Dataframe row and column organization

Question

Hello,

I have the following situation:

import pandas as pd
import numpy as np


l=[]

l.append(('Mod1',0,70))
l.append(('Mod1',1,88))
l.append(('Mod1',2,97))
l.append(('Mod2',0,44))
l.append(('Mod2',1,93))
l.append(('Mod2',2,100))
l.append(('Mod3',0,99))
l.append(('Mod3',1,71))
l.append(('Mod3',2,33))

I would like the Dataframe to be as follows:

   Mod1    Mod2    Mod3
0   70      44      99
1   88      93      71
2   97     100      33

where 0,1,2 would be the index but the way I did Mod1,Mod2,Mod3 are in different columns

the way I did:

df = pd.DataFrame(data=l)
df=df.transpose()

gets that way:

Answer 1

The code below has been tested with the pandas-1.0, and using the function pivot_table [Pandas-Docs] allows you to convert the dataframe for the desired format:

df.columns=['col1', 'col2', 'col3'] #adicionados títulos às diferentes colunas
df = df.pivot_table(index=['col2'], columns='col1', values='col3').reset_index()
df.drop(df.columns[0], axis=1, inplace=True)
df = df.rename_axis(None, axis=1)
df

Out[11]: Mod1 Mod2 Mod3 0 70 44 99 1 88 93 71 2 97 100 33

Answer 2

Third Edit: I believe that the code can now process the data in the expected way, that is, with automation of the generation of the lists and with correct inclusion of each "Mod", regardless of the nomenclature of each string. As for the problem of irregularity of the amount of data per "Mod", I used the feature to include the data 0 in each empty field (check if this brings harm to the data processing in your case).

Room Edit: There was an error in the logic of this last posted version, I will resend the version that I believe is working properly. To better organize the answer, I will replace the previous versions with the code below:

import pandas as pd

dados = {}
l = []
lista_mods = []
lista_valores = []
maximo = 0

l.append(('Mod1',0,70))
l.append(('Mod1',1,88))
l.append(('Mod1',2,97))
l.append(('Mod1',3,44)) # linha incluída para testar o funcionamento com diferentes quantidades de índices
l.append(('Mod2',0,44))
l.append(('Mod2',1,93))
l.append(('Mod2',2,100))
l.append(('Mod3',0,99))
l.append(('Mod3',1,71))
l.append(('Mod3',2,33))

for pos, c in enumerate(l):
    if l[pos][0] not in lista_mods:
        lista_mods.append(l[pos][0])

for pos, c in enumerate(lista_mods):
    for n in range(0, len(l)):
        if l[n][0] == c:
            lista_valores.append([])
            lista_valores[pos].append(l[n][2])

for c in lista_valores:
    if len(c) > maximo:
        maximo = len(c)

for pos, c in enumerate(lista_valores):
    while True:
        if len(c) < maximo:
            lista_valores[pos].append(0)
        else:
            break    

dados = dict(zip(lista_mods, lista_valores))
df = pd.DataFrame(dados)
print(df)
df.to_html('temp.html')