Modin SQL API Demo

modin_sql_demo01.py

import modin.pandas as pd
gstore_apps_df = pd.read_csv("https://tinyurl.com/googleplaystorecsv")

modin_sql_demo02.py

"SELECT App, Category, Rating FROM gstore_apps WHERE Price = '0'"

modin_sql_demo03.py

# Select columns
result_df1 = gstore_apps_df.loc[,: ['App', 'Category', 'Rating']]
# From the resulting dataframe filter where Price = 0
result_df = result_df1['Price'] == 0

modin_sql_demo04.py

import modin.sql as mdsql

modin_sql_demo05.py

sql_str = "SELECT App,Category,Rating FROM gstore_apps WHERE Price = '0'"

modin_sql_demo06.py

result_df = mdsql.query(sql_str, gstore_apps=gstore_apps_df)

modin_sql_demo07.py

query_str = "SELECT App, Category, Rating WHERE Price = '0'"
result_df = mdsql.query(query_str, from=gstore_apps_df)

modin_sql_demo08.py

gstore_reviews_df = pd.read_csv("https://tinyurl.com/gstorereviewscsv")

modin_sql_demo09.py

sql_str = (
    """SELECT category,
              avg(sentiment_polarity) as avg_sentiment_polarity,
              avg(sentiment_subjectivity) as avg_sentiment_subjectivity
       FROM (
           SELECT category, sentiment, sentiment_polarity
           FROM google_items_df INNER JOIN gstore_reviews_df
           ON google_apps_df.app = gstore_reviews_df.app
       )
       GROUP BY category
       HAVING CAST(avg_sentiment_subjectivity as float) < 0.5
       ORDER BY avg_sentiment_polarity DESC
       LIMIT 10"""
)
result_df = mdsql.query(
    sql_str,
    google_apps_df=google_apps_df,
    gstore_reviews_df=gstore_reviews_df
)

modin_sql_demo10.py

# join the items and reviews
result_df = mdsql.query(
    """SELECT category, sentiment, sentiment_polarity 
       FROM google_items_df INNER JOIN gstore_reviews_df 
       ON google_apps_df.app = gstore_reviews_df.app""", 
    google_apps_df=google_apps_df,
    gstore_reviews_df=gstore_reviews_df
)
# group by category and calculate averages
result_df = mdsql.query(
    """SELECT category, 
              avg(sentiment_polarity) as avg_sentiment_polarity,
              avg(sentiment_subjectivity) as avg_sentiment_subjectivity 
       GROUP BY category
       HAVING CAST(avg_sentiment_subjectivity as float) < 0.5
       ORDER BY avg_sentiment_polarity DESC
       LIMIT 10""",
    from=result_df
)