Result export in excel

oemof
#1

Hello all,

I used the “processing.create_dataframe(om)” to generate a panda data frame and exported
it with the pd.ExcelWriter function.

The result looks like:

image
image.png1634×170 9.51 KB

Is it possible to generate a panda dataframe which contains labels of oemof elements instead of pyomo tuples by using your
“processing.create_dataframe(om)”, because I would like to analyse the results in excel format. I already tried my best but I wasn’t able to change the processing code so I can produce the panda frame I want to have.

Thanks a lot!

Greetz
Steffen

#2

You could try this to put the scalars/sequences of all nodes into single sheets of an Excel-file:

import pandas as pd
from oemof.outputlib import processing, views

# om is the optimization model which has been solved

result_data = processing.results(om)
result_data = views.convert_keys_to_strings(result_data)

xls_file = 'results.xls'

writer = pd.ExcelWriter(xls_file, engine='xlsxwriter')
# add regular optimization results
nodes = sorted(set([item for tup in result_data.keys() for item in tup]))
for n in nodes:
    node_data = views.node(result_data, n)
    n = n[:30]  # trim string length to allowed chars for a worksheet
    if 'scalars' in node_data:
        node_data['scalars'].to_excel(writer, sheet_name=n+'_scalars')
    if 'sequences' in node_data:
        for key in keys:
            if key in data['sequences']:
                node_data['sequences'][key] = data['sequences'][key]
        node_data['sequences'].to_excel(writer, sheet_name=n+'_sequences')
writer.save()

Cheers
Cord

#3

Thank you Cord! That helps.

Greetz
Steffen

Finally i just used this and got all results :slight_smile:

# om is the optimization model which has been solved

result_data = processing.results(om)
result_data = views.convert_keys_to_strings(result_data)

xls_file = 'TEST.xlsx'

writer = pd.ExcelWriter(xls_file, engine='xlsxwriter')

# add regular optimization results
nodes = sorted(set([item for tup in result_data.keys() for item in tup]))
for n in nodes:
    node_data = views.node(result_data, n)
    n = n[:10]  # trim string length to allowed chars for a worksheet
    if 'sequences' in node_data:
        node_data['sequences'].to_excel(writer, sheet_name=n+'sequences')