Viewing profile — nraynaud
nraynaud
HN member- Joined
- Wed, Mar 28, 2007, 9:37 PM UTC
- HN karma
- 3,832
- Public activity
- 2,101 items
- HN profile
- View on Hacker News ↗
About nraynaud
https://github.com/nraynaud
contact : nraynaud à gmail.com
Recent public activity
-
comment
Comment #48720427
Funny, is recently searched “match plate in French“ the LLM went straight to match sticks and dinner plates, and the dummizer in search fixed “plate“ to “play“ and went into soccer…
-
comment
Comment #48236003
To be ironic, maybe the list of the files is original :) It's a very open minded curation.
-
comment
Comment #48147549
I would venture that's it's the whole reason they use it. because it doesn't work.
-
comment
Comment #48105955
I have not clicked, but recently I was suggested a video whose title was more or less: “everybody thinks 1984 agrees with them”.
-
comment
Comment #48105410
Funny, along the same lines I asked an AI to write some wasm text. It was ridiculously bad and I had to intervene heavily to get something working as intended.
-
comment
Comment #47972137
For people interested, you can do FTS in DuckDB without using the official module. I copy-pasted the relevant parts and slightly changed it for my need. https://github.com/nraynaud…
- comment
-
comment
Comment #47971914
In France we had a case where the government tried to bring terrorist charges against someone, the problem is that the police couldn’t materially have seen what they wrote in their…
-
comment
Comment #47932727
So I gather that nobody is working on a search that stays on the current branch?
-
comment
Comment #47677451
I took it as the extracted spec was weird and they looked into it.
-
comment
Comment #47585793
I concur, “the Germans” have created an algorithm that completely “see through” subqueries/CTEs when planning a query. The way the query is written has no bearing on the execution.…
-
comment
Comment #47479091
Wouldn't she have the excuse of truth as defense?
-
comment
Comment #46994556
a continuous chain of memories is enough to consider it the same culture. They did not imply that the date was the carving date, but that the culture extended as far back as that d…
-
comment
Comment #46987602
I guess in modern time we can compute eclipses from the past?
-
comment
Comment #46987593
Sorry that was in French https://peuplesautochtones.wordpress.com/2022/05/21/sites-ar... “Il a été proposé qu’elle puisse commémorer une éclipse lunaire qui a précédé une éclipse s…
-
comment
Comment #46987286
Seems to be an eclipse at that date, if they weren’t able to predict them, they had to have seen it.
-
comment
Comment #46688967
They may not have realized their interpretation is ambiguous until after the incident, that’s the kind of stuff you realize after you find a bug and do a deep dive in the literatur…
-
comment
Comment #46334740
I would also add that there some slack for repairs.
-
comment
Comment #46132070
In France those asshole put those cars on a company books to avoid paying the CO2 overcharge and the VAT.
-
comment
Comment #46132054
Yeah, it’s a big topic in France too. I just saw the current pinacle of this stupidity: a camera in the grille at the front of a Peugeot car.
-
comment
Comment #45713724
Naive question: ML tensor libraries don’t use a Z-order memory layout like textures do? It’s not beneficial like it is for textures?
-
comment
Comment #45330989
I'm very happy that they are trying to re-classify the DSM, it's probably the sign of a maturing medical field. Let's also hope that re-clustering will help with the reproduction c…
-
comment
Comment #44887198
I think after the big training they do smaller training to change some details. I suppose they feed the system a bunch of training chat logs where the answers are warm and empathet…
-
comment
Comment #44857217
I'm a bit confused, are they suggesting that a cadnium coumpound to treat PFAS is a done deal?
-
comment
Comment #44364242
An interesting detail is that the instability had warning signs in the waveforms, that probably means that an algorithmic response is possible.