dask and varying stepsize between channels

[Christopher Snyder]

Hi Patrick,

First of all awesome work. I really like what you've done here. I very nearly went from not having a solution to having a plausible solution 10 minutes and 1 "pip install sima" later.

(1) You may find it useful as I have to use or at least support dask to manage your images. It seems to do a lot of the things your custom Sequence class does. I think you'll like it. It lets me pretend I'm using numpy even though my images can't fit in memory. But, currently to convert it to a Sequence, you have to convert it to a numpy array first, which is not always possible since, yeah, out-of-core.

(2) Do you have support for channels that have different voxel spacing? I didn't see anything like that, but it seems like a common use case. Maybe it's just me. I would like to use the output of HMM.estimate() to manipulate data with the same field of view, but possibly different spacing. I think I'm on my own with that one.

Best,

Chris

[Patrick Kaifosh]

Christopher Snyder wrote:
> Hi Patrick,
>
> First of all awesome work. I really like what you've done here. I very
> nearly went from not having a solution to having a plausible solution
> 10 minutes and 1 "pip install sima" later.

Thanks. Please note that a lot of credit also goes to the other members
of the project. Certainly not just to me.

>
> (1) You may find it useful as I have to use or at least support dask

> <http://dask.pydata.org/en/latest/> to manage your images. It seems to

> do a lot of the things your custom Sequence class does. I think you'll
> like it. It lets me pretend I'm using numpy even though my images
> can't fit in memory. But, currently to convert it to a Sequence, you
> have to convert it to a numpy array first, which is not always
> possible since, yeah, out-of-core.

Cool. I hadn't heard of dask. If you are interested, you could probably
build a Sequence subclass that uses dask quite easily, and then
contribute it back to the project.

>
> (2) Do you have support for channels that have different voxel
> spacing? I didn't see anything like that, but it seems like a common
> use case. Maybe it's just me. I would like to use the output of
> HMM.estimate() to manipulate data with the same field of view, but
> possibly different spacing. I think I'm on my own with that one.

We have some thoughts on the matter (we were more thinking about
different scan paths rather than voxel spacing) but we haven't coded up
too much there yet. If you want, we could point you in the direction of
how to work on adapting what we have toward that end.
>
> Best,
> Chris
> --
> You received this message because you are subscribed to the Google
> Groups "sima-users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to sima-users+...@googlegroups.com
> <mailto:sima-users+...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/sima-users/e8b9474c-292e-47b0-bcb0-3fc9b10c9581%40googlegroups.com
> <https://groups.google.com/d/msgid/sima-users/e8b9474c-292e-47b0-bcb0-3fc9b10c9581%40googlegroups.com?utm_medium=email&utm_source=footer>.
> For more options, visit https://groups.google.com/d/optout.