IF Statement Pyspark

Question

My data looks like the following:

+----------+-------------+-------+--------------------+--------------+---+
|purch_date|  purch_class|tot_amt|       serv-provider|purch_location| id|
+----------+-------------+-------+--------------------+--------------+---+
|03/11/2017|Uncategorized| -17.53|             HOVER  |              |  0|
|02/11/2017|    Groceries| -70.05|1774 MAC'S CONVEN...|     BRAMPTON |  1|
|31/10/2017|Gasoline/Fuel|    -20|              ESSO  |              |  2|
|31/10/2017|       Travel|     -9|TORONTO PARKING A...|      TORONTO |  3|
|30/10/2017|    Groceries|  -1.84|         LONGO'S # 2|              |  4|

I am attempting to create a binary column which will be defined by the value of the tot_amt column. I would like to add this column to the above data. If tot_amt <(-50) I would like it to return 0 and if tot_amt > (-50) I would like it to return 1 in a new column.

My attempt so far:

from pyspark.sql.types import IntegerType
from pyspark.sql.functions import udf

def y(row):
    if row['tot_amt'] < (-50):
        val = 1
    else:
        val = 0
        return val

y_udf = udf(y, IntegerType())
df_7 = df_4.withColumn('Y',y_udf(df_4['tot_amt'], (df_4['purch_class'], 
(df_4['purch_date'], (df_4['serv-provider'], (df_4['purch_location'])))
display(df_7)

Error message I'm receiving:

SparkException: Job aborted due to stage failure: Task 0 in stage 67.0 failed 
1 times, most recent failure: Lost task 0.0 in stage 67.0 (TID 85, localhost, 
executor driver): org.apache.spark.api.python.PythonException: Traceback (most 
recent call last):
File "/databricks/spark/python/pyspark/worker.py", line 177, in main
process()
File "/databricks/spark/python/pyspark/worker.py", line 172, in process
serializer.dump_stream(func(split_index, iterator), outfile)
File "/databricks/spark/python/pyspark/worker.py", line 104, in <lambda>
func = lambda _, it: map(mapper, it)
File "<string>", line 1, in <lambda>
File "/databricks/spark/python/pyspark/worker.py", line 71, in <lambda>
return lambda *a: f(*a)
TypeError: y() takes exactly 1 argument (2 given)

Answer 1

You shouldn't need a UDF for this - use the built-in function when instead. Here is an example with toy data similar to your tot_amt column:

spark.version
# u'2.2.0'

from pyspark.sql import Row
from pyspark.sql.functions import col, when

df = spark.createDataFrame([Row(-17.53),
                              Row(-70.05),
                              Row(-20.),
                              Row(-9.),
                              Row(-1.84)
                             ],
                              ["tot_amt"])

df.show()
# +-------+
# |tot_amt|
# +-------+
# | -17.53| 
# | -70.05|
# |  -20.0|
# |   -9.0|
# |  -1.84|
# +-------+

df.withColumn('Y', when(col('tot_amt') < -50., 1).otherwise(0)).show()
# +-------+---+ 
# |tot_amt|  Y|
# +-------+---+
# | -17.53|  0|
# | -70.05|  1|
# |  -20.0|  0|
# |   -9.0|  0| 
# |  -1.84|  0|
# +-------+---+

Answer 2

How to make it work (pass struct)

from pyspark.sql.functions import struct

df_4.withColumn("y", y_udf(
    # Include columns you want
    struct(df_4['tot_amt'], df_4['purch_class'])
))

What would make more sense

y_udf = udf(lambda y: 1 if y < -50 else 0, IntegerType())

df_4.withColumn("y", y_udf('tot_amt'))

How it suppose to be done:

from pyspark.sql.functions import when

df_4.withColumn("y", when(df_4['tot_amt'] < -50, 1).otherwise(0))