shhudspeth/lahman-baltimoreparks

Practicing with Geocoder and Bokeh using data from Lahman's baseball datasets

Lahman-BaltimoreParks

Practicing with Geocoder and Bokeh using data from Lahman's baseball datasets

lahman-baltimoreparks's Issues

Loading the HTML after cloning repo

I tried loading the html file to view the maps and I ended up getting the Expired Key Map Error

https://developers.google.com/maps/documentation/javascript/error-messages#expired-key-map-error

(Just an FYI.)

You might have generated a new API token at some point after which might be part of it.

Copy Warnings

You can fix the copy warnings you're getting by specifically setting the values as copies rather than referencing the dataframe objects.

For example:

geo_parks['long'][4] = geo_parks['long'][5].copy()
geo_parks['long'][8] = geo_parks['long'][7].copy()
geo_parks['long'][6] = geocoder.google("324 E 25th St Baltimore MD").lng

Also, one of the beauties of using an Jupyter notebook is that you don't need to print your dataframes. Just calling it as the final command in the cell with show it and leave it in a nicer format.

Edit:
I was thinking of a different error when I told you the above improvement. It should actually be:

geo_parks.loc[4,'long'] = geo_parks['long'][5]
geo_parks.loc[8,'long'] = geo_parks['long'][7]
geo_parks.loc[6,'long'] = geocoder.google("324 E 25th St Baltimore MD").lng
geo_parks.head()

Filling in Missing Data

Am I correct in assuming that the reason you're using Google Maps API is to fill in the missing data?
If so I would recommend filling it this way, so that you don't need to do it by hand.

for i,b in enumerate(geo_parks.lat.isna()):
    if b == True:
        geo_parks.loc[i,'lat'] = geo_parks.loc[i, 'geo_code'].lat
        
for i,b in enumerate(geo_parks.long.isna()):
    if b == True:
        geo_parks.loc[i,'long'] = geo_parks.loc[i, 'geo_code'].lng

enumerate() zips a list with it's index into a tuple. I use i to indicate index and b to indicate boolean.

Fixing the Over Query Limit

If you look at your data in the geo_park dataframe you'll see that you sometimes get an over query limit. That's because your app is making too many calls too quickly. You're data set is small so even if we just add in a delay it won't take too long.

To do that, I think you have no choice but to remove the nice list comprehension geo_code = [geocoder.google(geo_parks['address'][x]) for x in range(len(geo_parks['address'])) ] and replace it with:

[Note: This piece of code has been edited]

import time

geo_codes = []
for address in geo_parks['address']:
    geo_code = geocoder.google(address)
    if str(geo_code) == '<[OVER_QUERY_LIMIT] Google - Geocode [empty]>':
        print(address, geo_code)
        time.sleep(5)
        geo_code = geocoder.google(address)
    geo_codes.append(geo_code)
geo_parks['geo_code']= geo_codes
geo_parks

I also modified the for loop to use the address itself rather than a numerical indexer. In the code you had it would look like this:
geo_code = [geocoder.google(x) for x in geo_parks['address'] ]