idonov

8 ms·

31.
by idonov 4y ago
Wow that's cool, is there a v6 version already?
32.
YOLOv6 Interactive Demo (SOTA object detection)(yolov6.dagshubusercontent.com)
2 points by idonov 4y ago|0 comments
33.
The State of Data Engineering(lakefs.io)
3 points by idonov 4y ago|0 comments
34.
by idonov 4y ago
In Hebrew it sometimes shows "Bad service" as a menu item it if sees it in a lot of reviews.
35.
Ask HN: DVC, Git LFS or lakeFS? Which is the best for data versioning?
3 points by idonov 4y ago|0 comments
36.
Ask HN: How do you collaborate on ML projects?
2 points by idonov 4y ago|1 comments
37.
by idonov 4y ago
I want to work in Poland now
38.
by idonov 4y ago
They already added that to stories
39.
Ask HN: Is GitHub Good for ML?
2 points by idonov 4y ago|0 comments
40.
Show HN: DagsHub GitHub Integration – You've been reviewing ML projects wrong(dagshub.com)
2 points by idonov 4y ago|0 comments
41.
by idonov 5y ago
Is it open source? If not, is there any way to integrate with you?
42.
Datasets should behave like Git repositories (2021)(dagshub.com)
2 points by idonov 5y ago|0 comments
43.
“Men are visual creatures”(pnas.org)
3 points by idonov 5y ago|0 comments
44.
Language aware diff on GitHub PRs(difflens.com)
2 points by idonov 5y ago|0 comments
45.
by idonov 5y ago
This is awesome
46.
by idonov 5y ago
True! There is an Eliza implementation on GitHub ( https://github.com/codeanticode/eliza/blob/master/data/eliza... ) so I guess this how it knows about it.
47.
by idonov 5y ago
It does use the neighboring files for context. Check out the FAQ ( https://copilot.github.com/#faq-what-context-does-github-cop... )
48.
by idonov 5y ago
This is a really interesting use case! I'll try that, thanks for sharing :)
49.
by idonov 5y ago
Did you read the blog post? This is only one example from the first part of it. You can find more useful examples if you read a bit more. For example, if you write "Pseudo Code:" after a chunk of code, the copilot will explain th
50.
by idonov 5y ago
It's not a direct GPT-3 product. It's built on a GPT model called Codex that was trained and fine tuned for code.
51.
Github Copilot Wants to Play Chess Instead of Code(dagshub.com)
280 points by idonov 5y ago|164 comments
52.
by idonov 5y ago
It does! I didn't test it on huge functions yet but technically speaking, Codex is able to take into account about 14kb of memory, which translate to a few hundreds lines of code :)
53.
GitHub Copilot Wants to Play Chess Instead of Code(dagshub.com)
2 points by idonov 5y ago|2 comments