Meer data ophalen

Feature Engineering met PySpark

John Hogue

Lead Data Scientist, General Mills

Over externe datasets

PRO'S

Belangrijke predictors toevoegen
Waarden aanvullen/vervangen
Goedkoop of makkelijk te krijgen

Data combineren

CONTRA'S

Kan analyse vertragen
Snel data leakage veroorzaken
Word expert in de dataset

Over joins

Onze datarichtingen oriënteren

Links: onze startdataset
Rechts: nieuwe dataset om toe te voegen

SQL-joins

PySpark DataFrame-joins

DataFrame.join(

    other,      # Other DataFrame to merge

    on=None,    # The keys to join on

    how=None)   # Type of join to perform (default is 'inner')

PySpark-join: voorbeeld

# Inspect dataframe head
hdf.show(2)

+----------+--------------------+
|        dt|                  nm|
+----------+--------------------+
|2012-01-02|        New Year Day|
|2012-01-16|Martin Luther Kin...|
+----------+--------------------+
only showing top 2 rows

# Specify join conditon
cond = [df['OFFMARKETDATE'] == hdf['dt']]

# Join two hdf onto df
df = df.join(hdf, on=cond, 'left')

# How many sales occurred on bank holidays?
df.where(~df['nm'].isNull()).count()

SparkSQL-join

Pas SQL toe op je dataframe

# Register the dataframe as a temp table
df.createOrReplaceTempView("df")
hdf.createOrReplaceTempView("hdf")

# Write a SQL Statement
sql_df = spark.sql("""
                      SELECT 
                        *
                      FROM df
                      LEFT JOIN hdf
                      ON df.OFFMARKETDATE = hdf.dt
                   """)

Laten we wat data joinen!

Feature Engineering met PySpark