Maxwell Flitton

Thanks for the response, I'm getting matrix dimensions errors. I loop through the `state_dict` and append the weights to a list with the following structure: I have the outcome categories...

thanks for your issue. I've been working on sorting out some other issues but will get on this today

@jarodium it turned out to be the `cbor` tagging. I've added a test script to recreate your problem. The following [pull request](https://github.com/surrealdb/surrealdb.py/pull/178/files) has the fix and test script.

@jarodium the pull request was passing all tests and i think the issue is solved. I'm now closing this issue but feel free to open another in the future if...

did you get a solution to this?

@KarateSnowMachine with the new python client we give 100% control over the coroutines to the user. From what I understand, the python client in the interim was executing async coroutines...

Due to inactivity I'm closing this issue. We are working on a solving a batching async problem [here](https://github.com/surrealdb/surrealdb.py/issues/164). If this addresses your problem please engage with this issue. If not...

@vladimirrotariu thanks for raising this, at this point in time the surrealML will does an execution per input which is not optimal but will be working on vector calculations once...

@LaurentMazare thanks for your reply. I'm working at SurrealDB which is a database written in rust and we're building support for inference for traced models exported in pytorch. We have...

@LaurentMazare As long as we can do inference on models alongside neural networks with ONNX I'm down for that. A lot of models in production are random forests and [hummingbird](https://github.com/microsoft/hummingbird)...