CODEWITHSUNDEEP
pythonpandasdataframedask
KEERTHANA 15CS26
KEERTHANA 15CS26

Reputation: 11

AttributeError: 'DataFrame' object has no attribute 'compute'

I have tried this code but this shows attribute error.

from dask.base import compute
import dask.dataframe as dd
import pandas as pd
def dfWithTripTimes(df):
    startTime=datetime.now()
    duration=df[["tpep_pickup_datetime","tpep_dropoff_datetime"]].compute()
    pickup_time=[timeToUnix(pkup) for pkuo in duration["tpep_pickup_datetime"].values]
    dropoff_time=[timeToUnix(pkup) for pkuo in duration["tpep_dropoff_datetime"].values]
    trip_Duration=(np.array(dropoff_time)-np.array(pickup_time))/float(60)
    NewFrame=df[['passanger_count','trip_distance','pickup_longitude','pickup_latitude','dropoff_longitude','dropoff_latitude']]
    NewFrame=df["trip_duration"]=trip_duration
    NewFrame=df["pickup_time"]=pickup_time
    NewFrame=df["speed"]=(NewFrame["trip_distance"]/NewFrame["Trip_Duration"])*60
    print("Time taken for creation of dataframe is {}".format(datetime.now()-startTime))
    return NewFrame
new_frame=dfWithTripTimes(data)

Upvotes: 1

Views: 3679

Answers (1)

MRocklin
MRocklin

Reputation: 57319

Only Dask DataFrame objects have a .compute method. The error that you get is consistent with your dataframe being a Pandas DataFrame instead. If you are using Pandas then there is no need to call .compute()

Upvotes: 3

Related Questions