jo_

5 ms·

1.
by jo_ 5y ago
Another voice in the fray, I also did well enough to bypass the technical screen. Interviewed with four companies they recommended. Picked one. Still with them two(-ish) years later and an outspoken advocate. It's a tricky situation
2.
by jo_ 9y ago
For what kind of tagging are you looking? Per-image labels? Single-tag-per-image? Image rectangles? Most of the tagging programs I've encountered really aren't that awesome. For such a relatively common task (for ML, at least)
3.
by jo_ 9y ago
How many people do you have on your ml team so far? I'm doing ML research (NLP for determining writing quality and similarity, amusingly) for my company and it's getting a bit lonely.
4.
by jo_ 9y ago
Perhaps they were fixed as part of unrelated bug fixes? That would explain why they didn't credit anyone as having reported them. Something like, "Fixed bad return value leak," or "refactoring old method" could b
5.
by jo_ 10y ago
Forgive my ignorance, but it seems like this is just attempting to take advantage of the optimization done by LLVM, yes? What I would love is a simple way of writing standalone functions that compile into a cross-platform LLVM file that I c
6.
by jo_ 10y ago
I'm a current WebPass subscriber and I'm dreading the move to another ISP when I change apartments. May I inquire what it takes to set up and maintain the transmitter?
7.
by jo_ 11y ago
People buckle because they are victims of NEEDS. If I didn't have marketable skills, or if I existed in a market which was oversaturated with people as skilled as myself, sure, I could argue with my boss or exercise my powers, but tha
8.
by jo_ 12y ago
This looks exceptionally helpful. I have around 50 tabs open, nominally, and my CPU fan is almost perpetually on high for it. Just before installing it, however, I found myself wondering, "Why am I using an app which allows me to kee
9.
by jo_ 12y ago
From you description, you do sound kinda' like a bot. Disabled cookies. Disabled Javascript. Irregular searches. I understand the frustration with saying, "You have to have these features supported to use the product," bu
10.
by jo_ 12y ago
They might not, but that doesn't mean we should make concessions to them. I agree that it's important to consider the practical aspects of a law's use, but we shouldn't avoid making laws because some groups will ignore
11.
by jo_ 12y ago
Not just the title, but the article itself. "New research, using ultrasound, has developed an invisible 3D haptic shape that can be seen and felt."
12.
by jo_ 12y ago
I'm not sure I follow. The same reasoning kinda' applies now: I can work more than 40 hours per week to look like a better applicant, but this becomes both counterproductive and untenable at a certain point. (There are a finite
13.
by jo_ 12y ago
Nothing special. The authors have implemented a machine learning algorithm using a quantum instruction set. DWave has had open-source implementations of some machine learning algorithms for a while. This is not exactly an easy feat, but
14.
by jo_ 12y ago
I'm guilty of abhorring UI changes. Very rarely do I see them as justified. I largely prefer incremental UI changes rather than grand overhauls. The latter seems to be a minimal increase in functionality and a maximal increase in gl
15.
by jo_ 12y ago
I don't know if I'd be able to maintain my calm in those situations, but yes, even when someone is treating you like shit, we have an obligation as humans to be decent to other people. Not excellent, mind you, just basic human de
16.
by jo_ 12y ago
What is your data type for inputHash and images.phash? ByteArray? Character array? Blob?
17.
by jo_ 12y ago
I'm not sure if I value my confidentiality more than the unlimited talk/data/text plan on which I'm grandfathered. It's a hard change to make, especially considering I no longer see the tracking data after I disabl
18.
by jo_ 12y ago
I just finished writing about distance hashing functions with a slightly different angle. I visualized the distances between a bunch of images using two different techniques, one of which was pHash (discussed in the parent article). Mine
19.
by jo_ 12y ago
This makes me rather unhappy. I'm seeing this on Verizon. Can someone with an alternative mobile provider like Sprint or T-Mobile test this, too?
20.
by jo_ 12y ago
I certainly agree that startups over-promise in the beginning, but they need to be very explicit and very clear as far as human safety is concerned. I work (up until the end of the day today) for a medical startup. We speak highly of our
21.
by jo_ 12y ago
Short answer: Python and R for academic. Java or C++ for production. Julia has some followers, LUA has a Deep Learning Implementation. Python+CUDA+C++ is very common in deep learning. --- Chiming in as a machine learning researcher. My
22.
by jo_ 12y ago
Oh! Oooh! I see what you're saying. I thought you were asserting the mean _UNCORRECTED_ vision of an American Adult was 20/20. It's certainly much more plausible that the corrected vision is beyond 20/20. I would rec
23.
by jo_ 12y ago
Thank you. I think I see where you're coming from. That article has some interesting phrasing which I think is causing confusion. (For me at least.) It looks like they're using 'normal' to mean 'average in the ran
24.
by jo_ 12y ago
Could I impose for a citation for this? A quick check indicates that in the industrialized world, between 50% and 70% of individuals require corrective lenses.[1][2] That seems to indicate that the _average_ human does NOT have 20/20
25.
by jo_ 12y ago
How is this different from the heretofore prolific Word2Vec? I see they mention it but don't provide information about how it is distinct from their approach. EDIT: My fault. I was only reading through the site instead of the paper.
26.
by jo_ 12y ago
There's also an issue of anonymity and trust. There are plenty of websites from which I'd like to make purchases (some of an unflattering disposition) that may continue to bill after cancellation requests (because they know nobod
27.
by jo_ 12y ago
There are enough factors to make this answer little more than speculation. Please take it with a grain of salt and not as evidence for a course of action. In theory, the Javascript engine is significantly faster in Chromium than in Firefox
28.
by jo_ 12y ago
I'd completely forgotten about the single-payer negotiation aspect. That makes sense.
29.
by jo_ 12y ago
Thank you for the reply. I understand much better now. An intuitive understanding of the price/demand/negotiation dynamics still eludes me, but this has helped a lot.
30.
by jo_ 12y ago
I've been digging through this thread and can't seem to find any information on the impact of 'free tuition' on tuition costs. In the US, we saw an increase in loan availability result in a significant uptick in the pri

More ›