TypeError: 'DataFrame' object does not support item assignment when using division

Question 1

TypeError: 'DataFrame' object does not support item assignment when using division.

Dividing two integer columns in a pyspark dataframe and receiving this error. Any ideas?

Reproducible:

simpleData = (((1, 2), (2, 6), (3, 6))
columns= ["sales","users"]
df = spark.createDataFrame(data = simpleData, schema = columns)
df = df.withColumn("users", df["users"].cast(IntegerType()))
df = df.withColumn("sales", df["sales"].cast(IntegerType()))
df['buy_rate'] = df['sales']/ df['users']

Question 2

just small tweak .

from pyspark.sql.types import IntegerType
import pyspark.sql.functions as F
simpleData=((1, 2), (2, 6), (3, 6))
columns= ["sales","users"]
df = spark.createDataFrame(data = simpleData, schema = columns)
df = df.withColumn("users", df["users"].cast(IntegerType()))
df = df.withColumn("sales", df["sales"].cast(IntegerType()))

df1 = df\
    .withColumn("buy_rate", (F.col("sales") / F.col("users")))
display(df1)

Karthikeyan Rasipalay Durairaj · Answer 1 · 2021-11-23T20:01:03

just small tweak .

from pyspark.sql.types import IntegerType
import pyspark.sql.functions as F
simpleData=((1, 2), (2, 6), (3, 6))
columns= ["sales","users"]
df = spark.createDataFrame(data = simpleData, schema = columns)
df = df.withColumn("users", df["users"].cast(IntegerType()))
df = df.withColumn("sales", df["sales"].cast(IntegerType()))

df1 = df\
    .withColumn("buy_rate", (F.col("sales") / F.col("users")))
display(df1)

Thanks Karthik, and very good approach with the assignment to the new column within the withColumn() this worked beautifully.

TypeError: 'DataFrame' object does not support item assignment when using division

In other languages

This page is in other languages

Popular in the category