Pandas interface

Pandas 1 is one of the key packages used in most data science projects done in Python. The associated import interface is one of the slowest, but you can harness the good data exploration and manipulation capabilities of this Python package.

Import from Pandas

Using the from_pandas() class method, you can create a new DataFrame based on the provided pandas.DataFrame. The read_pandas() method will replace the content of the current DataFrame instance or append further rows.

Export to Pandas

In addition to reading data from a pandas.DataFrame, you can also write an existing getml.data.DataFrame back into a pandas.DataFrame using to_pandas(). Due to the way data is stored within the getML engine, the dtypes of the original pandas.DataFrame can not be restored properly and their might be inconsistencies in the order of microseconds being introduced into timestamps.

1

https://pandas.pydata.org/