Hi, it seem to me this is a not a BigQuery limitation but rather related to the behavior of JSON_EXTRACT and JSON_EXTRACT_SCALAR. see this old discussion [1]. [1]:https://issuetracker.google.com/38021949

Hi, It seems the doc[1] you shared is the only one I can find that describes the feedforward NN model architecture. Keep in mind that AutoML Tables is covered by the Pre-GA Offerings Terms of the Google Cloud Terms of Service. But I encourage you to ...

Hi, The issue you are reporting seem very similar to this stackoverflow post[1], Can you try the suggested solution offered by Ernesto? Best Regards /Samar [1]: https://stackoverflow.com/questions/70172711/pandas-df-from-cloud-functions-to-bigquery-o...

Hello, I agree that using more slots (10K vs 2K) would yield a faster query processing time if you run the same query on the same data. Unfortunately we cannot see the root cause of the behavior you are reporting unless we have the Job ID of the quer...