techwizrd

10 ms·

91.
by techwizrd 3y ago
Can you provide an example of where this has been successful? I've spoken with many researchers and grad students only to find that there was a critical typo in the algorithm or undescribed setting (e.g., only converges when a learning
92.
by techwizrd 3y ago
As a researcher, I've even had difficulty making sure the code matches the algorithm as described in the paper. Getting old code running and validating that it's correct so you can compare is really non-trivial.
93.
by techwizrd 3y ago
I agree. I suppose it is meant for different folks. I like that Fira Code is easy to set up, readable with good ligatures and powerline, and something I just don't need to think about.
94.
by techwizrd 3y ago
I also haven't experienced this much. I use the lazy loading functionality for vim-plug, along with async where I can. My neovim starts up in only a few milliseconds and avoids loading most plugins until I open specific file types or u
95.
by techwizrd 3y ago
I can't make this any clearer: we agree! The goal isn't maximum safety at all costs. It is finding that appropriate balance.
96.
by techwizrd 3y ago
Unfortunately, many AMEs are not quite correct in the guidance they offer (hence the AME Guide), especially as the guidance updates and improves regularly. Vasovagal syncope is not necessarily disqualifying unless a cause cannot be determin
97.
by techwizrd 3y ago
We know the system is far too conservative and encourages pilots to conceal their medical conditions, and I'm thankful you're choosing to be honest. For loss of consciousness, you can be granted a Special Issuance since you have n
98.
by techwizrd 3y ago
I help lead this work because I agree with you! And we have been working hard on this for the last couple of years (and most of the research is publicly available). Historically, aircraft are certified for a generic pilot, pilots are certif
99.
by techwizrd 3y ago
Incredible is right because it's not true [0]. That said, the current system isn't perfect and doesn't incentivise pilots to disclose or appropriately manage their conditions. 0: https://www.faa.gov/ame_guide&
100.
by techwizrd 3y ago
The FAA's Office of Aerospace Medicine is actively researching can improve or rethink the aeromedical certification process. We need to do a substantial amount of research with very limited resources, especially when we have limited da
101.
by techwizrd 3y ago
This seems quite similar to plotnine [0], which also provides a grammar of graphics interface for Python. That said, I love ggplot and I can't wait to use this in my research! I hope we can port/re-implement ggthemes, scientificpl
102.
by techwizrd 3y ago
I agree completely. I use it for inspiration and maybe a quick prototype to understand something, but I usually implement the pieces myself. Debugging LangChain performance and bugs is just an exercise in frustration.
103.
by techwizrd 3y ago
Settling for almost twice their net annual income (and similarly large lawsuits against Dupont's subsidiaries) does not strike me as the government not touching them. This is certainly a better result than the Purdue opioid lawsuit whe
104.
by techwizrd 3y ago
Going from this to "we can't trust regulators" is quite a jump. The EPA and other regulators are quickly tightening regulations, but how do we currently filter PFAS up the food chain? If anything, distrust should be placed in
105.
by techwizrd 3y ago
(Can we avoid the editorialized titles?) Relative imports and mucking with sys.path should be avoided. This StackOverflow answer is providing the right recommendation: make your code into a library. You can do this via pyproject.toml or set
106.
by techwizrd 3y ago
You should care. My point is give me coarser info or no info, but not fake information—and I say this as a researcher working on synthetic data generation methods to improve privacy. Fake information is a pain for developers _and_ users. Un
107.
by techwizrd 3y ago
I can see why this would be useful, but we already spend a lot of time chasing down data quality issues in AI/ML models and statistics as-is. We may be requesting permissions to workaround/correct GPS accuracy issues, comply with
108.
by techwizrd 3y ago
I agree that it's a little silly, but I mostly use it to abstract over BYO LLMs and extract information from documents. It's nice to be able to quickly prototype something and swap out the underlying language model than set up a w
109.
by techwizrd 3y ago
Close enough. I work for for MITRE and the FAA leading our efforts to identify aviation safety hazards and also improve aeromedical certification, so I do work closely with the airlines, OEMs, unions, trade orgs, and other stakeholders. We
110.
by techwizrd 3y ago
For example, let's say we implement support for a web standard or vendor prefix. We can mark it as deprecated. But how do we know the code path is no longer in use? Are people still using this CSS property (e.g., a vendor prefix)? Are
111.
by techwizrd 3y ago
I agree that the usability research should be transparent and data-driven. As a statistician, I'd rather have something I can actually critique instead of "we just didn't think it was user-friendly."
112.
by techwizrd 3y ago
I disagree wholeheartedly. Privacy-preserving technologies like including privacy-preserving AI (e.g., federated learning, homomorphic encryption) and privacy-preserving data linkage/fusion are really important. They're crucial in
113.
by techwizrd 3y ago
I work in aviation safety research, and there are many problems we reduce to "task overload." Pilots and ATC are under an incredible workload and often required to multiple things at once. We have more and more things to monitor a
114.
by techwizrd 3y ago
I created a Mastodon account yesterday before I even knew about this and it was surprisingly easy and user-friendly. The big thing it's missing, for me, is a way to find the Mastodon accounts for people you follow on Twitter. If more o
115.
by techwizrd 3y ago
I'm excited by the idea, but I'm honestly pretty confused by Lemmy & Kbin. Reddit accounts and subreddits give an incentive to protect the account and "build up a history". My Reddit account is 13+ years old, and I d
116.
by techwizrd 3y ago
This has definitely spooked us as well. We have numerous images based on our rocky base image, so we're looking at what moving away from all things RHEL, Rocky, etc. as well.
117.
by techwizrd 3y ago
I have YouTube Red so I'm not affected, but I really hate how it's implemented for Google accounts with your own domain. Bringing your own domain and paying Google for the privilege makes you a second-class citizen with Google ser
118.
by techwizrd 3y ago
Source for folks: https://ncees.org/ncees-discontinuing-pe-software-engineerin...
119.
by techwizrd 3y ago
Honestly, this is really wonderful news from Meta. I'm sufficiently impressed.
120.
by techwizrd 3y ago
I work for MITRE on AI assurance. I'd love to reach out!

More ›