KeyError when using boolean filter on pandas data frame

问题

Trying to combine two data frames when a datetime object from one dataframe is within a datetime object range in the other.

Keep getting: KeyError: 'cannot use a single bool to index into setitem' on this line of code in the second chunk I posted.

gametaxidf.loc[arrivemask, 'relevant'] = 1

I'm assuming it would happen on the following line with a similar command as well.

This is the part giving me trouble:

with open('/Users/benjaminprice/Desktop/TaxiCombined/Data/combinedtaxifiltered.csv', 'w') as csvfile: 
    fieldnames1 = ['index','pickup_datetime', 'dropoff_datetime', 'pickup_long', 'pickup_lat','dropoff_long','dropoff_lat','passenger_count','trip_distance','fare_amount','tip_amount','total_amount','stadium_code'] 
    writer = csv.DictWriter(csvfile, fieldnames=fieldnames1) 
    writer.writeheader()

for index, row in baseballdf.iterrows(): 
    gametimestart = row['Start.Time'] 
    gametimeend = row['End.Time'] 
    arrivemin = gametimestart - datetime.timedelta(minutes=120) 
    arrivemax = gametimeend - datetime.timedelta(minutes = 30) 
    departmin = gametimeend - datetime.timedelta(minutes = 60) 
    departmax = gametimeend + datetime.timedelta(minutes = 90)

    gametaxidf = combineddf[combineddf.DATE==row.DATE]
    gametaxidf['relevant']=0

    for index, row in gametaxidf.iterrows():
        arrivemask = (arrivemin < row['dropoff_datetime']) and (row['dropoff_datetime'] < arrivemax)
        departmask = (departmin < row['pickup_datetime']) and (row['pickup_datetime'] < departmax) 
        gametaxidf.loc[arrivemask, 'relevant'] = 1
        gametaxidf.loc[departmask, 'relevant'] = 1

        with open('/Users/benjaminprice/Desktop/TaxiCombined/Data/combinedtaxifiltered.csv','a') as combinedtaxi:
            gametaxidf.to_csv(combinedtaxi,header=None)
    print(str(index) + "done")

Gametaxidf.head(5):

   index     pickup_datetime    dropoff_datetime  pickup_long  pickup_lat  \
0    195 2014-04-01 00:08:13 2014-04-01 00:15:32   -73.922218   40.827557   
1    344 2014-04-01 00:16:30 2014-04-01 00:20:38   -73.846046   40.754566   
2    558 2014-04-01 00:28:59 2014-04-01 00:36:36   -73.921692   40.831394   
3    744 2014-04-01 00:42:00 2014-04-01 00:49:46   -73.938080   40.804646   
4    776 2014-04-01 00:43:54 2014-04-01 00:53:22   -73.952652   40.810577   

   dropoff_long  dropoff_lat  passenger_count  trip_distance  fare_amount  \
0    -73.900620    40.856174                1           2.30          9.0   
1    -73.890259    40.753246                1           0.56          4.5   
2    -73.942719    40.823257                1           1.53          7.0   
3    -73.928490    40.830433                1           2.96         11.0   
4    -73.924332    40.827320                1           2.28         10.5   

   tip_amount  total_amount  stadium_code       DATE  relevant  
0           0          10.0           1.1 2014-04-01         0  
1           0           5.5           2.1 2014-04-01         0  
2           0           8.0           1.1 2014-04-01         0  
3           0          12.0           1.0 2014-04-01         0  
4           0          11.5           1.0 2014-04-01         0

Also getting this warning: A value is trying to be set on a copy of a slice from a DataFrame.

Try using .loc[row_indexer,col_indexer] = value instead

But it's letting me continue through that... any help would be great.

回答1:

Here

gametaxidf.loc[arrivemask, 'relevant'] = 1

you're trying to set dataframe values by .loc operator. Pandas docs for selecting rows says:

.loc is primarily label based, but may also be used with a boolean array. .loc will raise KeyError when the items are not found. Allowed inputs are:

A single label, e.g. 5 or 'a', (note that 5 is interpreted as a label of the index. This use is not an integer position along the index)

A list or array of labels ['a', 'b', 'c']

A slice object with labels 'a':'f', (note that contrary to usual python slices, both the start and the stop are included!)

A boolean array

You're trying to use the last type of input, but this

arrivemask = (arrivemin < row['dropoff_datetime']) and 
    (row['dropoff_datetime'] < arrivemax)

is scalar boolean, not array.

You need not to iterate through dataframe. Pandas does it for you. Just use:

gametaxidf.loc[
   (arrivemin < gametaxidf['dropoff_datetime'])
   &
   (gametaxidf['dropoff_datetime'] < arrivemax)
   , 'relevant'] = 1

来源：https://stackoverflow.com/questions/33817842/keyerror-when-using-boolean-filter-on-pandas-data-frame

标签

python

pandas

boolean

dataframe

keyerror