How can I change hvplot.heatmap format to New Dataframe?

HoloViews
, ,
1

I am making heatmap from dataframe using hvplot.heatmap. And then I have a question.
I can use hvplot.heatmap and show new heatmap. But other person want to see heatmap in excel. so I want to transform from hvplot.heatmap to excel file or dataframe. How should I do change format from hvplot.heatmap to New Dataframe.
Could someone help me?

Screen shot 1 Heat Map

20220127_001

Screen shot 2 original dataframe (try_heat_df)

20220127_002

Screen shot 3 What I wanted to. change from heat map to New Dataframe (columns – x , index – y)

20220127_003
20220127_003773×296 35.3 KB

My simple code

import numpy as np
import pandas as pd
import panel as pn
import holoviews as hv
import hvplot.pandas  # noqa
hv.extension('bokeh')

try_heat_df = pd.DataFrame (data = [[0,0,np.NaN],[1,0,0],[2,0,np.NaN],\
                                    [0,1,0]     ,[1,1,2],     [2,1,1],\
                                    [0,2,np.NaN],[1,2,1],     [2,2,1]], \
                            columns = ['x','y','val'] )

plot_heat = try_heat_df.hvplot.heatmap(  
            x='x', 
            y='y', 
            C='val',
            title=str('try_heat'), 
            cmap=["#000080","#0000FF","#33CCCC", "#99CCFF", "#CCFFFF", "#FFFF99", "#FFCC00", "#FF9900", "#FF7878", "#DD0b1d", "#CC0b1d"], 
            xaxis='bottom',  width=300 , height=300 , 
            ).opts(
            bgcolor='black',
            xlim=(0,2), ylim=(2,0) , colorbar=True, clim=(0,2),clipping_colors={'NaN': 'black','-': 'black'},\
            tools=['hover'],                          
            colorbar_opts = {'background_fill_color':'black'})

print ('plot_heat',plot_heat.data.values)

# ----------------------------------------------------------------------
#   Main Function
# ----------------------------------------------------------------------
if __name__ == "__main__":
    app = plot_heat
    pn.serve(app)

elif __name__.startswith("bokeh"):
    app = plot_heat
    app.servable()
2

I think you want to pivot your dataframe, maybe with something like below?

df = try_heat_df.pivot_table(values='val', index='y', columns='x')

1 Like
3

Thank you for quick reply. This is what I wanted to do.

1 Like