dimatura

28 ms·

301.
by dimatura 6y ago
I haven't looked into this drone in depth, but I think it may be feasible. The DJI Mini2, which any civilian can buy for $449, has a similar size, weighs 249g, and advertises a flight time of 30 minutes. (In my experience with the Mini
302.
by dimatura 6y ago
Good to hear Theano has come back to life, it was my first deep learning library :). These days I do PyTorch, and while I really appreciate its debuggability and flexibility, I definitely can see in retrospect some of the advantages of Thea
303.
by dimatura 6y ago
Yeah, no. I don't think I've ever seen any kind of support for Linux from Sony. I have to dust off my ancient windows laptop whenever I want to upgrade the firmware on my cameras.
304.
by dimatura 6y ago
No answers here... for me the meta-question is why are they so bad at user-facing software? (I don't really know about the quality of the firmware, it gets the job done, I guess). I like their cameras -- I own an NEX5N, a6000 and an RX
305.
by dimatura 6y ago
It's not really the same. True, MIDI won't work for say, iTunes, because it doesn't capture a lot of the information needed to recreate the audio of your average song. But for certain types of video, you can capture everythin
306.
by dimatura 6y ago
I think this trend may be limited to fields relying heavily on natively compiled dependencies, which may not even have any python at all. Conda packages can be C++, Fortran, pretty much anything. That's the main draw for me -- otherwis
307.
by dimatura 6y ago
Yeah, so while conda can be annoyingly slow, comparing it directly to pip in that department is unfair. A more fair comparison would be pip plus poetry or pipenv, which also have speed complaints :/
308.
by dimatura 6y ago
Qualitative opinion as someone who flies drones for fun and profit: yeah, drones, specially the bigger ones (e.g., M600), can be really loud. Once they get above a certain altitude you won't notice the noise, specially in a crowded urb
309.
by dimatura 6y ago
That's exactly the idea. Drones are relatively energy-efficient and are ill-suited for long range transport, specially if they have to haul a package. But for "last mile" delivery it might make sense.
310.
by dimatura 6y ago
Funny how that happens. In 2006 I was taken a grad-level Intro to ML course. After the first (and only) lecture on neural nets, I asked the prof on recommendations for learning more - since I had mostly a cogsci background, I was pretty int
311.
by dimatura 6y ago
From someone who used some SVMs in computer vision, before CNNs took over: * If you don't know the "shape" of your data, you can at least try the "kitchen sink" approach; just try out a bunch of different tranformat
312.
by dimatura 6y ago
Mostly S3. I used to do SSH, but these days I can afford to keep the data in the cloud. I do appreciate the possibility of migrating to other stores if needed in the future, though - might have to soon, for $reasons.
313.
by dimatura 6y ago
Yeah, git-annex has a lot of cool features that I have yet to see in other systems. I still use it for some things. My main pain point on MacOS was that the symlink mode didn't work well with some apps that didn't understand symli
314.
by dimatura 6y ago
It's subjective, naturally. I don't know if I'd be that enthusiastic, but I definitely like it a lot. My main languages are Python and C++. I've also used Java, C#, Javascript but not Typescript, among others. In my op
315.
by dimatura 6y ago
I have been interested in "git for binary data" for a while, mostly for ML/computer vision purposes. I've tried quite a few systems. Of course, there's git-lfs (which keeps "pointer" files and blobs in a c
316.
by dimatura 6y ago
I'm guessing the economy of scale isn't quite there, as there's much less demand for DIY IOT devices than phones. But there's some products out there that approach this idea. E.g., https://www.seeedstudio.com&
317.
by dimatura 6y ago
I've mainly used gunicorn and uwsgi, without seeing any large differences. But my apps have fairly light network requirements. It seems like the new trend for Python servers is ASGI ( https://asgi.readthedocs.io/en/
318.
by dimatura 6y ago
While on a work trip to Berlin, I walked into a random lebanese place in Neukolln to get a quick bite. Looked at the price of a shawarma - 2 euros - and thought, damn, this is going to be awful. But I was in a hurry and got one (actually tw
319.
by dimatura 7y ago
I've used moviepy ( https://github.com/Zulko/moviepy ) with some success. Doesn't do Final Cut export but it uses ffmpeg under the hood, so it could be used to create raw clips in some lossless format.
320.
by dimatura 7y ago
In the past I have used moviepy ( https://github.com/Zulko/moviepy ) to programatically create videos in Python. At a very high level, it manages creation of image frames that are ultimately fed into ffmpeg. It is (was?)
321.
by dimatura 7y ago
In my opinion, the point about automatic differentiation isn't quite correct. Automatic differentiation will help you compute gradients for complicated functions involving multiple operations, as long as each of these operations are th
322.
by dimatura 7y ago
A critical piece of wisdom I find myself repeating to younger students, handed down from an advisor: "always look at the pixels". (Sometimes replacing pixels with voxels/lidar points/meshes, etc.)
323.
by dimatura 7y ago
Seems like they are using photogrammetry, most likely with a large amount of high resolution images. Photogrammetry under these conditions can be fairly computationally expensive, specially if they use aim for high quality reconstructions (
324.
by dimatura 7y ago
Similar situation here, ended up with the same solution, after an initial attempt with HLS. jsmpeg ( https://github.com/phoboslab/jsmpeg ) made it pretty easy.
325.
by dimatura 7y ago
Yeah, you can do that as an inference step, like you described, and that has its uses without any learning required. But you can also make the output of the renderer depend (differentiably) on parameters apart from whatever you have as inpu
326.
by dimatura 7y ago
Sure, I can imagine that. Would be surprised if there isn't SIGGRAPH papers to that effect. At least initially, it would be more of an optimization problem, but ML could help as well.
327.
by dimatura 7y ago
This could be used that way, and there is some precedent, e.g. in RenderGAN ( https://arxiv.org/abs/1611.01331 ). Note the GAN part: that's why the "differentiable" part is important, so you can integrat
328.
by dimatura 7y ago
Nice! I played with OpenDR ( http://files.is.tue.mpg.de/black/papers/OpenDR.pdf ) a few years ago, and got really excited about it. Unfortunately it uses a custom autodiff implementation that made it hard it to inte
329.
by dimatura 7y ago
I'd be surprised if gmail.com wasn't one of the worst performers on this hypothetical list.
330.
by dimatura 7y ago
Update: read the paper. It's spot on, great read.

More ›