CODEWITHSUNDEEP
python-3.xpandas
Hellyeah
Hellyeah

Reputation: 145

Moving data from rows to created columns using Pandas - Python

I want to use pandas to move the data around on a text file so that it is easy to parse data for the user. So far I am able to import several text files and append the data to a data frame along with adding headers. What I want to do is move the data to the correct column, but the issue is all the data is on the same column.

Here is my data: 

test2218
math-science-physics
00:00:00:00
00:00:30:00
03-21 04:00:00
28
test2228
math
00:00:00:00
00:00:30:00
03-21 04:00:00
26
test2317
reading-comprehension
00:00:00:00
00:00:30:00
03-21 20:02:00

This is what I want my output to look like:

Test ID     Test Info               Duration_A  Duration_B  Next Use        Participants    
test2218    math-science-physics    00:00:00:00 00:00:30:00 03-21 14:00:00  28
test2228    math                    00:00:00:00 00:00:30:00 03-21 14:00:00  26
test2317    reading-comprehension   00:00:00:00 00:00:30:00 04-11 13:30:00  2

I've looked everywhere and can't find a clear answer. Can someone assist?

Here is my code so far:

import os, glob, pandas as pd
d_frame = []
c_names = ['Test ID', 'Test Info', 'Duration_A', 'Duration_B', 'Next 
           Use', 'Participants']
files_list = glob.glob(os.path.join('C:\\test', '*.txt'))

for file in files_list:
    if os.stat(file).st_size != 0:
    df = pd.read_csv(file, delimiter='\t',header=None, names = c_names)

Any insight on this would be greatly appreciated. Thanks in advance!

Upvotes: 4

Views: 528

Answers (3)

Chris Adams
Chris Adams

Reputation: 18647

Assuming your data is a pandas.DataFrame object and those 6 pieces of information are always present in that specific order, you might try:

df = pd.DataFrame({0: ['test2218', 'math-science-physics', '00:00:00:00', '00:00:30:00', '03-21 04:00:00', '28', 'test2228', 'math', '00:00:00:00', '00:00:30:00', '03-21 04:00:00', '26', 'test2317', 'reading-comprehension', '00:00:00:00', '00:00:30:00', '03-21 20:02:00']})

columns = ['Test ID', 'Test Info', 'Duration_A', 'Duration_B', 'Next Use', 'Participants']

df_new = pd.DataFrame(df.groupby(df.index // len(columns))[0].apply(list).values.tolist(), columns=columns)
print(df_new)

    Test ID              Test Info   Duration_A   Duration_B        Next Use    Participants
0  test2218   math-science-physics  00:00:00:00  00:00:30:00  03-21 04:00:00             28 
1  test2228                   math  00:00:00:00  00:00:30:00  03-21 04:00:00             26 
2  test2317  reading-comprehension  00:00:00:00  00:00:30:00  03-21 20:02:00           None

Or alternatively

df_new = pd.DataFrame(df.values.reshape(-1, len(columns)), columns=columns)

Upvotes: 3

C. Braun
C. Braun

Reputation: 5201

Here's a simple way to do it with numpy.reshape:

import numpy as np
import pandas as pd

pd.DataFrame(np.reshape(df.values, (len(df) // 6, 6)),
             columns=['Test ID', 'Test Info', 'Duration_A', 'Duration_B', 'Next Use', 'Participants'])


    Test ID              Test Info   Duration_A   Duration_B        Next Use    Participants
0  test2218   math-science-physics  00:00:00:00  00:00:30:00  03-21 04:00:00             28 
1  test2228                   math  00:00:00:00  00:00:30:00  03-21 04:00:00             26 
2  test2317  reading-comprehension  00:00:00:00  00:00:30:00  03-21 20:02:00              2

Upvotes: 3

Matt W.
Matt W.

Reputation: 3722

import pandas as pd

x= pd.Series(['test2218',
'math-science-physics',
'00:00:00:00',
'00:00:30:00',
'03-21 04:00:00',
'28',
'test2228',
'math',
'00:00:00:00',
'00:00:30:00',
'03-21 04:00:00',
'26',
'test2317',
'reading-comprehension',
'00:00:00:00',
'00:00:30:00',
'03-21 20:02:00',
'55'])

Loop through to find the indices needed

indices = []
for i in range(6):
    indices.append(list(range(i, len(x), 6)))

create a columns list and empty dataframe, then loop through to subset the indexes, and assign to the dataframe. 

columns=['Test ID', 'Test Info', 'Duration_A', 'Duration_B', 'Next Use', 'Participants']
df = pd.DataFrame({})
for col, ixs in zip(columns, indices):
    df[col] = x[ixs].reset_index(drop=True)

Upvotes: 1

Related Questions