I have a dataset of this type:

i want to convert this dataframe into an ascending list. I want to check the whole data into ascending order and then create a list using the ascending order numbers in the dataframe. and ascending order should be of numbers before ':'

desired output:

CodePudding user response：

You can unstack (or stack) to flatten to Series, then extract the number, convert to integer and keep the unique values in order:

For a python list you can try:

sorted(df.unstack()
         .str.extract('(\d ):', expand=False)
         .astype(int)
         .unique().tolist())

output: [0, 57, 166]

As Series:

out = (df.unstack()
         .str.extract('(\d ):', expand=False)
         .astype(int)
         .drop_duplicates()
         .sort_values().reset_index(drop=True)
      )

output:

0      0
1     57
2    166
dtype: int64