CODEWITHSUNDEEP
pythonpython-3.xpandas
kikee1222
kikee1222

Reputation: 2016

For each loop on Pandas, each category

I have a question.

I have a table like this

TAC | Latitude | Longitude
1 | 50.4 | -1.5

In Pandas, I wanted to say:

For each TAC, give me a zipped list of latitude and longitude (each TAC can have many rows).

I've tried things like the below, but I am doing something wrong! Can you help?

df1['coordinates'] = list(zip(df1.Lat, df1.Long))
new_df = df1.iloc[ : , : ].groupby('TAC').agg(df1['coordinates'])

For reference, DF1 is created as below

df = pd.read_csv('tacs.csv')
df1 = df[['magnet.tac','magnet.latitude', 'magnet.longitude']]
df1.columns = ['TAC','Lat','Long']

Upvotes: 1

Views: 47

Answers (1)

jezrael
jezrael

Reputation: 862731

First add usecols parameter for avoid SettingWithCopyWarning and then use GroupBy.apply with lambda function:

df = pd.read_csv('tacs.csv', usecols=['magnet.tac','magnet.latitude', 'magnet.longitude'])
df1.columns = ['TAC','Lat','Long']

#sample data
print (df1)
   TAC   Lat  Long
0    1  50.4  -1.5
1    1  50.1  -1.4
2    2  50.2  -1.8
3    2  50.9  -1.3

new_df = df1.groupby('TAC').apply(lambda x: list(zip(x.Lat, x.Long))).reset_index(name='coord')
print (new_df)
   TAC                         coord
0    1  [(50.4, -1.5), (50.1, -1.4)]
1    2  [(50.2, -1.8), (50.9, -1.3)]

Your solution should be changed:

df = pd.read_csv('tacs.csv')
df1 = df[['magnet.tac','magnet.latitude', 'magnet.longitude']].copy()
df1.columns = ['TAC','Lat','Long']

df1['coordinates'] = list(zip(df1.Lat, df1.Long))
new_df = df1.groupby('TAC')['coordinates'].agg(list).reset_index()

Upvotes: 2

Related Questions