Differences between revisions 47 and 48

Log of Kamioka Activity during 2020/08/17 ~ 27

Participants

If you are unfamiliar with the simulation operations, there is a possibility that errors may occur in the script description and so on.
By creating a GUI, such errors can be prevented and the reliability of the simulation results can be increased.

1. Installation of eel

note that:

We installed several tools to run simulations using Finesse.
We initially considered using eel, but found it difficult to create gui using eel and finesse at the same time, so we decided to use pysimplegui instead of eel.
This eliminates the need to install eel (and chromium).
PySimpleGUI is a module for python that creates gui.
eel is a Python module used to create a GUI.
I didn't found any instructions on how to install eel without using pip.
Therefore, we use pip to install eel in conda's virtual environment.

Activate the conda virtual environment.

% conda activate XXXXX

And type the following command

% pip install eel

To use eel, you need to install a specific browser like "chromium".

You can install chromium by typing the following command.

% sudo apt install -y chromium

2. Learning HTML and javascript

3. Learning Python

Python is used to run simulations using "finesse2
Creation of GUI In "finesse2", we use character strings which describe the simulation settings.
Create a selection box (GUI) and load the settings you entered into the box by "finesse2" and execute it.

4. Creating a simple GUI with python and HTML.

An error occurs when sending the input value to Finesse, so we are looking for an improvement.
We gave up to use eel.

how we installed package PySimpleGUI?
- We use conda. https://anaconda.org/conda-forge/pysimplegui
  - ```
  conda install -c conda-forge pysimplegui 
```
what is the name of virtual environment on control workstation?
- MIFsim37
- Now there are many kinds of environments on control workstation. Before leaving, we should delete unnecessary virtual environment.
- Yuzurihara likes the name of 'mifsim37'
you can choose which DoF to run.
You can now choose where to put the PD.
You can now choose the laser power.
overplot ga mondai.We can't split plots.
in jupyter-notebook, two plots are displayed separately. .py dato gamenn kuroi
- (Yuzurihara) one of the solution is to use 'matplotlib' and summarize all the result on one figure. If so, we don't need second plot.
- Koyama-san solved this by using matplotlib.
pd1 ga hituyoukadouka ("Transfer function" uses "pd2". need "pd1" for "diff"?)
- Based on lecture by Kokeyama-san, Koyama-san implemented.
Yuzurihara implemented the feature to output the plot data and the kat file.

make the plot display large
- Figure size changed large. But, the label and number on axis are very small.
TODO : measure the computational time
feature to be requested
- select option overplot/not overplot
- add new configuration f interferometer (MI/PRFPMI/DRFPMI)
- prepare manual
- read external data and overplot
- added all nodes to GUI (on GUI, we use REFL/AS. But, in the source we use just n1 / n2 /....)
  - solved
Request from Yuzurihara
- Remove exit button
  - solved
- Make the outputted figure large
  - solved(probably)
- Add the xlabel and ylabel on figure automatically
  - next
- Use the variable name which we can easily understand
- Add options to select the plotting scale(log/linear)

resize the figure of the interferometer

convert Fabry_Perot_MI_eom_drawing.png -resize 800x600 hoge.png

KAGRA/Subgroups/MIF/Simulation/KamikokaActivity20200817 (last edited 2020-10-18 17:32:36 by naoki.koyama)

-  ⇤ ← Revision 47 as of 2020-08-21 17:56:07 → 
  Size: 4954
  Editor: naoki.koyama
  Comment:
+   ← Revision 48 as of 2020-08-21 17:56:31 → ⇥
  Size: 4957
  Editor: naoki.koyama
  Comment:
-Deletions are marked like this.
+Additions are marked like this.
 Line 104:
-  * Add the xlabel and ylabel on figure automatically -> next
+  * Add the xlabel and ylabel on figure automatically
    *next