01-28-2022 05:57 AM
I have created a feature table (Databricks runtime ML 10.2) that includes a timestamp column as a primary key, that is not used as a feature but as a column to join on.
I have then created a model that trains from this feature table and some additional data, which excludes the primary keys. I tried excluding them, both using the feature store api, and from the sklearn api. The model is being trained fine, but when use the score_batch() method, I get an error claiming that 'TypeError: float() argument must be a string or a number, not 'Timestamp''.
This error is coming from sklearn, so is there some incompatibility there, or is this a bug in feature store?
Steps to reproduce :
01-28-2022 07:15 AM
01-28-2022 07:15 AM
maybe you can just try to cast timestamp as int
01-29-2022 05:20 AM
Thanks for your reply Hubert. Yes, casting it to long or int does solve the issue, but it is a workaround and I would like to keep the data as-is, with directly interpretable timestamps, especially when there is no reason why they should trigger an error during the prediction step since it is not being used at that stage.
03-11-2022 01:39 AM
Hi @Thibault Daoulas , Databricks released runtime ML 10.2 in December 2021. Here are the important improvisations. You can also refer to the documentation here.
Databricks Runtime ML includes AutoML, a tool to automatically train machine learning pipelines.
The FeatureStoreClient interface has been simplified.
For more information, see Work with feature tables and Databricks Feature Store Python API.
03-14-2022 05:06 PM
Hi @Thibault Daoulas ,
Did @Kaniz Fatma response help you to resolved your question? if yes, please mark it as best response. If not, please let us know.
03-15-2022 12:56 AM
Hi, it did not, but at least I know they are not fully supported so a workaround is to avoid timestamps, so I suppose you can mark this as resolved
03-15-2022 11:34 PM
Thank you @Thibault Daoulas for the update. Can you mark one of the answers whichever you feel is the best?
Join our fast-growing data practitioner and expert community of 80K+ members, ready to discover, help and collaborate together while making meaningful connections.
Click here to register and join today!
Engage in exciting technical discussions, join a group with your peers and meet our Featured Members.