Alex Baden

I can already se a problem with the above approach, I think. Converting the "0" cuboids is probably creating a lot of 0 cuboids... Might be worthwhile to write a...

It should be pretty easy, especially since we only have to migrate the base resolution. The catch is that I can't do the migration in place -- meaning, we'll need...

FYI, checked in on the job again this morning. The migration is still running on res0.

Let's hold off on taking any action until I've had a chance to test the S3 stack. We'll add his to the infrastructure agenda Tuesday. On Fri, Sep 2, 2016...

That's one option. But I was thinking the histogram would be computed on cuboid creation before the cuboid is stored in the DB. This would solve the problem of computing...

I have some skeletons you can use as test data

@kunallillaney think this is legit? I'm having a lot of problems with the bounding box code. Fixing this didn't fix my problem, but maybe it is a contributing factor...

Odd. I'll dig into it. Alex On Thu, Jun 9, 2016 at 01:11 Anish [email protected] wrote: > Hello! > > The data in Token Ex14R58 has the following dimensions: 4750x3410x41....

I haven't had a chance to look yet. I'm assuming it's a result of long-ago changes to the way imagesize / offset works. We had a similar issue with CAJAL...

Right -- I'm not entirely sure how we left off multiple channels in one HDF5. But, if we did support multiple channels in one HDF5, we could assume adding the...