You can get older version of TorchCodec that work with older version of PyTorch, but it unfortunately will not have the new features (HDR video decoding; fast Wav decoding) in the latest release. See the compatbility matrix: https://github.com/meta-pytorch/torchcodec#compatibility-wit...
That's not what the author means. Multiple times a day, I have conversations with LLMs about specific code or general technologies. It is very similar to having the same conversation with a colleague. Yes, the LLM may be wrong. Which is why I'm constantly looking at the code myself to see if the explanation makes sense, or finding external docs to see if the concepts check out.
Importantly, the LLM is not writing code for me. It's explaining things, and I'm coming away with verifiable facts and conceptual frameworks I can apply to my work.
Yes, and that peer review happens through the ACM. It serves an organizing function. The conferences themselves are also in-person events, and most of the important research papers come out of those conferences.
It doesn't. arXiv is exclusively a pre-print service. The ACM digital library is for peer-reviewed, published papers. All of the peer-review happens through the ACM, as well as the physical conferences where people present and publish their papers.
IEEE may do it, as it's a professional organization. That is, they're a non-profit dedicated to the furtherance of the field. Being open access fits their mission, and the costs can be handled by dues and fees. Springer and Elsevier are for-profit publishers. I don't know how if they can have an open-access business model.
Agreed. In grad school, I used Perl to script running my benchmarks, post-process my data and generate pretty graphs for papers. It was all Perl 5 and gnuplot. Once I saw someone do the same thing with Python and matplotlib, I never looked back. I later actually started using Python professionally, as I believe lots of other people had similar epiphanies. And not just from Perl, but from different languages and domains.
I think the article's author is implicitly not considering that people who were around when Perl was popular, who were perfectly capable of "understanding" it, actively decided against it.
This is one of those times that this adage applies. In the worst case, "begging for forgiveness" does not mean asking your manager for an exception. It means losing an expensive court case with your former employer about who owns your IP.
The presumption here, as I understand is, is that a zero-click search happened because Google used one of their inline apps to answer a question. (For example, if I search "population of new york city", I get a big and bold answer along with a graph that shows NYC's population over time, as well as LA and Chicago.)
But I'm not convinced that's necessarily true. I frequently search for things to get a general idea of what is out there, and I only skim the results. This is also something I do when I'm refining my search terms. It may take me several iterations to refine my terms until I find something I want to click through to. In this process, I may have three or four no-click searches before I land on a good query, and then I start clicking.
Solutions that shell out to the `ffmpeg` binary are not going to perform well.