ChrisFoster

9 ms·

31.
by ChrisFoster 8y ago
I'd argue the opposite. Vanilla CS theory assumes a very simple system model for the measurement process (ie, linear measurement) and shows that with a few assumptions you need far fewer measurements than you might expect. This by itse
32.
by ChrisFoster 8y ago
Waste heat must be radiated away, though thermodynamics would be quite content if that radiation is anisotropic. For example, consider a perfectly insulating black box with small hole. Observers on the opposite side from the hole will not o
33.
by ChrisFoster 8y ago
This is one of the main things which has driven the development of julia's logging system; it seems that some people just don't like debuggers very much (I'm one of them) and find staring at logs and attempting to run the cod
34.
by ChrisFoster 8y ago
I second that, the CMOG channel is great. They have high production quality (especially consider the constraints of the glass shop) and really talented artists / craftspeople. Bill Gudenrath's series on Venetian glassware is parti
35.
by ChrisFoster 8y ago
Thanks! At a quick glance looks like this may not be useful for the problem I had in mind, but I'll have to read it in more detail.
36.
by ChrisFoster 8y ago
My feeling is that Flux is a fantastic tool for playing with innovative models which can't be expressed in the usual frameworks. And quite possibly already the best thing that exists for open ended experimentation in ML. On the other h
37.
by ChrisFoster 8y ago
Hah, that someone was me, I'm glad it clicked for you! I had a similar experience of it "just working" when trying a few weeks ago. I particularly enjoyed being able to build and run Flux models interactively just like any ot
38.
by ChrisFoster 8y ago
> faster than the standard way of calculating these derivatives via sensitivity analysis This is quite shocking indeed, I'll look for the paper when it comes out. I'm particularly interested in whether this is true for the sens
39.
by ChrisFoster 8y ago
I've just been trying out a few deep learning frameworks recently, and found the julia framework Flux to be enormously fun, flexible and generally just extremely hackable. If you like lisp you might also find it a pleasing language - e
40.
by ChrisFoster 8y ago
Not really. ML systems are not created in isolation by a single team; they form an ecosystem with many players. I'm sure the current crop of cybercriminals would like an army of bots which manipulates people into giving away all their
41.
by ChrisFoster 8y ago
True, but it does provide at least some measure of reproducibility. Quality of implementation and reproducibility are orthogonal and both very valuable in their own right.
42.
by ChrisFoster 8y ago
It strikes me that even with a perfectly regular starting schedule, buses might clump together in time because the schedule is probably dynamically unstable. To explain, picking up passengers from a stop costs time and a long time between b
43.
by ChrisFoster 8y ago
It definitely does. I've practiced this a little myself and in louder environments you need a louder click. On the other hand, ambient noise can give rich information about close obstacles without yourself or that object needing to act
44.
by ChrisFoster 8y ago
This is hard to get right. Personally I try not to be a grumpy maintainer but I'm frequently a picky one. Lately I've been an absent one. I think maintainers can frequently come across as grumpy simply because their goals are diff
45.
by ChrisFoster 8y ago
It's not that people are uninterested; it's more that inventing a system for specifying and enforcing these kind of generic interfaces is a really hard design problem. As a comparison, consider that the C++ standards body has be
46.
by ChrisFoster 8y ago
Sometimes low level debugging is a surprisingly pleasant experience as the julia JIT generates proper DWARF debug info. So for instance, you can break in gdb and see the julia source code for any julia generated stack frames, neatly intertw
47.
by ChrisFoster 8y ago
In 0.7 I put quite a lot of design effort into the logging system to allow log records to be redirected in a sane way. It's by no means a complete system, but I think the changes were worthwhile and the core infrastructure is now in pl
48.
by ChrisFoster 8y ago
I've been using it in production since version 0.5. It's a joy for creating high performance numerical code without having to leave the comfort of a productive and interactive scientific computing environment. Perhaps surprisingly
49.
by ChrisFoster 8y ago
Dependency management for packages just got hugely better with the release of Pkg3 (now simply Pkg) as part of 0.7 / 1.0.
50.
by ChrisFoster 8y ago
Or, to see what's really going on at a low level you have immediate access to the native assembly from the REPL: julia> square(x) = x^2 square (generic function with 1 method) julia> @code_native square(1)
51.
by ChrisFoster 8y ago
Some people seem to be objecting to this as a fluff piece, but it speaks to me on a deep level. I develop free software because I like to learn and create; because good tools make work a joy; because I like to make useful things. And I&#x27
52.
by ChrisFoster 8y ago
As a slight variation on the same theme - just the anticipation of a social event can make me quite anxious and edgy regardless of whether I've been part of the planning. The thing that fixes it for me is going to events with a clear g
53.
by ChrisFoster 9y ago
Yes, the joy of a great design is in the delicate balance it strikes between simplicity and conflicting needs. There's plenty of anguish in finding the balance even in the head of a single person. The space of solutions to a design pro
54.
by ChrisFoster 9y ago
These are good points about computational complexity, but there's more to it: the NN architecture regularizes the representations which may be learned. A fully connected architecture may be fully general, but without powerful regulari
55.
by ChrisFoster 9y ago
Previous discussion, from 2013: https://news.ycombinator.com/item?id=5279307
56.
by ChrisFoster 9y ago
Maturity of the ecosystem and a growing awareness of the problem that this solves, perhaps? It's a kind of ripeness for change which, I think occurs in every large project. There's that gnawing feeling that something is more diff
57.
by ChrisFoster 9y ago
You're crazy if you're spending millions on matlab licenses. Just buy a copy of the matlab "compiler" mcc (not much of a compiler - more of packaging system after compilation to byte code, as I understand it). It'
58.
by ChrisFoster 9y ago
The preprocessor is a workable but somewhat poor tool for inlining code due to several reasons (no overloading, no sensible macro scoping, template argument lists clash with macro argument lists, etc). However, it's quite a reasonable
59.
by ChrisFoster 10y ago
Wow, this is such a beautiful example of basic science turning into powerful technology. Technology which looks like it will be both cheaper and better for the environment than any current alternative. Does anyone have an idea of how long
60.
by ChrisFoster 10y ago
I've had fun debugging this very problem in the past, as the root cause of a very puzzling intermittent failure. The symptoms were intermittent segfaults in some high level scientific python code running on a large cluster in AWS. Give

More ›