CODEWITHSUNDEEP
pythonpandaslistsorting
Juank
Juank

Reputation: 55

Apply sorting to a list of dataframes Python

I've struggling trying to insert a sorting in this code:

all_files = glob.glob(path + "/*.CSV") # To get all csv files disorganized
all_csv = [pd.read_csv(f, sep=',') for f in all_files] # List of dataframes

# I want to sort it by the values of the first column of each dataframe in the all_csv list.

for f in all_csv:
    goal = pd.DataFrame.sort_values(by=(f.iloc[:,0])) #Maybe something like this??

So, Anyone has an idea how can I do this? I've looking on other post but does not apply to a undefined column name (a.k.a. f.iloc[:,0] ) or a list of dataframes (I also thought of using dictionaries but I'd like to see if is posible to use with lists).

Thank you :)

May be useful this ideas: link, link

Upvotes: 0

Views: 50

Answers (2)

user7864386
user7864386

Reputation:

You can index df.columns for individual dataframes:

goal = df.sort_values(by=df.columns[0])

For the entire list of dataframes, you can use list comprehension:

all_csv = [df.sort_values(by=df.columns[0]) for df in all_csv]

Suppose you had a dataframe that looked like:

   a  b
0  2  1
1  3  2
2  1  3

Then when you run:

df = df.sort_values(by=df.columns[0])

df becomes:

   a  b
2  1  3
0  2  1
1  3  2

Upvotes: 0

bsauce
bsauce

Reputation: 644

This uses enke's code to sort each dataframe by the first column, but returns all dataframes in a list as you requested:

all_csv = [df.sort_values(by=df.columns[0]) for df in all_csv]

Upvotes: 1

Related Questions