olavolav

5 ms·

1.
On agile processes for AI research(olavstetter.com)
1 points by olavolav 3y ago|1 comments
2.
by olavolav 3y ago
Is agile the right approach for cross-functional teams of AI researchers and software developers? Here's what I learned from 10 years of leading such teams
3.
by olavolav 5y ago
Very cool library! As alternatives I might recommend hipsterplot (the name) or uniplot (very similar but lighter, I'm the author).
4.
by olavolav 5y ago
That's right, I'm using the 2x2 characters and some off-center horizontal line characters. Note aware of Unicode expansions, though more and more fonts seem to support the above. Sparklines in the terminal sounds very cool, did yo
5.
by olavolav 5y ago
Thanks for the question! So the labels only depend on the limits of the plotting window, which you can supply as a keyword option. In that sense, and together with using the same gridlines options, you could have the same background in some
6.
by olavolav 5y ago
Thanks! Do let me know if you are still looking for any particular feature
7.
by olavolav 5y ago
Very cool to see this – but will only be successful with great terminal plotting tools. The ones the author mentions like the matplotlib interface clearly won't do[0] due to lack of resolution A perfect use case for unicode plotting [
8.
by olavolav 5y ago
Author here: Version 0.4 added nice axis labels via an extended Wilkinson algorithm. For comparison: - gnuplot to the terminal https://www.datafix.com.au/BASHing/2019-06-21.html - UnicodePlots for Julia https:/&#
9.
Show HN: Beautiful and interactive Unicode plotting for the terminal(github.com)
5 points by olavolav 5y ago|2 comments
10.
by olavolav 6y ago
Speaking as a pytorch user, many of the steps in your Readme example remind me of the usual setup except the pipeline.run() handoff which is replaced by eager evaluation in pytorch. Are you seeing something like an eager mode for your libra
11.
by olavolav 6y ago
Thanks!
12.
by olavolav 6y ago
Hi, author here. This was & is a fun project, and maybe it's useful to some of you! Let me know if you have questions/comments/ideas, or check out the ASCII video: https://asciinema.org/a/Ldgn5pHOgxPJ
13.
Show HN: uniplot - Unicode plotting for debugging and ML pipelines(github.com)
8 points by olavolav 6y ago|3 comments
14.
by olavolav 7y ago
I'm guessing that's because they measured the return rate of institutions: "Wallets were returned to one of five societal institu- tions: (i) banks, (ii) theaters, museums, or other cultural establishments, (iii) post offic
15.
by olavolav 8y ago
Whatever test you do, you would need to know the total number of visitors in each group, right? And unless I missed it the article doesn't state those numbers. Intuitively, the numbers you quoted would be more significant the bigger th
16.
by olavolav 8y ago
Should any of you care for a more philosophical treatment of the idea of "cause", then I can wholeheartedly recommend: Bertrand Russel, On The Notion Of Cause Proceedings of the Aristotelian Society New Series, Vol. 13 (1912 - 191
17.
by olavolav 9y ago
Well, I think the OP should give you the idea that if your goal is (for example) to reduce violent crime rates overall, then the answer might be more complicated than that...
18.
by olavolav 9y ago
Dear HN audience. For almost all of the topics discussed here, if you haven't already, please go read this book: Michel Foucault, Discipline and Punish: The Birth of the Prison [1] The seminal book on questions like: Where do prisons c
19.
by olavolav 9y ago
The actual article may be less garbage than the headline of the news post suggests: Our model controlled for experiences with other classes of psychoactive substances (cannabis, dissociatives, empathogens, popular legal drugs) as well as co
20.
by olavolav 9y ago
That, great news, I'm especially curious about this line here: [ExUnit] Show code snippet from test source file in case of test errors ExUnit backtraces have historically been a bit hard to read sometimes IMHO.
21.
by olavolav 9y ago
So amwesome! :-) Nostalgia + sillyness. Made my morning
22.
by olavolav 10y ago
Regarding that last bit: You will clearly get better results by using A/B tests than by having humans take a guess as to what works best – but as you already pointed out, those are not the only two possibilities. Different varieties of
23.
by olavolav 10y ago
I would argue that A/B testing is close to it's end for 99% of it's current business applications. Right now there is a very limitied range of AI products that aim to replace A/B testing, but that will change soon enough