Weiyuan Wu

No, the overlapped patches will be de-overlapped. E.g. for the 1d case, the original image is [a,b,c], patchily with step size 1 produces two patches. patch 1 is [a,b] and...

> Alright, I got it. > However, in my segmentation experiment, after applying unpatchify, the output shows a shifted results. Although its de-overlapping, the segmentation results should not shift, right?...

Does step size 2 work for you?

I don't think so. Since the generic API assumes the return type as a `Map`, so the compiler will infer `T` as a `Map`. Then everything falls back to the...

Actually using generic is just a workaround. Finally, I think we need to type-ify all the return types. Something like [this](https://transform.now.sh/json-to-rust-serde) is pretty useful.

On my side, the problematic machine has two NICs, `ens3` for external IP and ens4 for talking to the swarm manager. By forcing the docker to use the `ens4` for...

@softprops It is still not published yet to the crates.io. I can help you maintain this repo if you are too busy and don't mind to hand over this to...

I second the timeout solution or maybe expose the wait time/# messages to the user. Currently `MissingSubscriptionConfirmation` is guaranteed to be triggered if you subscribe to too many channels. What...

Thanks @Abdelgha-4 for the suggestion! Indeed we once considered the PhiK correlation at https://github.com/sfu-db/dataprep/pull/145. However, PhiK is generally very slow comparing to other correlations so we decide to defer the...

> I see! sorry then, I didn't notice it was already discussed. No worries! If you think this is an important feature then we can certainly add it.