Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.featureform.com/llms.txt

Use this file to discover all available pages before exploring further.

@spark.df_transform(inputs=[upstream_data])
def transformation_name(upstream):
  ...
client.dataframe(transformation_name)
In scenarios where the dataset is sizable enough to exceed memory limits or when a sample suffices, you can include the rows parameter to restrict the number of pulled rows. 
client.dataframe(transformation_name, rows=10000)
Fetching training sets as dataframes is also feasible. In this case, Featureform’s serving API equips the training set object with a .dataframe() method. 
client.training_set("name", "variant").dataframe()
These mechanisms offer an intuitive and efficient approach to delve into your data using the familiar Dataframe APIs.