Which JXL supports, by the way. Tiling is mandatory for images bigger than 2048x2048, and you can construct images based on an 8x downscaled version, recursing that up to 4 times for up to 4096x downscaling.
That is awesome. In my domain, images (TIFFs usually) are up to 1m x 1m pixels and scaling usually goes 4x so that if you need 2x scaling you can just read 4 times as many tiles from the higher resolution level and downscale. With 8x scaling you need to go a level further - reading 16 pixels from the image to create 1 pixel of output. Not great but it would work and 4096 scaling would make the lowest resolution image 256 x 256 which is just what you need.
By the "g as in graphics" logic for "gif", for "jpeg" we have "j as in joint", "p as in photographic", "e as in experts", and "g as in group", so "je-feck"?
Qalculate! has been my go-to calculator on my laptop for years, very happy to have it on my phone now too!
And it definitely knows planet radii, try `planet("earth"; "radius")`. Specifically, it knows a bunch of attributes about atoms (most importantly to me, their atomic mass) and planets (including niche things like mean surface temperature). You can see all the data here: https://qalculate.github.io/manual/qalculate-definitions-fun...
The issue is not so much that it's cellulosic as that it's lignocellulosic; we have a pretty good grasp these days on cellulose (and hemicellulose, but we can handle that too so I'll skip over it). Lignin is (oversimplifying, but not by too much) what makes the difference between soft paper pulp and hard woodchips. The lignin crosslinks the (hemi)cellulose and makes it much harder to access for the enzymes we'd use to break down the cellulose, while itself being very difficult to break down (it's thought that the whole reason we have coal is due to how long it took for fungi to evolve the enzymes needed to break down lignin effectively in order to degrade wood).
You need a bunch of equipment and/or biochemical processing to break down lignocellulosic plant matter into something that can be efficiently fermented (keyword: "lignocellulosic biomass pretreatment"), so while thay may be available at a lab scale, it's not necessarily possible on an industrial scale (yet).
As one example, there's a method called "steam explosion" where you apply very hot (around 160-260 °C) and high-pressure (tens of atmospheres) steam to the biomass, then release the pressure relatively fast. The hot steam causes chemical reactions like hydrolysis, and the pressure release breaks down the material physically as the steam expands. Imagine the sort of equipment you'd need to do that on an industrial scale. Nowhere out of reach for modern chemical engineering, but someone has to build it.
Now, as you might imagine from how relatively non-woody switchgrass is, it doesn't have a ton of lignin compared to trying to ferment woodchips, but it still has enough to be problematic, and in fact there's research being done on how to reduce its lignin content, such as by genetic engineering.
I see, it sounded like a buzzword term (which I guess it still might be). The point is that the current title makes it sound like a general comparison, while the original title makes it clear that it's comparing three implementations of a single tool.
Dane here. Yes, it's because of limited testing capacity (new machines arrived recently and they are ramping up to 1000 tests per day, additional capacity will among other things go to surveillance of cases with mild symptoms, modeled on our existing system for influenza-like illness surveillance where samples are, well, sampled among a specific subset of GPs), but it's not only those with severe symptoms who are tested, it's anyone who is hospitalized (which also includes members of vulnerable populations with only moderate symptoms) and health care workers, with the goal being to prevent hospital-acquired infections. In a single sentence, the strategy could be summed up as "people who aren't in the hospital should assume it's COVID-19 if they have symptoms and act accordingly, but once they get to the hospital we can't afford to assume".
Have you considered learning stenography? The Open Steno Project is very DIY, and I'm sure they'd be thrilled to help think up an even more mobile design for a steno writer (though AFAIK they're already very usable when walking around with a harness). When it comes to novel input methods, I really think it's a great idea far too few people know about.
I noticed that the paper mentions constructing a triangulation of a set of initial sample points, which would allow for the optimization domain to be any convex polytope. Is this something you are planning to explore/incorporate?
EDIT: I see now that this is not an implemention of the algorithm in the paper, but its own algorithm entirely. Even so, it appears to me the same technique could be used in this case.
Yes, I hope to add that feature soon-ish. The reason I left it out of this first release is that computing Delaunay triangulations scales very poorly with dimensionality. That was why TRIOPT was restricted to low dimensional spaces. I will have to implement a more efficient way to compute nonoverlapping triangulations. In the mean time, you can simply make the optimization domain bigger to encapsulate the shape of your problem domain.
I think the point is that you can build a boat out of bricks, and you can build neural networks out of linear regression. But that doesn't make modern AI "just fancy linear regression" the same way modern boats aren't "just fancy bricks".
The theoretical property those hacks were aiming at showing is called the transitive reduction. And you don't need any hacks to calculate it, graphviz comes with the `tred` tool to do so.
