Create Dataframe from unknown number of Series pandas

Question

I have a list of (user defined) objects known as "Instrument Measurements". These, have a title, accessible by Instrument_Measurement.name and a series of data accessible by Instrument_Measurement.data (both user defined methods).

These "Instrument Measurement" objects are items in a list called list_instr_objects. The number of items ("Instrument Measurements") changes/cannot be hard-coded.

I need to create a Dataframe with column titles Instrument_Measurement.name for each "Instrument Measurement" and data for that column of Instrument_Measurement.data for that respective "Instrument Measurement".

I'm trying to do this by creating a dictionary of these objects and then converting that to a Dataframe:

from collections import defaultdict
testdict = defaultdict(list)

for i in range(len(list_instr_objects)):
    testdict[list_instr_objects[i].name].append(list_instr_objects[i].data)

This doesn't work though, only the first data entry for each instrument is kept and the entries come out in a seemingly random order. I think this is because I'm passing a Series to the dictionary where I should be passing a list, but I can't think how else to do it/how to fix this.

Any thoughts?

Many thanks in advance.

The columns would definitely be in an arbitrary order (as dictionaries do not have orders): is that a problem? The other issue is that you don't want to append it to a list, you want to just set it as testdict[list_instr_objects[i].name] = list_instr_objects[i].data (and it doesn't have to be a defaultdict, just a dictionary) — David Robinson
– David Robinson, Commented Mar 25, 2014 at 19:21
Unfortunately after using just a dictionary and not appending to list, I'm still getting the same issue of having only the first element. Unfortunately I do require the order to be preserved so I'll look into a non-dictionary based method. Many thanks. — vbastrangledpython
– vbastrangledpython, Commented Mar 25, 2014 at 19:31

kimal · Accepted Answer · 2014-03-25 19:40:36Z

1

I think you can use OrderedDict to keep the order and use below code to generate the dataframe.

import pandas as pd
from collections import OrderedDict

testdict = OrderedDict()

for i in range(len(list_instr_objects)):
    testdict[list_instr_objects[i].name] = (list_instr_objects[i].data)

combined_data = pd.DataFrame(testdict)

answered Mar 25, 2014 at 19:40

kimal

6926 silver badges5 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Create Dataframe from unknown number of Series pandas

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related