CODEWITHSUNDEEP
pythoncsv
Nikhil Gupta
Nikhil Gupta

Reputation: 561

Switching row header and column header in python

I have a .csv file that has data something like this:

#file...out/houses.csv
#data...sun may 1 11:20:43 2011
#user...abs12
#host...(null)
#group...class=house
#property..change_per_hour
#limit...0
#interval..10000000
#timestamp,house_0,house_1,house_2,house_3,.....,house_1000
2010-07-01 00:00:00 EDT,1.2,1.3,1.4,1.5,........,9.72
2010-07-01 01:00:00 EDT,2.2,2.3,2.4,2.5,........,19.72
2010-07-01 02:00:00 EDT,3.2,3.3,3.4,3.5,........,29.72
2010-07-01 05:00:00 EDT,5.2,5.3,5.4,5.5,........,59.72
2010-07-01 06:00:00 EDT,6.2,,6.4,,..............,
...

I want to convert this and save to a new .csv and the data should look like:

#file...out/houses.csv
#data...sun may 1 11:20:43 2011
#user...abs12
#host...(null)
#group...class=house
#property..change_per_hour
#limit...0
#interval..10000000
#EntityName,2010-07-01 00:00:00 EDT,2010-07-01 01:00:00 EDT,2010-07-01 02:00:00 EDT,2010-07-01 05:00:00 EDT,2010-07-01 06:00:00 EDT
house_0,1.2,2.2,3.2,5.2,6.2,...
house_1,1.3,2.3,3.3,5.3,,...
house_2,1.4,2.4,3.4,5.4,6.4,...
house_3,1.5,2.5,3.5,5.5,,...
...
house_1000,9.72,19.72,29.72,59.72,

I tried to use pandas: convert to a dictionary that looks like dtDict={'house_0':{'datetimestamp_1':'value_1','datetimestamp_2':'value_2'...}...}but I am not able to convert to a dictionary and use panda's DataFrame such as pandas.DataFrame(dtDict) to do that conversion. I dont have to use pandas (can you anything in python) but thought pandas is good for csv manipulation. any help?

Upvotes: 0

Views: 117

Answers (1)

Nelewout
Nelewout

Reputation: 6554

Assuming it is in a pandas dataframe already, this works:

df = pd.DataFrame(
    data=[[1, 3], [2, 5]],
    index=[0, 1],
    columns=['a', 'b']
)

Output:

>>>print(df)
      a  b
   0  1  3
   1  2  5

Then, transpose the dataframe:

>>>print(df.transpose())
      0  1
   a  1  2
   b  3  5

Upvotes: 1

Related Questions