HackerTrans
TopNewTrendsCommentsPastAskShowJobs

Dobiasd

no profile record

Submissions

[untitled]

1 points·by Dobiasd·قبل شهرين·0 comments

LLM Agents Demystified

github.com
2 points·by Dobiasd·قبل 8 أشهر·0 comments

Simple collaborative filtering in pure PostgreSQL

github.com
1 points·by Dobiasd·السنة الماضية·0 comments

Feeling deceived when receiving (non-labeled) LLM-generated messages

github.com
2 points·by Dobiasd·السنة الماضية·0 comments

Internals of the async/await pattern from first principles

github.com
57 points·by Dobiasd·قبل سنتين·15 comments

Treebomination: Convert a scikit-learn decision tree into a Keras model

github.com
79 points·by Dobiasd·قبل 3 سنوات·6 comments

comments

Dobiasd
·قبل سنتين·discuss
While I did not succeed in making the matmul code from https://github.com/Mozilla-Ocho/llamafile/blob/main/llamafil... work in isolation, I compared eigen, openblas, and mkl: https://gist.github.com/Dobiasd/e664c681c4a7933ef5d2df7caa87...

In this (very primitive!) benchmark, MKL was a bit better than eigen (~10%) on my machine (i5-6600).

Since the article https://justine.lol/matmul/ compared the new kernels with MLK, we can (by transitivity) compare the new kernels with Eigen this way, at least very roughly for this one use-case.
Dobiasd
·قبل سنتين·discuss
Are there any benchmarks on the performance of these new matrix multiplication kernels compared to the Eigen library (ideally for float32)?
Dobiasd
·قبل سنتين·discuss
https://github.com/Dobiasd/articles/commit/b5342dd5072480800...

Better? :)
Dobiasd
·قبل سنتين·discuss
Thanks for the feedback! I just extended the comment to touch on the "why" part. :) https://github.com/Dobiasd/articles/commit/53f360259ad8e64ca...
Dobiasd
·قبل سنتين·discuss
Thanks for the feedback! My code snippets in the article don't use any real/existing language. C# for example, is quite explicit with the transformation of generated to state machines, but also does not provide such methods, as far as I know. I've just added a comment explaining this choice: https://github.com/Dobiasd/articles/commit/f44b897f2a4d20aa9...
Dobiasd
·قبل سنتين·discuss
Thanks! Yeah, that was one of the intentions. :)
Dobiasd
·قبل 3 سنوات·discuss
The interview for my current job first went mediocre, but by talking about frugally-deep (a side project of mine) I was able to excite my (now) employer. :-)

https://github.com/Dobiasd/frugally-deep
Dobiasd
·قبل 3 سنوات·discuss
Hoogle is really amazing!

Inspired by it, I implemented something similar for FunctionalPlus (a functional-programming library for C++): https://www.editgym.com/fplus-api-search/

I'd love to see more projects taking this path too. :)
Dobiasd
·قبل 3 سنوات·discuss
> I spent some time over those 30 years looking at new IDEs, trying them out, configuring them, and each and every time this was time wasted, because the IDE was discontinued, abandoned, or otherwise became useless.

"wasted" sounds a bit too hard to me. If you learn and use some tool productively for some years, even if this stops, the years were not wasted. Ephemeral things can also have value.
Dobiasd
·قبل 3 سنوات·discuss
Yeah, this one does something much less insane, i.e., it converts the paths to the tree outputs into their corresponding DNS (disjunctive normal form) and represents each term as a node (side by side in the same layer) in the NN, as described by Arunava Banerjee in "Initializing Neural Networks using Decision" [1]. The resulting NN architecture is much more reasonable than the one that treebomination produces.

[1]: https://www.cise.ufl.edu/~arunava/papers/clnl94.pdf
Dobiasd
·قبل 3 سنوات·discuss
Thank! Yes, in contrast to treebomination, using TF-DF can actually make sense. ;)
Dobiasd
·قبل 3 سنوات·discuss
Thanks! This looks interesting. Some of the main differences I can spot so far are: - Hummingbird does not construct a NN with an architecture isomorphic to the source decision tree but instead cleverly compiled it into other (more sane) tensor computations. - Hummingbird is actually useful. ;)