yuanchuan·9 anni fa·discussTensorflow might not be the fastest in terms of computation speed, but it can be used from research to production with Tensorflow Serving.As such you won't need to implement/convert your model in another format for usage.
yuanchuan·10 anni fa·discussIt is definitely doable. You can refer to this blog post by AWS (https://aws.amazon.com/blogs/big-data/join-amazon-redshift-a...) to set up FDW to Redshift.What is more exciting is you can leverage Redshift MPP architecture with this method.
As such you won't need to implement/convert your model in another format for usage.