Fill DataFrame in for loop with Python involving a function

Question

I want to retrieve geo data (long/lat) from Mapzen for addresses in Germany. Mapzen provides an API which asks for a key. Each request returns a Json.

This following code returns the long/lat and address name for one address:

import pandas as pd
import request

BASE_URL = 'https://search.mapzen.com/v1/search'
txt = 'Stübekamp 33, Hamburg, Germany'
resp = requests.get(BASE_URL, params = {'api_key': "YourKey", 'size': 1, 'text': txt})
data = resp.json()
Full = pd.DataFrame(columns=["Long", "Lat", "Street"])
LongLat = data["bbox"][0:2]
Street = data["features"][0]["properties"]["label"]
Full.loc[1] = pd.Series({"Long": LongLat[1], "Lat": LongLat[0], "Street": Street})

I tried to replace the txt argument to loop over it, but as long as I understand the request.get method cannot be looped over. Therefore, I followed this approach and defined a function which I use in a for loop.

What I want the for loop to do is to paste the string of one row in addresses in the txt argument in the function. This should be done n times, whereas n is the length of the addresses vector. The retrieved information (long/lat/address) should be added to a new row in the AllAddresses DataFrame. So in the end I have a DataFrame with three Columns ("Long", "Lat", "Street") and in this case 3 rows.

def Getall(Input):
    resp = requests.get('https://search.mapzen.com/v1/search', params = {'api_key': "YourKey", 'size': 1, 'text': Input})
    data = resp.json()
    LongLat = data["bbox"][0:2]
    Street = data["features"][0]["properties"]["label"]
    Full = pd.DataFrame(columns=["Long", "Lat", "Street"])
    Full.loc[1] = pd.Series({"Long": LongLat[1], "Lat": LongLat[0], "Street": Street})

    return Full


addresses = pd.DataFrame(["Stübekamp 33, Hamburg, Germany", "Mesterfeld 28, Hamburg, Germany","Beutnerring 2, Hamburg, Germany"])


AllAddresses = []
for index, row  in addresses.iterrows(): 
    Input = row("0")
    data = Getall(Input)
    AllAddresses.append = data

This code however, returns the error:

TypeError: 'Series' object is not callable

I read that iterrows is the way to go, but I am coming from R and feel a little lost here.

simon · Accepted Answer · 2016-12-31 17:10:30Z

1

Addresses is a pandas dataframe for no apparent reason. Then you iterate over it which is generally a bad idea if you needed a pandas dataframe which you don't. Then you take "row" which is a series and call it like a function row("0"). As it is not a function you get an error. Just make addresses a list to solve your first problem.

Then of course you will find you have a problem with Full which also does not need to be a dataframe; you cannot add a row like that; and you are returning a dataframe for each row which is likely not what you want either.

answered Dec 31, 2016 at 17:10

simon

2,9112 gold badges20 silver badges30 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

CFM Over a year ago

I see my problem now. I was too focused on pandas dataframes. Thanks!

Collectives™ on Stack Overflow

Fill DataFrame in for loop with Python involving a function

1 Answer 1

1 Comment

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Linked

Related