Nummer', 1)
 data = data.drop('SE-Type', 1)
 data = data.drop('Psi', 1)
 data = data.drop('Delta', 1)
 #finally, reflecting the above-mentioned comments we need to relabel column lf. with Psi and Proben with Delta
 data = data.rename(columns={"lf.": "Psi (degree)", "Proben": "Delta (degree)"})
 ```
 
 %% Cell type:code id: tags:
 
 ``` python
 data
 ```
 
 %% Output
 
           Energie  Einfallswinkel  Psi (degree)  Delta (degree)
     0      1930.0            50.0     40.010899      140.596375
     1      1940.0            50.0     40.070530      140.679840
     2      1950.0            50.0     40.052708      140.847351
     3      1960.0            50.0     40.038048      141.239487
     4      1970.0            50.0     40.086514      141.575409
     ...       ...             ...           ...             ...
     4347  16900.0            65.0     14.899776      178.085114
     4348  16925.0            65.0     14.861681      178.171539
     4349  16950.0            65.0     14.917637      178.184357
     4350  16975.0            65.0     14.960825      178.107681
     4351  17000.0            65.0     14.887267      177.860275
     
     [4352 rows x 4 columns]
 
 %% Cell type:markdown id: tags:
 
 It remains to split the tables again into the different spectra for different "Einfallswinkel", i.e. angle of incidence.
 
 %% Cell type:code id: tags:
 
 ``` python
 incidence_angles = np.unique(data['Einfallswinkel'].to_numpy())
 print(incidence_angles)
 ```
 
 %% Output
 
     [50. 55. 60. 65.]
 
 %% Cell type:markdown id: tags:
 
 ### Now, visualizing the "spectra".
 
 %% Cell type:code id: tags:
 
 ``` python
 fig, ax1 = plt.subplots( figsize=[20, 10] )
 for i in incidence_angles:
     #i = 50.
     energy = data[np.abs(data['Einfallswinkel'] - i) < (np.finfo(np.float32).eps*10)]['Energie']
     psi = data[np.abs(data['Einfallswinkel'] - i) < (np.finfo(np.float32).eps*10)]['Psi (degree)']
     #delta = data[np.abs(data['Einfallswinkel'] - i) < (np.finfo(np.float32).eps*10)]['Delta (degree)']
     plt.scatter( energy, psi )
     #plt.scatter( energy, delta )
 plt.legend(incidence_angles)
 plt.xlabel(r'Energy (Angstrom)') #Chris explained they call the respective equivalent wavelength an energy in their domain
 plt.ylabel(r'Psi (degree)')
 #plt.ylabel(r'Delta (degree)')
 ```
 
 %% Output
 
     Text(0, 0.5, 'Psi (degree)')
 
 
 %% Cell type:markdown id: tags:
 
 ### Next steps
 
 %% Cell type:markdown id: tags:
 
 * Define and use a NeXus application definition for optical spectroscopy
 * Develop parser for proprietary formats
 * Reorganize/rewrite plotting functions
 * Use widgets (e.g. bokeh or ipywidgets) to make figure interactive
 * Use widgets to make spectra selectable
 * Make spectrum ranges clickable/selectable to define
 * Interface with auto-indexing tool
 
 %% Cell type:code id: tags:
 
 ``` python
 ```