[PYTHON] I want to visualize the transfer status of the 2020 J League, what should I do?

Motivation

• Many players transfer to soccer every season.
• 1650 players (as of January 31, 2020) are registered on the official website for this season.
• In addition, transfer information is posted on the site J'S GOAL'S site that publishes transfer information.
• I've always wondered if it would be possible to visualize the flow status of this player.
• Even if I searched Google teacher for "inflow / outflow graph", I couldn't find anything that could be used.
• If you call it "inflow / outflow visualization", you will discover a "sankey diagram"! This may be usable!

About the sankey diagram

The Sankey diagram is a diagram that represents the flow rate between processes. The thickness of the arrow indicates the amount of flow. In particular, it is used to represent the displacement of energy, supplies, expenses, etc.

[Source: Free encyclopedia "Wikipedia"](https://ja.wikipedia.org/wiki/%E3%82%B5%E3%83%B3%E3%82%AD%E3%83%BC_ % E3% 83% 80% E3% 82% A4% E3% 82% A2% E3% 82% B0% E3% 83% A9% E3% 83% A0)

• Sankey diagram in python
• Let's try analysis! Chapter 7: Data Visualization Using Google Charts Sankey Diagram
• github：ipysankeywidget
• demo：Exporting Images.ipynb
• plotly: Sankey Diagram in Python
• matplotlib: sankey

I don't understand the original matplotlib at all, and plotly is very beautiful but I don't understand at all. Try with the simplest demo: Exporting Images.ipynb!

Copying sutras

sankey.py

from ipysankeywidget import SankeyWidget
from ipywidgets import Layout

• Install the library in advance with pip install ipysankey widget.
• Also, when using with jupyter lab,

$ jupyter labextension install jupyter-sankey-widget @jupyter-widgets/jupyterlab-manager

And register the extension.

sankey.py

links = [
    {'source': 'start', 'target': 'A', 'value': 2},
    {'source': 'A', 'target': 'B', 'value': 2},
    {'source': 'C', 'target': 'A', 'value': 2},
    {'source': 'A', 'target': 'C', 'value': 2},
]

Since this is the data part, replace it with the transfer dummy data.

sankey.py

links = [
    {'source': 'Hokkaido_in', 'target': 'overseas_out', 'value': 1},
    {'source': 'Hokkaido_in', 'target': 'J2_out', 'value': 1},
    {'source': 'Hokkaido_in', 'target': 'University_out', 'value': 3},
    {'source': 'Shonan_in', 'target': 'Hokkaido_out', 'value': 1},
    {'source': 'J3_in', 'target': 'Hokkaido_out', 'value': 1},
]

sankey.py

layout = Layout(width="600", height="400")
sankey = SankeyWidget(links=links, layout=layout)
sankey.auto_save_png('./img/test.png')

The size of the output image is 600 pixels wide and 400 pixels high and is saved in PNG format.

Needless to say, I was able to express the flow of 5 players who joined and 2 players who transferred. (Look at it, but it seems that you can express the flow for the time being)

Summary

• "Sankey diagram in python" @deaikei and demo: Exporting Images.ipynb /Exporting%20Images.ipynb) I haven't studied enough yet.
• The J'S GOAL'S site clearly states that the copyright is prohibited from copying documents, images, videos, etc. ** Therefore, we will not share the scraping method or visualization results.
• Knowledge was obtained regarding the visualization of Sankey diagrams.