Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
dfhg
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
4 ms
·
1.
▲
by
dfhg
3y ago
Would you kindly elaborate a little bit the difference between training on own documents vs analyzing documents for answers?
2.
▲
by
dfhg
3y ago
Another question, which one is preferred, LlamaIndex or Langchain, for RAG? Thanks in advance for your insights.
3.
▲
by
dfhg
3y ago
Are there public examples of working products using RAG, compared with fine-tuning or training from scratch?
4.
▲
Why we're not raising funds?
(typesense.org)
3 points
by
dfhg
3y ago
|
0 comments