Pandas: TypeError: sort_values() missing 1 required positional argument: 'by'

Question

I am facing TypeError: sort_values() missing 1 required positional argument: 'by'

for i in range(0,len(data_sims.index)):
for j in range(1,len(data_sims.columns)):
    user = data_sims.index[i]
    serial = data_sims.columns[j]

    if dataUser.ix[i][j] == 1:
        data_sims.ix[i][j] = 0
    else:
        serial_top_names = data_neighbours.ix[serial][1:10]
        serial_top_sims = dataSim.ix[serial].sort_values(ascending=False)[1:10]
        user_preferences = dataUser.ix[user,serial_top_names]

        data_sims.ix[i][j] = getScore(user_preferences,serial_top_sims)

Detailed Error:

> ---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
 in ()
      9         else:
     10             serial_top_names = data_neighbours.ix[serial][1:10]
---> 11             serial_top_sims = dataSim.ix[serial].sort_values(ascending=False)[1:10]
     12             user_preferences = dataUser.ix[user,serial_top_names]
     13 

TypeError: sort_values() missing 1 required positional argument: 'by'

I tried using following but didn't work

pd.DataFrame.sort_values(dataSim,ascending=[0])[1:10]

Is there anything that I'm missing?

Answer 1

The documentation for Pandas mentions:

DataFrame.sort_values(by,
                      axis=0,
                      ascending=True,
                      inplace=False,
                      kind='quicksort',
                      na_position='last')

by : str or list of str Name or list of names which refer to the axis items.

In my example:

df.groupby('product').agg({'weight':'sum'}).sort_values(by='weight')

Adding by='weight' solved the TypeError: sort_values() missing 1 required positional argument: 'by'

Answer 2

The dataframe may have multiple columns, so specify the column you want to sort by.

df = df.sort_values(by='sentence_length').reset_index(drop=True)