The transform function in groupby is used to perform transformation operations on a groupby object. For example, we could replace NaN values in the groupby object using the fillna method. The resultant object after using transform has the same size as the original groupby object.
Let's introduce NAs into the sample sales data. The following code injects NAs into 25% of the records:
na_df = sales_data[["Sales", "Quantity", "Discount", "Profit", "Category"]].set_index("Category").mask(np.random.random(sales_data[["Sales", "Quantity", "Discount", "Profit"]].shape) < .25)na_df.head(10)
The following will be the output:
Snapshot of data with NAs inserted
Now, the four quantitative variables contain NAs in 25% of ...
