Geoffrey Bolmier

> Yeah, I think so. Now we do not provide such SDK or sdk. I am not sure if we should implment in Python or Go. Are you using Python...

You mean ORAS go client? [oras-project/oras-py](https://github.com/oras-project/oras-py) is still empty so I can't say 😄 (only using ORMB python sdk)

I'm actually getting more confused now. I run the following: ```shell $ docker pull python:3.7-slim $ docker run --rm -it --entrypoint /bin/bash python:3.7-slim root@:/$ apt-get update root@:/$ apt-get install vim-tiny...

Hi @guedes-joaofelipe! Glad to see you interest with this PR :) Usually the utility matrix is so large and sparse that we don't want to store it in its plain...

If I understand well, you need this format for compatible issues with your other models. It's not really efficient in most recommendation settings to use that format, are you using...

Which `pandas` version are you using? No problem for me with `1.2.1` and `1.3.4`

Great point @BrandonKMLee! We could either implement a new `predict_n_item` method or adapt the `predict` method to also handle this case. Feel free to open a pull request otherwise I'll...

Can you elaborate more on this, e.g. wdym by pre-rated items?

I was thinking of passing a list of the items' ids in parameter, so that it's easy to use as a mask on the item latent factor matrix

If we mask the item latent factor matrix we'll only compute the estimated ratings of interest, which will save time compared to computing all the estimated ratings and then mask...