Provide PySpark code to compute the acceptance rate per date based on the given table and code.

Question

Assisting AI · Accepted Answer

Here is a concise PySpark solution:

```python
from pyspark.sql import functions as F

# Assume df is the original DataFrame with columns: date, action
sent = df.filter(F.col("action") == "sent")
    .groupBy("date")
    .agg(F.count("*" ).alias("sent_cnt"))

accepted = df.filter(F.col("action") == "accepted")
    .groupBy("date")
    .agg(F.count("*" ).alias("accepted_cnt"))

# Join the two aggregates on date
rate_df = sent.join(accepted, "date", "inner")
    .withColumn("acceptance_rate", (F.col("accepted_cnt") / F.col("sent_cnt")) * 100)
    .select("date", "acceptance_rate")

rate_df.show()
```

This code groups by date, counts sent and accepted actions, joins the counts, and calculates the percentage. Complexity is O(n) for the filters and groupBy, and O(m) for the join, where n is the number of rows and m is the number of distinct dates.

Provide PySpark code to compute the acceptance rate per date based on the given table and code.

💡 Model Answer

🎤 Get questions like this answered in real-time