Pyspark has a to_date function to extract the date from a timestamp. In your example you could create a new column with just the date by doing the following:
from pyspark.sql.functions import col, to_date
df = df.withColumn('date_only', to_date(col('date_time')))
If the column you are trying to convert is a string you can set the format parameter of to_date specifying the datetime format of the string.
You can read more about to_date in the documentation here.