Pandas Settingwithcopywarning: A value is trying to be set on a copy of a Slice from a Dataframe

Question

I want to copy an element of a dataframe and insert it into another dataframe.

In essence there is a dataframe with name x area and another that I need to load with the data of the area, from the comparison of the name in the first dataframe. The code is working, but gives this title Warning.

What am I doing:

for i in range(0,len(NOVOS_ABERTOS_VIVO)):
    for j in range(0,len(AREA_VERSUS_COLABORADOR)):
        if NOVOS_ABERTOS_VIVO.loc[i]['Originator'] == AREA_VERSUS_COLABORADOR.loc[j]['Originator']:
            TEMP=AREA_VERSUS_COLABORADOR.loc[j]['Área']
            NOVOS_ABERTOS_VIVO.loc[i]['Área']=str(TEMP)

I also made:

for i in range(0,len(NOVOS_ABERTOS_VIVO)):
    for j in range(0,len(AREA_VERSUS_COLABORADOR)):
        if NOVOS_ABERTOS_VIVO.loc[i]['Originator'] == AREA_VERSUS_COLABORADOR.loc[j]['Originator']:
            AREA_VERSUS_COLABORADOR.loc[j]['Área']=NOVOS_ABERTOS_VIVO.loc[i]['Área'] 

Answer 1

Hello, Guido. Welcome back!

That one Warning can be easily solved by modifying your code in this way:

for i in range(0,len(NOVOS_ABERTOS_VIVO)):
    for j in range(0,len(AREA_VERSUS_COLABORADOR)):
        if NOVOS_ABERTOS_VIVO.loc[i]['Originator'] == AREA_VERSUS_COLABORADOR.loc[j]['Originator']:
            AREA_VERSUS_COLABORADOR.loc[j,'Área']=NOVOS_ABERTOS_VIVO.loc[i,'Área']

Note that instead of writing df.loc[linha][coluna], was written df.loc[linha,coluna]. This is the recommended way to do this type of operation for performance.

In the first form (df.loc[linha][coluna]) what is being done is a linear and independent call for operations, or a chain call. First the pandas returns all columns of the dataframe of the desired row to only return the column.

On the other hand, the second form (df.loc[linha,coluna]) allows the pandas treat your search as a single entity, greatly improving speed.

References:

http://pandas.pydata.org/pandas-docs/stable/indexing.html#Indexing-view-versus-copy