Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Prefect Community

is this relevant- can I override a serializer task param for numpy arrays?
<https://github.com/PrefectHQ/prefect/issues/2639#issuecomment-635002663>

good question, you may need to build a custom serializer e.g. NumpyArraySerializer and pass it there… I will check

So there is a <https://docs.prefect.io/api/latest/engine/serializers.html#pandasserializer|PandasSerializer> and it allows to pass an arbitrary file type analogically to Pandas dataframe’s “to_XYZ” method e.g. `<http://df.to|df.to>_csv()`, `<http://df.to|df.to>_parquet()`, etc. And there is a “to_numpy” method: <https://pandas.pydata.org/docs/reference/api/pandas.DataFrame.to_numpy.html>  Not sure if this will work with numpy, but perhaps it would be worth exploring?

You can make your own serializer. The <https://github.com/PrefectHQ/prefect/blob/ca19b43947eb16e36fe9ea777e4e65c15baaf0b8/src/prefect/engine/serializers.py#L97-L122|JSON> one is very easy to follow (and the PickleSerializer also)

got it... what I was missing was this
<https://docs.prefect.io/core/concepts/results.html#choose-a-serializer>